Drinking_Water_Distribution_Systems_NoRestriction.docx

This File is Uploaded B
y
4MechEngineer.com
You can Follow US
Facebook/4MechEngineer Linked-in/4MechEngineer
Instagram/4MechEngineer YouTube/4MechEngineer
Google+/4MechEngineer Twitter/4MechEngineer

Drinking Water Distribution Systems: Assessing and R
e
d
u
c
i
n
g
Risks
http://www.nap.edu/catalog/11728.html
Copyright © National Academy of Sciences. All r
i
g
h
t
s
reserved.
DRINKING WATER DISTRIBUTION SYSTEMS
A S S E S S I N G A N D R E D U C I N G R I S K S
Committee on Public Water Supply Distribution Systems:
Assessing and Reducing Risks
Water Science and Technology
Board Division on Earth and Life
Studies
THE NATIONAL ACADEMIES PRESS
Washington, D.C.
www.nap.edu

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
vi
THE NATIONAL ACADEMIES PRESS 500Fifth Street, N.W. Washington,DC 20001
NOTICE: The project that is the subject of this report was approved by the
Governing Board of the National Research Council, whose members are drawn
from the councils of the National Academy of Sciences, the National Academy
of Engineering, and the Institute of Medicine. The members of the committee
responsible for the report were chosen for their special competences and with
regard for appropriate balance.
Support for this project was provided by EPA Contract No. 68-C-03-081. A
n
y
opinions, findings, conclusions, or recommendations expressed in t
h
i
s
publication are those of the author(s) and do not necessarily reflect the views of
the organizations or agencies that provided support for the project.
International Standard Book Number-10 0-309-10306-1 (Book)
International Standard Book Number-13 978-0-309-10306-0 (Book)
International Standard Book Number-10 0-309-66432-2 (PDF)
International Standard Book Number-13 978-0-309-66432-5 (PDF)
Drinking Water Distribution Systems: Assessing and Reducing Risks is available
from the National Academies Press, 500 Fifth Street, N.W., Lockbox 2
8
5
,
Washington, DC 20055; (800) 624-6242 or (202) 334-3313 (in the
Washingtonmetropolitan area); Internet, http://www.nap.edu.
Illustration on the book cover is designed by Alight Tsai, MWH Soft.
Copyright 2006 by the National Academy of Sciences. All rights r
e
s
e
r
v
e
d
.Printed
in the United States of America

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
The National Academy of Sciences is a private, nonprofit, s
e
l
f
-
p
e
r
p
e
t
u
a
t
i
n
gsociety of
distinguished scholars engaged in scientific and engineering research,
dedicated to the furtherance of science and technology and to their use for
the general welfare. Upon the authority of the charter granted to it by the
Congress in 1863, the Academy has a mandate that requires it to advise the
federal government on scientific and technical matters. Dr. Ralph J.
Cicerone is president of the National Academy of Sciences.
The National Academy of Engineering was established in 1964, under t
h
e
charter of the National Academy of Sciences, as a parallel organization
o
f
outstanding engineers. It is autonomous in its administration and in the selection
of its members, sharing with the National Academy of Sciences the
responsibility for advising the federal government. The National Academy of
Engineering also sponsors engineering programs aimed at meeting national
needs, encourages education and research, and recognizes the superior
achievement of engineers. Dr. Wm. A. Wulf is president of the National
AcademyofEngineering.
The Institute of Medicine was established in 1970 by the National Academy o
f
Sciences to secure the services of eminent members of appropriate professions i
n
the examination of policy matters pertaining to the health of the public. T
h
e
Institute
acts under the responsibility given to the National Academy of Sciences by its
congressional charter to be an adviser to the federal government and, upon its
own initiative, to identify issues of medical care, research, and education. Dr.
Harvey V. Fineberg is president ofthe Institute ofMedicine.
The National Research Council was organized by the National Academy
o
f
Sciences in 1916 to associate the broad community of science and technology
with the Academy’s purposes of furthering knowledge and advising the
federal government. Functioning in accordance with general policies determined
by the Academy, the Council has become the principal operating agency of
both the National Academy of Sciences and the National Academy of
Engineering in providing services to the government, the public, and the
scientific and engineering communities. The Council is administered jointly by
both Academies and the Institute of Medicine. Dr. Ralph J. Cicerone and Dr.
Wm. A. Wulf are chair and vice-chair, respectively, of the National Research
Council.
www.national-academies.org

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
vi

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
COMMITTEE ON PUBLIC WATER SUPPLY DISTRIBUTION SYSTEMS:
ASSESSING AND REDUCING RISKS
VERNON L. SNOEYINK, Chair,University of Illinois, Urbana-Champaign
CHARLES N. HAAS, Vice-Chair, Drexel University, Philadelphia,
Pennsylvania
PAUL F. BOULOS, MWH Soft, Broomfield, Colorado
GARY A. BURLINGAME, Philadelphia Water Department, P
h
i
l
a
d
e
l
p
h
i
a
,
Pennsylvania
ANNE K. CAMPER, Montana State University, Bozeman
ROBERT N. CLARK, Environmental Engineering and Public Health
Consultant,Cincinnati, Ohio
MARC A. EDWARDS, Virginia Polytechnic and State University, B
l
a
c
k
s
b
u
r
gMARK
W. LECHEVALLIER, American Water, Voorhees, New Jersey
L. D. MCMULLEN, Des Moines WaterWorks, Des Moines, I
o
w
a
CHRISTINE L. MOE, Emory University, Atlanta, Georgia
EVA C. NIEMINSKI, Utah Department of Environmental Quality, Salt L
a
k
e
City
CHARLOTTE D. SMITH, Charlotte Smith and Associates,Inc., O
r
i
n
d
a
,
California
DAVID P. SPATH, California Department of Health Services (
R
e
t
i
r
e
d
)
,
Sacramento
RICHARD L. VALENTINE, University of Iowa, Iowa City
National Research Council Staff
LAURA J. EHLERS, Study Director
ELLEN A. DE GUZMAN, Research Associate
v

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
WATER SCIENCE AND TECHNOLOGY BOARD
R. RHODES TRUSSELL, Chair, Trussell Technologies, Inc., P
a
s
a
d
e
n
a
,
California
MARY JO BAEDECKER, U.S. Geological Survey (Retired), Vienna, V
i
r
g
i
n
i
a
JOAN G. EHRENFELD, Rutgers University, New Brunswick, New Jersey
DARA ENTEKHABI, Massachusetts Institute ofTechnology,Cambridge
GERALD E. GALLOWAY, Titan Corporation, Reston,Virginia
SIMON GONZALES, National Autonomous University of Mexico, Mexico
CHARLES N. HAAS, Drexel University, Philadelphia, Pennsylvania
KIMBERLY L. JONES, Howard University, Washington,DC
KAI N. LEE, Williams College, Williamstown, Massachusetts
JAMES K. MITCHELL, Virginia Polytechnic Institute and State U
n
i
v
e
r
s
i
t
y
,
Blacksburg
CHRISTINE L. MOE, Emory University, Atlanta, Georgia
ROBERT PERCIASEPE, National Audubon Society, New York, New Y
o
r
k
LEONARD SHABMAN, Resources for the Future, Washington,DC
HAME M. WATT, Independent Consultant,Washington,DC
CLAIRE WELTY, University of Maryland, Baltimore County
JAMES L. WESCOAT, JR., University of Illinois, Urbana-Champaign
GARRET P. WESTERHOFF, Malcolm Pirnie, Inc., White Plains, New Y
o
r
k
Staff
STEPHEN D. PARKER, Director
LAUREN E. ALEXANDER, Senior Staff O
f
f
i
c
e
r
LAURA J. EHLERS, Senior Staff Officer
JEFFREY W. JACOBS, Senior Staff Officer
STEPHANIE E. JOHNSON, Senior Staff O
f
f
i
c
e
r
WILLIAM S. LOGAN, Senior Staff Officer
M. JEANNE AQUILINO, Financial and Administrative A
s
s
o
c
i
a
t
e
ANITA A. HALL, Senior Program Associate
ELLEN A. DE GUZMAN, Research A
s
s
o
c
i
a
t
e
JULIE
VANO, Research Associate DOROTHY K.
WEIR, Research Associate MICHAEL J.
STOEVER, Project Assistant
vi

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
Preface
The distribution systemis a critical component of every drinking w
a
t
e
r utility.
Its primary function is to provide the required water quantity and quality at a
suitable pressure, and failure to do so is a serious system deficiency. Water
quality may degrade during distribution because of the way water is treated or
not treated before it is distributed, chemical and biological reactions that take
place in the water during distribution, reactions between the water and
distribution system materials, and contamination from external sources that
occurs because of main breaks, leaks coupled with hydraulic transients, and
improperly maintained storage facilities, among other things. Furthermore,
special problems are posed by the utility’s need to maintain suitable water
quality at the consumers tap, and the quality changes that occur in consumers’
plumbing, which is not owned or controlled by the utility.
The primary driving force for managing and regulating distribution systems is
protecting the health of the consumer, which becomes more difficult as o
u
r
nation’s distribution systems age and become more vulnerable to main b
r
e
a
k
s
and
leaks. Certainly factors that cause water of poor aesthetic quality t
o bedelivered
to the tap, or that increase the cost of delivering water, are also important.
Possibly because they are underground and out of sight, it is easy to delay
investments in distribution systems when budgets are considered. Rather than
wait for further deterioration, however, there is an urgent need for new
science that will enable cost-effective treatment for distribution, and design,
construction, and management of the distribution systemfor protection of public
health and minimization of water quality degradation.
This report was undertaken at the request of the U.S. Envi
ronmental Protection
Agency (EPA) and was prepared by the Water Science and Technology
Board (WSTB) of the National Research Council (NRC). The committee
formed by the WSTB conducted a study of water quality issues associated
with public water supply distribution systems and their potential risks to
consumers. Although the report focused on public systems that serve at least 25
people, much that is said in the report is also applicable to private, individual
distribution systems. The study considered regulations and non-regulatory
approaches to controlling quality; the health effects of distribution system
contamination; physical, hydraulic, and water quality integrity; and premise
plumbing issues. Important events that constitute health risks, such as cross
connections and backflow, pressure transients, nitrification and microbial
growth, permeation and leaching, repair and replacement of water mains, aging
infrastructure, corrosion control, and contamination in premise plumbing, were
examined. The activities of the Committee included the following tasks:
vii

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
viii PREFACE
1—As background and based on available information, identification o
f
trends
relevant to the deterioration of drinking water in water supply distribution
systems.
2—Identification and prioritization of issues of greatest concern f
o
r
distribution
systems based on review of published material.
3—Focusing on the highest priority issues as revealed by task
#2, (
a
)
evaluation of different approaches to characterizat ion of public health
risks posed by water-quality deteriorating events or conditions that may occur
in public water supply distribution systems; and (b) identification and evaluation
of the effectiveness of relevant existing codes and regulations and identification
of general actions, strategies, performance measures, and policies that could be
considered by water utilities and other stakeholders to reduce the risks posed by
water-quality deteriorating events or conditions. Case studies were identified
and recommendations were presented in their context.
4—Identification of advances in detection, monitoring and
modeling, analytical methods, information needs and technologies,
research and development opportunities, and communication strategies that
will enable the water supply industry and other stakeholders to further reduce
risks associated with public water supply distribution systems.
The Committee prepared an interim report entitled “Public Water S
u
p
p
l
y
Distribution Systems: Assessing and Reducing Risks, First Report” in March
2005 that dealt with the first two tasks listed above; the interim report has been
incorporated into this report in order to make this report a complete compilation
of Committee’s activities. The third and fourth tasks constitute the subject
matter of the present report; an explanation of where individual issues are
discussed in the report can be found at the end of Chapter 1.
The EPA is in the process of considering changes to the Total C
ol
i
form
Rule (TCR), which is one of the existing rules governing water quality in
distributions systems. This report does not include a comprehensive evaluation
of the science behind the TCR, a critique of that science, or specific suggestions
on how to change the Rule. However, the Committee believes that this report
should be considered when developing changes to the Rule, in order
to determine whether the revised Rule could better encompass distribution
system integrity.
When preparing the report the committee made a series of assumptions that
affected the outcome of the report. First, it was assumed that both treated and
distributed water has to meet U.S. water quality standards. Second, water
distribution will almost certainly be accomplished with the existing
infrastructure in which the nation has invested billions of dollars and which is
continuously being expanded. Thus, the report focuses on how to best use

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
PREFACE ix
traditionally designed distribution systems in which potable water is distributed
for all uses. These assumptions led the Committee to devote only a sm
al
l
section of the report to non-traditional distribution systemdesign (such as dual
distribution systems), investigation of which was not in the Committee’s charge.
The Committee believes that alternative methods of distributing water,
including dual distribution systems, point-of-use and point-of-entry treatment
systems, and community-based treatment systems need more research and
evaluation to determine their effectiveness and applicability, both in the United
States and elsewhere in the world. The Committee did not consider lead
and copper corrosion because this subject is part of the Lead and Copper Rule
and for this reason was intentionally excluded from the committee’s charge
by the study sponsor. Corrosion in distribution systems, in general, has very
important impacts on water quality in distribution systems, and the committee
believes that state-of-the-art internal and external corrosion control
procedures should be made available to the industry, perhaps in the form
of a manual of practice. Finally, at the request of EPA, the committee
did not consider issues surrounding the security of the nation’s distribution
systems, including potential threats and monitoring needed for security
purposes.
In developing this report, the Committee benefited greatly from the
advice and input of EPA representatives, including Ephraim King, Yu-Ting
Guilaran, Elin Betanzo, and Kenneth Rotert and from presentations by Russ
Chaney, IAPMO; Barry Fields, CDC; Johnnie Johannesen, Matt Velardes, and
Chris Kinner, Irvine Ranch Water District; Laura Jacobsen, Las Vegas Valley
Water District; Dan Kroll, HACH HST; Kathy Martel, Economic and
Engineering Services; Pankaj Parehk, LA Department of Water and Power;
Paul Schwartz, USC Foundation for Cross-Connection Control and Hydraulic
Research; and Walter J. Weber, Jr., University of Michigan. We also thank all
those who took time to share with us their perspectives and wisdom about the
various issues affecting the water resources research enterprise.
The Committee was ably served by the staff of the Water
Science a
n
d
Technology Board and its director, Stephen Parker. Study director
Laura Ehlers kept the Committee on task and on time, provided her own
valuable insights which have improved the report immeasurably, and did
a superb job of organizing and editing the report. Ellen de Guzman
provided the Committee with all manner of support in a timely and cheerful
way. This report would not have been possible without the help of these people.
This report has been reviewed in draft form by individuals chosen for thei
r
diverse perspectives and technical expertise, in accordance with p
r
o
c
e
d
u
r
e
sapproved
by the NRC’s Report Review Committee. The purpose of t
h
i
sindependent
review is to provide candid and critical comments that will assist the
institution in making its published report as sound as possible and to ensure that
the report meets institutional standards for objectivity, evidence, and
responsivenessto the study charge. The review comments and draft manuscript

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
x PREFACE
remain confidential to protect the integrity of the deliberative process.We w
i
s
h
to
thank the following individuals for their review of this report:
Gunther F. Craun, Gunther F. Craun and Associates;
Stephen Estes-Smargiassi, Massachusetts WaterResources Authority;
Timothy Ford, Montana State University;
Jerome B. Gilbert, J. Gilbert, I
n
c
.
;
Gregory J. Kirmeyer, HDR;
Michael J. McGuire, McGuire Environmental Consultants,Inc.;
Danny D. Reible, University of Texas;
Philip C. Singer, University of North Carolina;
and James Uber, University ofCincinnati.
Although the reviewers listed above have provided many constructive
comments and suggestions, they were not asked to endorse the conclusions and
recommendations nor did they see the final draft of the report before its release.
The review of this report was overseen by Edward Bouwer, Johns Hopkins
University. Appointed by the National Research Council, he was responsible
for making certain that an independent examination of this report was
carried out in accordance with institutional procedures and that all review
comments were carefully considered. Responsibility for the final content of
this report rests entirely with the authoring committee and institution.
Vernon Snoeyink,
Committee Chair

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
Contents
SUMMARY, 1
1 INTRODUCTION, 15
Introduction to Water Distribution Systems, 1
7
Distribution SystemIntegrity, 39
Impetus for the Study and Report Roadmap, 4
0
References, 43
2 REGULATIONS, NON-REGULATORY APPROACHES,
AND THEIR LIMITATIONS, 47
Regulatory Environment, 47
Limitations of Regulatory Programs, 70
Voluntary and Non-regulatory Programs that Influence
Distribution SystemIntegrity, 73
Conclusions and Recommendations, 82
References, 83
3 PUBLIC HEALTH RISK FROM DISTRIBUTION SYSTEM
CONTAMINATION, 87
Introduction to Risk, 87
Evidence from Pathogen Occurrence Measurements, 9
2
Evidence from Outbreak Data, 103
Epidemiology Studies, 112
Risks from Legionella,
125
References, 132
4 PHYSICAL INTEGRITY, 142
Factors Causing Loss of Physical Integrity, 144
Consequences ofa Loss in Physical Integrity, 154
Detecting Loss of Physical Integrity, 162
Maintaining Physical Integrity, 170
Recovering Physical Integrity, 180
References, 187
xi

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
xii CONTENTS
5 HYDRAULIC INTEGRITY, 192
Factors Causing Loss of Hydraulic Integrity, 194
Consequences ofa Loss in Hydraulic Integrity, 198
Detecting Loss of Hydraulic Integrity, 203
Maintaining Hydraulic Integrity, 206
Recovering Hydraulic Integrity, 212
References, 218
6 WATER QUALITY INTEGRITY, 221
Factors Causing Loss of Water Quality Integrity and
theirConsequences, 221
Detecting Loss of Water Quality Integrity, 237
Maintaining WaterQuality Integrity, 247
Recovering Water Quality Integrity, 252
References, 258
7 INTEGRATING APPROACHES TO REDUCING RISK
FROM DISTRIBUTION SYSTEMS, 269
Monitoring, 273
Distribution SystemModeling, 290
Data Integration, 298
Feasibility of Adopting G200 for Small Systems, 303
How to Provide Incentives to Adopt G200, 304
References, 310
8 ALTERNATIVES FOR PREMISE PLUMBING, 316
Key Characteristics of Premise Plumbing, 316
Gaps in Research and Monitoring, 323
Why Home Treatment Devices Are Not Always the Answer, 326
Policy Alternatives, 328
References, 336
ACRONYMS, 341
APPENDIXES
APPENDIX A Public Water Supply Distribution Systems:
Assessing and Reducing Risks, First Report, 3
4
5
APPENDIX B Committee Biographical Information, 386

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
Summary
Water distribution systems carry drinking water from a centralized t
r
e
a
t
-ment
plant or well supplies to consumers’ taps. These systems consist of pi
pes,
pumps, valves, storage tanks, reservoirs, meters, fittings, and other hydraulic
appurtenances. Spanning almost 1 million miles in the United States, distribu-
tion systems represent the vast majority of physical infrastructure for water sup-
plies, and thus constitute the primary management challenge from both an op-
erational and public health standpoint. Public water supplies and their distribu-
tion systems range in size from those that can serve as few as 25 people to those
that serve several million.
The issues and concerns surrounding the nation’s public water supply di
s-
tribution systems are many. Of the 34 billion gallons of water produced daily by
public water systems in the United States, approximately 63 percent is used by
residential customers. More than 80 percent of the water supplied to residences
is used for activities other than human consumption such as sanitary service and
landscape irrigation. Nonetheless, distribution systems are designed and oper-
ated to provide water of a quality acceptable for human consumption. Another
important factor is that in addition to providing drinking water, a major function
of most distribution systems is to provide adequate standby fire -flow. In order to
satisfy this need, most distribution systems use standpipes, elevated tanks,
storage reservoirs, and larger sized pipes. The effect of designing and operating
a distribution system to maintain adequate fire flow and redundant capacity is
that there are longer transit times between the treatment plant and the consumer
than would otherwise be needed.
The type and age of the pipes that make up water distribution systems r
a
n
g
e
from
cast iron pipes installed during the late 19th
century to ductile iron pipe a
n
d
finally to
plastic pipes introduced in the 1970s and beyond. Most water system
s and
distribution pipes will be reaching the end of their expected life spans in t
h
enext
30 years (although actual life spans may be longer depending on ut
i
l
i
t
y
practices and local conditions). Thus, the water industry is entering an era
where it will have to make substantial investments in pipe assessment,
repair, and replacement.
Most regulatory mandates regarding drinking water focus on enforcing w
a
-ter
quality standards at the treatment plant and not within the distribution s
y
s
-tem.
Ideally, there should be no change in the quality of treated water from thetime it
leaves the treatment plant until the time it is consumed. However, in reality
substantial changes can occur to finished water as a result of complex
physical, chemical, and biological reactions. Indeed, data on waterborne disease
outbreaks, both microbial and chemical, suggest that distribution systems remain
1

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
2 DRINKING WATER DISTRIBUTION SYSTEMS:ASSESSING AND REDUCING RISKS
a source of contamination that has yet to be fully addressed. As a consequence,
the U.S. Environmental Protection Agency (EPA) has renewed its interest in
water quality degradation occurring during distribution, with the goal of defin-
ing the extent of the problem and considering how it can be addressed during
rule revisions or via non-regulatory channels. To assist in this process, EPA
requested that the National Academies’ Water Science and Technology Board
conduct a study of water quality issues associated with public water supply dis -
tribution systems and their potential risks to consumers. The following state-
ment of task guided the expert committee formed to conduct the study:
1) Identify trends relevant to the deterioration of drinking water in w
a
t
e
r
supply distribution systems,as background and based on available information.
2) Identify and prioritize issues of greatest concern for distribution s
y
s-tems
based on a review of published material.
3) Focusing on the highest priority issues as revealed by task #2
,(a)
evaluate different approaches for characterization of public health risks posed by
water quality deteriorating events or conditions that may occur in public water
supply distribution systems; and (b) identify and evaluate the effectiveness of
relevant existing codes and regulations and identify general actions, strategies,
performance measures, and policies that could be considered by water utilities
and other stakeholders to reduce the risks posed by water-quality deteriorating
events or conditions. Case studies, either at the state or utility level, where dis-
tribution system control programs (e.g., Hazard Analysis and Critical Control
Point System, cross-connection control, etc.) have been successfully designed
and implemented will be identified and recommendations will be presented in
their context.
4) Identify advances in detection, monitoring and modeling, anal
yti
cal
methods, information needs and technologies, research and development oppor-
tunities, and communication strategies that will enable the water supply industry
and other stakeholders to further reduce risks associated with public water sup-
ply distribution systems.
The committee addressed tasks one and two in its first report, which is i
n-
cluded as Appendix A to this report. The distribution system issues given high-
est priority were those that have a recognized health risk based on clear epide-
miological and surveillance data, including cross connections and backflow;
contamination during installation, rehabilitation, and repair activities; improp-
erly maintained and operated storage facilities; and control of water quality
in premise plumbing. This report focuses on the committee’s third and fourth
tasks and makes recommendations to EPA regarding new directions and
priorities to consider.
This report considers service lines and premise plumbing to be part of the
distribution system. Premise plumbing and service lines have longer residence
times, more stagnation, lower flow conditions, and elevated temperatures com-
pared to the main distribution system, and consequently can have a profound

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
SUMMARY 3
effect on the quality of water reaching the consumer. Also, the report focuses o
n
traditional distribution system design, in which water originates from a central-
ized treatment plant or well and is then distributed through one pipe network t
o
consumers. Non-conventional distribution system designs including decentral-
ized treatment and dual distribution systems are only briefly considered. Such
designs, which would be potentially much more complicated than traditional
systems, require considerably more study regarding their economic feasibility,
their maintenance and monitoring requirements, and how to transition from an
existing conventional system to a non-conventional system. Nonetheless,
many of the report recommendations are relevant even if an alternative
distribution systemdesign is used.
REGULATORY CONSIDERATIONS
The federal regulatory framework that targets degradation of distribution
system water quality is comprised of several rules under the Safe Drinking Wa-
ter Act, including the Lead and Copper Rule (LCR), the Surface Water
Treat- ment Rule (SWTR), the Total Coliform Rule (TCR), and the
Disinfec- tants/Disinfection By-Products Rule (D/DBPR). The LCR establishes
monitoring requirements for lead and copper within tap water samples, given
concern over their leaching from premise plumbing and fixtures. The SWTR
establishes the minimum required detectable disinfectant residual and the
maximum allowed heterotrophic bacterial plate count, both measured within
the distribution system. The TCR calls for distribution system monitoring
of total coliforms, fecal coliforms, and/or E. coli. Finally, the D/DBPR
addresses the maximum disinfectant residual and concentration of disinfection
byproducts like total trihalomethanes and haloacetic acids allowed in
distribution systems. A plethora of state regulations and plumbing codes also
affect distribution system water quality, from requirements for design,
construction, operation, and maintenance of distribution systems to cross -
connection controlprograms.
Despite the existence of these rules, programs, and codes, current regulatory
programs have not removed the potential for outbreaks attributable to distribu-
tion system-related factors. Part of this can be attributed to the fact that existing
federal regulations are intended to address only certain aspects of distribution
system water quality and not the integrity of the distribution system in its total-
ity. Most contaminants that have the potential to degrade distribution
system water quality are not monitored for compliance purposes, or the
sampling requirements are too sparse and infrequent to detect
contamination events. For example, TCR monitoring encompasses only
microbiological indicators and not in real time. With the exception of
monitoring for disinfectant residuals and DBPs within the distribution system
and lead and copper at the customer’s tap, existing federal regulations do not
address otherchemical contaminants.
Although it is hoped that state regulations and local ordinances would con-
tribute to public safety from drinking water contamination in areas where federal

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
regulations are weak, the considerable variation in relevant state program
s makes
this impossible to conclude on a general basis. For cross-connection con-trol
programs, for the design, construction, operation, and maintenance of distri-
bution systems, and for plumbing code components, state programs range from
an absolute requirement to simply encouraging a practice to no provision what-
soever. Voluntary programs do exist to fill gaps in the federal and state regula-
tory requirements for distribution systemoperation and maintenance, most nota-
bly the G200 standard of the American Water Works Association. These pro-
grams, if adopted, can help a utility organize its many activities by unifying all
of the piecemeal requirements of the federal, state, and local regulations. The
following select conclusions and recommendations regarding the
effectiveness of existing regulations and codes and the potential for their
improvement are made, with additional detail found in Chapter 2.
EPA should work closely with representatives from states, water sys-
tems, and local jurisdictions to establish the elements that constitute an ac -
ceptable cross-connection control program. State requirements for cross-
connection control programs are highly inconsistent, and state oversight of such
programs varies and is subject to availability of resources. If states expect
to maintain primacy over their drinking water programs, they should adopt a
cross- connection control programthat includes a process for hazard assessment,
the selection of appropriate backflow devices, certification and training of
backflow device installers, and certification and training of backflow device
inspectors.
Existing plumbing codes should be consolidated into one uniform na-
tional code. The two principal plumbing codes that are used nationally h
a
v
e
different contents and permit different materials and devices. In addition to i
n-
tegrating the codes, efforts should be made to ensure more uniform imp lementa-
tion of the plumbing codes, which can vary significantly between jurisdictions
and have major impacts on the degree of public health protection afforded.
For utilities that desire to operate beyond regulatory requirements,
adoption of G200 or an equivalent program is recommended to help utili-
ties develop distribution system management plans. G200 has advantages
over other voluntary programs, such as HACCP, in that it is more easily adaptedto
the dynamic nature of drinking water distribution systems.
PUBLIC HEALTH RISK OF
DISTRIBUTION SYSTEM CONTAMINATION
Three primary approaches are available to better understand the hum
a
n
health risks that derive from contamination of the distribution system: risk as-
sessment methods that utilize pathogen occurrence data, waterborne disease out-
break surveillance, and epidemiology studies.Chapter 3 extensively reviews the

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
SUMMARY 5
available information in each of these categories and its implications for deter-
mining public health risk. In the case of pathogen occurrence measurements,
our understanding of the microbial ecology of distribution systems is at an early
stage. Microbial monitoring methods are expensive, time consuming, require
optimization for specific conditions, and currently are appropriate only for the
research laboratory. Methods do not exist for routine detection and quantifica -
tion of most of the microbes on the EPA’s Contaminant Candidate List. Until
better methods, dose-response relationships, and risk assessment data are avail-
able, pathogen occurrence measurements are best used in conjunction with other
supporting data on health outcomes, such as data on enhanced or syndromic
surveillance in communities, or from microbial or chemical indicators of poten-
tial contamination.
Outbreak surveillance data currently provide more information on the pub-
lic health impact of contaminated distribution systems. In fact, investigations
conducted in the last five years suggest that a substantial proportion of water-
borne disease outbreaks, both microbial and chemical, is attributable to prob-
lems within distribution systems. The reason for these observations is not clear;
outbreaks associated with distribution system deficiencies have been reported
since the surveillance system was started. However, there may be more atten-
tion focused on the distribution system now that there are fewer reported out-
breaks associated with inadequate treatment of surface water. Also, better out-
break investigations and reporting systems in some states may result in in -
creased recognition and reporting of all the risk factors contributing to the out-
break, including problems with the distribution systemthat may have been over-
looked in the past. Contamination from cross-connections and backsiphonage
were found to cause the majority of the outbreaks associated with distribution
systems, followed by contamination of water mains following breaks and con-
tamination of storage facilities. The situation may be of even greater
concern because incidents involving domestic plumbing are less recognized and
unlikely to be reported. In general the identified number of waterborne disease
outbreaks is considered an underestimate because not all outbreaks are
recognized, investigated, or reported to health authorities.
A third approach for estimating public health risk is to conduct an epidemi-
ology study that isolates the distribution system component. The body of evi-
dence from four epidemiological studies does not eliminate the consumption of
tap water that has been in the distribution system from causing increased risk of
gastrointestinal illness. However, differences between the study designs, the
study population sizes and compositions and follow-up periods, and the extent
of complementary pathogen occurrence measurements make comparisons diffi-
cult. Although all four cohort studies used similar approaches for recording
symptoms of gastrointestinal illness, different illness rates were observed, with
some more than twice as high as others. One of the major challenges for design-
ing an epidemiology study of health risks associated with water quality in the
distribution system is separating the effect of source water quality and treatment
from the effect of distribution systemwater quality.

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
Although there is a lack of definitive estimates, the available information
seems to be implicating contamination of the distribution systemin public health
risk. This is particularly true for Legionella pneumophila in water systems, for
which occurrence data, outbreak data, and epidemiological data are available. In
fact, since Legionella was incorporated into the waterborne disease outbreak
surveillance system in 2001, several outbreaks have been attributed to the mi-
croorganism, all of which occurred in large buildings or institutional settings.
As discussed in Appendix A, the committee relied on the limited available out-
break and epidemiological data as well as its best professional judgment to pri-
oritize distribution system contamination events into high, medium, and low
priority. Better public health data could help refine distribution systemrisks and
provide additional justification for the prioritization. The following select con-
clusions and recommendations regarding the public health risks of distribution
systems are made, with additional detail found in Chapter 3.
The distribution system is the remaining component of public water
supplies yet to be adequately addressed in national efforts to eradicate wa-
terborne disease. This is evident from data indicating that although the number
of waterborne disease outbreaks including those attributable to distribution sys -
tems is decreasing, the proportion of outbreaks attributable to distribution sys-
tems is increasing. Most of the reported outbreaks associated with distribution
systems have involved contamination from cross-connections and backsipho-
nage. Furthermore, Legionella appears to be a continuing risk and is the single
most common etiologic agent associated with outbreaks involving drinking wa-
ter. Initial studies suggest that the use of chloramine as a residual disinfectant
may reduce the occurrence of Legionella, but additional research is necessary
to determine the relationship between disinfectant usage and the risks of
Legionella and other pathogenic microorganisms.
Distribution system ecology is poorly understood, making risk assess-
ment via pathogen occurrence measurements difficult. There is very l
i
tt
l
e
information available about the types, activities, and distribution of microorgan-
isms in distribution systems, particularly premise plumbing. Limited heterotro-
phic plate count data are available for some systems, but these data are not r
o
u
-
tinely collected, they underestimate the numbers of organisms present, and they
include many organisms that do not necessarily present a health risk.
Epidemiology studies that specifically target the distribution system
component of waterborne disease are needed. Recently completed epidemi-
ological studies have either not focused on the specific contribution of distribu-
tion system contamination to gastrointestinal illness, or they have been unable t
o
detect any link between illness and drinking water. Epidemiological studies o
fthe
risk of endemic disease associated with drinking water distribution systems need
to be performed and must be designed with sufficient power and resources to
adequately address the deficiencies of previous studies.

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
SUMMARY 7
PHYSICAL, HYDRAULIC, AND WATER QUALITY INTEGRITY
One of the options being considered during revision of the TCR is that i
t
more
adequately address distribution system integrity—defined in this report as
having three components: (1) physical integrity, which refers to the maintenance
of a physical barrier between the distribution system interior and the external
environment, (2) hydraulic integrity, which refers to the maintenance of a desir-
able water flow, water pressure, and water age, taking both potable drinking
water and fire flow provision into account, and (3) water quality integrity, which
refers to the maintenance of finished water quality via prevention of internally
derived contamination. The three types of integrity have different causes of
their loss, different consequences once they are lost, different methods for de-
tecting and preventing a loss, and different remedies for regaining integrity.
Protection of public health requires that water professionals take all three integ-
rity types into account in order to maintain the highest level of water quality.
Physical Integrity
The loss of physical integrity of the distribution system—in which the s
y
s
-tem
no longer acts as a barrier that prevents external contamination from dete-
riorating the internal, drinking water supply—is brought about by physical a
n
d
chemical deterioration of materials, the absence or improper installation of c
r
i
t
i
-cal
components, and the installation of already contaminated components. When
physical integrity is compromised, the drinking water supply becomes
exposed to contamination that increases the risk of negative public health out-
comes. Most documented cases of waterborne disease outbreaks attributed
to distribution systems have been caused by breaches in physical integrity, such
as a backflow event through a cross connection or contamination occurring
during repair or replacement of distribution system infrastructure. Selected
conclusions and recommendations for maintaining and restoring physical
integrity to a distribution systemare given below. Additional detail is found in
Chapter4.
Storage facilities should be inspected on a regular basis. A disciplined
storage facility management program is needed that includes developing an i
n
-
ventory and background profile on all facilities, developing an evaluation a
n
d
rehabilitation schedule, developing a detailed facility inspection process, per-
forming inspections, and rehabilitating and replacing storage facilities when
needed. Depending on the nature of the water supply chemistry, every three to
five years storage facilities need to be drained, sediments need to be removed,
appropriate rust-proofing needs to be done to the metal surfaces, and repairs
need to be made to structures. These inspections are in addition to daily or
weekly inspections for vandalism, security, and water quality purposes (such as
identifying missing vents,open hatches,and leaks).

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
Better sanitary practices are needed during installation, repair, re -
placement, and rehabilitation of distribution system infrastructure. All
trades people who work with materials that are being installed or repaired a
ndthat
come in contact with potable water should be trained and certified for t
h
elevel of
sanitary and materials quality that their work demands. Quality w
o
r
k
- manship for
infrastructure materials protection as well as sanitary protection o
fwater and
materials are critical considering the increasing costs of infrastructure failure and
repair and increasingly stringent water quality standards.
External and internal corrosion should be better researched and con-
trolled in standardized ways. There is a need for new materials and c
o
r
r
o
s
i
o
nscience
to better understand how to more effectively control both external and internal
corrosion, and to match distribution system materials with the soil environment
and the quality of water with which they are in contact. At present the best
defense against corrosion relies on site-specific testing of materials, soils, and
water quality followed by the application of best practices, such as cathodic
protection. Indeed, a manual of practice for external and internal corrosion con-
trol should be developed to aid the water industry in applying what is known.
Corrosion is poorly understood and thus unpredictable in occurrence.
Insuffi- cient attention has been given to its control, especially considering its
estimated annual direct cost of $5 billion in U.S. for the main distribution
system, not counting premise plumbing.
Hydraulic Integrity
Maintaining the hydraulic integrity of distribution systems is vital to ensur-
ing that water of acceptable quality is delivered in acceptable amounts. The
most critical element of hydraulic integrity is adequate water pressure inside the
pipes. The loss of water pressure resulting from pipe breaks, significant leak-
age, excessive head loss at the pipe walls, pump or valve failures, or pres sure
surges can impair water delivery and will increase the risk of contamination of
the water supply via intrusion. Another critical hydraulic factor is the length of
time water is in the distribution system. Low flows in pipes create long
travel times, with a resulting loss of disinfectant residual as well as sections
where sediments can collect and accumulate and microbes can grow and be
protected from disinfectants. Furthermore, sediment deposition will result in
rougher pipes with reduced hydraulic capacity and increased pumping costs.
Long detention times can also greatly reduce corrosion control effectiveness by
impact- ing phosphate inhibitors and pH management. A final component of
hydraulic integrity is maintaining sufficient mixing and turnover rates in storage
facilities, which if insufficient can lead to short circuiting and generate pockets
of stagnant water with depleted disinfectant residual. Fortunately, water
utilities can achieve a high degree of hydraulic integrity through a combination
of proper system design, operation, and maintenance, along with monitoring
and model-

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
SUMMARY 9
ing. The following select conclusions and recommendations are made, w
i
t
h
additional detail found in Chapter 5.
Water residence times in pipes, storage facilities, and premise plumbing
should be minimized. Excessive residence times can lead to low disinfectant
residuals and leave certain service areas with a less protected drinking w
a
t
e
rsupply.
In addition, long residence times can promote microbial regrowth a
nd the
formation of disinfection byproducts. From an operational viewpoint it m
a
ybe
challenging to reduce residence time where the existing physical infrastructure
and energy considerations constrain a utility’s options. Furthermore, lim
- ited
understanding of the stochastic nature of water demand and water a
g
e
makes it
difficult to assess the water quality benefits of reduced residence time. Research
is needed to investigate such questions, as well as how to achieve
minimization of water residence time while maintaining other facets of hydrau-
lic integrity (such as adequate pressure and reliability of supply).
Positive water pressure should be maintained. Low pressures in the dis-
tribution systemcan result not only in insufficient fire fighting capacity but c
a
nalso
constitute a major health concern resulting from potential intrusion of con-
taminants from the surrounding external environment. A min imum residual
pressure of 20 psi under all operating conditions and at all locations (including at
the systemextremities) should bemaintained.
Distribution system monitoring and modeling are critical to maintain-
ing hydraulic integrity. Hydraulic parameters to be monitored should
include inflows/outflows and water levels for all storage tanks, discharge
flows and pressures for all pumps, flows and/or pressure for all regulating
valves, and pressures at critical points. An analysis of these patterns can
directly determine if the system hydraulic integrity is compromised. Calibrated
distribution system models can calculate the spatial and temporal variations of
flow, pressure, velocity, reservoir level, water age, and other hydraulic and
water quality parameters throughout the distribution system. Such results can,
for example, help identify areas of low or negative pressure and high water age,
estimate filling and draining cycles of storage facilities, and determine the
adequacy of the systemto sup-ply fire flows under a variety of conditions.
Water Quality Integrity
Breaches in physical and hydraulic integrity can lead to the influx of c
o
n
-
taminants across pipe walls, through breaks, and via cross connections. T
h
e
s
e
external contamination events can act as a source of inoculum, introduce nutri-
ents and sediments, or decrease disinfectant concentrations within the distribu-
tion system, resulting in a degradation of water quality. Even in the absence of
external contamination, however, there are situations where water quality is de-

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
graded due to transformations that take place within piping, tanks, and premi
se
plumbing. These include biofilm growth, nitrification, leaching, internal
corrosion, scale formation, and other chemical reactions associated with
increasing water age.
Maintaining water quality integrity in the distribution system is challenging
because of the complexity of most systems. That is, there are interactions b
e
-
tween the type and concentration of disinfectants used, corrosion contr
ol
schemes, operational practices (e.g., flow characteristics, water age, flushing
practices), the materials used for pipes and plumbing, the biological stability of
the water, and the efficacy of treatment. The following select conclusions and
recommendations are made, with additional details found in Chapter 6.
Microbial growth and biofilm development in distribution systems
should be minimized. Even though the general heterotrophs found in biofilms
are not likely to be of public health concern, their activity can promote the pro-
duction of tastes and odors, increase disinfectant demand, and may contribute to
corrosion. Biofilms may also harbor opportunistic pathogens (those
causing disease in the immunocompromised). This issue is of critical
importance in premise plumbing where long residence times promote
disinfectant decay and subsequent bacterialgrowth and release.
Residual disinfectant choices should be balanced to meet the overall
goal of protecting public health. For free chlorine, the potential residual loss
and DBP formation should be weighed against the problems that may be intro-
duced by chloramination, which include nitrification, lower disinfectant efficacy
against suspended organisms, and the potential for deleterious corrosion prob-
lems. Although some systems have demonstrated increased biofilm control with
chloramination, this response has not been universal. This ambiguity also exists
for the control of opportunisticpathogens.
Standards for materials used in distribution systems should be updated
to address their impact on water quality, and research is needed to develop
new materials that will have minimal impacts. Materials standards have his-
torically been designed to address physical/strength properties including t
he
ability to handle pressure and stress. Testing of currently available materials
should be expanded to include (1) the potential for permeation of contaminants,
and (2) the potential for leaching of compounds of public health concern as well
as those that contribute to tastes and odors and support biofilm growth. Also,
research is needed to develop new materials that minimize adverse water quality
effects such as the high concentrations of undesirable metals and deposits that
result from corrosion and the destruction of disinfectant owing to
interactions with pipe materials.

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
SUMMARY 11
INTEGRATING APPROACHES TO REDUCING
PUBLIC HEALTH RISK FROM DISTRIBUTION SYSTEMS
Because only a few regulations govern water quality in distribution s
y
s
t
e
m
s
,public
health protection from contamination arising from distribution syst
emevents will
require that utilities independently choose to design and operate t
h
e
i
rsystems
beyond regulatory requirements. One voluntary standard in parti
cu-lar—the
G200 standard for distribution system operation and management— directly
addresses the issues highlighted by EPA and characterized as high priority by
this committee (see Appendix A).
As for any voluntary program, it may be necessary to create incentives f
o
r
utilities to adopt G200, for which several options exist. An extreme would be t
o
create federal regulations that require adherence to a prescribed list of activities
deemed necessary for reducing the risk of contaminated distribution systems;
this list could partly or fully parallel the G200 standard. Another mechanism to
capture elements of G200 within existing federal regulations would be via the
sanitary surveys conducted by the state and required for some systems every
three to five years. Sanitary surveys encompass a wide variety of activities, and
could capture those felt to be of highest priority for reducing risk. Several other
options are discussed, including (1) making some of the elements of G200 fall
under existing federal regulations through the Government Accounting
Stan- dards Board, (2) state regulations that require adherence to G200
including building and plumbing codes and design and construction
requirements, (3) linking qualification for a loan from the State Revolving
Fund to a utility demon- strating that it is adhering to G200, and (4)
implementation of G200 as a way to improve a drinking water utilities’ access to
capital via betterbond ratings.
For small water systems that are resource limited, adherence to the G
2
0
0
standard or its equivalent may present financial, administrative, and technologi-
cal burdens. Thus, its adoption should occur using the following guidelines: (
1
)
implement new activities using a step-wise approach; (2) provide technical a
s
-
sistance, education, and training; and (3) develop regulatory, financial, and so-
cial incentives. Training materials, scaled for small-size systems, are e
s
s
e
n
t
i
a
l for
operators and maintenance crew. Public education can result in an increased
awareness and emphasis on the significance of implementing proactive
voluntary efforts, which could help to justify increased actions.
Certain elements of G200 deserve more thoughtful consideration because
emerging science and technology are altering whether and how these elements
are implemented by a typical water utility. Much of the current scientific thrust
is in the development of new monitoring techniques, models, and methods to
integrate monitoring data and models to inform decision making. The following
select conclusions and recommendations relate specifically to these techniques
and methods,with additional detail found in Chapter 7.
Distribution system integrity is best evaluated using on-line, real-time
methods to provide warning against any potential breaches in sufficient

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
time to effectively respond and minimize public exposure. This will require
the development of new, remotely operated sensors and data collection systems
for continuous public health surveillance monitoring. These types of syst
em
sshould
be capable of accurately (with sufficient precision) determining the na-ture, type,
and location/origin of all potential threats to distribution system integrity. The
availability, reliability, and performance of on-line monitors are improving,
with tools now available for detecting pressure, turbidity, disinfectant residual,
flow, pH, temperature, and certain chemical parameters. Although these
devices have reached the point for greater full-scale implementation, additional
research is needed to optimize the placement and number of monitors.
Research is needed to better understand how to analyze data from on-
line, real-time monitors in a distribution system. A number of companies are
selling (and utilities are deploying) multiparameter analyzers. These companies,
as well as EPA, are assessing numerical approaches to convert such data into a
specific signal (or alarm) of a contamination event—efforts which warrant fur-
ther investigation. Some of the data analysis approaches are proprietary,
and there has been limited testing reported in “real world” situations.
Furthermore, when multiple analyzers are installed in a given distribution
system, the pattern of response of these analyzers in space provides additional
information on sys- temperformance, but such spatially distributed information
has not been fully utilized. To the greatest degree possible, this research
should be conducted openly (and not in confidential or proprietary
environments).
ALTERNATIVES FOR PREMISE PLUMBING
Premise plumbing includes that portion of the distribution system associated
with schools, hospitals, public and private housing, and other buildings. It is
connected to the main distribution systemvia the service line. The quality of
potable water in premise plumbing is not ensured by EPA regulations, with the
exception of the Lead and Copper Rule which assesses the efficacy of corrosion
control by requiring that samples be collected at the tap after the water has been
allowed to remain stagnant.
Virtually every problem previously identified in the main water
transmission system can also occur in premise plumbing. However, unique
characteristics of premise plumbing can magnify the potential public health
risk relative to the main distribution system and complicate formulation of
coherent strategies to deal with problems. These characteristics include:
 a high surface area to volume ratio, which along with other factors c
a
n
lead to more severe leaching and permeation;
 variable, often advanced water age, especially in buildings that are i
r
-
regularly occupied;

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
SUMMARY 13
 more extreme temperatures than those experienced in the main di
stri
bu-
tion system
 low or no disinfectant residual, because buildings are unavoi
dable “dead
ends” in a distributionsystem;
 potentially higher bacterial levels and regrowth due to the lack of pe
r
-
sistent disinfectant residuals, high surface area, long stagnation times, and
warmer temperatures. Legionella in particular is known to colonize premise
plumbing, especially hot water heaters;
 exposure routes through vapor and bioaerosols in relatively confi
ned
spaces such as home showers;
 proximity to service lines, which have been shown to provide the gr
e
a
t
-est
number of potential entry points for pathogen intrusion;
 higher prevalence of cross connections, since it is relatively comm
onfor
untrained and unlicensed individuals to do repair work in premise plumbing;
 variable responsible party, resulting in considerable confusion over who
should maintain water quality in premise plumbing.
Premise plumbing is a contributor to the degradation of water quality, par-
ticularly due to microbial regrowth, backflow events, and contaminant intrus ion,
although additional research is needed to better understand its magnitude. In
particular, more extensive sampling of water quality within premise plumbing
by utilities or targeted sampling via research is required. The following detailed
conclusions and recommendations are given.
Communities should squarely address the problem of Legionella, both
via changes to the plumbing code and new technologies. Changes in
the plumbing code such as those considered in Canada and Australia that
involve mandated mixing valves would seem logical to prevent both
scalding and microbial regrowth in premise plumbing water systems. On -
demand water heating systems may have benefits worthy of consideration versus
traditional large hot water storage tanks in the United States. The possible
effects of chloramination and other treatments on Legionella control should be
quantified to a higher degree of certainty.
To better assess cross connections in the premise plumbing of privately
owned buildings, inspections for cross connections and other code violations
at the time of property sale could be required. Such inspection of privately
owned plumbing for obvious defects could be conducted during inspection upon
sale of buildings, thereby alerting future occupants to existing hazards and hi
gh-
lighting the need for repair. These rules, if adopted by individual states, mi
ght
also provide incentives to building owners to follow code and have repairs con-
ducted by qualified personnel, because disclosure of sub-standard repair
could affect subsequent transferof theproperty.

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
EPA shouldcreate a homeowner’s guide and website that highlights the
nature of the health threat associated with premise plumbing and mitiga-
tion strategies that can be implemented to reduce the magnitude of the risk.
As part of this guide, it should be made clear that water quality is regulated onl
y
to the property line, and beyond that point responsibility falls mainly on
consumers. Whether problems in service lines are considered to be the
homeowner’s responsibility or the water utility’s varies from systemto system.
Research is needed that specifically addresses potential problems aris-
ing from premise plumbing. This includes the collection of data quantifying
water quality degradation in representative premise plumbing systems in
geographically diverse regions and climates. In addition, greater attention
should be focused on understanding the role of plumbing materials.
Furthermore, the role of nutrients in distributed water in controlling regrowth
should be assessed for premises. Finally, the potential impacts of representative
point-of-use and point- of-entry devices need to be quantified. An
epidemiological study to assess the health risks of contaminated premise
plumbing should be undertaken in high risk communities.

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
1
Introduction
The first municipal water utility in the United States was established in B
o
s
-ton
in 1652 to provide domestic water and fire protection (Hanke, 1972). T
h
e Boston
system emulated ancient Roman water supply systems in that it was m
ul-
tipurpose in nature. Many water supplies in the United States were subsequently
constructed in cities primarily for the suppression of fires, but most have been
adapted to serve commercial and residential properties with water. By
1860, there were 136 water systems in the United States, and most of these
systems supplied water from springs low in turbidity and relatively free from
pollution (Baker, 1948). However, by the end of the nineteenth century
waterborne disease had become recognized as a serious problem in
industrialized river valleys. This led to the more routine treatment of water prior
to its distribution to consumers. Water treatment enabled a decline in the
typhoid death rate in Pitts- burgh, PA from 158 deaths per 100,000 in the
1880s to 5 per 100,000 in 1935 (Fujiwara et al., 1995). Similarly, both typhoid
case and death rates for the City of Cincinnati declined more than tenfold during
the period 1898 to 1928 due to the use of sand filtration, disinfection via
chlorination, and the application of drinking water standards (Clark et al.,
1984). It is without a doubt that water treatment in the United States has
proven to be a major contributorto ensuring the nation’s public health.
Since the late 1890s, concern over waterborne disease and uncontrolled wa-
ter pollution has regularly translated into legislation at the federal level. T
hefirst
water quality-related regulation was promulgated in 1912 under the Inter- state
Quarantine Act of 1893. At that time interstate railroads made a common cup
available for train passengers to share drinking water while on board—a
practice that was prohibited by the Act. Several sets of federal drinking
water standards were issued prior to 1962, but they too applied only to interstate
carriers (Grindler, 1967; Clark, 1978). By the 1960s, each of the states and
trust territories had established their own drinking water regulations,
although there were many inconsistencies among them. As a consequence,
reported waterborne disease outbreaks declined from 45 per 100,000 people in
1938−40 to 15 per 100,000 people in 1966−70. Unfortunately, the annual
number of waterborne disease outbreaks ceased to fall around 1951 and
may have increased slightly after that time, leading, in part, to the passage of
the Safe Drinking Wa- ter Act (SDWA) of 1974 (Clark, 1978).
Prior to the passage of the SDWA, most drinking water utilities c
onc
e
n-
trated on meeting drinking water standards at the treatment plant, even though it
had long been recognized that water quality could deteriorate in the distribution
system—the vast infrastructure downstream of the treatment plant that delivers
15

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
water to consumers. After its passage, the SDWA was interpreted by the U.S.
Environmental Protection Agency (EPA) as meaning that some federal water
quality standards should be met at various points within the distribution system
rather than at the water treatment plant discharge. This interpretation forced
water utilities to include the entire distribution systemwhen considering compli-
ance with federal law. Consequently water quality in the distribution system
became a focus of regulatory action and a major interest to drinking water utili-
ties.
EPA has promulgated many rules and regulations as a result of the SDWA
that require drinking water utilities to meet specific guidelines and numeric
standards for water quality, some of which are enforceable and collectively re -
ferred to as maximum contaminant levels (MCLs). As discussed in greater de-
tail in Chapter 2, the major rules that specifically target water quality within the
distribution system are the Lead and Copper Rule (LCR), the Surface
Water Treatment Rule (SWTR), the Total Coliform Rule (TCR), and the
Disinfec- tants/Disinfection By-Products Rule (D/DBPR). The LCR
established monitoring requirements for lead and copper within tap water
samples, given concern over their leaching frompremise plumbing and fixtures.
The SWTR establishes the minimum required detectable disinfectant
residual, or in its absence the maximum allowed heterotrophic bacterial plate
count, both measured within the distribution system. The TCR calls for the
monitoring of distribution systems for total coliforms, fecal coliforms, and/or E.
coli. Finally, the D/DBPR addresses the maximum disinfectant residual and
concentration of disinfection byproducts (DBPs) like total trihalomethanes
and haloacetic acids that are allowed in distribution systems.
Despite the existence of these rules, for a variety of reasons most c
o
n
t
a
m
i
-nants
that have the potential to degrade distribution system water quality are not
monitored for, putting into question the ability of these rules to ensure public
health protection from distribution system contamination. Furthermore, some
epidemiological and outbreak investigations conducted in the last five years
suggest that a substantial proportion of waterborne disease outbreaks, both mi-
crobial and chemical, is attributable to problems within distribution systems
(Craun and Calderon, 2001; Blackburn et al., 2004). As shown in Figure 1-1, the
proportion of waterborne disease outbreaks associated with problems in the
distribution system is increasing, although the total number of reported water-
borne disease outbreaks and the number attributable to distribution systems have
decreased since 1980. The decrease in the total number of waterborne disease
outbreaks per year is probably attributable to improved water treatment practices
and compliance with the SWTR, which reduced the risk from waterborne proto-
zoa (Pierson et al., 2001; Blackburn et al., 2004).
There is, however, no evidence that the current regulatory program has r
e
-
sulted in a diminution in the proportion of outbreaks attributable to distribution
systemrelated factors. Therefore, in 2000 the Federal Advisory Committee f
orthe
Microbial/Disinfection By-products Rule recommended that EPA evaluate
available data and research on aspects ofdistribution systems that may c
r
e
a
t
e

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
INTRODUCTION 17
FIGURE 1-1 Waterborne disease outbreaks in community w ater systems (CWS)
associated w ith distribution system deficiencies. Note that the majority of the reported
outbreaks have been in small community systems and that the absolute number of
outbreaks has decreased since 1982. SOURCE: Data from Craun and Calderon
(2001), Lee et al., (2002), and Blackburn et al. (2004).
risks to public health. Furthermore, in 2003 EPA committed to revising t
he
TCR—not only to consider updating the provisions about the frequency
and location of monitoring, follow-up monitoring after total coliform-positive
samples, and the basis of the MCL, but also to address the broader issue of
whether the TCR could be revised to encompass “distribution system integrity.”
That is, EPA is exploring the possibility of revising the TCR to provide a
comprehensive approach for addressing water quality in the distribution
system environment. To aid in this process, EPA requested the input of the
National Academies’ Wa- ter Science and Technology Board, which was asked
to conduct a study of water quality issues associated with public water supply
distribution systems and their potential risks to consumers.
INTRODUCTION TO WATER DISTRIBUTION SYSTEMS
Distribution system infrastructure is generally the major asset of a w
a
t
e
r
utility. The American Water Works Association (AWWA, 1974) defines t
hewater
distribution system as “including all water utility components for the dis-
tribution of finished or potable water by means of gravity storage feed or pumps
though distribution pumping networks to customers or other users, including
distribution equalizing storage.” These systems must also be able to provide
water for nonpotable uses,such as fire suppression and irrigation of landscaping.

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
They span almost 1 million miles in the United States (Grigg, 2005b) and i
n-
clude an estimated 154,000 finished water storage facilities (AWWA, 2003). As
the U.S. population grows and communities expand, 13,200 miles (21,239 km)
of new pipes are installed each year (Kirmeyer et al., 1994).
Because distribution systems represent the vast majority of physical i
n
f
r
a
-
structure for water supplies, they constitute the primary management challenge
from both an operational and public health standpoint. Furthermore, their repair
and replacement represent an enormous financial liability; EPA estimates t
h
e
20-year water transmission and distribution needs of the country to be $183
.6
billion, with storage facility infrastructure needs estimated at $24.8 billi
on
(EPA, 2005a).
Infrastructure
Distribution system infrastructure is generally considered to consist of t
h
e
pipes, pumps, valves, storage tanks, reservoirs, meters, fittings, and other h
y
-
draulic appurtenances that connect treatment plants or well supplies to consum-
ers’ taps. The characteristics, general maintenance requirements, and desirable
features of the basic infrastructure components in a drinking water distribution
systemare briefly discussed below.
Pipes
The systems of pipes that transport water fromthe source (such as a trea
t-ment
plant) to the customer are often categorized from largest to smallest as
transmission or trunk mains, distribution mains, service lines, and premise
plumbing. Transmission or trunk mains usually convey large amounts of water
over long distances such as froma treatment facility to a storage tank within the
distribution system. Distribution mains are typically smaller in diameter than the
transmission mains and generally follow the city streets. Service lines carry
water from the distribution main to the building or property being served. Ser-
vice lines can be of any size depending on how much water is required to serve a
particular customer and are sized so that the utility’s design pressure is main-
tained at the customer’s property for the desired flows. Premise plumbing refers
to the piping within a building or home that distributes water to the point of use.
In premise plumbing the pipe diameters are usually comparatively small, leading
to a greater surface-to-volume ratio than in other distribution systempipes.
The three requirements for a pipe include its ability to deliver the
quantity of water required, to resist all external and internal forces acting upon it,
and to be durable and have a long life (Clark and Tippen, 1990). The materials
commonly used to accomplish these goals today are ductile iron, pre -stressed
concrete, polyvinyl chloride (PVC), reinforced plastic, and steel. In the
past, unlined cast iron and asbestos cement pipes were frequently installed in
distribu-

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
INTRODUCTION 19
tion systems, and thus are important components of existing systems (see Fi
gure
1-2). Transmission mains are frequently 24 inches (61 cm) in diameter or
greater, dual-purpose mains (which are used for both transmission and distribu-
tion) are normally 16–20 inches (40.6–50.8 cm) in diameter, and
distribution mains are usually 4–12 inches (10.0–30.5 cm) in diameter.
Service lines and premise plumbing may be of virtually any material and are
usually 1 inch (2.54 cm) in diameter or smaller (Panguluri et al., 2005).
It should be noted that this report considers service lines and pr
e
m
i
seplumbing
to be part of the distribution system, and it considers the effects of service
lines and premise plumbing on drinking water quality. If premise plumbing
is included, the figure for total distribution system length would increase
from almost 1 million miles (Grigg, 2005b) to greater than 6 million miles
(Edwards et al., 2003). Premise plumbing and service lines have longer
residence times, more stagnation, lower flow conditions, and elevated tempera-
tures compared to the main distribution system (Berger et al., 2000). Inclusion
of premise plumbing and service lines in the definition of a public water supply
distribution system is not common because of their variable ownership, which
ultimately affects who takes responsibility for their maintenance. Most drinking
water utilities and regulatory bodies only take responsibility for the water deliv -
ered to the curb stop, which generally captures only a portion of the service line.
The portion of the service line not under control of the utility and all of the
premise plumbing are entirely the building owner’s responsibility.
Pipe-Network Configurations
The two basic configurations for most water distribution systems are t
h
e
branch
and grid/loop (see Figure 1-3). A branch system is similar to that of a tree
branch, in which smaller pipes branch off larger pipes throughout the service
area, such that the water can take only one pathway from the source to the
consumer. This type of system is most frequently used in rural areas. A
grid/looped system, which consists of connected pipe loops throughout the area
to be served, is the most widely used configuration in large municipal areas. In
this type of system there are several pathways that the water can follow fromthe
source to the consumer. Looped systems provide a high degree of
reliability should a line break occur because the break can be isolated with little
impact on consumers outside the immediate area (Clark and Tippen, 1990;
Clark et al., 2004). Also, by keeping water moving looping reduces some of
the problems associated with water stagnation, such as adverse reactions with
the pipe walls, and it increases fire-fighting capability. However, loops can be
dead-ends, especially in suburban areas like cul-de-sacs, and have
associated water quality problems. Most systems are a combination of both
looped and branched portions.
Design of water networks is very much dependent on the specific topogra-
phy and the street layout in a given community. A typical design might consist

i
g
h
t
s
reserved.
MATERIAL JOINT Corrosion Protection
1990s 1910s 1920s 1930s 1940s 1950s 1960s 1970s 1980s 1990s
INTERIOR EXTERIOR
Steel Welded None None
Steel Welded Cement None
Cast Iron (pit c
a
s
t
) Lead None None
Cast Iron Lead None None
Cast Iron Lead Cement None
Cast Iron Leadite None None
Cast Iron Leadite Cement None
Cast Iron Rubber Cement None
Ductile Iron Rubber Cement None
Ductile Iron Rubber Cement
PE
Encasement
Asbestos Cem
ent Rubber Material Material
Reinforced
Concrete (RCP)
Rubber Material Material
Prestressed
Concrete (RCP)
Polyvinyl C
h
l
o
r
i
d
e Rubber Material Material
High Density
Polyethylene
Fused Material Material
Molecular Oriented
PVC
Legends: Commercially Available Predominantly in Use
FIGURE 1-2 Timeline of pipe technology in the United States. SOURCE: Reprinted, w ith permission, fromAWWSC (2002). © 2
0
0
2
American Water.
20

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
INTRODUCTION 21
A B
FIGURE 1-3 Tw o Basic Configurations for Water Distribution Systems. (
A
)
Branched con-
figuration. (B) Looped configuration.
of transmission mains spaced from 1.5 to 2 miles (2,400 to 3,200 m) apart with
dual-service mains spaced 3,000 to 4,000 feet (900 to 1,200 m) apart. Service
mains should be located in every street.
Storage Tanksand Reservoirs
Storage tanks and reservoirs are used to provide storage capacity to m
e
e
t
fluctuations in demand (or shave off peaks), to provide reserve supply for f
i
r
e
-
fighting use and emergency needs, to stabilize pressures in the distribution sys-
tem, to increase operating convenience and provide flexibility in pumping,
to provide water during source or pump failures, and to blend different
water sources. The recommended location of a storage tank is just beyond the
center of demand in the service area (AWWA, 1998). Elevated tanks are used
most frequently, but other types of tanks and reservoirs include in-ground tanks
and open or closed reservoirs. Common tank materials include concrete and
steel.
An issue that has drawn a great deal of interest is the problem of low w
a
ter
turnover in these facilities resulting in long detention times. Much of the w
a
t
e
r
volume in storage tanks is dedicated to fire protection, and unless utilities prop-
erly manage their tanks to control water quality, there can be problems attribut-
able to both water aging and inadequate water mixing. Excessive water age c
a
nbe
conducive to depletion of the disinfectant residual, leading to biofilm growth,
other biological changes in the water including nitrification, and the emergence
of taste and odor problems. Improper mixing can lead to stratification and large
stagnant (dead) zones within the bulk water volume that have depleted disinfec-
tant residual. As discussed later in this report, neither historical designs nor op-
erational procedures have adequately maintained high water quality in storage
tanks (Clark et al., 1996).

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
Security is an important issue with both storage tanks and pumps because
o
f
their potential use as a point of entry for deliberate contamination of distribution
systems.
Pumps
Pumps are used to impart energy to the water in order to boost it to hi
gher
elevations or to increase pressure. Pumps are typically made from steel or cast
iron. Most pumps used in distribution systems are centrifugal in nature, in that
water from an intake pipe enters the pump through the action of a
“spinning impeller” where it is discharged outward between vanes and into the
discharge piping. The cost of power for pumping constitutes one of the major
operating costs for a water supply.
Valves
The two types of valves generally utilized in a water distribution system a
r
e
isolation valves (or stop or shutoff valves) and control valves. Isolation valves
(typically either gate valves or butterfly valves) are used to isolate sections for
maintenance and repair and are located so that the areas isolated will cause
a minimum of inconvenience to other service areas. Maintenance of the valves is
one of the major activities carried out by a utility. Many utilities have a regular
valve-turning program in which a percentage of the valves are opened and
closed on a regular basis. It is desirable to turn each valve in the system at least
once per year. The implementation of such a program ensures that water can be
shut off or diverted when needed, especially during an emergency, and that
valves have not been inadvertently closed.
Control valves are used to control the flow or pressure in a distribution sys-
tem. They are normally sized based on the desired maximum and minimum
flow rates, the upstream and downstream pressure differentials, and the flow
velocities. Typical types of control valves include pressure-reducing, pressure-
sustaining, and pressure-relief valves; flow-control valves; throttling valves;
float valves; and check valves.
Most valves are either steel or cast iron, although those found in pr
em
i
se
plumbing to allow for easy shut-off in the event of repairs are usually b
r
a
s
s
.
They exist throughout the distribution system and are more widely spaced in t
he
transmission mains compared to the smaller-diameter pipes.
Other appurtenances in a water system include blow-off and a
i
r
-
release/vacuum valves, which are used to flush water mains and release
en- trained air. On transmission mains, blow-off valves are typically located
at every low point, and an air release/vacuum valve at every high point on
the main. Blow-off valves are sometimes located near dead ends where water
can

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
INTRODUCTION 23
stagnate orwhere rust and other debris can accumulate. Care must be taken
a
t
these locations to prevent unprotected connections to sanitary or storm sewers.
Hydrants
Hydrants are primarily part of the fire fighting aspect of a water s
y
s
t
e
m
.Proper
design, spacing, and maintenance are needed to insure an adequate f
l
o
w to satisfy
fire-fighting requirements. Fire hydrants are typically exercised and tested
annually by water utility or fire department personnel. Fire flow tests are
conducted periodically to satisfy the requirements of the Insurance Services Of-
fice or as part of a water distribution system calibration program (ISO,
1980). Fire hydrants are installed in areas that are easily accessible by fire
fighters and are not obstacles to pedestrians and vehicles. In addition to being
used for fire fighting, hydrants are also for routine flushing programs,
emergency flushing, preventive flushing, testing and corrective action, and for
street cleaning and construction projects (AWWA, 1986).
Infrastructure Designand Operation
The function of a water distribution system is to deliver water to all custom-
ers of the system in sufficient quantity for potable drinking water and fire p
r
o
-
tection purposes, at the appropriate pressure, with minimal loss, of safe and a
c
-
ceptable quality, and as economically as possible. To convey water, pum
p
s
must provide working pressures, pipes must carry sufficient water, storage
facilities must hold the water, and valves must open and close properly.
Indeed, the carrying capacity of a water distribution system is defined as its
ability to supply adequate water quantity and maintain adequate pressure
(Male and Walski, 1991). Adequate pressure is defined in terms of the
minimum and maximum design pressure supplied to customers under specific
demand conditions. The maximum pressure is normally in the range of 80 to
100 psi; for example, the Uniform Plumbing Code requires that water pressure
not exceed 80 psi (552 kPa) at service connections, unless the service is
provided with a pressure-reducing valve. The minimum pressure during peak
hours is typically in the range of 40 to 50 psi (276–345 kPa), while the
recommended minimum pressure during fire flow is 20 psi (138 kPa).
Residential Drinking Water Provision
Of the 34 billion gallons of water produced daily by public water systems i
n
the United States, approximately 63 percent is used by residential customers f
or
indoor and outdoor purposes. Mayer et al. (1999) evaluated 1,188 homes from
14 cities across six regions of North America and found that 42 percent of an-

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
nual residential water use was for indoor purposes and 58 percent for outdoor
purposes. Outdoor water use varies quite significantly from region to region a
n
d
includes irrigation. Of the indoor water use, less than 20 percent is for c
o
n-
sumption or related activities, as shown below:
Human Consumption or Related Use – 17.1 %…… Faucet use – 15.7 %
Dishwasher – 1.4 %
Human Contact Only – 18.5 %…………………… Shower – 16.8 %
Bath – 1.7 %
Non-Human Ingestion or Contact Uses – 64.3 %… Toilet – 26.7 %
Clothes Washer– 21.7 %
Leaks – 13.7 %
Other – 2.2 %
Most of the water supplied to residences is used primarily for laundering, show-
ering, lawn watering, flushing toilets, or washing cars, and not for consumption.
Nonetheless, except in a few rare circumstances, distribution systems are as-
sumed to be designed and operated to provide water of a quality acceptable for
human consumption. Normal household use is generally in the range of
200 gallons per day (757 L per day) with a typical flow rate of 2 to 20 gallons
per minute (gpm) [7.57–75.7 L per minute (Lpm)]; fire flow can be orders of
magnitude greater than these levels, as discussed below.
Fire Flow Provision
Besides providing drinking water, a major function of most distribution sys-
tems is to provide adequate standby fire flow, the standards for which are gov-
erned by the National Fire Protection Association (NFPA, 1986). Fire -flow re-
quirements for a single family house vary from 750 to 1,500 gpm (2,839–5,678
Lpm); for multi-family structures the values range from 2,000 to 5,000 gpm
(7,570–18,927 Lpm); for commercial structures the values range from 2,000 to
10,000 gpm (7,570–37,854 Lpm), and for industrial structures the values range
from 3,000 to over 10,000 gpm (11,356–37,854 Lpm) (AWWA, 1998). The
duration for which these fire flows must be sustained normally ranges from three
to eight hours.
In order to satisfy this need for adequate standby capacity and pr
e
ssur
e
, most
distribution systems use standpipes, elevated tanks, and large storage res-
ervoirs. Furthermore, the sizing of water mains is partly based on fire protection
requirements set by the Insurance Services Office (AWWA, 1986; Von Huben,
1999). (The minimum flow that the water system can sustain for a specific pe-
riod of time governs its fire protection rating, which then is used to set the fire
insurance rates for the communities that are served by the system.) As a conse-

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
INTRODUCTION 25
quence, fire-flow governs much of the design of a distribution system, especially
for smaller systems. A study conducted by the American Water Works Associa-
tion Research Foundation confirmed the impact of fire-flow capacity on the
operation of, and the water quality in, drinking water networks (Snyder et
al., 2002). It found that although the amount of water used for fire fighting is
generally a small percentage of the annual water consumed, the required rates
of water delivery for fire fighting have a significant and quantifiable impact on
the size of water mains, tank storage volumes, water age, and operating and
maintenance costs. Generally nearly 75 percent of the capacity of a typical
drinking water distribution system is devoted to fire fighting (Walski et al.,
2001).
The effect of designing and operating a system to maintain adequate f
i
r
e
flow
and redundant capacity is that there are long transit times between the
treatment plant and the consumer, which may be detrimental to meeting drinking
water MCLs (Clark and Grayman, 1998; Brandt et al., 2004). Snyder et al.
(2002) recommended that water systems evaluate existing storage tanks to de-
termine if modification or elimination of the tanks was feasible. Water efficient
fire suppression technologies exist that use less water than conventional
standards. In particular, the universal application of automatic sprinkler
systems provides the most proven method for reducing loss of life and
property due to fire, while at the same time providing faster response to the fire
and requiring significantly less water than conventional fire -fighting techniques.
Snyder et al. (2002) also recommended that the universal application of
automatic fire sprin- klers be adopted by local jurisdictions for homes as well as
in otherbuildings.
There is a growing recognition that embedded designs in most urban a
r
e
a
s have
resulted in distribution systems that have long water residence times due t
othe
large amounts of storage required for fire fighting capacity. More than ten years
ago, Clark and Grayman (1992) expressed concern that long residence times
resulting from excess capacity for fire fighting and other municipal uses would
also provide optimum conditions for the formation of DBPs and the regrowth
of microorganisms. They hypothesized that eventually the drinking water
industry would be in conflict over protecting public health and protecting
public safety.
Non-conventional water distribution system designs that might addre
sssome
of these issues are discussed below including decentralized treatment, d
u
a
l
distribution systems, and an approach that utilizes enhanced treatment to solve
distribution system water quality problems. These alternative concepts were not
part of the committee’s statement of task, such that addressing them extensively
is beyond the scope of the report. However, their potential future role in abating
the problems discussed above warrants mention here and further consideration
by EPA and water utilities.
Decentralized Treatment
Distributed or decentralized treatment systems refer to those in which a cen-

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
tralized treatment plant is augmented with additional treatment units that a
r
e
located
at various key points throughout the distribution system. Usually, the
distributed units provide advanced treatment to meet stringent water quality re-
quirements at consumer endpoints that would otherwise be in violation. Distrib -
uted units would be located either at the point-of-entry of households, for exam-
ple, or at a more upstream location from which different water use could be
served. This might be at the neighborhood or district level, depending on tech-
nological and financial requirements.
How the decentralized treatment concept might be implemented in w
a
t
e
r
systems worldwide is still at a theoretical stage (e.g., Norton, 2006 and Weber,
2002, 2004). Weber’s approach involves having distributed networks (Distrib -
uted Optimal Technologies Networks or DOT-Nets) in which water supply
is optimized by separately treating several components of water and
wastewater streams using decentralized treatment units. The approach largely
views water supply, treatment, and waste disposal as different aspects of the
same integrated system. Box 1-1 describes the concepts in detail.
BOX 1-1
Distributed Optimal Technologies Networks
DOT-Net is a decentralized treatment concept in which water suppl
i
es are segregated
based on uses (or use functions) and levels of quality, to w hich a qualitative ranking on a
scale of 1 to 10 is assigned, w ith 1 being the best quality and 10 the worst. The use func-
tions include potable water, black water, gray water, various industrial discharges, etc. For
example, w ater extracted from a local surface water source might be given a rank of 6.
Follow ing centralized treatment, the water would have a rank of 2. There is then an as-
sumption that this supply will be degraded in distribution systems to a level that is generally
not acceptable as potable w ater (say 3). To address this, advanced treatment technologies
such as membranes and super-critical water treatment w ould be located as satellite
systems close to the point of use, producing a w ater of ranking 1.
This concept hinges upon segregating water into the various u
s
efunctions and devel-
oping and deploying the technology needed to bring about the desired w ater quality for
each function. For example, w ater for drinking, showering, and cooking would require the
highest level of quality and should be treated appropriately using satellite systems and
advanced technologies. Advanced technologies exist for the treatment, analysis, and con-
trol of personal water including sophisticated electromechanical systems for rapid monitor-
ing and feedback. The existing distribution systemw ould still be used, but w ould be sup-
plemented w ith treatment units to treat a portion of the water supply. For example, satellite
treatment units may be located in large buildings with a high population density or distrib-
uted over neighborhoods.
The concept extends to the waste streams generated by each t
y
p
e of water use, as
shown in Figure 1-4. Thus, advanced water treatment would be used not only prior to wa-
ter delivery, but also upon water disposal but before it is discharged into a centralized col-
lection system. For example if a certain commercial enterprise produced a highly degraded
waste stream (with a ranking of 10), a satellite unit could be used to raise the quality to that
of the other common w aste streams (say 7). Such advanced and other wastewater treat-
ment w ould be implemented in a manner to eventually resupply the source w aters or to
continues

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
Residential
INTRODUCTION 27
The principal trade-off associated with utilizing such systems is the alterna-
tive cost associated with upgrading large centralized treatment facilities and dis-
tribution networks. On a per person basis, it is less expensive to build one large
treatment system than to build several small ones. In addition to these
costs, multiple or new pipe networks are a necessary part of the design
framework for these satellite systems. That is, new piping would be needed
from the advanced water treatment system into the household (or industry),
although it would travel a short distance and would be a small percentage of the
total plumbing for the building. It is possible that investing in larger satellite
systems with separate piping might offer a cost advantage compared to small
satellite systems, based on economies of scale (Norton, 2006). Clearly, there
would have to be a policy to avoid social injustice such that decentralized
treatment when implemented is affordable to the average user.
BOX 1-1 Continued
producewater for another use function (e.g., recreational or i
n
d
u
s
t
r
i
a
luse). Also envisioned is
the potential recovery of energy from the treatment of black water as well as some
industrial sources.
Collection System
Use Function
Distribution System
7 1 AWT 3
7 EXP 10 3
1 AWT 3
7
7
Central 5
5
EXP
WTP
1 AWT 3
2
4
WWTP 6 Central
WTP
FIGURE 1-4 Distributed Optimal Technologies Netw orks. SOURCE: R
e
p
r
i
n
t
e
d
,w ith permission,
fromWeber (2005). © 2005 by Weber.
Recreational B
Industrial
Natural Surface and/or Subsurface Raw Water Source(s)
Commercial &
Recreational A
Strategic Storage and
ConditioningReservoir(s)

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
A second important consideration is the need to monitor the satellite s
y
s
-
tems, and whose responsibility that monitoring would be. Maintenance activi-
ties, such as repair and replacement of a new piping system associated with sat-
ellite treatment, would also have to be well planned in order to prevent contami-
nation of the distribution system downstream of the treatment unit. Incorpora-
tion of remote control technologies and other monitoring adaptations could re-
duce the need for human intervention while ensuring that the units operate satis -
factorily.
The decentralized treatment concept was tested on a limited basis in the
field (from October 1995 to September 1996) by Lyonnaise des Eaux-CIRSEE
in the municipality of Dampierre, France. A one-year study was carried
out using an ultrafiltration/nanofiltration system to treat water for 121
homes through 13,123 ft (4,000 m) of pipe at an average flow of 22.0 gpm (5
m3
/h) and a peak flow of 44.0 gpm (10 m3
/h). The ultrafiltration/nanofiltration
system was fully automatic and monitored by remote control. Results from the
study were very satisfactory from a quality perspective, and the cost
calculations showed that the system was cost competitive with centralized
treatment if production volumes were greater than 5,284,020 gal/year
(20,000 m3
/year) (Levi et al., 1997).
A more prospective example is provided by the Las Vegas Valley W
a
t
e
r
District (LVVWD) and the Southern Nevada Water Authority (SNWA), whi
ch
serve one of the most rapidly growing areas in the United States (see Box 1
-
2
)
.
Because of concerns over proposed MCLs for DBPs and the compliance frame-
work being established by the Stage 2 D/DBPR, Las Vegas is investigating t
he
application of decentralized or satellite water treatment systems within its distri-
bution network. Currently only about 10 percent of the network is having trou-
ble with compliance but it is anticipated as the system expands, more and more
of the network will be out of compliance.
Enhanced Treatment
A third approach to slowing water quality deterioration involves centralized
treatment options that can improve the quality of water to such a degree
that formation of DBPs and loss of disinfectant residual are minimized.
This approach is practiced by the Greater Cincinnati Water Works, which
serves a large metropolitan area consisting of urban and suburban areas with
potable water and fire flow protection. The distribution system is served by two
treatment plants, the largest being the Miller Plant, which has a design
capacity of 220 mgd (833,000 m3
/day) with an average water production of
133 mgd (428,000 m3
/day). The Miller Plant (Figure 1-5) has 12 granular
activated carbon contactors, each containing 21,000 ft3
(600 m3
) of GAC.
During normal plant operation, between seven and 11 of these contactors are
used (in parallel) to process water. Once GAC becomes spent, it is reactivated
using an on-site reactivation system. GAC treatment reduces total organic
carbon (TOC) levels from an

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
INTRODUCTION 29
BOX 1-2
Application of DecentralizedTreatmentwithin the Las Vegas Valley Water District
Betw een 1989 and 2004, Las Vegas grew faster than any other m
etropolitanarea in the
U.S. As a result, during this period LVVW D has more than doubled its service area
population. In 1989, the service area population was 558,000 but by 2004 it had grown to
1,209,000 (Jacobsen and Kamojjala, 2005).
The LVVWD receives its water on a wholesale basis from the S
o
u
t
h
e
r
n
Nevada Water Authority
(SNWA), which operates two water treatment plants with a combined design capacity of
900 mgd (3.41 mil m3
per day). The source of water is Lake Mead. The treatment train at
both plants is nearly identical, consisting of ozone and direct filtration. Chlorine is utilized as
the finaldisinfectant. Coagulation dosages are limited and TOC removals
through the biologically active filters range from 10 to 30 percent. The distribution system
consists of 3,300 miles (5,280 km) of pipe and 29 water storage reservoirs. The system
experiences long residence times (in some case greater than a w eek), resulting in an in-
crease in water temperature as it moves through the system. Consequently it is difficult to
maintain chlorine residuals in some parts of the system, necessitating the addition of chlo-
rine at many locations. Currently the system is in compliance with all Safe Drinking Water
Act regulations. However, based on distribution system hydraulic modeling estimates of
detention time and know n formation rates for trihalomethanes and haloacetic acids, it is
expected that some areas in the LVVWD will not comply with the DBP MCLs and compli-
ance framework being established by the Stage 2 D/DBPR. In order to meet Stage 2 regu-
lations, the LVVWD/SNWA evaluated several alternatives to change its treatment and/or its
residual disinfectant. Advanced oxidation, granular activated carbon (GAC) adsorption,
enhanced coagulation (including addition of clarification), and nanofiltration were consid-
ered possible changes to treatment that could be helpful. In addition, operational and
residual disinfectant changes were considered such as conversion from free chlorine to
chloramine and a reduction in distribution system detention time. A more unconventional
option considered by LVVWD/SNWA evaluated the potential for targeted or “hot-spot”
treatment using several smaller-scale treatment systems that would reduce the
concentration of DBPs in those areas of the distribution system that might exceed the
MCLs established by the Stage 2 D/DBPR.
Ultimately, the LVVWD/SNWA chose to use the “hot-spot” treatment approach for the
follow ing reasons. It w ould provide a cost-effective approach by only treating water where
needed at specific locations, instead of treating water for the entire system. It w ould reduce
residuals production from treatment as compared to intensive organics removal. And it
would provide for the continuous use of chlorine and avoid potential nitrification problems.
The decentralized treatment options being considered are (1) DBP and natural organic
material (NOM) removal by GAC adsorption, (2) DBP and NOM removal by biologically
active carbon (BAC), and (3) control of DBP reformation after treatment by GAC and BAC.
The American Water Works Association Research Foundation has funded a project thatw ill
test the concept of decentralized treatment and its application to the LVVWD (Jacobsen et
al., 2005).

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
DRINKING WATER DISTRIBUTION SYSTEMS: ASSESSING AND REDUCING RISKS
30
FIGURE
1-5
Schematic
of
the
GCWW
Ohio
River
Treatment
Plant
(
M
iller
Plant).

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
INTRODUCTION 31
annual average of 1.5 mg/L prior to GAC treatment to a combined f
i
v
e
-
y
e
a
r
average of 0.6 mg/L after GAC treatment. The plant is one of the world’s l
a
r
g
-est
municipal GAC potable water treatment systems (Moore et al., 2003).
Las Vegas and Cincinnati have chosen distinctly different approaches t
o
meeting and solving the residence time and excess capacity problem. The Las
Vegas system is currently conducting studies to explore the possibility of apply-
ing treatment at various locations. Decentralized treatment units would be
installed at points in the system where DBPs might exceed the Stage 2
D/DBPR. Research is being conducted that would focus primarily on removing
the precursor material in the water in order to keep the DBP formation
potential below regulated limits. A key aspect of this strategy is to use
distribution system models and GIS technology to monitor residence time and
DBP formation potential in the system. Cincinnati, on the other hand, has
chosen the more traditional but very effective approach of removing DBP
precursor material prior to distribution and thereby minimizing the potential
formation of DBPs throughout the system. Although the Greater Cincinnati
Water Works has a very large distribution system composed of a wide variety
of pipe materials, the utility routinely provides water well below the total
trihalomethane level of 80 g/L and the total haloacetic acid level of 60 g/L at
all locations in thesystem.
Dual Distribution Systems
Another option for design and operation of distribution systems is the cre
a-
tion of dual systems in which separate pipe networks are constructed for potabl
e
and nonpotable water. In these types of systems, reclaimed wastewater or waterof
sub-potable quality may be used for fire fighting and other special purposes
such as irrigation of lawns, parks, roadway borders and medians; air condition-
ing and industrial cooling towers; stack gas scrubbing; industrial processing;
toilet and urinal flushing; construction; cleansing and maintenance, including
vehicle washing; scenic waters and fountains; and environmental and recrea-
tional purposes. The design of these systems differentiates dual systems from
most community water supplies, in which one distribution system provides po-
table water to serve all purposes.
Most dual systems in use today were installed by adding reclaimed waterlines
alongside (but not connected to) potable water lines already in place. F
o
rexample,
in St. Petersburg, Florida, a reclaimed water distribution system w
a
s placed into
operation in 1976, and fire protection is provided from both the p
o
- table and
reclaimed water lines. San Francisco has a nonpotable system, c
o
n
-structed after
the 1906 earthquake, that serves the downtown area to augm
ent fire
protection. Rouse Hill, Australia was the first community to plan a dual
water system with the reclaimed water lines to serve all nonpotable uses, includ-
ing fire protection, such that the potable water line can have much smaller pipe
diameters. Both the potable and nonpotable systems have service reservoirs for

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
meeting diurnal variations in demand, and if a shortage of water for fire protec-
tion occurs, potable water can be transferred to the nonpotable system.
In a recent exchange of letters in the Journal of the American Water W
o
r
k
s
Association, Dr. Dan Okun (Okun, 2005) and Dr. Neil Grigg (Grigg, 2
005
a
)
addressed the merits of dual distribution systems for U.S. drinking water utili-
ties, especially given that ingestion and human consumption are minor uses in
most urban areas (see above). The argument is that because existing water dis -
tribution systems are designed primarily for fire protection, the majority of the
distribution systemuses pipes that are much larger than would be needed if the
water was intended only for personal use. This leads to residence times of
weeks in traditional systems versus potentially hours in a system comprised of
much smaller pipes. In the absence of smaller sized distribution systems, utili-
ties have had to implement flushing programs and use higher dosages of disin-
fectants to maintain water quality in distribution systems. This has the unfortu-
nate side effect of increasing DBP formation as well as taste and odor problems,
which contribute to the public’s perception that the water quality is poor.
Fur- thermore, large pipes are generally cement-lined or unlined ductile
iron pipe typically with more than 300 joints per mile. These joints are
frequently not water tight, leading to water losses as well as providing an
opportunity for external contamination of finished water.
From an engineering perspective it seems intuitively obvious that it is m
ost
efficient to satisfy all needs by installing one pipe and to minimize the number
of pipe excavations. This philosophy worked well in the early days of
water systemdevelopment. However, it has resulted in water systems with long
residence times (and their negative consequences) under normal water use
patterns and a major investment in above-ground (pumps and storage tanks) and
below- ground (transmission mains, distribution pipes, service connections, etc.)
infrastructure. Therefore as suggested in Okun (2005) it may be time to look at
alternatives for supplying the various water needs in urban areas such as dual
distribution systems. The water reuse aspect of dual systems is particularly
attractive in arid sections of the U.S. that otherwise require transportation
of large quantities of water into these areas.
Although there are many examples of water reuse in the United S
t
a
t
e
s(EPA,
1992), not many of them involve the use of a dual distribution system.The City of
St. Petersburg, which operates one of the largest urban reuse s
y
s
- tems in the
world, provides reclaimed water to more then 7,000 residential homes and
businesses. In 1991, the city provided approximately 21 mgd (79,500
m3
/day) of reclaimed water for irrigation needs of individual homes,
condominiums, parks, school grounds, and golf courses; cooling tower make -up;
and supplemental fire protection. In Irving, Texas, advanced secondary treated
wastewater and raw water from the Elk Fork of the Trinity River are used to
irrigate golf courses, medians, and greenbelt areas, and to maintain water levels
at the Las Colinas Development. The reclaimed water originates from the
11.5 mgd (43,500 m3
/day) Central Regional wastewater treatment plant.
A third example is provided in Hilton Head, South Carolina, where about 5
mgd

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
INTRODUCTION 33
(18,900 m3
/day) of wastewater is being used for wetlands applications and golf
course irrigation. All of the wastewater treatment systems have been upgraded to
tertiary systems, and an additional flow rate of the same size as the first is
being planned. Perhaps the most famous water reuse operation using a
dual distribution system is the Irvine Ranch Water District in Irvine, California
(see Box 1-3). There is a recent trend in California toward the use of more dual
distribution systems, particularly in new developments, as a result of statutory
requirements to use reclaimed water in lieu of domestic water for non-potable
uses (California Water Code Section 13550-13551) and because of the need to
conserve water to meet increasing local and regional water demands.
The potential advantages of using dual distribution systems include the fact
that much smaller volumes of water would need be treated to high standards,
which would result in cost savings at the treatment plant if all water supplied
were to be treated in this fashion. Another advantage is that flow in the potable
line would be expected to be relatively constant compared to a traditional system
where large quantities of water would need to be transferred over short time
periods (e.g., during fires). The associated flow and pressure changes in a pipe
carrying the total water needs for a community are expected to be much greater
than in the potable line of a dual distribution system. As discussed later in this
document, there is evidence that pressure transients may result in intrusion of
contaminated water. Furthermore, use of improved materials in the newer,
smaller distribution system would minimize water degradation, loss, and intru-
sion.
However, the creation of dual distribution systems necessitates the retrofit-
ting of an existing water supply system and reliance on existing pipes to provi
de
non-potable supply obtained from wastewater or other sources. Large costs
would be incurred when installing the new, small diameter pipe for potable wa-
ter, disconnecting the existing system from homes and other users so that
it could be used reliably for only nonpotable needs, and other retrofitting
measures. These costs can be reduced if a new system is used only for
reclaimed water distribution, as was done at Irvine Ranch, but this of course
would not decrease the extent of quality degradation now experienced in
existing systems. It is also critical to differentiate between full and partial
adoption of dual distribution systems, the latter of which has occurred in
several cities. For example, if a new nonpotable line is installed alongside an
existing potable line, the nonpotable line can draw demand away from the
potable line, thereby increasing its detention time and aggravating water quality
deterioration in the potable line. Furthermore, if the potable systemis still used
for fire flow, which generally governs pipe sizing, many of the advantages of the
dual systemwill not be realized.
Dual systems may be most advantageous in new communities where neither
type of distribution system currently exists. New communities could better op-
timize their systems because both types of piping systems could be built simul-
taneously. The cost savings fromthe need to treat a much smaller portion of the
total water to a higher quality could partially offset the costs ofconstructing two

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
BOX 1-3
Irvine Ranch Water District
The Irvine Ranch Water District is one of the first water d
i
s
t
r
i
c
t
sin the United States to
practice wastewater reuse. It serves 316,287 people over 133 square miles (344.5 km2
),
making it about a quarter the size of the Los Angeles Department of Water and Pow er.
There are 85,500 domestic connections and 3,700 recycled connections. As of 2003, there
are 1,075 miles (1,730 m) of pipe for the potable system. As part of t
h
e potable system,
there are 28 above-ground and below-ground storage tanks that range in volume from 0.75
million gallons (0.0028 million m3
) to 16 million gallons (0.061 million m3
) and have a total
storage capacity of 131.75 million gallons (0.50 million m3
). Much of the District’s infra-
structure is below grade due to aesthetic considerations.
The most unusual aspect of the District system is the recycled (
r
e
claim ed) water net-
work. There are 350 m iles (563.2 km) of reclaim ed w ater lines compared to 1,075 miles
(1729.7 km) of potable network lines. Domestic water tanks sit side by side w ith reclaimed
water tanks. The recycled water is used only for toilet flushing in a few high-rise buildings,
for cooling towers, for landscape irrigation especially at golf courses and condominium
complexes, for food crops, and by one carpet manufacturer. Recycled water for toilet flush-
ing is not used in residences, only in businesses. Recycled water itself is tertiary treated
wastewater. It meets all of the water quality standards for drinking w ater, but it is high in
salt. Interestingly, in the summer the recycled water has a much low er retention time in the
distribution system than the potable water because of greater demand for the recycled
water for landscaping. How ever, when the demand for recycled water is less than the input
from WWTPs, the recycled water is put it in long-term storage. Indeed, one of the reasons
dual systems were installed in high rises and other buildings was to make demand for recy-
cled water more level throughout the year. There are no hydrants on the recycled system,
so the reclaimed w ater is not used for fighting fires, and the pipe sizes in the recycled sys-
tem are generally smaller than in the potable system. Chloramine provides residual disin-
fection in the potable system but chlorine is used in the recycled system (as mandated by
California regulations). The SCADA system, w hich consists of 6,000 sampling points, pro-
vides minute-by-minute monitoring of chlorine residuals in the recycled system.
The potable system is required to meet all SDWA regulatory r
e
q
uirem ent ssuch as the
TCR, SW TR, D/DBPR, LCR, and source w ater monitoring on the imported w ater sources
and the well w ater. Special purpose monitoring includes a nitrification action plan that re-
quires tank sampling. For the recycled system, how ever, there are no specific monitoring
objectives required by regulations because the NPDES permit has been met at the end of
the WWTP. Internal requirements include bi-monthly sampling of conductivity, turbidity,
color, pH, chlorine residual, total coliform, and fecal coliform and total suspended solids (at
special locations). The w ater uses for the recycled water are very specific, and it is the
goal of the utility to make sure the water is of an acceptable quality for those uses. Domes-
tic potable water costs are 64 cents per thousand gallons for domestic water and 59 cents
per thousand gallons for recycledwater.
As might be expected, Irvine Ranch has a very extensive cross-connection control (CCC)
program. There are approximately 13,000 CCC devices in place throughout the system.
The District conducts an annual cross-connection shut-down test for the recycled irrigation
water, and only one cross connection has been found in the last 10 years. For backflow
prevention, a reduced pressure principle assembly at the meter is used, as required by
the state of CA. Additionaldevices are installed if found to be needed.
SOURCE: Johannessen et al. (2005).

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
INTRODUCTION 35
systems.Clearly, better understanding the technologicalpotential and economic
consequences ofdualdistribution systems is an important research goal.
***
Non-traditional options for drinking water provision present many unan-
swered questions but few case studies from which to gather information.
The primary concerns include determining their economic feasibility and the
existence of unknown costs, developing a plan for transition and
implementation (which are expected to be very significant undertakings in
existing communities), and maintenance of quality assurance and quality
control in systems that would be potentially much more complicated than the
current system. Further- more, it is not clear how alternative distribution system
designs will affect water security, an important consideration since September
11, 2001. The potential for cross connections or misuse of water supplies of
lesser quality is greatly increased in dual distribution systems and
decentralized treatment. Larger-scale questions involve potential social
inequities and the extent to which nontradi- tional approaches will transfer costs
to the consumer. These issues will have to be considered carefully in
communities that decide to adopt these new designs for water provision.
The previous discussion raises a number of research issues, some of whi
chare
already noted. With regard to the influence of fire fighting requirements,
distribution systems are frequently designed to supply water to meet maximum
day demand and fire flow requirements simultaneously. This affects minimum
pipe diameters, minimum system pressures (under maximum day plus fire flow
demand), fire hydrant spacing, valve placement, and water storage. Generally,
agencies that set fire flow requirements are not concerned about water quality
while drinking water utilities must be concerned about both quality and fire flow
capacity. It will be important to better evaluate the effectiveness of alternative
fire suppression technologies including automatic sprinkler systems in a wide
range of building types, including residences. Such systems have rarely been
evaluated for their positive and negative features with respect to water quality.
Furthermore, if fire suppression technologies were improved, it might be possi-
ble to rely on smaller sized pipes in distribution systems, as is being tested in
Europe (Snyder et al., 2002), rather than moving to dual distribution systems.
If alternatives such as satellite systems and dual systems are not used, con-
tinued efforts will be required to upgrade existing distribution systems and to
treat water to acceptable levels of quality, so that quality does not deteriorate
during distribution. The balance of this report is focused on traditional distribu-
tion system design, in which water originates from a centralized treatment plant
or well and is then distributed through one pipe network to consumers. None-
theless, many of the report recommendations are relevant even if an alternative
distribution systemdesign is used.

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
Water System Diversity
Water utilities in the United States vary greatly in size, ownership, and type
of operation. The SDWA defines public water systems as consisting of commu -
nity water supply systems; transient, non-community water supply systems; and
non-transient, non-community water supply systems. A community water sup-
ply system serves year-round residents and ranges in size from those that serve
as few as 25 people to those that serve several million. A transient, non-
community water supply system serves areas such as campgrounds or gas sta-
tions where people do not remain for long periods of time. A non-transient,
non-community water supply system serves primarily non-residential customers
but must serve at least 25 of the same people for at least six months of the year
(such as schools, hospitals, and factories that have their own water supply).
There are 159,796 water systems in the United States that meet the federal defi-
nition of a public water system (EPA, 2005b). Thirty-three (33) percent
(52,838) of these systems are categorized as community water supply systems,
55 percent are categorized as transient, noncommunity water supplies, and
12 percent (19,375) are non-transient, non-community water systems
(EPA, 2005b). Overall, public water systems serve 297 million residential
and commercial customers. Although the vast majority (98 percent) of
systems serves less than 10,000 people, almost three quarters of all Americans
get their water from community water supplies serving more than 10,000 people
(EPA, 2005b). Not all water supplies deliver water directly to consumers,
but rather deliver water to other supplies. Community water supply systems are
defined as “consecutive systems” if they receive their water from another
community water supply through one or more interconnections (Fujiwara et al.,
1995).
Some utilities rely primarily on surface water supplies while others r
e
l
y
primarily on groundwater. Surface water is the primary source of 22 percent o
f
the
community water supply systems, while groundwater is used by 78 percent of
community water supply systems. Of the non-community water supply systems
(both transient and non-transient), 97 percent are served by groundwater. Many
systems serve communities using multiple sources of supply such as a
combination of groundwater and/or surface water sources. This is important
because in a grid/looped system, the mixing of water from different sources can
have a detrimental influence on water quality, including taste and odor, in
the distribution system(Clark et al., 1988, 1991a,b).
Some utilities, like the one operating in New York City, own large areas o
f
the watersheds from which their water source is derived, while other utiliti
es
depend on water pumped directly from major rivers like the Mississippi River or
the Ohio River, and therefore own little if any watershed land. The SDWA was
amended in 1986 and again in 1996 to emphasize source water protection
in order to prevent microbial contaminants from entering drinking water
supplies (Borst et al., 2001). Owning or controlling its watershed provides an
opportunity for a drinking water utility to exercise increased control of its
source water quality (Peckenham et al., 2005).

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
INTRODUCTION 37
The water supply industry in the United States has a long history of l
ocal
government control over operation and financial management, with varying de-
grees of oversight and regulation by state and federal government. Water suppl
y
systems serving cities and towns are generally administered by departments o
f
municipalities or counties (public systems) or by investor owned com
panies
(private systems). Public systems are predominately owned by local municipal
governments, and they serve approximately 78 percent of the total
population that uses community water supplies. Approximately 82 percent of
urban water systems (those serving more than 50,000 persons) are publicly
owned. There are about 33,000 privately owned water systems that serve
the remaining 22 percent of people served by community water systems.
Private systems are usually investor-owned in the larger population size
categories but can include many small systems as part of one large
organization. In the small- and medium-sized categories, the privately
owned systems tend to be owned by homeowners associations or developers.
Finally, there are several classifications of state chartered public corporations,
quasi-governmental units, and municipally owned systems that operate
differently than traditional public and private systems. These systems include
special districts, independent non-political boards, and state chartered
corporations.
Infrastructure Viability over the Long Term
The extent of water distribution pipes in the United States is estimated to b
e
a
total length of 980,000 miles (1.6 x 106
km), which is being replaced a
t an
estimated rate of once every 200 years (Grigg, 2005b). Rates of repair and re -
habilitation have not been estimated. There is a large range in the type and age
of the pipes that make up water distribution systems. The oldest cast iron pipes
from the late 19th
century are typically described as having an expected average
useful lifespan of about 120 years because of the pipe wall thickness (AWWA,
2001; AWWSC, 2002). In the 1920s the manufacture of iron pipes changed to
improve pipe strength, but the changes also produced a thinner wall. These
pipes have an expected average life of about 100 years. Pipe
manufacturing continued to evolve in the 1950s and 1960s with the introduction
of ductile iron pipe that is stronger than cast iron and more resistant to corrosion.
Polyvinyl chloride (PVC) pipes were introduced in the 1970s and high-density
polyethylene in the 1990s. Both of these are very resistant to corrosion but
they do not have the strength of ductile iron. Post-World War II pipes tend to
have an expected average life of 75 years (AWWA, 2001; AWWSC, 2002).
In the 20th
century, most of the water systems and distribution pipes were
relatively new and well within their expected lifespan. However, as is obvi
ous
from the above paragraph and recent reports (AWWA, 2001; AWWSC, 2
0
0
2
)
, these
different types of pipes, installed during different time periods, will all b
ereaching
the end of their expected life spans in the next 30 years. Indeed, a
n
estimated 26
percent of the distribution pipe in the country is unlined and in

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
poor condition. For example, an analysis of main breaks at one large Midwest-
ern water utility that kept careful records of distribution system management
documented a sharp increase in the annual number of main breaks from 1970
(approximately 250 breaks per year) to 1989 (approximately 2,200 breaks per
year) (AWWSC, 2002). Thus, the water industry is entering an era where it
must make substantial investments in pipe repair and replacement. As shown in
Figure 1-6, an EPA report on water infrastructure needs (EPA, 2002c) predicted
that transmission and distribution replacement rates will rise to 2.0 percent per
year by 2040 in order to adequately maintain the water infrastructure, which is
about four times the current replacement rate according to Grigg (2005b).
These data on the aging of the nation’s infrastructure suggest that
utilities will have to engage in regular and proactive infrastructure assessment
and replacement in order to avoid a future characterized by more frequent
failures, which might overwhelm the water industry’s capability to react
effectively (Beecher, 2002). Although the public health significance of
increasingly frequent pipe failures is unknown given the variability in utility
response to such events, it is reasonable to assume that the likelihood of external
distribution system contamination events will increase in parallel with
infrastructure failure rates.
FIGURE 1-6 Projected annual replacement needs for transmission l
i
n
e
sand distribution mains,
2000–2075. SOURCE: EPA (2002c).

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
INTRODUCTION 39
DISTRIBUTION SYSTEM INTEGRITY
Many factors affect both the quantity and quality of water in di
stri
buti
on
systems. As discussed in detail in Appendix A, events both internal and external
to the distribution system can degrade water quality, leading to violation of wa-
ter quality standards and possible public health risks. Corrosion and leaching of
pipe materials, growth of biofilms and nitrifying microorganisms, and the for-
mation of DBPs are events internal to the distribution systemthat are potentially
detrimental. Furthermore, most are exacerbated by increased water age within
the distribution system. External contamination can enter the distribution sys -
tem through infrastructure breaks, leaks, and cross connections as a result of
faulty construction, backflow, and pressure transients. Repair and replacement
activities as well as permeable pipe materials also present routes for
exposing the distribution system to external contamination. All of these events
act to compromise the integrity of the distribution system.
For the purposes of this report, distribution system integrity is defined as
having three basic components: (1) physical integrity, which refers to the main-
tenance of a physical barrier between the distribution system interior and
the external environment, (2) hydraulic integrity, which refers to the
maintenance of a desirable water flow, water pressure, and water age, taking
both potable drinking water and fire flow provision into account, and (3)
water quality integrity, which refers to the maintenance of finished water quality
via prevention of internally derived contamination. This division is important
because the three types of integrity have different causes of their loss, different
consequences once they are lost, different methods for detecting and preventing
a loss, and different remedies for regaining integrity. Factors important in
maintaining the physical integrity of a distribution system include the
maintenance of the distribution system components, such as the protection
of pipes and joints against internal and external corrosion and the presence of
devices to prevent cross-connections and backflow. Hydraulic integrity
depends on, for example, proper system operation to minimize residence time
and on preventing the encrustation and tuberculation of corrosion products and
biofilms on the pipe walls that increase hydraulic roughness and decrease
effective diameter. Maintaining water quality integrity in the face of internal
contamination can involve control of nitrifying organisms and biofilms via
changes in disinfection practices.
In addition to the distinctions mentioned above, there are also
commonalities between the three types of integrity. All three are subject to
system specificity, in that they are dependent on such site-specific factors as
local water quality, types of materials present, area served, and population
density. Further- more, certain events involve the loss of more than one
type of integrity—for example, backflow due to backsiphonage involves the
loss of both hydraulic and physical integrity. Materials quality is important for
both physical and water quality integrity. In order for a law or regulation to
adequately address distribution system integrity—one of the options being
considered during revision of the TCR—it must encompass physical, hydraulic,
and water qualityintegrity.

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
IMPETUS FOR THE STUDY AND REPORT ROADMAP
Water supply systems have historically been designed for efficiency in w
a-ter
delivery to points of use, hydraulic reliability, and fire protection, while most
regulatory mandates have been focused on enforcing water quality standards at
the treatment plant. Ideally, there should be no change in the quality of treated
water from the time it leaves the treatment plant until the time it is consumed,
but in reality substantial changes may occur as a result of complex physical,
chemical, and biological reactions. Distribution systems are the final barrier to
the degradation of treated water quality, and maintaining the integrity of these
systems is vital to ensuring that the water is safe for consumption.
The sections above have discussed the aging of the nation’s water i
nfr
a-
structure and the continuing contribution of distribution systems to public health
risks from drinking water. For the last five years, EPA has engaged experts and
stakeholders in a series of meetings on the topic of distribution systems, with the
goal of defining the extent of the problem and considering how it can be
addressed during revisions to the TCR. As part of this effort, EPA led in the
creation of nine white papers that summarized the state-of-the-art of research
and knowledge in the area of drinking water distribution systems:
 Cross-Connections and Backflow (EPA, 2002a)
 Intrusion of Contaminants from Pressure Transients (LeChevallier e
t
al.,
2002)
 Nitrification (AWWA and EES, Inc., 2002e)
 Permeation and Leaching (AWWA and EES Inc., 2002a)
 Microbial Growth and Biofilms (EPA, 2002b)
 New or Repaired Water Mains (AWWA and EES Inc., 2002e)
 Finished Water Storage (AWWA and EES, Inc., 2002c)
 Water Age (AWWA and EES, Inc., 2002b)
 Deteriorating Buried Infrastructure (AWWSC, 2002)
Additional activities are ongoing, including consideration of a revision o
f
the
TCR to provide a more comprehensive approach for addressing the integrity of
the distribution system. To assist in this process, EPA requested that the Na-
tional Academies’ Water Science and Technology Board conduct a study of
water quality issues associated with public water supply distribution systems and
their potential risks to consumers. An expert committee was formed in October
2004 with the following statement of task:
1) Identify trends relevant to the deterioration of drinking water in w
a
t
e
r
supply distribution systems,as background and based on available information.
2) Identify and prioritize issues of greatest concern for distribution s
y
s-tems
based on review of published material.

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
INTRODUCTION 41
3) Focusing on the highest priority issues as revealed by task #2
,(a)
evaluate different approaches for characterization of public health risks posed by
water quality deteriorating events or conditions that may occur in public water
supply distribution systems; and (b) identify and evaluate the effectiveness of
relevant existing codes and regulations and identify general actions, strategies,
performance measures, and policies that could be considered by water utilities
and other stakeholders to reduce the risks posed by water-quality deteriorating
events or conditions. Case studies, either at state or utility level, where distribu-
tion system control programs (e.g., Hazard Analysis and Critical Control
Point System, cross connection control, etc.) have been successfully designed
and implemented will be identified and recommendations will be presented in
their context.
4) Identify advances in detection, monitoring and modeling, anal
yti
cal
methods, information needs and technologies, research and development oppor-
tunities, and communication strategies that will enable the water supply industry
and other stakeholders to further reduce risks associated with public water sup-
ply distribution systems.
The NRC committee addressed tasks one and two in its first report (NR
C
,
2005), which is included as Appendix A to this report. The following tr
e
nds
were identified as relevant to the deterioration of water quality in distribution
systems:
 The aging distribution system infrastructure, including increasing n
u
m
-
bers of main breaks and pipe replacement.
 Decreasing numbers of waterborne outbreaks reported per year s
i
nce
1982, but an increasing percentage attributable to distribution systemissues.
 Increasing host susceptibility to infection and disease in the U.S. po
p
u-
lation.
 Increasing use of bottled water and point-of-use treatmentdevices.
It was recommended in NRC (2005) that EPA consider these trends as it revises
the TCR to encompass distribution system integrity. The committee was made
aware of another important trend subsequent to the release of NRC (2005)—
population shifts and how they have affected water demand. Older
industrial cities in the northeast and Midwest United States no longer have
industries that use high volumes of water, and they have also experienced
major population shifts from the inner city to the suburbs. As a consequence,
the utilities have an overcapacity to produce water, mainly in the form of
oversized mains, at central locations, while needing to provide water to suburbs
at greater distances from the treatment plant. Both factors can contribute to
problems associated with high water residence times in the distribution system.
As part of its second task, the NRC committee prioritized the issues that a
r
e
the
subject of the nine EPA white papers, and it identified several si
gni
fi
cant
issues that were overlooked in previous reports. The highest priority issues were

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
those that have a recognized health risk based on clear epidemiological and sur-
veillance data. These include cross connections and backflow; contamination
during installation, rehabilitation, and repair of water mains and appurtenances;
improperly maintained and operated storage facilities; and control of water qual-
ity in premise plumbing.
This report focuses on the committee’s third and fourth tasks and m
ak
e
s
recommendations to EPA regarding new directions and priorities to consider.
All of the issues discussed in NRC (2005) are presented here, but considerably
more information is presented on the higher priority issues when recommending
detection, mitigation, and remediation strategies for distribution systems.
The report is intended to inform decision makers within EPA, public water
utilities, other government agencies and the private sector about potential
options for managing distribution systems.
It should be pointed out that this report is premised on the assumption tha
t
water entering the distribution system has undergone adequate treatment. [As
recognized in the SDWA, adequate treatment is a function of the quality of
source water. For example, some lower quality source waters may require filtra -
tion to achieve a product entering the distribution system that is of the
same quality (and hence poses the same risk) as a cleaner source water that
was treated only with disinfection.] There is not, therefore, an in-depth
discussion of drinking water treatment in the report except where it is pertinent
to mitigating the risks of degraded water quality in the distribution system.
For example, if the lack of disinfectant residual in the distribution system is
identified as a risk, the options for mitigating that risk must first consider
whether the root cause is inadequate treatment (e.g., insufficient reduction in
disinfectant demand), or causes attributable to the distribution system (e.g.,
excessive water age in storage facilities). It should also be noted that
deliberate acts of distribution system contamination are not considered, at the
request of the studysponsor.
Chapter 2 reviews the legal and regulatory environment in which distribu-
tion systems are designed, operated, and monitored, including federal, state, a
n
d
local regulations. The limitations and possibilities associated with n
o
n-regulatory
approaches are also mentioned. Chapter 3 presents the three primary
approaches for assessing the public health risk of contaminated distribution sys -
tems, focusing on short-term acute risks from microbial pathogens.
Chapters 4, 5, and 6 consider the physical, hydraulic, and water quality i
n
-
tegrity of distribution systems, respectively. For each type of integrity, t
he
chapters consider what causes its loss, the consequences if it is lost, and how to
detect, maintain, and recover the type of integrity. In most cases, the events that
compromise distribution system integrity are discussed only once, in the earliest
chapter to which they are relevant. Many of the common themes from these
chapters are brought together in Chapter 7, which presents a holistic framework
for distribution system management, highlighting those activities felt to be of
greatest importance to reducing public health risks. Areas where emerging sci-
ence and technology can play a role are discussed, including real-time, on-line
monitoring and modeling. The report concludes in Chapter 8 by considering the

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
INTRODUCTION 43
importance of premise plumbing to overall water quality at the tap, the need for
additional monitoring of premise plumbing, and the need for greater involve-
ment by regulatory agencies in exercising authority over premise plumbing.
Premise plumbing is an issue not generally considered to be the responsibility of
drinking water utilities, but there is growing interest—in terms of public health
protection—about the role of premise plumbing in contributing to water quality
degradation.
REFERENCES
American Water Works Association (AWWA). 1974. Water distribution research a
n
d
applied development needs. J. Amer. Water Works Assoc. 6:385–390.
AWWA. 1986. Introduction to Water Distribution Principles and Practices of W
a
t
e
r
Supply Operations. Denver, CO: AWWA.
AWWA. 1998. AWWA Manual M31:Distribution systemrequirements for fire p
r
o
t
e
c
- tion.
Denver, CO: AWWA.
AWWA. 2001. Reinvesting in Drinking Water Structure: Dawn of the Replacement Era.
Denver, CO: AWWA.
AWWA. 2003. Water Stats 2002 Distribution Survey CD-ROM. Denver, CO: AWWA.
AWWA and EES, Inc. 2002a. Permeation and leaching. Available on-line at
http://www.epa.gov/safewater/tcr/pdf/permleach.pdf. Accessed May 4, 2006.
AWWA and EES, Inc. 2002b. Effects of water age on distribution systemwater quality.
http://www.epa.gov/safewater/tcr/pdf/waterage.pdf. Accessed May 4, 2006.
AWWA and EES, Inc. 2002c. Finished water storage facilities. Available on-line at
http://www.epa.gov/safewater/tcr/pdf/storage.pdf. Accessed May 4, 2006.
AWWA and EES, Inc. 2002e. New or repaired water mains. Available on-line at
http://www.epa.gov/safewater/tcr/pdf/maincontam.pdf. Accessed May 4, 2006.
American Water Works Service Co., Inc. (AWWSC). 2002. Deteriorating buried i
n
f
r
a
-
structure management challenges and strategies. Available on-lineat
http://www.epa.gov/safewter/tcr/pdf/infrastructure.pdf. Accessed May 4, 2006.
Baker, M. H. 1948. The quest for purewater. The American Water Works Association.
Lancaster, PA:Lancaster Press.
Beecher, J. A. 2002. The infrastructure gap: myth, reality, and strategies. In: Assessing
the Future: Water Utility InfrastructureManagement. D. M. Hughes (ed.). Denver,
CO: AWWA.
Berger, P. S., R. M. Clark, and D. J. Reasoner. 2000. Water, Drinking. In:
Encyclope- dia of Microbiology 4:898–912.
Blackburn, B. G., G. F. Craun, J. S. Yoder, V. Hill, R. L. Calderon, N. Chen, S. H. Lee,
D. A. Levy, and M. J. Beach. 2004. Surveillance for waterborne-disease outbreaks
associated with drinking water—United States, 2001–2002. MMWR 53(SS-8):23–
45.
Borst, M., M. Krudner, L. O’Shea, J. M. Perdek, D. Reasoner, and M. D. Royer. 2
0
0
1
.
Source water protection: its role in controlling disinfection by-products (DBPs) a
n
d
microbial contaminants. In: Controlling Disinfection By-Products and Microbial
Contaminants in Drinking Water. R. M. Clark and B. K. Boutin (eds.). EPA/600/R-
01/110. Washington, DC: EPA Office of Research and Development.

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
Brandt, M., J. Clement, J. Powell, R. Casey, D. Holt, N. Harris, and C. T. Ta. 2
0
0
4
.
Managing
Distribution Retention Time to Improve Water Quality-Phase I. Denver CO:
AwwaRF.
Clark, R. M. 1978. The Safe Drinking Water Act: implications for planning. Pp. 1
1
7
–
137
In: Municipal Water Systems—The Challenge for Urban Resources M
a
n
a
g
e
-
ment. D.
Holtz and S. Sebastian (eds.). Bloomington, IN: Indiana University P
r
e
s
s
.
Clark, R. M., and D. L. Tippen. 1990. Water Supply. Pp. 5.173–5.220 In: Standard
Handbook of Environmental Engineering. R. A. Corbitt (ed.). New York:
McGraw-HillPublishing Company.
Clark, R. M., and W. M. Grayman. 1992. Distribution system quality: a trade-off b
e
-
tween public health and public safety. J. Amer. Water Works Assoc. 84(7):18.
Clark, R. M., and W. M. Grayman. 1998. Modeling water quality in drinking w
a
t
e
r
distribution
systems. Denver, CO: AWWA.
Clark, R. M., E. E. Geldreich, K. R. Fox, E. W. Rice, C. H. Johnson, J. A. Goodrich, J
.
A
.
Barnick, and F. Abdesaken. 1996. Tracking a Salmonella Serovar T
y
p
h
i
m
u
r
i
u
m
Outbreak in
Gideon, Missouri: role of contaminant propagation modeling. Journal of Water
Supply Research and Technology–Aqua 45(4):171–183.
Clark, R. M., J. A. Goodrich, and J.C. Ireland. 1984. Cost and benefit of drinking w
a
t
e
r
treatment. Journal of Environmental Systems 14(1):1–30.
Clark, R. M., W. M. Grayman, and R. M. Males. 1988. Contaminant propagation
in distribution systems. Jour. of Environ. Eng. ASCE 114(2):929–943.
Clark, R. M., W. M. Grayman, and J. A. Goodrich. 1991a. Water quality modeling: i
t
s
regulatory implications. In: Proceedings of the AwwaRF/EPA Conference on Water
Quality Modeling in Distribution Systems, Cincinnati, OH.
Clark, R. M., W. M. Grayman, J. A. Goodrich, R. A. Deininger, and A. F. Hess. 1
9
9
1
b
. Field
testing of distribution water quality models. J. Amer. Water Works Assoc.
83(7):67–75.
Clark, R. M., G. S. Rizzo, J. A. Belknap, and C. Cochrane. 1999. Water quality
and t
h
e
replacement of drinking water infrastructure: the Washington, DC case study.
Jour- nal of Water Supply Research and Technology – Aqua48(3):106–114.
Clark, R. M., W. M. Grayman, S. G. Buchberger, Y. Lee, and D. J. Hartman. 2
0
0
4
.
Drinking
water distribution systems: an overview. Pp. 4.1–4.49 In: Water Supply
Systems Security. L. W. Mays (ed.). New York: McGraw-Hill.
Craun, G. F., and R. Calderon. 2001. Waterborne disease outbreaks caused by d
i
s
t
r
i
b
u
-tion
systemdeficiencies. J. Amer. Water Works Assoc. 93(9):64–75.
Edwards, M., D. Bosch, G. V. Loganathan, and A. M. Dietrich. 2003. The Future C
h
a
l
-lenge
of Controlling Distribution System Water Quality and Protecting Plumbing I
n
-
frastructure: Focusing on Consumers. Proceedings of the IWA Leading Edge Con-
ference in Noordwijk, Netherlands.
Environmental Protection Agency (EPA). 1991. Water Conservation Guidelines. W
a
s
h
-ington,
DC: EPA.
EPA. 1992. Guidelines for Water Reuse. EPA/625/R-92/004. Washington, DC: EPA.
EPA. 2000. National primary drinking water regulations for lead and copper:final rule.
Federal Register 65:1950. Washington, DC: EPA.
EPA. 2002a. 2000 Community water system survey. EPA 815-R-02-005A. W
a
s
h
i
n
g
-ton,
DC: EPA Office of Water.
EPA. 2002b. Potential contamination due to cross-connections and backflow and t
h
e
associated health risks: an issues paper. Available on-line at http://www.epa.gov/
safewater/tcr/pdf/ccrwhite.pdf. Accessed May 8, 2006.

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
INTRODUCTION 45
EPA. 2002c. The clean water and drinking water infrastructure gap analysis. W
a
s
h
i
n
g
-ton, DC:
EPA.
EPA. 2005a. Drinking water infrastructure needs survey. EPA 816-R-05-001. W
a
s
h
-ington,
DC: EPA Office of Water.
EPA. 2005b. Factoids: drinking water and ground water statistics for 2003. EPA 8
1
6
-K-05-
001. Washington, DC: EPA Office of Water.
Fujiwara, M., J. M. Manwaring, and R. M. Clark. 1995. Drinking water in Japan
and t
h
e
United States: conference objectives. In: Drinking water quality management.
R.
M. Clark and D. A. Clark (eds.). Lancaster, PA: Technomic Publishing C
o
m
p
a
n
yInc.
Grigg, N. S. 2005a. Letter to the editor: designing future water distribution systems. J
.
Amer. Water Works Assoc. 97(6):99–101.
Grigg, N. S. 2005b. Assessment and renewal of water distribution systems. J. Amer.
Water Works Assoc. 97(2):58–68.
Grindler, B. J. 1967. Water and water rights: a treatise on the laws of water and a
l
l
i
e
d
problems: eastern, western, federal. Vol 3. Indianapolis, IN: The Allan S
mith
Company.
Hanke, S. H. 1972. Pricing Urban Water. Pp. 283–306 In: Public Prices for P
u
b
l
i
c
Products. S. Mushkin (ed.). Washington, DC: TheUrban Institute.
Insurance Services Office. 1980. Fire suppression rating schedule. New York: I
n
s
u
r
-ance
Services Office.
Jacobsen, L. 2005. Las Vegas Valley Water District. April 18, 2005. Presented to t
h
e NRC
Committee on Public Water Supply Distribution Systems. Washington,DC.
Jacobsen, L., and S. Kamojjala. 2005. Full SystemModels and GIS Integration. In:
Proceedings of theAWWA Annual Conference and Exposition, San Francisco, CA.
Jacobsen, L., S. Kamojjala, and M. Fang. 2005. Integrating hydraulic and water q
u
a
l
i
t
y models
with other utility systems: a case study. In: Proceedings of the AWWA I
n
- formation
Management and Technology Conference, Denver, CO.
Johannessen, J., C. Kinner, and M. Velardes. 2005. Dual Distribution Systems: T
h
e
Irvine
Ranch Water District Experience. January 13, 2005. Presented to the NRC
Committee on Public Water Supply Distribution Systems. Irvine, CA.
Kirmeyer, G., W. Richards, and C. D. Smith. 1994. An assessment of water d
i
s
t
r
i
b
u
t
io
nsystems
and associated research needs. Denver, CO: AwwaRF.
LeChevallier, M., R. Gullick, and M. Karim. 2002. The potential for health risks f
r
o
m
intrusion of contaminants into the distribution system from pressure transients.
Draft Distribution System White Paper. Washington, DC:EPA.
Lee, S. H., D. A. Levy, G. F. Craun, M. J. Beach, and R. L. Calderon. 2002. S
u
r
v
e
i
l
- lance
for waterborne-disease outbreaks in the United States, 1999–2000. M
M
W
R
51(No. SS-
8):1−49.
Levi, Y., S. Pernettes, O. Wable, and L. Kiene. 1997. Demonstration unit of s
a
t
e
l
l
i
t
e treatment
in distribution system using ultrafiltration and nanofiltration. Pp. 5
8
1
–
5
9
5
In: Proceedings of
the AWWA Conference on Membrane Technology Conference. New Orleans, LA.
Mayer, P., W. B. DeOreo, E. M. Opitz, J. C. Kiefer, W. Y. Davis, B. Dziegielewski, and
J. O. Nelson. 1999. Residential End Uses of Water. Denver, CO: AwwaRF.
Male, J. W., and T. M. Walski. 1991. Water Distribution Systems: A T
r
o
u
b
l
e
s
h
o
o
t
i
n
g
Manual. Chelsea,
MI:Lewis Publishers, Inc.
Moore, B. C., F. S. Cannon, D. H. Metz, and J. DeMarco. 2003. GAC pore structure in
Cincinnati during full-scale treatment/reactivation. J. Amer. Water Works Assoc.
95(2):103–118.

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
National Fire Protection Association (NFPA). 1986. Fire Protection Handbook, 1
6
t
h
edition. A. E. Cote and J. L. Linville (eds.). Batterymarch Park Quincy, MA: Na-
tional Fire Protection Association.
National Research Council (NRC). 2005. Public Water Supply Distribution
Systems: Assessing and Reducing Risks, First Report. Washington, DC: The
National Acad- emies Press.
Norton, J. W., Jr. 2006. Cost advantages of implementing distributed treatment t
e
c
h
-
nologies for reduction of water-borne risk factors. Pp. 78-105 In: Financial a
n
d
Technical Feasibility of Distributed Advanced Technology Water Treatment Sys-
tems. Ph.D. Dissertation. University of Michigan, Department of Environmental
and Water Resources Engineering.
Okun, D. A. 2005. Letter to the editor: designing future water distribution systems. J
.
Amer. Water Works Assoc. (6):99–101.
Panguluri, S., W. M. Grayman, and R. M. Clark. 2005. Distribution system water q
u
a
l
- ity
report: a guide to the assessment and management of drinking water
quality i
n
distribution systems. Cincinnati, OH: EPA Office of Research and
Development.
Peckenham, J. M., C. V. Schmitt, J. L. McNelly, and A. L. Tolman. 2005. Linking w
a
t
e
r
quality to the watershed: developing tools for source water protection. J. Amer. Wa-
ter Works Assoc. 97(9):62–69.
Pierson, G., K. Martel, A. Hill, G. Burlingame, and A. Godfree. 2001. Methods to prevent
microbiological contamination associated with main rehabilitation and
replacement. Denver, CO:AwwaRF.
Snyder, J. K., A. K. Deb, F. M. Grablutz, S. B. McCammon, W. M. Grayman, R
. M.
Clark, D. A. Okun, S. M. Tyler, and D. Savic. 2002. Impacts of fire flow on distri-
bution systemwater quality, design and operation. Denver, CO: AwwaRF.
Von Huben, H. (Tech. Ed). 1999. Water distribution operator training handbook,
2nd
edition. Denver, CO: AWWA.
Walski, T. M., D. V. Chase, and D. A. Savic. 2001. Water Distribution Modeling, 1
st
Edition. Waterbury, CT: Haestad Press.
Weber, Jr., W. 2002. Distributed optimal technology networks: a concept and strateg
y for
potablewater sustainability. Water Science and Technology 46(6–7):241–246.
Weber, Jr., W. 2004. Optimal uses of advanced technologies for water and w
a
s
t
e
w
a
t
e
rtreatment
in urban environments. Water Science and Technology: Water S
u
p
p
l
y
4(1):7–12.
Weber, Jr., W. 2005. Distributed Systems. April 18, 2005. Presented to the N
R
C
Committee on
Public Water Supply Distribution Systems. Washington, DC.

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
2
Regulations, Non-regulatory A
p
p
r
o
a
c
h
e
s
,and their
Limitations
This chapter provides an overview of the existing regulatory framework a
s
well as non-regulatory approaches that are intended to protect drinking
water quality within water distribution systems. Included is a discussion of
federal and state statutes and regulations and local codes, along with their
limitations. In addition, several non-regulatory programs are described that are
intended to complement existing regulations.
REGULATORY ENVIRONMENT
Federal and state statutes and regulations along with local codes are used t
o
establish requirements intended to protect the drinking water quality within dis -
tribution systems. The federal Safe Drinking Water Act (SDWA) is the vehicle
used nationally to address drinking water quality issues. Prior to the passage of
the SDWA, federal involvement in water supply had been limited to
development of large multi-purpose water projects and regulation of water
quality with respect to interstate carriers. After passage of the SDWA, the
federal government became involved in developing national drinking water
regulations pursuant to the new law and in conducting research to support
these regulations. States implement the federal mandates but also utilize their
own statutory and regulatory requirements to protect drinking water quality.
For example, the states play a significant role in oversight functions ranging
from licensing of water treatment plant operators to the approval of new
sources of supply and the approval of new treatment facility design. Local
agencies such as health departments, environmental health programs, and
building departments implement codes and ordinances that address water
distribution systems, most often that portion of the infrastructure not controlled
by public water systems. This section provides an overview of the
various statutory and regulatory approaches that apply to distribution systems.
Safe Drinking Water Act
The SDWA (Public Law 93-523), enacted in 1974 and amended in 1
9
8
6
(Public Law 99-339), 1988 (Public Law 100-572), and 1996 (Public Law 104-
182), provides the statutory bases by which public water systems are regulated.
47

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
Pursuant to the SDWA, the U.S. Environmental Protection Agency (EPA) is
mandated to establish regulations for drinking water in the form of either maxi-
mum contaminant levels (MCL) or maximum contaminant level goals
(MCLGs). MCLs are water quality standards that must be met by utilities and
are enforced by state or federal agencies. Unlike MCLs, MCLGs are non-
enforceable and are set at a level at which no known or anticipated adverse hu-
man health effects occur. Where it is not economically or technologically feasi-
ble to ascertain the level of a contaminant, a treatment technique is prescribed by
EPA in lieu of establishing an MCL. For example, because the viable concen-
tration of Giardia lamblia is difficult to measure, it has been established that if
water is treated at a given pH, temperature, and chlorine concentration for a
specified length of time (all of which are verified by the water utility), a fixed
level of Giardia inactivation will takeplace.
The SDWA also provides EPA with the authority to delegate the implemen-
tation of the SDWA requirements to the states through the process of primacy.
Forty-nine (49) of the 50 states have accepted primacy, with Wyoming being the
exception. The SDWA applies to public water systems, which can be publicly or
privately owned. Public water systems are defined as providing drinking water
to at least 25 people or 15 service connections for at least 60 days per year. As
mentioned in Chapter 1, there are approximately 160,000 public water sys - tems
in the United States, providing water to more than 290 million people.
Currently, 51 organic chemicals, 16 inorganic chemicals, seven di
si
nfec- tants
and disinfection byproducts (DBPs), four radionuclides, and coliform b
a
c
-teria are
monitored for compliance with the SDWA (EPA, 2005a). Standards for most
contaminants are required to be met at the point of entry to the distribution
system, such that the SDWA does not directly address distribution system
contamination for most compounds. Despite these spatial restrictions, the
SDWA does provide EPA with the authority to regulate contaminants within
distribution systems—an authority that EPA has used to promulgate several
regulations that address distribution system water quality including the Total
Coliform Rule (TCR), the Lead and Copper Rule (LCR), the Surface
Water Treatment Rule (SWTR), and the Disinfectants/Disinfection
Byproducts Rule (D/DBPR).
The 1996 amendments to the SDWA mandated that EPA conduct research
to strengthen the scientific foundation for standards that limit public exposure to
drinking water contaminants. Specific requirements were given for research on
waterborne pathogens such as Cryptosporidium and Norovirus, DBPs,
arsenic, and other harmful substances in drinking water. EPA was also directed
to conduct studies to identify and characterize population groups, such as
children, that may be at greater risk from exposure to contaminants in drinking
water than is the general population. In response to that mandate EPA has
developed a Multi- Year Plan that describes drinking water research program
activities and plans for fiscal years 2003–2010 (see Box 2-1).

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
REGULATIONS, NON-REGULATORY APPROACHES, AND THEIR LIMITATIONS 49
BOX 2-1
EPA Multi-Year Plan for Drinking Water
The Multi-Year Plan establishes three long-term goals:
1. By 2010, develop scientifically sound data and approaches to a
ssessand m anage
risks to hum an health posed by exposure to regulated w aterborne pathogens and chemi-
cals, including those addressed by the Arsenic, M/DBP, and Six-Year Review Rules.
2. By 2010, develop new data, innovative tools, and improved technol
ogi
es to support
decision making by the EPA Office of Water on the Contaminant Candidate List and other
regulatory issues, and to support implementation of rules by states, local authorities, and
w ater utilities.
3. By 2009, provide data, tools, and technologies to support m
a
n
agem ent decisions by
the EPA Office of W ater, state, local authorities, and utilities to protect source water and the
quality of w ater in the distribution system.
Some of the tasks in the Multi-Year Plan related to distribution systems include:
 Collect data to assess the stability of arsenic in w ater distribution system
s.
 Prepare a report on chlorine and chloramines to control b
i
o
f
i
l
m
s in model distribution
systems.
 Prepare a report on the mechanisms and kinetics of c
h
l
o
r
a
m
i
n
eloss and DBP formation
in distribution systems. This work includes the modeling of n-
nitrosodimethylamine formation.
 Prepare a report on the effect of oxidizing conditions on m
e
t
a
l
releases, corrosion rate,
and scale properties of distribution systemmaterials.
 Prepare a report on biofilm formation rates in pilot-scale distribution s
y
stems.
 Report on the characterization and prediction of scale f
o
r
m
a
t
i
o
n
(including aluminum) in
distribution systems.
 Prepare a report on the detection of opportunistic p
a
t
h
o
g
e
n
s
(E. coli, Aeromonas,
Mycobacterium) in biofilms using molecular detection techniques.
 Collect data on the treatment conditions w hich may enhancethe solubilization of
arsenic-containing iron oxides w ithin the distribution system.
 Prepare a report on the link betw een the distribution system and Mycobacterium
avium complex (MAC) found in clinicalcases.
 Prepare a report on characterization of drinking w aterd
i
s
t
r
i
b
u
t
i
o
nsystembiofilm microbial
populations using molecular detection methods.
 Prepare a report on corrosion chemistry relationships and treatment a
p
p
r
o
a
c
h
e
s
.
 Prepare a report on the impact of change from conventional treatment of surface
water to alternative treatment (membrane) on biofilm grow th in water distribution systems in
support of regulation development.
 Improve methods for rapid detection of w ater quality changes.
 Conduct leaching studies to characterize organotin c
o
n
c
e
n
t
r
a
t
i
o
n
sin distribution systems.
SOURCE: EPA (2003a).

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
Associated Federal Regulations
There are several federal regulations that are designed to address spe
ci
fi
c
distribution system water quality issues, although none of these regulations deal
wholly with the integrity of distribution systems as defined in Chapter 1. The
following provides a brief description of each of these regulations.
National Interim Primary Drinking Water Regulations
Following the passage of the SDWA, EPA adopted the National Interi
m
Primary Drinking Water Regulations (NIPDWR) on December 24, 1975 and on
July 9, 1976. The NIPDWR established the first national standards for drinking
water quality. These standards included limits for ten inorganic chemicals, six
organic pesticides, turbidity, and five radionuclides. In addition, the NIPDWR
established standards for microbiological contamination based on total coliform
organisms.
Total Coliform Rule
The primary purpose of the TCR is to ensure public health protection from
microbial contamination of drinking water, and it applies to all public water sys -
tems. It is the only regulation that is intended to measure the microbiological
quality of water within that part of the distribution systemcontrolled by the pub-
lic water supply. In 1989 EPA promulgated the TCR as a revision to the exist-
ing regulation that required public water systems to monitor for coliform organ-
isms in the distribution system. The TCR changed the concept of monitoring for
coliform organisms from one based on measuring the concentration of coliforms
to determining the presence or absence of coliforms. In addition, the TCR es -
tablished an MCL based on the presence or absence of total coliforms, modified
monitoring requirements including testing for fecal coliforms or E. coli, required
the use of a sample siting plan, and also required sanitary surveys for water sys -
tems collecting fewer than five samples per month. The MCL for total coli-
forms is as follows:
 For a system serving more than 33,000 people and collecting more t
h
an40
samples per month, a non-acute violation occurs when more than 5.0 percent of
the samples collected during the month are total coliform positive.
 For systems serving 33,000 people or less and collecting less than 4
0
samples per month, a non-acute violation occurs when more than one sample is
total coliform positive in a given month.

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
 Any fecal coliform positive repeat sample, E. coli positive repeat s
a
m
-ple,
or any total coliform positive repeat sample following a fecal coliform or E
.coli
positive routine sample constitutes an acute violation of the MCL for total
coliforms.
The sampling frequency ranges from one sample per month for water sys-
tems serving 25 people to 480 samples per month for the largest of water
systems serving greater than 3,960,000 people (40 CFR 141.21 & 141.63).
Sam- pling locations, identified in the sample siting plan, are required to be
representative of water throughout the distribution system, including all
pressure zones and areas supplied by each water source and distribution
reservoir.
Trihalomethane Rule
In 1979 EPA promulgated a rule that established a drinking water standard
for trihalomethanes (THMs), a group of chemicals produced as a consequence of
chlorine disinfection. These chemicals are regulated because of the
concern over their potential carcinogenic risk. The drinking water standard
set at 0.10 mg/L addressed the total concentration of four specific THMs:
chloroform, di- chlorobromomethane, dibromochloromethane, and bromoform.
This rule was the first to regulate the chemical quality of drinking water in the
distribution system. The rule affected public water systems serving greater
than 10,000 people because EPA was concerned that smaller systems would
not have sufficient expertise available to deal with elevated levels of THMs
without compromising microbiological safety. Water systems were required
to sample quarterly at a minimum of four points in the distribution system
and determine the average concentration of the four sample points.
Compliance with the standard was based on the running average of any four
consecutive quarterly results (EPA, 1979).
Surface Water Treatment Rule
On June 29, 1989, the EPA published the SWTR in response to Congress’
mandate to require systems that draw their water from surface water sources
(rivers, lakes, and reservoirs) and groundwater under the influence of surface
water to filter, where appropriate, and to disinfect their water before distribution.
The SWTR seeks to reduce the occurrence of unsafe levels of disease-causing
microbes, including viruses, Legionella bacteria, and the protozoan Giardia
lamblia. The SWTR requires water systems that filter to meet specific turbidity
limits, and it assumes that this will achieve reductions in Giardia lamblia cysts
(99.9 per cent) and viruses (99.99 per cent). Also, water systems are required to
continuously monitor the residual disinfection concentration entering the distri-
bution system, except those serving less than 3,300 people, which are allowed
to

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
52 DRINKING WATER DISTRIBUTION SYSTEMS: ASSESSING AND REDUCING RISKS
collect grab samples. Furthermore, water systems (both filtered and unfiltered)
are required to ensure a residual disinfectant concentration of not less than 0
.
2
mg/L entering the distribution system and to maintain a detectable residual dis-
infectant concentration in the distribution system measured as total chlorine,
combined chlorine, or chlorine dioxide. The use of the heterotrophic
bacteria plate count (HPC) is allowed as a surrogate for a detectable disinfectant
in the distribution system provided that the concentration of heterotrophic
bacteria is less than or equal to 500 colony forming units/milliliter (EPA, 1989).
Samples for measuring residual disinfectant concentrations or heterotrophic
bacteria must be taken at the same locations in the distribution systemand at the
same time as samples collected for totalcoliforms.
Lead and Copper Rule
The LCR was published in June 1991 and is intended to address the concern
over chronic exposure of young children to lead in drinking water, the lead be-
ing principally from the leaching of the chemical from premise plumbing, fix-
tures, solder, and flux, and acute effects from copper. Indeed, since June 19,
1986, the use of solder and flux with more than 0.2 percent lead and the use of
pipes and pipe fittings with more than 8.0 percent lead in the installation or re -
pair of any public water system or plumbing in residential or non-residential
facilities has been prohibited. States are required to enforce these requirements
through state or local codes.
Unlike the TCR, which is intended to assess water quality that is representa-
tive of the entire distribution system in a dynamic or flowing state, the
LCR i
s
predicated on assessing water quality that represents worst case conditions.
The LCR established monitoring requirements for tap water at “primary”
locations— homes that contain lead pipes or copper pipes with lead solder
installed after 1982. These homes were generally identified through a review
of permits and records in the files of the building department(s) that indicate the
plumbing materials installed within publicly and privately owned structures
connected to the distribution system and the material composition of the
service connections. The number of required samples depends on the size of the
water system. Sam- ples are collected from interior taps where water is
typically drawn for consumption and after the tap has been left unused in a
static state for a minimum of six hours. Table 2-1 describes the standard and
reduced monitoring requirements of the LCR.
The LCR also established requirements for corrosion control treatment,source
water treatment, lead service line replacement, and public education. T
h
eLCR
establishes “action levels” in lieu of MCLs. The action level for lead w
a
s
established at 0.015 mg/L while the action level for copper was set at 1.3 mg/L.
An action level is exceeded when greater than 10 percent of samples collected
from the sample pool contain lead levels above 0.015 mg/L or copper levels
above 1.3 mg/L. Water systems exceeding the respective action level are

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
TABLE 2-1 Standard and Reduced Monitoring Requirements of the Lead and Copper R
u
l
e
System size
(number of people served)
Standard monitoring
requirements
(number of sites)
Reduced monitoring
requirements*
(number of sites)
100,000 100 50
10,001 to 100,000 60 30
3,301 to 10,000 40 20
501 to 3,300 20 10
101 to 500 10 5
< 100 5 5
*Utilities can reduce the number of sampling sites and the f requency of m
o
n
i
toring
f rom the required
semi-annual f requency to a lesser f requency if their water sy stem meets the f ollowing conditions:
Reduce to Annual monitoring if:
 the sy stem serves less than 50,000 people and the lead and copper l
e
ve
l
sare less than the
action lev el f or two consecutiv e six-month monitoring periods or,
 the sy stem meets Optimal Water Quality Parameter (OWQP) specificationsfor two consecu-
tiv e six-month monitoring periods
Reduce to Triennial Monitoring if:
 the sy stem serves more than 50,000 people and the lead and copper l
e
v
e
l
sare less than the
action lev el f or three consecutiv e y ears or,
 the sy stem meets OWQP specif ications for three consecutiv e y ears of monitoring o
r
,
 the sy stem has 90
th
percentile lead lev els less than 0.005 mg/L and 90
th
percentile copper
lev els less than 0.65 mg/L f or two consecutiv e six-month monitoring periods or,
 The sy stem has demonstrated optimized corrosion control
Reduce to Monitoring once every nine years if:
 the sy stem serves less than 3,300 people, the distribution sy stem, the s
e
r
v
ice
lines, and the
premise plumbing are f ree of lead-containing and copper-containing materials and,
 the sy stem has 90
th
percentile lead lev els less than 0.005 mg/L and 90
th
percentile copper
lev els less than 0.65 mg/L f or one six-month monitoring period.
required to install corrosion control treatment and conduct lead service line r
e
-
placement and mandatory lead education.
Information Collection Rule
In May 1996, EPA promulgated the Information Collection Rule (IC
R),
which established monitoring and data reporting requirements for large
public water systems including surface water systems serving at least
100,000 people and groundwater systems serving at least 50,000. The rule was
intended to provide EPA with information on the occurrence in drinking water
of (1) DBPs and
(2) disease-causing microbes including Cryptosporidium (EPA, 1996). E
P
A
used
the information generated by the rule to develop new regulations for disin-
fectants and DBPs (EPA, 2006a).

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
Operator Certification
Pursuant to the SDWA amendments of 1996, EPA in cooperation with the
states was directed to issue guidelines specifying minimum standards for certifi-
cation and recertification of the water treatment and distribution system opera-
tors of all public water systems. The guidelines were required to take into ac-
count the size and complexity of the system, existing state programs, and other
factors aimed at providing an effective program at reasonable cost to states and
public water systems (EPA, 1999). EPA, through grants to the states allocated
on the basis of “reasonable costs,” was required to reimburse training and certi-
fication costs for operators of systems serving 3,300 persons or fewer, including
an appropriate per diem for unsalaried operators who had to undergo training as
a result of the federal requirement. States are required to adopt and implement a
program for the certification of operators of public water systems that meet or
are equivalent to the requirements of the EPA guidelines.
Stage 1 Disinfection and Disinfection Byproducts Rule
On December 16, 1998, EPA published the Stage 1 D/DBPR, making m
o
r
e
stringent the existing standard for trihalomethanes as well as establishing ne
w
standards for disinfectants and other DBPs (EPA, 1998a). The rule, which a
p
- plies
to all public water systems, lowers the existing TTHM standard from 0
.1
0mg/L to
0.080 mg/L and establishes new standards for five haloacetic a
c
i
d
s(HAAs) at
0.060 mg/L, bromate at 0.010 mg/L, and chlorite at 1.0 mg/L. In addition,
the Rule establishes limits for disinfectants including chlorine,
chloramine, and chlorine dioxide within the distribution system (via Maximum
Residual Disinfectant Levels or MRDLs). For chlorine and chloramines, sam-
ples for measuring residual disinfectant must be taken at the same locations in
the distribution systemand at the same time as samples collected for total coli-
forms. For chlorine dioxide, samples must be taken daily at the entrance to the
distribution system. Compliance with the MRDLs for chlorine and chloramines
is based on the annual running average of all monthly samples collected, while
compliance with the MRDL for chlorine dioxide is based on each daily sample.
Finally, the Rule requires enhanced coagulation for certain systems in order to
achieve specific reductions of DBP precursor material (as measured by total
organic carbon concentrations).
Interim Enhanced Surface Water Treatment Rule
In December 1998, EPA promulgated the Interim Enhanced Surface Water
Treatment Rule (IESWTR) that applied to public water systems serving greater
than 10,000 people that were subject to the original SWTR. The IESWTR es -
tablished a requirement for the reduction of Cryptosporidium and a more strin-

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
gent turbidity requirement for filtered water supplies, among other provi
sions. The
IESWTR also requires certain water systems to evaluate their d
i
s
i
n
f
e
c
t
i
o
npractices to
ensure that there will be no significant reduction in microbial protection as the
result of modifying disinfection practices to meet MCLs specified by the Stage 1
D/DBPR. In addition, the IESWTR requires that all finished water storage
facilities, for which construction began after February 16, 1999, be covered.
EPA further indicated that it would consider whether or not to require the
covering of existing reservoirs during the development of subsequent microbial
regulations (EPA, 1998b).
Long Term 1 Enhanced Surface Water Treatment Rule
In 2002 EPA promulgated the Long Term 1 Enhanced Surface Wa
ter
Treatment Rule (LT1ESWTR). The LT1ESWTR applies to public water
systems that use surface water or groundwater under the direct influence of
surface water and serve fewer than 10,000 persons. The purposes of the
LT1ESWTR are to improve control of microbial pathogens, specifically
Cryptosporidium, in drinking water and to address risk trade-offs with DBPs.
The LT1ESWTR requires systems to meet strengthened filtration
requirements as well as to calculate benchmark levels of microbial inactivation
to ensure that microbial protection is not jeopardized if systems make changes
to comply with requirements of the Stage 1 D/DBPR (EPA, 2002a). The only
difference between this rule and the IESWTR is the size of the affected
community.
Stage 2 Disinfectants and Disinfection Byproducts Rule
On January 4, 2006, EPA adopted the Stage 2 D/DBPR that makes m
ore
stringent the previous rule regulating certain DBPs. Under the Stage 1 D/DBPR
water systems are allowed to average the DBP sample results from across
the distribution system. As a result some customers could be exposed to levels
of DBPs that consistently exceeded the MCLs and that might escape detection.
The new rule requires that water systems meet the MCLs for THMs and HAAs
at each sampling location based on the running annual average of any four con-
secutive quarterly sample results at that location. The intent of this change is to
reduce DBP exposure and provide more equitable health protection and to lower
potential cancer, reproductive, and developmental risks (EPA, 2006a).
To determine the locations within the distribution system where the highest
levels of THMs and HAAs are expected to occur, the Rule requires water sys-
tems to conduct an Initial Distribution System Evaluation. Initial Distribution
System Evaluations are studies that evaluate THM and HAA levels at various
points within the distribution system. The results from these studies along with
existing compliance monitoring information will be used to determine future
compliance monitoring locations.

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
Long Term 2 Enhanced Surface Water Treatment Rule
On January 5, 2006, EPA adopted the Long Term 2 Enhanced Surface W
a
-ter
Treatment Rule (LT2ESWTR). The LT2ESWTR applies to public w
a
t
e
rsystems
that use surface water or groundwater under the direct influence of surface
water. The purpose of the LT2ESWTR is to reduce disease incidence asso-
ciated with Cryptosporidium and other pathogenic microorganisms in drinking
water. The LT2ESWTR supplements existing regulations by targeting addi-
tional Cryptosporidium treatment requirements to higher risk systems based on
actual monitoring data of source water quality.
The LT2ESWTR also contains provisions to mitigate risks from uncovered
finished water storage facilities. Water systems with uncovered finished water
storage reservoirs are required to cover the reservoir or treat the reservoir dis -
charge to the distribution system to achieve inactivation and/or removal of at
least 2-log Cryptosporidium,3-log Giardia, and 4-log virus (EPA, 2006b).
Finally, to ensure that systems maintain microbial protection as they t
a
k
e
steps to reduce the formation of DBPs the LT2ESWTR requires water systems
that proposed to modify their disinfection process to reduce THMs and HAAs to
assess the existing levels of disinfection that the system provides. Systems are
required to establish a benchmark, which is the system’s lowest monthly average
microbial inactivation. If the benchmark is more than the required inactivation
of 3-log removal for Giardia and 4-log removal for viruses, the system may
consider decreasing the amount of disinfectant added or the contact time, or al-
tering other disinfection practices to lower THM and HAA levels (EPA, 2006b).
Unregulated Contaminant Monitoring Rule 2
On August 22, 2005, EPA proposed the second of two Unregulated C
on
-
taminant Monitoring Rules (UCMR2), which will require monitoring for a list of
26 chemical contaminants suspected to be present in drinking water. The pur-
pose of the UCMR2 is to develop data on the occurrence of these contaminants
in drinking water, the size of the population exposed to these contaminants, and
the levels of the exposure. This information will be used along with health ef -
fects information to determine whether or not drinking water standards should
be established for these contaminants. All community water systems and non-
transient, non-community water systems serving more than 10,000 people will
be required to monitor, while a representative sample of 800 community water
systems and non-transient, non-community water systems serving less than
10,000 people will have to carry out monitoring. The monitoring is proposed to
begin in 2007.
Unlike the first UCMR (which is not discussed above), the UCMR2 will i
n-
clude contaminants that are considered potential DBPs and for which monitoring
will be conducted in the distribution system. These contaminants include the
nitrosamines N-nitroso-diethylamine (NDEA), N-nitroso-dimethylamine

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
(NDMA), N-nitroso-di-n-butylamine (NDBA), N-nitroso-di-n-propylamine
(NDPA), N-nitroso-methylethylamine (NMEA) and N-nitroso-pyrrolidine
(NPYR). Nitrosamines are considered potential human carcinogens, a
n
d
NDMA has
been shown to form in chlorinated or chloraminated water as a re - sult of
disinfection (EPA, 2005b).
Water Security-related Directives and Laws
Although not a new issue, security has become paramount to the water uti
l
-ity
industry since the events of September 11, 2001. The potential for natural,
accidental, and purposeful contamination of water supply has been present f
o
r
decades whether in the form of earthquakes, floods, spills of toxic chemicals, o
r
acts of vandalism. For example, in May 1998, President Clinton issued Presi
-
dential Directive (PDD) 63 that outlined a policy on critical infrastructure pro-
tection, including our nation’s water supplies. However, it was not until
after September 11, 2001, that the water industry truly focused on the
vulnerability of the nation’s water supplies to security threats. In recognition
of these issues, President Bush signed Public Health Security and Bioterrorism
Preparedness and Response Act of 2002 (the “BioterrorismAct”) into law in
June 2002 (PL107-188). Under the requirements of the Bioterrorism Act,
drinking water utilities are required to prepare vulnerability assessments and
emergency response plans for water systems serving at least 3,300 people.
***
Table 2-2 summarizes the key requirement(s) of federal rules and r
e
gul
a-tions
from a distribution systemperspective.
State Regulatory Programs
State regulatory programs that address water distribution systems can v
a
r
y
significantly. In general most states have statutory and regulatory requirements
that cover (1) design, construction, operation, and maintenance of di
stri
bution
systems, (2) cross-connection control, and (3) plumbing products certified for
use pursuant to American National Standards Institute/ NSF International
(ANSI/NSF) standards 60 and 61. Furthermore, most states have adopted a
plumbing code that dictates the types of materials that can be used for premise
plumbing, although these codes are not generally enforced from a state statutory
or regulatory standpoint but rather are implemented at the local county and/or
municipal level.

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
TABLE 2-2 Summary of Regulated Distribution SystemRequirements
Law/Rule/
Regulation
Key DistributionSystemRequirements
SDWA  Established national primary and secondary drinkingwater regulations
(MCLs and MCLGs)
 Allow ed EPA to establish point of compliance
NIPDWR
 Adopted at the passage of the SDWA and required t
h
a
t
representative
coliformsamples be collected throughout the distribution system
THM Rule
 Established a standard for totalTHMs of 0.10 mg/L
 Compliance based on the annual average of THM l
e
v
e
l
sat all monitoring
locations w ithin the distribution system
TCR
 Regulates coliformbacteria, w hich are used as “
s
u
r
r
o
g
a
t
e
”organisms to indicate
w hether or not systemcontamination is occurring
 Compliance based on results fromrepresentative m
o
n
i
t
o
r
i
n
g
locations w ithin the
distribution system
SWTR
 Requires that a detectable disinfectant residual be m
a
i
n
t
a
i
n
e
d
at repre-
sentative locations in the distribution system
 Requires continuous monitoring of disinfectant r
e
s
i
d
u
a
l
entering the distribution
systemfor w atersystems serving greater than 3,300 people
LCR
 Requires that lead and copper concentration be b
e
l
o
w
action levels in
samples taken at the w orst caseor highest riskconsumer's tap
ICR
 Provides monitoring data to support the interim and l
o
n
g
-
t
e
r
menhanced SWTR
and Stage 2 DBP rule
1996
SDWAA
 Focused on the role that surface waterquality can p
l
a
yin influencing the
quality of distributed w ater
 Established requirement for certification of operat
or
sof watersystems
including w ater distribution systemoperators
IESWTR
 Enhances protection from pathogens, including C
r
y
p
t
o
s
p
o
r
id
i
u
m
,and tries to prevent
increases in microbial risk for large systems w hile they comply w ith the
Stage 1 D/DBPR
 Prohibits the construction of new uncovered f
i
n
i
s
h
e
dwater storage facilities
Stage 1
D/DBPR
 Low ersthe standard for totalTHMs from0.10 mg/L t
o 0.08 mg/L. This
standard applies to all community w ater supplies in the U.S.
 Set an MCL for 5 HAAs of 0.06 mg/L.
continues

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
TABLE 2-2 Continued
Law/ Rule/
Regulation
Key DistributionSystemRequirements
LT1ESWTR
 Enhances protection from pathogens, including C
r
y
p
t
o
s
p
o
r
i
d
i
u
m
, and tries to
prevent increases in microbial risk for systems serving less than
10,000 people w hile they comply w ith the Stage 1 D/DBPR
Stage 2
D/DBPR
 Requires an Initial Distribution SystemEvaluation (
I
D
S
E
s
)
 Compliance based on the locational running annual average of total
THM and HAA levels at each monitoring location w ithin the distribu-
tion system
LT2ESWTR
 Requires additional Cryptosporidium treatment for h
i
g
h
risk systems
and maintenance of microbial protection while reducing the formation
of DBPs
 Requires uncovered finished water storage fac
i
l
i
ti
esto be covered or the
discharge from the finished w ater storage facilities to the distribution
system to be treated to achieve inactivation and/or removal of at
least 4-log virus, 3-log Giardia, and 2-log Cryptosporidium
UCMR2
(Proposed)
 Will require distribution systemmonitoring for n
it
rosamines to deter-
mine their occurrence as DBPs
Requirements for Design, Construction,Operation,and Maintenance
Using their existing statutory authority, many states have established r
e
-
quirements for the design, construction, operation, and maintenance of distribu-
tion systems. This was revealed in a survey of state drinking water
programs conducted by the Association of State Drinking Water Administrators
(AS- DWA) in March 2003. Of the 34 states responding, the majority reported
having some requirements for water-main design and construction, storage
facilities and pump station design and construction, and distribution system
operation and maintenance (ASDWA, 2003). A summary of the responses is
provided in Ta- bles 2-3, 2-4, and 2-5, respectively.
There appears to be less consistency between states, however, regarding the
individual elements that each state requires be met. For example, most states
have requirements for minimum operational pressures and the types of pipes that
can be used, while less than half the states have requirements for storage and
handling of pipes and distribution system maintenance plans. Only a small
number of states have requirements for nitrification control and storage tank
water quality monitoring. States also use different approaches for establishing
these requirements. In some cases states have established their own require -
ments, while in others requirements are based on third party standards such as

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
TABLE 2-3 Summary of Results fromthe ASDWA Distribution Systemand Total C
o
l
i
f
o
r
m
Rule Survey: Water Main Design and Construction
Numbers of States
Element
Required Encouraged
Not
Addressed
Minimum pipe diameter (set m
i
ni
m
um or
size based on flow, number of service
connections, etc.)
26 3 5
Design for an operational p
r
e
s
s
u
r
e
of at
least 20 psi under all flow conditions
32 0 2
Minimum flow velocity through p
i
p
e
s 9 6 19
Maximum flow velocity through pipes 9 8 17
Pipe material 30 2 2
Storage and handling of pipes 16 7 9 (2 NR)
Minimum depth of cover over pipes t
o 25 7 2
prevent freezing and damage
Pressure/leakage testing before p
l
a
c
i
n
g 26 7 1
new mains into service
Disinfection, flushing, and m
i
c
r
o
b
i
a
ltest- 29 5 0
ing before placing new mains into service
Looping of pipes/minimization of dead 17 15 2
ends
Proper flushing devices at dead e
n
d
s 23 9 2
Protection of air-release and air v
a
c
u
u
m 22 9 1 (1 NR)
valves
Isolation valves at intersections a
n
d
over 23 8 3
lengthy stretches of water main
Separation of w ater mains and sanitary 29 4 1
sew ers to protect the watermain from
contamination
Protection of w ater main at surfacewater 21 11 2
crossings
Exterior corrosion protection of w
a
t
e
r 14 12 8
mains
Cross connection c
o
n
t
r
o
l
/
b
a
c
k
f
l
o
w
prevention (through
the d
r
i
n
k
i
n
g
w ater
29 2 3
program)
NR: No Response
Note: State practices may hav e changed since 2003. This surv ey is not a complete census of all state
drinking water programs, rather it is indicativ e of the practices of the 34 states that responded to the
surv ey .
SOURCE: Reprinted, with permission, f rom ASDWA (2003). © 2003 by ASDWA.

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
o
l
i
f
o
r
m
Rule Survey: Storage Facilities and Pump Station Design and Construction
Numbers of States
Element
Required Encouraged
pumps
surv ey .
those developed by the American Water Works Association (AWWA)or t
he
Ten
State Standards (ASDWA, 2003).
Cross-Connection Control Requirements
One of most common means of contaminating distribution systems i
s
through a cross connection. Cross connections occur when a nonpotable water
source is connected to a potable water source. Under this condition contami-
nated water has the potential to flow back into the potable source. Backflow can
occur when the pressure in the distribution systemis less than the pressure in the
nonpotable source, described as backsiphonage. Conditions under which
backsiphonage can occur include water main breaks, firefighting demands,
and pump failures. Backflow can also occur when there is increased pressure
from the nonpotable source that exceeds the pressure in the distribution system,
described as backpressure. Backpressure can occur when industrial operations
connected to the potable source are exerting higher internal pressure than the
pressure in
Not
Addressed
Standards for tankdesign and c
o
n
s
t
r
u
c
t
i
o
n 28 5 1
Tanks designed to ensure adequate t
u
r
n
o
v
e
r 15 16 3
Storage tank vents, screens, over
fl
o
ws
,and 30 4 0
access hatches
Telemetry or other means for control- 15 15 4
ling/monitoring the storage f
a
c
i
l
i
t
y
Provisions for draining the storage f
a
c
i
l
i
t
y 22 10 2
Standards for paints and coatings a
n
d
provi- 31 3 0
sions for testing before p
l
a
c
i
n
g
the storage
facility in service
Cathodic protection for storage facilities 15 12 7
Standards for pump station design a
n
d
con- 26 6 2
struct
Drainage of underground pump stations and 22 8 3 (1 NR)
valve vaults
Minimum inlet pressure for in-line b
o
o
s
t
e
r 25 7 2

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
Element
o
l
i
f
o
r
m
Rule Survey: Distribution System Operation and Maintenance
Numbers of States
Required Encouraged
Not
Addressed
Operational pressure ≥ 20 psi under a
l
l
flow 30 3 1
conditions
Distribution systemmaintenance plan 16 11 7
Routine distribution systemflushing, c
l
e
a
n
i
n
g 11 20 3
and/or pigging
Valve and hydrant exercise/ m
a
i
n
t
e
n
a
n
c
e 10 19 5
plan
Telemetry or other means for control- 7 14 13
ling/monitoring the DS
Unaccounted for w aterrequirements 12 13 9
Disinfection, flushing, testing, and o
t
h
e
r
fol- 26 8 0
low -up action before returning a w ater
main to service after repairs
Tank flushing 5 19 10
Tank inspection and maintenance 13 16 5
Tank cleaning 8 18 8
Provisions for testing before p
l
a
c
i
n
g
the stor- 24 8 2
age facility backin service fo
l
l
o
wi
n
g
clean-
ing/maintenance
Maintaining a minimum disinfectant r
e
s
i
d
u
a
l 21 7 6
in groundw ater systems (if d
i
s
i
n
f
e
c
t
i
o
n
is provided)
Storage tank w ater quality monitoring 5 11 18
Nitrification control 4 7 23
Other w ater quality monitoring in the distribu-
tion system(beyond the SWTR, TCR, and
17 8 8 (1 NR)
LCR)
surv ey .

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
the distribution systemor when irrigation systems connected to the potable s
y
s
-
tem are pumping from a separate water source and the pump pressure exceeds
the distribution system pressure.
Of 30 states surveyed by ASDWA, the vast majority required some sort o
f
cross-connection control program, either through regulations (23 states) o
r
guidelines, that is administered by the Drinking Water Program or as part of the
State’s Plumbing Code (ASDWA, 1999). However, these requirements and the
authority to implement them vary considerably in terms of how detailed a water
system’s program must be, the types of systems (community and/or non-
community) required to have a program, and the role the states play in imple -
menting and maintaining a program. Some states rely solely on plumbing codes
to address cross connections and backflow, which is problematic because
plumbing codes, in most cases, do not require testing and follow-up inspections
of backflow prevention devices.
A similar assessment of state cross-connection control requirements by E
P
A
(EPA, 2002b), which is summarized in Table 2-6, demonstrates the variability i
n
state requirements. Based on the EPA review, there are 48 states which ha
ve
some minimum requirement relating to cross connections in their state adminis -
trative code or state law (EPA, 2002b). A number of states do not go beyond
these minimum requirements or require public water systems to administer any
type of cross-connection control program at the local level. These states tend to
rely on community water systems to implement cross-connection control pro-
grams. In a few cases, states specify that systems which serve a population of a
certain size category must implement a cross-connection controlprogram.
There are five primary elements of an effective cross-connection control
program. The first is authority; effective cross-connection control programs
must have the legal authority to implement program requirements. Legislation
must provide the authority to: (1) enter premises and inspect facilities to deter-
mine hazards; (2) install, repair, and test backflow devices; (3) license inspectors
to test assemblies; and (4) terminate water service in case of non-compliance.
According to the American Backflow Prevention Association State Program
Survey (ABPA, 1999), 16 of 26 states require utilities to have the authority to
implement program requirements. However, on average only 55 percent of sys -
tems required to have an enforceable program actually have one in place.
The second requirement is to inspect facilities and test devices. It is impor-
tant to conduct site inspections, and the right of entry enables the
inspector t
o
identify where a high hazard might exist. The frequency of inspections
and testing is typically based on the degree of hazard. A testing program must
identify the appropriate standards that a backflow prevention device must meet,
and assemblies must be tested by a certified backflow assembly tester.
Many states require in regulation some of the critical components that make up a
testing program. For example, 35 of 50 states specified a list of design
standards that backflow assemblies must meet, and 34 of 50 states stipulated a
testing frequency interval for various backflow assemblies in their regulations
(EPA, 2002b). A

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
TABLE 2-6 State Cross-Connection Control Requirements
Number of
Requirement
Does the state have a requirement for the control o
f
cross-connections
and/or backflow prevention?
Is it specified in the requirement that the systemmustimplement or
develop a cross-connection controland/or backflow prevention
program?
Does the state require authority to implement a l
o
c
a
l
ordinance or rule for
cross-connection controland/or backflow prevention?
States with
Requirement
50
32
33
Must the authority cover testing of backflow prevention assemblies? 27
Must the authority cover the use of only licensed or certified backflow
assembly testers?
16
Must the authority cover the entry of the premises for the sake of in-
specting the premises?
14
Must the authority cover the entry of the premises for the sake of in-
specting and/or installing backflow prevention assemblies?
15
Does the state require training, licensing, or certification of backflow
prevention assembly testers?
26
prevention assembly and/or device installers?
6
prevention assembly and/or device repairers?
10
Does the state require training, licensing, or certification of cross-
connection controlinspectors?
19
Does the state require inspection of backflow prevention devices
and/or testing of backflow prevention assemblies?
37
Does the state require the systemto include recordkeeping as part of
cross-connection control?
34
Does the requirement include keeping records of hazard assessment
surveys?
11
Does the state require the systemto notify the public follow ing the
occurrence of a backflow event?
3
Does the state require the local rule or ordinance to allow the system
to take enforcement action against customers w ho do not comply 23
w ith the cross-connection controland backflow prevention require-
ments?
Does the state conduct periodic reviews of cross-connection control
programs?
3
Does the state regulation or plumbing code require public education
regarding cross-connection controland/or backflow prevention?
7
SOURCE: EPA (2002b).

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
fewer number of states included certification specifications for testers in regula-
tion.
A third issue is training and certification. The testing of backflow p
r
e
v
e
n
-tion
assemblies by a certified tester is necessary to ensure that the assembly is
functioning properly and will prevent backflow. The EPA survey revealed that
26 of 50 states require certification of backflow assembly testers (EPA, 2002b).
The states often require the tester to pass a proficiency test and written exam to
qualify for certification. A smaller number of states expand their training
requirements to program managers, installers, and/or repairers. States rely on
plumbers for cross-connection control testers/repairers, survey inspectors, and
program managers. Twenty-seven (27) percent of the training was conducted by
plumber-affiliated organizations, 15 percent by AWWA-affiliated organizations,
12 percent by state agencies, 6 percent by others, and 40 percent did not specify
the source of training.
A fourth important element is record keeping following inspections a
n
d
testing. According to the ABPA survey, 17 of 26 states require record keeping,
and 10 of 26 states indicated a requirement for water systems to report backflow
incidents to the state. Additional details are found in Table2-7.
Public education is a final critical element. According to the ABPA survey,
five of 26 states required public awareness of backflow potential as an element
of their cross-connection control program. Public education is usually a func-
tion of the local water purveyor which may educate the public through bill in -
serts and special mailings. States also maintain internet sites that educate con-
sumers about cross-connection control programs and the role they play in pro-
tecting the public’s drinking water.
TABLE 2-7 ABPA State Survey Results on Record Keeping Requirements
Record Keeping Requirement PercentofStates
Number of States that require record keeping (17 of 26) 65%
Records of inventory of backflow assemblies i
n
service (14 of 26) 53%
Records of reports of routine testing of a
s
s
e
m
b
l
i
e
s (16 of 26) 61%
Records of hazard assessment surveys (9 of 26) 34%
Records of enforcement activities (8 of 26) 30%
Number of States w hich require annual reporting t
o
the States (6 of 26) 23%
Number of States w hich require reporting of b
a
c
k
f
l
o
w
incidents (10 of 26) 38%
SOURCE: Reprinted, with permission, f rom The American Backf low Prev ention Association (ABPA)
State Program Surv ey (1999). © 1999 by ABPA.

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
At the current time, there is no unified basis from which cross-connection
control programs are designed, adopted, and implemented, which is reflected i
n
the immense variability in programs discussed above. EPA has not adopte
d
national cross-connection control program requirements, although the agency
has provided guidance on cross-connection control issues for approximately two
decades through its Cross-Connection Control Manual. In 2003 EPA published
the third edition (EPA, 2003b), which is designed as a tool for health officials,
waterworks personnel, plumbers, and any others involved directly or indirectly
in water supply distribution systems. It is intended to be used for educational,
administrative, and technical reference in conducting cross-connection control
programs. Interestingly, the states that have strong cross-connection control
programs are generally not in favor of greater EPA involvement because their
programs might be compromised. Those states with programs that are lacking,
however, could benefit greatly from EPA directives.
An indirect benefit of a cross-connection control program that has an effec-
tive inspection aspect is its ability to identify improper customer account infor-
mation, missing water meters, unauthorized use of water, and illegal
connections. This can result in a reduction in lost water and in the generation
of more revenue.
Requirements for Drinking Water Products, Components,and Materials
Because of the potential for drinking water products, components, and m
a
-
terials to add contaminants to drinking water, EPA initiated the development o
fa
Drinking Water Additives third party certification program in 1985. The pur-
pose was to establish standards by which products, components, and materials
would be tested to ensure that contaminants of health concern would not intro-
duced into drinking water at levels that imposed a risk to the public. The result-
ing standards—ANSI/NSF Standard 60 and ANSI/NSF Standard 61—were ini-
tially adopted by NSF through a consensus standards development process
in October 1988. These standards are designed to test products that are added to
drinking water (Standard 60) and products, components, and materials that come
into contact with drinking water (Standard 61).
ANSI/NSF Standard 61 is the more relevant standard with regards to water
distribution systems. Thirty-six (36) states have adopted ANSI/NSF Standard 61
by either statute or regulation and thus require water systems to use only w
a
-ter
distribution system products, components, and materials that are certi
fi
edpursuant
to the standard. Eight additional states have policies (but not requirements)
that water systems use products, components, and materials that meet the
standard (ASDWA, 2004). Standard 61 applies to all distribution system mate-
rials (including pipes, valves, coatings, storage tank materials, etc.) as well as to
premise plumbing including home water faucets. These standards can be used by
water utilities (along with AWWA industry standards) in the specification of
materials they purchase or allow to be installed in theirsystems.

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
Plumbing Codes
Plumbing codes are used by states, territories, counties, local governments,
and any other formof governance which has a responsibility to protect their
constituents’ health and safety. Plumbing code requirements do not
generally apply to the utility-owned portion of public water systems but rather to
residential and non-residential property. Accountability in enforcing the codes
primarily resides with the inspection entity, though in many states the licensed
plumber and design professionals are also held accountable. Once adopted the
codes are used by all sectors of the plumbing industry and public,
including inspectors/plan reviewers; contractors/masters;
journeymen/apprentices; engineers/architects; material, pipe, and product
manufacturers; and certification organizations and test labs. Plumbing codes are
usually implemented by the “Authority Having Jurisdiction”, which can be a
state agency, county commission, or local building department. In some cases
plumbing codes are implemented by agencies of the federal government such
as the Army Corps of Engi- neers, Air Force, or the Department of
Housing and Urban Development (Chaney, 2005).
The major plumbing codes include the Uniform Plumbing Code (UPC), the
International Plumbing Code (IPC), and the Southern Building Code Congress
International. As indicated in Table 2-8, by 1999 47 states had adopted plumb-
ing codes, with the UPC, developed and maintained by the International Asso-
ciation of Plumbing and Mechanical Officials (IAPMO), being the most com-
monly used code (14 states) (EPA, 2002b). More recent information indicates
that the various codes were amalgamated by the year 2000 into the three codes
that are in use today: the UPC, the IPC, developed and maintained by the Inter-
national Code Council (ICC), and the National Standard Plumbing Code
TABLE 2-8 Plumbing Codes Adopted by the States by 1999
Plumbing Code Number of States Adopting
Statew ide Code 47
No Statew ide Code 3
Statewide Codes Adopted
Uniform Plumbing Code 14
State Code 7
International Plumbing Code 5
National Standard Plumbing Code 4
Southern Building Code Congress I
n
t
e
r
n
a
t
i
o
n
a
l 4
Other 13
SOURCE: EPA (2002b).

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
(NSPC). NSPC, published by the Plumbing, Heating, and Cooling
Contractors National Association, is adopted in New Jersey and some counties of
Maryland but is otherwise not used widely. The UPC has now been adopted in
approximately 28 states (Chaney, 2005).
The UPC and IPC have different contents and permit different materials a
n
d
devices. The UPC, for instance, allows for some piping material that is
n
o
t
permitted under the IPC. The IPC permits air admittance valves not permitted in
the UPC. Some venting configurations are permitted in one code and not the
other. Both the UPC and the IPC include important cross -connection
control requirements intended to prevent contamination of the domestic water
supply that is internal to the property as well as to the drinking water delivered
by the public water system. Both codes also establish minimum requirements
for the separation of water and sewer lines as well as requirements for the
disinfection of new or repaired potable water systems. Both codes, however,
have certain shortcomings. For examples, the UPC does not prohibit the
installation of water service or water distribution pipe in soil contaminated with
solvents, fuels, organic compounds, or other detrimental material which could
cause permeation, corrosion, degradation, or structural failure of the piping
material. The UPC does not require that water service and distribution pipe and
fittings conformto ANSI/NSF Standard 61, which is intended to prevent the use
of materials that will leach contaminants into drinking water at levels that may
constitute a health risk. The IPC requires that all cross-connection control
devices be inspected annually including devices that cannot be tested and air
gaps, while the UPC only requires inspection of testable devices. Inspection of
all devices is prefer- able to ensure that tampering has not occurred. Both the
IPC and UPC have established minimum distances between water supply wells
and sewage disposal systems. The distances established by the IPC are less
conservative and may not provide adequate protection from potential
contamination. A comparison of the two codes with regard to the principal
requirements within the codes that address water distribution system integrity
is contained in Table 2-9.
The major difference between the UPC and IPC is the procedural
process by which the codes are maintained. IAPMO uses an American National
Stan- dards Institute (ANSI) consensus development process for the UPC, while
the ICC uses a government or inspector only process for the IPC. The ICC pre-
dominantly consists of building inspectors from three organizations (Building
Officials and Code Administrators, Southern Building Code Congress
Interna- tional, and International Conference of Building Officials) that have
been widely involved in developing structural and fire codes for years. The
ANSI consensus code development and maintenance process used by IAPMO is
open to all interested parties, it is balanced to prevent any one sector of the
industry from domi- nating, and it provides for due process (participants have
appeal rights to ANSI) (Chaney, 2005). Given the disparities between the codes,
and the possible resulting confusion, efforts are underway to combine the UPC
and the IPC into a single model code (IAPMO, 2005).

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
TABLE 2-9 Comparison of UPC and IPC: Requirements for the Protection of Water Sys-
tem Distribution Systems
Element UPC IPC
Code Maintenance ANSI Consensus Process Inspectors from Specific Process
Organizations
Cross-Connection Control Requirements
Devices Similar device requirements for
degree of hazard, but I
P
C
more
detailed regarding type of
device and application
Minimum R
e
q
u
i
r
e
dAir Gaps
Protection f
r
o
m
Law n
Irrigation Systems
Protection f
r
o
m
Fire
Sprinkler Systems
Inspections a
n
d
Testing
Similar requirements b
u
t
UPC
provides more detail
UPC requires i
n
s
p
e
c
t
i
o
n
s
of testable
devices only
Same requirements except f
o
r
¾ inch
openings affected by side w all
w here IPC more restrictive
Similar requirements but IPC is
more specific as to requirements
for systems not under constant
pressure
IPC requires inspection of t
e
s
t
a
b
l
e
and
non-testable devices and air gaps
Additional Distribution System Requirements
Separation of W
ater
and Sewer Lines
Disinfection ofNew
or Repaired Water
Pipe
Identification o
f
Po-
table and Nonpo-
table Water Systems
Requires minimum 12 i
n
c
h
vertical separation
Flushing w ith potable
w ater; 50 parts per million
(ppm) of chlorine
solution/24 hours or 200
ppm for 3 hours; flush to
purge chlorine; bacterio-
logical analysis
UPC requires color c
o
d
i
n
g
of each
system
Require minimum 12-inch v
e
r
t
i
c
a
l
separation but IPC is more r
e
s
t
r
i
c
-
tive on
horizontalclearance w here
verticalclearance is less than 12
inches
Flushing with potable w
a
t
e
r
;50 parts per
million (ppm) of c
h
lo
r
inesolution/24
hours or 200 ppm for 3 hours;
flush to purge chlorine;
bacteriologicalanalysis
IPC requires color coding or metal
tags
Pipe Materials UPC does not require pipe
material to meet A
N
S
I
/
N
S
F
61
IPC requires pipe m
a
t
e
r
i
a
l meet ANSI/NSF
Standard 61
Pipe Placement UPC does not address IPC prohibits placement of w ater
pipe in soils contaminated w
ith
contaminants that could adversely
affect the pipe
Water Supply Protection Requirements
Water Supply W
e
ll
Protection
UPC requires 50 feet b
e
-
tween
w ater supply w
e
l
l
s
and sewage
disposal s
y
s-tems such as
septic tanks and 100 feet
betw een watersupply wells
and disposal fields
IPC requires 25 feet betweenwa-ter
supply w ells and sewage dis-
posalsystems such as septic
tanks and 50 feet betw een water
supply w ells and disposal fields
Note: Where certain entries are blank, the two codes are similar and the sma
l
ldif f erence is mentioned
f or only one of the codes. SOURCES: IPC (2003); UPC (2003); Chaney (2005).

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
In the United States, plumbing codes are adopted in one of two ways: (1
)
through statutory adoption which usually occurs through the enactment of legis -
lation or (2) through regulatory adoption which occurs upon the implementation
of regulations or procedures. At the state level, codes are usually adopted
through a public hearing process that allows interested parties to present testi-
mony (Chaney, 2005).
Although states will adopt the UPC or IPC as their base plumbing c
o
d
e
,
they may amend the code to address specific issues. In addition, pl
um
bi
ng
codes may also be adopted at the local county and municipal level that are at
least as stringent as the state plumbing code. For example, in Iowa, the state
adopted the UPC as the plumbing code but then amended the UPC to add addi-
tional backflow prevention provisions including a requirement that cities with
populations of 15,000 or greater enact a backflow prevention program with con-
tainment by January 1, 1996. Although local jurisdictions in Iowa must adhere
to the provisions of the state plumbing code, these jurisdictions may adopt local
ordinances or rules and regulations that provide for higher but not lower stan-
dards than those found in the state plumbing code (State of Iowa, 2005).
As examples, the City of Des Moines, and Linn County, Iowa have adopted the
UPC with some modifications. In the case of Linn County the modifications
require the examination, qualification, and licensing of plumbing
contractors, plumbers, and the registration of apprentice plumbers (Linn County,
2004). In addition, homeowners are prohibited fromcarrying out plumbing work
on their residence unless they pass the County’s homeowners examination.
LIMITATIONS OF REGULATORY PROGRAMS
Existing federal regulations such as the TCR, SWTR, LCR, L
T
1
E
S
W
T
R
,and the
Stage 1 and Stage 2 D/DBP Rules are intended to address only certain aspects
of distribution system water quality and are not designed to address the
integrity of the distribution systemin its totality. Of these regulations, only the
TCR may provide some indication of potential problems with distribution sys-
tem integrity related to microbial contamination. However, the TCR has signifi-
cant limitations that affect its use as an indicator of distribution systemintegrity.
TCR sampling requirements are based on water systemsize and as a result vary
widely, from as many as hundreds of samples per month to one sample per
month. Each water system is required to develop a sample siting plan that is
approved by the state regulatory agency.For larger water systems even a sam-
ple siting plan that results in hundreds of samples per month may not adequately
cover the myriad of potential points where contamination could occur, such as
storage tanks,premise plumbing, and service connections.For smaller systems
the sampling is so infrequent that contamination would be easily missed. Al-
though most reported outbreaks associated with distribution systems have oc-
curred in community water systems because of their greater size and complexity,
there have been a number of outbreaks associated with noncommunity
water

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
systems that have been attributed to deficiencies in the distribution
system. I
n
addition to the problems associated with sample locations and the
frequency of sampling, TCR monitoring does not provide real-time
information. There are inherent delays between sampling and reporting of
coliform results that do not allow for sufficient time to recognize a
contamination event and to prevent public exposure and disease transmission.
(It generally takes about 24 hours to obtain results from the time of sample
collection to the completion of coliform analysis using presently available
analytical methods.)
The TCR encompasses only microbiological indicators. With the exception
of monitoring for disinfectant residuals and DBPs within the distribution system
and lead and copper at the customer’s tap, existing federal regulations do not
address other chemical contaminants within the distribution system. Yet there
have been a number of examples of waterborne outbreaks associated with
chemical contamination (chlordane, ethylene glycol) of the distribution system
as a result of cross connections, contamination of water mains during construc-
tion, and contamination of storage facilities (Craun and Calderon., 2001; Black-
burn et al., 2004).
Some federal regulations are inherently contradictory to one another, a
s
they
relate to distribution integrity and maintenance of water quality, such that water
suppliers have found it difficult to be in compliance with both simultaneously.
For example, the SWTR and TCR recommend the use of chlorine to minimize
risk from microbiological contamination. However, chlorine or other
disinfectants interact with naturally occurring organic matter in treated water to
form DBPs. As a result many water systems have changed disinfectants (gener-
ally from chlorine to chloramine) in order to be in compliance with the MCLs
for DBPs in the distribution system. The increased reliance on chloramine can
be problematic if close attention is not paid to controlling nitrifying bacteria
in the distribution system. Biological nitrification can result in the loss of
chloramine residual, which may then present a health threat to the consumer (as
discussed in Appendix A). Simultaneous compliance with the D/DBPR and the
LCR can also create problems for the maintenance of distribution integrity and
water quality. Raising the pH of treated water will assist in controlling corro -
sion (and hence reduce lead concentrations) but may increase the formation of
THMs.
In areas where federal regulations are weak, state regulations and local or-
dinance contribute to public safety from drinking water contamination. States
have adopted requirements that address certain aspects of distribution
system integrity. All states appear to have provisions for the control of cross
connections and/or backflow prevention, although there is considerable
variation in how they are implemented and by whom. The majority of
states have established regulations within their drinking water programs
requiring cross- connection control programs to be implemented by water
systems or local authorities, while some have adopted plumbing codes that
included the requirements and others have established only guidelines for
cross-connection control programs (ASDWA, 2003; EPA, 2002b). In general,
very few states provide

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
dedicated resources for implementing a cross-connection control program b
u
t
rather incorporate the program activities into the overall public water sys
t
e
m
supervision program. At best, most states attempt to assess that a water systemhas
an effective cross-connection control program when carrying out a sanitary
survey of the water system. However, because sanitary surveys may occur only
once every several years, it is difficult to ascertain the level of compliance. A
few states track the number of cross-connection control devices that are annually
installed and tested while others determine programs effectiveness by the num-
ber of backflow incidents reported (ASDWA, 1999).
Although most states have also established requirements for the desi
gn,
construction, operation, and maintenance of distribution systems, as di
scussed
previously these requirements vary significantly and some states only encourage
certain contamination prevention activities while others do not address themat
all. For example, some states only encourage the separation of water mains and
sanitary sewers to protect the water main from contamination or the disinfection,
flushing, testing, and other follow-up actions before returning a water main to
service after repairs. Even where states have established extensive require-
ments, the onus for ensuring implementation is placed on the water system.
States do not dedicate resources to routinely oversee that implementation occurs.
Local regulatory programs are implemented through the plumbing code.
Because local plumbing codes must be consistent with the provisions of the s
t
a
t
e
plumbing codes, local regulatory programs should have the authorities to a
d-dress
certain distribution system integrity issues including cross-connection control,
use of appropriate pipe and other plumbing materials, and separation of
water and sewer lines. However, program implementation can vary from
one local jurisdiction to another. For example, licensing of plumbing
contractors and plumbers is normally part of the local jurisdictions regulatory
program. Neither of the two prominent plumbing codes—the UPC and the
IPC—address licensing requirements, and there is no national system for
licensing of plumbers or plumbing inspectors. There also appears to be no
uniformity regarding the training and licensing of personnel who install,
maintain, and inspect backflow prevention devices. Yet there are numerous
organizations such as AWWA, New England Water Works Association,
American Society of Safety Engineers, American Backflow Prevention
Association, Backflow Prevention Institute, University of Southern California
Foundation for Cross-Connection Control and Hydraulic Research, and IAPMO
that offer personnelcertifications that address competency.
There also is a significant difference between the approach taken by stat
e
drinking water regulatory programs and water systems to ensure high w
a
t
e
rquality
within premises, particularly residential dwellings, versus utility-owned portions
of the distribution system. Plumbing codes (UPC and IPC) address
requirements for the installation of plumbing fixtures, appurtenances, and back-
flow prevention devices within premise plumbing where necessary such as
to prevent contamination of the public water system(UPC, 2003; IPC, 2003).
However, there are no provisions for ongoing inspections or surveillance to en-

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
sure that modifications to the premise plumbing by the homeowner will not ad-
versely affect the quality of the drinking water, either within the premise or
within the water distribution system. Plumbing codes (UPC and IPC) have also
never addressed ongoing water quality within the premise. Provisions for peri-
odic premise inspections to check for cross contamination, to ensure that the
integrity of the system is being maintained, and to assess premise water quality
could be required by local ordinances, but funding mechanisms would have to
be created (Chaney, 2005).
Finally, there is no incentive for homeowners to keep their premise plumb-
ing in compliance with codes. Houses are built to code but many fall out of
compliance due to age and as the code changes. In addition there are no organi-
zations that advise homeowners on how to maintain their plumbing systems such
as when flushing is necessary, water temperature recommendations, home
treatment devices, etc. (Chaney, 2005). A further discussion of issues associated
with premise plumbing and possible solutions can be found in Chapter 8.
VOLUNTARY AND NON-REGULATORY PROGRAMS THAT
INFLUENCE DISTRIBUTION SYSTEM INTEGRITY
Voluntary and non-regulatory programs exist that are designed to
provi
de public water systems with approaches for maintaining and improving
distribution system integrity. There are several objectives of these non-
regulatory water quality improvement programs for water supplies, foremost
among them being to further protect public health and to engage in risk
management efforts beyond what is provided by federal, state, and local
regulations and the enforcement system developed for primacy agencies. A
related motivation for a utility to implement such programs is to help organize
their many activities—i.e., to have a unifying umbrella that encompasses all of
the piecemeal requirements of the federal, state, and local regulations. A second
important objective of these programs is to increase customer satisfaction,
which is based largely on a perception of the quality of service and the cost
and quality of the delivered product. One common theme among these
programs is their intent to assist utilities in identifying best practices and
then affirm that the utility is employing these practices. Examples of best
practices include continuing or expanding monitoring of water quality and
setting up water quality goals, engagement in plant optimization projects,
studies on applicability of emerging technologies, and proactive preparation
for upcoming regulations—activities that, along with routine operation,
compliance monitoring, and maintenance, are often collectively described in a
utility’s distribution system management plan (if one exists). Voluntary and non-
regulatory programs can also help utilities to improve efficiency, as manifested
in responsiveness and cost. Performing services at a low cost is desirable but
customers and others require a high level of service. A balance must be
achieved to satisfy the expectations of regulators, customers, and owners at
a reasonable cost.

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
Voluntary programs are attractive because although public water s
y
s
t
e
m
s
recognize the need for health and environmental regulations to protect the pub-
lic, utilities (particularly larger ones) seek the flexibility to undertake activities
that will achieve these goals within the broader existing regulatory framework
while reducing the need for intensive regulatory oversight. Programs such as
voluntary accreditation are being designed that will allow water systems to im-
plement industry best practices that go beyond regulatory requirements to pro-
duce a drinking water quality that exceeds the minimum established by law.
Given the need to improve public confidence in drinking water quality, w
a
-ter
systems can use the recognition that they receive from implementing these
voluntary programs to promote these efforts to their customers. In particular,
water systems can communicate how they are achieving their water quality
goals along with an increased level of service without the need for a significant
increase in cost to their customers. Water systems are also able to demonstrate
that the product that they are providing not only exceeds regulatory require-
ments but competes equally with other sources such as bottled water, vended
water, and home treatment devices,at far less cost.
A few select voluntary, non-regulatory programs are described below, i
n-
cluding accreditation, Hazard Analysis and Critical Control Points (HACCP)
Plans, and Water Safety Plans, that can serve as guides to water utilities
that want to improve their distribution system management. Note that the
Partner- ship for Safe Water and QualServe, two voluntary AWWA programs
that target drinking water quality, are not discussed because distribution
systems are not their primary focus. QualServe uses self-assessment and peer-
review methods to identify opportunities for improvement in water and
wastewater utility services, while the Partnership for Safe Water focuses
on water treatment plant optimization.
Accreditation Standards
Currently, there is no nationwide system that accredits water u
t
i
l
i
t
i
e
s
.However,
a voluntary, nationwide accreditation program for all water utilities,
including small utilities, is currently under development by AWWA. The basis
of the program is to verify the application of standards and best practices that
will ensure the delivery of high quality services, exceeding regulatory compli-
ance. The program will be carried out by independent auditors who will verify
conformation with the accreditation standards on-site. The goals of the program
are not only to improve customer satisfaction, but also to provide a tool for regu-
latory agencies to use in evaluation of water utilities and to encourage utilities to
evolve beyond seeking compliance with existing regulations to seeking the best
strategies to protect public health.
The accreditation standards developed so far are water treatment plant o
p
-
eration and management (G100), distribution system operation and management
(G200), and source water management and protection (G300). (After piloting

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
the implementation of these standards at both large and small systems in August
2005, other areas of accreditation standards will be developed.) The Distribu-
tion System Operation and Management Standards (G200) (AWWA/ANSI,
2004), published in May 2004, are intended to improve distribution systems’
water quality and utility’s management efficiency by voluntarily adhering to
standards that exceed current regulatory requirements and by performing inde-
pendent audits to verify performance. The standards call for development of
water quality sampling plans at prescribed sites in distribution systems. Nitrifi-
cation control; booster chlorination; internal corrosion monitoring and
control; reduction of the formation of DBPs; and color, taste, and odor
monitoring and control are defined as programs that should have individual
goals and action plans established specifically for each utility. Distribution
system management activities listed in the standard include system pressure
monitoring, backflow prevention, permeation prevention, water loss
minimization, valve exercising and replacement, fire hydrant maintenance and
testing, maintenance of coatings and linings, water use metering, external
corrosion control, water quality monitoring, and energy management. The
verification step of the standard includes providing certain required documents
and records. For those utilities that decide to develop a distribution system
management plan that meets the AWWA G200 standard, conformance would
be verified on a periodic basis. Because G200 provides a comprehensive
framework in which a water utility can manage distribution system integrity
and it targets those activities felt by the committee to be of highest priority in
reducing public health risks, it is further discussed in Chapter 7.
Hazard Analysis and Critical Control Points
Voluntary programs that deal with water quality and management i
ssue
s
fromthe perspective of risk evaluation and reduction are being adapted to drink-
ing water treatment, operations, and distribution from other branches of the in-
dustry. An example is the Hazard Analysis and Critical Control Points
(HACCP) program, which was developed by NASA in the 1960s for the U.S.
space program, later transferred to food safety, and recently formatted for drink-
ing water quality. The program relies on three steps, which are addressed con-
tinuously in a cycle: hazard identification, remediation, and verification.
HACCP for the drinking water industry is based around the same seven princi-
ples as were developed for NASA and other industries (NASA, 1991;
Codex Alimentarius Commission, 1993, 1997; Mucklow, 1997). The HACCP
principles are to:
 Identify hazards and controlmeasures
 Identify critical controlpoints
 Establish critical limits
 Identify monitoring procedures

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
 Establish corrective action procedures
 Verify and validate the HACCP Plan
 Establish record keeping andresponsibility
HACCP is a risk management program because utilities use it to first i
den-
tify and evaluate hazards/risks, and then to establish control systems to minimize
the occurrence and effects of incidents that may impact the safety and quality of
the water. A water utility can choose to apply HACCP to any one “process”—
i.e., watershed protection, treatment, or the distribution system. Some utilities
may already have good watershed protection programs and good control over
treatment facilities, and so may view the distribution system as a priority. How-
ever, because HACCP is a proactive approach to system management that helps
the utility to identify “hazards” further upstream, it works quite well as a com-
prehensive system plan, from source to tap. For maximum benefits, it is impor-
tant to leave the decision to individual utilities and not be too prescriptive about
how to apply HACCP (Friedman et al., 2005).
A recently completed project sponsored by the AWWA Research F
o
u
n
d
a
-tion
(Friedman et al., 2005) describes HACCP pilot studies conducted with three
utilities’ distribution systems—Greater Cincinnati Water Works, Cincinnati,
Ohio; Calgary Water Works, Calgary, Alberta; and the City of Everett, Everett,
Washington. Training workshops were held at each utility location to explain
HACCP terminology and to initiate development of the utility’s HACCP plan.
Each participating utility formed a HACCP team to further develop the HACCP
plan and to guide its implementation. The goal was for each utility to imple-
ment their HACCP plan over a 12-month period during which certain opera-
tional and water quality parameters would be monitored. The participating utili-
ties found that the implementation of HACCP to water supply distribution was
feasible and practical, but that the time and resource requirements were greater
than originally anticipated. The development of the HACCP plan was useful in
honing in on the most important risks and process controls for water quality
management. Within the 12-month pilot study period, none of the three partici-
pating utilities developed a fully implemented HACCP program for certification.
A longer period of time and/or a greater resource commitment was likely to be
required before the HACCP systems would be considered fully
implemented, complete, and certifiable. Box 2-2 describes two other HACCP
case studies in detail, for Austin,Texas, and Burwick, Maine.
NSF International provides HACCP certification to water utilities in t
he
United States through its HACCP-9000 registration program. The program con-
sists of third-party verification of utility HACCP plans, combined with a
registration with ISO 9000 standards. However, adoption of the HACCP ap-
proach need not be tied formally to such administrative programs. HACCP
could be an integral part of a utility’s distribution system management plan, ei-
ther in addition to or in lieu of G200 (given the substantial similarities between
the two programs). In particular, HACCP is useful for improving a
utility’s awareness of its existing databases and how it can better manage the
information

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
contained within, and for promoting record keeping and reporting. Critics con-
tend that HACCP is little more than properly operating a distribution system.
Indeed, there may be little value added in the United States where utilities are
relatively heavily regulated compared to other countries where HACCP has been
successfully adopted (such as Australia, which has no national water quality
standards). However, advocates contend that the part of HACCP that most utili-
ties do not already engage in is checking to verify that actions are working (Mar-
tel, 2005). Furthermore, HACCP puts an increased focus on operator training,
which can be ignored in the face of so many other competing activities,
like compliance monitoring. The program is more likely to be adopted by
larger-size utilities because of the need for a larger staff and budget to carry out
HACCP.
Nonetheless, there is another practical consideration that makes G200 a
more attractive organizing program for distribution systems than HACCP. Pro -
grams like HACCP are ideally suited to industries that experience little variation
on a day-to-day basis (such as food and beverage processing plants) and are not
as easily adapted to the dynamic nature of drinking water distribution systems
that may experience changes in water quality depending on season, source of
supply, and changing daily demands. Furthermore, unplanned disruptions such
as water main breaks require immediate responses in areas that may not be con-
sidered critical control points, making it very difficult to proactively control
contamination events. Finally, the vast number of locations within a distribution
systemthat could be potential critical control points (presumably every resi-
dence where a cross connection exists) argues against the formal adoption of
HACCP.
The cost of creating a HACCP plan for a community of 10,000 may be i
n
the range of $10,000, including a day- or two-day-longworkshop.
Water Safety Plans
In 1994, the World Health Organization (WHO) adapted the HACCP pro-
gram through Water Safety Plans, which can be prepared for individual water
systems. The WHO’s Guidelines for Drinking Water Quality (2004) describe an
approach to follow in preparing Water Safety Plans. The approach is to identify,
prioritize, and prevent risks arising from hazards associated with distribution of
drinking water. The three critical components of a water safety plan are:
 System assessment regarding both the quantity and quality of suppli
ed
water
 Identification of controlmeasures
 Management plans describing actions during both normal and ex
tr
em
e
conditions and documenting, monitoring, communication, and improvement
efforts.

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
BOX 2-2
HACCP Case Studies
There are few case studies of where HACCP has been applied to d
i
s
t
r
i
b
u
t
i
o
n system
management. One involves a relatively small utility, the South Berw ick Water District, in
South Berw ick, Maine, w hich serves about 4,000 people. At this utility, a HACCP training
workshop was held on June 2003 to assemble the HACCP team, w hich included the
superintendent, foreman, and a service person, as well as outside experts such as an
engineer familiar w ith the South Berwick system, a microbiologist from EPA, a state
regulator who was an expert on cross-connection control, and a risk manager from the
bottled water industry. As in other cases where HACCP has been applied, assembling a
team that has as many people from different cross sections of the water utility as possible is
one of the benefits of doing HACCP, but because of the small size of the utility this
required outside assistance. The process flow diagram for the entire water system is
show n in Figure 2-1.
Agamenticus Wells
Agamenticus
Pumping
Station
NaOCl
Distribution
System
NaOCl
Blackmore W ells
and
Pumping Station
UV
NaOCl
W illow Drive
W ells and
Pumping Station
FIGURE 2-1 Process Flow Diagram for the South Berw ickWater D
i
s
t
r
i
c
t
.
SOURCE: Re- printed,
w ith permission, by Martel (2005).© 2006 by AwwaRF.
Three priority hazards were identified by the HACCP team, tw o of w
hichinvolve the
distribution system: (1) backflow through unprotected cross connections, (2) long dead-end
mains w ith zero or poor disinfectant residual, and (3) unintentional contamination of
shallow wellpoints at the Agamenticus Wellfield. It should be noted that it w as verydifficult
to gather enough information to determine the frequencyof occurrence or the severity of
these hazards, given the utility’s lack of data. For this reason, South Berw ick’s initial
HACCP plan focused on monitoring activities to further characterize these hazardsand
improve existing controlmeasures. Unfortunately, the HACCP plan w as not fully imple-
mented because of a lack of manpow er and because of other priorities. With only three
full-time employees at the utility, daily systemoperation and maintenance took priority over
HACCP plan implementation. Furthermore, the utility personnelw ere involved with
building a new treatment facility, developing a new rate structure, and addressing local
and state political issues. This case study illustrates the need for sufficient manpower to
successfully implement a HACCP Plan.
A second case study is from Austin, Texas, a much larger water s
u
p
p
l
y
that serves
approximately 770,000 people. The interdisciplinary HACCP team consisted primarily of in-
house staff: the w ater quality manager, the w ater laboratory supervisor, an engineer
continues
Concrete
cistern
13,00 0 gal
Grav el well
Bedrockwell
Well points
Bedrock
# 1
Bedrock
# 2
Bedrock
# 3
Bed-
rock
Well #
1
Junction
Road
Bedrock W ell Bed-
rock
1 MG
Reservoir

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
/planner, a construction inspector, the cross-connection control s
u
p
e
r
v
i
s
o
r
,
the Assistant Director of
Treatment, the Infrastructure Superintendent, and a state regulator. A HACCP pilot study
was conducted from May 2003 to September 2004. The team focused on one pressure
zone within the distribution system for the HACCP pilot study (see the flow process
diagram below in Figure 2-2):
FIGURE 2-2 Flow Process Diagram for the Austin Water Supply. S
O
U
R
C
E
:Reprinted, w ith permission,
fromMartel (2005).© 2006 by AwwaRF.
Austin’s HACCP team identified two high priority hazards: b
a
c
k
f
l
o
w
through unprotected
cross connections (focusing specifically on irrigation and hydrant vandalism) and
contamination from new construction sites (primarily via improper valve turning). Austin
found that HACCP is more complex than initially envisioned. Originally, the utility thought
that HACCP w ould involve identifying critical flow paths within the distribution system and
monitoring these flow paths more intensively to assure water quality to downstream sites.
Instead, by nature of the selected hazards, the measures used to control these hazards
focused on operations and maintenance activities rather than water quality monitoring.
This approach added layers of complexity to the existing monitoring program. On a posi-
tive note, the HACCP approach helped the utility (1) improve understanding of their distri-
bution system hazards; (2) heighten employee aw areness of pressure zone
boundaries, pressure transients, the need to maintain pressure and to respond quickly to
main breaks in small pressure zones; (3) improve awareness of existing databases and
monitoring programs; (4) improve data management skills; (5) identify needed
improvements to existing databases; and (6) improve reporting procedures for acceptance
of new mains.
SOURCE: Martel et al. (2006).

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
Water safety plans present an affordable risk management tool for all drinking
water suppliers, regardless of size. While some critical elements of the pl
a
n
should be assured by all systems, more costly or time -consuming elements,
characterized as not critical, may be added to the plans based on budgetary and
staff availability. The most critical elements of the water safety plan documents
include system description, water flow diagrams, hazard identification, identifi-
cation of a team, and contingency plan. Additional items include specification of
chemicals and materials, job descriptions for staff responsible for individual
operations, corrective actions for deviations, record-keeping procedures, valida-
tion data, and incident documentation procedures. Finally, optional elements
may include manuals for hygiene, preventive maintenance, and equipment cali-
bration; job descriptions for all staff; training programs and records; documenta-
tion of corrective actions, audits, and verification procedures; and consumer
complaint policy and procedures.
Clearly, the elements of a Water Safety Plan closely resemble the elements
of a HACCP Plan: (1) source-to-tap systemassessment; (2) control measures for
identified hazards and operational monitoring of control measures; and (3) a
management plan that documents the system assessment, control measures,
monitoring plan, corrective action procedures to address water quality incidents,
communication plan, and supporting programs such as standard operating pro-
cedures, employee training, and risk communication. Both HACCP and Water
Safety Plans should be used continuously.
A 2004 conference sponsored by NSF International examined a variety o
f
risk management approaches, including HACCP, ISO certification, Water
Safety Plans, and Environmental Management Systems. Not only were
many commonalities among these programs evident, the distinctions between
them were unclear. The conference presented a number of domestic and
international case studies where water utilities had utilized one of these risk
management systems, but no case studies targeting the distribution system
were discussed. In- deed, the choice of the “right” program for any given water
utility may present a challenge, specifically because there is no precedence for
using these programs for distribution system management, but also because of a
lack of coordination between the programs, a lack of tangible benefits beyond
what a utility already accomplishes, and inefficient communication to the public
about the programs. It is up to an ambitious utility manager and staff to learn
about the programs, evaluate their applicability, and select one.
Training for Operators, Inspectors, and Related Personnel
While utilities endeavor to optimize their infrastructure and operate the di
s-
tribution system to minimize degradation, an integral component not to be i
g
-
nored are the operators, inspectors, and related personnel charged with running
and monitoring the system. Inevitably, the operators and field personnel serve
as guardians to minimize degradation in the distribution systemand ensure wa-

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
ter quality is maintained for the consumer.
Training of distribution system operators was identified as a high pri
ori
tyissue
for reducing risk in drinking water distribution systems (NRC, 2005). The need
for the continuing and intensive training of operators of distribution sys -
tems has increased recently for three reasons. First, as federal and state regula-
tions become increasingly stringent and more complex, they require enhanced
skills for proper sample collection and preservation, as well as better under-
standing of aquatic chemistry and biology for proper implementation and inter-
pretation of results. Second, in many systems the D/DBPR (EPA, 1998a) cre -
ated a shift in the use of disinfectants in the distribution systems from a rela -
tively simple application of chlorine to the rather complicated application
and maintenance of chloramine. Finally, with an increase in the importance of
security of drinking water pipes, pumps, reservoirs, and hydrants, there is a
corresponding increase in the responsibility of operators to make decisions
during perceived securityevents.
Typically distribution system operators, mechanics, and field crews are w
e
l
l
trained in the mechanical aspects of water delivery (such as pipe repl
acement
and repair; pump, valve, and storage facility operation; etc.) and safety. In cases
where contractors are used to repair or maintain the infrastructure (for example,
many utilities allow certified plumbers to perform the tasks related to backflow
prevention and cross-connection control), diligence of construction inspectors in
providing oversight is of paramount importance because the contractor may or
may not be following standard practices. A case in point regarding the impor-
tance of training plumbers is the ban on lead solder implemented in the late
1980s. Because the responsibility for high lead levels in drinking water falls on
the utility, many utilities were actively engaged in training plumbers about the
dangers of lead from the use of lead solder and about the new requirements of
the LCR. This training was critical to reducing the risk of lead exposure from
drinking water.
The importance of operator training in protecting public health from c
o
n
-
taminated drinking water cannot be overstated. A recent critique of the Walker-
ton, Ontario Inquiry Report (Hrudey and Walker, 2005) claims that lives could
have been saved had operators been properly trained. Failure to perform basic
monitoring duties and understand the vulnerability of the systemto a contamina-
tion event in May 2000 led to more than 2,300 cases of waterborne disease in a
system of only 5,000 people. “Water system operators must be able to recog-
nize that the threats to their system contrasted with the system’s capability to
cope. They have a professional responsibility to ensure deficiencies are identi-
fied, made known to management, and effectively remedied. Pending necessary
improvements, operators must increase their vigilance and develop contingency
plans to cope with periods of stress. Contingency plans should be practiced us -
ing simulated incidents before a real crisis develops” states Hrudey. Justice
O’Connor who led the multi-million dollar inquiry into the Walkerton
tragedy concluded that “Ultimately, the safety of drinking water is protected by
effective

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
management systems and operating practices, run by skilled and well-trainedstaff”
(Hrudey and Walker, 2005).
Operator training classes and seminars are offered through industry associa-
tions (e.g., AWWA, the National Rural Water Association) and third party con-
tractors. The International Association for Continuing Education and Training
(IACET) has recently developed certification for trainers, which is a positive
step toward ensuring the quality of instructors who are providing operator train-
ing. However, it is well recognized that nationally there is a paucity of adequate
training facilities, instructors, and apprentice programs to replace an experienced
workforce who will be retiring in the coming decade (Brun, 2006; Eaton, 2006;
McCain and Fahrenbruch, 2006; Pomerance and Means, 2006).
As discussed earlier, there are existing EPA guidelines for the certification of
treatment plant operators and distribution system operators (EPA, 19
99
),which
have subsequently been implemented by states (leading to state require-
ments for certification). However, these requirements are not always enforced,
particularly on small systems. Stronger enforcement of the distribution system
operator certification requirements developed by individual states could be a
mechanism to support training and apprentice programs. Also, future regula-
tions need to include mechanisms to fund training and apprentice programs spe-
cifically for distribution system operators. Finally, while existing certification
exams test generic knowledge, future requirements should ensure that operators
understand the system in which they work and are familiar with portions of op-
erating plans that apply to performance of their daily activities.
CONCLUSIONS AND RECOMMENDATIONS
The Total Coliform Rule, the Surface Water Treatment Rule, the Disinfec-
tants/ Disinfection By-Products Rule, and the Lead and Copper Rule are the
federal regulations that address water quality within the distribution system, and
they do so in a piecemeal fashion. These rules were not intended to address
distribution system integrity as defined in Chapter 1, which consists of physical,
hydraulic, and water quality integrity. For example, the TCR considers only that
microbial contamination indicated by fecal parameters. Nor does the SDWA
contemplate federal actions that would address premise plumbing, with the ex-
ception of lead in plumbing materials. As a result a more comprehensive ap-
proach needs to be taken to ensure that the overall integrity of distribution sys -
tems is maintained. The following regulatory recommendations are made.
EPA should work closely with representatives from states, water sys-
tems, and local jurisdictions to establish the elements that constitute an ac -
ceptable cross-connection control program. Although states, either
through drinking water regulations or state plumbing codes, have cross -
connection control requirements in place, these requirements are inconsistent
amongst states. State oversight of cross-connection control programs varies
and is subject to

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
availability of resources. If states expect to maintain primacy over their drink-
ing water programs, they should adopt a cross-connection control program that
includes a process for hazard assessment, the selection of appropriate backflow
devices, certification and training of backflow device installers, and certification
and training of backflow device inspectors. Although tracking compliance by
water systems is also an important element, the resource implications of tracking
and reporting requirements should be carefully considered. EPA may need to
allow use of federal funds for training of backflow prevention device inspectors
for small water systems.
Existing plumbing codes should be consolidated into one uniform na-
tional code. Although similar with regard to cross-connection control
requirements and other premise plumbing protection measures, the two
principal plumbing codes that are used nationally, the UPC and the IPC,
have different contents and permit different materials and devices. These
differences appear to be addressable, recognizing that the two code developing
organizations may have other issues that would need to be resolved. In addition
to integrating the codes, efforts should be made to ensure more uniform
implementation of the plumbing codes. Their implementation can vary
significantly between jurisdictions, which can have major impacts on the
degree of public health protection afforded to their constituents.
For utilities that desire to operate beyond regulatory requirements,
adoption of G200 or an equivalent program is recommended to help utili-
ties develop distribution system management plans. G200 has advantages
over other voluntary programs, such as HACCP, in that it is more easily adaptedto
the dynamic nature of drinking water distribution systems.
More attention should be paid to having adequate facilities, instructors,
and apprentice programs to train utility operators, inspectors, foremen,
and managers. The need for the continuing and intensive training of operators of
distribution systems has increased as a result of more sophisticated federa
l
and state regulations, the shift in the use of disinfectants in the distribution sys -
tem, and the increase in importance of security of drinking water
distribution systems. Recent development of IACET certification for trainers is
a positive step toward the quality of instructors providing operator training.
Future regulations need to include mechanisms to fund training and apprentice
programs.
REFERENCES
American Backflow Prevention Association (ABPA). 1999. American Backflow P
r
e
-
vention Association State Program Survey. Available on-line at: http://www.
abpa.org/originalsite/ABPA_Survey_Report.pdf. Accessed May 4, 2006.
AWWA/ANSI. 2004. G-200 Distribution Systems Operation and Management. D
e
n
-ver,
CO: AWWA.

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
Association of State Drinking Water Administrators (ASDWA). 1999. Survey of S
ta
t
eCross-
Connection Control Programs. September 29, 1999. Washington, DC: A
S
-DWA.
ASDWA. 2003. Summary of Results from the ASDWA Distribution System & T
C
RSurvey,
Design and Construct & Operation and Maintenance. Washington, D
C
:ASDWA.
ASDWA. 2004. Survey of State Adoption of ANSI/NSF Standards 60 and 61. W
a
s
h
-ington,
DC: ASDWA.
45.
Brun, P. 2006. Is it workforce planning or succession planning? Source 20:6.
Chaney, R. 2005. The Uniform Plumbing Code: development, maintenance and a
d
m
i
n
i
-stration
as a pathway to reducing risk. April 18, 2005. Presented to the NRC Com- mittee on
Public Water Supply Distribution Systems. Washington,DC.
Codex Alimentarius Commission. 1993. Guidelines for the Application of the H
a
z
a
r
dAnalysis
Critical Control Point (HACCP) System, CAC/GL 18-1993. Rome, Italy: Codex
Alimentarius Commission and the FAO/WHO Food Standards Program, Food and
Agriculture Organization of the United Nations and World Health Organi- zation.
Codex Alimentarius Commission. 1997. Guidelines for the Application of the H
A
C
C
P
System.
Rome, Italy: Codex Alimentarius Commission and the FAO/WHO F
o
o
d Standards
Program, Food and Agriculture Organization of the United Nations and World
Health Organization.
Craun, G. F., and R. L. Calderon. 2001. Waterborne disease outbreaks caused by distri-
bution systemdeficiencies. J. Amer. Water Works Assoc. 93(9):64–75.
Eaton, G. 2006. San Diego County Water Authority prepares for the future. S
o
u
r
c
e 20:14–
15.
Environmental Protection Agency (EPA). 1979. National Interim Primary Drinking
Water Regulations for the Control of Trihalomethanes in Drinking Water, Final
Rule. Federal Register 44:68641.
EPA. 1989. National Primary Drinking Water Regulations: Filtration, Disinfection,
Turbidity, Giardia lamblia, Viruses, Legionella, and Heterotrophic Bacteria; Final
Rule (SWTR). Federal Register 54:27486.
EPA. 1991. National Primary Drinking Water Regulation: Lead and Copper Rule, F
i
n
a
lRule.
Federal Register 56:26460.
EPA. 1996. National Primary Drinking Water Regulations: Monitoring
Requirements for Public Drinking Water Supplies; Final Rule. Federal Register
61:24353.
EPA. 1998a. National Primary Drinking Water Regulations: Disinfectants and Disinfec-
tion Byproducts, FinalRule. Federal Register 63:69389.
EPA. 1998b. National Primary Drinking Water Regulations; Interim Enhanced S
u
r
f
a
c
e
Water
Treatment Rule; Final Rule. Federal Register 63:69477.
EPA. 1999. Final guidelines for the certification and recertification of the operators
o
f
community and nontransient noncommunity public water systems. Federal Register
64:5915–5921.
EPA. 2002a. National Primary Drinking Water Regulations; Long Term 1 E
n
h
a
n
c
e
d
Surface
Water Treatment Rule, Final Rule. Federal Register 67:1811.

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
EPA. 2002b. Potential contamination due to cross-connections and backflow and t
h
e
associated health risks, an issue paper. Washington, DC: EPA Office of Ground
Water and Drinking Water.
EPA. 2003a. Drinking Water Research Program, Multi-Year Plan, 2003. W
a
s
h
in
g
to
n
,DC: EPA
Office of Research and Development. Available on-line at:
http://www.epa.gov/osp/myp/dw.pdf. Accessed May 4, 2006.
EPA. 2003b. Cross-Connection Control Manual. Washington, DC: EPA.
EPA. 2005a. FACTOIDS: Drinking Water and Ground Water Statistics for 2004.
Washington, DC: EPA.
EPA. 2005b. Unregulated Contaminant Monitoring Regulation (UCMR) for P
u
b
l
i
c
Water
Systems Revisions. Federal Register 70:49093.
EPA. 2006a. National Primary Drinking Water Regulations: Stage 2 Disinfectants a
n
d
Disinfection Byproducts Rule; National Primary and Secondary Drinking W
a
t
e
r
Regulations, Final Rule. Federal Register 71:387.
EPA. 2006b. National Primary Drinking Water Regulations: Long Term 2 E
n
h
a
n
c
e
d
Surface
Water Treatment Rule, Final Rule. Federal Register 71:653.
Friedman, M., G. Kirmeyer, G. Pierson, S. Harrison, K. Martel, A. Sandvig, and A. H
a
n
-son.
2005. Development of distribution system water quality optimization pl
a
n
s
.Denver,
CO: AwwaRF.
Hrudey, S. E., and R. Walker. 2005. Walkerton—5 years later tragedy could have b
e
e
n
prevented. OpFlow 31:1–7.
International Association of Plumbing and Mechanical Officials (IAPMO). 2003. U
n
i
-form
Plumbing Code, 2003 edition. Ontario, CA: IAPMO.
IAPMO. 2005. Can we make it work? Available on-line at http://www.iapmo.org
/iapmo/news/code-release.html. Accessed April26, 2006.
International Code Council. 2003. International Plumbing Code, 2003 Edition. F
a
l
l
s
Church, VA: International Code Council.
Linn County. 2004. Linn County Plumbing Regulations. Available on-line at
http://www.linncountyauditor.org/Ordinances/Plumbing--%5B5%5D.pdf. Accessed
April26, 2006.
MacPhee, M. J. (ed.). 2005. Distribution system water quality challenges in the 2
1
s
t
century:a strategic guide. Denver, CO: AWWA.
Martel, K. 2005. HACCP Applied to Distribution Systems. January 13, 2005. Pre-
sented to the NRC Committee on Public Water Supply Distribution Systems. W
a
s
h
-
ington, DC.
Martel, K., G. Kirmeyer, A. Hanson, M. Stevens, J. Mullenger, and D. Deere. 2
0
0
6
.
Application of HACCP for Distribution SystemProtection. Denver, CO: A
w
w
a
R
F
.
McCain, K., and M. Fahrenbruch. 2006. Succession planning: thebabies and boomers.
Source 20:16–17.
Mucklow, R. 1997. Where did HACCP come from? In: Heads Up for HACCP. N
a
-tional
Meat Association. Available on-line at http://www.nmaonline.org/files/
headsup12-1.htm. Accessed May 4, 2006.
National Aeronautics and Space Administration (NASA). 1991. A dividend in f
o
o
d safety.
Spinoff 1991. NASA Technical Report ID 20020086314. Washington, D
C
:
NASA.
National Research Council (NRC). 2005. Public Water Supply Distribution
Systems: Assessing and Reducing Risks, First Report. Washington, DC: National
Academies Press.
Pomerance, H., and E. G. Means. 2006. Succession planning: leveraging theinevitable.
Source 20:10–13.

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
State of Iowa. 2005. State Plumbing Code. Available on-line at http://www.legis.state
.ia.us/Rules/2002/iac/641iac/64125/64125.pdf. Accessed April26, 2006.
World Health Organization (WHO). 2004. Guidelines for drinking water quality, t
h
i
r
d
edition. Available on-line at http://www.who.int/water_sanitation_health/dwq/
gdwq3/en/. Accessed April 26, 2006. Geneva, Switzerland: WHO.

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
3
Public Health Risk from DistributionSystem
Contamination
One of the most challenging facets of reducing the risk of contaminated d
i
s
-
tribution systems is being able to quantify the existing risk. This is made c
om
-
plicated not only by the plethora of factors that can constitute public health r
i
s
k
s
,
including a diversity of microbial pathogens and chemical compounds, but al
so
by the varying response that a given individual will have when exposed to those
factors. This chapter describes three primary mechanisms used to assess the
acute public health risk of distribution system contamination, the limitations of
these methods, and what conclusions can be derived from currently available
data.
INTRODUCTION TO RISK
The process of risk assessment involves determining the likelihood and se
-
verity of different adverse impacts given exposure of a population to a h
a
z
a
r
d
.Risk
analysis includes the process of risk assessment, as well as risk management
activities to decide what an acceptable risk level is and to take actions to reduce
risk (NRC, 1983). Risk assessment requires the activities of hazard iden-
tification, exposure assessment, and dose-response (or exposure-response) as-
sessment. Hazard identification is the determination of what adverse agents
might be present and what adverse impacts they might cause. Exposure assess -
ment is the quantitative determination of the levels of contaminants (in the case
of environmental exposures) individuals may consume/inhale/contact over a
specific time period. Dose-response assessment is the quantitative determina-
tion of the likelihood of an individual having a particular adverse effect from a
given exposure. Alternatively, this can be viewed as the proportion of persons
in a population who are expected to have the adverse effect were they to have
the particular exposure.
Various federal agencies, including the U. S. Environmental Protecti
on
Agency (EPA), have developed specific guidelines and procedures for perform-
ing risk assessment, particularly for carcinogens and for substances that result in
non-carcinogenic toxic effects. In the case of infectious agents (which are fre-
quently the concern in drinking water), methodologies are at a developmental
stage.
One of the goals of performing risk assessment within a regulatory frame-
work is to develop regulatory guidance or standards (or decide not to undertake
such action) based on the results. This process,which is part of risk analysis,
87

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
requires additional considerations such as cost and equity. Under the S
a
f
e
Drinking
Water Act, EPA is required to set a maximum contaminant level goal(MCLG)
for certain contaminants that is absolutely protective against all adverse
health effects, given available risk assessment information. For most con-
taminants with MCLGs, a regulatory level is then established—a maximum con-
taminant level (MCL)—or a treatment technique is required, both of which in -
corporate considerations of feasibility (see Box 3-1).
In determining a regulatory level such as an MCL, implicitly or
explicitly the acceptable residual risk (after the implementation of any
interventions) must be decided upon. The empirical evidence is that, for human
carcinogens, EPA has regarded a window of residual lifetime risk of 1/1,000,000
to 1/10,000 to be acceptable (see Box 3-2 for an explanation of the origins of
this value and its extension to infectious agents). In other words, a residual
risk resulting in no more than 1 extra cancer in the lifetime of a population of
10,000 to 1,000,000 persons is regarded as being acceptable.
Risks from Drinking Water
Drinking water can serve as a transmission vehicle for a variety of hazard-
ous agents: enteric microbial pathogens from human or animal fecal contamina-
tion (e.g., noroviruses, E. coli O157:H7, Cryptosporidium), aquatic microorgan-
isms that can cause harmful infections in humans (e.g., nontuberculous myco-
bacteria, Legionella), toxins from aquatic microorganisms (such as cyanobacte-
ria), and several classes of chemical contaminants (organic chemicals such as
benzene, polychlorinated biphenyls, and various pesticides; inorganic chemicals
such as arsenic and nitrates; metals such as lead and copper; disinfection by-
products or DBPs such as trihalomethanes; and radioactive compounds).
Contaminants in drinking water can produce adverse effects in humans due
to multiple routes of exposure. In addition to risk from ingestion, exposure can
also occur from inhalation and dermal routes. For example, inhalation of drop-
lets containing respiratory pathogens (such as Legionella or Mycobacterium) can
result in illness. It is known that DBPs present in drinking water may volatilize
resulting in inhalation risk, and these compounds (and likely other organics)
may also be transported through the skin (after bathing or showering) into the
bloodstream (Jo et al., 1990). Reaction of disinfectants in potable water with
other materials in the household may also result in indoor air exposure of con-
taminants; for example Shepard et al. (1996) reported on release of volatile or-
ganics in indoor washing machines. Thus, multiple routes of exposure need to be
considered when assessing the risk presented by contaminated distribution
systems. It should be noted, however, that the report will not consider such indi-
rect routes of exposure as (1) the loss of pressure and subsequent inadequate fire
protection, (2) loss of water for hospitals and dialysis centers, and (3) leaks in
household plumbing that lead to toxic mold growth.

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
PUBLIC HEALTH RISK FROM DISTRIBUTIONSYSTEM CONTAMINATION 89
It has been recognized for some years that consumers face risk from multi-
ple hazards, and that action to reduce the risk from one hazard may increase the
risk fromother hazards given the same exposure. There are prominent examples
of this phenomenon in the drinking water arena that have greatly complicated
efforts to reduce overall risk from distribution systems. Havelaar et al. (2000)
assessed the relative changes in risk fromswitching to ozone treatment of drink-
ing water in the Netherlands. In this case, there was a projected reduction in risk
from waterborne infectious disease (such as Cryptosporidium) while there was a
projected increase in risk from DBP formation (the primary one examined was
bromate). To compare the net change in overall risk, it is necessary to place the
multiple risks (with their different endpoints in terms of disease severity) on the
same scale. Havelaar et al. (2000) did this comparison using the methodology of
disability adjusted life years (DALY’s). In this approach, the severity of an
adverse health effect is quantitatively weighted by an index (disability weight)
reflecting the proportional degradation in health (a weight of 0 is reflective of
absence of an effect, while a weight of 1 is reflected in total impairment); the
integral of the years of diminished functioning multiplied by the disability
weight is summed with the reduction in lifespan due to premature mortality
to get the aggregate impact to a population. In principle, using such an approach
one can optimize for the overall net reduction in risk, considering competing
hazards. It is noted that the DALY framework has not been adopted for U.S.
regulatory practice and remains controversial for a number of technical and pol-
icy reasons (including age equity) (Anand and Hanson, 1997).
When risk is assessed for chemical or microbial exposure, it should be con-
sidered that not all segments of the population are at the same degree of risk.
This may be due to differences in exposure in terms of either consumption
(Gerba et al., 1996) or in concentrations (due to heterogeneity in the environ-
1
Paragraph (5) allows departure upwards from setti ng the MCL as cl ose to the MCLG as f
e
a
s
i
b
l
eif doing so would result i n an
increase in risk from other contami nants, or would i nterfere with the perfor mance of processes used to address other
contaminants. Paragraph (6) allows departure upward from the “as close as feasi ble” criterion in certain circumstances if
the benefits would not justify the cost of compliance at that standard.
BOX 3-1
U.S. Code, Title 42(6A)(XIIB)§300g-1
(Safe DrinkingWater Act as Amended)
(A) Maximum contaminant level goals. Each maximum contaminant level goal e
s
t
a
blishedunder
this subs ection shall be set at the level at which no know n or anticipated adverse effects
on the health of persons occur and which allowsan adequate margin of safety.
(B) Maximum contaminant levels. Except as provided in paragraphs (5) and (6)1
, each
national primary drinking w ater regulation for a contaminant for which a maximum contami-
nant level goal is established under this subsection shall specify a maximum contaminant
level for such contaminant w hich is as close to the maximum contaminant level goal as is
feasible.

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
ment, e.g., in the distributed water), or to intrinsic differences in susceptibility
(Balbus et al., 2000). Unfortunately, our ability to assess quantitative differ-
ences in intrinsic susceptibility remains poor, and therefore protection of suscep-
tible subpopulations often relies upon the imposition of safety factors.
Methods for Characterizing Human Health Risk
Characterization of human health risks may be performed using an epide-
BOX 3-2
Origin of the 1/10,000 Acceptable Risk Level for Carcinogens and Infectious Agents
EPA has been at the forefront of the issue of acceptable risk in v
i
r
t
u
a
l
l
yall of its pro-
grammatic areas, primarily as the result of court challenges to its regulations. In response
to the 1987 Section 112 Clean Air Act decision (Natural Resources Defense Council vs.
U.S. Environmental Protection Agency 824 F. 2nd 1146 [1987]), EPA decided it w ould b
a
s
e
its
regulatory decisions on quantitative risk assessments using the general policy that a
lifetime added cancer risk for the most exposed person of 1 in 10,000 (1 × 10-4
) might con-
stitute acceptable risk and that the margin of safety required by statute and reinforced by
the court should reduce the riskfor the greatest number of persons to an added lifetime risk
of no more than 1 in 1 million (1 × 10-6
). How ever, EPA (along w ith the courts) has
not viewed “safe” as the equivalent of risk-free and has determined that standards should
protect against significant public health risks (EPA 49 Fed. Reg. 8386 [1984]; Rodricks et
al. 1987; Industrial Union Department, AFL-CIO v. American Petroleum Institute et al.
448
U.S. 607 [1980]). EPA has repeatedly rejected the opinion that it can e
s
t
a
b
l
i
s
ha universal (i.e.,
brightline) acceptable risk that should never be exceeded under any circumstances, and
they maintain that guidance provided under one statute might have little relevance to others
because of differing program goals. In practical terms, EPA almost never regulates at a
theoretical risk below 1 × 10-6
(de minimis) and almost alw ays regulates at a theoretical risk
below 1 × 10-4
(de manifestis)” (NRC, 2004).
Policy w ith respect to acceptable levels of risk from exposure to infectious agents is
less well developed than for chemical carcinogens. How ever, in framing the Surface Water
Treatment Rule (Federal Register, June 29, 1989, page 27486), the rule for reduction of
risk from Giardia and viruses was set to achieve a residual estimated risk of infection below
1/10,000 per year. This number derived from the then average waterborne illness rate
associated w ith reported waterborne outbreaks (Regli et al., 1991). How ever it is now rec-
ognized that the waterborne illness rate is substantially greater than this value—due to
underreporting of outbreaks, as well as to substantial endemic illness. The use of infection
rather than illness as an endpoint w as intended to compensate for secondary cases and
also for presumed heightened infectivity amongst sensitive subpopulations.
The use by EPA of an acceptable risk window for microorganisms in the 10-6
to 10-4
range as one factor in setting standards continues. Asrecently as the promulgation of the
Long Term 2 Enhanced Surface Water Treatment Rule (Federal R
e
g
i
s
t
e
r
, January 5, 2006),
EPA has stated: “EPA and Advisory Committee deliberations focused on mean source
water Cryptosporidium concentrations in the range of 0.01–0.1 oocysts/L as thresh- old
levels for requiring additional treatment…these levels are estimated to result in an an-
nual infection risk in the range of 1.7x10–4
– 6 x 10–3
… for a treatment p
l
a
n
t
achieving 3-log
Cryptosporidium removal (the treatment efficiency estimated for conventional plants under
existing regulations).”

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
miological approach or using a risk assessment approach. These methods a
re
complementary and have different strengths and limitations, and each has been
used for assessment of drinking water risks in various applications. Epidemiol-
ogical approaches study the relationship between exposures and disease in actual
populations and are descriptive, correlational, or analytic. In the descriptive
study, population surveys or systematic disease surveillance (monitoring)
describe disease patterns by various factors such as age, seasonality, and
geographic location. Correlational (also called “ecologic”) studies collect
population level data on disease rates and exposures and look for correlations.
Ana- lytical studies (whether experimental or observational) are those in which
individual-level data is collected and the investigator tests a formal hypothesis
about the association between exposure and disease.
Risk assessment methods, on the other hand, follow the hazard
identification, dose-response assessment, exposure assessment, and risk
characterization paradigm noted above. Frequently, but not always, the dose-
response assessment is based upon extrapolation from results of trials in
animals (although results from human exposure may be used where
available—for example, in human feeding trials of infectious agents or from
studies in populations exposed in occupational or other settings to particular
agents ofconcern).
Epidemiological studies have the advantage of involving human popul
a-tions,
often experiencing the exposure of interest and representing a range of
variability in susceptibility and behavior. However to detect a small increase in
risk from the baseline, epidemiological studies require very large sample sizes,
and thus considerable expense and effort. Epidemiological studies cannot pro-
vide direct information on the potential for risk reduction from a proposed
change in treatment practice that has not yet been implemented since by defini-
tion there is not yet human exposure to conditions expected from the proposed
change. However, epidemiological studies can be designed to measure the di-
rect impact of a treatment intervention after it has been implemented. This is
very powerful tool and it has provided the evidence base that changes in water
treatment have had a positive impact on community health. For example,
the recent meta-analysis by Fewtrell and Colford (2004) demonstrates the body
of evidence linking improvements in community and household water quality
to health.
Risk assessment approaches have the advantage of being flexible in their
application to potential (but not yet experienced) situations. A risk assessment
can be performed even when the projected risk from a particular exposure
or change of exposure is very small. They have the disadvantage of requiring ex-
tensive measurement or modeling to ascertain exposure, and also of the need for
dose-response studies. Often these dose-response studies are in animals or at
higher doses, thereby requiring extrapolation with respect to dose (via a formal
mathematical dose-response curve) and/or between species. Generally, whether
animal or human data are used to establish the dose-response relationship, the
range in variability in susceptibility is small (compared to a full human popula-

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
tion) and therefore some margin of safety may need to be explicitly used to a
c
-
count for more susceptible subpopulations.
This chapter discusses what is known about the human health risks that d
e
-
rive from contamination of the distribution system, relying on three pri
m
ary
approaches: risk assessment methods that utilize data on pathogen occurrence
measurements, outbreak surveillance data, and epidemiology studies. A special
section is devoted to Legionella, for which all three types of activities have oc-
curred, leading to greater understanding of the risks inherent fromgrowth of this
organism in distribution systems. Because the impetus for this study was revi-
sion of the Total Coliform Rule, the report focuses primarily on acute risks from
microbial contamination of the distribution system. However, there are short-
and long-term risks fromchemicals that merit mention (particularly DBPs—lead
and copper were outside the scope of the study). DBP concentrations in the dis-
tribution system can vary significantly depending on water residence time,
the types of disinfectants used, and biological and chemical reactions, among
many other factors (see Chapter 6). The concentrations of trihalomethanes in
finished water tend to increase with increasing water age, while certain
haloacetic acids tend to decrease in concentration over time (see Chapter 6;
Arbuckle et al., 2002). A number of epidemiologic studies have examined
the health significance of DBP exposure and have reported significantly
increased risks of bladder, rectal, and/or colon cancers in some populations
(King et al., 1996; Koi- vusalo et al., 1997; Doyle et al., 1997; Cantor et al.,
1998; Yang et al., 1998; King et al., 2000) as well as adverse reproductive
outcomes (Waller et al., 1998; Dodds et al., 1999; Klotz and Pyrch, 1999; King
et al., 2000). However, determining and classifying DBP exposure in these
studies has been extremely challenging and has made it difficult to interpret
the findings of these studies (Ar- buckle et al., 2002, Weinberg et al., 2006).
Furthermore, the contribution of distribution systems to the reported risk, as
opposed to drinking water treatment or other processes, has not been elucidated.
Because epidemiological studies of DBP exposure have been extensively
reviewed by others (Boorman et al., 1999; Nieuwenhuijsen et al., 2000; Graves
et al., 2001), they are not reviewed here.
EVIDENCE FROM PATHOGEN OCCURRENCE MEASUREMENTS
The risk assessment approach relies on being able to measure or pr
e
di
c
t (e.g.,by
modeling) the concentration of an etiologic agent in the water supply. Certain
microbial pathogens are indicative of distribution system contamination
stemming from both internal and external sources. These include bacteria
known to form biofilms—a physiological state in which organisms attach to and
grow on a surface (Characklis and Marshall, 1990)—and bacteria that indicate
an external contamination event such as intrusion. In distribution systems, the
interior pipe walls, storage tanks, sediments, and other surfaces in contact with
finished water are colonized by bacteria, which can survive, grow, and
detach depending on local conditions. Other types of bacteria (such as
coliforms) as

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
well as enteric viruses and protozoa (Quignon et al., 1997; Piriou et al., 2000)
are also found in biofilms. However, their presence can also be attributable to an
external contamination event or break through of the treatment barrier.
The microbiology of distribution systems can be influenced by a variety o
f
factors (e.g., poor quality source water, inadequate treatment, unsanitary
activity, backflow). Given this report’s assumption of adequate treatment, a
discussion of all source water microbes and those that would be eliminated
during treatment is not warranted. Furthermore, virtually any microorganism in
close enough proximity to a vulnerable part of the distribution system (e.g., a
cross connection, main break, or leak) could enter during an external
contamination event. Control of these events—see Chapters 4 and 5—is
important for reducing the risks of not only microbial pathogens but also
chemicals that might enter distribution system. Because the complexity of
microbes from such diverse sources is beyond the scope of this report, the
following section focuses on those organisms most likely to indicate either
internal or external contamination of the distribution system.
The Microbiology of Bulk Water
The microbiology of distribution systems essentially consists of two differ-
ent environments—microorganisms in the bulk water column and those in
biofilms attached to the surfaces of pipes, sediments, and other materials. Mi-
croorganisms in the bulk water column originate from either the source water,
from bacterial growth within the treatment process (e.g., within the treatment
filters), from biofilms within the distribution system, or from recontamination of
the water from cross connections, intrusion, pipe breaks, or other external
sources.
Heterotrophic Bacteria
Heterotrophic bacteria (a broad classification that takes into account all b
a
c
-
teria that utilize organic carbon) are commonly found in the bulk water of distri-
bution systems because they readily form biofilms in such systems. They are
measured by using heterotrophic plate counts (HPC). Heterotrophs have tradi-
tionally been divided into two primary groups based on their cell wall character-
istics—Gram-negative andGram-positive.
The presence of a disinfectant residual in drinking water has a
tremendous selective effect, particularly on Gram-negative bacteria, which are
relatively sensitive to inactivation by disinfectants. Identification of bacteria
using fatty acid analysis (Norton and LeChevallier, 2000) showed that
chlorination resulted in a rapid shift frompredominately Gram-negative bacteria
(97 percent) in the raw water to mostly Gram-positive organisms (98 percent)
in the chlorinated water (see Table 3-1). Bacteria in the raw water were diverse,
with Acinetobacter

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
TABLE 3-1 Bacterial Populations Isolated fromthe Water Column During Treatment
Bacterial Identification
Percentage of
Population
in
Raw Water
Percentage of
Population in
Ozone
Contactor
Percentage of
Population in
Filter
Effluent
Percentage of
Population in
Distribution
System
Influent
Gram Negative
Acidovorax spp. 2 4 7
Acinetobacter spp. 29 6
Alcaligenesspp. 12 2 1
Alteromonasspp. 2
Comamonasspp. 1 3
Enterobacter spp. 2 5
Flavobacterium spp. 2 5
Hydrogenophaga 8 3 1
spp.
Klebsiella spp. 10 1 3
Methylobacterium
1 2
spp.
Pseudomonasspp. 14 53 22
Rhodobacter spp. 2 1
Sphingomonasspp. 2 2 19
Stenotrophomonasspp. 2 1 2
Xanthobacter spp. 3
Others* 2 1 5
Gram Positive
Bacillusspp. 7
Nocardia spp. 1 3 7 53
Rhodococcusspp. 16 4
Staphylococcusspp. 1 1
Others* 1 1 1
Unidentified 3 9 16 33
* Includes organisms isolated f rom only one site at a frequency of 1%. 100 i
s
o
l
a
t
e
swere identif ied from each
site.
SOURCE: Adapted f rom Norton and LeChev allier (2000).
spp., Pseudomonas spp., and Klebsiella spp. predominate among the 20
genera identified. Ozonation of the raw water reduced the microbial diversity
to 13 genera, dominated by Pseudomonas spp. and Rhodococcus spp. However,
following biologically active granular activated carbon filtration, 19 genera
were identified in the filter effluent, the majority of which (63 percent) matched
isolates observed in the raw water. The predominant genera were
Pseudomonas spp. and Sphingomonas spp., which are known to grow attached
to the carbon fines of the filter while utilizing natural organic compounds found
in the aquatic environment. Final chlorination of the filtered water resulted in a
shift to No- cardia spp. as the water entered the pipe system. Nocardia spp.
possess characteristic fatty acids that are closely related to Rhodococcus,
Mycobacterium, and Corynebacterium. Its partially acid-fast cell wall and
possession ofthe catalase

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
enzyme, which breaks down hydrogen peroxide, are important factors that en-
able the organismto survive disinfection. Other Gram-positive bacteria found in
chlorinated drinking water include Bacillus and Staphylococcus spp. Bacillus
spp. form environmentally resistant spores that can withstand prolonged contact
with chlorine. Some strains of Bacillus and Staphylococcus aureus can produce
toxins when contaminated water is used in food preparation (LeChevallier and
Seidler, 1980).
Treated drinking water will include a mixture of Gram-negative and Gram-
positive bacteria. In the absence of a disinfectant residual, Gram-negative bacte-
ria will out grow Gram-positive bacteria and dominate the bacterial population.
These organisms typically include Pseudomonas, Acinetobacter,
Flavobacte- rium, and Sphingomonas spp. For the most part, these organisms
have limited public health significance, except for Pseudomonas aeruginosa,
which is a possible opportunistic pathogen in drinking water and in the
biofilms of water systems. It is known to colonize point-of-use carbon filters
in drinking water systems (de Victoria and Galvan, 2001; Chaidez and Gerba,
2004). Pseudomonas aeruginosa is of concern in bathing waters, especially in
swimming pools and spas, where skin infections may result due to exposure. In
the case of drinking water, there are a few studies that suggest a relationship
between the presence of this organism in the water and disease. In one hospital
setting, five of 17 patients with a Pseudomonas infection carried a genotype
also detected in the tap water (Trautmann et al., 2001). In another outbreak of
pediatric P. aeruginosa urinary tract infections, two isolates had genotypes
similar to those in the water. The outbreak was resolved when the taps in the
unit were changed (Ferroni et al., 1998).
Despite these specific incidences, a workgroup recently convened by t
h
e
World
Health Organization (WHO) to address this issue concluded that HPC
bacteria were not associated with any adverse health effect (Bartram et al.,
2003). “Some epidemiological studies have been conducted into the relation-
ship between HPC exposures from drinking water and human health
effects. Other studies relevant to this issue include case studies, especially
in clinical situations, and compromised animal challenges using heterotrophic
bacteria obtained fromdrinking water distribution systems. The available body
of evidence supports the conclusion that, in the absence of fecal contamination,
there is no direct relationship between HPC values in ingested water and human
health effects in the population at large. This conclusion is also supported
indirectly by evidence from exposures to HPC in foodstuffs where there is
no evidence for health effects link in the absence of pathogen contamination.
There are a small number of studies that have examined possible links between
HPC bacteria and non-intestinal outcomes in general populations. The
conclusions of these studies do not support a [health] relationship” (WHO,
2002).
One of the difficulties in interpreting the significance of HPC data is tha
t
test methods involve a wide variety of conditions that lead to a wide range of
quantitative and qualitative results. For this reason, the EPA has not yet is sued a
health-based standard. However, the Surface Water Treatment Rule
requires

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
that distribution system locations without a detectable disinfectant residual
maintain HPC levels at or below 500 colony forming units (CFU)/mL in at least
95 percent of the samples each month (EPA, 1989).
Coliform Bacteria. Total coliform bacteria (a subset of Gram-negative
bacteria) are used primarily as a measure of water treatment effectiveness and
can occasionally be found in distribution systems. The origins of total coliform
bacteria include untreated surface water and groundwater, vegetation, soils, in-
sects, and animal and human fecal material. Typical coliform bacteria found in
drinking water systems include Klebsiella pneumoniae, Enterobacter aerogenes,
Enterobacter cloacae, and Citrobacter freundii. Other typical species and gen-
era are shown in Table 3-2. Although most coliforms are not pathogenic, they
can indicate the potential presence of fecal pathogens and thus in the absence of
more specific data may be used as a surrogate measure of public health risk.
Indeed, the presence of coliforms is the distribution system is usually interpreted
to indicate an external contamination event, such as injured organism passage
through treatment barriers or introduction via water line breaks, cross connec-
tions, or uncovered or poorly maintained finished water storage facilities
(Gel- dreich et al., 1992; Clark et al., 1996). However, biofilms within
distribution systems can support the growth and release of coliforms, even when
physical integrity (i.e., breaches in the treatment plant or distribution system)
and disinfectant residual have been maintained (Characklis, 1988; Haudidier et
al., 1988; Smith et al., 1990), such that their presence may not necessarily
indicate a recent external contamination event. Coliform regrowth in the
distribution system is more likely during the summer months when temperatures
are closer to the optimum growth temperatures of these bacteria.
Thermotolerant coliforms (capable of growth at 44.5 o
C), also termed “fecal
coliforms” have a higher association with fecal pollution than total coliforms.
And Escherichia coli is considered to be even more directly related to fecal pol
-
lution as it is commonly found in the intestinal track of warm-blooded animals.
Although most fecal coliform and E. coli strains are not pathogenic, some strains
are invasive for intestinal cells and can produce heat-labile or heat-stable toxins
(AWWA, 1999). E. coli and most of the thermotolerant coliforms do not grow
in biofilms, although they most likely can be trapped and retained w
i
t
hi
n
biofilms.
TABLE 3-2 Coliform Isolates Typically Found in Drinking Water
Citrobacter Enterobacter Escherichia Klebsiella
C. freundii E. aerogenes E. coli K. pneumonia
C. diversus E. agglomerans K. oxytoca
E. cloacae K. rhinoscleromatis
K. ozaena
SOURCE: Adapted fromGeldreich and LeChevallier (1999).

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
Aeromonas. Aeromonas spp. are Gram-negative bacteria found in fr
e
shand
salt water and cause a wide variety of human infections including septice-
mia, wound infections, meningitis, pneumonia, respiratory infections, hemolytic
uremic syndrome, and gastroenteritis (Carnahan and Altwegg, 1996; Alavandi et
al., 1999). The ability of these microorganisms to grow at low temperatures and
low nutrient conditions are important in their occurrence in drinking water sup-
plies. Through the Unregulated Contaminant Monitoring Rule (see Chapter 2),
EPA examined the occurrence of Aeromonas spp. in 308 drinking water
systems and found detectable concentrations in 2.6 percent of 5,060 samples and
in 13.6 percent of the systems. In a 16-month study conducted on the presence
of A. hydrophila in drinking water in Indiana, 7.7 percent of the biofilm samples
were positive for A. hydrophila (Chauret et al., 2001). The health significance
of detecting aeromonads in drinking water is not well understood. Some
countries (such as the Netherlands) have set standards for aeromonads in
drinking water leaving the treatment plant (< 20 CFU/200 mL) and in the
distribution system(< 200 CFU/100 mL).
Mycobacteria. Organisms of the genus Mycobacteria are also found i
n
drinking water. Of particular concern is the MAC, or Mycobacterium avium
complex. Studies have detected M. avium complex organisms in drinking water
distribution systems with concentrations ranging between 0.08 and 45,000
CFU/mL (Haas et al., 1983; duMoulin and Stottmeir, 1986; Carson et al., 1988;
duMoulin et al., 1988; Fischeder et al., 1991; von Reyn et al., 1993; Glover et
al., 1994; von Reyn et al., 1994; Covert et al., 1999). M. avium are resistant to
disinfectants, especially free chlorine (Taylor et al., 2000). Indeed, it is postu-
lated that they may in fact be selected for in distribution systems as a result of
their resistance to chlorine (Collins et al., 1984; Schulze-Robbecke and Fische-
der, 1989; Briganti and Wacker, 1995). However, there is also evidence
that MAC are susceptible to chlorine dioxide and chloramine (Vaerewijck
et al., 2005).
Falkinham et al. (2001) examined eight, well characterized drinking w
a
t
e
r
systems and reported that 20 percent of the water isolates and 64 percent of t
he
biofilm isolates were identified as M. avium or M. intracellulare. Additionally,
8 percent of the water isolates were identified as M. kansasii. Most of
these isolates were detected in raw water samples, with M. avium complex
organisms detected in five of six surface water sites ranging from 6 to 35
percent of the organisms isolated. M. avium complex organisms were not
detected in any plant or well effluent sample, but were occasionally detected at
low levels (< 1 CFU/mL) in drinking water systems. However, M. avium and
M. intracellulare were recovered frequently from drinking water biofilm
samples, indicating that
M. avium levels were increasing in the distribution system. Increases in M
.avium
levels in drinking water were correlated to levels of AOC (r2
= 0.65, p = 0.029)
and BDOC (r2
= 0.64, p = 0.031) (Falkinham et al., 2001; LeChevallier, 2004).

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
The greatest increase in M. avium complex infections have been with a
c
-
quired
immunodeficiency syndrome (AIDS) patients; approximately 25 to 50 percent
of these patients suffer debilitating and life-threatening MAC infections
(Horsburgh, 1991; Nightingale et al., 1992), although the availability of highly
active antiretroviral therapy has reduced the incidence of MAC in AIDS patients
in recent years. Members of the MAC are known opportunistic pathogens, with
symptoms of pulmonary infection mimicking that of M. tuberculosis (Wolinsky,
1979). The organism infects the gastrointestinal or pulmonary tract, suggesting
that food or water may be important routes of transmission for AIDS patients
(Singh and Lu, 1994). It should be pointed out that epidemiology studies have
not yet identified drinking water as a risk factor for MAC, except perhaps in
hospital water systems.
Free-Living Protozoa
Of the genera of protozoa present in distribution systems, Acanthamoeba,
Hartmanella and Naegleria are known to feed on bacteria and biofilms by g
r
a
z
-ing.
Previous research has shown that all coliforms as well as bacterial patho-gens and
opportunistic pathogens may be ingested by protozoa. Ingested bacte-ria, if not
digested, may survive within the protozoa and be protected from residual
disinfectant. The survival of Legionella has been the subject of numerous
reports in the literature with regards to its increased resistance to disinfectants
while in the intracellular state (Levy, 1990).
Of the eucaryotes mentioned above, two are known to be pathogeni
c—
Naegleria spp. and Acanthamoeba. These are usually associated with re
cre
a-tional
rather than drinking waters, although Acanthamoeba was included as p
a
r
t of the
first Contaminant Candidate List (EPA, 1998) as an opportunistic pathogen
affecting contact lens wearers. Previous studies have shown that these or-
ganisms are usually found at the source. However, cysts have also been isolated
from drinking water distribution systems in France (Jacquemin et al., 1981;
Gel- dreich, 1996).
Routine monitoring for free-living protozoa is rarely done. Isolation a
nd
identification of these organisms are accomplished only when there is evidence
for disease outbreak or when research studies are being conducted. As interest in
the ability for protozoa to harbor bacterial pathogens increases, it is probable
that more effort will be expended in determining their presence in
distribution systems,including premise plumbing.
Fungi
Although many fungi have been found in drinking water systems, their l
ev-
els are typically low and the organisms have not been directly associated with
disease (Kelley et al., 2003). The origin of fungi in drinking water systems has

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
not been well characterized, but it is assumed that they come from environ-
mental sources including surface water and groundwater, soils, and vegetation.
The four most frequently occurring genera of filamentous fungi isolated from
chlorinated and unchlorinated distribution systems in southern California were
Penicillium, Sporocybe, Acremonium, and Paecilomyces (Nagy and Olson,
1982). Aspergillus fumigatus was the predominant species detected in the dis-
tribution system water supplies in Finland (Niemi et al., 1982). A variety of
fungi (Cephalosporium sp., Verticillium sp., Trichodorma sporulosum, Nectria
veridescens, Phoma sp., and Phialophora sp.) were identified from water service
mains in England (Bays et al., 1970; Dott and Waschko-Dransmann, 1981).
Outside of specialized research studies, potable water supplies are not routinely
tested for fungi.
The Microbiology of Distribution System Biofilms
Biofilms in drinking water pipe networks contain all of the organisms m
en-
tioned above that are found in bulk distribution system water, as well as others.
The microbial composition of any given pipe segment can be highly
variable, and in most cases is poorly, if ever, characterized. The pipe surface
itself can influence the composition and activity of biofilm populations. Studies
have shown that biofilms developed more quickly on iron pipe surfaces than on
plastic PVC pipes, despite the fact that adequate corrosion control was applied,
that the water was biologically treated to reduce AOC levels, and that chlorine
residuals were consistently maintained (Haas et al., 1983; Camper, 1996).
In addition to influencing the development of biofilms, the pipe surface ha
s
also been shown to affect the composition of the microbial communities present
within the biofilm (Figure 3-1). Iron pipes supported a more diverse microbial
population than did PVC pipes (Norton and LeChevallier, 2000). Undoubtedly
part of the reason that certain bacteria associate with certain pipe types is be-
cause materials may leach compounds that support bacterial growth. For exam-
ple, pipe gaskets and elastic sealants (containing polyamide and silicone) can be
a source of nutrients for bacterial proliferation. Colbourne et al. (1984) reported
that Legionella were associated with certain rubber gaskets. Organisms associ-
ated with joint-packing materials include populations of Pseudomonas aerugi-
nosa, Chromobacter spp., Enterobacter aerogenes, and Klebsiella pneumoniae
(Schoenen, 1986; Geldreich and LeChevallier, 1999). Coating compounds for
storage reservoirs and standpipes can contribute organic polymers and solvents
that may support regrowth of heterotrophic bacteria (Schoenen, 1986; Thofern et
al., 1987). Liner materials may contain bitumen, chlorinated rubber, epoxy
resin, or tar-epoxy resin combinations that can support bacterial regrowth
(Schoenen, 1986). PVC pipes and coating materials may leach stabilizers that
can result in bacterial growth. Studies performed in the United Kingdom re -
ported that coliform isolations were four times higher when samples were col-
lected from plastic taps than from metallic faucets (cited in Geldreich and

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
LeChevallier, 1999). The purpose of these studies was not to indicate that c
e
r
-tain
pipe materials are preferred over another, but to demonstrate the importance of
considering the type of materials that come into contact with potable water.
Although procedures are available to evaluate the growth stimulation potential
of different materials (Bellen et al., 1993), these tests are not applied in the
United States by ANSI/NSF.
Pseudomonas
16.0%
Stenotrophomonas
4.0%
Acidovorax
24.0%
Other - Gram -
8.0%
A
Xanthobacter
22.0%
Nocardia
26.0%
Stenotrophomonas
74.0%
B
Other - Gram +
1.0%
Agrobacterium
4.0%
Other - Gram -
1.0%
Nocardia
20.0%
FIGURE 3-1 Microbial populations isolated from iron pipe (A) or P
V
C
(B) surfaces.
SOURCE: Adapted fromNorton and LeChevallier (2000).

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
***
For both bulk drinking water and biofilms, the identification of microor-
ganisms typically relies on culturing bacteria from potable supplies, which h
a
s
important limitations. Culture methods do not detect all microbes that may exi
stin
water, such that only a fraction of viable organisms is recovered (Amann e
t
al.,
1995). In addition, most culture methods only detect relatively rapidly
growing heterotrophic bacteria, and slowly growing organisms, fastidious or
autotrophic organisms, and anaerobes are generally not examined.
Diagnostic kits are unreliable for many heterotrophic bacteria because the
methodology often requires the analyst to perform a Gram stain, which is
difficult because of the slow growth and acid-fast or partially acid-fast nature of
bacteria surviving in disinfected drinking water.
An alternative method includes fatty acid profiling. As shown above, thi
s
approach can be used to identify organisms from drinking water (Norton a
n
d
LeChevallier, 2000) but in this study the organisms were cultured prior to identi-
fication and therefore the limitations associated with culturing are still present.
Additionally, for identification, the lipid profile must match an established pro-
file in a database; these databases are predominated by medical (and not envi-
ronmental) organisms. The use of fatty acid profiles was further developed by
Smith et al. (2000) who used biofilm samples without prior culturing to demon-
strate that predominantly Gram-negative bacteria were present, but no further
identification was accomplished. A similar approach was taken by Keinanen et
al. (2004) who compared profiles from two drinking water systems and showed
that they differed, but again, no identifications were obtained. Although
fatty acid profiling has been used in these studies to provide some insight on
microbial ecology, the limitations associated with the method preclude it from
extensive use in characterizing mixed microbial communities.
Molecular methods offer the promise of a more complete determination o
f
the microbiology of drinking water (see Chapter 6 for details). DNA extraction
coupled with polymerase chain reaction (PCR) amplification can be used to
identify waterborne microbes (Amann et al., 1990, 1995). These procedures can
be combined with quantitative real-time PCR, fluorescence in-situ hybridization,
or flow cytometry to provide quantitative assessments of bacterial populations.
However, careful quality assurance is necessary to ensure complete extraction
and recovery of environmental DNA. Martiny et al. (2003) utilized
terminal restriction fragment length polymorphisms to identify members of a
biofilm consortium over a three-year time period. In this study, several
organisms were identified (Pseudomonas, Sphingomonas, Aquabacterium,
Nitrospira, Plancto- myces, Acidobacterium) but for the majority of the
peaks no sequence match could be made.
It is telling that there is very little published information about the micro-
bial ecology of distribution systems. At this point in time, the detection methods
are expensive, are time consuming, require optimization for specific conditions,
and are appropriate only for the research laboratory. As a consequence,there is

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
a lack of information about the types, numbers, and activities of microorganisms in
drinking water. It is also unknown how the ecology of the main distribution
systemis related to that in premise plumbing, how the populations vary between
distribution systems in different locations, and how the populations respond to
water quality changes within a distribution system. This translates into a lack of
understanding about whether organisms of potential public health concern may
be present in water systems and further complicates the ability to assess risk due
to their presence.
As mandated by the Safe Drinking Water Act, the EPA has issued a s
e
c
- ond
Contaminant Candidate List that includes 10 microbes (or microbial p
r
o
d
- ucts) for
potential future regulation (EPA, 2004) (see Table 3-3). For most o
f
these
microbes, methods do not exist for routine testing of drinking water supplies,
and basic research is needed on their occurrence, survival, and importance in
potable water. Where the current list includes organisms that are not dis -
cussed above, they are considered to be of primary concern in untreated or in -
adequately treated source waters and not in distribution systems, such that a
more detailed discussion is beyond the scope of the report.
It can be hard to determine whether the detection of frank or opportunistic
pathogens in drinking water poses an unacceptable risk. In addition to the moni-
toring techniques being difficult, time-consuming, expensive, and of poor sensi-
tivity, the methods do not detect specific virulence determinants, such that many
environmental isolates (e.g., E. coli, Aeromonas, Legionella, etc.) are indistin-
guishable from their clinical strains. Therefore even when monitoring for poten-
tially pathogenic organisms is done, the public health significance of the results
is often in question. Furthermore, there is insufficient supporting information
(in terms of occurrence data for exposure assessment, dose-response data, health
effects, and models that can predict pathogen occurrence for different distribu-
tion system contamination scenarios such as contamination via cross connec-
tions, main breaks, or intrusion) to conduct a risk assessment for many water-
borne microbes. For all these reasons, measurement of the microbe itself
is
TABLE 3-3 Contaminant Candidate List Microbes
Bacteria Mycobacterium avium
Helicobacter
Aeromonas
Viruses Caliciviruses
Echovirus
Coxsackieviruses
Adenovirus
Protozoa Microsporidium
Toxins Cyanobacterial toxins

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
typically insufficient to make a public health determination. Until better moni-
toring methods, pathogen occurrence models, dose-response data, and risk a
s
-
sessment data are available, pathogen occurrence measurements are best used in
conjunction with other supporting data on health outcomes. Such supporting
data could include enhanced or syndromic surveillance in communities, as well
as the use of microbial or chemical indicators of potential contamination.
EVIDENCE FROM OUTBREAK DATA
Most information on the risks of waterborne disease in the United S
t
a
t
e
s
comes
from surveillance and investigation of waterborne disease outbreaks. Apassive
voluntary surveillance system for waterborne disease outbreaks s
t
a
r
t
e
din 1971 and is
a collaboration between the Centers for Disease Control and Pre- vention
(CDC), the EPA, and state and regional epidemiologists. This surveillance
system includes outbreaks associated with both drinking and recreational water,
and outbreaks due to both microbial and chemical agents. The objectives of the
surveillance system are to (1) characterize the epidemiology of waterborne
disease outbreaks, (2) identify the etiologic agents that cause the outbreaks, (3)
determine the risk factors that contributed to the outbreak, (4) inform and train
public health personnel to detect and investigate waterborne disease
outbreaks, and (5) collaborate with local, regional, national and international
agencies on strategies to prevent waterborne diseases (Stanwell-Smith et al.,
2003).
From 1971 through 2002, 764 drinking water outbreaks have been reported
through this surveillance system. Although this is believed to be an u
n
d
e
r
e
s
t
i
-mate of
the true number of outbreaks that occurred during this period, the information
collected in this surveillance system has been extremely valuable for
improving our understanding of the agents that cause waterborne disease and the
risk factors involved in waterborne disease outbreaks. The data collected in this
surveillance systemincludes:
 Type of exposure (drinking water or recreational water)
 Location and date of outbreak
 Actual or estimated number of persons exposed, ill, hospitalized,dead
 Symptoms, incubation period, duration of illness
 Etiologic agent
 Epidemiological data (attack rate, relative risk or odds ratio)
 Clinical laboratory data (results of fecal and serology tests)
 Type of water system
◦ Community, non-community, or individual homeowner drinking w
a
-ter
supply
◦ Swimming pool, hot tub,water park, or lake for recreational water

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
 Environmental data (results of water analyses, sanitary survey, w
a
t
e
r
plant inspection)
 Factors contributing to contamination ofwater
The surveillance data are summarized in biannual reports (Morbidity a
n
d
Mortality Weekly Report Surveillance Summaries) that are published by the
CDC and distributed to public health authorities and practitioners throughout the
country. The information is also available on the Internet at http://www.cdc.gov/
mmwr. These reports (Herwaldt et al., 1991; Moore et al., 1993; Kramer et al.,
1996; Levy et al., 1998; Barwick et al., 2000; Lee et al., 2002a; Blackburn et al.,
2004) indicate three main trends:
1. The overall number of reported waterborne disease outbreaks associ-
ated with drinking water is declining from a peak of over 50 reported outbreaks
in 1980 to eight reported outbreaks in 2002.
2. For a substantial portion of drinking water outbreaks, the pathogen i
snot
identified and the outbreaks are classified as “acute gastrointestinal illness of
unknown etiology” (AGI). From 1986 through 2002, approximately 41 percent
of the over 250 outbreaks reported during this period were classified as
AGI, and this proportion varies by reporting period from a peak of 68 percent
in 1991–1992 to 17 percent in 1993–1994. Overall, Giardia and
Cryptosporidium are the most commonly reported etiologic agents of
waterborne disease when a pathogen is identified and are associated with about
20 percent of reported outbreaks associated with drinking water since the mid -
1980s. However, with the recent addition of Legionella outbreaks to the
surveillance system, Legionella is now the single most common cause of
outbreaks involving drinking water (as discussed below).
3. Most drinking water outbreaks involve groundwater systems, e
s
p
e
c
i
a
l
l
y
untreated groundwater systems. Forty (40) percent of the 25 drinking w
a
t
e
r
outbreaks reported between 2001 and 2002 involved untreated groundwater sys -
tems (Blackburn et al., 2004).
Declining Number of Drinking Water Outbreaks
Since the mid-1980s, the number of waterborne outbreaks has decl
i
ned
(Figure 3-2). The reason for the decrease is largely attributed to the promulga-
tion of more stringent drinking water regulations, including the Surface Water
Treatment Rule, the Total Coliform Rule, and others. In addition, many water
utilities have made voluntary improvements, such as the Partnership for Safe
Water program to reduce the risk of waterborne cryptosporidiosis. The Partner-
ship program entails a comprehensive evaluation of treatment practices with a
focus on achieving filtered drinking water turbidities less than 0.1 nephelometric
turbidity units (NTU). The number of reported outbreaks began to decrease
sharply beginning with the 1985–1986 reporting period; this was attributable

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
FIGURE 3-2 Number of drinking w ater disease outbreaks in the United S
t
a
t
e
s
, 1971–2002.
Individual—private or individual w ater systems (9 percent of U.S. population or 24
million users); Community—systems that serve > 25 users year round (91 percent of U.S.
population or 243 million users); Noncommunity—systems that serve < 25 users and
transient water systems such as restaurants, highway rest areas, parks (millions of users
yearly). SOURCE: Blackburn et al. (2004).
primarily to fewer community and noncommunity outbreaks. With the
institu-tion and enforcement of better regulations that chiefly affect these types of
water systems (particularly community systems), a marked drop in the number
of outbreaks was seen. In contrast, the increase in outbreaks reported
during 1999– 2000 was attributable primarily to individual homeowner systems,
which affect fewer persons, are less regulated, or are more subject to changes in
surveillance and reporting. In 2001–2002, individual homeowner systems
comprised 40 percent of the waterborne outbreaks (Figure 3-3).
Etiologic Agents Associated With Drinking Water Outbreaks
The agents responsible for waterborne disease outbreaks were p
r
e
d
o
m
i
-nantly
undefined, microbial (parasitic, bacterial, or viral), or chemical. Indeed,
surveillance data on waterborne disease outbreaks associated with drinking wa -
ter in the United States from 2001 to 2002 indicate that almost 30 percent
of reported outbreaks were due to bacterial agents, 16 percent were due to proto-
zoa, 16 percent were due to viral agents, 16 percent were due to chemical con-
taminants, and 23 percent had an unidentified etiology. Figure 3-4 shows the

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
FIGURE 3-3 Waterborne outbreaks by etiological agent, water system, water s
o
u
r
c
e
,and
deficiency—United States, 2001–2002. SOURCE: Blackburn et al. (2004).
etiology of waterborne disease outbreaks over time. The large number of water-
borne disease outbreaks associated with protozoa in the early 1980s was mostly
caused by Giardia and was greatly reduced by the implementation of the Sur-
face Water Treatment Rule in 1989 (Barwick et al., 2000). Relatively few out-
breaks due to viruses have been reported, in part because of the difficulty of the
detection methodologies for these organisms. However, the number of reported
viral outbreaks has increased significantly since 1999 with the development of
better diagnostic techniques for noroviruses. Nine of the 15 drinking water out-
breaks associated with noroviruses that have been reported since 1986 occurred
between 1999 and 2002 (Herwaldt et al., 1991; Moore et al., 1993; Kramer et
al., 1996; Levy et al., 1998; Barwick et al., 2000; Lee et al., 2002a; Blackburn
et al., 2004).

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
FIGURE 3-4 Agents responsible for waterborne outbreaks. SOURCE: Blackburn et al.
(2004).
Over the past 30 years, there has been a wide range of chemical agents a
s
-
sociated with drinking water outbreaks, including arsenic, benzene, chlordane,
chlorine, chromate, copper, cutting oil, developer fluid, ethyl acrylate, ethylene
glycol, fluoride, fuel oil, furadan, lead, leaded gasoline, lubricating oil, kerosene,
nitrate, nitrite, phenol, polychlorinated biphenyls, selenium, sodium hydroxide,
toluene, xylene, and unidentified herbicides. From 1993 through 2002, most
drinking water outbreaks associated with chemical agents have been due to cop-
per (eight outbreaks, usually related to premise plumbing) followed by ni-
trates/nitrites (six outbreaks, usually related to contamination of groundwater)
(Kramer et al., 1996; Levy et al., 1998; Barwick et al., 2000; Lee et al., 2002a;
Blackburn et al., 2004).
Outbreaks Associated With Groundwater Systems
In recent years, as treatment of surface water supplies has improved, water-
borne outbreaks have increasingly involved groundwater supplies (Figure 3-
3). There is increasing recognition that many groundwater supplies have
microbial contamination, yet the use of untreated groundwater continues in
many small communities and by individual homeowners. A survey of 448
wells in 35 states reported that 31 percent of the sites were positive for at least
one virus, and en- terovirus RNA was detected in approximately 15 percent,
rotavirus RNA in 14 percent, and hepatitis A virus RNA in 7 percent of the
wells by reverse-tran-

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
scription-polymerase chain reaction (RT-PCR) (Abbaszadegan et al.,
2003).Fout et al. (2003) examined 321 samples from 29 groundwater sites by RT -
PCR and reported that 72 percent of the sites were virus positive. Borchardt
et al. (2004) collected monthly samples from four municipal wells in one city in
Wis- consin for a 12-month period and detected enteric viruses by RT-PCR in
50 percent of the samples. Two studies in Ontario, Canada examined the
relationship between E. coli in well water and acute gastrointestinal illness in
households using the water for drinking (Raina et al., 1999, Strauss et al., 2001).
In the first study of 181 households with untreated well water, water samples
were collected five times during the one-year study, and E. coli was detected in
20 percent of the household wells. The second study included 235 households
in four rural communities (Strauss et al., 2001) and reported that 20 percent
of the households had at least one water sample that exceeded the national
standards for total coliforms or E. coli.
Outbreaks Associated With Distribution Systems
Among the seven outbreaks associated with community water systems in
2001–2002, four (57.1 percent) were related to problems in the water distribu-
tion system. Preliminary results from the 2003–2004 surveillance report indi-
cate that distribution systems were associated with 38 percent of the outbreaks
associated with drinking water systems during this period (Liang et al., 2006).
Other epidemiological and outbreak investigations conducted in the last five
years suggest that a substantial proportion of waterborne disease outbreaks, both
microbial and chemical, is attributable to problems within distribution systems
(Craun and Calderon, 2001; Blackburn et al., 2004) (see Figure 1-1). Craun and
Calderon (2001) examined causes of reported waterborne outbreaks from 1971
to 1998 and noted that, in community water systems, 30 percent of 294 out-
breaks were associated with distribution systemdeficiencies, causing an average
of 194 illnesses per outbreak. Distribution system contamination was observed
to be the single most important cause of outbreaks in community water systems
over that time period.
The reason for the apparent increase in the proportion of outbreaks associ
-ated
with water distribution systems is not entirely clear. Outbreaks associated with
distribution systemdeficiencies have been reported since the surveillance system
was started. However, there may be more attention focused on the distribution
system now that there are fewer outbreaks associated with inadequate treatment
of surface water. Also, better outbreak investigations and reporting systems
in some states may result in increased recognition and reporting of all the risk
factors contributing to the outbreak, including problems with the distribution
system that may have been overlooked in the past. Although waterborne
disease outbreaks in general are still under-reported, the surveillance systemhas
become more mature, and outbreak investigations and analyses are becoming
more sophisticated.

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
The CDC surveillance system for waterborne disease outbreaks attempts t
o
collect information on outbreaks and their contributing causes. For example,
from 1981 to 1998, the CDC documented 57 waterborne outbreaks related
to cross-connections, resulting in 9,734 detected and reported illnesses (Craun
and Calderon, 2001). Contamination fromcross-connections and backsiphonage
were found to cause the majority of the outbreaks associated with distribution
systems (51 percent), compared with contamination of water mains following
breaks (39 percent) and contamination of storage facilities (the remaining 10
percent). A separate compilation by the EPA of backflow events revealed many
more incidents of backflow and resulting outbreaks—a total of 459 incidents
resulting in 12,093 illnesses from backflow events from 1970 to 2001
(EPA, 2002). The situation may be of even greater concern because incidents
involving premise plumbing are even less recognized.
Most reported outbreaks associated with distribution systems occur i
n
community water systems because of their greater size and complexity. For
example, from 1999 to 2002 there were 18 reported outbreaks in community
water systems, and nine (50 percent) of these were related to problems in the
water distribution system (Lee et al., 2002b; Blackburn et al., 2004). However,
there have been a number of reported outbreaks associated with noncommunity
water systems that have been attributed to deficiencies in the distribution sys -
tem. Finally, the magnitude and severity of reported outbreaks associated with
distribution systems vary, with an average about almost 200 illnesses per
outbreak (Craun and Calderon, 2001) and a total of 13 deaths.
The Extent of Underestimation
The number of identified waterborne disease outbreaks is considered an u
n
-
derestimate because not all outbreaks are recognized, investigated, or reported t
o
health authorities (Blackburn et al., 2004). For example, outbreaks occurring i
n
national parks, tribal lands, or military bases might not be reported to state o
r
local authorities. Factors influencing whether a waterborne outbreak is
recog-nized include awareness of the outbreak, availability of laboratory testing,
and resources available for surveillance and investigation of outbreaks. The
detection and investigation of waterborne outbreaks is primarily the
responsibility of the local, state, and territorial public health departments with
varying resources and capacities. Differences in the capacity of local and state
public health agencies and laboratories to detect an outbreak might result in
reporting and surveillance bias, such that the states with the majority of
outbreaks might not be the states with the majority of waterborne disease.
Outbreaks are more likely to be recognized when they involve acute illnesses
with symptoms requiring medical treatment, or when sensitive laboratory
diagnostic methods are readily available. These and other limitations are
discussed below.

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
Underreporting of Outbreaks Involving Individual Homeowner Systems
Although the surveillance system has always included outbreaks associated
with individual homeowner water systems, it is likely that most sporadic c
a
s
e
s and
small clusters of waterborne disease associated with individual homeowner water
systems are not recognized or reported because small numbers of peopl
e are
involved. Furthermore, a cluster of cases of gastroenteritis within a si
ngle
household may easily be attributed to food contamination or person-to-person
transmission, such that the possibility of waterborne transmission may not
be considered or investigated. From 1971 to 1980, 37 (11.6 percent) of the
320 reported drinking water outbreaks were associated with individual
homeowner systems, and most of these outbreaks involved chemical agents
when an etiologic agent was identified (Craun, 1986). From 1993 to 2002, 41
(28.7 percent) of the 143 reported drinking water outbreaks were associated with
individual homeowner water systems, suggesting that there may be increased
recognition and reporting of these smaller outbreaks in the past ten years of
surveillance.
Underreporting of Outbreaks Involving Premise Plumbing
Outbreaks associated with premise plumbing are not specifically
identified in the CDC surveillance reports. Adverse health effects associated with
premise plumbing problems are less likely to be recognized and reported in this
surveillance system, especially if they occur within a single household.
However, a number of outbreaks associated with drinking water have been
reported from public building settings such as schools, restaurants, churches,
factories, and apartment buildings. Some of these outbreaks were due to
contamination of a private well that serves the building. Other outbreaks in
public buildings were classified as due to distribution system deficiencies and
appeared to involve cross-connections and/or backsiphonage problems.
Examples of the latter type of outbreak include:
 an outbreak of copper poisoning in the early 1980s that occurred w
h
e
n
“backsiphonage of corrosive water containing carbon dioxide from a soda-
mixing dispenser caused copper to be leached from piping in a building (Craun,
1986);
 a norovirus outbreak in 1995 at a high school in Wisconsin t
h
a
taf-
fected 148 persons. The school was connected to the community water supply.
However, water in the school became contaminated from backsiphonage of wa-
ter from hoses submerged in a flooded football field (Levy et al., 1998);
 a chemical outbreak in 1995, in which 13 persons in a healthcare f
a
c
i
l
- ity
in Iowa became ill after drinking water that was contaminated with concen-
trated liquid soap. A valve on the water supply hose to the soap dispenser had
been left open and allowed the soap to enter the water supply in the building.

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
Although the building had vacuum breakers to prevent backsiphonage,
these were installed incorrectly at the soap dispensers (Levy et al., 1998);
 a chemical outbreak in 1999, in which four residents of an apartm
ent
building in Florida had acute gastroenteritis that was attributed to
unidentified chemical poisoning. A cross-connection was discovered between
their drinking water and an improper toilet flush-valve. Residents of the
apartment had no- ticed on several occasions that their tap water was blue before
the onset of illness (Lee et al., 2002a).
 a small waterborne disease outbreak at a middle school in Fl
ori
da in
2001 due to a cross-connection between the air conditioning unit and the potable
water supply. A maintenance worker used the potable water systemto dilute the
ethylene glycol solution in the chiller unit. The higher water pressure in the
chiller unit forced the diluted ethylene glycol into the school’s water supply and
pink-colored water was observed in the school bathrooms. Three students be-
came ill with gastrointestinalsymptoms (Blackburn et al., 2004).
Underreporting of Outbreaks Involving Chemical Agents
From 1971 to 1980, 38 (11.9 percent) of the 320 reported drinking w
a
t
e
r
outbreaks were attributed to chemical agents (Craun, 1986), and from 1993 to
2002, 25 (17.5 percent) of the 143 reported drinking water outbreaks were
attributed to chemical agents (Kramer et al., 1996; Levy et al., 1998; Barwick
et al., 2000; Lee et al., 2002a; Blackburn et al., 2004). The CDC believes that
waterborne chemical poisonings are underreported for many reasons. First,
most of these are probably due to copper and lead leaching from plumbing in
private residences and affect relatively few people and are consequently unlikely
to be recognized by public health authorities. Furthermore, exposure to
chemicals in drinking water can often cause non-specific symptoms that may not
be recognized as chemical poisoning or may not be linked to a specific
chemical. The detection, investigation, and reporting of waterborne disease
outbreaks linked to chemical exposures are not as well established as the
methods for dealing with outbreaks associated with infectious agents. Finally,
many physicians may have difficulty recognizing and diagnosing chemical
poisonings unless they have had additional training in this area (Barwick et al.,
2000).
Revisionsof the CDC Waterborne Disease Outbreak Surveillance System
The CDC is making several changes to its waterborne disease outbreak sur-
veillance system that are relevant to better understanding the role of distribution
systems, including premise plumbing. Previously, the risk factors or deficien-
cies that contributed to a waterborne disease outbreak were classified as: (1) use
of untreated surface water, (2) use of untreated groundwater, (3) treatment defi-
ciency, (4) distribution systemproblem, or (5) miscellaneous. The 2003–2004

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
MMWR Surveillance Summary will use a new and detailed classification s
y
s
-tem
for risk factors that contributed to the outbreak, and it will distinguish b
e
-tween
deficiencies before or after entry into a building or home. This distinction is
important because drinking water before it enters a building is usually man-
aged by the water utility and subject to EPA drinking water regulations. How-
ever, drinking water problems that occur after entry into a building, such as
those due to Legionella colonization in premise plumbing, cross-connections,
point-of-use devices, or drink mix machines, may not be the responsibility of the
water utility or regulated by EPA (lead and copper are an exception—see Chap-
ter 2). Preliminary results from the surveillance system for 2003–2004 indicate
that 48 percent of the outbreaks associated with drinking water were associated
with deficiencies in source water, water treatment, and the distribution system
and 52 percent of the outbreaks were due to deficiencies after the point of entry.
In this latter group of outbreaks, approximately 47 percent involved Legionella
and 35 percent involved chemical agents (including copper) (Liang et al., 2006).
In addition, the surveillance system will now report all the identified deficien-
cies that contributed to the waterborne disease outbreak rather than reporting
only the primary deficiency. Finally, CDC is moving toward a web-based sys-
tem for reporting outbreaks and developing a public access database on water-
borne disease outbreaks that will allow investigators to examine and
analyze these data.
EPIDEMIOLOGY STUDIES
Three basic epidemiological study designs can be used to assess the p
u
b
l
i
c
health risk of contaminated water supplies (Steenland and Moe, 2005): descrip-
tive, correlational or ecological, and analytic. In the descriptive study, popula-
tion surveys or systematic disease surveillance describe disease patterns by vari-
ous factors such as age, seasonality, and geographic location. These studies do
not test a formal hypothesis about the relation between a specific exposure (or
risk factor) and disease, but they can help identify specific populations or geo-
graphic regions for further study. This category includes the systematic surveil-
lance of outbreaks discussed in the previous section as well as endemic cases.
Surveillance systems are useful for showing trends in the causes and risk factors
of waterborne disease, but they are not very sensitive and cannot serve as a rapid
warning system of a water-related health problem in a specific community be-
cause of reporting delays. In addition to the waterborne disease outbreak
surveillance system, there is also a national system of notifiable diseases
in the United States that mandates that health care providers report specific
infections, including a number of potentially waterborne infections such as
cholera, cryptosporidiosis, E. coli O157:H7, giardiasis, hepatitis A virus,
legionellosis, polio- myelitis, salmonellosis, shigellosis, tularemia, and typhoid
fever. Like the outbreak surveillance system, the surveillance for notifiable
diseases is a voluntary passive surveillance system with low sensitivity and
reporting delays. Finally,

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
the descriptive framework has been used in the Foodnet surveillance program t
o
assess occurrence of common gastroenteric illnesses in the population a
n
dgather
information on the prevalence of various risk factors for d iarrheal disease (such
as food consumption habits, water consumption habits, and recreational water
contact). Although the notifiable disease surveillance system and the Foodnet
program provide valuable data on disease occurrence, they provide no
information on what proportion of these diseases are related to drinking water.
Correlational or ecologic studies collect population level data on disease rates
and exposures and look for correlations. For example, bladder cancer r
a
t
e
sin cities
with chlorinated surface water can be compared to cities with chlorinated
groundwater to see if there may be a correlation between chlorination of
surface water, formation of DBPs, and bladder cancer. However, these studies
do not collect information on individual risk factors or confounders that may be
related to risk of disease, such as smoking. Correlational studies do not test a
formal hypothesis and are considered weaker than studies that collect
individual- level data. But they can provide valuable information for generating
hypothe- ses. Time-series studies are another example of correlational studies
and have been used to examine the relationship between changes in water
quality indicators (such as turbidity) and disease rates in the population served
by the water supply (such as emergency department visits for gastroenteritis)
(Schwartz et al., 1997). These studies have the advantage of comparing the
same population at different points in time (thus controlling for confounding) so
that only the variables that change are those that are being studied—i.e., water
quality and disease rates.
Analytical studies are those in which individual-level data are collected, and
the investigator tests a formal hypothesis about the association between e
x
p
o
-
sure and disease. Analytical studies can be experimental, such as a clinical t
r
i
a
l
where some households are given bottled water to drink and other h
o
u
s
e
h
o
l
d
sare asked
to drink tap water, and then disease rates between the two study groups
are compared to determine the risk of disease attributable to drinking
water. In these clinical trials, study participants are randomly assigned to
a study group in order to ensure that other potential risk factors for disease
are equally distributed among the study groups. An example of this design is
the study of Colford et al. (2002) in which home water purification devices
were installed in the homes of a test group of study participants and the control
group consisted of homes in which “sham” devices were installed. Both groups
kept health diaries to record symptoms of gastroenteritis and other health
effects. At the end of the observation period, incident rates of disease were
compared as a ratio, e.g., diarrhea episodes per person-year in the “exposed
group” (those with the sham device) divided by diarrhea episodes per person-
year in the “unexposed group” (those with additionalpurification).
Other analytical studies can be observational or natural experiments, where
the investigator examines disease rates over time in study groups that have dif-
ferent exposures. Observational studies can use a cohort design, case-control
design,or cross-sectionaldesign.In the cohort design,all study participants are

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
disease-free at the beginning of the study and disease rates over time are c
om
-
pared between study participants who are exposed to various risk factors v
s
.
those
who are not exposed. This design allows the consideration of multiple
health outcomes and can be either prospective or retrospective. The cohort de-
sign is useful for rare exposures because the study deliberately recruits a cohort
of individuals who are more likely to become exposed because of their occupa-
tion or geographic location. An example of this is the study of Frost et al.
(2005), who assessed the illness rate of cryptosporidiosis and the presence of
antibodies to Cryptosporidium in two populations (one exposed to surface water
and one to groundwater). They concluded that populations receiving surface-
derived water had higher antibody prevalence (but not higher illness rate) than
individuals receiving groundwater. Cohort studies are not well suited for rare
diseases because the purpose of this study design is to compare how frequently
the disease occurs in the exposed group vs. the unexposed group. If the disease
is rare, then a very large cohort must be recruited in order to make a meaningful
comparison.
Case-control studies are often used to study rare diseases and start with r
e
-
cruiting a group of individuals with the disease of interest (cases) and another
group of individuals without the disease (controls). The study individuals a
r
e then
queried about their past exposure to the specific risk factors of interest. I
na case-
control study, the measure of association is the “risk odds ratio” which
compares the odds of exposure to a specific risk factor among the cases to the
odds of exposure among the controls. In contrast to the cohort study, a
case- control study can look at only one health outcome but can examine
multiple risk factors. An example of the case-control design is the study of
Steinmaus et al. (2003) who examined associations of risk factors with bladder
cancer in the western U.S. This study found no association of bladder cancer
with daily arsenic ingestion in drinking water below 80 µg/day and found some
association in smokers at ingestions of greater than 200 µg/d of arsenic.
Cross-sectional studies are similar to ecologic studies in that exposure r
a
t
e
sand
disease rates are measured at the same time. However, cross-sectional studies
collect individual-level data whereas ecologic studies collect population- level
data. Seroprevalence surveys are a form of cross-sectional study where, for
example, prevalence of antibodies to Cryptosporidium can be measured in
populations served by different types of water supplies. The use of epidemiol-
ogical methods to study health risks associated with drinking water has been
reviewed by Savitz and Moe (1997).
Descriptive Studies of Endemic Waterborne Disease
The risk of endemic waterborne disease (sporadic cases) is difficult to e
s
t
i
-
mate, although various authors have made educated guesses. Bennett et al.
(1987) estimated that the incidence of waterborne disease in the United States
was 940,000 cases per year and resulted in 900 deaths.Although the purpose of

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
the study was to rank the importance of various disease categories (water ranked
next to the last, above zoonotic diseases) and to define the opportunities for pre-
vention, the study has been criticized as little more than an exercise in guess
work. Morris and Levin (1995) used incidence rates for enteric diseases and
prevalence rates for specific groups of pathogens detected in water to give wa -
terborne infectious disease estimates of 7.1 million mild infections, 560,000
cases of moderate or severe illness, and 1,200 deaths annually in the United
States. The authors concluded, however, that available data were inadequate to
refine the estimates.
Recent data on the incidence of diarrheal disease in the U.S. is a
v
a
i
l
a
b
l
e
from
the FoodNet population-based surveillance system (managed by the C
D
C
)
.The
disease estimates from the FoodNet system are based on telephone s
u
r
v
e
y
sthat used
random-digit-dialing and interviewed one individual per household to recall
their occurrence of diarrhea in the four weeks prior to the interview. As shown
in Table 3-4, the overall diarrhea prevalence rates from these surveys range from
5 to 11 percent, resulting in an estimated incidence of around 0.7 to
1.4 episodes/person/year. Diarrhea prevalence rates were consistently higher i
n
children under five years of age.
Other CDC estimates based on the FoodNet data and other sources suggest
that there are 211 million episodes of acute gastroenteritis in the United States
each year that result in over 900,000 hospitalizations and 6,000 deaths (Mead et
al., 1999). Mead et al. (1999) estimated the incidence of gastrointestinal illness
to be 0.79 episodes/person/year. These FoodNet data are valuable for providing
a measure of baseline diarrhea incidence in the U.S. population and the public
health and economic burden associated with diarrheal diseases in an industrial-
ized country. However, it is important to point out that these data offer no in -
formation on the proportion of diarrheal disease attributable to drinking water.
Furthermore, these data probably underestimate the total burden of acute gastro-
enteritis in the population because cases with only vomiting were not included in
the estimate (Imhoff, 2004), and vomiting is a common symptom for most gas -
troenteritis due to noroviruses and otherviral agents.
TABLE 3-4 Burden of Diarrheal Disease in the U.S. based on FoodNet Telephone S
u
r
v
e
y
Data
Year
No. of
States
Total #
respondents
in analysis
Overall prevalence
of acute diarrheal
illness in past four
weeks
Estimated
incidence of
episodes/person/
year
Diarrhea
prevalencein
children
< five years
NA = the authors did not report an estimate of the incidence rate.
SOURCES: Herikstatd et al. (2002); Imhof f et al. (2004) ; Hawkins et al. (2002) ; McMillan et al. (2004).
old
1996–1997 5 8,624 11% 1.4 10%
1998–1999 7 12,075 6% 0.72 9%
2000–2001 8 14,046 5% NA 9%
2002–2003 9 15,578 5% NA 9%

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
Analytical Epidemiological Studies
Determining the proportion of diarrheal disease that is attributable to w
a
t
e
r
contamination is best done through analytical, experimental epidemiological
studies. There have been four analytical epidemiological studies of acute gas -
troenteritis and drinking water systems relevant to distribution systems, all of
which focused on risks from microbiological agents.
Laval Studies
Payment et al. conducted two epidemiology studies (Payment et al,
1991; Payment et al, 1997) in a suburb of Montréal known as Laval that
examined the health of people who drank tap water and compared the group to
people receiving water treated by reverse osmosis to determine which group
had higher levels of gastrointestinal illness. In the 1991 study, reverse
osmosis units were installed in 299 households (1,206 persons), and another
307 households (1,202 persons) were followed as controls with no device
installed. Both groups were monitored for a 15-month period. Highly credible
gastrointestinal illness (HCGI) was defined as (1) vomiting or liquid diarrhea
with or without confinement to bed, consultation with a doctor, or
hospitalization, or (2) nausea or soft diarrhea combined with abdominal cramps
with or without absence from school or work, confinement to bed, consultation
with a doctor, or hospitalization. The water source for the study area was a
river that was contaminated by human sewage discharges, including
combined sewer overflows. The community had a single water treatment plant
with pre-disinfection, alum flocculation, rapid sand filtration, ozonation, and
final disinfection with chlorine or chlorine dioxide. The quality of the
finished water leaving the plant included an average of 0.6 mg/L total
chlorine and approximately 0.4 mg/L free chlorine, an average turbidity of
0.26 NTU, and no detection of indicator bacteria or human enteric viruses in
weekly samples (Payment et al., 1991). The overall incidence of highly credible
gastroenteritis was 0.66 episodes/person/year and was highest in children five
years of age and younger. The authors concluded that approximately 35 percent
of the self-reported gastrointestinal illnesses was attributed to tap water
consumption.
The 1997 study included groups receiving (1) regular tap water, (2) tap w
a
-ter
from a continuously purged tap, (3) bottled plant effluent water, or (4) bot-tled
plant effluent water purified by reverse osmosis. Differences in gastroen-
teritis rates between groups 1 and 2 versus group 3 was assumed to be due to
changes in water quality that occurred between the time the water left the treat-
ment plant and the time the water reached the household. The water ingested by
group 1 represented tap water that had gone through the distribution system and
also had residence time in the household plumbing. The water ingested by
group 2 represented tap water quality in the distribution system without any sig-
nificant residence time in the household plumbing. It should be noted that be-

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
tween the time of the first and second study, the water treatment plant was si
g-
nificantly upgraded with higher disinfection doses and better filtration. Esti-
mated Giardia removal/inactivation exceeded 7.4 logs, and estimated virus inac-
tivation by chlorine exceeded 10 logs. The average turbidity of the finished
water was 0.1 NTU and never exceeded 0.5 NTU. However, periods of “micro -
failures” in individual filters were reported (Susan Shaw, EPA, personal com-
munication, 2006).
This second study attributed 14 percent to 40 percent of the gastrointestinal
illness to the consumption of tap water (which met Canadian guidelines). Pay-
ment et al. (1997) concluded that the distribution system played a role in water-
borne disease because the rates of HCGI were similar for group 3 (ingested puri-
fied bottled water) and group 4 (ingested bottled water from the treatment plant),
but groups 1 and 2 (ingested water from the distribution system) had higher
HCGI rates than group 4. Interestingly, there appeared to be no correlation be-
tween the relatively short residence time of the water in the distribution system
(which varied from 0.3 to 34 hours) and the incidence of HCGI in a family.
Furthermore, microbiological testing of the water in the distribution system did
not indicate any bacterial indicators of contamination, but these water samples
were not tested for viruses or protozoa. Contrary to their expectation, the inves -
tigators observed higher HCGI rates in families that ingested water from the
continuously purged taps compared to families with regular tap water that may
be subject to bacterial regrowth in household pipes. The investigators suggested
that the shorter residence time for water from the continuously purged taps may
have transported pathogens in the distribution system to the household sooner
than regular tapwater and that there may have been inadequate contact time with
residual chlorine in the distribution system to inactivate any introduced patho-
gens.
Transient pressure modeling (Kirmeyer et al., 2001) found that the distribu-
tion system studied by Payment et al. was extremely prone to negative pressures,
with more than 90 percent of the nodes within the system drawing ne
ga
ti
ve
pressures under certain modeling scenarios (e.g., power outages). The system
reported some pipe breaks, particularly during the fall and winter when tempera -
ture changes placed added stresses on the distribution system. Although the
system employed state-of-the-art treatment, the distribution network suffered
from low disinfectant residuals, particularly at the ends of the system. Low dis -
infectant residuals and a vulnerability of the distribution systemto pressure tran-
sients (suggesting intrusion as a possible mechanism of contamination)
could account for the observed illnesses.
Melbourne Study
A double-blinded, randomized trial was recently completed in Melbourne,
Australia, to determine the contribution of drinking water to gastroenteritis (
H
e
l
-
lard
et al., 2001). Melbourne, with a population of about 3 million, draws i
t
s

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
drinking water from protected forest catchments of the Upper Yarra and Thom
-
son rivers. The catchments, which are approximately 1,550 square
kilometers (600 square miles) in area, are closed to public access and have no
permanent human habitation or activity except for logging in limited areas.
Water from these catchments is stored in two major reservoirs (Silvan and
Cardinia) with detention times of approximately two and 33 months,
respectively. Water from both reservoirs is treated by chlorination, fluoridation
(slurry or acid), and pH adjustment with lime.
Routine water quality monitoring at sampling points in the distribution s
y
s
-tem
included total and fecal coliforms, HPC bacteria, and total and free chlorine. Free
chlorine levels in the distribution system ranged from 0 to 0.94 mg/L, with a
median of 0.05 mg/L, and 90 percent of samples had < 0.20 mg/L. Total coli-
formbacteria were detected in 18.9 percent of 1,167 routine 100-mL water sam-
ples, but fecal coliform bacteria were not detected. Median HPC concentrations
were 37 CFU/mL with 13 percent of samples greater than 500 CFU/mL. During
the study, water quality monitoring included testing a weekly composite sample
from four water mains for selected pathogens: Campylobacter sp.,
Aeromonas sp., Clostridium perfringens, Cryptosporidium sp. and Giardia sp.
These distribution system samples were positive for Aeromonas spp. (50
percent of 68 weekly samples), Campylobacter (one occasion), and Giardia
(two positive samples by reverse transcriptase-PCR). No samples had
detectable C. perfringens spores or Cryptosporidiumparvumoocysts.
The study area in Melbourne is a growing area with relatively new houses
and many families with young children. Six hundred (600) families (with
at least two children one to 15 years of age) were recruited into the study.
Ap- proximately one third of the study households lived in areas of the
distribution system with average water residence times of one to 1.5 days.
Approximately two thirds of the study households lived in areas of the
distribution system with average water residence times of three to four days
(maximum six days).
Study households were randomly assigned to receive either a real or p
l
a
-
cebo water treatment unit installed under the kitchen sink. Functional units w
e
r
e
designed to remove viruses, bacteria, and protozoa using microfiltration a
nd
ultraviolet light treatment. The study participants completed a weekly health
diary reporting gastrointestinal symptoms during the 68-week observation pe-
riod. The rates of HCGI ranged from 0.79/person/year for those with functional
treatment units and 0.82/person/year with the sham devices. The study con-
cluded that the water was not a source of measurable gastrointestinal
disease (the ratio of illness rates between the group drinking treated water
compared to the normal tap water was 0.99, with a 95 percent confidence
interval of 0.85– 1.15; p = 0.85). Analysis of 795 fecal specimens from
participants with gastroenteritis did not reveal any difference in pathogen
detection rates between the two groups.
This study was not designed to examine the risks fromthe distribution s
y
s
-tem
separately from the risks associated with the entire water system. However, since
there appeared to be no measurable contribution to illness due to drinking

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
water, one may assume that the risks from degraded water quality in the
distribution systemwere also below the detection limit of the study.
Davenport Study
The 1996 amendment to the Safe Drinking Water Act included a mandate t
o
the CDC and the EPA to conduct studies to determine the occurrence of w
a
t
e
r
- borne
disease. To address this mandate, EPA scientists conducted several epi-
demiological studies of waterborne disease, and EPA funded several studies by
external investigators, including the pilot study and full-scale study in
Daven- port, Iowa.
As a preliminary trial to the subsequent epidemiology study, a randomized,
triple-blinded, home drinking water intervention trial of 77 households was con-
ducted for four months in Contra Costa County, California (Colford et a
l
.,
2002). The drinking water was treated using an under-the-kitchen-sink device
that incorporated ultraviolet light and microfiltration. Although the purpose of
the trial was to evaluate the “blinding” of the study (e.g., could the
participating households detect the active and identical-looking placebo
devices), analysis of the data showed that the incidence rate ratio of highly
credible gastrointestinal illness (HCGI) (incidence rate of the placebo group
divided by the active device group, adjusted for clustering) was 1.32, with a 95
percent confidence interval of
0.75 to 2.33. Given the small study size, the higher rate of HCGI among t
he
placebo group was not statistically significant. The authors concluded, however,
that the relative rates of HCGI were consistent with those observed by Payment
et al. (1991, 1997). This pilot study is interesting because it provides another
estimate of self-reported HCGI rates in a cohort of households followed over
time, and it confirmed that study subjects could successfully be blinded to the
type of water treatment device they had during the intervention trial.
The full-scale Water Evaluation Trial was conducted in Davenport, Iowa t
o
determine the incidence of gastrointestinal illness associated with consumption of
drinking water meeting all federal and state treatment guidelines (LeCheval- lier
et al., 2004; Colford et al., 2005). The municipal water systemused a si
ngl
esource
(the Mississippi River) and was treated at a single plant with c
o
n
ve
n- tional treatment
consisting of coagulation, flocculation, sedimentation, prechlorination,
filtration (dual filters with granular activated carbon and sand), and post-
filtration chloramination. The average turbidity of the finished water was
0.05 NTU.
A total of 456 households with 1,296 participants were randomized into two
groups. One group received a household water treatment device with a 1-
micron absolute ceramic filter and UV light with 35,000–38,000 uW-
second/cm2
output. The other group received a sham device that was identical to
the active device but had an empty filter chamber and a UV light that was
shielded to block the transmission of radiation but still generated the same light
and heat as the active unit. Each study household had an active device for six

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
months and a sham device for six months and was blinded to the status of t
he
i
r
device during the study. Study participants recorded the occurrence
of a
n
y
symptoms in daily health diaries. HCGI was defined as in the previous
studies as (1) vomiting, (2) watery diarrhea, (3) soft diarrhea and abdominal
cramps, or
(4) nausea and abdominal cramps.
Incidence of HCGI varied by season and ranged during the study
period from 1.64 to 2.80/person/years at risk (Wade et al., 2004). The overall
HCGI rate for households with the sham device was 2.12
episodes/person/yearand
2.20 episodes/person/year for households with the active device. The over
al
l HCGI
rate for the entire study population was 2.16 episodes/person/year. Mul-tivariate
analyses showed no effect of the household water treatment device on illness
rates during the 12-month study period. As in the studies by Payment et al., the
highest illness rates were in children five years of age and younger. The overall
conclusion was that less than 11 percent of the gastrointestinal illness observed
in this community was due to drinking water. Unlike the studies by Payment et
al., this study included households without children, and it is possible that the
number of young children in the study was too small to be able to detect an
effect in this more vulnerable group.
United KingdomStudy
A study conducted in Wales and northwest England from 2001 to 2
0
02found a
very strong association (p < 0.001) between self-reported diarrhea and reported
low water pressure at the home tap based on a postal survey of 423
subjects (Hunter et al., 2005). This study was part of a larger case-control study
of risk factors associated with sporadic cryptosporidiosis and was not specifi-
cally designed to study waterborne disease. Cryptosporidiosis cases and con-
trols were identified from family physician practices in Wales and northwest
England, and a postal survey asking a number of questions about potential risk
factors for diarrhea was mailed to 662 cases of cryptosporidiosis and 820 con-
trols. The survey included questions on travel outside the U.K., eating habits,
food preparation habits, contact with animals, contact with young children, con-
sumption of unboiled water, contact with other persons with diarrhea, and age.
Questionnaires were returned by 427 controls, and 423 were included in the
analyses. Of these, 28 (6.6 percent) reported having diarrhea in the two weeks
before receiving the survey.
Four risk factors for diarrhea in the control group remained significant in
the logistic regression model using a stepwise comparison strategy: feeding a
child under five years old, contact with another person who had diarrhea, loss of
water pressure at home, and how often the subject ate yogurt. The first three
risk factors had a positive association with diarrhea (Odds Ratios of 2.5, 7.0, and
12.5, respectively, after adjusting for the effects of the other variables in
the model). Yogurt consumption had a protective effect against diarrhea
and showed a dose-response relationship (more frequent consumption was
associ-

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
ated with lower risk). The investigators suggested that the strength of the asso-
ciation between loss of water pressure and risk of diarrhea indicates that this was
not a spurious association and was not likely to be affected by recall bias be-
cause it was just one of many potential risk factors that was investigated.
The study populations were drawn from two large regions that include bot
h
heavily industrialized areas and rural areas and about 240 water treatment plants.
The overall microbiological water quality for the utilities in these regions w
a
s
described to be excellent with less than 0.05 percent of water samples
positive for E. coli during this study period. The investigators hypothesized that
most of the reported episodes of pressure loss were due to main breaks in
which contamination entered the distribution system. However, no attempt
was made to collect information on recorded main breaks in the systems where
the controls lived. The investigators concluded that up to 15 percent of
gastrointestinal illness may be associated with consumption of drinking water
that was contaminated from main breaks or other pressure loss events,
and that the associated costs of this illness should be taken into account
when weighing the costs of replacing aging water supply distribution systems.
Although there had previously been concern about possible health risks from
pressure loss and pathogen intrusion in water distribution systems (LeChevallier
et al., 2003), this was the first study to provide solid evidence of that risk, with
policy implications for how to manage low pressure events in public water
supplies.
***
The body of evidence from these epidemiological studies does not e
l
i
m
i
n
a
t
e
consumption of tap water that has been in the distribution system from causi
ng
increased risk of gastrointestinal illness. The conflicting results between t
he
Laval and U.K. studies, which indicated risk associated with distribution system
water, versus the Melbourne and Davenport studies, which showed no increased
risk of gastrointestinal illness associated with tap water, may be due to a number
of differences between the study designs and the individual water systems.
With respect to the latter, all four cohort studies were in cities that used s
u
r
-face
water supplies. In Laval and Davenport, the rivers received upstream s
e
w
- age
discharges and were known to be contaminated. With the Davenport studyin
particular, it is possible that the reason they found no contribution to disease
from the water supply was because the investigators chose a well-operated and
maintained system. In Melbourne, the source water came from a highly
protected watershed. In Laval and Davenport, the water treatment plants used
conventional filtration and disinfection—indeed, Laval had both ozonation
and chlorination although the average turbidity of the finished water during the
first study was quite high (0.26 NTU). The water treatment plant in Melbourne
did not practice filtration. There is no information on the water supplies in the
U.K. study. Little to no information on the distribution systems was provided in
the descriptions of the Laval or Melbourne studies except that the residence time
in

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
the Laval systemwas relatively short (0.3 to 34 hours), while the residence t
i
m
e
for most of the study families in the Melbourne study was 72 to 96 hours.
Differences in study design such as population size and composition a
nd
follow-up period also played a role. As shown in Table 3-5, the size of the study
population in the Davenport study is approximately half of the study population
in the Laval and Melbourne studies (although the Davenport study used a cross-
over design to try to compensate for the smaller sample size). The Davenport
study also had the shortest follow-up period of the four studies. Unlike the La-
val and Melbourne studies that only recruited households with children, house-
holds enrolled in the Davenport study were not required to have children, and
the average household size was smaller in the Davenport study (2.84 persons)
compared to the Laval and Melbourne studies (Laval 1988–1989: 3.97 persons;
Laval 1993–1994: 3.84 persons; Melbourne: 4.69 persons). The smaller sample
size, shorter follow-up period, and possibly lower proportion of children (a vul-
nerable sub-population), may be reasons why the Davenport study did not detect
a significant risk of waterborne illness.
TABLE 3-5 Comparison of Population Parameters fromthe Epidemiology Studies
Study Laval
1988-1989
Laval
1993-1994
Melbourne
1997-1999
Davenport
2000-2002
# households in
tapw ater
group
307 346 (tap w ater)
330 (tap w
/
v
a
l
v
e
)
300 229
# of persons in 1,202 1,296 (tap w ater) 1,399 650
tapw ater group 1,300 (tap
w /valve)
# households i
n 299 339 (purified) 300 227
purified w
ater
group
354 (bottled pl
ant)
# of people in 1,206 1,360 (purified) 1,412 646
purified w
ater
group
1,297
(bottled p
l
a
n
t
)
% children in 6.2 <6 yrs 12.8 <6 yrs (tap) 40.2 < 10 NA
tapw ater group 16.5 <6 yrs (tap yrs
valve)
% children in 9.6 <6 yrs 15.1 <6 yrs 40.9 < 10 NA
purified w
ater
group
(purified)
15.4 <6 yrs
yrs
(bottled plant)
Weeks of Approx 60 Approx 69 68 54
observation time

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
Statistical power in a cohort study is determined by the size of the st
udy
population, the follow-up time, and the frequency of the health outcome of in-
terest (incidence of HCGI), with the number of outcomes being more relevant
that the size of the study population (Hulley and Cummings, 1988). The Dav-
enport study was designed to have the statistical power to detect an 11 percent or
greater risk of HCGI due to water (Colford et al., 2005). The Melbourne study,
with the larger sample size and longer follow-up period, was designed to detect
a 15–20 percent reduction in the overall rate of HCGI in the group with the ac-
tive point-of-use treatment devices. However, the total number of HCGI epi-
sodes measured in both study populations was very similar (tap water:
Mel- bourne = 1,500 episodes, Davenport = 1,431 episodes; purified water:
Mel- bourne = 1,459 episodes, Davenport = 1,476 episodes). Thus, the higher
HCGI rates detected in the Davenport study and the cross-over design appear to
have mitigated the effects of the smaller sample size and shorter follow-up
period on the statistical power of the study. As shown in Table 3-5, all of
these studies had relatively large study populations and measured thousands
of illness episodes,and thus had similar statisticalpower.
There was limited assessment of exposure among the studies. All of t
he
studies monitored water quality at the treatment plant, but there was a wide
range in the amount of sampling and analyses of water in the distribution sys -
tem. For example, monitoring in the Davenport study was extensive, with tap
water samples and treatment device samples collected from about one-fourth of
the study households at three times during the study. They documented higher
coliform and HPC levels in water from the treatment devices compared to tap
water (LeChevallier et al., 2002). None of the studies reported pathogen detec-
tion in the tap water, except for three occasions in the Melbourne study. It
should be noted that the microbiological analyses of water differed for each
study. Finally, all four studies attempted to measure the volume of tap water
ingested via surveys, and these surveys indicated that subjects in the
purified water groups also consumed regular tap water (reported range 14.5 to
40 percent).
All four cohort studies used similar approaches for recording symptoms o
f
gastrointestinal illness and similar definitions of HCGI. Different rates of HCGI
were observed in the four cohort studies. It is striking that the rates reported by
the Davenport study and the Contra Costa County pilot study are more than
twice as high as the rates reported by the Laval and Melbourne studies and about
three times higher than the FoodNet rates of diarrheal disease (see Table 3-6).
The reason for these higher rates is unknown because the investigators state that
they used similar case definitions as the Laval and Melbourne studies. If there
were several significant transmission routes of enteric pathogens in these com-
munities that were responsible for these higher reported illness rates, then an
intervention study targeted only to waterborne disease transmission may not
show any effect (see Briscoe, 1984). However, the use of the cross-over design
in Davenport should have been valuable in this regard because the effect
of other transmission routes is better controlled for using this design.

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
TABLE 3-6 Rates of Highly Credible GastrointestinalIllness fromthe Epidemiology S
t
u
d
i
e
s
Study
Laval
Estimated rate* of
HCGI in tap water
groups
Estimated rate* of
HCGI in purified
water group
Estimated rate* of
HCGI in all study
participants
1988-1989 0.76 0.50 0.66
Laval
1993-1994
0.66 (tap)
0.70 (tap valve)
0.58 0.60
Melbourne 0.82 0.79 0.80
1997-1999
Contra Costa 3.48 2.63 3.05
County, CA
1999
Davenport 2.12 2.20 2.16
2000-2002
FoodNet ND ND Approx 0.72
* rate expressed as episodes/person/y ear
The conflicting results of these epidemiological studies raise a number
o
f
questions. The fact that these were carefully conducted studies by research
teams with considerable experience implies that there are detectable elevated
risks of waterborne disease associated with some water systems and not others.
However, not enough information was gathered to know what characteristics of
the water systems posed increased risk, whether it be the source water, the
treatment plant, or the distribution system.
For the studies that showed no detectable association between gastrointesti-
nal symptoms and consumption of tap water (Melbourne and Davenport), it is
not clear if they suffered from an inadequate design and sample size in order to
detect an association, or if there simply was no association. The randomized
clinical trial design used in Laval, Melbourne, and Davenport is one of the most
rigorous analytical study designs and is less likely to be affected by error and
confounding. However, it is possible that selection bias in the recruitment of the
study population, misclassification of drinking water exposure, or inaccurate
reporting of health outcomes may have affected the results of these studies. It
must be kept in mind that epidemiological studies are not able to prove that there
is zero risk associated with a specific exposure; they can only report that the risk
is below the level that the study had the power to detect, which was 15 to
20 percent (Melbourne) or 11 percent (Davenport).
For the studies that did show an association between gastrointestinal symp-
toms and consumption of tap water (Laval study), or an association
between gastrointestinal symptoms and a water pressure drop (UK study), it is
not clear what portion of the observed risk was due to water contamination in
the distribution system as opposed to water contamination at the source and/or
inadequate

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
water treatment. The second Laval study examined the risks associated with t
h
e
distribution systemby including a study group that received bottled plant e
f
f
l
u
-ent as
well as groups that ingested tap water and continuous-flow tap water (“tap
valve” group). Tap water drinkers had elevated risk of HCGI compared to those
who ingested bottled water from the treatment plant or purified bottled
water, suggesting that water in the distribution system posed an increased
health risk (although routine water quality monitoring of the distribution system
did not provide evidence of compromised quality). However, there was also an
indication of some increased risk of illness from water with reduced residence
time in the distribution system (tap valve group) compared to water with
average residence times (from 0.3 to 34 hours in this system). This suggests
that additional contact time with disfinectants in the distribution system may be
helpful in reducing risks. The UK study suggests that pressure drops in the
distribution system was associated with increased gastrointestinal illness, but
this association needs to be tested more systematically and rigorously in
furtherstudies.
One of the major challenges for designing an epidemiology study of heal
th
risks associated with water quality in the distribution system is separating t
he
effect of source water quality and treatment from the effect of distribution sys-
tem water quality. Knowledge of how water distribution systems become con-
taminated from anecdotal evidence and outbreak data (main breaks, sudden
changes in pressure and intrusion, backpressure or backsiphonage, etc.) suggests
that the exposure to contamination in the distribution system is likely to be in-
termittent and may be very difficult to capture in an epidemiological
study. Nonetheless, new approaches to deal with this challenge were tested in a
pilot study in the southeastern U.S. and a third approach is being tested in a
study in the Midwestern U.S. These studies were designed by
multidisciplinary teams of university and research foundation scientists with
input from outside experts including EPA and CDC staff. Support for
these studies came from the EPA STAR Grant Program, and they are part of
a series of studies funded by or conducted by the EPA to develop a
national estimate of waterborne disease risks. These three approaches are
described in Box 3-3 as examples. Other study designs may also be useful for
addressing the question of endemic disease risks associated with water quality
in the distribution system.
RISKS FROM LEGIONELLA
The role of biofilms and microbial risk can best be illustrated by the ex
am-
ple of the bacterium Legionella pneumophila in water systems, for which occur-
rence data, outbreak data, and epidemiological data are available. Legionella are
widely distributed in the aqueous environment and have been found in drinking
water (Stout et al., 1985; Rogers et al., 1994) and biofilms (Rogers et al., 1994;
Pryor et al., 2004; Thomas et al., 2006). Although the bacteria have been iso-
lated from biofilms in water distribution systems,there is evidence that the

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
Tap Water HH
vs.
BOX 3-3
Three Approaches to Designingan Ideal Epidemiology Study thatwould Determine
the Distribution System Component to Waterborne Disease
Method 1
This method relies on conducting a vulnerability assessment of t
hewater distribution
system and identifying areas in the distribution system that are more vulnerable and less
vulnerable to contamination—based on pipe age and composition, history of main
breaks, history of coliform detections, estimates of residence time, and chlorine residual.
The study population (families w ith one or more children < six years old) should be
recruited in the most vulnerable and the least vulnerable geographic areas of the
distribution system. It is important to randomize the study population in each geographic
area into tw o groups. The researchers would provide purified bottled w ater to half of the
study households, and ask the other half of the study population to drink tap w ater. All
study households would be asked to record health symptoms in a health diary. The
difference in the rates of reported gastrointestinal symptoms (GI) for families drinking tap
w ater to the rates for families drink-
ing purified bottled water would then be compared. This difference (
G
I
t
a
p
-
G
I
b
o
t
t
l
e
)represents the risk of
GI symptoms due to source water and distribution system w ater. Part of the analysis
would be to compare this difference (GItap-GIbottle) for the study populations in the most
vulnerable areas (where the degradation of distribution system water quality would be the
greatest) to the difference (GItap-GIbottle) for the study populations in the least vulnerable
areas (where there should be little or no impact from degradation of w ater quality in the
distribution system). This difference between the study groups in different parts of the dis-
tribution system should represent the impact of the distribution system on risk of GI illness
(see Figure 3-5). Although the study is not blinded, the technique of “comparing the differ-
ence of the difference” controls for lack of blinding. This “double-difference methodology” is
commonly used in economics studies and program evaluation to assess the impact of a
specific intervention by comparing the differences between intervention and control groups
at baseline and at a follow -up time point (Maluccio and Flores, 2005).
FIGURE 3-5 Study Design to Examine Risks from Water Quality in the D
is
t
r
ib
u
t
ionSys- tem: M ethod
1.
Water Treatment P
l
a
n
t
A
Least vulnerable d
i
s
t
r
i
b
ution
system areas
B
Most vulnerable d
i
s
t
r
i
b
ution
system areas
Bottled water HH Bottled water HH
vs.
Tap Water HH
A vs. B

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
Method 2
This approach is identical to the first, except that the study p
o
p
u
lationin each geographic
area is random ized into three groups. The researchers would provide purified bottled
water to one-third of the study households, bottled finished water directly from the
treatment plant to the other third of the study households, and bottled w ater from the most
vulnerable part of the distribution system to the final third of the study population. As be-
fore, study households would be asked to record health symptoms in a health diary. This
study, which has a cross-over design, is shown in Figure 3-6. The advantage of
this approach over the first approach is that the study is blinded because everyone
receives bottled water. Furthermore, one can recruit study subjects in any geographic
location because drinking w ater is delivered to their home. This design is similar to a
human challenge study because the investigators control exposure to the study water. The
disadvantages are that bottled distribution system water will not capture temporal changes
in w ater quality. Also, possible changes in water quality during bottling and storage may not
reflect quality of distribution system w ater. How ever, these disadvantages could be
mitigated by detailed microbiological studies of distribution system w ater quality in the
study site prior to starting the epidemiologic study, bottling the distribution system water
more frequently, bottling composite samples of the distribution system w ater over time
and geographic area, and characterizing changes in distribution system w ater quality
during bottling and storage.
A B C D
First 6
m
onths
Last 6
m
onths
176 households per group (average); Total704
Power estimates:
GI risk due to source w ater quality and treatment
efficacy: 91% GI riskdue to distribution system: 86%
Assumptions:
20% attrition
20% variance inflation due to clustering
FIGURE 3-6 Study Design to Examine Risks fromWater Quality in the D
is
t
r
ibution System:
Method 2 Cross-over Study.
continues
Purified Bottled
Water
Purified Bottled
Water
Bottled Plant
Water
Bottled DS
Water
Bottled Plant
Water
Bottled DS
Water
Purified Bottled
Water
Purified Bottled
Water

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
organism must be taken up by protozoa to proliferate (Nahapetian et al., 1991;
Barbaree et al., 1986; Barbaree, 1991; Murga et al., 2001). Some studies have
reported that the presence of amoebae is a predictor of Legionella colonization
in plumbing systems (Moore et al., 2006).
Levels of legionellae in potable water systems are typically low, but ampli-
fication can occur in cooling towers, recirculating hot water systems, and
hot tubs (EPA, 1999). Legionella species have been shown to proliferate in
biofilms in institutional and premise plumbing (Pryor et al., 2004; Thomas et
al., 2006) and can be found in water heaters, shower heads, and cooling
towers (Wad- owsky and Yee, 1983, 1985; Stout et al., 1985; Rogers et al.,
1994). Indeed, in a study of legionellosis in the United Kingdom, 528 of the
examined 604 cases were attributed to contaminated cooling towers, 70 (or 12
percent) were caused by contaminated drinking water, and six were caused by
contaminated whirl- pools (VROM, 2005).
BOX 3-3 Continued
Method 3
A third approach is being attempted in the Wisconsin groundwater study (WAHTER) in
several communities that use untreated groundwater. This study uses a community level
intervention where UV disinfection is added at the wellhead, and community gastrointesti-
nal symptom rates are compared before and after the UV intervention. The risk from
the distribution system w ill be estimated using a risk assessment approach. Enteric virus
concentrations are being measured in water samples from w ell heads (representing
contamination in the groundwater) and compared to virus concentration measurements in
water samples from study households (representing contamination from both the
groundwater and the distribution system). The difference in virus concentration will be
attributed to the distribution system. In those study communities w ith UV disinfection
installed at the w ellheads, viruses measured at the households could only have originated
from intrusions into the distribution system. Note that the feasibility of this approach
depends on studying a water supply where pathogens are detected with some
frequency. For a water supply where a high proportion of water samples do not have
detectable pathogens, the application of this study design is uncertain.
The study also measures the incidence of gastrointestinal s
y
m
p
tom sin a cohort of children
in the study com munities using a health diary. The researchers intend to model the
illness rate in the study population as a function of household pathogen concentration using
dose-response models where incidence of acute gastrointestinal illness in the study
population is a function of the pathogen dose in the household water (calculated as con-
centration of virus in the volume of water ingested over a defined period of time). The in-
vestigators w ill then use quantitative risk assessment to estimate the community illness
rates if the population drank water directly from the wellhead. The difference between the
measured illness rates in the study population and the estimated illness rates
associated w ith source water willrepresent the riskfrompathogens in distribution system.
One of the challenges of this approach is that there are d
i
f
f
e
r
e
n
t dose-response rela-
tionships for different waterborne viruses. Thus, information on the etiology of the pre-
dominant viralinfections in the community w illbe used to guide the modeling analyses.
SOURCE: Available online at http://cfpub.epa.gov/ncer_abstracts/index.cfm/fuseaction/
display.abstractDetail/abstract/7430/report/0. Accessed August 10, 2006.

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
Legionella is an example of an organism that is an efficient pulmonary
pathogen when inhaled as large aggregates or biofilm fragments. Inhalation of
large numbers of the bacteria overwhelms the pulmonary defenses, and Pontiac
fever results. Aspiration of smaller numbers of organisms as biofilm fragments
may cause Legionnaire’s disease. Epidemiological studies have linked water
contaminated with both Legionella and protozoa to outbreaks of legionellosis
(Fields et al., 1989; Breiman et al., 1990). A review paper by Lin et al. (1998)
suggests that hospitals take routine samples for the organism in their distribution
systems and determine the efficacy of any disinfection processes by measuring a
reduction in Legionella counts.
Legionella are specifically mentioned in the EPA’s Surface Water Treat-
ment Rule, with the MCLG set at zero. For this reason, the bacterium was not
included on the Contaminant Candidate List for methods development and po-
tential future regulation. However, there is little evidence that filtration and
disinfection of surface water prevents the growth of Legionella species in distri-
bution system plumbing. In fact, since Legionella was incorporated into the
waterborne disease outbreak surveillance system starting in 2001, several
outbreaks have been attributed to the microorganism. During 2001–2002, the
six drinking water outbreaks attributed to Legionella species (19.4 percent
of the total) caused illness in 80 persons and resulted in 41 hospitalizations and
four deaths. All of these outbreaks occurred in large buildings or institutional
settings and were related to multiplication of Legionella species in the
respective distribution systems. As mentioned previously, Legionella is now
the single most common cause of outbreaks involving drinking water (Liang et
al., 2006). These outbreaks underscore the importance of remaining vigilant
about the possibility of growth of Legionella species in building complexes
and the need to take measures to reduce this threat (see Chapter 8).
In an epidemiological study, Kool and colleagues (1999)
examined 3
2
nosocomial outbreaks of Legionnaires’ disease from 1979 to 1997
where drinking-water was implicated and tabulated the characteristics of the
hospital (size, transplant program) and the primary disinfectant treatment,
disinfectant residual, water source, community size, and pH of the water. The
researchers found that the odds of a nosocomial Legionella outbreak was 10.2
(95 percent confidence interval of 1.4–460) times higher in systems that
maintained free chlorine than in those using a chloramine residual. They
estimated that 90 percent of waterborne Legionella outbreaks could be
prevented if chloramine was universally used. Heffelfinger et al. (2003)
reported that 25 percent (38) of 152 hospitals surveyed had definite reported
cases or outbreaks of hospital-acquired Legion- naires’ disease during the
period 1989 to 1998. However, hospitals supplied with drinking water
disinfected with monochloramine were less likely (odds ratio 0.20; 95
percent confidence interval, 0.07 to 0.56) to have hospital- acquired
Legionnaires’ disease than other hospitals. Cunliffe (1990) reported that
suspensions of Legionella pneumophila were more sensitive to
monochloramine disinfection, with a 99 percent level of inactivation when ex-
posed to 1.0 mg monochloramine/L for 15 minutes, compared with the 37-

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
minute contact time required for Escherichia coli inactivation under s
i
m
i
l
a
r
conditions. Donlan et al. (2002) reported that monochloramine was s
i
g
n
i
f
i
c
a
n
t
l
ymore
effective than free chlorine at eradicating laboratory-grown biofilms of L.
pneumophila.
Legionella has also been the subject of pathogen occurrence measurements.
Researchers at the CDC conducted a study of Legionella occurrence in 53 p
u
b
l
i
c
buildings before and after the conversion of the San Francisco water s
u
p
p
l
y
from free chlorine to chloramine (Fields, 2005; Flannery et al., 2006). T
h
e
y
showed that the concentration of legionellae was reduced more than 20-fold b
ythe
conversion from free chlorine to chloramine. Interestingly, the i
nci
dencerate
of Legionella infections was low (only one laboratory-confirmed case in the two
years prior to the switch to chloramine) despite the fact that the major sero- type
detected included the clinically significant Legionella pneumophila sero- group
1. The results illustrate the difficulty in relating the detection of microbes in
drinking water to a documented risk of waterborne disease.
Another recent study examined the impact of switching from chlorine to
monochloramine disinfection on Legionella occurrence in Pinellas County, Flor-
ida (Moore et al., 2006). In this study, water samples were collected from 96
buildings (public buildings and individual homes) for a four-month period when
chlorine was the primary disinfectant and from the same sampling sites for a
four-month period after monochloramine was introduced into the municipal wa -
ter system. In the first period, 20 percent of the buildings were colonized with
Legionella in at least one sampling site. Legionella colonization was reduced by
69 percent within a month after monochloramine introduction. Monochloramine
appeared to be more effective in reducing Legionella in hotels and single-family
homes than in county government buildings, perhaps because of more consistent
water usage. As in the San Francisco study, the reported incidence of legionel-
losis in the study area during this time was too low (nine cases) to determine if
the change to monochloramine had an impact on human disease.
Given that 20 percent of reported outbreaks involving drinking water are a
t
-
tributed to Legionella, additional attention should be given to the control of thi
s
potential pathogen, especially in institutional and premise plumbing (see Chap-
ter 8).
Accurate estimates are not yet available for the prevalence of adverse health
effects attributable to deficiencies in distribution systems from pathogen occur-
rence measurements, waterborne disease outbreak surveillance, or epidemiol-
ogical studies. Pathogen occurrence measurements are rare due to limitations in
detection methods and cost issues. Models to quantitatively predict pathogen
occurrence in distribution systems (e.g., by cross-connections, main breaks,
or intrusion) have not yet been developed. Despite under-reporting and limited
data on risk factors, the voluntary waterborne disease outbreak surveillance sys-

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
tem provides the best available evidence of public health risks associated w
i
t
h
distribution systems in the United States. These data suggest that about o
n
e
-
third to one-half of reported waterborne disease outbreaks are associated w
i
t
h
distribution system problems. To date, only one epidemiological study (the sec-
ond Laval study) has been specifically designed to examine the contribution of
the distribution system to endemic disease occurrence. Until better data are
available fromthese three approaches, it will not be possible to accurately assess
the magnitude of the health impacts resulting from distribution system deficien-
cies. The following conclusions and recommendations are made.
The distribution system is the remaining component of public water
supplies yet to be adequately addressed in national efforts to eradicate wa-
terborne disease. This is evident from data indicating that although the number
of waterborne disease outbreaks including those attributable to distribution sys-
tems is decreasing, the proportion of outbreaks attributable to distribution sys-
tems is increasing. Most of the reported outbreaks associated with distribution
systems have involved contamination from cross-connections and backsipho-
nage. Furthermore, Legionella appears to be a continuing risk and is the single
most common etiologic agent associated with outbreaks involving drinking wa-
ter. Initial studies suggest that the use of chloramine as a residual disinfectant
may reduce the occurrence of Legionella, but additional research is necessary
to determine the relationship between disinfectant usage and the risks of
Legionella and other pathogenic microorganisms.
Distribution system ecology is poorly understood. There is very little in-
formation available about the types, activities, and distribution of
microorganisms in distribution systems. Limited HPC data are available for
some systems, but these data are not routinely collected, they underestimate
the numbers of organisms present, and they include many organisms that do
not necessarily present a health risk. To more adequately assess risk, more
information on the microbial ecology of distribution systems, including
premise plumbing, is needed.
There is inadequate investigation of waterborne disease outbreaks as-
sociated with distribution systems, especially in premise plumbing. Le-
gionella has only recently been added to the outbreak surveillance system. Ex-
isting data on outbreaks due to other etiologic agents would rarely implicate
premise plumbing because backflow and regrowth events likely would not be
recognized and reported unless an institutional building with large numbers of
people was affected. The Centers for Disease Control and Prevention are com-
mended for revising the format used to report waterborne disease outbreaks
to the surveillance system such that outbreaks arising from events in premise
plumbing are now more clearly identified.

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
Epidemiology studies that specifically target the distribution system
component of waterborne disease are needed. Recently completed epidemi-
ological studies have either not focused on the specific contribution of distribu-
tion system contamination to gastrointestinal illness, or they have been unable t
o
detect any link between illness and drinking water. Epidemiological studies o
fthe
risk of endemic disease associated with drinking water distribution systems need
to be performed and must be designed with sufficient power and resources to
adequately address the deficiencies of previous studies.
This chapter highlights the lack of information available to assess the public
health risk of contaminated distribution systems. One of the consequences of
this fact is that the committee was forced to rely heavily on its best professional
judgment to prioritize contamination events into high, medium, and low priority
(see Appendix A). Better public health data, including data on waterborne out-
breaks, from epidemiological studies, and on distribution system water quality,
could help refine distribution system risks and provide additional justification
for the rankings.
The following three chapters consider the roles of physical, hydraulic, a
n
d
water quality integrity. Protection of public health requires that water
professionals incorporate approaches that combine all three into a
comprehensive program of best practices to maintain the highest level of water
quality.
REFERENCES
Abbaszadegan, M., M. LeChevallier, and C. Gerba. 2003. Occurrence of viruses in U
.
S
.
groundwaters. J Amer. Water Works Assoc. 95(9):107–120.
Alavandi, S. V., M. S. Subashini, and S. Ananthan. 1999. Occurrence of haemolytic a
n
d
cytotoxic Aeromonas species in domestic water supplies in Chennai. Indian J. M
e
d
.Res.
110:50–55.
Amann, R. I., B. J. Binder, R. J. Olson, S. W. Chisholm, R. Devereux, and D. A. S
ta
h
l
.
1990. Combination of 16S rRNA-targeted oligonucleotide probes with flow c
y
-
tometry for analyzing mixed microbial populations. Appl. Environ. Microbiol.
56:1919–1925.
Amann, R. I., W. Ludwig, and K.-H. Schleifer. 1995. Phylogenetic identification and i
n
situ
detection of individual microbial cells without cultivation. Microbiol. Rev.
59:143–169.
American Water Works Association (AWWA). 1999. Waterborne pathogens: manual o
f
water supply practices—M48, First Edition. Denver, CO: AWWA.
Anand, S., and K. Hanson. 1997. Disability-adjusted life years: a critical review.
J. Health Econ. 16(6):685–702.
Arbuckle, T. E., S. E. Hrudey, S. W. Krasner, J. R. Nuckols, S. D. Richardson, P. Singer,
P. Mendola, L. Dodds, C. Weisel, D. L. Ashley, K. L. Froese, R. A. Pegram, I. R
.
Schultz, J. Reif, A. M. Bachand, F. M. Benoit, M. Lynberg, C. Poole, and K. Waller.
2002. Assessing exposure in epidemiologic studies to disinfection by-products in
drinking water: Report from an international workshop. Environ. Health Perspect.
110 (Suppl1):53–60.

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
Balbus, J., R. Parkin, and M. Embrey. 2000. Susceptibility in microbial risk a
s
s
e
s
s
m
e
n
t
:definitions
and research needs. Environmental Health Perspectives 108(9):901–905. Barbaree, J.
M. 1991. Controlling Legionella in cooling towers. ASHRAE J
o
u
r
n
a
l
33:38–42.
Barbaree, J. M., B. S. Fields, J. C. Feeley, G. W. Forman, and W. T. Martin. 1986. I
s
o
-
lation of protozoa from water associated with a legionellosis outbreak and
demonstration of intracellular multiplication of Legionella pneumophila. Appl.
Environ. Microbiol. 51:422–424.
Bartram, J., J. Cotruvo, M. Exner, C. Fricker, and A. Glasmacher (eds). 2003. H
e
t
e
r
o
-trophic
Plate Counts and Drinking-water Safety. London: IWAPublishing.
Barwick, R. S., D. A. Levy, G. F. Craun, M. J. Beach, and R. L. Calderon. 2000. S
ur-
veillance for waterborne-disease outbreaks—United States, 1997–1998. MMWR
49(SS-4):1–34.
Bays, L. R., N. P. Burman, and W. M. Lavis. 1970. Taste and odour in water s
u
p
p
l
i
e
sin Great
Britain: a survey of the present position and problems for the future. J
o
u
r
n
a
lSociety Water
Treatment Examination 19:136–160.
Bellen, G. E., S. H. Abrishami, P. M. Colucci, and C. Tremel. 1993. Methods for a
s
s
e
s
s
-ing the
biological growth support potential of water contact materials. Denver, CO:
AwwaRF.
Bennett, J. V., S. D. Holmberg, M. F. Rogers, and S. L. Soloman. 1987. Infectious a
n
d
parasitic diseases. In: Closing the Gap: The Burden of Unnecessary Illness. R. W
.
Amler and H. B. Dull (eds.). New York: Oxford University Press.
45.
Borchardt, M. A., N. L. Haas, and R. J. Hunt. 2004. Vulnerability of d
r
i
n
k
i
n
g
-
w
a
t
e
rwells in La
Crosse, Wisconsin, to enteric-virus contamination from surface water contributions.
ApplEnviron Microbiol. 70(10):5937–5946.
Boorman, G. A., V. Dellarco, J. K. Dunnick, R. E. Chapin, S. Hunter, F. Hauchman, H
.
Gardner, M. Cox, and R.E. Sills. 1999. Drinking water disinfection by-
products: review and approach to toxicity evaluation. Environ. Health Perspect.
107:207–17.
Breiman, R. F., B. S. Fields, G. N. Sauden, L. Volmer, A. Meier, and J. S. Spilka. 1
9
9
0
.
Association of shower use with Legionnaires' disease. J. Amer. Med. A
s
s
o
c
. 263:2924–
2926.
Briganti, L. A., and S. C. Wacker. 1995. Fatty acid profiling and the
identification of environmental bacteria for drinking water utilities. Denver,
CO:AWWA.
Briscoe, J. 1984. Intervention studies and the definition of dominant transmission
routes. Am. J. Epidemiol. 120:449–55.
Camper, A. K. 1996. Factors Influencing Microbial Growth in the Distribution System:
Laboratory and Pilot Experiments. Denver, CO: AwwaRF.
Cantor, K. P., C. F. Lynch, M. E. Hildesheim, M. Dosemeci, J. Lubin, M. Alavanja, and
G. Craun. 1998. Drinking water source and chlorination byproducts—I. R
i
s
kof
bladder cancer. Epidemiology 9:21–28.
Carnahan, A. M., and M. Altwegg. 1996. Taxonomy. In: The genus Aeromonas. S
.
Joseph (ed.). New York: John Wiley & Sons.
Carson, L. A., L. A. Bland, L. B. Cusick, M. S. Favero, G. A. Bolan, A. L. Reingold, and
R. A. Good. 1988. Prevalence of nontuberculous mycobacteria in water supplies
o
f
hemodialysis centers. Appl. Environ. Microbiol. 54:3122−3125.

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
Chaidez, C., and C. Gerba. 2004. Comparison of the microbiological quality of p
o
i
n
t
-
o
f
-use
(POU)-treated water and tap water. Int. J. Environ. Health Res. 14:253–261.
Characklis, W. G. 1988. Bacterial Regrowth in Distribution Systems. Denver, C
O
:AWWA
and AwwaRF.
Characklis, W. G. and K. C. Marshal. 1990. Biofilms. New York: John Wiley & S
o
n
s
, Inc.
Chauret, C., C. Volk, R. Creason, J. Jarosh, J. Robinson, and C. Warnes. 2001. D
e
t
e
c
-tion
of Aeromonas hydrophila in a drinking-water distribution system: a field and pilot
study. Can. J. Microbiol. 47:782–786.
Clark, R. M., E. E. Geldreich, K. R. Fox, E. W. Rice, C. H. Johnson, J. A. Goodrich, J. A.
Barnick, and F. Abdesaken. 1996. Tracking a Salmonella serovar typhimurium out-
break in Gideon, Missouri: role of contaminant propagation modeling. Journal of
Water Supply Research and Technology - Aqua. 45(4):171–183.
Colford, J. M., Jr., T. J. Wade, S. K. Sandhu1, C. C. Wright, S. Lee, S. Shaw, K. F
o
x
, S.
Burns, A. Benker, M. A. Brookhart, M. van der Laan, and D. A. Levy. 2005. A
randomized, controlled trial of in-home drinking water intervention to reduce gastro-
intestinal Illness. Am. J. Epidemiol. 161(5):472–482.
Colford, J. M., J. R. Rees, T. J. Wade, A. Khalakdina, J. F. Hilton, I. J. Ergas, S. Burn
s
,A,
Benker, C. Ma, C. Bowen, D. C. Mills, D. J. Vugia, D. D. Juranek, and D. A
.Levy.
2002. Participant blinding and gastrointestinal illness in a randomized, control trial
of an in-home drinking water intervention. Emerging Infectious Diseases 8(1):29–
36.
Colbourne, J. S., D. J. Pratt, M. G. Smith, S. P. Fisher-Hoch, and D. Harper. 1984. W
a
-ter
fittings as sources of Legionella pneumophila in hospital plumbing system. L
a
n
-cet
1(8370):210–213
Collins, C. H., J. M. Grange, and M. D.Yates. 1984. Mycobacteriain water. J. Appl.
Bacteriol. 57:193–211.
Covert, T. C., M. R. Rodgers, A. L. Reyes, and G. N. Stelma. 1999. Occurrence of n
o
n
-
tuberculous mycobacteria in environmental samples. Appl. Environ. Microbiol.
65(6):2492–2496.
bution systemdeficiencies. J. Amer. Water Works Assoc. 93(9):64–75.
Craun, G. F., ed. 1986. Waterborne Diseases in the United States. Boca Raton, FL: C
R
C
Press,
Inc.
Cunliffe, D. A. 1990. Inactivation of Legionella pneumophila by monochloramine. J
.Appl.
Bacteriol. 68(5):453–459.
De Victoria, J., and M. Galvan. 2001. Pseudomonas aeruginosa as an indicator of health
risk in water for human consumption. Water Sci. Technol. 43:49–52.
Dodds, L., W. King, C. Woolcott, and J. Pole. 1999. Trihalomethanes in public w
a
t
e
r
supplies and adverse birth outcomes. Epidemiology 10:233–237.
Donlan, R., R. Murga, J. Carpenter, E. Brown, R. Besser, and B. Fields. 2
0
0
2
.
Monochloramine disinfection of biofilm-associated Legionella pneumophila in a po-
table water model system. Pp. 406–410 In: Legionella. R. Marre, Y. Abu Kwaik,
C. Bartlett, N. P. Cianciotto, B. S. Fields, M. Frosch, J. Hacker, and P. C. L
uck
(eds.).
Washington, DC: American Society for Microbiology.
Dott, W., and D. Waschko-Dransmann. 1981. Occurrence and significance of actinomy-
cetes in drinking water. Zbl. Bakt. Hyg. I., Abt. Orig. B, 173:217–232.
Doyle, T. J., W. Zheng, J. R. Cerhan, C.-P. Hong, T. A. Sellers, L. H. Kushi, and A. R
.
Folsom. 1997. The association of drinking water source and chlorination by-

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
products with cancer incidence among postmenopausalwomen in Iowa: a p
r
o
s
p
e
c
-tive
cohort study. Am. J. Public Health 87:1168–1176.
duMoulin, G. C., and K. D. Stottmeir. 1986. Waterborne mycobacteria: an i
n
c
r
e
a
s
i
n
gthreat to
health. Am Soc. Microbiol. News 52:525–529.
duMoulin, G. C., K. D. Stottmeier, P. A. Pelletier, T. A. Tsang, and J. Hedley -
Whyte. 1988. Concentration of Mycobacterium avium by hospital hot water
systems. J. Amer. Medical Assoc. 260:1599−1601.
Environmental Protection Agency (EPA). 1989. Drinking water; national primary drink-
ing water regulations; total coliforms (including fecal coliforms and E. coli); final
rule. Federal Register 54:27544–68.
EPA 1998. Announcement of the drinking water contaminant candidate list; notice.
Federal Register 63:10274–87.
EPA. 1999. Legionella: human health criteria document. EPA-822-R-99-001. W
a
s
h
-ington,
DC: EPA Office of Science and Technology.
EPA. 2002. Potential contamination due to cross-connections and backflow
and t
h
e
associated health risks: an issues paper. Available on-line at:
http://www.epa.gov/ safewater/tcr/pdf/ccrwhite.pdf. Accessed May 11, 2006.
EPA. 2004. Drinking water contaminant candidate list 2; notice. Federal R
e
g
i
s
te
r69(64):17406–
17415.
Falkinhan, J. O., III, C. D. Norton, and M. W. LeChevallier. 2001. Factors i
n
f
l
u
e
n
c
i
n
gnumbers
of Mycobacterium avium, Mycobacterium intracellulare, and other myco-
bacteria in drinking water distribution systems. Appl. Environ. Microbiol.
67(3):1225–1231.
Ferroni, A., L. Nguyen, B. Pron, G. Quesne, M. C. Brusset, and P. Berche. 1998. O
ut-
break of nosocomial urinary tract infection due to Pseudomonas aeruginosa in a pe-
diatric surgical unit associated with tap-water contamination. J. Hosp. Infect.
39:301–307.
Fewtrell, L. and J. M. Colford, Jr. 2004. Water, Sanitation and Hygiene: Interventions
and Diarrhoea - A Systematic Review and Meta-analysis. 2004. Washington, DC:
World Bank.
Fields, B. S., G. N. Sanden, J. M. Barbaree, W. E. Morril, R. M. Wadowsky, E. H. White,
and J. C. Feeley. 1989. Intracellular multiplication of Legionella pneumophila in
amoebae isolated from hospitalhot water tanks. Current Microbiol. 18:131–137.
Fields, B. 2005. Legionella in distribution system biofilms. January 13, 2005. P
r
e
-
sented to
the NCR Committee on Public Water Supply Distribution Systems. Irvine, CA.
Fischeder, R. R., R. Schulze-Robbecke, and A. Weber. 1991. Occurrence of mycobacte-
ria in drinking water samples. Zbl. Hygiene 192:154−158.
Flannery, B., L. B. Gelling, D. J. Vugia, J. M. Weintraub, J. J. Salerno, M. J. Conroy, V.
A. Stevens, C. E. Rose, M. R. Moore, B. S. Fields, and R. E. Besser. 2006. R
e
d
u
c
- ing
Legionella colonization of water systems with monochloramine. Emerg. I
n
f
e
c
t
. Dis.
[serial on the Internet]. Available on-line at http://www.cdc.gov
/ncidod/EID/vol12no04/05-1101.htm. Accessed April13, 2006.
Fout, G. S., B. C. Martinson, M. W. N. Moyer, and D. R. Dahling. 2003. A multiplex
reverse transcriptase-PCR method for detection of human enteric viruses in ground-
water. Appl. Environ. Microbiol. 69: 3158–3164.
Frost, F. J., M. Roberts, T. R. Kunde, G. Craun, K. Tollestrup, L. Harter, and T. M
u
l
l
e
r
. 2005.
How clean must our drinking water be: the importance of protective i
m
m
u
- nity. Journal
of Infectious Diseases191:809–814.

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
Geldreich, E. E., K. R. Fox, J. A. Goodrich, E. W. Rice, R. M. Clark, and D. L. S
w
e
r
d
- low.
1992. Searching for a water supply connection in the Cabool, Missouri diseaseoutbreak
of Escherichia coli 0157:H7. Water Research 26(8):1127–1137.
Geldreich, E. E. 1996. Microbial Quality of Water Supply in Distribution Systems.
Boca Raton, FL: CRC Press, Inc.
Geldreich, E. E., and M. W. LeChevallier. 1999. Microbial water quality in distribution
systems. Pp. 18.1–18.49 In: Water Quality and Treatment. 5th
edition. R. D.
Let- terman (ed.). New York: McGraw-Hill.
Gerba, C. P., J. B. Rose, and C. N. Haas. 1996. Sensitive populations: who is at t
h
e
greatest
risk? International Journal of Food Microbiology 30(1–2):113–123.
Glover, N. A., N. Holtzman, T. Aronson, S. Froman, O. G. W. Berlin, P. Dominguez, K.
A. Kunkel, G. Overturf, G. Stelma, Jr., C. Smith, and M. Yakrus. 1994. The isola-
tion and identification of Mycobacterium avium complex (MAC) recovered
from Los Angeles potable water, a possible source of infection in AIDS patients.
Interna- tional J. Environ. Health Res. 4:63−72.
Graves, C. G., G. M. Matanoski, and R. G. Tordiff. 2001. Weight of evidence of an
association between adverse reproductive and developmental effects and exposure to
disinfection byproducts:acritical review. Regul. Toxicol. Pharmacol. 34:103–24.
Haas, C. N., M. A. Meyer, and M. E. Paller. 1983. The ecology of acid-fast organisms
in water supply, treatment, and distribution systems. J. Amer. Water Works Assoc.
75:39−144.
Haudidier, K., J. L. Paquin, T. Francais, P. Hartemann, G. Grapin, F. Colin, M. J. Jour-
dain, J. C. Block, J. Cheron, O. Pascal, Y. Levi, and J. Miazga. 1988. Biofilm
growth in drinking water networks: a preliminary industrial pilot plant experiment.
Water Sci. Technol. 20:109–115.
Havelaar, A. H., A. E. Hollander, P. F. Teunis, E. G. Evers, H. J. Van Kranen, J
.F.
Versteegh, J. E. Van Koten, and W. Slob. 2000. Balancing the risks and benefits of
drinking water disinfection: disability adjusted life-years on the scale. Environ-
mental Health Perspective108(4):315–321.
Hawkins, M. A., S. M. DeLong, R. Marcus, T. Jones, S. Shallow, S. M. Zansky, K. G
.
McCombs, A. K. Courtney, C. Medus, B. C. Imhoff, and the EIP FoodNet W
o
r
k
i
n
g
Group.
2002. The Burden of Diarrheal Illness in FoodNet, 2000–2001 (Draft r
e
- port).
Atlanta, GA: CDC.
Heffelfinger, J. D., J. L. Kool, S. K. Fridkin, V. J. Fraser, J. C. Carpenter, J. Hageman, J.
Carpenter, and C. G. Whitney. 2003. Risk of hospital-acquired Legionnaires ’ dis-
ease in cities using monochloramine versus other water disinfectants. Infection
Control and HospitalEpidemiol. 24(8):569–574.
Hellard, M. E., M. I. Sinclair, A. B. Forbes, and C. K. Fairley. 2001. A
randomized, blinded, controlled trial investigating the gastrointestinal health effects of
drinking water quality. Environ. Health Perspect. 109:773–778.
Herikstad, H., S. Yang, T. J. Van Gilder, D. Vugia, J. Hadler, P. Blake, V. Deneen, B
.
Shiferaw, F. J. Angulo, and The FoodNet Working Group. 2002. A
population- based estimate of the burden of diarrhoeal illness in the United
States: FoodNet, 1996–97. Epidemiol. Infect. 129:9–17.
Herwaldt, B. L., G. F. Craun, S. L. Stokes, and D. D. Juranek. 1991. Waterborne-disease
outbreaks, 1989–1990. MMWR 40(No. SS-3):1–21.
Horsburgh, C. R. 1991. Mycobacterium avium complex infection in the acquired immu-
nodeficiency syndrome. New England Journal of Medicine 324:1332−1338.
Hulley, S. B., and S. R. Cummings. 1988. Designing Clinical Research: An E
p
i
d
e
m
i
-
ologic Approach. Baltimore, MD:Williams and Wilkins.

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
Hunter, P. R., R. M. Chalmers, S. Hughes, and Q. Syed. 2005. Self-reported diarrhea i
n
a
control group: a strong association with reporting of low-pressure events in tap
water. Clinical Infectious Diseases 40:e32–34.
Imhoff, B., D. Morse, B. Shiferaw, M. Hawkins, D. Vugia, S. Lance-Parker, J. Hadler, C
.
Medus, M. Kennedy, M. R. Moore, and T. Van Gilder. 2004. Burden of
self- reported acute diarrheal illness in FoodNet surveillance areas, 1998–1999.
For the Emerging Infections Program (EIP) FoodNet Working Group. Clinical
Infectious Diseases 38 Suppl3:S219–26.
Jacquemin, J. L., A. M. Simitzia, and N. Chaneoau. 1981. Free-living amoebae in f
r
e
s
hwater:
a study of the water supply of the town of Poiters. Bull. Soc. Pathol. Exot.74:521–
534.
Jo, W. K., C. P. Weisel, and P. J. Lioy. 1990. Routes of chloroform exposure and b
o
d
y
burden from showering with chlorinated tap water. Risk Analysis 10(4):575–580.
Keinanen, M. M., P. J. Martikainen, and M. H. Kontro. 2004. Microbial community
structure and biomass in developing drinking water biofilms. Can. J. Microbiol.
50(3):183–191.
Kelley, J., G. Kinsey, R. Paterson, D. Brayford, R. Pitchers, and H. Rossmoore. 2
0
0
3
.
Identification and control of fungi in distribution systems. Denver CO: AWWA a
n
d
AwwaRF.
King, W. D., and L. D. Marrett. 1996. Case-control study of bladder cancer and c
h
lo
r
i
- nation
by-products in treated water (Ontario, Canada). Cancer Causes Control7:596–
604.
King, W. D., L. Dodds, and A. C. Allen. 2000. Relation between stillbirth and specific
chlorination by-products in public water supplies. Environ. Health P
e
r
s
p
e
c
t
.108:883–
886.
Kirmeyer, G. J., M. Friedman, K. Martel, D. Howie, M. LeChevallier, M. Abbaszadegan,
M. Karim, J. Funk, and J. Harbour. 2001. Pathogen intrusion into the
distribution system. #90835. Denver, CO: AWWA and AwwaRF.
Klotz, J. B., and L. A. Pyrch. 1999. Neural tube defects and drinking water d
i
s
i
n
f
e
c
t
i
o
n by-
products. Epidemiology 10:383–390.
Koivusalo, M., E. Pukkala, T. Vartiainen, J. J. K. Jaakkola, and T. Hakulinen. 1
9
9
7
.
Drinking water chlorination and cancer – a historical cohort study in Finland. Can-
cer Causes Control 8:192–200.
Kool, J. L., J. C. Carpenter, and B. S. Fields. 1999. Effect of monochloramine disinfec-
tion of municipal drinking water on risk of nosocomial Legionnaires’ disease. Lan-
cet 353(9149):272–277.
Kramer, M. H., and T. E. Ford. 1994. Legionellosis: ecological factors of an e
n
v
i
r
o
n
- mentally
'new' disease. Zentralbl Hyg Umweltmed.195(5–6):470–482.
Kramer, M. H., B. L. Herwaldt, G. F. Craun, R. L. Calderon, and D. D. Juranek. 1
9
9
6
.
Surveillance for waterborne-disease outbreaks—United States, 1993–1994.
MMWR 45(No. SS-1):1–33.
LeChevallier, M. W. 2004. Control, treatment, and disinfection of Mycobacterium
avium complex in drinking water. Pp. 143–168 In: Pathogenic Mycobacteria in Wa-
ter. S. Pedley, J. Bartram, G. Rees, A. Dufour, J. Cotruvo (eds.). ISBN:
1843390590. Geneva: World Health Organization.
LeChevallier, M. W., R. W. Gullick, M. R. Karim, M. Friedman, and J. E. Funk. 2
0
0
3
. The
potential for health risks from intrusion of contaminants into distribution sys-
tems from pressuretransients. J. Water and Health 1(1):3–14.
LeChevallier, M. W., and R. J. Seidler. 1980. Staphylococcus aureus in rural d
r
i
n
k
i
n
gwater.
Appl. Environ. Microbiol. 39:739–742.

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
LeChevallier, M. W., T. J. Wade, S. E. Shaw, D. Levy, R. L. Calderon, J. M. Colford, Jr
.
2004. Results of the big WET: an epidemiology study of the microbiological quality
of drinking water in Davenport, Iowa. AWWA Water Quality and Technology Con-
ference, San Antonio, Texas, November 14–17, 2004.
Lee, S. H., D. A. Levy, G. F. Craun, M. J. Beach, and R. L. Calderon. 2002a. S
u
r
v
e
i
l
-lance
for waterborne disease outbreaks—United States, 1999–2000. M
M
W
R
51(SS-8):1–47.
Lee, S., D. Levy, A. Hightower, B. Imhoff, and the EIP FoodNet Working G
r
o
u
p
.2002b.
Drinking water exposures and perceptions among 1998–1999 FoodNet s
u
r
- vey
respondents. International Conference on emerging Infectious Diseases. A
t
- lanta,
GA, March 2002.
Levy, D. A., M. S. Bens, G. F. Craun, R. L. Calderon, and B. L. Herwaldt. 1998. S
u
r
-
veillance for waterborne-disease outbreaks—United States, 1995–1996. M
M
W
R
47(SS-
5):1–34.
Liang, J. L., E. J. Dziuban, G. F. Craun, V. Hill, M. R. Moore, R. J. Gelting, R
. L.
Calderon, M. J. Beach, and S. L. Roy. 2006. Surveillance for Waterborne Disease
and Outbreaks Associated with Drinking Water and Water not Intended for Drink-
ing—United States, 2003–2004. MMWR (in print).
Lin, Y. E., R. D. Vidic, J. E. Stout, and V. L. Yu. 1998. Legionella in water distribution
systems. J. Amer. Water Works Assoc. 90:112–121.
Maluccio, J. A., and R. Flores. 2005. Impact evaluation of a conditional cash t
r
a
n
s
f
e
r program.
The Nicaraguan Red de Protección Social. Research Report No. 141.
Washington, DC: International Food Policy Research Institute.
Martiny, A. C., T. M. Jorgensen, H. J. Albrechtsen, E. Arvin, and S. Molin. 2003. L
o
n
g
-term
succession of structure and diversity of a biofilm formed in a model drinking water
distribution system. Appl. Environ. Microbiol. 69(11):6899–6907.
McMillan, M., T. F. Jones, A. Banerjee, D. Vugia, A. Cronquist, S. Segler, P. Ryan, C
.
Medus, P. Smith, B. Shiferaw, and F. Angulo. 2004. The Burden of Diarrheal Ill-
ness in FoodNet, 2002–2003. In: Proceedings of the International Conference
on Emerging Infectious Diseases, Atlanta, GA.
Mead P.S., L. Slutsker L, V. Dietz, L. F. McCaig, J. S. Bresee, C. Shapiro, J. M. Griffin,
and R. V. Tauxe. 1999. Food-related illness and death in the United States. Emerg.
Infect. Dis. 5(5):607–25.
Moore, A. C., B. L. Herwaldt, G. F. Craun, R. L. Calderon, A. K. Highsmith, and D. D
.
Juranek. 1993. Surveillance for waterborne disease outbreaks—United States,
1991–1992. MMWR 42(SS-5):1–22.
Moore, M. R., M. Pryor, B. Fields, C. Lucas, M. Phelan, and R. E. Besser. 2006. Intro-
duction of monochloramine into a municipal water system: Impact on colonization
of buildings by Legionella spp. ApplEnviron Microbiol 72(1):378–383.
Morris, R. D., and R. Levine. 1995. Estimating the incidence of waterborne infectious
disease related to drinking water in the United States. Assessing and
Managing Health Risk from Drinking Water Contamination: Approaches and
Application. Proceedings of the Rome Symposium, September 1994. IAHS 233:
75–88.
Murga, R., T. S. Forster, E. Brown, J. M. Pruckler, B. S. Fields, and R. M. donlan. 2
0
0
1
.Role
of biofilms in the survival of Legionella pneumophila in a model potable w
a
t
e
rsystem.
Microbiology 147:3121–3126.
Nagy, L. A., and B. H. Olson. 1982. The occurrence of filamentous fungi in drinking
water distribution systems. Canadian Journal Microbiol. 28:667–671.

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
Nahapetian, K., O. Challemel, D. Beurtin, S. Dubrou, P. Gounon and F. Squinazi. 1
9
9
1
.The
intracellular multiplication of Legionella pneumophila in protozoa from hos
pi
ta
l
plumbing systems. Res. Microbiol. 142:677–685.
National Research Council (NRC). 1983. Risk Assessment in the Federal G
o
v
e
r
n
m
e
n
t
:Managing the
Process. Washington, DC: National Academy Press.
NRC. 2004. Review of the Army’s Technical Guides on Assessing and M
a
n
a
g
i
n
g
Chemical
Hazards to Deployed Personnel. Washington, DC: National Academy Press.
Niemi, R. M., S. Knuth, and K. Lundstrom. 1982. Actinomycetes and fungi in s
u
r
f
a
c
ewaters
and in potablewater. Appl. Environ. Microbiol. 43:378–388.
Nightingale, S. D., L. T. Byrd, P. M. Southern, J. D. Jockusch, S. X. Cal, and B
. A. Wynne.
1992. Mycobacterium avium-intracellulare complex bacteremia in human
immunodeficiency virus positive patients. Journal of Infectious Disease
165:1082−1085.
Nieuwenhuijsen, M. J., M. B. Toledano, N. E. Eaton, J. Fawell, and P. Elliott. 2
0
0
0
.
Chlorination disinfection byproducts in water and their association with adverse r
e
-
productiveoutcomes: a review. Occup. Environ. Med. 57:73–85.
Norton, C. D., and M. W. LeChevallier. 2000. A pilot study of bacteriological popula-
tion changes through potable treatment and distribution. Appl. Environ. Microbiol.
66(1):268–276.
Payment, P., J. Siemiatycki, L. Richardson, G. Renaud, E. Franco, and M. Prevost. 1
9
9
7
.A
prospective epidemiological study of gastrointestinal health effects due to the
consumption of drinking water. Intern. J. Environ. Health Res. 7:5–31.
Payment, P., L. Richardson, J. Siemiatycki, R. Dewar, M. Edwardes, and E. Franco.
1991. Randomized trial to evaluate the risk of gastrointestinal disease due to con-
sumption of drinking water meeting microbiological standards. Am. J. Pub. Health
81(6):703–708.
Piriou, P., K. Helmi, M. Jousset, N. Castel, E. Guillot, and L. Kiene. 2000. Impact of
biofilm on C. parvum persistence in distribution systems. In: Proceedings of an In-
ternational Distribution System Research Symposium, June 10–11. Denver, CO:
AWWA.
Pryor, M., S. Springthorpe, S. Riffard, T. Brooks, Y. Huo, G. Davis, and S. A. Satter.
2004. Investigation of opportunistic pathogens in municipal drinking water
under different supply and treatment regimes. Water Sci. Technol. 50(1):83–90.
Quignon, F., M. Sardin, L. Kiene, and L. Schwartzbrod. 1997. Poliovirus-1 in
a
c
t
i
v
a
t
io
n and
interaction with biofilm: a pilot-scale study. Appl. Environ. Microbiol. 63(3):978–
982.
Raina, P. S., F. L. Pollari, G. F. Teare, M. J. Goss, D. A. Barry, and J. Wilson. 1
9
9
9
. The
relationship between E. coli indicator bacteria in well-water and gastrointestinal
illnesses in rural families. Can. J. Public Health 90(3):172–175.
Regli, S., J. B. Rose, C. N. Haas, and C. P. Gerba. 1991. Modeling risk from Giardia
and viruses in drinking water. J. Amer. Water Works Assoc. 83(11):76–84.
Rogers, J., A. B. Dowsett, P. J. Dennis, J. V. Lee, and C. W. Keevil. 1994. I
n
fl
u
e
n
ceof
materials on biofilm formation and growth of Legionella pneumophila in potable
water systems. Appl. Environ. Microbiol. 60:1842–1851.
Savitz, D., and C. L. Moe. 1997. Water: chlorinated hydrocarbons and infectious a
g
e
n
t
s
. Pp.
89–118 In: Topics in Environmental Epidemiology. N. K. Steenland and D. A.
Savitz (eds.). Oxford: Oxford University Press.

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
Schoenen, D. 1986. Microbial growth due to materials used in drinking water systems.
In: Biotechnology, Vol. 8. H. J. Rehm and G. Reed (eds.). Weinheim: VCH
Verlagsgesellschaft.
Schulze-Robbecke, R., and R. Fischeder. 1989. Mycobacteria in biofilms. Zbl. Hyg.
88:385–390.
Schwartz J., R. Levin, and K. Hodge. 1997. Drinking water turbidity and pediatric h
o
s
-pital
use for gastrointestinal illness in Philadelphia. Epidemiology 8:615–620.
Shepherd, J. L.; R. L. Corsi, and J. Kemp. 1996. Chloroform in indoor air and w
a
s
t
e
w
a
-ter: the
role of residential washing machines. Journal of the Air & Waste Manage- ment
Association 46(7):631–642.
Singh, N., and V. Yu. 1994. Potable water and Mycobacterium avium complex in H
I
V
patients:is prevention possible? Lancet 343:1110–1111.
Smith, D. B., A. F. Hess, and S. A. Hubbs. 1990. Survey of distribution system c
o
l
i
f
o
r
m
occurrences in the United States. In: Proceedings of the Water Quality Technology
Conference. Denver, CO: AWWA.
Smith, C. A., C. Phiefer, S. J. Macnaughton, A. Peacock, R. S. Burkhalter, R. Kirkegaard,
and D. C. White. 2000. Quantitative lipid biomarker detection of unculturable mi-
crobes and chlorine exposure in water distribution system biofilms. Water Research
34(10):2683–2688.
Stanwell-Smith, R., Y. Anderson, and D. Levy. 2003. National surveillance
systems. Pp. 25–40 In: Drinking Water and Infectious Disease: Establishing the
Links. P. R. Hunter, M. Waite, and E. Ronchi (eds.). Boca Raton, FL: CRC Press
Steenland, K., and C. Moe. 2003. Epidemiology and drinking water—are we r
u
n
n
i
n
g dry?
Epidemiology 14(6):635–636.
Steinmaus, C., Y. Yuan, M. N. Bates, and A. H. Smith. 2003. Case-control s
tudy of
bladder cancer and drinking water arsenic in the western United States. American
Journal of Epidemiology 158:1193–1201.
Stout, J. E., V. L. Yu, and M. G. Best. 1985. Ecology of Legionella pneumophila w
i
t
h
i
nwater
distribution systems. Appl. Environ. Microbiol. 49:221–228.
Strauss, B., W. King, A. Ley, and J. R. Hoey. 2001. A prospective study of rural d
r
i
n
k
-ing
water quality and acute gastrointestinal illness. BMC Public Health 1:8.
Taylor, R. H., J. O. Falkinham, III, C. D. Norton, and M. W. LeChevallier. 2000. C
h
l
o
-rine,
chloramine, chlorine dioxide, and ozone susceptibility of M
y
c
o
b
a
c
t
e
r
i
u
mavium. Appl.
Environ. Microbiol. 66(4):1702–1705.
Thofern, E., D. Schoenen, and G. J. Tuschewitzki. 1987. Microbial surface c
o
l
o
n
i
z
a
t
i
o
n and
disinfection problems. Off Gesundh.-wes., 49:Suppl:14.
Thomas, V., K. Herrera-Rimann, D. S. Blanc, and G. Greub. 2006. B
i
o
d
i
v
e
r
s
i
t
y of
amoebae and amoeba-resisting bacteria in a hospital water network. Appl. Environ.
Microbiol.72(4):2428–2438.
Trautman, M., T. Michalsky, H. Wiedeck, V. Radlosavljevic, and M. Ruhnke. 2
0
0
1
.Tap
wate colonization with Pseudomonas aeruginosa in a surgical intensive careunit
(ICU) and relation to Pseudomonas infections of ICU patients. Infect. Control.
Hosp. Epidemiol. 22:49–52.
Vaerewijck, M. J. M., G. Huys, J. P. Palomino, J. Swings, and F. Portaels. 2005. Myco-
bacteria in drinking water distribution systems: ecology and significance for human
health. FEMS Microbiology Review 29:911–934.
von Reyn, C. F., R. D. Waddell, T. Eaton, R. D. Arbeit, J. N. Maslow, T. W. Barber, R
.J.
Brindle, C. F. Gilks, J. Lumio, J. Lahdevirta, A. Ranki, D. Dawson, and J.
O. Falkinham, III. 1993. Isolation of Mycobacterium avium complex from water in
the

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
United States, Finland, Zaire, and Kenya. Journal of Clinical Microbiology
31:3227−3230.
von Reyn, C. F., J. N. Maslow, T. S. Barber, J. O. Falkinham, III, and R. D. A
r
b
e
i
t
.
1994. Persistent colonisation of potable water as a source of Mycobacterium avium
infection in AIDS. Lancet 343:1137−1141.
VROM. 2005. Legionella in Europe: Problems and Prevention: Summary by the C
h
a
i
r
-man.
International Congress September 28–29, Amsterdam, The Netherlands.
Available on-line at http://international.vrom.nl/docs/internationaal/congres%
20verslag%20Engels%2025-11-04.pdf. Accessed May 11, 2006.
Wade, T. J., S. K. Sandhu, D. Levy, S. Lee, M. W. LeChevallier, L. Katz, and J
.M.
Colford, Jr. 2004. Did a severe flood in the Midwest cause an increase in the inci-
dence of gastrointestinal symptoms? Amer. J. Epidemiol. 159(4):398–405.
Wadowsky, R. M., and R. B. Yee. 1983. Satellite growth of Legionella p
n
e
u
m
o
p
h
i
l
awith an
environmental isolate of Flavobacterium breve. Appl. Environ. Microbiol. 46:1447–
1449.
Wadowsky, R. M., and R. B. Yee. 1985. Effect of non-legionellaceae bacteria on t
h
e
multiplication of Legionella pneumophila in potable water. Appl. Environ. M
i
c
r
o
-biol.
49:1206–1210.
Waller, K., S. H. Swan, G. DeLorenze, and B. Hopkins. 1998. Trihalomethanes in
drinking water and spontaneous abortion. Epidemiology 9:134-140.
Wolinsky, E. 1979. Nontuberculous Mycobacteria and associated diseases. Am. Rev.
Respir. Dis. 119:107–159.
World Health Organization (WHO). 2002. Heterotrophic Plate Count Measurements i
n
Drinking Water Safety Management. WHO/SDE/WSH/02.10. Geneva, S
w
i
t
z
e
r
- land:
WHO.
Yang, C. Y., H. F. Chiu, M. F. Cheng, and S. S. Tsai. 1998. Chlorination of d
r
i
n
k
i
n
g water
and cancer mortality in Taiwan. Environmental Research 78:1–6.

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
4
Physical Integrity
This chapter focuses on physical integrity—the ability of the distribution
system to act as a physical barrier that prevents external contamination from
affecting the quality of the internal, drinking water supply. Water distribution
system engineers have defined the physical integrity of the distribution system
to be its ability to handle external and internal stresses such that the physical
material of the system does not fail (Male and Walski, 1991). Here failure is
interpreted more broadly to encompass the absence of a critical component, the
improper installation of a component, or the installation of an already contami-
nated component.
The physical integrity of the distribution system is always in a state of
change, and the aging of the nation’s distribution systems and eventual need for
replacement are growing concerns. Maintaining such a vast physical infrastruc-
ture is a challenge because of the complexity of individual distribution systems,
each of which is comprised of a network of mains, fire hydrants, valves, auxil-
iary pumping or booster disinfection substations, storage reservoirs, standpipes,
and service lines along with the plumbing systems in residences, large housing
projects, high-rise buildings, hospitals, and public buildings. This is further
complicated by factors that vary from system to system such as the size of the
distribution network for the population served, the predominant pipe material
and age of pipelines, water pressure, the number of line breaks each year, water
storage capacity, and water supply retention time in the system. When consider-
ing the replacement of a given component of the distribution system,
decision makers must weigh its potential remaining life versus the potential that
the component will fail, which could result in costly consequences and
compromise the water utility’s service.
The physical integrity of the distribution system, from the entry point to the
customer’s tap, is a primary barrier against the entry of external
contaminants and the loss in quality of the treated drinking water. This barrier
includes such materials as the pipe wall and reservoir cover as well as physical
connections to nonpotable water sources. The barrier must be non-permeable
since contaminants can enter through breaks or failures in materials as well
as through the materials themselves. Table 4-1 gives examples of the
infrastructure components that constitute this physical barrier, what they
protect against, and the materials of which they are commonly constructed.
A variety of components and materials make up this physical barrier.
Four major component types are delineated and referred to repeatedly in this
chapter:
(1) pipes including mains, services lines, and premise plumbing; (2) fittings a
n
d
appurtenances such as crosses, tees, ells, hydrants, valves, and m
e
t
e
r
s
;
142

i
g
h
t
s
reserved.
TABLE 4-1 Infrastructure Components, What They Protect Against, and Common Materials
Component External Contamination the Barrier Protects Against Materials Used
Pipe Soil, groundwater, sewer exf iltration, su
r
fa
c
erunoff, human
activ ity , animals, insects, and other lif e f orms
Asbestos cement, reinf orced concrete, steel, lined and unlined cast
iron, lined and unlined ductile iron, PVC, poly ethy lene and
HDPE, galv anized iron, copper, poly buty lene
Pipe wrap and coatings Supporting role in that it preserv es the pipe integrity Poly ethy lene, bitumastic, cement-mortar
Pipe linings Supporting role in that it preserv es the pipe integrity Epoxy , urethanes, asphalt, coal tar, cement-mortar,plastic inserts
Serv ice lines Soil, groundwater, sewer exf iltration, su
r
fa
c
erunoff, human
Premise plumbing Air contamination, human activ ity , sewage a
n
d
industrial
nonpotable water.
Galv anized steel or iron, lead, copper, chlorinate
dPVC, cross-
linked poly ethy lene, poly ethylene, poly butylene, PVC, brass,
cast iron
Copper, lead, galv anized steel or iron, iron, steel, c
h
l
o
r
i
n
a
t
e
d
PVC, PVC,
cross-linked poly ethy lene, poly ethy lene, poly butylene
Fittings and a
p
p
u
r
t
e
n
a
n
c
e
s(meters,
v alv es, hy drants, ferrule
s)
Soil, groundwater, sewer exf iltration, su
r
fa
cerunoff, human
Brass, rubber, plastic
Storage f acility w
a
l
l
s
,
roof ,
cov er, v ent hatch
Air contamination, rain, algae, surf ace r
u
n
o
f
f
,human activ ity ,
animals, birds, and insects
Concrete, steel, asphaltic, epoxy , plastics
Backf low p
r
e
ve
n
t
i
o
n
dev ices
Nonpotable water Brass, plastic
Liquids Not applicable Oils, greases, lubricants
Gaskets and joints Soil, groundwater, sewer exf iltration, su
r
fa
cerunoff, human
activ ity , animals, insects, and other lif e f
o
r
m
s
Rubber, leadite, asphaltic, plastic
143

i
g
h
t
s
reserved.
e
d
u
c
i
n
g
Risks
(3) storage facilities including reservoirs (underground, open, and covered), e
l
e
-
vated storage tanks, ground level storage tanks, and standpipes; and (4) b
a
c
k
-
flow prevention devices. The materials used by the water industry for t
he
se
components, particularly pipes, have changed significantly over time (AWWA,
1986; Von Huben, 1999). For example, cast iron pipe (lined or unlined)
has been largely phased out due to its susceptibility to both internal and
external corrosion and associated structural failures. Ductile-iron pipe (with or
without a cement lining) has taken its place because it is durable and strong, has
high flexural strength, and has good resistance to external corrosion from soils.
It is, however, quite heavy, it might need corrosion protection in certain soils,
and it requires multiple types of joints. Concrete, asbestos cement, and polyvinyl
chloride (PVC) plastic pipe have been used to replace metal pipe because
of their relatively good resistance to corrosion. Polyethylene pipe is growing
in use, especially for trenchless applications like slip lining, pipe bursting,
and directional drilling (Morrison, 2004). High-density polyethylene pipe is
the second most commonly used pipe. It is tough, corrosion resistant both
internally and externally, and flexible. The manufacturer estimates its service
life to be 50 to 100 years (AWWA, 2005a). Chapter 1 discusses the rate of pipe
replacement in the United States and notes that much of the current
infrastructure is nearing the end of its usable lifetime.
FACTORS CAUSING LOSS OF PHYSICAL INTEGRITY
Losses in physical integrity are caused by an abrupt or gradual alteration i
n
the structure of the material barrier between the external environment and
the drinking water, by the absence of a barrier, or by the improper installation or
use of a barrier. These mechanisms are summarized in Table 4-2.
Infrastructure components break down or fail over time due to chemical i
n
-
teractions between the materials and the surrounding environment, eventually
leading to holes, leaks, and other breaches in the barrier. These processes c
a
n
occur over time scales of days to decades, depending on the materials and condi-
tions present. For example, plastic pipes can be very rapidly compromised b
y
nearby hydrophobic compounds (e.g., solvents in the vadose zone that r
e
sul
t
from surface or subsurface contamination), with the resulting permeation of
those compounds into the distribution system through the pipe materials. Both
internal and external corrosion can lead to structural failure of pipes and joints,
thereby allowing contaminants to infiltrate into the distribution system via leaks
or subsequent main breaks. Materials failure can be hastened if the distribution
systemwater pressure is too high, from overburden stresses on pipes, and during
natural disasters. Indeed, hurricanes and earthquakes have caused extensive
sudden damage to distribution systems, including broken service lines and fire
hydrants, pipes disconnected or broken by the uprooting of trees, cracks in ce-
ment water storage basins, and seam separations in steel water storage tanks
(Geldreich, 1996).

i
g
h
t
s
reserved.
TABLE 4-2 What Causes a Loss in Physical Integrity?
Mechanism of Integrity Loss
Component
Alteration in material structure leading to
failure
Absence of the barrier or material
Improper application or installation of the
barrier
Pipe  Corrosion
 Permeation
 Too high internal water p
r
e
s
s
u
re
or surges
 Shif ting earth
 Exposure to UV light
 Stress f rom ov erburden
 Temperature fluctuations, f reezing
 Absence of external or i
n
te
r
n
a
llinings,
wraps, coatings to protect the pipe
 Unsanitary activ ity during c
o
n
s
t
r
u
c
t
i
o
n
,
replacement, or repair
 Unintentional creation of cracks and
breaks
 Use of f aulty materials
Fitting and
appurtenance
Storage facility wall,
roof, cover, vent, hatch
 Corrosion
 Permeation
 Corrosion
 Permeation
 Natural disasters
 Failure due to aging and w
eathering
 Appurtenance in a f looded m
e
t
e
r
or v alv e pit
(absence of appropriate structures)
 Missing cov er, roof , hatch, v
ent,can
lead to unprotected access to the stor-
age f acility . Could be unintentional or
intentional (v andalism)
o
n
s
t
r
u
c
t
i
o
n
,
 Unintentional creation of cracks and
breaks
 Contact between dissimilar m
e
t
a
l
s
o
n
s
t
r
u
c
t
i
o
n
,
 Unintentional or intentional creation of
cracks and breaks
 Poor drainage f or runof f
Backflow prevention
device
 Corrosion  Missing dev ice will allow a
backf low
ev ent v ia a cross connection
 Improper installation
 Inadequate drainage of meter p
i
t
 Operational f ailure
145

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
A second major contributor to the loss of physical integrity is when certain
critical components are absent, either by oversight or due to vandalism.
For example, the absence of backflow prevention devices and covers for
storage facilities can allow external contaminants to enter distribution systems.
For the purposes of this discussion,pipes are assumed to always be present.
Finally, human activity involving distribution system materials can a
l
l
o
w
contamination to occur such as through unsanitary repair and replacement prac-
tices, unprotected access to materials, or the improper handling of materials
leading to unintentional damage. One must even consider the installation of
flawed materials, which might, for example, be brought about because of a lack
of protection of materials during storage and handling.
Structural Failure of Distribution System Components
Metallic pipe failures are divided generally into two categories:
corrosion failures and mechanical failures. Common types of failures for iron
mains include (Male and Walski, 1991; Makar, 2002):
 Bell splits or cracks that require cutting out the joint and replacing it
with a mechanical fitting; these are typical for leaditejoints
 Splits at tees and offsets and other fittings that require replacement
 Circumferential cracks or round cracks and holes, more typical in
smaller diameter pipe (< 10 in.). These can result from a lack of soil support,
causing the pipe to be called upon to act as abeam
 Splits or longitudinal cracks or spiral cracks that will blow out. L
ong
i
-
tudinal cracks are more common for larger pipe (> 12 in.) and can result f
r
o
m
crushing under external loads or from excessive internal pressure
 Spiral failures in medium diameter pipe
 Shearing failures in large diameter pipe
 Pinholes (corrosion hole) caused by internal corrosion
 Tap or joint blowout
 Crushed pipe
A simpler categorization can be found in Romer et al. (2004), who summarized
three types of pipe failures as weeping failures, pipe breaks, and sudden failures.
A weeping failure is where a leak allows an unnoticeable exchange of water to
and from the surrounding soil. A pipe break includes a hole in the pipe or a dis -
engagement of a bell-and-spigot joint. A sudden failure is the bursting of a pipe
wall or shear of the pipe cross section, as would occur for a concrete pipeline, or
a blow out, which refers to a complete break in a pipe.
Pipe breaks can occur for a myriad of reasons such as normal materials d
e
-
terioration, joint problems, movement of earth around the pipe, freezing a
n
d
thawing, internal and external corrosion, stray DC currents, seasonal changes in
internal water temperature, heavy traffic overhead including accidents that dam-

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
PHYSICAL INTEGRITY 147
age fire hydrants, changes in system pressure, air entrapment, excessive o
v
e
r
-
head loading, insufficient surge control (such as with water hammer and pres-sure
transients), and errors in construction practices (Male and Walski, 1
9
9
1
)
.This last
factor is especially troubling since it should be entirely preventable.
Nonetheless, there is evidence that poor quality workmanship during initial pipe
installation can lead to early structural failure of pipes (Clark and Goodrich,
1989). Burlingame et al. (2002) reported on premature (within one year of in -
stallation) failures in service lines that resulted from the combination of using
hard copper tubing and poor workmanship during cutting and flaring of the
ends. AwwaRF (1985) has also reported that failures with copper tubing can be
due to poor workmanship. One of the goals of proper installation of water
mains is to account for and circumvent these issues; unfortunately, failure to do
so translates into a substantialnumber of unnecessary main breaks.
One overriding factor in determining the potential for pipe failure is t
heforce
exerted on the water main. Contributors to this force include changes in
temperature, which cause contraction and expansion of the metal and the
surrounding soil, the weight of the soil over the buried main, and vibrations on
the main caused by nearby activities such as traffic. An important consideration
in this regard is the erosion potential of the supporting soil beneath the buried
main. In the construction of a main, special sand and soil can be laid beneath it
to help it bear external forces. But the movement of water in the ground beneath
the main can wash away the finer material and create small or large caverns un-
der the pipe. The force now bearing down on top of the pipe must be taken by
the pipe itself, without the help of supporting material underneath. If
these forces exceed the strength of the pipe, the main breaks. Most often these
breaks occur at the weakest part of the main, i.e., thejoint.
The factors that cause pipe failures can compound one another,
hastening the process. For example, if a main develops small leaks because of
corrosion, water within the distribution system can exfiltrate into the area
surrounding the pipe, eroding away the supporting soil. Leakage that
undermines the foundation of a water main can also occur from nearby sewer
lines, go on essentially unno- ticed, and eventually lead to water main collapse
(Morrison, 2004).
Table 4-3 summarizes common problems that lead to pipe failures for pi
pesof
differing materials. These are some of the principal factors, but they are n
o
tthe
only factors that act individually or in combination to lead to a main break.
Other factors could include a street excavation that accidentally disturbs a water
main and the misuse of fire hydrants. At most utilities, overall pipe break rates
have been relatively low and stable (Damodaran et al., 2005) even though the
infrastructure is aging.
Other components of distribution system also experience structural failure,
although they have not historically received the attention afforded to pipes. F
o
r
storage facilities, structural failure is less of a problem than external contamina-
tion due to the absence or failure of an essential component such as a cover o
r
vent. Fittings and appurtenances can suffer from the effects of corrosion
and permeation.

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
TABLE 4-3 Most Common Problems that Lead to Pipe Failure for Various Pipe M
a
t
e
r
i
a
l
s
Pipe Material
(commonsizes)
PVC and P
ol
yethyl
ene
(4−36 in.)
Cast/Ductile I
r
o
n
(4−64
in,) (lined a
n
d
unlined)
Steel
(4−120 i
n
.
)
Asbestos-Cem ent
(4−35 in.)
Concrete
(12−16 to 144−168 i
n
.
)
(prestressed o
r
rein-
forced)
Problems
Excessive deflection, joint misalignment and/or leakage, leak-
ing connections, longitudinal breaks fromstress, exposure to
sunlight, too high internal w ater pressure or frequent surges in
pressure, exposure to solvents, hard to locate w hen buried,
damage can occur during tapping
Internal corrosion, joint misalignment and/or l
e
a
k
a
g
e
,
external
corrosion, leaking connections, casting/manufacturing flaws
Internal corrosion, externalcorrosion, e
x
cessive deflection, joint
leakage, imperfections in w elded joints
Internal corrosion, cracks, joint misalignment and/or
leakage, small pipe can be damaged during handling or
tapping, pipe must be in proper soil, pipe is hard to locate
w hen buried
Corrosion in contact w ith groundwaterhigh in s
u
l
f
a
t
e
s
and chlorides,
pipe is very heavy, alignment can be difficult, settling of the
surrounding soilcan cause joint leaks, manufacturing flaws
SOURCES: Morrison (2004) and AWWA (1986).
Corrosion as a Major Factor
Corrosion is the degradation of a material by reaction with the local envi-
ronment. In water distribution systems, the term corrosion refers to dissolution
of concrete linings and concrete pipe, as well as to the deterioration of metallic
pipe and valves via redox reactions (e.g., iron pipe rusting). Degradation origi-
nating from the inside of the pipe via reactions with the potable water is termed
internal corrosion. Degradation originating outside the pipe on surfaces
contacting moist soil is referred to as external corrosion. Both internal and
external corrosion can cause holes in the distribution system and cause loss of
pipeline integrity. In some cases holes are formed directly in pipes by corrosion,
as is the case with pinholes, but in many other instances corrosion weakens the
pipe to the point that it will fail in the presence of forces originating from the
soil environment.
The type of corrosion and mode of failure causing loss of physical integrity
are highly system specific. External corrosion can be exacerbated by a low soi
l
redox potential, low soil pH, stray currents, and dissimilar metals or gal
vani
c
corrosion (Von Huben, 1999; Szeliga and Simpson, 2002; Romer et al., 2004;
Bonds et al., 2005). The life of the pipe is also influenced by the material used,
thickness of the pipe wall, use of protective outer wraps or coatings, application
of cathodic protection, and backfill materials and techniques. Internal corrosion

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
is influenced by pH, alkalinity, disinfectant type and dose, type of bacteria p
r
e
- sent
in biofilms, velocity, water use patterns, use of inhibitors, and many other
factors.
Corrosion is not well understood, particularly at the level of the local w
a
t
e
r
utility, such that insufficient attention has been given to its control (see a l
a
t
e
r
section in this chapter). Some utilities have tried to avoid the issue by u
s
i
ngplastic
pipe. Even so, unprotected metal materials are regularly used at the present
time, illustrating the water industry’s lack of attention to the problem. Ac-
cording to Romer et al. (2004), “approximately 72 percent of the materials re -
ported in use for water mains are iron pipe, approximately two-thirds of the re-
ported corrosion is in corrosive soils, and approximately two-thirds of the corro-
sion is on the pipe barrel.” In addition, metallic or cementitious pipe are often
designed on the basis of their hydraulic capabilities first and foremost, and cor-
rosion resistance is often a secondary consideration. The annual direct costs of
corrosion are estimated to be $5 billion (Romer et al., 2004) for the main distri-
bution system(not counting premise plumbing).
Issues with Service Lines
Recent evidence indicates that service lines (the piping between the water
main and the customer’s premises) and their fittings and connections (ferrules,
curb stops, corporation stops, valves, and meters) can account for a significant
proportion of the leaks in a distribution system (AWWA Water Loss
Control Committee, 2003). However, much less is known about what causes
structural failures in service lines compared to distribution mains and other
system components. Possibilities include improper techniques used during
installation that damage materials, improper tapping and flaring to make
connections, lack of corrosion prevention or use of corrosive backfill material,
damage during handling to plastic tubing, and kinks in copper tubing, and
excessive velocity. The Uniform Plumbing Code and International Plumbing
Code do not clearly address these issues, and local plumbing codes may not
either.
Many galvanized and lead pipe service lines are being replaced with c
o
p
p
e
ror
plastic pipe (chlorinated polyvinyl chloride or CPVC) (Von Huben, 1999).
CPVC and copper each have their benefits and weaknesses.
Installation of CPVC requires less skill compared to installation of copper,
although if workers are not careful installation can result in cracking and
damage to CPVC pipe. CPVC is better for corrosive soils and waters, while
copper is more resistant to internal biofilm growth. Buried CPVC pipe is
difficult to locate compared to metal or copper pipe because it does not conduct
electrical current for tracing. CPVC can impart a “plastic” flavor to water while
the copper pipe can impart a “metallic” flavor. With CPVC, low levels of
vinyl chloride can leach into the water. If manufacturers follow American
Society for Testing and Materials (ASTM) standards and are ISO 9002
certified, and certification includes NSF

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
International standards 14 and 61, the adverse conditions above can be m
i
n
i
-
mized.
Permeation
Permeation refers to a mechanism of pipe failure in which contaminants ex-
ternal to the pipe materials and non-metallic joints compromise the structural
integrity of the materials and actually pass through theminto the drinking water.
Permeation is generally associated with plastic pipes and with chemical solvents
such as benzene, toluene, ethylbenzene, and xylenes (BTEX) and other hydro-
carbons associated with oil and gasoline, all of which are easily detected using
volatile organic chemical gas chromatography analyses. These chemicals can
readily diffuse through the plastic pipe matrix, alter the plastic material, and
migrate into the water within the pipe. Such compounds are common in soils
surrounding gasoline spills (leaking storage tanks), at abandoned industrial sites,
and near bulk chemical storage, electroplaters, and dry cleaners (Glaza and Park,
1992; Geldreich, 1996). Permeation incidents have occurred at high-risk
sites, such as industrial sites and near underground chemical storage tanks, as
well as at lower risk residential sites (Holsen et al., 1991). In some cases the
integrity of the pipe has been irreversibly compromised, requiring the complete
replacement of the contaminatedsection.
Common pipe materials such as PVC, polybutylene, and polyethylene differ
in their chemical and physical structure, and thereby differ in their susceptibility
to being altered upon exposure to solvents and in permeation rates. In studying
BTEX and 1,3-dichlorobenzene, PVC pipe was found to be more
permeable than polyethylene pipe unless the polyethylene pipe was altered by
the solvents in contact, after which it can become more permeable to the
pollutants (Burlin- game and Anselme, 1995).
Human Activities that Lead to Contamination
A second major cause of physical integrity loss is human activity surround-
ing construction, repair, and replacement that can introduce contamination into
the distribution system. Any point where the water distribution systemis opened
to the atmosphere is a potential source of contamination. This is particularly
relevant when laying new pipes, engaging in pipe repairs, and rehabilitating
sites. For example, a Midwestern water utility experienced a noticeable in-
crease in the heterotrophic bacterial population of water from a newly installed
pipe and identified Pseudomonas fluorescens, Ps. Maltophilia, and Ps. putida as
the bacteria responsible for the increase (Geldreich, 1996). The same strains of
Pseudomonas were recovered from the sand used as an aggregate in making the
concrete lining for the new ductile iron pipe, implicating contamination during
construction and installation. More recently, workers in Camden, New Jersey,

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
were cleaning and lining a 30-inch water main when a parallel sewer line from
the post-Civil War era broke. Because of the proximity of the sewer line and the
possibility of contamination, officials decided to issue a boil-water alert until
water quality testing could show that no external contamination had entered the
main. Between 1997 and 1999, the Philadelphia water supply measured ele-
vated turbidity (>1 NTU) in about 12 to 14 percent of the samples that were col-
lected from newly installed water mains. This turbidity, or the particulate debris
captured on filters, was found to be largely iron oxides and rust (from the exist-
ing water mains still in service), vegetable material such as plant roots, and
backfill sand.
Incidents like these are not uncommon, as revealed in a survey by Pierson
e
t
al. (2002), who point out that pipe repair and installation have not been accom-
plished using the best available sanitary practices. This is captured generally in
Table 4-4, which summarizes the survey of distribution system workers at three
different utilities (eastern U.S., western U.S., and western Canada) on the poten-
tial for external contamination to occur during water main repair and replace -
ment activities. Given that the average number of main repairs a year for a sin-
gle utility ranges from 66 to 901 (which corresponds to 7.9–35.6 repairs per 100
miles of pipe per year) (Clark and Goodrich, 1989), it is clear that exposure of
the distribution systemto contamination during repair is an inescapablereality.
Unsanitary activity during construction, replacement, or repair can also lead
to the contamination of fittings and appurtenances. The use of inappropriate or
inferior materials, and the contact between dissimilar metals within fittings, can
also cause failures where they should not occur. Appurtenances can be improp-
erly installed in a flooded meter or valve pit which can allow contaminants to
enter under intrusion or can create corrosive conditions.
Backfill sand contaminating a new pipe at a water main construction site. Photo courtesy of Bu-
reau of Laboratory Service, PhiladelphiaWater Department.

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
TABLE 4-4 Potential for Contaminant Entry during Water Main Activities
Percentof Responses from Workers
at 3 DifferentUtilities (A, B, C)
Activity
Occurs Often Occurs
Sometimes
A B C A B C
Broken service line fills trench during installation 46 75 56 39 25 33
Pipe gets dirty during storage before i
n
s
t
a
l
l
a
t
i
o
n 53 75 22 43 25 33
Trench dirt gets into pipe during installation 24 100 39 37 0 44
Rainw ater fills trench during i
n
s
t
a
l
l
a
t
i
o
n 20 25 5 60 75 83
Street runoff gets into pipe before installation 30 0 11 61 38 67
Pipe is delivered dirty 4 25 17 33 63 22
Trash gets into pipe before i
n
s
t
a
l
l
a
t
i
o
n 24 0 0 56 50 11
Vandalism occurs at the site 15 0 0 35 0 5
Animals get into pipe before i
n
s
t
a
l
l
a
t
i
o
n 0 0 0 11 0 11
SOURCE: Reprinted, with permission, f rom Pierson et al. (2002). © 2002 by American Water Works
Association.
New pipe materials are not sterile, whether they have been kept well pro-
tected or not. Indeed, according to a survey (Geldreich, 1996) about 18 percent
of new pipe, irrespective of pipe material and size, failed upon testing the water
to approve it for release. In one case, Geldreich reported the finding of a piece
of wood construction material embedded in a new main that contributed to coli-
form contamination. Thus, new materials need inspection and some form of
disinfection before they are exposed to drinking water. The physical cleanliness
of new pipe is important to guarantee that post-installation disinfection will be
successful (Geldreich, 1996). The installation or rehabilitation of facilities such
as storage reservoirs with floating covers must include water quality checks for
health and aesthetic considerations and not assume that new materials and their
installation will be free of contaminants (Krasner and Means,1986).
The installation process for buried pipe is not the only place where c
on-
tamination can occur. The storage of pipe, pipe fittings, and valves along road-
ways or in pipe yards prior to installation can expose them to contamination
from soil, stormwater runoff, and pets and wildlife. Damage to pipes prior to
their installation is also possible, such as during pipe storage and handling
or actual manufacturing defects such as surface impurities or nicks.
Regardless of where and how materials become contaminated, the hope i
s
that post-installation disinfection will be sufficient to kill any introduced bacte-
ria. This is not always the case, however, as evidenced by a coliform event in
Florissant, Missouri in 1984 (Geldreich, 1996). The coliforms detected in a
storage tank were thought to be the result of inadequate disinfection following
new pipe installation or repair. Unfortunately, contaminated water subsequently
passed into the distribution network. No direct public health outcome was re-
ported; however, the “repeated reissuance of boil-water orders caused a loss of
confidence” in the water utility by the public (Geldreich, 1996).

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
It is unclear how often faulty materials are installed or good materials a
r
e
improperly installed because most utilities do not keep records that would f
a
c
i
l
i
-tate
the evaluation of this problem. Sufficient standards exist for materials quality
and for the certified testing of materials quality. Water utilities can incorporate
existing standards into contracts and specifications for materials and materials
installations, and most if not all water utilities already do this. Water utilities can
also certify and decertify manufacturers and contractors.
Absence of a Barrier
Points in a plumbing system where nonpotable water comes into contact
with the potable water supply are called cross connections, and a backflow event
occurs when nonpotable water flows into the drinking water supply through a
cross connection. The use of backflow prevention devices can be extremely
effective in eliminating this type of contamination event. The absence of such
devices, which is widespread given the highly variable nature of cross -
connection control programs across the country, constitutes a potential threat to
the physical integrity of distribution systems. Backflow protection devices are
seldom installed on domestic service lines and even on many small business
service lines. Operational failure of devices that are in place is akin to having
the device not be present.
Similar issues surface for storage facilities that do not have adequate p
r
o
t
e
c
-tion
to prevent their contamination. There are 154,000 treated water storage facilities
in the United States (AWWA, 2003) encompassing a variety of types including
elevated tanks, standpipes, open and covered reservoirs, underground basins,
and hydropneumatic storage tanks. Storage facilities are susceptible to external
contamination from birds, insects, other animals, wind, rain, and algae. Indeed,
coliform occurrences have been associated with birds roosting in the vent
ports of covered water reservoirs (Geldreich, 1996). This is most problematic
for uncovered storage facilities, although storage facilities with floating
covers are also susceptible to bacterial contamination due to rips in the
cover from ice, vandalism, or normal operation. Even with covered storage
facilities, contaminants can gain access through improperly sealed access
openings and hatches or faulty screening of vents and overflows. Four reported
waterborne disease outbreaks have been associated with covered storage tanks,
in particular, a Salmonella typhimurium outbreak due to a bird
contamination of a covered municipal water storage tank (Clark et al.,
1996). Such events can be aggra- vated by the loss of disinfectant residual that
storage tanks typically experience with increasing water age.

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
Preparing to sample stored drinking water from the access hatch of a floating cover on a distribu-
tion system reservoir. Photo courtesy of Bureau of Laboratory Service, Philadelphia Water Depart-
ment.
CONSEQUENCES OF A LOSS IN PHYSICAL INTEGRITY
A loss of physical integrity implies a breakdown in the barrier that prevents
contact between the external, unsanitary environment and the internal, drinking
water environment. The water quality effects that can result include the
introduction into the distribution system of microbial and chemical
contaminants, debris, and particulate matter, sometimes accompanied by
changes in water color, turbidity, taste, and odor. Whether a breach in physical
integrity results in exposure of the public to contaminants at levels posing an
unacceptable risk is dependent on site-specific conditions. As revealed in
Chapter 3 and Appendix A, most documented cases of waterborne disease
outbreaks that can be attributed to distribution systems have been
caused by breaches in physical integrity. For example, a review of 619
reported waterborne disease outbreaks in the U.S. between 1971 and 1998 found
that over one-half of the outbreaks in distribution systems were due to cross
connections and backflow (Craun and Calderon, 2001). Of the 12 largest
outbreaks, seven were associated with cross connections, three with
contaminated storage tanks, and two with water main contamination during
installation or repair. Overall, in community water systems, cross
connections were the number one cause of distribution systemrelated
outbreaks, contaminated mains were number two, and contaminated storage
facilities were number three. In non-community water systems, con-
taminated storage facilities were the second leading cause. The
contaminants

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
involved have ranged from pathogens such as Giardia, Norwalk virus-
like agents, hepatitis A virus, Campylobacter, Salmonella, Shigella, and E.
coli 0157:H7 to chemical contaminants such as copper (the most commonly
reported chemical), chlordane, nitrite, ethylene glycol, and oil (Craun and
Calderon, 2001).
Not all of what can enter a distribution system from a failure in a physical
barrier will have a known or direct health impact. Particulate matter and other
debris can gain entry during main breaks; reservoir cover, hatch, or vent failures;
and during repair, installation, and maintenance activities. Utilities have re-
ported particulates in distribution system water that included such things as
sand, patina, pipe joint materials, rubber gasket chunks, insect pieces, plant
fibers, and glass chips, many of which are likely to have no direct health
impacts (Booth and Brazos, 2005).
Changes in taste and odor, turbidity, and color typically provoke customers
to complain (Burlingame, 1999a,b; McGuire et al., 2004), but may present little
direct public health risk. This is because aesthetic problems often occur at con-
taminant concentrations far below the known health effects levels. For example,
color problems derived from iron or manganese introduced into drinking water
during a backflow event from a fire service connection or a heating system are
unlikely to pose a health risk. On the other hand, color problems can also indi-
cate backflow events that have health risks associated with them such as with
ethylene glycol or corrosion inhibitors from HVAC and fire service connections.
The sections below discuss the typical consequences of the loss of physical in-
tegrity in pipes, fittings and appurtenances; storage facilities; and backflow pre -
vention devices.
Contamination of Mains, Fittings, and Appurtenances
Pipe interior, appurtenances, and related materials can be exposed to micro-
bial and chemical contaminants in the external environment (1) during
water main failures and breaks and (2) due to human activities to install new,
rehabilitate old, or repair broken mains and appurtenances. When a pipe break
or failure occurs, there is immediate potential for external contamination
from soil, groundwater, or surface runoff to enter the distribution system or
come into contact with the pipe interior in the area of the failure. Other less
dramatic types of structural failure, such as the development of cracks or leaks
in pipe, pipe joints, or appurtenances, can also provide avenues for distribution
system contamination during periods of low pressure or a pressure transient—
an event known as intrusion. Intrusion refers to the flow of nonpotable water
into drinking water mains through leaks, cracks, submerged air valves, faulty
seals, and other openings resulting from low or negative pressures. Discussed
in greater detail in the next chapter, intrusion can exist undetected for long
periods of time. A prominent example of a waterborne disease outbreak
being caused by a main break and intrusion is presented in Box4-1.

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
BOX 4-1
Waterborne Disease Outbreak Associatedwith Main Breaksand Intrusion:
Cabool, Missouri
In the w inter of 1989–1990, Cabool, Missouri, a town of app
ro
xi
m
atel
y 2,100 people,
experienced a large outbreak of E. coli O157:H7. A total of 243 cases were reported, with
32 hospitalizations and four deaths. This was the first documented waterborne outbreak
of
E. coli O157:H7 and the largest w aterborne outbreakof E. coli O157:H7 b
e
f
o
r
e the 2000
outbreakin Walkerton, Canada.
The tow n’s water system (untreated groundwater) was i
m
p
l
i
c
a
t
e
din the outbreak. Two of the
tow n’s four wells were operating at the time of the outbreak: one was 305 meters deep
and the other was 396 meters deep. Both wells had protected w ellheads, and the
monitoring data from the ten years before the outbreak indicated that no coliforms had been
detected in either well. Investigation of the outbreak indicated that the distribution
system was not well maintained and w as vulnerable to sewage contamination at several
points. Approximately 35 percent of the total flow was lost in the system—suggesting
leaks, inaccurate meters, or unmetered connections. The town sewer system w as also in
poor condition and operating beyond capacity, resulting in regular sewage back-ups and
overflows.
As with most waterborne disease outbreaks, a constellation of r
i
s
kfactors contributed to
this outbreak. In mid-December 1989, unusually cold w eather caused two large water
mains and 45 in-ground water meters to fail (Figure 4-1). Ten cases of bloody diarrhea
were reported to the local health department on January 4, 1990. A boil-water order was
issued on January 5, and water chlorination was initiated on January 12. Analyses of the
temporal distribution of the cases indicated that the first cases occurred seven days before
the first water main break (December 23), and the last case occurred three days after the
implementation of w ater chlorination (Figure 4-2). The early cases may have been due to
leaks and holes that developed prior to the main break. There was a small increase in the
incidence of diarrhea after the first main break and a large increase in diarrhea cases
about four days after the second main break on December 26.
FIGURE 4-1 Map of Cabool, Missouri w ith sites of watermain breaks.
SOURCE: Reprinted, with permission, f rom Swerdlow et al. (1992). © 1992 by A
m
e
r
i
c
a
n College of
Phy sicians.

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
In addition, replacement of the failed water meters may have f
u
r
t
her contributed to
contamination of the distribution system. During the replacement of the meters and main
break repairs, the lines were subjected to “limited flushing” but were not disinfected, and no
water samples were tested for microbial indicators to examine the water quality before
bringing the lines back into service. Although sewage overflow into the distribution system
via the main breaks and intrusionwas believed to be responsible for the outbreak, microbial
contamination of the distribution system could not be confirmed. Only tw o water samples
from the distribution systemw ere collected (on December 18 and January 3) and analyzed,
but neither sample w as collected from the areas with the highest concentration of cases.
Hydraulic modeling of the system by Geldreich et al. (1992) reinforced the evidence that
the second main break had the potential to contaminate a greater portion of the distribution
system, including the northern part of the tow n w here 36 percent of the cases occurred.
This outbreak illustrates how, despite a clean groundwater s
o
u
r
c
e
, lack of disinfection
combined w ith poorly maintained water and sewer lines, unusually cold weather, and cas-
ual line replacement practices led to a large drinking water outbreakwith fatalities in a small
tow n in an industrialized country.
FIGURE 4-2 Cases of diarrheal disease among city residents. SOURCE: S
w
e
r
d
lowet al. (1992).
SOURCE: Reprinted, with permission, from Swerdlow et al. (1992). © by American College
of Physicians.
SOURCES: Swerdlow et al. (1992), Geldreich et al. (1992), Hrudey and Hrudey (2004).

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
The storage, installation, rehabilitation, and repair of water mains, fittings,
and appurtenances provide another opportunity for microbial and chemical c
o
n
-
tamination of materials that come into direct contact with drinking water. Pi
er-
son et al. (2001) noted that this was particularly prevalent during the handling
and storage of distribution system materials and during their installation in the
trench. Indeed, a survey of water utilities found that about 14 percent
experienced positive coliform samples from 1 to 10 percent of the time that
new mains are checked before they are released (Haas et al., 1998).
Studies have demonstrated that the soil surrounding buried pipe can be c
on-
taminated with fecal indicator microorganisms and pathogens (Kirmeyer et al.,
2001). Besides contaminated soil, runoff from streets and agricultural land can
be highly concentrated with microbiological and chemical contaminants (Make-
peace et al., 1995), and this runoff can contaminate pipes during a main break,
during the unprotected storage of pipe materials, and even during pipe installa-
tion in the trench. One of the culprits in this regard are sewer lines that run in
close proximity to distribution system mains. Leaking sewers can contaminate
the soil and groundwater in the area of a water main or a trench where main ac-
tivity will take place. The general rule is that there should be a horizontal sepa-
ration of at least 10 ft (3 m) between water and sewer lines, and that the water
line should be at least 1 ft (0.3 m) above the sewer (although variations to this
general rule may occur from state to state). This rule, however, is fairly recent
in comparison to the average age of the nation’s buried infrastructure.
A second major mechanism of pipe failure is permeation, where contami-
nants external to the pipe materials and non-metallic joints compromise t
he
structural integrity of the materials and actually pass through them into the
drinking water. Taste and odor events are common consequences of permeation
of plastic pipe given the types of contaminants involved. For example, in one
case solvents trapped beneath a polyethylene wrap and soil migrated
through plastic pipe and pipe connections to contaminate the drinking water in
the service lines (Burlingame and Anselme, 1995). Because the solvents were
derived from a hot-butyl rubber coating applied to the external surface and
ferrules of a ductile iron main, they included toluene, indan, indene,
naphthalene, xylene, and benzofuran. The event was initially detected first by
customer complaints about off odors. In addition to the taste and odor issues,
continued exposure to solvents can change a pipe’s integrity and eventually
lead to pipe failure.
Although there is the potential for water quality degradation as a result o
f
the permeation of plastic pipe, the health impacts associated with such permea-
tion are not well documented nor are they expected to be significant. In some
permeation incidents, the concentrations of certain chemicals have been shown
to reach levels in the low parts per million, which are well above their respective
maximum contaminant levels (MCLs) (AWWA and EES, Inc., 2002). How-
ever, these MCLs are based on long-term exposure, and the short-term risk lev-
els for these chemicals are generally much higher. In the case of permeation by
gasoline components, the taste or odor thresholds of the majority of these
chemicals are below the levels that would pose a short-term risk
(EPA,

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
2002a,b,c,d), such that customers would notice an objectionable taste or odor i
n
the water before significant exposure. In addition, these high concentrations
would be expected to occur during worst case situations where water has been in
contact with the affected pipe for a considerable length of time. During periods
of normal water use, these concentrations would be expected to be much lower.
It should be noted that the taste and odor thresholds for some contaminants may
be above the MCL, in which case permeation of these chemicals could result in
undetected long-term exposure if monitoring of these chemicals in the distribu-
tion systemis not conducted.
Contamination of Storage Facilities
Although they may suffer from structural failures, storage facilities are most
susceptible to external contamination due to the absence or failure of an essen-
tial component, such as a cover, vent, hatch, etc. The complete absence of a
cover or vent on a storage facility can allow birds access to the tank and subse-
quently introduce microbial pathogens such as bacteria and parasites to the wa-
ter within. For example, in the winter of 1993 a waterborne disease outbreak of
salmonellosis in Gideon, Missouri, was traced back to the contamination by
birds of the public water supply’s distribution systemstorage tank (see Box 4-2;
Clark et al., 1996). Indeed, one water storage tank connected to the distribution
systemwas found to have holes in the top and bird feathers floating in the water.
Two other storage tanks were found to be in similar need of maintenance, and
pigeons were found roosting on the tanks. Birds, and consequently bird excre-
ment, are probably the biggest concern for storage tanks and reservoirs
with floating covers. Sea gulls, for example, can be found roosting at storage
facilities. Open reservoirs also offer the opportunity for detrimental changes in
water quality because of exposure to the atmosphere or sunlight, such as changes
in pH, dissolved oxygen, and algal growth.
Even when covered, storage facilities can suffer from algal growth on the
tops of floating covers that can gain entry into the tank through rips and tears or
missing hatches. Algae can also be airborne or carried by birds and gain entry
into storage tanks through open hatches and vents. Algae increase the chlorine
demand of the stored water, reduce its oxygen content upon their degradation,
affect taste and odor, and in some cases release byproducts.
Chemical contaminants gain access to storage facilities via air pollution a
n
d
surface-water runoff into open storage reservoirs. For example, accidental s
p
i
l
l
sof
chemicals during truck transport on highways adjacent to reservoirs are a
potential threat, and can be very serious if the chemicals are present in a concen-
trated form and highly toxic (Geldreich, 1996). Surface-water runoff into open
reservoirs can also introduce pesticides, herbicides, fertilizers, silt, and
humic materials from nearby land. The potential for chemical contamination of
storage facilities continues to be overlooked in regulations in comparison to
microbial

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
BOX 4-2
Waterborne Disease Outbreak Associated with
Contamination of Water Storage Tanks:Gideon, Missouri
In 1993, the town of Gideon, Missouri (pop. 1,100) suffered from anoutbreak of sal-
monellosis that affected more than 650 people and caused seven deaths. The Gideon
water system consisted of two deep wells (396 meters) with no treatment or disinfection
and a w ater distribution system that dated to the 1930s. In early November 1993, a
cold snap caused a thermal inversion in the water storage tanks that resulted in taste and
odor problems. In response, the water system was systematically flushed on November
10. The first cases of acute gastroenteritis were reported on November 29 and
diagnosed as Salmonella typhimurium. How ever, the outbreak investigation later revealed
that diarrhea cases in Gideon started around November 12 with a peak incidence around
November 20. By early December, there was a 250 percent increase in absenteeism in the
Gideon schools and a 600 percent increase in anti-diarrheal medication sales. Over 40
percent of nursing home residents suffered from diarrhea and seven people died. The
outbreak was not linked to the water system until December 15 when the water system
samples were reviewed and investigative water sampling w as initiated. A boil-water
advisory was issued on December 18. On December 22nd, emergency chlorination was
added to the production well and the two municipal storage tanks were
superchlorinated. The last reported cases occurred on December 28.
Water samples collected from a hydrant in the distribution sy
stemon December 16,
17, 20, and 21 w ere positive for total coliforms, and the samples from December 20 and 21
were also positive for fecal coliforms. Inspection of the two municipal w ater storage tanks
suggested that the outbreakwas probably caused by bird feces in one or more of the tanks.
The larger of the tanks was in disrepair and had birds roosting on the roof. A third private
storage tank had an unscreened overflow pipe and a hole at the top of the tank that w as
large enough for birds to enter. This private tank had been drained on December 30, but
tJhe outbreak strain of S. typhimurium was detected in samples of sediment collected
on January 5, 1994. The remaining w ater on the bottom of the tank was described as
black
contamination. For example, the Long Term 2 Enhanced Surface Water T
r
e
a
t
-ment
Rule requires that water systems with uncovered finished water stor
a
ge
reservoirs cover the reservoir or treat the reservoir discharge to the distribution
systemto achieve a 4-log virus, 3-log Giardia, and 2-log Cryptosporidium inac-
tivation, the latter of which would not protect against chemical contamination.
However, it should be noted that EPA has published a Guidance Manual on Un -
covered Finished Water Reservoirs (EPA, 1999) that addresses chemical con-
tamination. Although the actions contained in the manual are not mandated,
some states (such as California) are requiring water systems to implement them.
Contamination Due to the Absence or
Operational Failure of Backflow Prevention Devices
Backflow events via unprotected domestic, commercial, industrial, and fire
connection services can introduce contaminants into the potable water supply,
with potentially profound health implications. A recent survey (USC, 2002)
found that more than 95 percent of sampled homes had direct or indirect cross

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
connections (“direct” meaning a cross connection subject to both
backpressure and backsiphonage while an “indirect” cross connection is subject to
backsiphonage only). Because of the enormous range of contaminant sources
involved, as well as the number of unprotected cross connections, backflow
events collectively constitute the greatest potential health risk from
distribution system contamination. Whether an individual backflow event
poses a risk depends on the type of the contamination, the length of an
individual’s exposure to the contaminated water, and other factors. A survey
of water utilities in North America found that 28.8 percent of cross connections
resulted in bacteriological contamination whereas 26.1 percent resulted in
chemical contamination, and 29.8 percent resulted in both bacteriological and
chemical contamination (Lee et al., 2003).
Although their potential to occur is high in all systems, backflow events a
r
ea
particular concern in dual distribution systems where one line carries a n
o
n
p
o
-table
water source that may become connected with a potable source in the otherline.
Generally, the nonpotable line is a substantialhealth risk because it c
a
r
r
i
e
s
and very turbid, w ith rust, suspended particles, and bird feathers f
l
o
a
t
i
n
g on the top. Ini- tially
attention was focused on the private tank as the source of the outbreak as reported by
Skala (1994). However, an in-depth hydraulic analysis of the Gideon system, conducted as
part of the outbreak investigation, raised questions about the possibility of the private tank
being the source of the outbreak. A subsequent review of as-built draw ings of the Gideon
system by Missouri Department of Natural Resources personnel revealed that the
private tankwas separated from the municipal system by a functioning backflow prevention
valve. In a subsequent hydraulic analysis the private tank w as eliminated as a
contamination source for the outbreak, w hich led to results that w ere consistent w ith
the behavior of the system as observed during the outbreak scenario. This analysis also
pointed to the largest municipal tank as the most likely source of the outbreak. A visual
inspection of the large municipal tank revealed broken and rusted hatches and bird parts
and feathers on the top of the tank and floating on the surface of the tank w ater. Both
Clark et al. (1996) and Angulo et al. (1997) concluded that the large municipal tank
w as the source of the outbreak.
In the end, the outbreak investigation concluded that the cold w
e
ather in early No-
vem ber caused a thermal inversion in the w ater storage tanks that mixed the contaminated
upper layers of stored water with the water entering the distribution system. The w ide-
spread flushing program on November 10 served to draw more contaminated storage tank
water into the distribution system than under normal operation. The large discharge of the
stored water over a short period of time may also have stirred up sediments in the tank and
introduced them into the distribution system. Hydraulic modeling indicated that the part of
the distribution system that served the school and the contaminated fire hydrantwould have
received water from the problem municipal tank w ithin the first six hours of flushing. Other
contributing factors included late recognition of the outbreak by the public health
authorities, late recognition that the outbreak was linked to the public water supply, and a
low rate of compliance w ith the boil-w ater order.
SOURCES: Skala (1994), Clark et al. (1996), Angulo et al. (1997), Hrudey and Hrudey (2004).

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
162 DRINKING WATER DISTRIBUTION SYSTEMS: ASSESSING AND REDUCINGRISKS
reclaimed water containing chemicals and microbial pathogens at levels
exceed-ing water quality standards.
One of the most well known backflow events affected two Chicago h
o
t
e
l
sin
1893 during the World’s Fair (Columbian Exhibition). A loss in water pres - sure
in the distribution system caused backsiphonage through cross connections
which contaminated the hotels’ drinking water. An amoebic dysentery outbreak
resulted in over 1,400 illnesses and at least 98 deaths (Von Huben, 1999). It is
likely that the frequency and magnitude of contamination events due to cross -
connections is underreported, especially where premise plumbing is involved.
Box 4-3 describes a waterborne disease outbreak associated with an unprotected
cross connection.
DETECTING LOSS OF PHYSICAL INTEGRITY
In some cases, a loss in physical integrity might actually be observed, such
as a hole or tear in a reservoir cover, a missing vent or hatch on a storage facil-
ity, or a flooded meter or valve pit. Other structural failures, such as pipe leaks,
tend to be much less obvious. The ability to predict and detect a failure in
a material barrier is a desired capability for any water supplier. Structural failure
is predictable for all major infrastructure components given information about
materials composition and age and the surrounding environment. Structural
integrity and operational performance should be confirmed on a regular basis via
testing and inspections, particularly for backflow devices and storage facilities.
The lack of standards and proper training can be predictive of a loss of physical
integrity due to the improper installation, repair, or replacement of infrastructure
components.
Predictions of structural failure can often be made based on historical i
n-
formation. For example, much is known about iron pipe based on years of ac -
tual experience. Cast iron pipe has been shown, under the right conditions, to
last 100 years and more. When first introduced, cast iron pipe had no internal
lining or external coatings to protect it from corrosion. After 1860, most pipes
were lined with a molten tar pitch, and after 1922 some pipes were lined with a
cement-mortar lining which in turn is sometimes protected by an asphaltic seal-
coat. By the mid 1950s, ductile iron pipe came into use that has about twice the
strength as cast iron but with a reduced wall thickness. This thinner wall pipe is
more forgiving during installation and is more resistant to damage (AWWA,
2005b). It can be protected from external corrosion by a polyethylene wrap.
AWWA/ANSI standards exist for pipes, joints, wraps and epoxy coatings, and
fittings, and they provide information on the lifetimes of these materials.
The drive to predict and prevent failures varies depending on the c
onse-
quence of the failure (Makar and Kleiner, 2002). For example, a branched dis-
tribution systemhas greater consequences associated with failure compared to a
grid/looped system. There has been much attention given to predicting pipe
failures, and more attention is needed in order to better predict overall system

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
reliability (Grigg, 2005). Most useful would be a user-friendly guidance manual
for utilities regarding the failure mechanisms of different types of pipes and h
o
w
to
use the various types of information on the current condition of the pipe t
o
determine its expected lifetime.
Table 4-5 summarizes common methods used to detect a failure in a m
ate-rial
barrier based on the types of failures that can occur. Inspection, direct t
e
st
-ing, and
consumer complaints play a significant role. Water quality testing m
a
ybe the least
effective means for detecting a loss in physical integrity, and thus i
s
not discussed
extensively.
The sections below discuss the role of inspections, condition assessment o
f
infrastructure, leak detection, main break monitoring, and water quality monitor-
ing for both prediction and detection of physical integrity. It is hoped that water
utilities will embrace these activities and keep appropriate records in order
to identify those factors that lead to failures, recognize early warning
conditions, and improve their overall prediction capabilities. Integrating all of
these data streams in order to plan how and where to rehabilitate, repair, or
replace infrastructure is a significant challenge for water utilities and yet
essential to being proactive in deterring contamination events that would
pose a risk to public health (Martel et al., 2005). This chapter does not focus
on failure analysis, which is the systematic investigation into the causes of
pipe failure by visual
TABLE 4-5 Examples of Ways to Detect a Loss in Physical Integrity
Component Mechanismof
Integrity Loss
Detection by
Pipe Permeation VOC testing, investigate customer complaints
about taste/odor
Structural f
a
i
l
u
r
e
(leak)
Structural f
a
i
l
u
r
e
(break)
Leak detection, investigate customer complaints
Investigate customer c
o
m
p
l
a
i
n
t
s
,pressure monitoring
Fitting and
appurtenance
Improper installation Inspection
Unsanitary activity Inspection, w ater quality testing
Structural failure Inspection, pressure monitoring, investigation of
customer complaints, leak d
e
t
e
c
t
i
o
n
,
detection of
operational failures
Improper installation I
n
s
p
e
c
t
i
o
n
Unsanitary activity I
n
s
p
e
c
t
i
o
n
Storage f
a
c
i
l
-
ity
w all, r
oof,
Structural f
a
i
l
u
r
e
(crack, hol
e)
Inspection, w ater quality testing
cover, v
e
n
t
,
hatch
Backflow
prevention
Absence of Inspection, w ater q
u
a
l
i
t
y
testing
Improper installation Inspection
Unsanitary activity Inspection, w ater quality testing
Absence of Inspection, investigate customer c
o
m
p
l
a
i
n
t
s
Improper
installation Inspection, investigate customer complaints
devices
Operational failure Inspection, investigate customer complaints

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
BOX 4-3
Waterborne Disease Outbreak Associatedwith a Cross-Connection
in the Water Distribution System:The Netherlands
A new housing development in the central part of the Netherlands w as built w ith a dual
distribution system. One set of pipes carried drinking w ater and a second set of pipes car-
ried w ater from the same source that received partial treatment and w as designed to be
used for toilet flushing, laundry, and garden irrigation (“economy water”). Both the drinking
water and the economy water originated from a surface water source and were treated by
coagulation, flocculation, sedimentation, and rapid sand filtration. The drinking w ater was
further treated by dune filtration. Approximately 30,000 households were served by
this dual distribution system.
On December 3, 2001, the w ater utility received complaints from t
w
opeople living in
one neighborhood of the development that the drinking water had an unusual taste and
odor. Drinking water samples collected on December 4 indicated unusually high
coliform levels. On December 5 and 6, the water utility issued a boil-water advisory. On
December 6, a local physician informed the public health service that he had seen an
unusually number of cases of gastroenteritis with nausea, vomiting, and diarrhea in his
clinic over the past few days. Further investigation of the water system revealed that when
maintenance w ork was done on November 29, the drinking water system had been
connected to the economy water system in order to flush and clean it, and that the w orkers
had failed to remove the cross-connection when the economy water systemwas put on-line
again. In addition, the economy water supply lines were under higher pressure than
the drinking w ater lines, which forced the economy water to enter the drinking water
distribution system. This cross- connection was removed on December 6, E. coli
concentrations in the drinking water system dropped to below detection limits on
December 12, and the boil-w ater advisory was lifted on December 17. A 1,000-liter
sample of the economy w ater collected on December
20 w as found to contain approximately 1.6x103
PCR-detectable units o
f norovirus. South- ern
blot hybridization identified the norovirus isolate as a genogroup I virus.
Tw o retrospective studies were conducted to determine the effect of this cross-
connection on the rates of gastroenteritis in the housing d
e
v
e
lo
p
m
e
n
t
. T he first was a retrospective
cohort study that compared the incidence of gastrointestinal symptoms and other health
symptoms during the period of November 29 through December 9 among 412 households
in the area exposed to the cross-connection to the incidence of symptoms among 486
households in an adjacent control area that also had the dual distribution system but was
not affected by the cross-connection. Data on symptoms and normal daily w ater
consumption w ere collected by a one-time questionnaire that was mailed out to over 900
households in the exposed area and over 1,600 households in the control area. In
addition, over 400 stool collection kits were mailed to randomly selected households
equally divided between the exposed neighborhood and the adjacent control neighborhood,
and the households were asked to provide a stool specimen from one member of the
household w ho had recently experienced gastroenteritis.
The results of this study indicated that during the period of N
o
v
e
m
ber 29 through De-
cem ber 9, households in the exposed area experienced significantly higher illness rates
than households in the adjacent control area. In the exposed area, the rates of
diarrhea, vomiting, nausea, abdominal pain, and chills w ere twice as high (38–54 percent of
exposed households) as the rates reported from the control area (19–28 percent). The
reports of blood in stoolwere about four times higher, and the reports of itching were over
six times higher in the exposed neighborhood compared to the control neighborhood.
How ever, these symptoms were rare (1–3 percent blood in stool, 1–5 percent itching).
Households in the exposed areawere 1.5 times more likely to seek medical care during the
period of November 29 to December 9 than households in the controlarea. There w as no

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
significant difference in the reported rates of coughing and sneezing in both neighborhoods
or in the reports of symptoms that occurred after December 9. Interestingly, the distribution
of symptom reports over time indicated that a peak in gastroenteritis symptoms occurred in
both communities around December 3–5, but the peakw as low er and shorter in the control
community. Although the adjacent control community experienced a lower incidence of
gastrointestinal symptoms, the rates during this time w ere still higher than normal. Also,
there was clear evidence of a dose-response effect in both communities w ith significantly
higher rates of households w ith diarrhea in those who reported higher water consumption
(chi square for trend of 51.26 for the exposed area and 23.47 for the adjacent control area,
p<0.01 for both communities).
Analyses of the 53 stool samples that were sent from 31 e
x
p
o
s
e
d households and 22
households in the adjacent control area yielded one norovirus genogroup I strain and one
Giardia lamblia isolate from the exposed neighborhood and one norovirus genogroup II
strain from the control neighborhood. The second norovirus strain came from a household
that reported gastroenteritis symptoms after December 9.
The second retrospective investigation was a survey of two h
e
a
lthcare facilities for cases
of gastroenteritis during the period of November 26 – December 12, 2001. One
health facility served both the exposed (pop. 1,866) and adjacent control (pop. 2,875) areas
in the cohort study. The other health facility was farther away and served a different part of
the housing development (pop. 5,788) that was not exposed to the cross-connection inci-
dent in the distribution system but w as still served by the dual distribution system. Based on
the computer database of date of visit, patient address, and diagnosis codes, the inci-
dence of gastroenteritis cases seeking medical care was compared between the three
communities. Residents in the exposed area had a rate of 19.8 cases per 1,000 inhabi-
tants compared with 7.0 cases per 1,000 in the adjacent control area and 3.3 cases per
1,000 in the more distant control area. The rate of gastroenteritis cases seeking medical
care increased markedly in the exposed area and moderately in the adjacent control area
on December 3 –5 and again on December 10–11 (weekdays). There was no change
in the rate of diagnosed cases of gastroenteritis in the distant control area during this
time period.
Taken together, the results of these two retrospective studiessuggest that an outbreak
of gastroenteritis, probably due to noroviruses, occurred shortly after the cross connection
between the economy water distribution system and the drinking water distribution system
was created. It is notable that there appeared to be an increased risk of gastroenteritis in
the adjacent control community that reportedly was not affected by the cross-
contamination incident. This may have been due to secondary transmission from the ex-
posed community to the adjacent control community by other routes (food, person-to-
person) because these two communities shared several facilities located in the control
community (schools, health center, supermarket) or consumption of contaminated w ater by
visitors from the control community to the exposed community. Studies of the surface wa-
ter source in the spring of 2001 indicated high concentrations of noroviruses of up to
1.4x104
PCR detectable units per liter, and it is unlikely that these v
i
r
u
s
e
sw ould have been
completely eliminated by the treatment processesused for the economy water. Norovi-
ruses (1.6x103
PCR detectable units per liter) were also detected in t
h
eeconomy water on
December 20. It is possible that exposure to the economy water in this system through
aerosols fromtoilets, laundry, or garden irrigation may have posed some risk to the inhabi-
tants in this development even w ithout the cross-connection incident.
SOURCE: Cooperativ e Research Centre f or Water Quality and Treatment (2003).

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
means and other inspection tools in order to determine the point wherethe f
a
i
l
- ure
started and the specific type of failure. Makar et al. (2005) discusses the actors
that can cause failure such as flaws in the material (inherent to its manufacture
or produced afterwards), forces that exceeded the design strength, design that
did not account for normal operating loads, or some combination of the above.
Inspections
Regular inspections of the distribution system, either visual examination o
f
the various structures or via acoustic leak detection and pressure m
oni
tori
ng
(discussed below), provide the most direct way to detect a failure in the material
barrier. Storage facilities need to be inspected on a routine basis for vandalism,
settling, cracking and spalling, seepage, leakage at seams and joints and in the
roof, missing hatches and vents, rust and corrosion, cathodic protection, and
failing structures (AWWA, 1986). A second critical type of inspection is to
check both the material integrity and the cleanliness of pipe prior to installation.
Even though it is often assumed that pipe is inspected before it leaves the fac-
tory, damage to the spigot end of the pipe, the exterior, and the internal
lining can occur during pipe storage and handling. Another reason to
inspect pipe prior to installation is to detect manufacturing defects such as
surface impurities or nicks, which are likely to induce corrosion and pitting
once installed (Von Huben, 1999). Finally, pipe should be examined before
installation for oil, dirt, grease, animals, and foreign matter; if found, the
pipe should be cleaned out with a strong hypochlorite solution.
An important opportunity in this regard is the sanitary survey, which i
sa
broad review and inspection program for a water utility that occurs once every
three to five years. The survey might reveal an absence of (1) training and certi-
fication, (2) use of standards, and (3) routine inspections, all of which could be
predictive of a loss of physical integrity. This is because a lack of training, cer-
tification, inspection, and standards often lead to the improper installation and
application of materials (for example, using the wrong backflow prevention de-
vice or installing plastic pipe in contaminated soils).
Monitoring the Condition of Buried Infrastructure
The various tools available for locating buried pipe include ground-probing
radar, metal detectors, magnetic locators, and radio transmission units for metal-
lic pipe (Von Huben, 1999). Similar methods can be used to detect non-metallic
pipe if metallic tapes or tracer wire was installed with the pipe. Locating
pipe i
s
the first step of condition assessment. A condition assessment is based on
the assumption that materials or infrastructure components deteriorate, with the
goal of gathering information to predict the need for repair, rehabilitation,
or re-

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
placement (Grigg, 2005). The three steps of condition assessment are (Morri-
son,2004):
1. Develop an up-to-date inventory of assets. With pipes, a Geographic
Information System (GIS) can be used to collect the following data:
diameter, material, classification/grade, wall thickness, joint type, installation
date, lining and coating types, corrosion protection system, depth of burial, soil
conditions, groundwater level, bedding classification, and history of problems
(Shamsi, 2005).
2. Inspect the internal pipe condition, pipe wall condition, pipe environ-
ment condition, and leakage (which can be difficult and costly to accomplish for
buried, in-use pipe).
3. Rate the condition of the asset.
There are five categories of pipe rating used during condition
assessment (Morrison, 2004):
Rank 5. In danger of immediate failure, requires emergency repair or r
e
-
placement as soon as possible to avoid jeopardizing public health and safety.
Rank 4. Severely deteriorated and in need or repair, renewal, or
replacement. Should be addressed immediately.
Rank 3. Mildly deteriorated, short-term performance just adequate; how
-
ever, will require renewal or replacement soon. Capital improvement plans are
needed with more frequent inspections.
Rank 2. Minor deterioration, performance adequate. An inspection or a
s
-
sessment plan should exist.
Rank 1. Little to no deterioration, performance more than adequate.
Condition assessment requires information from existing pipe to help p
r
e
-dict
the lifetime of pipe still in use. To make the exercise more economically feasible, it
might be done for selected pipes that represent a cross section of i
n-stalled pipe
materials and installation dates. Within some utilities, condition assessment
is conducted whenever pipe repairs are made or new pipe is installed,
because existing pipe is exposed, which facilitates the assessment. Other utilities
carry out a regularly scheduled assessment program independent of other
activities. Whatever the final outcome, how and when condition assessments
will be conducted should be determined and standardized at each utility.
The technologies available to carry out condition assessment are varied a
n
d
mature. Destructive testing includes the use of coupons or cuts fromactual s
e
c
-tions
of pipe and spot condition assessment. Nondestructive testing i
ncl
ude
magnetic, electromagnetic, sonic, acoustic, infrared thermography, and ground-
penetrating radar equipment for locating pipe; global positioning system
(GPS)/GIS databases for managing information; and ultrasonics, acoustic emis -
sion, magnetic flux leakage, and remote field eddy current for assessing
pipe. Finally, closed-circuit television has been used in some situations. For the
most

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
part there are sufficient tools available to allow utilities to conduct condi
ti
on
assessments and to utilize the data that are collected to guide decisions. What is
needed is for these tools to be utilized more uniformly as well as fine-tuned for
use in small systems that have more limited capabilities. It would be extremely
useful for the results of condition assessments to be shared among utilities, and
even benchmarked, so that utilities can build upon shared experience and
knowledge.
Future tools for assessing the condition of buried pipe include real-tim
e tools
that travel through pipe and collect information; small chips set in p
i
p
e
;sensors
to record sounds of breaks; fiber optics to record breaks in light; and
improved metering to identify leaks (Grigg, 2005). These tools are in develop-
ment and likely show promise for specific situations rather than globally for all
materials in all circumstances. For example, different sensors are needed for
plastic pipe than for iron or concrete pipe.
Leak Detection
The early detection of leaks and their remediation is a goal for water utili-
ties. Leak detectors include listening devices, such as an aquaphone or a more
complicated amplified detection kit that detects sound caused by flowing or es -
caping water (Von Huben, 1999). Another way of detecting leaks is to conduct a
water audit which uses flow meters around smaller districts of a systemat night
when water use should be low. Acoustic methods are easy to use and widely
applied on metallic pipes, with improvements being made for use on plastic
pipes (Lange, 2002). Morgan et al. (2005) recently used a fixed-based acoustic
monitor system called MLOG to scan the distribution system at night for
leaks. The system was highly effective, detecting 17 previously unknown
leaks within the first three months of use.
In addition to improvement in leak detection, water meters have been d
e
-
veloped to detect and record backwards flow through the meter in order to d
e
-
termine the magnitude and frequency of backflow events (Neptune
Technology Group, 2005). Although the majority of this flow may simply be
service line water, the use of advance meter reading can detect these backflows
in real time. The ability to detect and track backflow events will allow more
focused monitoring to determine their impact on drinking water quality.
Main Break Monitoring
Main break monitoring consists of utilities recording responses to w
a
t
e
r main
breaks such as time and date of response, location of break, valves oper-ated to
shut down the main, properties affected by the shut down, repaired or
replaced portion of main, and shut-down time. Transmission mains are given a
higher priority for main break monitoring and the prevention of failure than

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
smaller size distribution mains, given their potential seriousness (for example,
destruction to local structures such as roadways, bridges, and buildings or inter-
ruption of automobile traffic and the evacuation of residences).
If main break monitoring data are maintained in a computerized database,then
quarterly, annual, or five-year historical evaluations can be done, trends c
a
nbe
predicted, and both can be compared to changes in related practices (such
a
s
replacing cast iron with ductile iron pipe). If the utility can also collect data on
leaks and repairs leading up to a break, as well as failure analysis results during
the break, it becomes possible to develop better predictive models for the distri-
bution system’s pipe infrastructure. Water utilities have successfully trended
water main break rates and have adjusted their practices to minimize the occur-
rence of failures for various types of pipe in their systems. The trending of wa-
ter main breaks and leaks along with condition assessment provide an important
tool to minimize public health risk. Not only will a water utility reduce its risk
of serious consequences from an unexpected failure, but in reducing the serious-
ness of such failures the water utility will gain control in minimizing the poten-
tial for water quality contamination. This will happen in two ways, because the
severity of any single failure is reduced, and because the frequency of failure is
reduced.
Water Quality Testing
Much of the monitoring needed to assess the physical integrity of a distri
bu-
tion systemis accomplished by other means than water quality monitoring, suc
has
by leak detection, customer complaint response, inspections, or the exercising of
valves and hydrants. However, water quality testing can play a role. Typi- cally,
water quality analyses are limited to common chemical parameters (total
chlorine residual, pH), physical parameters (turbidity, color), and biological pa-
rameters (heterotrophic plate count, total coliform count) (see Chapter 6 for a
more thorough discussion). For those parameters that are routinely monitored
under the Safe Drinking Water Act, a detection of a change in a
parameter would not in itself identify the occurrence of a loss in physical
integrity that resulted in contamination, since water quality changes could be
from internal conditions in the system or fro m a treatment breakthrough or
failure. However, a thorough follow-up response to a change in water quality
(such as high turbidity, colored water, or non-detectable chlorine residual)
could include valve checks, hydrant flushing, and other techniques that might
identify the cause of the loss in water quality as being an external contamination
event.
Water quality data can also be useful in identifying problems with physical
integrity when integrated with others sets of data, such as customer complaints,
water main break occurrences, timing of newly installed water mains, cleaning
of storage facilities, or backflow events (see Chapter 7 for more discussion of
data integration). Water quality testing is particularly useful if it can be corre -
lated with customer complaints. For example, consumer complaints of chemi-

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
cal-type odors along with the utility’s detection of volatile organic
chemicals (VOCs) could signal that permeation of plastic pipe has occurred,
which could be further studied through groundwater and soil testing for the
same VOCs. Once remediation is put in place, the same water quality
parameters could be used to gauge the success of the remediation. Backflow
events are another example of where customer complaints used in
conjunction with water quality monitoring may be informative. Depending on
the contaminants present, backflow events can affect the color of water, can
introduce debris and particles, and can cause an off-odor or taste.
As with any environmental sampling, increasing the frequency of w
a
t
e
rquality
monitoring, for example going to on-line monitoring of storage f
a
c
i
l
i
t
y effluent as
opposed to daily or weekly grab sampling, will make it more likely that a
contamination event will be detected. For example, when doing water
quality sampling on a new water main prior to its release for use, a typical num-
ber of samples would be four to five. If this new addition or replacement in-
volved 100 feet of 6-inch pipe, the total volume would be around 150 gallons
(568 liters), such that four water quality samples of about 250 mL would only
test 1/568th
of the potentially contaminated water. Clearly, the approval of a
new water main should not rely solely on the final water quality check, but also
on inspections at every stage of the process guaranteeing that materials were
handled in a sanitary manner and protected from exposure to contamination.
Thus, in isolation water quality data are not sufficient to identify failures
in physical integrity. But combined with other data, they may be useful for
detecting external contamination events.
MAINTAINING PHYSICAL INTEGRITY
Every water supplier’s goal is to develop the means by which to better
maintain the physical integrity of its distribution system so that a failure or loss
rarely occurs, or when it does occur its impact is minimized. Table 4-6 summa-
rizes some common measures used to prevent a loss of physical integrity in the
distribution system.
The maintenance issues for pipes, fittings and appurtenances, storage facili-
ties, and backflow prevention devices are similar in a general sense. Material
s
selection must meet standards and best practices. Installations of all components
must be followed up with routine inspections. A regular program of valve o
p
-
eration and maintenance must be in place so that shut downs can be
effective when needed. Many valves and hydrants are unused for a number of
years, and debris within the distribution system may cause hydrants to become
heavily en- crusted leading to a significant reduction in discharge flow and fire
protection. Furthermore, valves and hydrants should be carefully manipulated
to maintain positive pressures and mitigate pressure transients that could result
in pipe breakage. Good construction practices, conducted by those with
training and certification and that follow standards and specifications, are
essential. Standard

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
Valves should be inspected and operated on a regular basis to prevent rust and encrustation from
interfering with their performance. Photo courtesy of Bureau of Laboratory Services, Philadelphia
Water Department.
parts should be used to ensure consistency in repairs, and they should be storedin a
sanitary fashion. Designing the distribution system to minimize sections o
fpipe
and appurtenances that cannot be adequately tested, flushed, and di
s
i
n- fected would
be beneficial (Pierson et al., 2002). Finally, funding and staffing must support
all of these activities. These and other preventive measures are discussed
below. It should be noted that maintaining the appropriate operating pressure to
prevent main breaks and intrusion is discussed more thoroughly in Chapter 5.
Materials Quality
Materials that make up drinking water infrastructure range in type a
n
d
value.
Pumps have various components from pipe to valves to impellers, all made of
differing materials. System piping includes valves and fittings, ferrules, and
hydrants. Storage facilities range in their composition from concrete to steel
with linings of cement, asphaltic, and epoxy. Customer premise plumbing in -
cludes meters, backflow prevention devices, valves, fittings, tubing, and faucets
made of a plethora of materials. Rubber gaskets and plastic seats can be found

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
TABLE 4-6 Examples of Ways to Maintain Physical Integrity
Component Mechanismof Integrity
Loss
Preventionby
Pipe Permeation Standards on pipe applications, l
ocalas-
sessments of soiland groundw ater forcon-
tamination
Structural f
a
i
l
u
r
e
(leak or
break)
Better design and installation, e
a
r
l
y
leak de-
tection w ith rehabilitation and repair, opti-
mized scheduling of pipe renew als, opti-
mized placement of valves for effective
shut-offsand isolations
Fitting and
appurtenance
Improper installation Standards and certification for installation,
follow ed by inspection
Unsanitary activity Strict requirements and inspection during
repair, rehab, installation
Structural failure Improved materials quality as w ellas quality
in the operating components of v
a
l
v
e
s
and
hydrants, periodic valve exercising fol-
low ed by maintenance or replacement as
needed
Improper installation Strict requirements on i
n
s
t
a
l
l
a
t
i
o
n
and design
Unsanitary activity Strict requirements during repair, rehab, in-
stallation and inspection
Storage f
a
c
i
l
-
ity
w all, r
oof,
Structural failure(crack,
hole)
Better design and installation, e
a
r
l
y
leak detection
w ith rehabilitation and repair
cover, v
e
n
t
,
hatch
Backflow
prevention
Absence of Inspection and better design w ith i
n
s
p
e
c
t
i
o
n
Improper
installation Strict requirements on i
n
s
t
a
l
l
a
t
i
o
n
and design
Unsanitary activity Strict requirements during repair, rehab, in-
stallation and inspection
Absence of Inspection and certification
Improper installation Strict requirements on installation and desi
gn
device
Operational failure Annualtesting and maintenance
in valves and meters and in the joints of mains. In addition to these solid mate-
rials there are greases, lubricants, fluxes, and coatings. The diversity, complex-
ity, and value of materials used in drinking water infrastructure are important to
distribution system management, especially given the increasing emphasis on
system reliability and more stringent water quality demands. The following
factors should be considered when choosing distribution systemmaterials:
 health effects of the material when in contact with drinking water;
 hazards and safety in working with the materials;
 structural capabilities of thematerial;
 water quality impacts of thematerial;
 cost and availability of the material;
 compatibility of the material with othermaterials in the systema
ndwith
the conveyed waterand surrounding soils;

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
 environmental effects of thematerial;
 whether the manufacturers of the material are ISO certified and m
e
e
t
NSF and ASTM standards; and
 future changes that could impact on the above.
Not all of these factors have been given equal weight. Materials selection
is typically based on tensile strength, flexural strength, durability, corrosion
resistance, roughness coefficient (Hazen Williams C value), and economy
(e.g., the cost of materials and installation lifetime value) (AWWA, 1986).
Indeed, economic considerations and the availability of the material can
weigh in heavily and may dominate the choice of material.
As shown in Table 2-3, 30 of 34 responding states have some basic re-
quirements for the types of pipe materials allowed in distribution systems.
In- deed, a variety of standards and guidelines are available to help utilities
choose the correct materials for their infrastructure, including the ASTM Annual
Book of Standards, standards from NSF International, AWWA standards, and
other publications (Nayyar, 1992). In practice, the larger water utilities tend to
apply material standards and test whether they are being met, but small water
utilities likely have no way to test for compliance. Furthermore, water suppliers
in the United States have underutilized the services that materials engineering
can provide such as manufacturer and supplier certification, development of
materials specifications for procurement, and evaluation of materials (chemical
and physical) according to specifications after procurement (Burlingame et
al., 2002). Testing of materials to ensure they meet the standards used for
procurement should be a broader practice within the water industry, and not
limited to only the largest water utilities.
In addition to making an informed choice of materials, water utilities shoul
d
strive to protect the quality of the materials after initial purchase. This includes
inspections during materials manufacture; proper storage, handling, and trans-
port of the material to the utility; inspection and testing of the material
upon delivery; protection during onsite storage; inspection during and after
installation; failure analysis to detect early failures; and finally replacement of
the material when its lifetime is exceeded (see section below on asset
management). Failure analysis involves using a standard approach to record
events around material failures; take soil and pipe samples and collect
background records; conduct a preliminary investigation to determine the
type of failure; and conduct structural analyses, visual examinations,
metallographic and mechanical testing, and inspections for graphitization and
manufacturing flaws on pipe surfaces (Makar, 2002). Because failure analysis
has not been widely embraced by the water community, there is limited
information on many of the materials in common use today. Thus, additional
support is needed for technology transfer about materials, funding of materials
testing programs, better materials development and information management,
better training, and better cross-industry networking. For example, there have
been no studies to date on the conse-

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
quences of material failures due to workmanship/installation errors or
manufac-turer variability.
Another consideration is that many materials still in use today were n
ot
originally designed to meet the system reliability and water quality standards
expected today and for the near future. Existing materials standards may not be
complete and up-to-date for all applications. Furthermore, manufacturers are not
always responsive to customer or end-user needs, especially as these needs
change due to water quality regulations. Although improvement is needed
in many areas, a substantial first step would be to improve installation
workmanship. This could be accomplished by requiring that all trades people
who work with materials being installed or repaired that come in contact with
potable water be trained and certified for the level of sanitary and
materials quality that their work demands.
Corrosion Control
The historical use of metallic pipes and the many environmental conditions
they come in contact with have made both external and internal corrosion an
issue for the water industry for some time. Although most utilities use some
form of internal corrosion control to minimize color and turbidity problems and
to meet the Lead and Copper Rule requirements, not all utilities practice external
corrosion control, even though it is important for maintaining the physical integ-
rity of their distribution systems, as acknowledged by 14 of 34 responding states
(see Table 2-3). There is no regulatory motivation for external corrosion control
in the water utility industry as there is in the oil and gas pipeline industry where
corrosion control such as cathodic protection of its pipelines is mandated (Ro-
mer et al., 2004). Nonetheless, understanding the conditions that lead to corro-
sion and implementing a consistent corrosion control methodology can result in
significant operation and maintenance savings because of the longer pipeline
life.
As mentioned previously, the extent of external corrosion depends on soi
l
conditions such as resistivity, pH, and water content; the occurrence of s
t
r
a
y
currents; contact between dissimilar metals; and bacterial activity in the e
n
v
i-
ronment surrounding the pipe. The testing and GIS mapping of soil conditions
can help water utilities predict and plan for corrosive problems and design cor-
rosion control (Romer et al., 2004). Unfortunately, the tools for analyzing soils
prior to making water main construction decisions require further development.
In addition, there is no standardized corrosivity testing method used by all water
utilities. The Ductile Iron Pipe Research Association has promoted a qualitative
corrosion evaluation system based on soil conditions of resistivity, pH, redox
potential, the presence of sulfides, and site drainage conditions, which has been
found to be dependable and accurate for determining when external corrosion
control should be applied for buried iron pipe (Bonds et al., 2005). The Ameri-
can Concrete Pressure Pipe Association provides recommendations based
on

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
soil chloride and resistivity. In general, methods for the analysis of
corrosion include a soil corrosivity survey, a close-interval potential survey, a
cell-to-cell potential survey, ultra-sonic measurements, pit depth analysis, visual
inspection, corrosion rate measurements, acoustic monitoring, and failure
analysis (Szeliga and Simpson, 2002).
External corrosion control methods include determining the soil
conditions and then (1) selecting the appropriate distribution system materials,
such as plastic pipe for use in very corrosive soils; (2) applying external
metallic corrosion prevention materials at the time of manufacture, such as
concrete, mortar, or asphaltic shop coat; (3) applying barrier coatings and
polyurethane encasements in the field; (4) using galvanic cathodic protection or
impressed cathodic protection; and (5) mitigating stray currents (Szeliga and
Simpson, 2002; Romer et al., 2004). For example, Edmonton, Alberta
proactively reduced the impacts of external corrosion using cathodic protection
and nondestructive testing of their cast iron mains (Seargeant, 2002). Proactive
measures are also important, since a variety of design options (such as using
rubber-gasket bell-and-spigot joints) can affect the extent of external corrosion
(Romer et al., 2004). Transmission mains are more frequently engineered for
external corrosion control than distribution mains because of the greater need
to prevent catastrophic failures in the larger diameter water mains.
Internal corrosion of pipe is caused by distribution system water that is c
or-
rosive to the materials with which it comes into contact. Internal
corrosion i
s
common in unlined cast-iron and steel mains and also occurs inside
steel water tanks, metal service lines, and premise plumbing and appliances.
Concrete pipe and cement mortar are also vulnerable to corrosion from low
alkalinity, low hardness waters. Internal corrosion is generally controlled by
feeding corrosion inhibitors, such as phosphates, to the water in combination
with pH adjustment and alkalinity control. The mechanism of action is generally
one of forming a stable scale on the pipe surface from corrosion products and
water constituents that both inhibits corrosion and reduces the release of metals
from scale dissolution. Inhibitors and water quality control procedures need to
be tested at each site of use because of differences in source water quality, pipe
materials, and pipe condition. Ductile-iron and steel pipe are generally lined
with a cement mortar lining to prevent internal corrosion or contact with
water. Linings can reduce the frequency of small leaks in pipes and pipe
connections as a result of the high resistance of cement mortar to pressure,
enhance the hydraulic characteristics of the mains, and prevent further
internal corrosion damage. Finally, steel water tanks are protected by internal
coatings and cathodic protection.
External and internal corrosion control practices need to be used more con-
sistently, universally and uniformly. A manual of practice for the industry
should be developed as an aid to implementing best practices. At present the
best defense against corrosion relies on site-specific testing and practical experi-
ence gained at individual utilities, given the variation in materials, soils, and
water quality from utility to utility. There is also a need for research to develop
new materials and corrosion science to better understand how to more effec-

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
tively control both external and internal corrosion, and to better match distribu-
tion system materials with the soil environment and the quality of water w
i
t
h
which they are in contact.
Permeation Prevention
Appropriate measures can be taken to minimize the occurrence of permea-
tion, such as issuing regulations or guidelines that define the conditions
under which plastic pipe should be used. The proper selection and use of PVC,
polybutylene, and polyethylene plastic pipe, such as according to the soil or
potential soil conditions in which the pipe will be buried, limits the potential for
permeation. For example, California precludes the use of plastic pipe in areas
subject to contamination by petroleum distillates (California Code of
Regulations, Title 22, Division 4, Chapter 16, Article 5, Section 64624f). In
addition to the pipe material, the environmental conditions around the buried
pipe are also important. Utilities that install plastic water mains need to
maintain an up-to-date knowledge base of the locations of underground
storage tank sites, industrial spills, other developments that could discharge
solvents, and their associated solvent plumes so as to avoid the contact of such
contaminants with the pipe. In general, if this information can be gathered
prior to laying new pipe, most if not all permeation incidents can be avoided.
Maintaining Storage Facilities
Storage facility issues are similar to other distribution system
components in that materials selection, system design following standards and
specifications, installation inspection, and good construction practices by
those with training and certification all play a role. Many states do have some
standards for storage tank design and construction, the use of vents, screens,
hatches, and overflows, and they even encourage tank inspection and
maintenance (see Table 2-4). However, perhaps because of their perceived
peripheral role in water supply, storage tanks have not historically received the
attention afforded to pipe maintenance.
Storage facilities have many purposes (see Chapter 1), such that a disci-
plined storage facility management program is critical to water utilities. Such a
program includes developing an inventory and background profile on all tanks,
developing an evaluation and rehabilitation schedule, developing a detailed tank
evaluation process, performing tank evaluations, making rehabilitations and
replacements when needed, and performing a one-year warranty inspection for
all tanks (Wallick and Zubair, 2002). More specifically, storage tanks should be
inspected for needed repairs, barrier screen replacements, and painting. De -
pending on the nature of the water supply chemistry, such detailed
inspections should be made every three to five years, and consist of tanks
needing to be

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
drained, sediment removed, and appropriate rust-proofing applied to the m
etal
surfaces (such as where the water level rises and falls more frequently) (
K
i
r
-
meyer et al., 1999). These inspections are in addition to daily or weekly inspec-
tions for vandalism, security, and water quality purposes (such as identifying
missing vents,open hatches,leaks, and so forth).
In one of the rare documents that addresses storage facilities, Von
Huben (1999) summarizes the use of air vents to allow air to enter and exit as the
water level rises and falls. These vents must be screened to keep out birds and
insects. In general, preventing access to the tank interior by wildlife and
sediment removal are important deterrents to possible pathogen contamination
and coliform colonization that should be undertaken for every tank.
Asset Management
Asset management refers to a strategy of operating, maintaining, rehabilitat-
ing, and replacing infrastructure in order to sustain a cost-effective level of ser-
vice to customers. For a water utility, asset management requires collecting and
analyzing data and information about all functions of the utility (customer
service and support, financial, engineering, operations, maintenance) in order
to make strategic decisions about the infrastructure (Paralez and Muto,
2002; Schwarzwalder, 2002; Allbee, 2004; Lockridge, 2004; Cagle, 2005).
When thought of with respect to maintaining physical integrity, it refers to
developing an inventory of distribution system components and determining
when repair should give way to rehabilitation or replacement (EPA, 2004).
Table 4-7 gives some of the typical life expectancies for pipe, storage, valves,
hydrants and service lines, although it is expected that properly installed
and well maintained pipes should have a service life much longer than their
design lives (Morrison, 2004).
TABLE 4-7 Material Life Expectancies
Distribution System Component Typical Life Expectancies
Concrete and metal storage tanks 30 years
Transmission pipes 35 years
Valves 35 years
Mechanical valves 15 years
Hydrants 40 years
Service lines 30 years
SOURCE: EPA (2004). EPA’s Note: These expected usef ul liv es are drawn f rom a v
a
r
i
e
tyof sources. The
estimates assume that assets hav e been properly maintained.
The adjusted usef ul lif e will be equal to or less than the ty pical usef ul lif e.

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
In order to do asset management, the water supplier needs to have condition
assessment data (see earlier discussion) and management tools (such as funding,
planning, and modeling tools) (Grigg, 2004). The goal of asset management is to
determine the time to failure and vulnerability of individual components (like
pipes) under varying scenarios. As mentioned previously, determining the con-
dition of in-use buried pipe is currently difficult and costly to accomplish be-
cause the pipe is usually still in use, the inside needs to be assessed, and the as -
sessment can only look at one small area of one pipe out of many associated
pipes. Thus, a water utility typically lumps pipes into classes and assigns to
them average failure information, and, using statistics about the system, then
predicts investment needs to maintain the assets.
Beyond maintaining physical integrity, there are many important reasons
for utilities to engage in asset management, including (Morrison, 2004):
 To maintain assets at a predetermined level of service, which requires
inspection and assessment in order to ascertain whether the assets are capable of
providing this level of service;
 To uncover performance issues that might hinder a utility’s abili
tyto
meet customer service expectations, or potentially lead to a catastrophic failure
endangering public health and safety;
 To control costs of rectifying or mitigating a problem, which a
r
e al-ways
much less just after inspection than after a rupture or other emergency event;
 To tailor maintenance practices to the actual condition of the asset, a
n
dnot
merely base them on habit, resulting in an overall reduction in expenditure;
 To properly plan for the retirement and/or replacement of the a
s
s
e
t
,
which, if done over a period of time, will avoid any unexpected surprises.
Westerhoffet al. (2004) found that most utilities engage in asset manage-
ment, although it ranges from simple maintenance programs to complicated
business planning processes.Indeed,terminology, data collection, reporting,
mapping, inventory control, records, and operational parameters are largely d
e
-
fined on a utility-by-utility basis (Grigg, 2005).
Cross-Connection Control
Proven technologies and procedures are available to mitigate the impact o
f
cross connections on potable water quality. Well-known backflow control de-
vices include air gaps, reduced-pressure-zone backflow preventors, doubl
echeck
valves, vacuum breakers, and complete isolation. Lists of approved backflow
prevention assemblies can be found with the University of Southern Cali-
fornia (USC), the American Society of Sanitary Engineering (ASSE),
Under- writers Laboratories, the International Association of Plumbing and
Mechanical

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
Officials, Factory Mutual, and the Canadian Standards Association, while t
he
three most commonly used guidance manuals are the USC Manual of Cross -
Connection Control (USC, 1993), AWWA’s Manual M14 (AWWA, 2004), and
EPA’s Cross-Connection Control Manual (EPA, 2003).
The application of backflow prevention devices is based largely on the d
e
-
gree of hazard thought to be present. A potential threat from chemical and bi
o-
logical contaminants that pose a human health risk would constitute a high haz-
ard. A low hazard would include incidents that alter the water’s aesthetic prop-
erties but do not constitute a health threat. Higher hazards are also related to the
type of facility from which the threat emanates, such as hospitals, funeral
homes, chemical manufacturing plants, laboratories, film processing
facilities, commercial laundromats, among many others. Low hazard facilities
include apartment complexes, warehouses, office buildings, and public
buildings. Table 4-8 gives the recommended applications of various backflow
protection devices according to the degree of hazard and whether those
hazards are due to either back-siphonage (negative pressure or suction on the
supply side of the device) or back-pressure (high pressure on the service side of
thedevice).
There are generally two types of cross-connection control programs: one i
sa
service-protection program and the other is an internal protection program
(AWWA, 2004). The service-protection program is the most common one
for water utilities to undertake, given their typical enforcement capabilities.
This program is one of “containment,” in that any backflow incident would be
contained within the customer’s facility and prevented fromentering the public
distribution system. This is accomplished by installing a backflow prevention
device at the water meter. Water utilities are typically effective with this type
of program because they readily have enforcement capability in the shut-off of
the water service at the curb stop. The internal protection program is
based on “elimination” or getting rid of the cross connection where it exists
within a customer’s plumbing. Because water utilities typically have no
authority within the premises of their customers, it is more likely that other
agencies such as the local health department or plumbing code agency would
maintain such a program.
Lee et al. (2003) found that more than 80 percent of responding water u
t
i
l
i
- ties
require approved backflow protection devices and field testing of t
h
e
i
rproper
operations. However, little if any information exists on whether the
sedevices
are present in customers’ premises, where 83 percent of cross connections are
known to exist (Lee et al., 2003). It is probable that they are absent for a very
large percentage of cross connections nationwide or not functioning properly
for a small percentage of cross connections. Clearly, their increased use and
regular inspection would do much to reduce public health risks fromdrinking
water distribution systems. Indeed, for utilities operating dual distribution
systems, the need for an effective cross-connection control program is
paramount.

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
TABLE 4-8 Use of Backflow Prevention Devices by Degree of Hazard and Mechanism
Degree of Hazard
Device Low Hazard High Hazard
Back-
siphonage
Back-
pressure
Back-
siphonage
Back-
pressure
Air Gap (AG) X X
Atmospheric vacuumbreaker (AVB) X X
Spill-resistant pressure-type vacuum- X X
breaker assembly (SVB)
Double checkvalve assembly (DC or X X
DCVA)
Pressure vacuum-breaker assembly X X
(PVB)
Reduced-pressure principle assembly X X X X
Reduced-pressure principle detector
assembly
Double checkvalve detector checkas-
sembly
Dual checkdevice (internalp
r
o
t
e
c
t
i
o
n
only)
Dual checkw ith atmospheric v
e
n
t
device
(internal protection only)
SOURCE: Adapted f rom Table 3-1 in AWWA (
2
0
0
4
).
X X X X
X X
X X
X X
Basically, every state has some requirement for cross connection c
o
n
t
r
o
l
(see
Tables 2-3 and 2-6), and state plumbing codes define the type of c
r
o
s
s
-
connection control devices that are approved for use. Unfortunately, as d
i
s
-
cussed in Chapter 2, the elements of such programs, their implementation, and
oversight vary widely, partly because of the variation in available resources. In a
few states, local jurisdictions are responsible for implementing a cross -
connection control program. In most states, testing of cross-connection control
devices is the responsibility of the customer while inspection of the devices
is the responsibility of the water system or the local jurisdiction. Given this
variability, Chapter 2 recommends that EPA explicitly define what an
acceptable cross-connection controlprogram should be.
RECOVERING PHYSICAL INTEGRITY
It is impossible for a distribution system of any significant size to be m
an-
aged in such a way as to prevent any loss of physical integrity over time. Even a
water utility with a good program of corrosion control and pipe replacement can
experience an annual pipe break rate of around 750 to 850 breaks per year (Fa-

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
larski, 2002). Damodaran et al. (2005) gave an industry average of 0.1 to 0
.3
breaks per mile of pipe per year, such that a low break rate would cause 1 to 3
breaks per year per 1,000 people served. Philadelphia tracks the number of
breaks experienced for each 1,000 miles of main using a five-year moving aver-
age to smooth out the effect of weather variations. Based on historical informa -
tion dating back to 1930, the average for 2001 was 212 breaks for every 1,000
miles of main—the lowest total in over 45 years and better than the national
average of 240 to 270 breaks per 1,000 miles. Nonetheless, even with a water
main replacement program that appears to be successful compared to the
national average, every year over 600 water main breaks occur. Therefore,
procedures need to be in place by which to recover from a failure in a material
barrier and minimize the effects on water quality.
Table 4-9 summarizes some of the common methods used today to recover
from a failure in a material barrier in order to prevent or minimize contamina-
tion of the water supply. There are several categories of recovery efforts. First,
compromised materials can be cleaned, repaired, rehabilitated, or replaced. For
example, leaks and small breaks can be repaired by repair sleeves or by
joint sealing compounds. Storage facilities might have to be drained and
cleaned following potential contamination. Another form of restoration is to
treat the contaminated water. Chlorine and other disinfectants have been used
to protect pipes and storage facilities against external microbial contamination,
prevent
TABLE 4-9 Ways to Recover from a Loss in Physical Integrity
Component Mechanismof Integrity
Loss
Recovery by
Pipe
Fitting and
appurtenance
Permeation Reline or replace and c
o
n
d
u
c
tw ater quality
testing
Structural failure (leak) Replace or repair or rehab
Structural failure (break) Replace or repair, flush or disinfect, c
o
n
-
duct w ater quality testing
Improper installation Replace, reinstall
Unsanitary activity Disinfect, flush, and w ater q
u
a
l
i
t
ytesting
Structural failure Replace, repair, rehab and disinfect
Improper installation Reinstall
Unsanitary activity Disinfect and flush
Storage f
a
c
i
l
-
ity
w all, r
oof,
Structural failure(crack,
hole)
Repair or rehab or replace, disinfect
cover, v
e
n
t
,
hatch
Backflow
prevention
Absence of Install
Unsanitary activity Disinfect, flush, and w ater q
u
a
l
i
t
ytesting
Absence of Install
device
Operational failure Replace or repair

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
regrowth of nuisance organisms in response to intruded chemicals, prevent fur-
ther contamination from the installation of a dirty main, and alleviate customer
complaints. Both continuous disinfectant residual maintenance throughout the
distribution system and dosing a section of the system with disinfectant are
common. Third, recovery is often brought about by flushing the contaminated
water from the system rather than treating it, generally using hydrant flushing.
Although flushing is mentioned sporadically here because it accompanies many
of the other recovery techniques, it is treated more comprehensively in Chapters
5 and 6 where hydraulic and water quality integrity are the focus,respectively.
In those situations where the absence of a component was the cause for t
h
e
lack of physical integrity, then simply installing the component is the recovery
effort. For example, the installation of backflow prevention devices or changing
covers on reservoirs (say from floating to hard covers) should restore integrity.
Finally, where operational failure is the problem, devices may also need to be
entirely replaced, along with instituting inspections to ensure that failure does
not recur.
Repairing, Rehabilitating, and Replacing Pipe
Common types of repair activities include cutting and plugging the portion
of pipe associated with a leak, installing a repair sleeve or clamp, eliminating
dead end mains, replacing and repairing valves, adding ferrules, and repairing or
replacing hydrants. These activities are discussed extensively in Grigg (2004)
and not considered further here. Improvements are being made in locating bur-
ied failure sites, excavation, and repair. For example, trenchless methods
are being developed and applied, although the technology development is slow.
Rehabilitation of pipe involves the recycling and reinforcing of the existing
infrastructure in order to prolong its useful life. For example, structural lining
can be used to improve the structural integrity of existing pipes and involves
placing a watertight structure in immediate contact with the inner surface of a
cleaned pipe (Selvakumar et al., 2002; Ellison et al., 2003). The most com-
monly used structural lining techniques include conventional slip lining (where
new PE pipe is structurally able to replace the existing pipe), cured-in-place re-
habilitation or inversion lining (which inserts a non-structural material) (Hughes
and Conroy, 2002), fold-and-formpipe, and close-fit slip lining (which can use a
structural or non-structural replacement material). Selvakumar et al. (2002)
provide a detailed description of all these methods along with their costs, bene-
fits, and limitations. Nonstructural rehabilitation of water mains, which does not
focus on recovering the physical integrity of distribution systems, includes
chemical dosing for corrosion control, cement mortar lining, epoxy resin lining,
and thin-walled PE lining (Hughes and Conroy, 2002; Grigg, 2004; Damodaran
et al., 2005). Such rehabilitation should be internally inspected to ensure that it
is done to standards.

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
Pipes are candidates for replacement when the pipe is severely deteriorated
(e.g., the pipe has suffered a series of breaks), or when additional hydraulic ca-
pacity is needed. Box 4-4 discusses the economic considerations that play into
the decision to replace a pipe rather than rehabilitate or repair it. Historically,
pipeline replacement involved the construction of a new pipeline normally paral-
lel to the one being replaced. Once constructed, the new pipeline was connected
to the pipe network and the old pipeline abandoned. This approach normally
involved digging a trench, installing the new pipe, backfilling the trench, and
final surface restoration. This construction can be very disruptive in built -up
areas plus it may be very difficult to find a location to construct a new
waterline. As a result, new trenchless technologies have developed which can
result in cost savings over the conventional construction methods.
Horizontal directional drilling has seen considerable growth as an alternative to
open trench construction, especially at crossings of waterways, rail lines, and
highways. A drilling bit bores a horizontal hole that is kept open using drilling
fluid. Once a predetermined length of hole is completed, a new pipe is pulled
back through the horizontal hole. This method is far less disruptive than open
trench construction, and in most cases would not interfere with business or
residential property access.
Another type of trenchless technology that is most useful in areas where
it i
s
difficult to install new pipe is pipe-bursting. This technology is similar to
horizontal directional drilling, but with pipe-bursting a new pipe is pulled
in the same location as the old pipe. A burster is pulled through the old pipe,
breaking it apart and making room for the new pipe. The only openings required
are at the two ends and at all active service locations. The equipment can install
pipe of the same size that is being replaced or a size or two larger. Selvakumar
et al. (2002) give a detailed description of pipe bursting, microtunneling, and
horizontal directional drilling methods along with their costs, benefits, and
limitations.
BOX 4-4
Decision-makingregarding Replacement vs. Ongoing Repair
There now exist fairly good models for making decisions about o
n
g
o
i
n
g repair vs. re-
placement of infrastructure pipe components (Damodaran et al., 2005), although they do
not incorporate public health risk and water quality deterioration. The traditional economic
life of a component is the point at which the cost of keeping it in use equals the cost
of replacing it. The “cost”, though, has been expanded beyond the utility’s internal costs
to include external costs, like the public’s costs associated w ith the failure of a component
(loss of water and business, traffic disruptions, etc.). Expectations for customer service are
rising at the same time that repair and replacement costs are rising. Decisions based
on internal costs alone often favor ongoing repair over replacement. When external
costs (such as the number of households affected by a failure) are counted, replacement
begins to be favored over repair. When the break rate for a 20-ft long pipe exceeds once
per year then it can become more economical to replace the pipe than repair it (Damodaran
et al., 2005). Utilities need guidance on including external costs along w ith internal costs,
and the advantages and disadvantages of replacement methods, so that they can make
up-to-date and sound decisions in a timely manner.

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
Regardless of whether the situation requires repair, rehabilitation, or r
e
-
placement, there are practices that can minimize the contamination potential,
such as maintaining a positive pressure until the repair site is unearthed and
cleared. Trench water should be removed before work is done, and street drain-
age should be provided to keep water and runoff out of the trench. New materi-
als and repaired materials can be sprayed or swabbed with chlorine or appropri-
ate sanitizing agents, as specified in ANSI/AWWA standards C600-99 for the
installation of ductile iron mains and C651-99 for the disinfection of mains.
During these activities, inspectors or engineers managing the site need to be
aware of all issues related to water quality including the type of pipe that can be
laid in soils suspected of contamination, the means by which to protect materials
during storage, the methods for working in trenches to prevent contamination of
materials, and what to do if materials do become contaminated.
Prior to the release for use of a new or replaced water main or facility, a w
a
-ter
utility will typically conduct water quality testing. Total coliform bacteria
have been the most common indicator that the new material is sanitary and did
not become contaminated during storage or installation. In addition to total coli-
formtesting, the water utility can also test for turbidity, HPC bacteria, total chlo-
rine residual, pH, and odor, as unsanitary and improper installation practices can
affect theseparameters.
As documented in Table 2-3, 16 of 34 responding states address the storage
and handling of pipes, while 29 of 34 address the need for disinfection and water
quality testing following installation. Experience has shown, unfortunately, that
sanitary practices vary widely. Even well-run utilities can experience a 30 per-
cent failure rate in the approval of new mains based on water quality testing
(Burlingame and Neukrug, 1993). Pipe design and construction is usually fo-
cused on existing codes (such as depth of installation to prevent freezing)
and corrosion protection (such as using plastic pipe or metallic pipe with
protective wrap in corrosive soils) but not on sanitary practices and rarely on
permeation concerns. Pierson et al. (2001) found that although the
ANSI/AWWA standards, particularly C600-99, attempt to address installation
or construction practices, there is a general lack of training and the use of
requirements for sanitary practices. It is possible for trenches where pipe is
being laid or repaired to fill partially with water from broken lines or from
precipitation or groundwater. This water can mobilize soil-related contaminants
as well as carry contamination itself. Clearly, during emergency repairs or
repairs made under less than favorable conditions, it becomes even more
difficult to prevent the exposure of materials to environmental contamination.
This could be addressed in part by requiring foremen or managers of
construction sites to be certified on a regular basis, as it is for the certification of
backflow installers and testers. Such training and certification can be provided
through third-party organizations (non-water utility agencies) such as the New
England Water Works Association and American Society of Sanitary Engineers.
Not only would foremen or managers have to know the engineering
requirements, but they would also have to record and un-

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
derstand the issues related to protecting the sanitary condition of the materials
and the water supply.
Disinfection
Haas et al. (1998) reported that interior pipe surfaces are not free of mi
cro-
bial contaminants even under best case conditions. Furthermore, the lack
of adequate distribution system maintenance (which includes flushing,
disinfecting, and coliform testing of all pipe repairs and pipe replacement
activities) has been found to contribute to higher coliform occurrence rates
(Clement et al., 2003). Thus, when a new main is installed or a valve is
repaired, it is advisable to act as if some level of contamination has occurred to
both the water and the materials and to address potential contamination before
the affected portion of the water system is returned to use. When the interior of
pipe has become contaminated or needs cleaning due to unsanitary activities,
disinfection becomes necessary.
Pipes can have a significant chlorine demand which reduces the e
f
f
e
c
t
i
v
e
-ness of
disinfection (Haas et al., 1999). Fortunately, there is a current AWWA standard
(C652) governing new pipe disinfection, which sets forth two options. The first
is to flush followed by filling the facility/pipe with a strong (> 25 mg/L)
chlorine solution and maintaining it for 24 hours providing that a residual of 10
mg/L remains. The second option is contacting the pipe or facility with a 100
mg/L free chlorine solution for at least three hours so that the residual re-
maining is at least 50 mg/L. The chlorine used for these disinfection operations
may be supplied either as solid calcium hypochlorite powder dissolved in water,
sodium hypochlorite (liquid bleach) dissolved in water, or gaseous chlorine dis -
solved in water.
These guidelines basically require that a “CT” (product of disinfectant and
contact time) of 14,000 (first option) or 9,000 (second option) mg-min/L be
achieved. Tests on actual mains indicate that these guidelines are sufficient to
yield four logs (99.99 percent) inactivation of heterotrophic plate count (HPC)
bacteria (Haas et al., 1998). Where unusually high levels of contamination are
suspected,the design “CT” for facility disinfection should be increased.
After disinfection, the chlorinated water must be flushed from the
system and the adequacy of disinfection checked by microbiological testing. In
flushing the heavily chlorinated water, attention must be paid to (1) preventing
leakage into the active distribution system if the newly disinfected pipe is
connected to the system, (2) the potential impacts on the sewer system if the
water is discharged to a sewer, or (3) dechlorinating the water (using sulfur
dioxide, sulfite, or bisulfite) if the water is discharged to a surface waterbody so
as to minimize adverse impacts to aquatic life.

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
The loss of physical integrity of the distribution system—in which the s
y
s
-tem
no longer acts as a physical barrier that prevents external contamination from
deteriorating the internal, drinking water supply—is brought about by
physical and chemical deterioration of materials, the absence or improper instal-
lation of critical components, and the installation of already contaminated com-
ponents. When physical integrity is compromised, the drinking water supply
becomes exposed to sources of contamination that increase the risk of negative
public health outcomes. The following primary conclusions and recommenda-
tions for maintaining and restoring physical integrity to a distribution systemare
made.
Storage facilities should be inspected on a regular basis. A disciplined
storage facility management program is needed that includes developing an i
n
-
ventory and background profile on all facilities, developing an evaluation a
n
d
rehabilitation schedule, developing a detailed facility inspection process, per-
forming facility inspections, and rehabilitating and replacing storage facilities
when needed. Depending on the nature of the water supply chemistry, every
three to five years storage facilities need to be drained, sediments need to be
removed, appropriate rust-proofing needs to be done to the metal surfaces, and
repairs need to be made to structures. These inspections are in addition to daily
or weekly inspections for vandalism, security, and water quality purposes (such
as identifying missing vents,open hatches,and leaks).
Better sanitary practices are needed during installation, repair, re -
placement, and rehabilitation of distribution system infrastructure. All
trades people who work with materials that are being installed or repaired a
ndthat
come in contact with potable water should be trained and certified for t
h
elevel of
sanitary and materials quality that their work demands. Quality w
o
r
k
- manship for
infrastructure materials protection as well as sanitary protection o
fwater and
materials should go hand-in-hand considering the increasing costs o
f
infrastructure
failure and repair and the increasingly stringent water quality standards.
Training and certification can be provided through third-party organizations
(non-water utility agencies) such as the New England Water Works Association
and American Society of Sanitary Engineers.
Although it is difficult and costly to perform, condition assessment of
buried infrastructure should be a top priority for utilities. Every water util-
ity should maintain a complete, up-to-date inventory of all infrastructure com-
ponents from storage facilities to pipes to valves to hydrants, including thei
r
current condition. Because failure analysis has not generally been embraced by
the water community, there is limited information on many of the materials in
common use today. Most useful would be a user-friendly guidance manual for
utilities regarding the failure mechanisms of different types of infrastructure

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
materials and how to use the various types of information on the current condi-
tion of the pipe to determine its expected lifetime. Finally, as an essential part
of condition assessment, every water utility should have in place a leak detection
program that includes checking service lines as well as transmission mains.
External and internal corrosion should be better researched and con-
trolledin standardized ways. There is a need for new materials and c
o
r
r
o
s
i
o
nscience to
better understand how to more effectively control both external and internal
corrosion, and to match distribution system materials with the soil environment
and the quality of water with which they are in contact. At present the best
defense against corrosion relies on site-specific testing of materials, soils, and
water quality followed by the application of best practices, such as cathodic
protection. Indeed, a manual of practice for external and internal corrosion con-
trol should be developed to aid the water industry in applying what is known.
Corrosion is poorly understood and thus unpredictable in occurrence.
Insuffi- cient attention has been given to its control, considering its
estimated annual direct cost of $5 billion for the main distribution system (not
counting premise plumbing).
Cross-connection control should be in place for all water utilities.
Every utility should have a uniform and consistent cross-connection control pro-
gram along with adequate support such as regulations or codes, and staffing. The
program should at the least provide for service-protection or containment (i.e.,
making sure that customers cannot backflow contaminants into the public
distribution system), and when possible should attempt to eliminate cross con-
nections on customer’s premises. Most if not all technical and
administrative information already exists upon which to institute a cross -
connection control program.
REFERENCES
Allbee, S. 2004. A center of excellence—a sensible step on the pathway to
excellence i
n
water utility infrastructure management. Underground Infrastructure
Management (Nov/Dec.):27–29.
American Water Works Association (AWWA). 1986. Introduction to Water Distribu-
tion Principles and Practices of Water Supply Operations. Denver, CO: AWWA.
AWWA. 1999. Water Audits and Leak Detection, Manual M36, 2nd
edition. D
e
n
v
e
r
,CO:
AWWA.
AWWA. 2003. Water Stats 2002 Distribution Survey CD-ROM. Denver, CO: AWWA.
AWWA. 2004. Recommended Practice for Backflow Prevention and Cross-Connection
Control, ManualM14, 3rd
AWWA. 2005a. Flexible, lightweight PE gaining ground. Opflow (July):24–25.
AWWA. 2005b. Ductile-Iron Pipe—Iron and Icon for Durability, Reliability. O
p
f
l
o
w
(February):14–15.

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
AWWA and EES, Inc. 2002. Permeation and leaching. Available on-line at
http://www.epa.gov/safewater/tcr/pdf/permleach.pdf. Accessed May 5, 2006.
American Water Works Association Research Foundation (AwwaRF). 1985. I
n
t
e
r
n
a
l
Corrosion of Water Distribution Systems. Cooperative report with DVGW F
o
r
-
schungsstelle. Denver, CO: AWWA Research Foundation.
AWWA Water Loss Control Committee. 2003. Committee report: applying w
o
r
l
d
w
i
d
eBMPs in
water loss control. J. Amer. Water Works Assoc. 95(8):65–80.
Angulo F. J., S. Tippen, D. J. Sharp, B. J. Payne, C. Collier, J. E. Hill, T. J. Barrett, R. M
.
Clark, E. E. Geldreich, H. D. Donnell, Jr., and D. L. Swerdlow. 1997. A c
o
m
m
u
-nity
waterborne outbreak of salmonellosis and the effectiveness of a boil water order.
American Journal of Public Health 87(4):580–584.
Bonds, R. W., L. M. Barnard, A. M. Horton, and G. L. Oliver. 2005. Corrosion a
n
d
corrosion
control of iron pipe: 75 years of research. J. Amer. Water Works Assoc. 97(6):88–
98.
Booth, S., and B. Brazos. 2005. Qualitative Procedures for Identifying Particles
in Drinking Water. Denver, CO: AwwaRF.
Burlingame, G. A. 1999a. Solving customers’ taste and odor complaints—part
1: t
h
e
importance of thefirst response. Opflow 25(10):10–11.
Burlingame, G. A. 1999b. Solving customers’ taste and odor complaints—part 2: tracking
odors to their source. Opflow 25(11):6–7.
Burlingame, G. A., and C. Anselme. 1995. Distribution system tastes and odor. P
p
.
281–319
In: Advances in Taste-and-Odor Treatment and Control. Denver, CO:
AwwaRF.
Burlingame, G. A., and H. M. Neukrug. 1993. Developing proper sanitation
requirements and procedures for water main disinfection. Pp. 137–146 In:
Proceedings of AWWA Annual Conference. Denver, CO: AWWA.
Burlingame, G. A. 2001. A balancing act: distribution water quality and operations.
Opflow 27(7):14–15.
Burlingame, G. A., J. Rahman, E. Navera, and J. E. Durrant. 2002. Pp. 83–101 In: A
s
-
sessing the Future: Water Utility Infrastructure Management. D. M. Hughes (ed.).
Denver, CO: AWWA.
Cagle, R. F. 2005. Daddy, are we there yet? Underground infrastructure management.
Jan/Feb:43–46.
Clark, R. M., and J. A. Goodrich. 1989. Developing a database on infrastructure needs.
J. Amer. Water Works Assoc. 81(7):81–87.
Clark, R. M., Geldreich, E. E., Fox, K. R., Rice, E. W., Johnson, C. H., Goodrich, J. A
.
,
Barnick, J. A., and Abdesaken, F. 1996. Tracking a Salmonella serovar typhi-
murium outbreak in Gideon, Missouri: Role of contamination propagation modeling.
Journal of Water Supply Research and Technology—Aqua 45(4):171–183.
Clement, J., C. Spencer, A. J. Capuzzi, A. Camper, K. V. Andel and A. Sandvig. 2
0
0
3.
Influence of Distribution System Infrastructure on Bacterial Regrowth. Denver, C
O
:
AwwaRF.
Cooperative Research Centre for Water Quality and Treatment. 2003. Setback for N
e
t
h
-
erlands Dual Supplies. Health Stream 30:5.
bution systemdeficiencies. J. Amer. Water Works Assoc. 93:9:64–75.
Damodaran, N., J. Pratt, J. Cromwell, J. Lazo, E. David, R. Raucher, C. H
e
r
r
i
c
k
, E.
Rambo, A. Deb, and J. Snyder. 2005. Customer acceptance of water main s
tr
uc-
tural reliability. Denver, CO: AwwaRF.

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
Donahue, E. J., III. 2002. GASB 34 and water utilities: deferred maintenance and c
o
n
-
tributed capital. In: Assessing the Future: Water Utility Infrastructure Management.
D. M. Hughes (ed.). Denver, CO: AWWA.
Ellison, D., S. J. Duranceau, S. Ancel, G. Deagle, and R. McCoy. 2003. Investigation
o
f
pipecleaning methods. Denver, CO: AwwaRF.
Environmental Protection Agency (EPA). 1999. Uncovered Finished Water R
e
s
e
r
v
o
i
r
sGuidance
Manual. EPA 815-R-99-011. Washington, DC: EPA Office of W
a
t
e
r
.Available on-line
at http://www.epa.gov/safewater/mdbp/pdf/uncover/ufw8p.pdf.
EPA. 2002a. Technical fact sheet on: Benzene. Available on-line at http://www.epa.gov/
OGWDW/dwh/t-voc/benzene.html. Accessed on May 8,2006.
EPA. 2002a. Technical fact sheet on: Xylenes. Available on-line at
http://www.epa. gov/OGWDW/dwh/t-voc/xylenes.html. Accessed on May 8, 2006.
EPA. 2002c. Technical fact sheet on: Toluene. Available on-line at http://www.epa.gov/
OGWDW/dwh/t-voc/toluene.html. Accessed on May 8, 2006.
EPA. 2002d. Technical fact sheet on: Ethylbenzene. Available on-line at h
t
t
p
:
/
/
w
w
w
.
epa.gov/ OGWDW/dwh/t-voc/ethylben.html. Accessed on May 8, 2006.
EPA. 2003. Cross-Connection Control Manual. Washington, DC: EPA. Available o
n
-line
at http://www.epa.gov/safewater/crossconnection.html. Accessed on May 8
,
2006.
EPA. 2004. Taking stock of your water system—a simple asset inventory for very s
m
a
l
l
drinking water systems. EPA 816-K-03-002. Washington, D.C.: EPA.
Falarski, M. R. 2002. East Bay Municipal Utility District’s Pipeline Replacement P
r
o
-gram.
In: Assessing the Future: Water Utility Infrastructure Management. D. M
. Hughes
(ed). Denver, CO:AWWA.
Geldreich, E. E. 1996. Microbial Quality of Water Supply in Distribution Systems.
Boca Raton, FL: CRC Press, Inc.
Glaza, E. C., and J. K. Park. 1992. Permeation of organic contaminants through g
a
s
- keted
pipejoints. J. Amer. Water Works Assoc.84(7):92–100.
Government Accounting Standards Board. 1991. GASB Statement 34: basic financial
statements and management’s discussion and analysis for state and local
governments issued in 1991.
Grigg, N. S. 2004. Assessment and Renewal of Water Distribution Systems.
Denver, CO: AwwaRF.
Grigg, N. S. 2005. Assessment and Renewal of Water Distribution Systems. J. Amer.
Water Works Assoc. 97:2:58–68.
Haas, C. N., M. Gupta, G. A. Burlingame, R. B. Chitluru, and W. O.Pipes. 1999. B
a
c
t
e
-rial
levels of new mains. J. Amer. Water Works Assoc. 91(5):78–84.
Haas, C. N., R. B. Chitluru, M. Gupta, W. O. Pipes, and G. A. Burlingame. 1998. D
e
-
velopment of disinfection guidelines for the installation and replacement of w
a
t
e
r
mains. Denver, CO: AwwaRF.
Holsen, T. M., Park, J. K., Bontoux, L., Jenkins, D. and Selleck, R. E. 1991. The e
f
f
e
c
t of
soils on the permeation of plastic pipes by organic chemicals. J. Amer. Water
Works Assoc.83(11):85–91.
Hrudey, S. E., and E. J. Hrudey. 2004. Safe Drinking Water: Lessons from Recent O
u
t
-
breaks in Affluent Nations. London: IWA Publishing.
Hughes, D. M., and P. J. Conroy. 2002. Matching deteriorating main conditions to r
e
-
placement/rehabilitation options. In: Assessing the Future: Water Utility Infrastruc-
ture Management. D. M. Hughes (ed.). Denver, CO:AWWA.

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
Kirmeyer, G. J., L. Kirby, B. M. Murphy, P. F. Noran, K. D. Martel, T. W. Lund, J.
L
.
Anderson, and R. Medhurst. 1999. Maintaining and operating finished water stor-
age facilities. Denver, CO: AwwaRF.
Kirmeyer, G. K., M. Freidman, K. Martel, D. Howie, M. LeChevallier, M.
Abbaszadegan, M. Karim, J. Funk, and J. Harbour. 2001. Pathogen Intrusion into
the Distribution System. Denver, CO: AwwaRF.
Krasner, S. W., and E. G. Means, III. 1986. Returning newly covered reservoirs to s
e
r
- vice:
health and aesthetic considerations. J. Amer. Water Works Assoc. 78(3):94–100.
Lange, G. 2002. Locating leaks in plastic water pipes. Opflow 28(7):1,4,5,10.
Lee, J. J., P. Schwartz, P. Sylvester, L. Crane, J. Haw, H. Chang, and H. J. Kwon. 2
0
0
3
.
Impacts of Cross-Connections in North American Water Supplies. Denver, C
O
:
AwwaRF.
Lockridge, R. 2004. The four C’s of asset management. Underground Infrastructure
Management Nov/Dec.:39–41.
Makar, J. 2002. Investigating large gray cast-iron pipe failures: a step-by-step approach.
In: Assessing the Future: Water Utility Infrastructure Management. D. M. Hughes
(ed.). Denver, CO: AWWA.
Makar, J., and Y. Kleiner. 2002. Maintaining water pipeline integrity. In: Assessing t
h
e
Future: Water Utility Infrastructure Management. D. M. Hughes (ed.). Denver, C
O
:
AWWA.
Makar, J., R. Rogge, S. McDonald, and S. Tesfamariam. 2005. The effect of corrosion
pittingon circumferential failures in grey cast iron pipes. Denver, CO: AwwaRF.
Makepeace, D. K., D. W. Smith, and S. J. Stanley. 1995. Urban stormwater q
u
a
l
i
t
y:summary
of contaminant data. Critical Reviews in Environmental Science a
n
dTechnology
25(2):93–127.
Male, J. W., and T. M. Walski. 1991. Water Distribution Systems: A T
r
o
u
b
l
e
s
h
o
o
t
i
n
g
Manual. Chelsea,
MI:Lewis Publishers, Inc.
Martel, K., A. Hanson, G. J. Kirmeyer, M. Besner, A. Carrier, M. Prevost, A. L
y
n
g
g
a
a
r
d
-Jensen
and N. Bazzurro. 2005. Data Integration for Water Quality Management.
Denver, CO: AwwaRF.
McGuire, M., N. Graziano, L. Sullivan, R. Hund, and G. Burlingame. 2004. Water Util-
ity Self-Assessment for the Management of Aesthetic Issues. Denver, CO:
Aw-waRF.
Morgan, W., R. Titus, and D. M. Hughes. 2005. Simultaneous C
o
m
m
u
n
i
c
a
t
i
o
nof Acoustic Data
and Meter Readings Automatically. In: Proceedings from the Interna- tional Water
Association Water Loss Task Force Conference, Halifax, Nova Scotia.
Morrison, R. 2004. Condition assessment—back to the basics. Underground I
n
f
r
a
s
t
r
u
c
- ture
Management Nov/Dec.:51–55.
Nayyar, M. L. 1992. Piping Handbook, Sixth Edition. New York: McGraw-Hill, Inc.
Neptune Technology Group. 2005. E-Coder. Available on-line at
http://www.neptunetg.
com/uploadedFiles/E-Coder%20Press%20Release.pdf. Accessed May 8, 2006.
Paralez, L. L., and D. Muto. 2002. Creating an asset management strategy: an
a
s
s
e
tmanagement template. In: Assessing the Future: Water Utility Infrastructure Man-
agement. D. M. Hughes (ed.). Denver, CO: AWWA.
Pierson, G. L., G. Burlingame, and K. Martin. 2002. Establishing a tradition of c
o
n
t
a
m
i
-nation
prevention. Opflow 28(7):6,7,11.
Pierson, G., K. Martel, A. Hill, G. Burlingame, and A. Godfree. 2001. Methods to prevent
microbiological contamination associated with main rehabilitation and
replacement. Denver, CO:AwwaRF.

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
Propato, M., and J. G. Uber. 2004. Vulnerability of water distribution systems to p
a
t
h
o
-gen
intrusion: how effective is a disinfectant residual? Environ. Sci. Technol.
38(13):3713–3722.
Romer, A. E., G. E. C. Bell, S. J. Duranceau, and S. Foreman. 2004. External Corrosion
and Corrosion Control of Buried Water Mains. Denver, CO:AwwaRF.
Schwarzwalder, R. 2002. Asset management for the new millennium: strategic a
p
-
proaches
for water utilities. In: Assessing the Future: Water Utility Infrastructure
Management. D. M. Hughes (ed.). Denver, CO:AWWA.
Seargeant, D. 2002. Using new technology to optimize management of cast iron p
i
p
eassets.
. Hughes
(ed.). Denver, CO:AWWA.
Selvakumar, A., R. M. Clark, and M. Sivaganesan. 2002. Costs for water supply d
i
s
t
r
i
- bution
systemrehabilitation. Jour. Water Resources Planning and M
a
n
a
g
e
m
e
n
tASCE128(4):303–306.
Shamsi, U. M. 2005. GIS Applications for Water, Wastewater and Stormwater Systems.
Boca Raton, FL: CRC Press.
Skala, M. F. 1994. Waterborne salmonella outbreak in southeastern Missouri. M
i
s
s
o
u
r
i
Epidemiologist 17(2):1–2.
Swerdlow, D. L., B. L. Woodruff, R. C. Brady, P. M. Griffin, S. Tippen, H. D. D
o
n
n
e
l
l
,Jr., E.
Geldreich, B. J. Payne, A. Meyer Jr., J. G. Wells, K. D. Greene, M. Bright, N.
H. Bean, and P. A. Blake. 1992. Waterborne outbreak in Missouri of Escherichia coli
O157:H7 associated with bloody diarrhea and death. Annals of Internal Medi- cine
117(10):812–819.
Szeliga, M. J., and D. M. Simpson. 2002. Evaluating the conditions of existing w
a
t
e
r mains.
. Hughes
(ed.). Denver, CO:AWWA.
University of Southern California (USC). 2002. Prevalence of cross c
o
n
n
e
c
t
i
o
n
sin household
plumbing systems. Available on-line at http://www.usc.edu/dept/fcchr/
epa/hhcc.report.pdf. Los Angeles, CA: USC Foundation for Cross-Connection Con-
trol and Hydraulic Research.
USC. 1993. Manual of Cross-Connection Control, 9th Edition. Los Angeles, CA: U
S
C
Foundation for Cross-Connection Control and Hydraulic Research.
Von Huben, H. (Tech. Ed). 1999. Water Distribution Operator Training Handbook, 2
nd
Wallick, P. C., and M. Zubair. 2002. Tank evaluation, rehabilitation, and replacement
decisions for water storage tanks. In: Assessing the Future: Water Utility Infrastruc-
ture Management. D. M. Hughes (ed.). Denver, CO: AWWA.
Westerhoff, G., P. Fahy, and S. Robinson. 2004. On the pathway to improved
a
s
s
e
t
management. Underground InfrastructureManagement Nov/Dec:35–37.

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
5
Hydraulic Integrity
The hydraulic integrity of a water distribution system is defined as its abil-
ity to provide a reliable water supply at an acceptable level of service—that is,
meeting all demands placed upon the system with provisions for adequate pres -
sure, fire protection, and reliability of uninterrupted supply (Cesario, 1995;
AWWA, 2005). Water demand is the driving force for the operation of munici-
pal water systems. Because water demands are stochastic in nature, water sys -
tem operation requires an understanding of the amount of water being used,
where it is being used, and how this usage varies with time. For most water
systems the ratio of the maximum day water demand to the average day water
demand ranges from 1.2 to 3.0, and the ratio of the peak hour to the average day
is typically between 3.0 and 6.0. Of course, these values are system
specific, and seasonal variations may make these ratios even more extreme
(Walski et al., 2003). Demands may be classified as follows (Clark et al., 2004):
 Baseline demands, which usually correspond to consumer demands a
n
d
unaccounted-for-water associated with average dayconditions.
 Seasonal variations in demand because water use typically varies overthe
course of the year with higher demands occurring in the warmer months.
 Fire demands, which may be the most important consideration for w
a
-ter
systemdesign.
 Diurnal variations due to the continuously varying demands which a
r
e
inherent in water systems.
There is a need for research that relates distribution systemdesign to demand i
na
stochastic framework. Pioneering work by Buchberger and Wu (1995),
Buchberger and Wells (1996), and Buchberger at al. (2003) has found that resi-
dential water use follows a Poisson arrival process with a time dependent rate
parameter. Variations in demand have an important influence on water distribu-
tion system operation and in the determination of water age which in turn influ-
ences water quality, as discussed later in the chapter.
From an infrastructure perspective, a water distribution system is an elabo-
rate conveyance structure in which pumps move water through the system, con-
trol valves allow water pressure and flow direction to be regulated, and
reservoirs smooth out the effects of fluctuating demands (flow equalization)
and provide reserve capacity for fire suppression and other emergencies. All
these distribution system components and their operations and complex
interactions can
192

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
HYDRAULICINTEGRITY 193
produce significant variations in critical hydraulic parameters, such that
many opportunities exist for the loss of hydraulic integrity and degradation of
service. This, in turn, may lead to serious water quality problems, some of which
may threaten public health.
One of the most critical components of hydraulic integrity is the m
a
i
nt
e
- nance
of adequate pressure, defined in terms of the minimum and m
ax
i
m
umdesign
pressure supplied to customers under specific demand conditions. L
ow
pressures, caused for example by failure of a pump or valve, may lead to inade-
quate supply and reduced fire suppression capability or, in the extreme, intrusion
of potentially contaminated water. High pressures will intensify wear on valves
and fittings and will increase leakage and may cause additional leaks or breaks
with subsequent repercussions on water quality. High pressures will also in-
crease external load on water heaters and other fixtures. Pipes and pumps must
be sized to overcome the head loss caused by friction at the pipe walls and thus
to provide acceptable pressure under specific demands, while sizing of control
valves is based on the desired flow conditions, velocity, and pressure differen-
tial. A related need is to ensure that pressure fluctuations associated with surge
conditions are kept below an acceptable limit. Excessive pressure surges gener-
ate high fluid velocity fluctuations and may cause resuspension of settled parti-
cles as well as biofilm detachment.
A second element of hydraulic integrity is the reliability of supply,
which refers to the ability of the system to maintain the desirable flow rate even
when components are out of service (e.g., facility outage, pipe break) and is
normally accomplished by providing redundancy in the system. Examples
include looping of the pipe network and the development of backup sources to
ensure multiple delivery points to all areas.
Many water quality parameters change with length of time in the distribu-
tion system, a factor directly related to the hydraulic design of the system. For
example, chlorine residuals decrease with the increasing age of water and may
be completely lost, and trihalomethanes concentrations may increase with time.
In addition, higher concentrations of substances may leach from pipe materials
and linings if the contact time with the water is increased. Low velocities
in pipes create long travel times, resulting in pipe sections where sediments can
collect and accumulate and microbes can grow and be protected from disinfec -
tants. Furthermore, sediment deposition will result in rougher pipes with re -
duced hydraulic capacity. If peak velocity is increased or flow reverses in these
pipe sections due to any operational change or shock loading, such as tank fill-
ing or draining, valve opening or closing, pump going on- or off-line,
unexpected higher system pressure, or hydrant flushing, there is a risk that
deposits will be suspended and carried to consumers. Long detention times can
also greatly reduce corrosion control effectiveness by effecting phosphate
inhibitors and pH management. Thus, reducing residence time is an important
hydraulic issue both in pipes and in storage facilities.
A final component of hydraulic integrity is maintaining sufficient
mixing and turnover rates in storage facilities. Insufficient turnover rates and
incom-

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
plete (uneven) hydraulic mixing in reservoirs can allow short-circuiting b
e
t
w
e
e
nthe
tank inlet and outlet and generate pockets of stagnant water with depleted
disinfectant residual. This can lead to bacterial regrowth and other
biological changes in the water, including nitrification and taste and odor
problems.
This chapter discusses the factors that can cause the loss of hydraulic i
n
t
e
g
-rity,
the consequences of losing hydraulic integrity, how to detect loss of h
y
-draulic
integrity, techniques for maintaining hydraulic integrity, and how to recover
systemhydraulic integrity once it is lost.
FACTORS CAUSING LOSS OF HYDRAULIC INTEGRITY
There are many different ways that a water distribution system can lose i
t
s
hydraulic integrity, such that water quality becomes impaired. A loss of hydrau-
lic integrity implies a loss of positive line pressures, flow reversals, rapid
changes in velocity, a reduction in hydraulic capacity, a detrimental increase in
water residence time, or a combination of these events. Factors causing a loss of
systemhydraulic integrity include (1) pipe leaks and breaks, (2) rapid changes in
pressure and flow conditions, (3) planned maintenance activities and emergen-
cies, (4) tuberculation and scale formation in pipes, and (5) improper operational
control.
Pipe Deterioration
Pipe deterioration resulting in leaks or breaks can lead to a loss of hydraulic
integrity because adequate pressures can no longer be maintained. As discussed
in detail in Chapter 4, all pipe materials are vulnerable to some kind of chemical
or physical deterioration, and all water mains will require rehabilitation and
eventual replacement. Aging pipe infrastructure and chronic water main breaks
are a common problem for many water utilities. Analysis of water industry data
showed that on average, main breaks occur 700 times per day in the United
States (Cromwell et al., 2001). The condition of distribution system pipes
is influenced by material type and age, line pressure, type of soil, installation
procedures, and many other factors, making it difficult to predict where breaks
and leaks will occur. Chapter 4 discusses the roles of leak detection and
condition assessment in determining the current condition of distribution
systeminfra- structure.
Pressure Transients and Changes in Flow Regime
Rapid changes in pressure and flow caused by events such as rapid val
ve
closures or pump stoppages and hydrant flushing can create pressure surges of
excessive magnitude. These transient pressures,which are superimposed on the

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
The after effects of a water main break that occurred beneath the side walk of an urban street. Photo
courtesy of Bureauof LaboratoryServices,PhiladelphiaWater Department.
normal static pressures present in the water line at the time the transient occurs,
can strain the systemleading to increased leakage and decreased system reliabil-
ity, equipment failure, and even pipe rupture in extreme cases. High-flow ve-
locities can remove protective scale and tubercles, which will increase the rate of
corrosion. Uncontrolled pump shutdown can lead to the undesirable occurrence
of water-column separation, which can result in catastrophic pipeline failures
due to severe pressure rises following the collapse of the vapor cavities.
Vacuum conditions can create high stresses and strains that are much greater
than those occurring during normal operating regimes. They can cause the col-
lapse of thin-walled pipes or reinforced concrete sections, particularly if these
sections were not designed to withstand such strains. In less drastic cases,
strong pressure surges may cause cracks in internal lining, damage connections
between pipe sections, and destroy or cause deformation to equipment such as
pipeline valves, air valves, or other surge protection devices. Sometimes the
damage is not realized at the time, but may cause the pipeline to collapse in the
future, especially if combined with repeated transients.
Transient pressure and flow regimes are inevitable. All systems will, a
t
some
time, be started up, switched off, or undergo rapid flow changes such as those
caused by hydrant flushing, and they will likely experience the effects of
human errors, equipment breakdowns, earthquakes, or other risky disturbances
(Wood et al., 2005). Figure 5-1 illustrates typical hydraulic events following a
pump trip.

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
Low pressure transients may promote the collapse of water mains, l
e
a
k
a
g
einto the
pipes at joints and seals under sub-atmospheric pressures, and backsiphonage
(see Chapter 4). There is also evidence that pressure transients can lead to the
intrusion of contaminants into the distribution system. LeChevallier et al.
(2003) reported the existence of low and negative pressure transients in a num-
ber of distribution systems. Gullick et al. (2004) studied intrusion occurrences
in distribution systems and observed 15 surge events that resulted in a negative
pressure. Most were caused by the sudden shutdown of pumps at a pump station
because of either unintentional (e.g., power outages) or intentional (e.g., pump
stoppage or startup tests) circumstances. Friedman et al. (2004) confirmed that
negative pressure transients can occur in the distribution system and that
the intruded water can travel downstream from the site of entry. Locations with
the highest potential for intrusion were sites experiencing leaks and breaks, areas
of high water table, and flooded air-vacuum valve vaults.
FIGURE 5-1 Hydraulic events following a pump trip. The system is p
u
m
p
i
n
g
drinking w a- ter to an
elevated storage tank while serving the intermediate customers w ith adequate pressures.
Due to an unexpected power failure, the pump quickly runs down (loses speed). This w ill
create a negative pressure wave (downsurge) that w ill propagate into the distribution
system, putting the customers at a potential intrusion risk due to negative pressures. In
addition, it is possible that the pressure drops to the point that a vapor pocket forms
adjacent to the pump. Subsequently, this cavity w ill collapse and produce a large
pressure spike that can damage the pipeline and the seals which w ill make the system
even more vulnerable to low pressure events.

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
Hydraulic Changes during Maintenance and Emergencies
Water distribution systems are occasionally subject to emergencies o
r
planned
maintenance activities in which certain components become inoperable and the
system can no longer provide the minimum level of service to customers
(AWWA, 2005). Planned maintenance activities include supplies going off line
(e.g., stopping the treatment plant or shutting down a well); reservoir shutdown
for inspection, cleaning, or repairs; installation of new pipe connections; pipe
rehabilitation or break repairs; and transmission main valve repairs. Examples of
emergency situations include earthquakes, hurricanes, power failures, equip-
ment failures, or transmission main failures. All these activities can result in a
reduction in system capacity and supply pressure and changes to the flow paths
of water within the distribution system.
Tuberculation and Scale
The hydraulic capacity of distribution systems can be compromised by de
-
posits on the internal surface of the pipelines. The deposition of corrosion p
r
o
d
-ucts
in the form of tubercles and other types of scales on the interior of the pi
pe
scan
seriously clog water lines and thus restrict the flow of water. Scales m
a
yalso form
because metal salts such as calcium carbonate, aluminum silicate, e
t
c
.(see Chapter
6) in treated water entering the network are supersaturated, leading to their
precipitation on the pipe walls. Excessive pressure may be necessary to deliver
the required flow of water in pipes with tuberculation and scales, further
weakening aging pipes. The reduction in hydraulic capacity is caused by the
increases in head loss due to the roughness of the deposits and to the decrease in
pipe diameter that they cause.
Inadequate Operational Control
Historically, utilities have focused on the quality of water leaving the t
r
e
a
t
-ment
plant, because of regulatory drivers, and on the quantity of water supplied by the
distribution system, because of their mission to satisfy water demand and
maintain system pressure. Thus, it is not surprising that distribution system op-
erations at many utilities and their associated professionals (designers, builders,
plumbers, inspectors, etc.) have been water quantity focused rather than water
quality focused.
There is now greater recognition of the water quality effects of how l
ong
water is retained in the various elements of the distribution system. Retention
time or water age is strongly related to the characteristics of the system and its
operation. For example pipe roughness, which affects water flow and residence
time, may be modified by repair or rehabilitation. Operational activities, such as

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
The effects of internal corrosion, shown as a build up of tuberculation, on an unlined cast iron water
main. Photocourtesy of Bureauof Laboratory Services, PhiladelphiaWater Department.
pump scheduling and planned maintenance, or unplanned effects, such as unex-
pected changes in demand, will all have an effect on water age. A particularly
important issue that demonstrates the interaction of system operation and water
quality is the ability or inability of utilities to ensure adequate mixing intensity
and time in storage tanks to minimize short circuiting and to limit residence
times to be within acceptable limits. Interestingly, the design of tanks to ensure
adequate turnover is required in only 15 of 34 states that responded to a survey
of drinking water programs conducted by the Association of State Drinking Wa-
ter Administrators in March 2003 (see Table 2-4). Dealing with these issues is
discussed in the context of systemoperation later in this chapter.
CONSEQUENCES OF A LOSS IN HYDRAULIC INTEGRITY
There are several detrimental consequences of losing system hydraulic in-
tegrity, including contamination of the distribution system via intrusion, sedi-
mentation, a reduction in hydraulic capacity, loosening of scale, and extended
water age. Each of these has attendant water quality implications, as described
below.

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
External Contamination
A distribution system can become contaminated by the external e
n
v
i
r
o
n
-ment
for several reasons. The most well documented contamination events a
r
ebackflow
and direct contamination at breaks and repair sites, discussed in Chap-ter 4. A
specific type of backflow event related to a loss of hydraulic integrity is called
intrusion, which refers to the entrance of contamination into the water
distribution system through leaks (caused by corroded areas, cracks, and loose
joints) because of sustained low or negative pressures or a pressure transient.
When a section of the distribution system is depressurized due to a normal shut-
down, failure of a main or a pump, routine flushing, or emergency fire-fighting
water drawdown, contaminated water can be pulled into the main. For example,
during a large fire, a pump is connected to a hydrant. High flows pumped out of
the distribution systemcan result in a significantly reduced water pressure
around the withdrawal point. A partial vacuum is created in the system, which
can cause suction of contaminated water into the potable water system through
nearby leaks. During such conditions, it is possible for water to be withdrawn
from nonpotable sources into the distribution system and subsequently distrib-
uted to homes and buildings located near the fire. The same conditions can be
caused by a water main break.
Sustained low pressure events and transient pressure events that lead to i
n-
trusion of contaminated water have the potential for substantial water quality
and health implications. The potential for intrusion of contaminated groundwa-
ter into pipes with leaky joints or cracks seems greatest in systems with
pipes below the water table and where pathogens or chemicals are in close
proximity to the pipe. As discussed in Chapter 4, two recent studies
(Kirmeyer et al., 2001; Karim et al., 2003) have established that soil and water
samples collected immediately adjacent to pipelines can contain high fecal
coliform concentrations and viruses. In the event of a large intrusion of
pathogens, the disinfectant residual normally sustained in drinking water
distribution systems may be insufficient to neutralize contaminated water (see
Chapter 6 discussion on Adequate Disinfectant Residual). Transient events
can also generate high intensities of fluid shear and may cause resuspension
of settled particles as well as biofilm detachment.
Sedimentation
When water is moving slowly through a pipe, particles suspended in the w
a
-
ter may settle out into the pipe. The accumulated sediment reduces the pipe’s
hydraulic capacity. They also serve as a food source for bacteria and create a
hospitable environment for microbial growth. If not removed these
materials may cause water quality deterioration, taste and odor problems, or
discoloration of the water. This is particularly evident if the sediments are
disturbed (stirred up) by changes in the flow of water, such as when a main
break occurs,a service

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
reservoir is filling or draining, a pump is going on or off line, or during normal
hydrant flushing activities. The normal flow of water through the system will
reduce some but not all sediment accumulation over time, and supplemental
measures are periodically needed to clear out the system.
Reduction in Hydraulic Capacity and
Associated Increase in Pumping Costs
As metal pipes age their roughness tends to increase and their cross s
e
c-tional
area tends to decrease due to encrustation and tuberculation of corrosion
products on the pipe walls. This increase in hydraulic roughness and decrease in
effective diameter will increase the resistance to flow and reduce the hydraulic
capacity of the affected mains. Other deposits such as microbial slimes can also
result in a significant decrease in the hydraulic capacity of water mains. The
reduction in the hydraulic capacity can lead to a subsequent unwanted reduction
in system pressure due to the higher head loss. The loss in system pressure can
result in a water system that cannot deliver the necessary fire flows and, in the
extreme, it provides the potential for backflow of contaminants.
A flooded transmission main and metering chamber. This is a prime location for intru-
sion to occur in the event of lowor negative pressure transients. Photo courtesy of Phila-
delphia Water Department’s Bureauof LaboratoryServices.

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
In order to meet demand in such systems, higher pumping rates are neededto
overcome the higher head losses and to avoid or postpone the replacement,
duplication, or rehabilitation of tuberculated mains. This can overload motors
and result in a significant increase in energy consumption and operational and
maintenance costs of a water utility. Furthermore, the additional pumping can
over-pressurize certain portions of the distribution system, thereby increasing
leaks and breaks, and it can lead to ineffective utilization of storage tanks and
reservoirs because high pressure in the mains prevents outflow from the
reservoirs. If these reservoirs are subsequently put back into service during
peak times when consumption is high, this may result in the provision of “old”
(poor quality) water.
Poor Water Quality from Sediment Suspension and Removal of Scales
Changes in flow (magnitude and direction) within the water distributi
on
system as a result of hydrant flushing and valve and pump operation can scour
sediments, tubercles, and scales from the interior pipe walls and degrade water
quality. For example, when the water velocity is increased or flow direction is
reversed, sediment deposited on the pipe walls during periods of low flow may
be re-suspended and scales may detach. These materials may cause the water to
be colored, turbid, and sometimes odorous. Also, it is possible that these parti-
cles have adsorbed contaminants such as arsenic and other metals that
originated in the source water, as discussed in Chapter 6.
Hydraulic Integrity and Water Age
As distribution system water ages, its quality degrades, such that delivering
“younger” water is a desirable operational goal for water utilities. However, the
concept of water age is complex. Water age at a given location and time in a
water distribution systemis actually a mixture of water parcels that have trav-
eled along different paths through the distribution system with
correspondingly different travel times. Therefore, the age of water at a given
point in the distribution network is not a single value, but rather a distribution
of values, termed a residence time distribution (Levenspeil, 2002). This concept
is illustrated in Figure 5-2, which shows the results of a study conducted by
EPA in collaboration with the Greater Cincinnati Water Works in which a
calcium chloride tracer was introduced into an isolated portion of the
distribution system(Clark et al., 2004; Panguluri et al., 2005). Figure 5-2 shows
the field results from 34 continuously recording specific conductivity meters
that were deployed at various nodes in the system, with an EPANET modeling
prediction superimposed on the data. The three concentration peaks represent the
different parcels of water that have taken different routes to the monitoring
point, resulting in a residence time distribution at that monitoring station at the
time the data were collected.

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
FIGURE 5-2 Field results of monitoring at location CM-18 from 34 c
o
n
t
i
n
u
o
u
s
l
yrecording specific
conductivity meters, with a detailed all-pipe (non-skeletonized) EPANET modeling
prediction superimposed on the data. SOURCE: Panguluri et al. (2005).
For the purposes of this report, water age at a specific point in the distribu-
tion system is assumed to be the mean of the residence time distribution. The
report uses the term “water age” as a surrogate for water quality. However, it
should be noted that while water quality may depend on the age of the water, it
may also depend on the specific residence time distribution at that point in the
network or on one of its statistics (such as its variance). These complexities are
infrequently considered in studies where water age is measured, making this an
area ripe for additional research.
In addition to water age at any one point in the network being a distributionof
values, the age of water delivered to all consumers is also a distribution of
values, the shape of which depends on the location of the consumer, seasonality,
whether the network is looped versus one way, the existence of storage facilities,
etc. A typical system may deliver water to consumers that has resided in the
network for a few days, but many systems have some portion of the
network where residence time is much longer. For example, in Blacksburg,
Virginia, 97 percent of the water in the main distribution systemhas a water age
of less than 7 days, but 1 percent of the system has a residence time longer
than 28 days. Premise plumbing adds another layer of complexity that is
addressed in Chapter 8.
Hydraulically, increased water age is a consequence of many factors, i
n-
cluding the inevitable loss of carrying capacity as pipes age. However, system
design and operation have the most significant impact on water age, particularly
where water storage facilities are concerned. For example, high residence time
in these facilities can allow the disinfectant residual to be completely depleted,
thereby preventing the protection of finished water from additional
microbial contaminants that may be present in the distribution system
downstream of the

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
facilities. A survey of water utilities found that bacterial regrowth became a
problem in free chlorine residual systems when water age reached three days
whereas in monochloramine residual systems regrowth was not a problem until
water age reached or exceeded seven days (Baribeau et al., 2005). Other nega-
tive consequences ofincreased water age are discussed in Chapter 6.
DETECTING LOSS OF HYDRAULIC INTEGRITY
Ideally, the verification of hydraulic integrity should involve r
e
a
l
-
t
i
m
emonitoring
of pressure, flow direction, and velocity based on telemetry da
t
a
. This type of data
can be transmitted electronically from permanently installed measurement
devices in the field. Typical measurement locations should include treatment
plants and wells, pump and booster stations, reservoirs, valves, and other critical
points in the systemsuch as elevated sites.
An effective system-wide monitoring program can capture local
variations in hydraulic behavior (e.g., pressure, flow) at a specific point in a
water distribution system but cannot provide an overall understanding of the
spatial and temporal changes, complex flow pathways, and interactions among
the various water system characteristics. Thus, water distribution system
network models are attractive supplements to monitoring for evaluating
hydraulic and water quality changes throughout the distribution system. By
combining telemetry data and modeling information, water utilities can gain a
more complete and accurate picture of their systems hydraulic and water quality
operation and performance capabilities. For example, the North Marin Water
Authority in North Marin, California, draws its water from two sources, one of
very poor quality with high levels of natural organic matter and another source
of very high quality. Be- cause of demand variations there is a great deal of
mixing between the water sources at various nodes in the system leading to
wide variations in trihalomethane (THM) values over a given day. On the
surface these variations in THM concentration were unexplainable until
hydraulic modeling techniques were applied which clearly showed that these
variations were the result of the mixing effect from the two sources of water
(Clark and Buchberger, 2004). Hy- draulic integrity is best measured by
monitoring and modeling of the system hydraulic parameters, as discussed
below.
System-Wide Monitoring
Monitoring the operation of water distribution system components
yields data used to detect the system hydraulic integrity. This can be
accomplished in real-time by means of a Supervisory Control and Data
Acquisition (SCADA) system, which provides local and remote (supervisory)
real-time control and monitoring of selected process equipment and
parameters at strategic locations throughout the water distribution system. Any
parameter with a proper sensor

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
and transmitter that can produce an analog signal (e.g., 4-20 mADC)
proportional to the variation of the measured parameter can be monitored in
real-time or historically via the SCADA system. The acquired data can be
viewed on a real-time basis and also archived in a database for historical
evaluation at a later date. The data generated fromthe sensors and transmitters
is conveyed to the central control system using various communication media
such as telephone lines, fiber optic cables, or radio and cellular systems. The
amount of data collected is determined by the polling frequency of the SCADA
system.
To detect changes in hydraulic integrity, certain hydraulic characteristics o
f
water system components should be monitored continually in the distributi
on
system via SCADA. These include reservoir inflow/outflow rates, water vol-
umes and levels (used to calculate daily volume turnover), pump station opera-
tion such as status and speed settings, pump discharge flows and pressures,
valve positions, regulating valve downstream (and /or upstream) pressures, pipe
flow rates, and pressures at strategic sites. In addition, disinfectant residual,
temperature, conductivity, turbidity, dissolved oxygen, and pH can be continu-
ously monitored at the treatment plant. Temperature in storage tanks and reser-
voirs could also be monitored via SCADA to detect thermal stratification that
results from poor mixing characteristics. Temperature differential between the
inflow and the bulk water in the reservoir can result in density gradients inside
the storage facility and cause stratification and poor hydraulic mixing and, thus,
the greatest potentialfor water quality deterioration (Mahmood et al., 2005).
Continuous system-wide monitoring provides insight into the patterns
o
f
operational characteristics throughout the distribution system. An analysis of
these patterns can directly determine if the systemhydraulic integrity is not
compromised and the system is operating as designed, or detect any
unantici- pated operational anomalies. For example, high night-time flows in
specific areas could be an indicator of high leakage. Sonic leak detection
equipment (discussed in Chapter 4) can be used to pinpoint the exact
location of those leaks, which can then be isolated and repaired. Similarly,
unexpected low pressure readings, excessive pumping, or a drop in reservoir
levels in a specific area could indicate a large main break that may increase the
potential forbackflow.
Another function of SCADA is the ability to monitor and remotely control
local conditions of water system components based on any desired range of op-
erating conditions or set points. For example, a pump can be set to turn on or
off automatically when the pressure at a critical location or the water level in a
reservoir drops to a specified lower limit or goes above a specified upper limit.
Alarms can be set to alert operators when a fault within the system equipment
(e.g., equipment operating out of its normal range or overheating of a pump) and
any breach in the system hydraulic integrity is detected. For example, extreme
fluctuations in pressure and flow readings could result from pressure surges
generated from a power failure at a pump station. SCADA could then
divert water to the affected region from a different pump station, thus
ensuring adequate supply and fire flow protection.

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
SCADA systems also contain pertinent system operational information r
e
-
quired for water distribution network modeling (Cesario, 1995), such as t
he
boundary conditions (e.g., tank water levels, valve and pump statuses and set-
tings) for the network model as well as local flow and pressure conditions.
These data can be used for calibrating network models (the process of adjusting
model parameters so that modeled values reasonably match with measured
data), confirming normal system operation, verifying daily variation in total
system demands (based on a mass balance of the flows from the treatment plant
and wells and in and out of the reservoirs), estimating water losses during main
breaks, and investigating and solving operational problems. Operating data can
be time specific or represent several consecutive points in time for comprehen-
sive dynamic (extended period simulation) network modeling (e.g., 24-hour
simulation) (see Chapter 7 for details on modeling). Clark et al. (2004) list
many benefits of remote monitoring and network modeling for water security
protection.
Beyond remote controlled, real-time monitoring provided by SCADA, a
c
-
tual field measurements can be made to detect any potential loss of system h
y
-
draulic integrity. Hydrant tests are performed to determine if fire flow require-
ments are met as an indicator of the hydraulic strength of the water system.
Head loss tests are conducted to determine the hydraulic capacity of pipes as an
indication of system hydraulic performance capability. Pump efficiency tests
can be used to determine whether or not pump performance (e.g., overall system
efficiency, electrical motor performance, and pump hydraulics) is degrading
with time and if replacement or upgrading of equipment is warranted. Hydraulic
grade line tests of a pipeline profile (stretches of pipes) help locate partially
closed valves and deteriorated pipes with poor hydraulic capacity (high rough-
ness). Field measurements of pressure, flow conditions, velocity, and other wa-
ter system characteristics can also be carried out using a variety of measurement
devices at any facility to verify questionable SCADA readings.
Network Modeling
Computer based mathematical models provide an effective and vi
a
bl
emeans
of analyzing hydraulic and water quality conditions in distribution s
y
s
- tems (see
Clark and Grayman, 1998; Lansey and Boulos, 2005; Panguluri et al., 2005;
Boulos et al., 2006). They can calculate the spatial and temporal variations of
flow, pressure, velocity, reservoir level, water age, source contribution,
disinfectant concentration, and other hydraulic and water quality parameters
throughout the distribution system. These predictive capabilities are useful for
detecting a loss of system hydraulic integrity. For example, model results can
help identify areas of low pressures, excessive head losses, and high water age;
compute water losses; locate partially closed valves; verify that the replacement
or addition of a new supply source (e.g., emergency service connections or add-
ing a new reservoir or well) will have little or no effect on the flow, velocity,

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
and pressure patterns and residence times; estimate filling and draining cycles o
f
reservoirs; detect oversized facilities; calculate interzone water transfers; a
n
d
determine the adequacy of the system to supply fire flows under a variety
of demand loading and operating conditions.
A few specific models are of particular importance to maintaining hydraulic
integrity. First, surge models can be used to assess the hydraulic adequacy of the
system under various transient conditions, identify weak spots, and evaluate the
efficacy of surge control devices. These models could be instrumental in
future research to better understand the potential for intrusion to contaminate
distribution systems. Second, computational fluid dynamics (CFD) modeling
has potential for investigating hydraulic mixing and transport characteristics in
storage facilities and pipes for a wide range of designs and system operational
conditions (Panguluri et al., 2005). CFD models predict flow patterns, heat
transfer, and chemical reactions via the solution of partial differential equations
that describe conservation of mass, momentum, and energy in a two- or three-
dimensional grid that approximates the pipe or tank geometry. CFD models are
used to simulate temperature profiles, unsteady hydraulic and water quality con-
ditions, and decay of constituents in bulk flow and in storage facilities. How-
ever CFD modeling requires experienced and skilled programmers for effective
application (Panguluri et al., 2005). Such network modeling applications greatly
enhance the ability of water utilities to effectively manage, operate, and main -
tain their water distribution systems and deliver an adequate level of service to
their customers.
MAINTAINING HYDRAULIC INTEGRITY
Water utilities often find themselves choosing between two approaches t
o
preserve systemhydraulic integrity: (1) reacting only to emergencies or (2) act-ing
to prevent problems from occurring. The desirable approach is to develop an
active program to prevent future problems and service interruptions.
To maintain the hydraulic integrity of water distribution systems and ensure
the highest possible water quality, travel times in the system should be kept as
short as possible and large fluctuations in the hydraulic regime and low flow and
pressure conditions should be avoided. This can be accomplished by imple -
menting best design, management, operational, and maintenance practices,
as discussed below. Hydraulic modeling, discussed in the previous section, is
also a critical component that can be used to identify problems areas within the
distribution system and to develop design and operational alternatives that
address the deficiencies. Those practices necessary to maintain both
physical and hydraulic integrity, such as preventing the formation of
leaks and cracks in pipe mains and using backflow prevention devices, are
discussed in the previous chapter.

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
System Redundancy
Reliability of water distribution systems, which is necessary to m
i
nimiz
e
outages, is provided by building redundancy in the system in the form of l
oopi
ng
and backup sources. A looped (as opposed to branched) multi-source sy
s
t
e
mhas
the hydraulic advantage of carrying water to any location from more thanone
direction when a high rate is required (e.g., a fire flow demand) or when a pipe
or source is out of service (see Chapter 1). Sufficient interconnections between
the distribution mains are necessary to improve the ability of the system to
maintain the normal supply by re-routing the water when a breakdown occurs.
Dead-end distribution lines should be avoided. A fire -flow demand or large
water use on a dead-end main can only draw water through a single pipe, with
the maximum flow dictated by the size and length of the pipe. In addition, dur-
ing scheduled maintenance or repairs on dead-end mains both the supplied cus-
tomers and available fire flows will be affected. Availability of back-up power
(e.g., generators in pump stations), extra pumps, additional reservoirs, standby
wells, and emergency interconnections with other systems will provide the nec-
essary redundant sources.
Redundancy can also be facilitated by ensuring an adequate number of op-
erable valves and hydrants, as well as their strategic placement to allow for con-
trol of the system and for shutdown of sections for emergency repair and
planned maintenance (Male and Walski, 1991).
Management of Pressure Zones
Water distribution systems work best with minimal fluctuations in pressure.
The pressure differential range, which specifies the operating values for m
a
x
i
- mum
and minimum pressure to be maintained, is based on local engi
neeri
ngstandards and
conditions. Many states have established requirements for t
he design,
construction, operation, and maintenance of drinking water distribution systems
that relate to hydraulic parameters. For example, 32 of 34 responding state
require that distribution systems be designed for an operational pressure of at
least 20 psi under all flow conditions (see Table 2-3). Further, nine of 34 re-
quire both a minimum and maximum velocity through pipes. These require-
ments determine the maximum and minimum ground elevations that can be sup-
plied. The minimum pressure establishes the highest ground elevation that can
be supplied, and the maximum pressure defines the lowest ground elevation.
The former criterion ensures that the highest customers will be supplied with at
least the minimum pressure, while the latter ensures that the lowest customers
will not experience objectionably high pressures.
To supply water at acceptable pressure, the distribution system is thus d
i
-
vided into a number of distinct pressure zones. The maximum change in el
eva-tion
across each zone is determined by the difference between the maximum a
n
d
minimum design pressure values. Adding new pressure zones or adjusting ex
i
st-

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
ing pressure zone boundaries is needed when pressure differentials are
outside their desirable values. Pressure zone boundaries are delineated through
the use of closed valves. To improve reliability, pressure-regulating valves (or
pumps) are normally installed between the zones (along the pressure zone
boundaries), and stretches of new pipe are added to eliminate dead ends.
Pressure zoning is desirable but requires careful planning and design (
f
o
r
details, see Boulos et al., 2006). Proper design of pressure zones will reduce
leaks (because leakage normally varies exponentially with pressure and will be
reduced with a fall in system pressures), breaks, and pumping costs;
improve reservoir turnover rates; and avoid over-pressurizing the system.
Existing facilities (e.g., reservoirs, pumps, pressure regulating valves) and
natural (e.g., rives, lakes) or political boundaries (e.g., city limits, county and
state boundaries) will influence the design and modification of pressure zones
(Cesario, 1995).
Surge Protection
Pressure events or surges that can allow intrusion to occur are caused by
sudden changes in water velocity due to loss of power, sudden valve or hydrant
closure or opening, a main break, fire flow, or an uncontrolled change in on/off
pump status (Boyd et al., 2004). Intrusion can be minimized by knowing the
causes of pressure surges, defining the system’s response to surges, and estimat-
ing the system’s susceptibility to contamination when surges occur (Friedman et
al., 2004). Pressure transients in distribution systems are usually most severe at
pump stations and control valves, in high-elevation areas, in locations with low
static pressures, and in remote locations that are distanced from overhead stor-
age (Friedman et al., 2005).
A number of devices can be used for controlling transients in pipeline sys-
tems (Boulos et al., 2005, 2006; Wood et al., 2005). The general principles of
pressure surge control devices are to store water or otherwise delay the change
of flow or to discharge water from the line so that rapid or extreme fluctuations
in the flow regime are minimized. Devices such as pressure-relief valves, surge
anticipation valves, surge vessels, surge tanks, pump bypass lines, or any com-
bination thereof are commonly used to control maximum pressures. Storage
tanks with a free water surface can be effective in controlling surges. Minimum
pressures can be controlled by increasing pump inertia or by adding surge ves-
sels, surge tanks, air-release/vacuumvalves, pump bypass lines, or any combina-
tion of these components. The overriding objective is to reduce the rate at
which flow changes occur. Figure 5-3 illustrates typical locations for the vari-
ous surge protection devices in a water distribution system.
Because no two distribution systems are hydraulically the same, there a
r
e no
general rules or universally applicable guidelines for eliminating objectionable
pressures in distribution systems. Any surge protection device must b
e
chosen
accordingly. The final choice will be based on the initial cause and location of
the transient disturbance(s), the system itself, the consequences if

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
FIGURE 5-3 Typical locations for various surge protection devices. S
O
U
R
C
E
: Reprinted, w ith
permission, fromBoulos et al. (2006). © 2006 by MWH Soft Pub.
remedial action is not taken, and the cost of the protection measures. A com
bi-
nation of devices may prove to be the most effective and economical. Final de-
termination of the adequacy and efficacy of the proposed measure should
be checked and validated using detailed surge modeling. Boulos et al. (2005,
2006) provide a detailed transient flow chart that offers a comprehensive guide
to the selection of components for surge control and suppression in distribution
systems. Good maintenance, pressure management, an adequate disinfectant
residual, and routine monitoring programs are also essential components of
transient protection.
Flushing Water Mains
Flushing is one of the most ubiquitous activities of water utilities for both
maintaining and recovering the integrity of distribution systems because it is t
h
e
primary means by which to remove contaminated water from the system. It w
a
s
discussed briefly in Chapter 4 in association with the cleaning and
disinfection of water mains following pipe installation, repair, and replacement. It
is a topic of Chapter 6, which focuses on water quality integrity, because
flushing is routine in areas with repeat customer complaints about color, taste,
or odor; in dead ends mains; and in storage facilities. Its importance with respect
to maintaining hydraulic integrity is that flushing removes accumulated
sediment and corrosion

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
products that reduce the hydraulic capacity of the pipe, improving the flow
o
fwater through the distribution system.
Flushing (discussed in greater detail in a subsequent section) is performed by
isolating sections of the distribution system and opening fire hydrants (
o
r
flushing valves) to cause a large volume of flow to pass through the isolated
pipelines so that a scouring action is created. Water is then discharged through a
hydrant, which in turn removes any material buildup from the pipe. When
flushing pipes, it is important to ensure that the flushing velocity is sufficient to
suspend loose sediments. Flushing should continue until the water has cleared
and disinfectant residual has reached normal expected levels. To minimize any
negative environmental impacts (as flushed water may be high in suspended
solids and other contaminants that can harm waterbodies), flushed water is nor-
mally discharged into sanitary or combined sewers or storm water management
facilities. It is important to optimize flushing programs, as excessive flushing
can waste significant volumes of water.
Operation and Design for Water Age Minimization
As discussed in Chapter 1, a primary reason for water quality pr
obl
em
s within
distribution systems is the advanced water age necessitated by the provi-sion of
adequate standby fire flow and redundant capacity. This requires that
utilities use standpipes,elevated tanks, and large storage reservoirs, as well as
Hydrant flushing.Photocourtesy of Bureauof Laboratory Services,PhiladelphiaWater
Department.

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
larger-sized pipes than would otherwise be necessary. The effect of
designing and operating a system to maintain adequate fire flow and
redundant capacity can result in long travel times and low velocities
between the treatment plant and the consumer, which can be detrimental to
water quality.
Brandt et al. (2004, 2006) have recently completed a two-volume st
udy
sponsored by the American Water Works Association Research Foundation
to suggest ways to minimize water age (retention times) while at the same
time controlling water quality degradation and providing the pressure and
quantity constraints that are required to maintain water service. In particular,
Brandt et al. (2006) have developed a diagnostic methodology by which a water
utility can assess and then minimize water quality problems associated with
excessive retention times. Best management practices for controlling
retention time can generally be categorized into storage and network
methods. Storage methods include adjusting pump schedules, reducing the
operational top water level, removing storage tanks from service, and
reconfiguring reservoir and storage tanks to avoid dead zones. Network
methods include altering network valving patterns, installing time actuated
valves, flushing (manual and automated), and abandoning and downsizing mains
(Brandt, 2006).
An important aspect of hydraulic integrity maintenance is to ensure s
u
f
f
i
- cient
mixing and to minimize water age in storage facilities—issues which if n
o
t
addressed can generate pockets of stagnant water with depleted disinfectant r
e
-
sidual and associated water quality problems. Mixing will eliminate i
nternal
dead zones within a storage facility and prevent short-circuiting between the
tank inlet and outlet. Completely mixed flow can be achieved by using a turbu-
lent (high velocity) inlet jet, mechanical mixers, or hydraulic circulation sys -
tems. Controlling pumping rates and fill and discharge rates can also provide
adequate intensities to achieve complete mixing. For example, Grayman et al.
(2000) recommend that to avoid stratification in distribution storage facilities,
the fill time should exceed the mixing time. A utility’s SCADA system can be
used to monitor the real-time mixing intensity within a storage facility, and as
such is useful for process control. It should be noted that utilities may be con-
strained in their ability to provide complete mixing due to the increased energy
requirements.
Both poor mixing and improper tank discharge management can i
ncrease
the residence time of water in a service reservoir. To combat this potential prob-
lem, frequent exercising of reservoirs (i.e., continuously mixing the water
and making sure that fresh water replaces stagnant water) is required. Grayman
et al. (2000) used various modeling techniques to develop a set of general guide-
lines for reducing water quality deterioration associated with inadequate mixing
and excessive water age in distribution storage facilities. They reviewed the
application of CFD, compartment, and physical scale models. A stand-alone
model called CompTank is presented which provides a wide range of alterna-
tives and allows the user to model water age and the concentrations of reactive
or conservative substances overlong time periods.

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
***
There is limited information about how to operationally reduce water age i
n
an
existing system while taking into account larger issues such as mi
nimizing
operational costs and maintaining the other aspects of hydraulic integrity, such
as reliability of supply and adequate pressure for all water uses. At the present
time, there is so much variability in the system parameters affecting distribution
system operation that it is not possible, for example, to quantify the
tradeoff between the risk of running out of water and the risk of delivering water
of poor quality. This quandary is manifested in our inability to optimally
maintain and operate storage facilities. The benefits of large storage tanks are
not clear, nor is it easy to determine whether to remove a tank from service or
reduce its volume. Answering such questions will require research that
quantifies how various actions (such as removing a storage tank from service)
will affect other aspects of hydraulic integrity (such providing fire flow and
minimizing water age) within a given distribution system.
RECOVERING HYDRAULIC INTEGRITY
When a distribution system experiences high head losses, inadequate pres-
sures or flows, high turbidity from scale loosening or resuspension of sediment,or
low disinfectant residual and high bacterial counts from advanced water, there are
several steps the utility should take. One of the first steps is to consider one or
more of the standard techniques available to remove any loose sediment,
biofilm, and tubercles that may be the cause of the problem. These procedures
can restore most of the pipes’ original hydraulic capacity, and include conven-
tional and unidirectional flushing, air scouring, swabbing, abrasive pigging,
chemical cleaning, mechanical cleaning and lining (nonstructural, cement or
epoxy applied linings), and structural lining. If the problems persist even after
the application of these techniques, replacement of the pipes should be consid-
ered (see Chapter 4). A brief discussion ofeach technique follows.
It should be noted that to overcome increasing head losses and local defi-
ciencies in system pressure and to increase the carrying capacity of water m
a
i
n
s
,
increased levels of pumping are usually needed. This will result in increases i
n
energy consumption and increased operational costs fora water utility.
Conventional Flushing
Conventional flushing generally involves opening hydrants in a specific area
of the distribution system until the water visually runs clear. While e
f
f
e
c
- tive in
quickly removing loose particles, this type of flushing is usually not e
f
- fective in
dislodging well-attached deposits and cannot remove scales and t
u
b
e
r
-culation.
Because in a looped systemthe water will flow to the hydrant from

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
multiple mains and directions, it becomes very difficult to achieve the h
i
gh
-
velocity flushing required to scour and remove deposits (as shown in Figure 5-
4). As a result, some sediment and biofilm may not be removed, and the
cleanup method requires a substantial quantity of water. In addition, because the
dynamics of the entire distribution system are not considered, it is possible that
the water used to the flush the system may come from a component that has not
been previously cleaned. Therefore, sediments, detached biofilm, etc., may
simply be transported from one part of the distribution systemto another.
Unidirectional Flushing
Unidirectional flushing involves the closure of valves and opening of h
y
-
drants to create a one-way flow in the water mains (see Figure 5-5). This in-
creases the speed of the water flow so that the shear velocity near the pipe wall
is maximized, producing a scouring action in the mains, effectively removing
sediment deposits and biofilm. Flushing should start at a clean water
source (e.g., pump station) and proceed outward in the system so that flushing
water is drawn from previously flushed reaches. This ensures that clean water is
always used to flush the mains. No special equipment is needed; however,
substantial planning time is required to define the flushing zones, the valves and
hydrants to be operated, the duration of the flush for each zone, the required
velocities, and the sequence of operation. A hydraulic model of the
distribution systemcan
FIGURE 5-4 Conventional flushing in a looped system results in water flow ing toward the
hydrant from all directions generating lower velocity and less scouring of the mains.
SOURCE: Reprinted, w ith permission, fromBoulos et al. (2006). © 2006 by MWH Soft Pub.

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
greatly simplify and expedite the planning process, especially for estimati
ng
pipe flow rates, velocities, and flushing times. While more costly and time con-
suming than conventional flushing, unidirectional flushing is more effective and
uses less water (Hasit et al., 2004). There are often long-term water quality
benefits because deposits and water of questionable quality are actually removed
rather than being re-routed to otherparts of the distribution system.
Work done by Slaats (2001) demonstrated the velocities needed to entrain sedi
-
ments, and these were within the range of velocities used for flushing. C
a
r
r
i
e
r
eet al.
(2002) showed that loose deposits could be removed by unidirectional
flushing as a function of time, pipe material, and water characteristics. Gauthier
et al. (1997) showed that loose deposits in a French system removed by flushing
contained organisms including invertebrates, protozoa, and bacteria (although it
should be noted that French distribution systems maintain no disinfectant resid-
ual such that their ecology is not representative of U.S. distribution systems).
The abiotic constituents were primarily iron, volatile solids, calcium, aluminum,
and other insoluble materials. Deposits flushed from four systems in the United
Kingdom were all high in iron and manganese (Marshall, 2000).
Not all systems can or will routinely flush. There may be water restrictions
that preclude flushing, and customers may be upset if they see water
being
FIGURE 5-5 Unidirectional flushing results in water flowing toward t
h
ehydrant in only one
direction resulting in higher velocity, more scouring, and better cleaning of the mains with
less water use. SOURCE: Reprinted, w ith permission, from Boulos et al. (2006). © 2006 by
MWH Soft Pub.

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
“wasted” while they are being told to conserve. Additionally, there is
often a
requirement that disinfectant residuals in the flushed water be neutralized,
and this may be more complicated if chloramines are present compared to
chlorine.
Air Scouring, Swabbing, and Abrasive Pigging
There is a long history of cleaning pipelines in order to remove accumulated
material resulting from corrosion, improper pH adjustment, post precipitation of
water treatment chemicals, and biofilm growth. Cleaning usually is a precursor
to another process like lining or insertion rather than a process onto itself. This is
due to the fact that cleaning potentially exposes unprotected metal pipe which
would result in additional water quality problems.
Scouring, swabbing, and abrasive pigging are progressively more aggres-
sive cleanup techniques that involve more specialized equipment and specialized
skills. Although a few water utilities have implemented these methods using
their own staff, typically these methods are contracted to specialty firms. Air
scouring involves the continuous injection of filtered, compressed air into the
pipe, along with a continuous but smaller flow of water. Given a continuous
supply of water and air in the right proportions, discrete “slugs” of water
are formed in the pipe and driven along by the compressed air at high velocity.
The high velocity slugs tend to remove silt, sediment, loose matter, and debris
from the base of the pipe. No disassembly of the pipe is necessary. Water
scouring involves the insertion of a high-pressure water jet into the pipe to
remove deposited materials. The water jet pressure can be adjusted to remove
the deposits without damaging the piping material. The jet will back flush
the deposited material to the insertion point in the pipeline. While jetting is very
effective, it is limited to the length of the jetting equipment, which will result
in frequent insertion points,and to small diameter pipes.
More aggressive techniques, such as swabbing and abrasive pigging, workto
varying degrees in removing heavy sediment, biofilm, adherent material, tu-
berculation, and even very hard scale (Ellison et al., 2003). Swabbing involves
driving cylindrical foam sponges (known as swabs) through pipes using water
pressure. The swabs travel along the water main and scrub the scale encrusta-
tions and slime build-up from the inner pipe walls. Loosened debris and swabs
are eventually flushed out at an exit point. Currently, pigging is used primarily
if there are hydraulic problems in the water mains, i.e., to improve the “C” factor
(roughness coefficient) of the pipes. It involves isolating a segment of the dis-
tribution system and passing a fluid-propelled object through the pipe. A styro-
foam plug is often used as the “pig,” which is normally the same or slightly big-
ger in diameter than the water main and is shaped like a torpedo (Deb, 1990).
Increasing sizes of pigs are passed through the pipeline to gradually
remove deposits within the pipeline. The abrasiveness of the pig results in
varying quality of water being discharged during the cleaning process. For
instance, higher

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
concentrations of suspended solids normally follow the more abrasive pigs, a
s
these scourthe inner lining of the pipe.
Both pigging and swabbing can be difficult to implement because they r
e
-
quire the removal of hydrants or the installation of new pipeline appurtenances
(e.g., pig launching and receiving stations). Few water utilities have i
m
pl
e
-mented
these methods using their own staff, such that these methods are usuall
ycontracted
to specialty firms.
Chemical Cleaning, Mechanical Cleaning, and Lining
Chemical cleaning to restore old pipes involves the recirculation in an iso-
lated pipe section of proprietary acids and surfactants to remove scale and de-
posit, while mechanical cleaning is accomplished by dragged scrapers. Scrapers
are devices that use springs to force blades against the wall of the pipe. As the
device moves through the pipe, the blades scrape the material off the walls
which can then be flushed from the pipe. These techniques are typically applied
in the rehabilitation of older unlined cast iron pipes that have become scaled and
tuberculated. In another example, a process using a cleansing solution of an
organic oxide scavenger and muriatic acid circulated through an isolated section
of distribution main worked effectively for small diameter pipelines (Estrand,
1995). Compared to air scouring and pigging, chemical clean ing is
infrequently used due to the cost of chemicals and their proper disposal after
cleaning.
It is common practice to reline a cleaned pipeline to protect the newly e
x
-
posed metallic pipeline material. The most common technique is to use concrete
mortar applied to the internal surface, a technology that has been used for over50
years. Spray-on epoxy lining is a newer method that is especially u
s
e
f
u
lwhen
the water is low in hardness, which can cause a cement lining to deterio - rate.
Most recently, polyurethane lining is becoming a competitive alternative to
concrete mortar lining especially in long pipelines with few service connections.
This type of chemical lining on the inner surface of the pipe is referred to
as nonstructurallining and does not increase the pipe’s structural integrity.
Maintaining the hydraulic integrity of distribution systems is vital to ensur-
ing that water of acceptable quality is delivered in acceptable amounts. The
most critical element of hydraulic integrity is adequate water pressure inside the
pipes. The loss of water pressure resulting from pipe breaks, significant leak-
age, excessive head loss at the pipe walls, pump or valve failures, or pressure
surges can impair water delivery and increase the risk of contamination of the
water supply via intrusion. In addition, slow moving water or changes in the
flow regime (including flow reversals) and advanced water age can negatively
impact finished water quality. Proper system design, operation, and mainte-

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
nance, along with monitoring and modeling, can help water utilities achieve a
high
degree of hydraulic integrity and reliability and extend the life of their dis -
tribution systems. The following conclusions and recommendations focus on
the highest priorityissues.
Water residence times in pipes, storage facilities, and premise plumbing
should be minimized. Excessive residence times can lead to low disinfectant
residuals and leave certain service areas with a less protected drinking w
a
t
e
rsupply.
In addition, long residence times can promote microbial regrowth a
nd the
formation of disinfection byproducts. From an operational viewpoint it m
a
ybe
challenging to reduce residence time where the existing physical infrastructure
and energy considerations constrain a utility’s options. Furthermore, lim
- ited
understanding of the stochastic nature of water demand and water a
g
e
makes it
difficult to assess the water quality benefits of reduced residence time. Research
is needed to investigate such questions, as well as how to achieve
minimization of water residence time while maintaining other facets of hydrau-
lic integrity (such as adequate pressure and reliability of supply).
Positive water pressure should be maintained. Low pressures in the dis-
tribution systemcan result not only in insufficient fire fighting capacity but c
a
nalso
constitute a major health concern resulting from potential intrusion of con-
taminants from the surrounding external environment. A minimum residual
pressure of 20 psi under all operating conditions and at all locations (including at
the system extremities) should be maintained. The minimum value could be
adjusted based on site specific conditions.
Where feasible, surge protection devices should be installed. Because
these devices provide the only practical opportunity to prevent intrusion of con-
taminants due to low or negative pressure events, surge tanks should be consid-
ered at all pump stations (to dampen negative pressure waves) and other surge
control devices at vulnerable locations in the system such as high points. This
can be aided by a comprehensive surge analysis on a representative network
model of the distribution system to select, locate, and size the most
effective combination of surge protection devices. Although looped networks
are generally less susceptible to objectionable pressure transients than single
long transmission main systems, they must still be protected against low or
negative pressure transients.
Distribution system monitoring and modeling are critical to maintain-
ing hydraulic integrity. Hydraulic parameters to be monitored should
include inflows/outflows and water levels for all storage tanks, discharge
flows and pressures for all pumps, flows and/or pressure for all regulating
valves, and pressures at critical points. An analysis of these patterns can
directly determine if the system hydraulic integrity is compromised or if the
systemis operating as designed, or detect any unexpected operational anomalies.
Calibrated distribu-

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
tion system models can calculate the spatial and temporal variations of f
l
o
w
,
pressure, velocity, reservoir level, water age, and other hydraulic and w
a
t
e
r quality
parameters throughout the distribution system. Such results can, for ex-ample,
help identify areas of low or negative pressure and high water age, estimate
filling and draining cycles of storage facilities, and determine the adequacy of
the system to supply fire flows under a variety of demand loading and operat-
ing conditions.
REFERENCES
American Water Works Association (AWWA). 1986. Introduction to Water Distribu-
tion Principles and Practices of Water Supply Operations. Denver, CO: AWWA.
AWWA. 2005. AWWA Manual M32: Computer modeling of water distribution s
y
s
-
tems. Denver, CO:AWWA.
Baribeau, H., N. L. Pozos, L. Boulos, G. F. Crozes, G. A. Gagnon, S. Rutledge, D. S
k
i
n
-ner,
Z. Hu, R. Hofmann, R. C. Andrews, L. Wojcicka, Z. Alam, C. Chauret, S. A
.
Andrews, R. Dumancis, and E. Warn. 2005. Impact of Distribution System Water
Quality on Disinfection Efficacy. Denver, CO:AwwaRF.
Boulos, P. F., B. W. Karney, D. J. Wood, and S. Lingireddy. 2005. Hydraulic t
r
a
n
s
ie
n
tguidelines
for protecting water distribution systems. J. Amer. Water Works Assoc. 97(5):111–
124.
Boulos, P. F., K. E. Lansey, and B. W. Karney. 2006. Comprehensive Water D
i
s
t
r
i
b
u
- tion
Systems Analysis Handbook for Engineers and Planners. Second edition. Pasa-
dena, CA: MWH Soft Pub.
Boyd, G. R., H. Wang, M. D. Britton, D. C. Howie, D. J. Wood, J. E. Funk, and M
. J.
Friedman. 2004. Intrusion within a simulated water distribution system due to hy-
draulic transients. 1: Description of test rig and chemical tracer method. J. Environ.
Eng. 130(7):774–783.
0
0
4
.
Managing
Distribution Retention Time to Improve Water Quality—Phase I. Denver CO:
AwwaRF.
0
0
4
.
Managing
Distribution Retention Time to Improve Water Quality—Phase II. Den- ver CO:
AwwaRF.
Buchberger, S. G., J. T. Carter, Y. H. Lee, and T. G. Schade. 2003. Random D
e
m
a
n
d
s
, Travel
Times and Water Quality in Deadends. Denver, CO:AWWA.
Buchberger, S. G., and G. J. Wells. 1996. Intensity, duration and frequency of residential
water demand. J. Water Resources Planning and Management 122(1):11–19.
Buchberger, S. G., and L. Wu. 1995. A model for instantaneous water demand. J. H
y
-
draulic Eng. 121(3):232–246.
Carriere, A., B. Barbeau, V. Gauthier, C. Morissette, R. Millette and A.Lalumiere. 2
0
0
2
.
Unidirectional flushing: loose deposits characterization in the test zones of four C
a
-
nadian distribution systems. In: Proceedings of the AWWA Water Quality
Tech- nology Conference. Denver, CO: AWWA.
Cesario, L. 1995. Modeling, analysis and design of water distribution systems. D
e
n
v
e
r
, CO:
AWWA.

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
Clark, R. M., and S. G. Buchberger. 2004 Responding to a contamination threat i
n a
drinking water network: the potential for modeling and monitoring. Pp 9.1-9.26 In:
Water Supply Systems Security. L. W. Mays (ed.). New York: McGraw-Hill.
Clark, R. M., W. M. Grayman, S. G. Bucberger, Y. Lee, and D. J. Hartman. 2
0
0
4
.
Drinking
water distribution systems: an overview. Pp 4.1-4.2 In: Water Supply Sys- tems
Security. L. W. Mays (ed.). New York: McGraw-Hill.
a
t
e
r
distribution
Clark, R. M., S. Panguluri, and R. C. Haught. 2004. Remote monitoring and n
e
t
w
o
r
k
models: their potential for protecting U.S. water supplies. Pp. 14.1–14.22 In: Water
Supply Systems Security. Mays, L. W. (ed). New York: Mc Graw-Hill.
Cromwell, J., G. Nestel, and R. Albani. 2001. Financial and economic optimization o
f
water
main replacement programs. Denver, CO: AwwaRF.
Deb, A. K, J. K. Snyder, J. J. Chelius, and D. K. O’Day. 1990. Assessment of E
x
i
s
t
i
n
gand
Developing Water Main Rehabilitation Practices. Denver, CO: AwwaRF.
Ellison, D., S. J. Duranceau, S. Ancel, G. Deagle, and R. McCoy. 2003. Investigation
o
f
PipeCleaning Methods. Denver, CO: AwwaRF.
Estrand, C., A. Hicatt, and J. Ludwidg. 1995. Chemical cleaning process for water p
i
p
e
systems. In: Proceedings of the Hydraulics of Pipelines Conference, ASCE, Phoe-
nix, AZ.
Friedman, M., L. Radder, S. Harrison, D. Howie, M. Britton, G. Boyd, H. Wang, R. G
u
l
-
lick, D. Wood and J. Funk. 2004. Verification and Control of Pressure T
r
a
n
s
i
e
n
t
sand
Intrusion in Distribution Systems. Denver, CO: AwwaRF.
Gauthier, V., C. Rosin, L. Mathieu, J. M. Portal, J. C. Block, P. Chaix, and D. G
a
te
l
.
1997. Characterization of the loose deposits in drinking water distribution systems.
In: Proceedings of the AWWA Water Quality Technology Conference.
Denver, CO: AWWA.
Grayman, W. M., L. A. Rossman, C. Arnold, R. A. Deininger, C. Smith, J. F. Smith, and
R. Schnipke. 2000. Water quality modeling of distribution systemstorage facilities.
Denver, CO: AwwaRF.
Gullick, R. W., M. W. LeChevallier, R. C. Svindland, and M. J. Friedman. 2004. O
c
c
u
r
-rence
of transient low and negative pressures in distribution systems. J. Amer. W
a
-ter Works
Assoc. 96(11):52–66.
Hasit, Y. J., A. J. DeNadai, H. M. Gorill, S. B. McCammon, R. S. Raucher, a
n
d J.
Whitcomb. 2004. Cost and Benefit Analysis of Flushing. Denver, CO: AwwaRF.
Karim, M., Abbaszadegan, M. and M. W. LeChevallier. 2003. Potential for p
a
t
h
o
g
e
n
intrusion during pressuretransients. J. Amer. Water Works Assoc.95(5):134–146.
Kirmeyer, G. J., M. Friedman, K. Martel, D. Howie, M. LeChevallier, M. Abbaszadegan,
M. Karim, J. Funk, and J. Harbour. 2001. Pathogen intrusion into the
distribution system. Report No. 90835. Denver, CO: AwwaRF and AWWA.
Lansey, K. E., and P. F. Boulos. 2005. Comprehensive Handbook on Water Q
u
a
l
i
t
y
Analysis for Distribution Systems. Pasadena, CA: MWH Soft Pub.
LeChevallier, M. W., R. W. Gullick, M. R. Karim, M. Friedman, and J. E. Funk. 2
0
0
3
. The
potential for health risks from intrusion of contaminants into distribution sys-
tems from pressuretransients. Jour. Water Health 1(1):3–14.
Levenspeil, O. 2002. Modeling in chemical engineering. Chemical Engineering S
c
i
e
n
c
e57:
4691–4696.
Mahmood, F., J. G. Pimblett, N. O. Grace, and W. M. Grayman. 2005. Evaluation of
water mixing characteristics in distribution system storage tanks. J. Amer. Water
Works Assoc. 97(3):74–88.

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
Marshall, G. P. 2000. Understanding and Preventing Discolored Water. UKWIR Report
#01/DW/03/17. London: UKWIR Ltd.
Panguluri, S., W. M. Grayman, and R. M. Clark. 2005. Distribution system water q
u
a
l
-ity
quality i
n
Development.
Slaats, N. (ed.). 2001. Processes Involved in the Generation of Discolored Water.
Nieuwegein, the Netherlands: KIWA.
Walski, T. M., D. V. Chase, D. A. Savic, W. M. Grayman, S. Beckwith and E. Koelle.
2003. Advanced Water Distribution Modeling and Management. Waterbury, CT:
Heastad Press.
Wood, D. J., S. Lingireddy, and P. F. Boulos. 2005. Pressure Wave Analysis of T
r
a
n
- sient
Flow in PipeDistribution Systems. Pasadena, CA: MWH Soft Pub.

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
6
Water Quality Integrity
As discussed in Chapters 4 and 5, breaches in physical and hydraulic i
nteg-
rity can lead to the influx of contaminants across pipe walls, through breaks, and
via cross connections. These external contamination events can act as a source
of inoculum, introduce nutrients and sediments, or decrease disinfectant concen-
trations within the distribution system, resulting in a degradation of water qual-
ity. Even in the absence of external contamination, however, there are situations
where water quality is degraded due to transformations that take place within
piping, tanks, and premise plumbing. Most measurements of water quality taken
within the distribution system cannot differentiate between the deterioration
caused by externally vs. internally derived sources. For example, decreases in
disinfectant concentrations with travel time through the distribution system
could be the result of demand from an external contamination event or it could
be due to disinfectant reactions with pipe walls and natural organic matter
remaining after treatment.
This chapter deals with the various internal processes or events occurri
ng
within a distribution system that lead to degradation of water quality, the conse-
quences of those processes, methods for detecting the loss of water quality, op-
erational procedures for preventing these events, and finally, how to restore wa-
ter quality integrity if it is lost. In many cases, the detection methods and recov-
ery remedies are similar to those discussed in previous chapters.
FACTORS CAUSING LOSS OF WATER QUALITY INTEGRITY
AND THEIR CONSEQUENCES
For water quality integrity to be compromised, specific reactions must occur
that introduce undesirable compounds or microbes into the bulk fluid of the dis -
tribution system. These reactions can occur either at the solid–liquid
interface of the pipe wall or in solution. Obvious microbial examples include the
growth of biofilms and detachment of these bacteria within distribution system
pipes and the proliferation of nitrifying organisms. Important chemical
reactions include the leaching of toxic compounds from pipe materials, internal
corrosion, scale formation and dissolution, and the decay of disinfectant residual
that occurs over time as water moves through the distribution system. All these
interactions are governed by a suite of chemical and physical parameters
including temperature, pH, flow regime, concentration and type of disinfectant,
the nature and abundance of natural organic matter, pipe materials, etc.
Many of these variables may be linked in distribution systems; for example,
seasonalincreases
221

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
in temperature may be accompanied by changes in organic matter, flow regimes,
and disinfectant concentrations. As a consequence, attempting to correlate the
occurrence of a given event (such as corrosion, microbial growth,
disinfectant decay, or DBP formation) within distribution systems to a single
variable (such as temperature) is difficult.
Biofilm Growth
One way in which water quality can be degraded in the distribution s
y
s
t
e
m
is due
to the growth of bacteria on surfaces as biofilms. Virtually every water
distribution system is prone to the formation of biofilms regardless of the purity
of the water, type of pipe material, or disinfectant used. The extent of biofilm
formation and growth, the microbial ecology that develops, and the subsequent
water quality changes depend on surface-mediated reactions (e.g., corrosion,
disinfectant demand, immobilization of substrates for bacterial growth), mass
transfer and mass transport processes, and bulk fluid properties (concentration
and type of disinfectants, general water chemistry, organic concentration, etc.).
These interactions can be exceedingly complex, which typically means that the
mechanisms leading to biofilm growth may not be obvious and are often system
specific.
Bacteria growing in biofilms can subsequently detach from the pipe walls.
Because these organisms must survive in the presence of the disinfectant r
e
s
i
d
-ual
present in the distribution system, the interaction between the suspended
organisms and residual is critical. If the residual has decayed due to reactions
with compounds in the water or with the pipe wall, intrusion, or other sufficient
external contamination, it is possible for attached bacteria to be released into
water that contains insufficient disinfectant to cause their inactivation. The po-
tential for this to occur is higher in premise plumbing, which generally has
longer water residence times that may lead to very low disinfectant concentra-
tions.
Pathogenic Microorganisms
An obvious risk to public health from distribution system biofilms is the r
e
-
lease of pathogenic bacteria. As discussed in Chapter 3, there are i
n
s
t
a
n
c
e
s
where
opportunistic pathogens have been detected in biofilms, including L
e
-
gionella, Aeromonas spp., and Mycobacterium spp. Assessing risk from t
h
e
s
e
organisms in biofilms is complicated by the potential for two modes of transmis-
sion. Aeromonas spp. causes disease by ingestion, while the other two organ-
isms cause the most severe forms of disease after inhalation. In the case of
Aeromonas spp., which is included as one of the unregulated “contaminants” to
be tested for in the Contaminant Candidate List, it has been shown that drinking

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
WATER QUALITY INTEGRITY 223
water isolates carry virulence factors directly involved in pathogenesis (Sen and
Rogers, 2004).
Coliforms and Heterotrophs
Another consequence of biofilms is their potential to support the growth a
nd
release of organisms of regulatory concern, especially coliforms. Coliforms
released from biofilms may result in elevated coliform detection even though
physical integrity (i.e., breaches in the distribution system) and disinfectant re -
sidual have been maintained (Characklis, 1988; Haudidier et al., 1988; Smith et
al., 1990). It should be noted that coliforms arising from biofilms are generally
considered to be low risk (see Chapter 2), which is also inferred by EPA’s vari-
ance to the Total Coliform Rule for coliforms emanating from biofilms (see
page 208). However, coliform regrowth may indirectly present a risk by mask-
ing the presence of bacteria introduced in a simu ltaneous contamination event.
If repeated occurrences of coliforms in the distribution system force a utility to
notify the public, there can be a loss of consumer confidence and trust in
the utility.
The regrowth of heterotrophs in biofilms can also be of concern, especially
for European communities that are required to monitor their presence. S
o
m
e
U.S. utilities routinely monitor heterotrophs using heterotrophic plate count
s
(HPC) as a general indicator of microbial quality, and may be required to assess
their numbers if chlorine residuals are too low. In general, heterotrophic bacte-
ria are usually not of public health concern, but with the growing immunocom-
promised population many utilities are interested in minimizing the presence of
these organisms in their water.
Corrosion and Other Effects
In addition to the regrowth issue, biofilms in distribution systems can cause
other negative effects on finished water quality. The processes listed here do not
require that the organisms detach from the surfaces, since the changes in water
quality are due to their metabolic activities as they grow on the surfaces.
Bacterial biofilms may contribute to the corrosion of pipe surfaces and t
h
e
i
r
eventual deterioration. Although a considerable amount of corrosion internal t
othe
pipe can be mediated by abiotic factors, it is known that bacteria can bot
h
directly and indirectly influence corrosion of metal surfaces. Of particular con-
cern is the pitting of copper that can lead to pinhole leaks in premise plumbing.
Geesey et al. (1993) reported that pitting of copper plumbing in four hospitals
around the world was likely attributable to bacterial activity. Wagner et
al. (1997) have said that biologically produced polymers typical of biofilms
create high and low chloride concentration cells, and consequently localized
corrosion cells, leading to increased copper corrosion. Laboratory studies have
shown that

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
the presence of bacteria on copper surfaces could accelerate corrosion w
h
e
n
compared to an abiotic system (Webster et al., 2000). In other studies, specific
organisms were correlated with copper corrosion and could be isolated from pi
ts
(Bremer and Geesey, 1991; Bremer et al., 1992). However, other research h
a
s
shown that organisms alone did not cause copper pitting, and that particulate
matter was also required (Walker et al., 1998).
Microbes may also influence iron surfaces in distribution systems. I
r
o
n
bacteria can grow on ferrous metal surfaces (Ridgway et al., 1981), and by v
i
r
- tue
of their metabolism may modify the local chemistry at the metal s
u
r
f
a
c
e
which in
turn promotes localized corrosion (Victoreen, 1974). As stated b
y
McNeill and
Edwards (2001), there are many possible effects of bacterial action and biofilm
formation on iron corrosion. These include the production of differential
aeration cells (Lee et al., 1980), soluble metal uptake by biofilm polymers
(Tuovinen et al., 1980), changes in iron speciation by oxidation or reduction
(Shair, 1975; Denisov et al., 1981; Kovalenko et al., 1982; Okereke and Stevens,
1991; Chapelle and Lovely, 1992; Nemati and Webb, 1997), and the production
of pH gradients (Tuovinen et al., 1980) or corrosive hydrogen sulfide (Tuovinen
et al., 1980; DeAraujo-Jorge et al., 1992). All of these factors can contribute to
increased localized corrosion and the deterioration of the pipe material, as well
as influencing water quality by causing the release of metal ions or corrosion
products and associated problems with water color.
Other effects of biofilms are worth noting. As demonstrated in the waste-
water industry, it is possible to have nitrifying bacteria present in biofilms, and
these organisms could result in nitrification episodes in distribution systems
where chloramine is used (Wolfe et al., 1990, and see the section below). Actin-
omycetes or fungi present in biofilms may result in taste and odor problems
(Burman, 1965, 1973; Olson, 1982), which then lead to consumer complaints.
Excess biofilm growth can result in the loss of hydraulic capacity by increasing
fluid frictional resistance at the pipe wall (see examples in Characklis et
al., 1990). Finally, growth of biofilms and the associated organics can create a
chlorine demand at the pipe wall.
Biologically Stable Water
Because this report focuses on distribution system events, it does not delve
into failures or breaches at the treatment plant that might allow a breakthrough
of contaminated water. Nonetheless, a brief discussion of biologically stable
water is warranted, given its potential to reduce the growth of bacteria in the
distribution system. Drinking water is generally considered to be biologically
stable if it does not support the growth of bacteria in the distribution system. In
its broadest sense, biologically stable water restricts growth because it lacks an
essential nutrient (nitrogen or phosphorus), is sufficiently low in utilizable or-
ganic carbon, or contains adequate disinfectant. Although all of these parame -
ters may influence biofilm growth, the U.S. drinking water industry has typically

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
viewed biologically stable water as sufficiently low in organic carbon as to l
i
m
i
tthe
proliferation of heterotrophic bacteria. In this context, the general concepts of
microbial stable water and maximum regrowth potential are relatively well
understood (Rittman and Snoeyink, 1984; Sathasivan et al., 1997).
Another mechanism for ensuring biological stability is the maintenance ofan
adequate disinfectant residual. However, since disinfectants decay
in t
h
e
distribution system, reliance on a residual to ensure biological stability may
not be entirely feasible. Within distal portions of the distribution system or
within stagnant portions of premise plumbing, disinfectants disappear via
reactions with pipe or bulk water or via nitrification. At these locations, any
available organics can then be freely utilized by the bacteria present.
The reduction of organic carbon to control microbial growth may al
l
ow
utilities to decrease their reliance on disinfectants. This approach also has the
advantage of decreasing the potential for the production of disinfectant by-
products (DBPs). Organic carbon removal is most often accomplished through
enhanced coagulation, granular activated carbon filtration, or biological
filtration. Although there is controversy surrounding target concentrations of
organics that will limit regrowth, some recommendations have been made. van
der Kooij et al. (1989) and van der Kooij and Hijnen (1990) showed a
correlation between assimilable organic carbon (AOC) and regrowth in a non-
disinfected distribution system, and provided evidence for biological stability in
the Nether- lands when the AOC concentration (Pseudomonas fluorescens P17 +
Spirillum NOX) is reduced to 10 µg acetate C eq/L (van der Kooij 1992).
LeChevallier et al. (1991) have suggested that coliform regrowth may be
controlled by influent AOC levels (P17 + NOX) below 50 µg acetate C eq/L.
Based on a field study, LeChevallier et al. (1996) subsequently recommended a
level below 100 µg C/L to control regrowth. Servais et al. (1991) have
associated biological stability with a biodegradable dissolved organic carbon
(BDOC) level of 0.2 mg/L, but Joret et al. (1994) have stated that the value is
0.15 mg/L at 20º C and 0.30 mg/L at 15º C.
It should also be noted that organic carbon may not be the limiting nutrient.In
Japan and Finland, evidence supports the concept that phosphorus is limiting
(Miettinen et al., 1997; Sathasivan et al., 1997; Sathasivan and Ohgaki,
1999; Lehtola and Miettinen, 2001; Keinanen et al., 2002; Lehtola et al.,
2002a,b, 2004). In these cases, the addition of phosphate-based corrosion
inhibitors may decrease the biological stability of the water and allow for
regrowth (Miettinen et al., 1997).
This discussion illustrates that the best strategy for creating and maintaining
biologically stable water is most likely to be system specific. Each water utility
should identify the limiting nutrient and best practices to attain and then main-
tain biological stability. Changing water quality goals should then keep
these factors in mind. For example, the dosing of ammonia during a switch to
chloramination would relieve nitrogen limitations to regrowth, whereas dosing
of phosphate corrosion inhibitors can relieve phosphate limitations.

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
Nitrification
Biological nitrification is a process in which bacteria oxidize reduced n
i
t
r
o
-gen
compounds (e.g., ammonia) to nitrite and then nitrate. It is associated w
i
t
h
nitrifying bacteria in distribution systems and long retention times in water sup-
ply systems practicing chloramination. One of the most important problems
exacerbated by nitrification is loss of the chloramine disinfectant residual. This
occurs because a reduction in ammonia results in an increased ratio of chlorine
to ammonia nitrogen. This ratio controls the stability of monochloramine,
which is governed by a complex set of reactions (Jafvert and Valentine, 1992;
also see following section on loss of disinfectant residual). As the ratio ap-
proaches 1.5 on a molar basis, a rapid loss of monochloramine occurs attribut-
able to the eventual oxidation of N(III) to primarily nitrogen gas and the release
of more ammonia. The released ammonia can then be further oxidized by the
nitrifying organisms, establishing what amounts to a positive feedback loop.
Furthermore, the loss of disinfectant residual removes one of the controls on the
activity of nitrifiers, and it may also lead to the increased occurrence of micro -
organisms such as coliforms (Wolfe et al., 1988, 1990) and heterotrophic bacte-
ria.
As discussed in NRC (2005), the loss of chloramine residual is the m
ost
significant health threat that can result from nitrification. It should be noted,
however, that there are other lesser health effects of nitrification that may
be important for certain populations. Nitrite and nitrate have been shown to
cause methemoglobinemia (blue baby syndrome), an acute response to nitrite
that results in a blockage of oxygen transport (Bouchard et al., 1992).
Methemoglo- binemia affects primarily infants below six months of age, but
it may occur in adults of certain ethnic groups (Navajos, Eskimos) and those
suffering from a genetic deficiency of certain enzymes (Bitton, 1994). Pregnant
women may also be at a higher risk of methemoglobinemia than the general
population (Bouchard et al., 1992). A second concern is that nitrate may be
reduced to nitrite in the low pH environment of the stomach, reacting with
amines and amides to form N-nitroso compounds (Bouchard et al., 1992; De
Roos et al., 2003). Nitrosa- mines and nitrosamides have been linked to
different types of cancer, but the intake of nitrate from drinking water and its
causal relation to the risk of cancer is still a matter of debate (Bouchard et al.,
1992). A study by Gulis et al. (2002) in Slovakia related increased colorectal
cancer and non-Hodgkin’s lymphoma to medium (10.1–20 mg/l) and high
(20.1–50 mg/l) concentrations of nitrate nitrogen in drinking waters.
Similarly, Sandor et al. (2001) showed a correlation between the consumption
of waters containing greater than 88 mg/l nitrate nitrogen and gastric
cancer. Despite numerous papers (Sandor et al., 2001; Gulis et al., 2002; Kumar
et al., 2002; De Roos et al., 2003; Coss et al., 2004; Few- trell, 2004), the
concentration at which nitrate nitrogen in drinking waters presents a health risk
is unclear (Fewtrell, 2004). Finally, a lesser but still significant water quality
effect of nitrification is a reduction in alkalinity and pH in low

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
alkalinity waters. This may cause the pH to decrease to the point that c
o
r
r
o
s
i
o
nof lead
or copper becomes a problem.
It is important to recognize that nitrate and nitrite may come from sources
other than nitrification. van der Leeden et al. (1990) found that 93 percent of all
U.S. water supplies contain less than 5 mg/l nitrate, but noted that these values
may be changing as a result of the increased use of nitrate-containing fertilizers.
Increased use of chloramination (up to 50 percent of the surface water systems
in the United States may use chloramination in the near future as a result of the
Stage 1 Disinfectants/Disinfection Byproducts Rule; EPA, 2003) may result in
higher levels of nitrate in drinking waters (Bryant et al., 1992), but the increment
in nitrate plus nitrite nitrogen from this source would typically be less than 1
mg/L, which is well below the current maximum contaminant level (MCL).
Thus, as stated earlier the concern may be predominantly for more
susceptible populations (pregnant women, infants, some ethnic groups).
Interestingly, although nitrification is a recognized potential problem in w
a
-ter
systems practicing chloramination, nitrification control is required or encour-
aged in only 11 of 34 states that responded to a survey of drinking water pro -
grams conducted by the Association of State Drinking Water Administrators in
March 2003 (see Table 2-5). This illustrates the need for state agencies to rec-
ognize the potential issues associated with chloramination and nitrification, and
thereby prepare their utilities to deal with this potentially problematic issue.
Leaching
All materials in the water distribution system, including pipes, fittings, lin-
ings, other materials used in joining or sealing pipes, and internal coatings leach
substances into the water. The processes that account for this include corrosion,
dissolution, diffusion, and detachment. Taste and odor problems (Burlingame et
al., 1994; Khiari et al., 2002) are the most likely outcome of leaching because
most substances leaching into water from materials in the distribution systemare
non-toxic, present only at trace levels, or are in a form unlikely to cause health
problems.
There are however, a few situations in which leaching may present a sub-
stantial health risk. By far the most significant is the leaching of lead from lead
pipe, lead-containing solder, and lead service connections. Monitoring of lead in
tap water and replacement of these lines are important components of the
Lead and Copper Rule. Other materials used in distribution systems that have
the potential for leaching include PVC pipes manufactured before about
1977. These are known to leach carcinogenic vinyl chloride into water at levels
above the MCL (AWWA and EES, Inc., 2002). Cement materials have, under
unusual circumstances, leached aluminum into drinking water at
concentrations that caused death in hemodialysis and other susceptible
patients (Berend et al., 2001). Because levels of aluminum normally present
in drinking water can also threaten this population, the FDA has issued guidance
for water purification pre-

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
228 DRINKING WATER DISTRIBUTION SYSTEMS: ASSESSING AND REDUCINGRISKS
treatments in the U.S. for dialysis and other patients
(http://www.gewater. com/library/ tp/1111_Water_The.jsp). Asbestos fibers
may also be released from asbestos cement; the content of asbestos in water
is regulated with an MCL, although utilities are not required to monitor for
asbestos in the distribution system. Finally, excessive leaching of organic
substances from linings, joints, and sealing materials have occasionally been
noted. Some of these substances may support the growth of biofilms (Shoenen,
1986), such that their use should be limited.
For new materials, NSF International establishes levels of allowable
contaminant leaching through ANSI/NSF Standard 61 (see Chapter 2).
However, this standard, which establishes minimum health effect requirements
for chemical contaminants and impurities, does not establish performance,
taste and odor, or microbial growth support requirements for distribution system
components. This is unfortunate because research has shown that distribution
system components can significantly impact the microbial quality of drinking
water via leaching. Procedures are available to evaluate growth stimulation
potential of different materials (Bellen et al., 1993), but these tests are not
applied in the United States by ANSI/NSF.
Internal Corrosion
Internal corrosion manifests as (1) the destruction of metal pipe interiors b
y
both uniform and pitting corrosion (see Chapter 4) and (2) the buildup of s
c
a
l
e
sof
corrosion products on the internal pipe wall that hamper the flow of water
(see Chapter 5). A large number of water quality parameters such as disinfec-
tant residual, temperature, redox potential, alkalinity, calcium concentration,
total dissolved solids concentration, and pH play an important role both in the
internal corrosion of pipe materials and the subsequent release of iron. The
products of corrosion may appear in water as dissolved and particulate metals,
and the particles may cause aesthetic problems because of their color and turbid-
ity if they are present in sufficient concentration. Metals such as lead and cop-
per in tap water are governed by the Lead and Copper Rule; asbestos particles
and iron particles with adsorbed chemicals such as arsenic (Lytle et al., 2004)
are of concern because of possible health effects. The quality of distributed wa-
ter must be controlled so that both corrosion and metal release do not cause wa-
ter quality problems.
Scale Formation and Dissolution
Scale on pipe surfaces may form in distribution systems for a variety of r
e
a
-
sons including precipitation of residual aluminum coagulant after filtration, pre-
cipitation of corrosion products, precipitation of corrosion inhibitors, and
precipitation of calcium carbonate and silicate minerals. Scale that forms in a
thin,

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
smooth coat that protects the metal pipe by reducing the rate of corrosion is ge
n-
erally desirable, whereas uncontrolled precipitation can reduce the effective di-
ameters of distribution pipes and can create rough surfaces, both of which
reduce the hydraulic capacity of the system (as discussed in Chapter 5)
and increase the cost of distributing water.
In terms of internal contamination events, rough surfaces and scales w
i
t
h
reduced metals such as ferrous iron can increase problems with
bi
ofilm
s (Camper et al., 2003). That is, ferrous iron reacts with chlorine
and monochloramine, reducing the effective concentration of disinfectant in the
vicinity of biofilms. Furthermore, rough surfaces contain niches where
microbes can grow without exposure to hydraulic shear. If the scale
material is loosely attached to the pipe wall, such as some aluminum
precipitates, hydraulic surges can result in substantial increases in the turbidity
of tap water. Scales are also important because they can dissolve under some
water quality conditions and release metals to the water in the distribution
system. For example, Sarin et al. (2003, 2004) showed that iron scales release
iron during flow stagnation, which then causes turbid and colored water. Dodge
et al. (2002), Valentine and Stearns (1994), and Lytle et al. (2002) showed
that uranium, radium-226, and arsenic, respectively, could be adsorbed to iron
corrosion scales found in distribution systems. (In order for these metals to
accumulate they must be present in the source water.) Lytle et al. (2002) showed
that arsenic would accumulate on iron solids in distribution systems even when
present in water at concentrations less than 10 µg/L. Aluminum and manganese
solids can also adsorb metal contaminants and may subsequently release them
because of changes in water quality. Research is needed to fully characterize
this potential source of contamination related to internal corrosion and scale
dissolution and to find ways to control it.
Other Chemical Reactions that Occur as Water Ages
Many water distribution systems in the United States experience long reten-
tion times or increased water age, in part due to the need to satisfy fire fighting
requirements. Although not a specific degradative process, water age is a char-
acteristic that affects water quality because many deleterious effects are
time dependent. The most important for consideration here are (1) the loss of
disinfectant residuals and (2) the formation of DBPs. The importance of water
age is recognized in part by the survey of state drinking water programs where
nearly all states that responded to the survey either required or encouraged
utilities to minimize dead ends and to have proper flushing devices at remaining
dead ends (Table 2-3).

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
Loss of Disinfectant Residual
Maintenance of a disinfectant residual throughout a distribution system is
considered an important element in a multiple barrier strategy aimed at main -
taining the integrity of a distribution system. It is generally assumed that
the presence of a disinfectant is desirable because it may kill pathogenic
organisms, and therefore the lack of a disinfectant is an undesirable situation.
The absence of a disinfectant residual when one is expected may also indicate
that the integrity of the systemhas been compromised, possibly by intrusion or
nitrification. If the disinfectant is chloramine, its decay will produce free
ammonia that could promote the onset of nitrification. Understanding the
nature of the processes leading to disinfectant losses, especially when those
processes lead to excessive decay rates,is important in managing water quality.
Loss of disinfectants in distribution systems is typically due to reduction r
e
-
actions in the bulk water phase and at the pipe–water interface that reduce d
i
s
i
n
-
fectant concentration over time, although nitrification (in the case of
chloramine) can also play a role. Dissolved constituents that can act as reduc-
tants in the aqueous phase include natural organic matter (NOM) and ferrous
Fe(II) and manganous Mn(II) ions. These substances may occur in the water
either as a result of incomplete removal during treatment, from the corrosion of
pipe material (e.g., cast iron), or from the reduction of existing insoluble
iron and manganese deposits. Disinfectants may also readily react with
reduced forms of iron and manganese oxides typically found on the surface of
cast iron pipes as well as with adsorbed NOM (Tuovinen et al., 1980, 1984;
Sarin et al., 2001, 2004). Benjamin et al (1996) found that the accumulation of
iron corrosion products at the pipe wall and the release of these products into
the bulk water led to a deterioration of water quality. There have been several
reports that the loss of chlorine residuals in corroded unlined metallic pipes
(particularly cast iron) increases with increasing velocity (Powell, 1998; Powell
et al., 2000; Grayman et al., 2002; Doshi et al., 2003). Correlative evidence for
the role of corrosion in reducing disinfectant residuals was produced by
Camper et al. (2003), who studied the interactions between pipe materials,
organic carbon levels, and disinfectants using annular reactors with ductile–iron,
polyvinyl chloride (PVC), epoxy, and cement-lined coupons at four field
sites. They found that iron surfaces supported much higher bacterial populations
than other materials.
Modeling efforts to understand disinfectant decay have been primarily em
-
pirical in nature or semi-mechanistic, and they have mostly addressed n
o
n
-
biological reactions. The primary purpose of these types of models is to serve a
s
a
predictive tool in managing water quality. Most modeling research has tar-
geted the relatively fast reactions of free chlorine in the aqueous phase, predict-
ing free chlorine decay versus hydraulic residence time using single system-
specific decay coefficients. For example, Vasconcelos et al. (1996) developed
several simple empirical mathematical models to describe free chlorine decay.
Clark (1998) proposed a chlorine decay and TTHM formation model based on a

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
competitive reaction between free chlorine and NOM. The model was validated
against the Vasconcelos et al. (1996) data sets and found to be as good or better
(based on r2
values) than the models examined by Vasconcelos et al. (1996).
More sophisticated models have improved predictive management capabili-
ties and are also useful as research tools in the elucidation of fundamental proc-
esses. Rossman et al. (1994) developed a chlorine decay model that i
n
c
l
u
d
e
s
first-
order bulk phase and reaction-limited wall demand coefficients; this model is
incorporated into EPANET1
. The model developed by Clark (1998) was e
x
-tended
to include a rapid and slow reaction component and to study the effect o
f
variables
such as temperature and pH (Clark and Sivaganesan, 2001). Further
extensions included the formation of brominated byproducts (Clark et al., 2001).
McClellan et al. (2000) modeled the aqueous-phase loss of free chlorine due to
reactions with NOM by partitioning the NOM into reactive and non-
reactive fractions. Other models have incorporated reactions with reactive pipe
surfaces that may dominate the loss pathways (Lu et al., 1995; Vasconcelos et
al., 1997) as well as bulk phase reactions. Clark and Haught (2005) were able to
predict free chlorine loss in corroded, unlined metallic pipes subject to changes
in velocity by modeling the phenomena as being governed by mass transfer to
the pipe wall where the chlorine was rapidly reduced.
Less studied has been the loss of monochloramine in distribution
systems. Monochloramine, while generally less reactive than free chlorine, is
inherently unstable because it undergoes autodecomposition. While
autodecomposition occurs via a complex set of reactions, the net loss of
monochloramine occurs according to the stoichiometry:
3 NH2Cl → N2 + NH3 + 3 Cl−
+ 3 H+
(1)
This reaction has been reasonably well studied (Valentine et al., 1
9
9
8
;
Vikesland et al., 2000) and can be approximated (in the absence of other reac-
tions) by a simple second-orderrelationship:
1/[NH2Cl] - 1/[NH2Cl]o = kvcsc t (2)
where kvcsc is a rate constant describing the second order l
ossof
monochloramine (Valentine et al., 1998). Its derivation involves the simplifying
assumption that monochloramine decays by a mechanism involving the rate
limiting formation of dichloramine that then rapidly decays. As such, kvcsc is a
combination of several fundamental rate constants and the Cl/N ratio. It can be
simply calculated and used to predict monochloramine decay in the aqueous
phase in the absence of other demand reactions. It should be pointed out
that
1
EPANET is a model developed by EPA that performs an extended period simulation of h
y
d
r
a
u
l
i
c and water quality
behavior within pressurized pipe networks (see Chapter 7).

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
3
chloramine will decay more rapidly than predicted by this approach if
significant amounts of demand substances other than ammonia are present in
solution or if reactions with pipe walls are considered. Other demand
substances can include NOM and reduced metals in the aqueous phase, as well
as Fe(II) in pipe deposits.
Wilczak (2001) found that a sequential first-order empirical model best fi
tthe
East Bay Municipal Utility District’s chloramine decay data. Palacios a
n
d Smith
(2002) found that chloramine decay in San Francisco’s water was consi
s-tent with
a sequential first-order model, but that a simple first-order decay r
a
t
e could be
applied to the data due to the low organic matter concentrations. Duirk et al.
(2002) developed a comprehensive aqueous-phase chloramine reaction model
that accounts for both monochloramine autodecomposition as well as
reduction by NOM that is similar in structure to that proposed by McClellan et
al. (2000). Reaction of trace levels of free chlorine that equilibrate with
monochloramine was a key mechanism accounting for slow
monochloramine loss due to reaction with NOM. As a consequence, loss of
chloramine should decrease with increasing pH because both autodecomposition
and its reaction with NOM become slower. Table 6-1 summarizes the
mechanisms for loss of a chloramine residual.
Disinfection Byproduct Formation
Formation of DBPs in distribution systems is attributable to reactions
o
f
chemical disinfectants with NOM either in bulk solution or associated with pipe
deposits (Rossman et al., 2001). The importance of NOM associated with pipe
deposits is based largely on evidence from controlled lab studies and is open to
speculation,and must certainly be very system specific.
TABLE 6-1 Reactions that Reduce Chloramine Residual
Reaction Stoichiometry
Chloramine auto-decomposition 3 NH2Cl  N2 + NH4
+
+ 3 Cl-
+ 2 H+
Oxidation of organic m
a
t
t
e
r
by
chloramine
0.1 C5H7O2N + NH2Cl + 0.9 H2O 
0.4 CO2 + 0.1 HCO -
+ 1.1 NH +
+ Cl-
3 4
Reaction of chloramine w ith c
o
r
r
o
sion
products at pipe w all
0.5 NH2Cl + H+
+ Fe2+

Fe3+
+ 0.5 NH4+
+ 0.5 Cl -
Oxidation of nitrite by chloramine NH2Cl + NO -
+ H O 
2 2
NH3 + NO -
+ HCl
SOURCE: Reprinted, with permission, f rom Woolschlager et al. (2001). © 2001 by IWA Publishing.

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
Most studies have focused on formation of halogenated DBPs produce
d
from reactions of NOM with free chlorine, especially those DBPs that are cur-
rently regulated by the EPA—trihalomethanes (THMs) and haloacetic acids
(HAAs). However, over 600 potentially harmful DBPs have been identified
(Richardson, 1998) including both chlorinated and brominated compounds.
Brominated compounds arise from the oxidation of bromide which can be
an important factor in determining DBP speciation even when found at the
sub- milligram per liter level. Many of the DBPs formed in chloraminated
systems are the same as those observed in systems practicing chlorination
(Figure 6-1). This may be a consequence of similar formation mechanisms
involving free chlorine or attributable to the practice of prechlorination prior to
ammonia addition and subsequent chloramine formation. However, the rates of
formation of most DBPs are much slower in chloraminated systems, resulting in
the reduced formation of many DBPs, especially THMs.
Table 6-2 lists some of the DBPs rated as high priority and observed in a r
e
-
cent comprehensive survey of 12 full-scale treatment plants in the United S
t
a
t
e
sin
2000. The halogenated DBPs detected in this study have included mono-, di-,
tri-, and/or tetra- species of halomethanes (HMs) (including iodinated species);
haloacetonitriles (HANs); haloketones (HKs); haloacetaldehydes (HAs); and
halonitromethanes (HNMs). The presence of bromide resulted in a shift in
speciation for the trihalomethanes (THMs) and haloacetic acids (HAAs). Bro-
minated DBPs for other classes of DBPs (HANs, HKs, HAs, HNMs) were also
detected. Chloramination formed certain dihalogen-substituted DBPs (HAAs,
HAs) preferentially over related trihalogenated species. In addition, chlorine
dioxide produced dihalogenated HAAs (Richardson et al., 2004). Recently sev-
eral DBPs have been identified as unique to chloraminated systems. These in-
clude N-nitrosodimethylamine (NDMA) (Choi and Valentine, 2002 a,b, Mitch
and Sedlak, 2002), cyanogen chloride, and several iodohaloacetic acids, none of
which are currently regulated at the federal level. The state of California has,
however, established a notification level of 10 ppb in drinking water for NDMA,
a potent carcinogen (http://www.dhs.ca.gov/ps/ddwem/chemicals/NDMA/
NDMAindex.htm). NDMA seems to be a relatively widespread DBP and may
become more prevalent as the use of chloramination increases.
Given the relatively high reactivity of free chlorine with NOM, it is not sur-
prising that a significant amount of DBPs is formed in the water treatment plant
as the result of primary disinfection (i.e., disinfection at the treatment plant to
meet CT requirements). DBP formation, however, continues in the distribution
system, as shown in Figure 6-2. Based on an evaluation of data from utilities
that participated in the Information Collection Rule (ICR) and that use surface
water as their source, TTHMs increased through distribution systems on average
about 50 percent when chlorine was used to maintain the distribution
system residual (McGuire and Graziano, 2002). Similar results were
obtained for chloraminated distribution systems, mainly because these systems
had water with higher TTHM precursors than those utilities that were using free
chlorine. Chloramine-specific DBPs (like N-nitrosodimethylamine
cyanogen chloride,

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
Chlorination
Chloramination
FIGURE 6-1 Comparison of Halogenated DBPs. SOURCE: Richardson et al. (2004).
HANs, 0.2%
Chloropicrin, 0
.
5%
HKs, 0.2%
TTHM, 35.6%
Unaccounted f or
TOX , 51.5%
THAA, 11.9%
HANs, 1.0%
TTHM, 3.9%
THAA, 10.8%
Chloropicrin, 0
.
7%
HKs, 0.6%
Unaccounted forTOX,
82.9%

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
TABLE 6-2 Chlorination and Chloramination Disinfection Byproducts
Chloroform Dichloroacetonitrile Acetaldehyde
THMs
Bromodichlorom ethane Bromochloroacetonitrile Propionaldehyde -CHCl3
Chlorodibrom om ethane Tribromoacetic a
c
i
d Butyraldehyde
HAAs
-CHCl2COOH
Bromoform Trichloroacetonitrile Valeraldehyde
HANs
-CHCl2CN
Chloroacetic acid Dichloroacetonitrile Hexanal Chloropicrin
(halonitro-
Bromoacetic acid Bromochloroacetonitrile Heptanal
methanes)
-CCl3NO2
Dichloroacetic a
c
i
d Dibromoacetonitrile Octyl aldehyde Cyanogen
Halides
Trichloroacetic a
c
i
d Dichloropropanone Benzaldehyde
-CNCl
Bromochloroacetic acid Trichloropropanone Nonyl aldehyde
Bromodichloroacetic Chloropicrin (
c
h
l
o
r
o
n
i
- Decrylaldehyde
acid tromethane) Formaldehyde
Dibromoacetic acid TOX, TOCl, TOBr (as Glyoxal
Cl)
Chlorodibrom oacetic Cyanogen Chloride Methylglyoxal
acid
Tribromoacetic a
c
i
d Dichloroaldehyde Acetaldehyde
Trichloroacetonitrile Propionaldehyde
Note: this table is not comprehensiv e, as new DBPs are discov ered on a regular basis.
SOURCE: Reprinted, with permission, f rom Richardson and Krasner (2003). © 2003 by Elsev ier Ltd.
iodohaloacetic acids) are expected to form primarily in distribution system
s since
chloramine is not usually used during primary disinfection (although am
-monia is
sometimes added in the treatment plant to stop THM and HAA form
a- tion).
Finally, haloacetic acid levels are also expected to increase in the distribution
system, but not to the same degree as THMs.
It should be noted that processes may occur in distribution systems that
cause a loss of DBPs. For example Baribeau et. al. (2006) showed that the for-
mation of several haloacetic acids did not increase with water age in a chlorin -
ated distribution system (Figure 6-3). Speight and Singer (2005) correlated
HAA reduction to a reduction in chlorine and suggested that the observed HAA
loss was due to biodegradation that was otherwise inhibited in the presence of

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
200
180
160
140
120
100
80
60
40
20 0 3.5 5.2 5.5 8.2 8.4 18.0 323
Estimated average water age (hour)
FIGURE 6-2 Changes in total trihalomethanes in a system w ith free c
h
l
o
r
i
n
e w ith water age.
SOURCE: Reprinted, w ith permission, from Baribeau et al. (2006). © 2006 by Ameri- can
Water Works Association.
60
50
40
30
20
10
0
POE 3.5 5.2 5.5 8.2 8.4 18.0 323
Estimated average water age (hour)
FIGURE 6-3 HAA concentrations in a chlorinated system as a f
u
n
c
t
i
o
nof w ater age. SOURCE:
Reprinted, w ith permission, from Baribeau et al. (2006). © 2006 by American Water Works
Association.
Chloroform Bromodichloromethane Dibromochloromethane Bromoform
Mono-haloacetic acids Di-haloacetic acids Tri-haloacetic acids
HAA
(mg/L)
THM4
(

g/L)

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
chlorine. However, interpreting field data can be difficult, and changes in D
B
P
concentrations may alternatively be attributable to changes in treatment pl
ant
operation (Pereira et al., 2004). The nature of DBP decay processes are not well
established in actual distribution systems, but laboratory studies suggest that
these may include biodegradation (Baribeau et al., 2005a), hydrolysis (Zhang
and Minear, 2002), and reduction by reduced forms of iron (Chun et al., 2005;
Zhang et al., 2004).
DBP modeling efforts can be categorized as empirical and s
e
m
i
-mechanistic.
Motivation for modeling includes estimating the extent of t
he
problem from easily
measured parameters, predicting the influence of treatment practices aimed at
reducing DBPs, and as a tool in establishing fundamental mechanisms. Amy
et al. (1987) correlated DBP formation to a number of important variables that
include chlorine dosages, DOC and bromide concentrations, temperature,
and contact time. Harrington et al. (1992) used a similar approach to develop
an empirical model to simulate THM and HAA formation during water
treatment. More recently, semi-mechanistic kinetic models have been
developed that couple disinfectant loss to the formation of selected DBPs.
McClellan et al. (2000) proposed a model for the formation of THMs in chlorin-
ated water assuming a fixed number of chlorine-consuming and THM-forming
sites per mg C in the aqueous phase. Duirk and Valentine (2002) used a similar
approach to model dichloroacetic acid formation from the reaction of
monochloramine with NOM. As already stated, no efforts have as yet included
the specific role of NOM on deposit/pipe surfaces that may be required to ade-
quately model DBP formation in distribution systems. In spite of the limita-
tions, considerable progress has been made in using models to explain observa-
tions and make simple predictions about the influence of treatment practices and
distribution systemresidence time. Continued effort is needed to refine these
models by including unifying principles that are not system specific and an im-
proved description of all pertinent phenomena.
It should be pointed out that while chlorine demand may be a useful meas-
urement to correlate with DBP formation in the bulk water (Gang et al., 2002),
this would not be the case if the demand were governed by inorganic constitu-
ents or by reactions with deposit materials (corrosion products, etc.).
DETECTING LOSS OF WATER QUALITY INTEGRITY
Distinguishing the loss of water quality integrity due to internal changes a
s
opposed to changes brought about by external events is very difficult b
e
c
a
u
s
e
there are few parameters that can be conclusively linked to internal contamina-
tion. Routinely monitored parameters such as temperature, pH, disinfectant re -
sidual, and even microbial constituents cannot differentiate between external and
internal contamination. Other less routinely monitored constituents, including
dissolved metals, turbidity, total organic carbon, synthetic organic compounds,
or nuisance organisms such as invertebrates may also not be definitive for exter-

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
nal vs. internal sources. However, if utilities have a good understanding of their
distribution system and its points of vulnerability, value judgments as to the po-
tential source of the detected contaminants can be made. With these limitations
in mind, the following sections on detecting physical, chemical, and biological
changes are presented. When there is a clear distinction between the ability of
the monitoring method to distinguish between internal and external contamina-
tion, these methods are thereby identified.
Detection of Physical and Chemical Changes
Taste and Odor
Tastes and odors detectable by the consumer are a common indication
of a
loss of water quality integrity (McGuire, 1995). In fact consumers may
only complain about the loss of water quality if they detect taste and odors
(Watson, 2004). Fortunately, methods exist to directly evaluate the flavor and
odor of tap water (Krasner et al., 1985; Dietrich et al., 2004; APHA, 2005) and a
guide exists to determine possible sources within the distribution system and
customers’ premises (McGuire et al., 2004).
Because most drinking waters in the United States contain a total c
h
l
o
r
i
n
e
residual, the taste and odor of tap water might be described as “chl
ori
nous.”
Whether this is noticeable to the water-consuming public depends on the chlo-
rine species present, the concentration of the residual, and the temperature of the
tap water. Other causative agents of tastes and odors in drinking water are usu-
ally metals, volatile organic chemicals, and microbial activity, with the
latter being the most prevalent (APHA, 2005). A very common cause is open
storage reservoirs where algae have been allowed to grow within the water as
well as along the sides of the basins. These algae can produce earthy, musty,
grassy, fishy, decaying vegetation and similar odors. Watson and Ridal (2004)
credited taste and odors to periphyton and more specifically to certain
cyanobacteria present in biofilms, as well as to the presence of dreissenid
mussels in the Great Lake region. Skjevrak et al. (2004) detected the presence
of ectocarpene, dicty- ipterenes, beta-ionone, menthol, menthone, and other
VOCs in biofilms within distribution systems. Furthermore, bacteria such as
actinomycetes can give rise to geosmin, and other microorganisms such as
certain fungi have been associated with consumer complaints about taste and
odor.
Much of the biological activity that causes taste and odor problems is i
ndi-
rect. Taste and odors problems may arise as a result of bacterial
processes i
n
certain types of pipes, such as iron, copper, and lead (Geldreich,
1996). In water systems with chlorophenols or bromophenols, biological activity
(particularly fungal) can convert these compounds to very odorous
chloro/bromo-anisoles that have much lower thresholds of odor detection than
the original compounds (Bruchet, 1999; Montiel et al., 1999). It is also possible
that other chlorinated

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
and oxidant-derived byproducts can be produced or allowed to increase in t
he
distribution system to the point where they begin to be detectable by customers.
Finally, if a source water contributes sulfur or iron to the distribution system
(such as from a groundwater supply), biological activity in the distribution sys -
tem can produce compounds that change the taste of the water.
Tastes and odors may be associated with external contamination events,
such as permeation and intrusion. Among the compounds most likely to present
a taste and odor problem stemming from an external contamination event
are gasoline additives or constituents, soluble components of soil, and
compounds found in sewage.
Changes in taste and odor can occur anywhere in the distribution sy
st
e
m
that the chlorine residual deteriorates and the water becomes stagnant, such as in
storage tanks, at dead-end water mains, and behind closed valves. Also in stag-
nant areas of the distribution system where corrosion conditions release iron into
the water, the iron may be detected by customers both visually and by taste.
Interestingly, most nuisance tastes and odors that cause customer
complaints originate within customers’ premises (except for those that come
from source water such as geosmin, 2-methylisoborneol, and certain chemical
spills) (Suffet et al., 1995; Khiari et al., 2002). Common causes are stagnant
plumbing (musty odors from biological growth), backflow events (various
types of chemical odors), hot water heater odors (hydrogen sulfide from
biological activity in hot water tanks), and corrosion of plumbing materials
(release of copper, zinc, and iron). New plastic pipe can leach odors for a
period oftime.
Within the main distribution system, new pipe and facilities need to be
checked for their contributions to potential off-odors before they are released for
use. Ductile iron pipe that is lined with cement-mortar might have an asphaltic
coating that can leach volatile organic chemicals into the water if it has not
cured sufficiently. New pipe joint lubricant can also impart aldehyde-type odors
to the water. New linings of storage tanks also need to be cured adequately be-
fore being placed into service.
Finally, the stability of the chlorine or chloramine residual is important t
o
controlling undesirable tastes and odors. Blending of source waters or boosting
of disinfectants that is not well controlled can produce dichloramine (which is
more odorous than monochloramine). It has also been shown that when a sys -
tem uses chlorine dioxide as a primary oxidant, chlorite in the distribution sys -
tem can react with free chlorine to reform chlorine dioxide that (1) can give a
strong chlorinous odor at the tap and (2) can be released into the air of a home
and react with volatile organic chemicals (such as fromnew carpet or paneling)
to create cat urine or kerosene type odors (Dietrich and Hoehn, 1991).
Colorand Turbid Water
Colored and turbid water at the tap is a strong indication that corrosion o
f
iron, iron release from scales, and post precipitation of aluminum salts are n
o
t

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
being controlled. The presence of colored or turbid water is therefore indicative
of changes in quality due to internal contamination. Iron released from the pipe
wall as Fe2+
will diffuse into the bulk water where it is oxidized by oxygen or
disinfectant, and then precipitates as ferric oxyhydroxide particles that cause
color and turbidity (Lytle and Snoeyink, 2002; Sarin et al., 2003). The effect of
this process may be made worse if the particles settle during periods of low
flow and are then resuspended by hydraulic surges. Post precipitation of
aluminum may result in particles that are loosely attached to the pipe wall
and are suspended during hydraulic surges. This type of aluminum
precipitate can be the cause of turbid or dirty water.
Dissolved and Particulate Metal Concentrations
If water leaving the treatment plant has metal concentrations that meet regu-
latory requirements, and if these levels increase in transit through the distribu-
tion system (typically iron) or in premise plumbing (lead, copper), then it may
be assumed that leaching and internal corrosion are occurring. Although it is
beyond the scope of this report to discuss the details of the Lead and Copper
Rule, this is the only example of a regulation that specifically addresses the in-
ternal degradation of water quality in a distribution system. In the case of iron,
elevated levels are more likely to be associated with secondary standards
and aesthetic concerns rather than with a specific public health threat.
Disinfectant Residual and Disinfection Byproduct Measurements
Measurements of disinfectant residuals and DBP concentrations often a
c
-
company one another and are routinely practiced using a number of standard
analytical methods. Sudden temporal increases in disinfectant loss indicate
a sudden change in water quality or system characteristics. For example,
this might be due to significant input of a reactive contaminant due to a
cross- connection or rapid onset of nitrification in the system. Unexpected
spatial losses point to problems associated with specific elements of the
distribution system such as a zone where internal corrosion is excessive or
where nitrification is occurring.
Identification of the causes of excessive disinfectant loss and DBP f
o
r
m
a
-tion
involves a combination of bench studies and field observations. The sig-
nificance of what is considered excessive loss must be gauged against what is
considered “normal” or not excessive. Free chlorine is stable for many days in
water containing no reactive constituents such as NOM. The applied dose
is then a bench mark for comparison. Simulated Distribution System (SDS)
jar testing using water obtained from the point of entry into a distribution system
or at other points can be used to determine the rate of disinfectant loss
attributable to bulk phase reactions as well as DBP formation. This requires
only measure-

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
ments of the free chlorine and DBP concentrations as a function of contact t
i
m
e.
The rate of chlorine loss and DBP formation in the SDS can then be compared
with losses observed in the systembetween points of known hydraulic residence
time (acknowledging that residence time at a given point is actually a distribu-
tion of values and not easy to determine—see Chapter 5). Differences between
the lab and field must be attributable to processes occurring inside the distribu-
tion system, most likely at the pipe–water interface.
Determining if chloramine loss is “excessive” is more complicated. T
h
e
SDS jar test will be indicative of how fast the chloramine disappears in the bul
k
phase but will not by itself reveal the mechanism if the loss is unexpectedly
high. This must also be compared to the rates of loss from autodecomposition,
which can be predicted using the second order relationship previously discussed
or perhaps measured in “distilled water” at the same pH as the system of inter-
est. If the bulk water reactions are much higher than expected in clean water or
than predicted, then one can presume the presence of significant amounts of
reactive substances such as ferrous iron or NOM. The loss rate in the bulk phase
can then be compared to values determined by measuring chloramine concentra-
tions in the system at points of known hydraulic residence times. If the rate of
loss determined by the field measurements are much higher than the bulk
loss rates, then presumably this is due to reactions at the pipe–water interface.
These include biological nitrification and reaction with reduced iron.
Indicatorsof Nitrification
Smith (2006) recently summarized important parameters (Table 6-3) t
ha
t
can be used as indicators of biological nitrification—one phenomenon that
is directly associated with internal changes in water quality. The most important
indicators (after loss of residual) are formation of nitrite and nitrate, loss of am-
monia, and a decrease in pH. An increase in heterotrophic plate count may also
TABLE 6-3 Usefulness of Water Quality Parameters for Distribution SystemNitrification
Monitoring
Parameter/Usefulness
Very Useful Useful Limited Usefulness
Total Chlorine
Nitrite-N
Free ammonia-N
Temperature
Free Chlorine*
Nitrate
Total a
m
m
o
n
i
a
-
N
HPC-R2A
pH
Dissolved O
x
y
g
e
n
TOC
Hardness
Alkalinity
AOB**
* Very usef ul during breakpoint chlorination (not f or routine monitoring)
** AOB=Ammonia oxidizing bacteria. Of limited usefulnessuntil rapid,
inexpensiv e enumeration methods become av ailable.
SOURCE: Reprinted, with permission, f rom Smith (2006). © 2006 by American Water W
o
r
k
sAssocia- tion.

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
indicate the growth of nitrifying organisms in the system. However, since bi
o-
logical nitrification can be associated with biofilms (Regan et al., 2003), an ab-
sence of nitrifying organisms in the bulk phase does not necessarily indicate the
absence ofnitrification.
Detection of Biological Changes
Several biological constituents, some of which are part of compliance moni-
toring, can be used to detect the loss of water quality integrity due to both inter-
nal and external contamination events. The applicability of each group of or -
ganisms for assessing internal changes in water quality is described in this sec-
tion.
Heterotrophic Plate Counts
Since the end of the 19th
century, heterotrophic plate counts (HPC) h
a
v
e
been used as an indicator of the proper functioning of treatment processes (Bar-
tram et al., 2003). By extension, HPC have also been used as an index of water
quality and safety in the distribution system, and the method continues to be
used in many countries as an index of regrowth (an internal event) of microor-
ganisms within the distribution system. Although it is difficult to establish the
exact contribution of suspended bacteria vs. proliferation and release of biofilm
cells if increases in HPC are observed, there is evidence that biofilm growth and
detachment can be the source of elevated bacterial numbers. Published accounts
by van der Wende et al. (1989) and LeChevallier et al. (1990) demonstrated that
elevated bacterial counts in water could not be attributed to replication of sus -
pended cells, but rather was due to biofilm growth on pipe surfaces. Accord-
ingly, the Surface Water Treatment Rule allows HPC levels to be used as a sur-
rogate for a “detectable” residual for regulatory compliance purposes, provided
that HPC is less than or equal to 500 colony forming units (CFU)/ml. Although
other countries do not set specific numerical limits for HPC (Robertson and
Brooks, 2003), the European Union has a recommendation of 100 CFU/ml. In
addition, the World Health Organization is presently debating whether or not
HPC counts should be included in their regulations on water quality.
Linking changes in HPC with a meaningful water quality variable can b
e
very
difficult. Many conditions, such as an increase or decrease in the organic
carbon concentration, stagnation, loss of disinfectant residual, and/or
nitrification, will result in an increase in HPC. Another cause of observed rises
in HPC might be that chorine-injured organisms regain culturability, even
though their actual numbers have not changed. It is also possible that increased
HPC is not due to bacterial growth in the distribution system but may originate
froman external contamination event. In a study where HPC was analyzed in
a distribution system with groundwater as its source, the concentrations of
HPC varied

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
with distance in the distribution system(Pepper et al., 2004), suggesting t
h
a
t
most of
the HPC bacteria originated from the distribution system rather than from the
source water, although it could not be determined with certainty. Intru- sion
could account for increased HPC, either by fomenting bacterial regrowth as a
result of nutrient intrusion, or simply by increasing HPC as a result of bacterial
intrusion. Therefore, differentiating between possible causes for changes in
HPC cannot be done without thorough monitoring of the system over a long
period of time, and without having a thorough knowledge of the microbial di-
versity, the physiology, and the ecology of the microbiota being detected
(Szewsyk et al., 2000).
A further complication is that the concentrations of HPC bacteria in a w
a
t
e
r
sample are dependent on the media being used for their enumeration. R2A a
ga
rhas
been shown to give the highest numbers; however, the importance of t
h
i
s with
regards to the use of different media vis-à-vis the detection of anomalies in the
system is yet to be determined. Because of the variety of incubation tem-
peratures and media used,it is difficult to compare within or between systems.
Regardless of the detection method employed, it is possible to use H
P
C
bacteria as a general indicator of distribution system hygiene and
performance. This approach requires that samples be taken at regular spatial
intervals along the distribution system at time points that reflect the hydraulic
residence time of the water in that section of the pipe. If increased bacterial
numbers are seen in the same “packet” of water in a plug flow system, it is
evidence that deterioration in water quality has occurred. With reasonable
forensic investigation, the utility can then determine if the increased counts are
due to internal vs. external events. This type of system monitoring is already
performed by industries (other than water supply) that rely on high-quality water
for manufacturing purposes.
Coliforms
Under ideal circumstances, the presence of coliforms in a drinking w
a
t
e
r
sample should indicate external fecal contamination of the water supply, w
h
i
c
his the
main premise behind the current Total Coliform Rule. Although this concept
has served the industry reasonably well, it is not without flaws. Methods may
not be sufficiently sensitive for detection, and sample collection may give
false positive (e.g., contaminated faucet screens) or false negative (disinfectant
residual not neutralized) results. Studies of coliform presence in distribution
systems indicate that coliforms may be introduced via treatment breakthrough as
well as by intrusion events, main breaks, and other external contamination
events (Besner et al., 2002). Furthermore, on occasion, coliforms have
been shown to multiply in biofilms, contributing to their detection in drinking
waters (LeChevallier et al., 1996). These same authors indicated that there was a
correlation between coliform occurrence and variables such as temperature,
AOC levels, and disinfectant type being used. Therefore, it is often not
possible to

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
determine if a coliform-positive sample is the result of an external
contamination event vs. regrowth of the organism within a biofilm. This
problem is acknowledged in the Total Coliform Rule, in which the EPA allows
for a variance from the regulation (40CFR, Code of Federal Regulations,
1993) “for systems that demonstrate to the State that the violation of the total
coliform MCL is due to a persistent growth of total coliforms in the distribution
system rather than fecal or pathogenic contamination, a treatment lapse or
deficiency, or a problem in the operation or maintenance of the distribution
system.”
Although the term coliform is used, it should not be forgotten that the group
includes several different genera (see Table 3-2) which may survive/regrow dif-
ferently under different conditions. When E. coli is found in drinking water, it is
generally believed to be associated with an external contamination event (see
references in Tallon et al., 2005) and linked to fecal contamination rather than
an internal/biofilm source. Consequently, E. coli is one monitoring tool that is
used to distinguish between internal and external contamination. However, it
should be noted that E. coli is less resistant to disinfectants than some other
pathogenic bacteria, viruses, and the protozoan cysts/oocysts. Thus, its absence
does not indicate an absence of pathogens.
Other Indicators of Fecal Contamination
There is a great deal of interest in identifying alternative indicators for w
a
-
terborne pathogens, as evidenced by the recent publication of the NRC r
e
po
r
t
Indicators for Waterborne Pathogens that summarizes the most recent insights
on the topic (NRC, 2004). An overview of some of these organisms (Clostrid-
ium perfringens, Enterococci and fecal streptococci, Bacteroides spp., Bacillus
subtilis spores, Pseudomonas spp., Aeromonas spp., Staphylococci, HPC bacte-
ria, hydrogen sulfide producers, and bacteriophages) is given in Table III of Tal-
lon et al. (2005). As noted in the table, most of these organisms are not entirely
specific to fecal contamination and/or suffer from difficulties in detection. As a
case in point, the spores of Clostridium perfringens have been proposed as an
indicator of fecal contamination of water and have been used as surrogates for
assessing the efficacy of water treatment processes designed to remove viruses
and the cysts/oocysts of Giardia and Cryptosporidium spp. (Payment and
Franco, 1993; Venczel et al., 1997). Because the spores are more resistant to
disinfection and the environment, their responses to these stresses are less pro-
nounced than vegetative bacteria. This indicator is not always specific for fecal
contamination, however, because it can be found in soils and sediments as part
of the natural flora. Nonetheless, to date it has not been identified as a part of the
natural flora of drinking water distribution systems, and as such, may be a
reasonable indicator of external contamination regardless of whether it
arises from fecal contamination or the soil.

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
Direct Detection of Pathogens
It would be optimal to directly measure pathogenic microorganisms
o
r
molecules specific for them instead of relying on indicator organisms. At the
present time, however, this approach is not realistic for several reasons. There is
a wide diversity of potential pathogens, ranging from viruses to bacteria to fungi
to protozoa, and each group of organisms represents a unique challenge for de-
tection. For many viruses and protozoa, there are no appropriate lab-based cul-
turing methods. If culturing methods are possible, selective media can reduce
the chances for recovering stressed organisms. Organisms can be present in
such low numbers that direct sampling is not sufficiently sensitive, and concen-
tration methods are also prone to error. With the advent of improved molecular
methods in the future, some of these limitations may be overcome. A review of
the issues associated with implementing these novel methods for pathogen de-
tection as well as problems associated with conventional approaches for assess -
ing microbial water quality have been published previously (NRC, 2004). This
report also points out that even if these new methods show promise, standardiza-
tion and validation are critical if the methods are to be used in a regulatory con-
text.
Summary
Water quality integrity needs to be evaluated rapidly and, if at all possible,
using in-line, real time methods (as discussed in Chapter 7). Unfortunately, cur-
rent microbial detection methods do not lend themselves to this approach,given
the many types of microbes possible,the limitations of individual indicators, and
the rapidity and sensitivity of certain methods. HPC can be a useful parameter,
but only if frequent monitoring is carried out and only if anomalous levels are
detected (Robertson and Brooks, 2003). The levels that have been proposed as
action levels or guidelines by various industries are too site-, season-,and
method-specific to be generally applied. In any case, HPC counts do not lend
themselves to on-line monitoring because it may take up to seven days for re-
sults,depending on the media being used. The concentration of HPC within the
distribution systemmay be an indirect measure of AOC and BDOC in the water
as organic carbon may be a reason for HPC regrowth (Robertson and Brooks,
2003). It may be tempting to suggest that AOC and/or BDOC be measured in
place of HPC, but these assays are not as easily applied by most utilities
and often take longer than the incubation period required for the HPC
measurements.
Because many of the organisms present within distribution systems c
a
n
n
o
t be
detected using the plate count methods typical for HPC (Block, 1992; Le -
clerc, 2003), efforts have been made to use rapid and relatively inexpensive mi-
croscopic techniques to visualize all microbes present employing fluorescent
DNA stains, such as acridine orange direct counts (AODC), 4'-6-diamidino-2-
phenylindole (DAPI), SYBER® Green, propidium iodide, etc.). Problems asso-

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
ciated with these methods are that both dead and live cells are detected, there i
s
interference from inorganic constituents, background autofluorescence can oc-
cur, and the level of detection may not be sufficiently low. Other dyes
have been used by some investigators (McFeters et al., 1999), but their
usefulness for routine monitoring remains to be seen. It is also possible to
use specialized equipment such as the ChemScan RDI/ Scan RDI™ or flow
cytometers to quantify fluorescently stained organisms. In these cases, the
equipment is expensive, and for flow cytometry, extensive optimization may be
required.
Indirect methods could be used to determine water quality integrity; for e
x
-
ample adenosine triphosphate (ATP) methods have been proposed, but they a
r
e
rather expensive and do not lend themselves to routine monitoring. Although
hand-held ATP photometers have been developed and are currently being used
in the food industry, the usefulness of these methods in the water industry
remains in question, given their low level of sensitivity and the relatively
dilute nature of finishedwater.
Molecular methods for the detection of microbes are still far from
routine.Although promising, PCR-based methods employing specific primers for
a suite of targeted organisms or for 16S ribosomal DNA may suffer fromthe
same problems as HPC and coliform counts. New analytical methods have been
used for early detection of chemical agents in water, and many approaches are
being developed in response to the need to detect chemical bioterrorism and
warfare agents. Calles et al. (2005) describe photoionization and quadrupole ion
trap, time-of-flight mass spectrometry as a means of detection of certain
hazardous compounds in a fast and sensitive manner. Similar approaches for
biological contaminants and indicator organisms are possible, but considerably
more research and development will be needed to ensure that the methods are
reliable. Even with the possibility for increased sensitivity and results that can
be obtained more quickly, the overall limitations on the use of coliform
bacteria to signify fecal contamination still exist.
Indeed, it is improbable that one indicator (such as coliforms) will serve a
l
l
needs because of the varied sources of contamination, the different types of or-
ganisms involved, the impact these events have on public health, and the chang-
ing regulatory climate. NRC (2004) proposes a tiered approach for microbial
monitoring. The first level is routine monitoring of common indicators to pro-
vide an early warning of a health risk or a change from background conditions
that could pose a health risk. These methods should be rapid, reasonably inex-
pensive, and low in cost. If a potential problem is identified at this level, it
should be followed by more detailed studies to assess the extent of public health
risk. This might involve expanded sampling and using a more tailored detection
method for indicators or even direct measurement of pathogens with molecular
methods. The third level is a detailed investigation of the source of contamina-
tion so that it can be ameliorated. The need for standardized methods decreases
as the investigation moves through the three phases, such that at the third level,
specialized research tools may berequired.

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
MAINTAINING WATER QUALITY INTEGRITY
Maintenance of water quality in the distribution system requires diligent a
t
-
tention by treatment plant personnel and those in charge of the distribution sys-
tem. A delicate balance must be achieved in order to comply with relevant regu-
lations while taking into account the detrimental effects that may occur as water
travels through miles of pipes to the consumer. As regulations become increas -
ingly more complicated and stringent, it is more difficult for utilities to balance
the requirements. For example, there may be a need to bolster disinfectant re-
siduals at various points throughout the distribution system, but this may lead to
unacceptable levels of DBPs. The issue is even more complicated in
premise plumbing where long periods of stagnation ultimately influence the
water quality that the consumer receives. The following section describes
methods and processes for maintaining the water quality integrity of potable
water. The committee believes that nitrification control is best accomplished by
maintaining an adequate disinfection residual and by booster disinfection, both
of which are discussed below. The reader is referred to AWWA (2006) for
further details on nitrification control.
Adequate Disinfection Residual
Maintenance of a disinfectant residual in the distribution system is required
under the Surface Water Treatment Rule and has been designated as the
b
e
s
t
available technology for compliance with the Total Coliform Rule. The practice
of carrying a disinfectant residual through the distribution system is also integral
to the control of biofilms. This residual is intended to act as a prophylactic in the
event of intrusion or backflow of a contaminant within the distribution sys -
tem. With regard to the latter, the difficulty arises in determining what consti-
tutes an adequate residual. There are examples of disease outbreaks caused by
external contamination of a distribution system with a virus (Levy et al., 1998)
and Giardia (Craun and Calderon, 2001); in both cases, a disinfectant residual
was present or required. A few studies have examined the persistence of patho-
gens introduced into water carrying a disinfectant residual. Using sewage at
various concentrations, Snead et al. (1980) demonstrated that there was no
pathogen inactivation by chorine at 0.2 mg/L when 0.01 percent sewage
was added. Payment (1999) reported inactivation of indigenous coliforms in
sewage only if the chlorine concentration was greater than 0.6 mg/L. These low
levels of chlorine are typical in sections of distribution systems with more
advanced water age and in premise plumbing. More recent work has involved
modeling of potential intrusion events to obtain insight on how chlorine
and monochloramine inactivate organisms under relevant scenarios. Propato
and Uber (2004), whose approach incorporates the variable factors of intrusion
location along with mixing and contact time prior to consumption to simulate
inactivation of pathogens in the distribution system, showed that
monochloramine did

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
not provide any protection against contamination by Giardia under r
e
a
l
i
s
t
i
c
conditions.
In another modeling study, chloramine and chlorine were evaluated for their
ability to inactivate Giardia and E. coli O157:H7 under a range of water
quality conditions in the distribution system (Baribeau et al., 2005b). This
group demonstrated that chlorine at a level of 0.5 mg/L would not
inactivate Giardia, but was sufficient to disinfect E. coli when a simulated
sewage intrusion event was 0.2 percent of the total flow. In contrast,
monochloramine under the same conditions performed poorly in reducing the E.
coli counts in a reasonable amount of time. In both of these modeling studies,
there were inherent assumptions that remain to be verified under field
conditions. However, the insights obtained, along with the laboratory and
disease outbreak data, demonstrate that criteria for disinfecting organisms
introduced during external contamination events is not well understood.
Federal regulations regarding the maintenance of a distribution system r
e
-
sidual require, for large systems, a detectable free or combined residual in 95
percent of the sample results analyzed during a one-month period, or demonstra-
tion of a heterotrophic plate count less than 500 CFU/mL. Smaller systems have
reduced monitoring requirements. Some states have chosen to define “detect-
able residual” including specific requirements for chlorinated and chloraminated
distribution systems. For example, Texas requires 0.2 mg/L for chlorinated wa-
ter and 0.5 mg/L for chloraminated water (TECQ, 2005). The North Carolina
regulations are even more stringent (NCDENR, 2004), in that when chlorine is
the single applied disinfectant, the residual disinfectant in the distribution sys -
tem must be at least 0.2 mg/1 as free chlorine in at least 95 percent of the sam-
ples each month. When ammonia and chlorine are applied together as disinfec-
tants, the residual disinfectant must be at least 2.0 mg/1 as combined chlorine in
at least 95 percent of the samples each month.
In addition to the state regulations mentioned above, the literature i
m
pl
i-
cates and in some cases makes suggestions for appropriate disinfectant levels.
For example, systems that maintained dead-end free chlorine levels of < 0.2
mg/liter or monochloramine levels of < 0.5 mg/liter had substantially more coli-
form occurrences than systems that maintained higher disinfectant residuals
(LeChevallier et al., 1996). This committee did not reach consensus on recom-
mending specific numbers, given the need for additional research on the level of
protection provided by maintenance of a disinfectant residual and the large vari-
ability in contact time between points of contaminant entry and consumers, both
within an individual distribution system and between systems. To date, most
studies have examined rather large amounts of contamination (1 percent or
more), and studies have not been done in flowing pipes, where the hydrodynam-
ics would be important. It is not clear what level of microbial inactivation
would be required during an event of a given magnitude, nor how that might
vary depending on the type of organisms in the vicinity of the distribution sys-
tem. Given that current federal regulations for surface water systems require a
“detectable” disinfectant level within the distribution system, each utility should
set targets depending on the expected loss of residual in the system. This loss

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
will depend on (1) the extent of treatment to minimize disinfectant demand i
n the
bulk water and (2) distribution system operational practices that minimize the
disinfectant demand of the pipe walls and minimize water age (such as turn-
over of stored water).
Booster Disinfection
Reactions can reduce disinfectant residuals within distribution system
s,
such that some utilities have chosen to use booster chlorination or booster
chloramination to increase residuals at susceptible locations. Using
additional points of disinfectant application in the distribution system can
reduce the amount of chlorine added at a treatment plant for the purpose of
maintaining the distribution system residual. This, in turn, has the potential to
limit DBP formation and subsequent exposure of those consumers’ drinking
water from taps close to the initial source. The booster disinfection
simultaneously increases protection (in terms of the presence of a residual) for
those drinking water from taps with longer hydraulic residence times.
An important consideration for implementing booster chlorination
o
r
chloramination is the proper location of facilities. Kirmeyer et al. (2000) list the
following criteria in selecting the location of boosterstations:
 The location should be such that a relatively large volume of water c
a
nbe
disinfected.
 The water to be treated travels in onedirection.
 The chlorine residual in the water has begun to decrease, but has n
o
tto-
tally dissipated.
 The chlorine can be applied uniformly into thewater.
 The location is acceptable by neighbors and is easily accessible for
chemical delivery vehicles with room for chemical storage and feed equipment.
 Power is readily available.
 Communications systems are readily available for the SCADA system.
 Flow and/orresidual pacing can be used.
 Safety concerns can be addressed.
 For a common inlet/outlet line, chlorine should be injected as the stor-age
facility is filling, although mixing the chlorine throughout the contents may be
difficult.
Booster Chlorination
Booster chlorination is an alternative for maintaining a residual in drinking
water systems where substantial disinfectant degradation occurs with travel
through the system. When Uber (2003) surveyed 4,000 drinking water utilities,
15 percent of the respondents reported currently using boosterchlorination for

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
(1) disinfectant residual maintenance, (2) prevention of biological regrowth, and
(3) disinfection after open reservoirs. Utilities practicing booster disinfection
ranged in size from 0.14 to 830 MGD, with an average of 55 MGD. Most (55
percent) of the booster stations operated with a constant delivery dose, although
35 percent used flow-pacing or residual pacing to adjust dose. A few stations
used a time-dependent set-point regime. Fifty-seven (57) percent of the stations
were controlled manually, 33 percent were automated, and ten percent were con-
trolled remotely with the aid of SCADA. Half of the stations with automatic
control also had remote alarms. Examples provided in the report show that in -
corporation of decay rate and THM formation data is fundamental to predicting
whether there will be any net gain in maintenance of residual and formation of
THMs when disinfectant application is changed from a single location to multi-
ple locations. Important products of the study were the Booster Disinfection
Design and Analysis software and network models, which aid in the placement
and operation of booster disinfection systems. The software is capable of pro-
viding information such as (1) setting the dosing schedules given the locations
are provided; (2) selecting of booster dose schedules and location; and (3) heu-
ristic screening of potential booster locations.
Booster Chloramination
Approximately 12 percent of the respondents to the survey
published i
n
Uber (2003) practice chloramination, by one of two methods. Most
reportedly use chlorine to bind excess ammonia—a useful approach if
chloramine decay results in the excess ammonia or if sufficient ammonia
remained during the initial formation of chloramine. Three booster stations
were identified in the survey using a second method in which both chlorine and
ammonia were applied at the same location. This approach is used when
there is a need to increase the overall concentration of chloramine present.
Wilczak et al. (2003) reviewed operating practices at utilities employing booster
chloramination with the addition of free chlorine (Martin and Cummings,
1993; Cohen, 1998; Ireland and Knudson, 1998) or both chlorine and
ammonia (Potts et al., 2001). Monitoring of chlorine and ammonia was
practiced by all utilities, in the majority of cases with on-line combined chlorine
analyzers. Nitrite, pH, and on-line free ammonia analyzers were also employed
by some of the utilities. Process control options included manual dose control,
dosage determined by the flow, dosage determined by the flow along with
measurements of the chlorine residual, and dosage set by the flow and
controlled by a desired feed level set point. In all cases, operators could
manually alter the chemical doses depending on water quality results. The goal
of all utilities was to maintain a total chlorine residual of at least 2.0 mg/L.

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
***
Several factors should be investigated before implementing booster di
si
n-
fection. Hatcher et al. (2004) described recommendations made to the Sweet-
water Authority, which could be made without capital expenditure in the distri-
bution system. Among these were (1) optimizing corrosion control, (2) improv-
ing the biostability of the water by reducing the free ammonia concentrations at
the treatment plant effluent to prevent nitrification (thereby avoiding disinfectant
loss), and (3) conversion of the last of three plants to chloramine to avoid chlo-
rine-chloramine blending in the system. Grayman et al. (2004) described the
methods to characterize and improve mixing within a storage reservoir so that
disinfectant decay in the reservoir could be minimized. If operations such as
these (improved tank mixing, optimized chloramine formation at the plant, im-
proved corrosion control to reduce disinfectant demand by pipe surfaces) can
improve the maintenance of the total chlorine residual, boosting may not be nec-
essary.
Corrosion Control
As discussed in Chapter 4, there are measures that can be taken to control
both internal corrosion and metal release, including materials selection for the
distribution system, addition of a corrosion inhibitor such as phosphates, control
of the chemistry of the water being distributed, or some combination of
these approaches. Materials selection is important because materials that are not
subject to corrosion can be used when it is very difficult to control water
composition. The use of phosphate inhibitors to control problems with iron,
lead, and copper is widespread, but care must be taken to ensure that the added
phosphate does not decrease the biological stability of the water or cause a
problem in municipal wastewater treatment plant discharges. Water stagnation
is an important cause of many iron release problems, so distribution systems
must be designed to maintain flowing water conditions to the extent possible
(Sarin et al., 2004). Also, the use of proper pH control and maintenance of an
acceptable alkalinity concentration are also effective ways to control both
corrosion and metal release (Sarin et al., 2003).
It should be noted that internal corrosion control can positively i
nfl
uence
the effectiveness of chlorine-based disinfectants for inactivation of bacteria in
biofilms. Corrosion products react with residual chlorine, preventing the biocide
from penetrating the biofilm and controlling coliform growth. Studies have
shown that free chlorine is impacted to a greater extent than monochloramine,
although the effectiveness of both disinfectants is impaired if corrosion rates are
not controlled (LeChevallier et al., 1990, 1993). Increasing the phosphate-based
corrosion inhibitor dose, especially during the summer months, can help reduce
corrosion rates.

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
Materials Specification to Control Leaching
Chapter 4 discusses how the quality of materials is critical to minimizing
the potential for external contamination to enter into the potable water system
via leaks, breaks, and permeation. Internal contamination processes like leach-
ing and corrosion are also related to the quality of the materials used in the dis-
tribution system. Most substances leaching into water from materials in the dis -
tribution system are non-toxic and unlikely to cause health problems. However,
PVC pipes manufactured before about 1977, cement materials, and the excessive
leaching of organic substances from linings and joining and sealing materials
have occasionally been noted. In addition to the direct effect of leaching, new
pipe materials can have a significant indirect effect on internal water quality. For
example, they may exert a chlorine demand that can reduce the residual dis -
infectant in the distribution system and hence degrade the microbial quality of
the drinking water (Haas et al., 2002). Standards for manufacture of materials
and guidance for specifying materials need to be updated to address water qual-
ity issues (e.g., leaching of emerging chemicals, biogrowth promoting potential,
leaching of non-health related but taste/odor-related chemicals, susceptibility to
permeation).
RECOVERING WATER QUALITY INTEGRITY
Recovering water quality integrity following an internal contamination
event in the distribution system revolves around a few specific activities. In
many cases, the same method is used to address several issues including
colored/turbid water, loss of residual, nitrification, and elevated microbial
counts. Options are limited to (1) flushing to remove the taste/odor/color/
turbidity or to restore disinfectant concentrations, (2) permanently switching
disinfectants to maintain a residual, typically from free chlorine to chloramine,
(3) periodically changing from chloramine to free chlorine to mitigate
nitrification, (4) implementing corrosion control to reduce corrosion and
leaching, or (5) changing water sources. These approaches are discussed in more
detail below. It should be noted that other distribution system maintenance and
repair options such as cleaning, relining, replacement, and localized disinfection
can also alleviate internally derived water quality problems; these methods are
described in Chapters 4 and 5.
Flushing
As shown in previous chapters, water main flushing is an operational acti
v-
ity that involves moving water through the distribution system, often at a rate
that facilitates scouring of the surfaces, and discharging it through hydrants or
blow-off ports.Many researchers and utility managers have suggested that op-

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
timized flushing is important in maintaining or recovering water quality (
P
a
t
t
i
-son,
1980; Emde et al., 1995, 1997; Smith et al., 1996; Antoun et al., 1997; Bar- beau
et al., 1999). Disinfectant residuals may be regained or maintained by
moving out “old” water and replacing it with water containing a measurable
residual.
Flushing can also be used to remove deposits as well as discolored w
a
-ter
resulting from suspended material, which can be an effective method f
o
r
combating
biofilm growth. Gauthier et al. (1997) showed that loose deposits in a French
system removed by flushing contained organisms including invertebrates,
protozoa, and bacteria. Ackers et al. (2001) characterized the sediments removed
by flushing as corrosion products, components of the pipe lining, treatment
breakthrough, animal and biomatter, and calcium deposits. Antoun et al.
(1997) recommended that flushing be used to reduce the potential for total
coliform positive samples in a distribution system, and the approach seemed to
alleviate the problem. Similarly, Emde et al. (1995 and 1997) suggested that
flushing at sufficient velocities could remove biomass from pipe surfaces, there-
fore controlling biofilms. This approach was utilized by the Zurich Water Sup-
ply in Switzerland to control regrowth in a distribution system supplying water
without a secondary disinfectant (Klein and Forster, 1998). In another situation,
temporary control of invertebrates through a program including flushing was
advocated (van Lieverloo et al., 1998).
A recent survey (Friedman et al., 2003) showed that of 23 U.S. utilities
that responded, 20 have regularly scheduled flushing programs, while the
remaining three flushed on an as-needed basis. In order of frequency cited, the
objectives used for flushing were to eliminate colored water, restore disinfectant
residual, reduce turbidity, eliminate tastes and odors, reduce the number of bac-
teria, reduce DBP precursors, remove sediment, comply with regulations, main-
tain water quality, decrease chlorine demand, respond to customer complaints,
reduce corrosion inhibitor build-up, eliminate stale water, respond to animal
activity, and eliminate lime deposits. Another survey (summarized in Table 2-5)
suggests that flushing is variously supported by state agencies. Of 34 respond-
ing states, in only 11 states are flushing/cleaning/pigging required, with 20 oth-
ers encouraging the practices by utilities. One of the difficulties in assessing the
efficacy of flushing for restoring or maintaining water quality is the lack of data
collection by utilities. A nation-wide survey (Chadderton et al., 1992) showed
that utilities typically implemented flushing in response to consumer complaints.
In this study they also found that less than half of the utilities collected water
samples during the flushing process for analysis of chlorine, turbidity, bacteria,
or other parameters. In most cases, flushing proceeded until the water was visu-
ally clear. In light of the benefits that can be attained by properly conducted
flushing (to minimize the amount of water wasted and appropriately discharge
the waste), more attention should be given to this approach for maintaining qual-
ity or resolvingproblems.

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
Change Disinfectant
For nearly 100 years, drinking water utilities in the United States and C
an-
ada have converted to chloramine for improved ability to maintain a residual at
dead ends or other areas with high hydraulic residence time due to the
lower decay rates of chloramine versus free chlorine. Systems also have
converted to chloramine for taste, odor, or DBP control (Kirmeyer et al., 1993).
Lowering of the THM standard from 100 to 80 µg/L in 2001 and the
forthcoming requirement to meet this value at each monitoring location have
encouraged additional utilities to switch to chloramine.
In recent years the ability of chloramine to penetrate biofilms and
controlLegionella has received more attention and helped those who have already
converted for other reasons (e.g., maintenance of residual, taste and odor, or
DBP control) justify their decision. Preliminary results show that although
Legionella is better controlled with chloramination (see Chapter 3 and Pryor et
al., 2004), it is possible that other organisms of public health concern such as
Mycobacteria could have a selective advantage under these conditions (Pryor
et al., 2004). The ability of chloramine to control biofilms has been
documented by several utilities that find it easier to meet the requirement of
the Total Coliform Rule using chloramine (Norton et al., 1997; Richard Mann,
Metropolitan Water Dis- trict of Southern California, Los Angeles, CA, personal
communication, 1998) although this increased control is not universal
(Muylwyk et al., 2001). In a multi-year study in pipe loops with a three-day
residence time, both chlorine and chloramine were found to be effective
disinfectants but chloramine persisted longer (Clark et al., 1994). Other
laboratory studies have not demonstrated an advantage of monochloramine over
free chlorine (Ollos et al., 1998; Clark and Sivaganesan, 1999; Camper et al.,
2003).
Meanwhile, issues regarding the disturbance of bacteria or metallic oxides on
the pipe walls during the disinfectant switch and their influence on w
a
t
e
rquality
have been and continue to be a concern. A recent example of how t
he
disturbance of pipe walls can affect water quality is described by Edwards and
Dudi (2004) regarding the District of Columbia Water and Sewer Authority’s
drinking water lead levels after the conversion from free chlorine to chloramine.
During the time when chlorine was used as the disinfectant in the distribution
system, lead dioxide formed on the lead pipes. When the redox potential
decreased because of the conversion from chlorine to monochloramine, the
lead dioxide was converted to a more soluble lead (+II) compound, and this
caused an increase in the lead concentration in the bulk water. While it is
likely that iron release and elevated coliform levels concomitant with the
conversion to chloramine will resolve as the distribution systems reequilibrate,
in some cases utilities have intervened with flushing to accelerate the transition,
but noted increases in water quality problems during the flushing event.
In addition to permanent changes in disinfectant, there are instances where
short-term switches are practiced. Drinking water utilities using chloramine as a
disinfectant residual sometimes temporarily switch to free chlorine both as a

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
preventative and control measure for nitrification. This can be accompli
shedsystem-
wide by simply turning off the ammonia feed pumps. To switch to f
r
e
echlorine in
an isolated pressure zone or storage facility, enough chlorine must b
e
added to pass
the breakpoint and achieve a free chlorine residual. There are differing views
about the practice of periodic chlorination in chloraminated systems. Indeed,
the practice has been abandoned by some utilities in Southern California as
a response to legitimate concerns over short-term exposure to elevated DBP
levels (Hatcher et al., 2004).
When nitrification occurs in storage tanks, and other strategies such as a
d
-
justments in chlorine to ammonia ratio, increased turnover, or flushing have not
solved the problem, breakpoint chlorination is a common response (Skadsen,
and Cohen, 2006). In this case chlorine is added to oxidize all of the n itrogen
species (ammonia and nitrite) and achieve a free chlorine residual. This strategy
is effective because ammonia oxidizing bacteria are sensitive to free chlorine
(Baribeau, 2006), but temporary since their populations will recover after the
weaker disinfectant (chloramine) is reestablished in the facility or system. This
is demonstrated by a need for repeated breakpoint chlorination of reservoirs es -
pecially in summer months.
Adequate mixing is important for efficient breakpoint events and for routine
maintenance of the disinfectant residual within storage facilities. Figure 6-4
shows the variability of a chloramine (total chlorine) residual at the sample tap
on a common inlet/outlet pipe to a storage tank. As the tank fills, water from the
distribution systemwith a total chlorine residual of about 2 mg/L enters the tank;
FIGURE 6-4 Variability in storage tank chlorine residual as a function o
f
filland drain cycle.
SOURCE: Reprinted, w ith permission, fromGuistino (2003). © 2003 by Joe Guistino.
East Reservoir
4.0 20.0
3.0 15.0
2.0 10.0
1.0 5.0
0.0 0.0
Date
ChlorineResidual Elevation
Chlorine
Residual
08/16/02
00
12
08/17/02
00
12
08/18/02
00
12
08/19/02
00
12
08/20/02
00
12
08/21/02
00
12
08/22/02
00
12
08/23/02
00
Elevation

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
as the tank drains the water is observed to have a total chlorine residual of about1
mg/L. Some intermediate points are seen on the figure that are characteristic of
distribution water mixing with tank water (Guistino, 2003). Grayman et al.
(2004) have described design and operational problems that lead to poor mixing
within storage facilities. Figure 6-4 also illustrates the utility of using continu-
ous disinfectant analyzers on storage facilities. The data reveal that the disinfec-
tant residual is a function of the fill and drain cycle (or elevation) of the tank
more than the water quality of the storage facility and distribution system.
Regardless of whether disinfectant changes are long- or short-term, utilities
should be aware that these changes may have implications for protecting publ
ic
health, especially during an intrusion event. Chloramine may be inadequate for
protection against microorganisms that enter the distribution system during
intrusion, as discussed previously for Giardia cysts but also for enteric
viruses with less susceptibility to chloramine than chlorine. Karim et al. (2003)
showed that over half of soil samples collected during pipe replacements tested
positive for enteric viruses.
Change Treatment/Corrosion Control
If corrosion or metal release is identified as a problem, one of the measures
listed in the previous section on corrosion control should be undertaken. Identi-
fication of the cause of the problem is most important relative to selection of the
best approach. For example, if the cause of the problem is variable pH, treat-
ment to control pH and possibly to add alkalinity for the purpose of increasing
buffer intensity is probably necessary. Adding alkalinity also benefits corrosion
control via the addition of carbonate (from the standpoint of developing a stable
scale).
Beyond contamination that enters potable water from external sources (
i
n
-
trusion, cross connections, etc.), there are processes within the distribution sys-
tem that contribute to degradation of water quality. The large surface area to
volume ratio of pipe surfaces, reactive pipe materials, advanced water ages, and
bulk water reactions all contribute to deleterious changes in water quality from
the treatment plant to the consumer. Maintaining water quality integrity in the
distribution system is challenging because of the complexity of the system.
There are interactions between the type and concentration of disinfectants, cor-
rosion control schemes, operational practices (e.g., flow characteristics, water
age, flushing practices), the materials used for pipes and plumbing, the biologi-
cal stability of the water, and the efficacy of treatment. In some cases, changes
to improve water quality may be reasonably easy, while others may be
extremely difficult. The following conclusions and recommendations are
made.

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
Prior to distribution, the quality of treated water should be adjusted to
minimize deterioration of water quality. For example, appropriate use o
f
phosphate inhibitors and control of pH and alkalinity can be used to
minimize both internal corrosion of lead, copper, and iron pipes and the formation
of colored water owing to the release of iron from corrosion scales.
Coagulation with aluminum salts should be done in a way that minimizes the
residual aluminum concentration in filtered water, thereby reducing the amount
of aluminum that precipitates in the distribution system. Ensuring that water is
biologically stable via removal of organic carbon will have a positive impact on
the preservation of water quality as it travels from the treatment plant to the
consumer. It should be kept in mind that other chemical adjustments may have
an impact on the biological stability of the water.
Microbial growth and biofilm development in distribution systems
should be minimized. Even though general heterotrophs are not likely to be of
public health concern, their activity can promote the production of tastes a
n
d
odors,
increase disinfectant demand, and may contribute to corrosion. Biofilms may
also harbor opportunistic pathogens (those causing disease in the immuno-
compromised); this is of greatest importance in premise plumbing where long
residence times contribute to disinfectant decay and subsequent bacterial growth
and release. Coliforms may also proliferate in biofilms. With perhaps the ex-
ception of E. coli, coliforms from biofilms are indistinguishable fromthose aris-
ing from external contamination. When these coliforms are detected, it is diffi-
cult to determine if a contamination event of public health significance has oc-
curred (fecal contamination) as opposed to the growth of indigenous organisms.
Residual disinfectant choices should be balanced to meet the overall
goal of protecting public health. For free chlorine, the potential residual loss
and DBP formation should be weighed against the problems that may introduced
by chloramination, which include nitrification, lower disinfectant e
f
f
i
c
a
c
y
against
suspended organisms, and the potential for deleterious corrosion problems.
Although some systems have demonstrated increased biofilm control with
chloramination, this response has not been universal. This ambiguity also exists
for the control of opportunistic pathogens.
Current microbial monitoring is limited in its ability to indicate distri-
bution system contamination events, such that new methods and strategies
are needed. Current methods are not specific for the source of contamination and
do not allow for good process control because of the substantial amount o
f time
needed to generate results. There are limits to the effectiveness of indicators
with respect to predicting the presence of pathogens. A tiered approach to
microbial monitoring should be considered in which common indicators are
used initially to provide an early warning of a potential health risk, followed by
more detailed studies to assess the extent of the risk using more specific meth-
ods.In concert with this, techniques for monitoring for specific pathogens with

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
known public health significance should be developed, ideally with resul
ts
available on-line and in real time. The implementation of best practices to main -
tain water quality (see Chapter 2) is needed until better monitoring approaches
can be developed.
Standards for materials used in distribution systems need to be up-
dated to address their impact on water quality, and research is needed to
develop new materials that will have minimal impacts. Materials standards
have historically been designed to address physical/strength properties including
the ability to handle pressure and stress. Testing of currently available m
a
t
e
r
i
a
l
sshould
be expanded to include (1) the potential for permeation of contaminants, and (2)
the potential for leaching of compounds of public health concern as w
e
l
las those
that contribute to tastes and odors and support biofilm growth. T
heresults of these
tests should be incorporated into the standards in a way that water quality
deterioration attributable to distribution system materials is minimized.
Also, research is needed to develop new materials that minimize adverse water
quality effects such as the high concentrations of undesirable metals and
deposits that result from corrosion and the destruction of disinfectant owing to
interactions with pipe materials.
REFERENCES
Ackers, J., M. Brandt, and J. Powell. 2001. Hydraulic characterization of deposits a
n
d
review of sediment modeling. Report REF. No. 01/DW/03/18. London: U
K
W
I
R
Ltd.
American Public Health Association (APHA). 2005. Standard Methods for the E
x
a
m
i
-nation
of Water and Wastewater, 21st
edition. A. D. Eaton, L. S. Clesceri, E. W
.Rice, and
A. E. Greenberg (eds.). Washington, DC: APHA.
Amy, G. L., P. A. Chadik, and Z. K. Chowdhury. 1987. Developing models for p
r
e
d
i
c
t
- ing
trihalomethane formation potential and kinetics. J. Amer. Water Works Assoc.
79(7):89–97.
Antoun, E. N., T. Tyson, and D. Hiltebrand. 1997. Unidirectional flushing: a
remedy t
o
water quality problems such as biologically mediated corrosion. In:
Proceedings of theAWWA Annual Conference. Denver, CO: AWWA.
American Water Works Association (AWWA). 1990. Position statement on c
h
l
o
r
i
n
e
residual. Pp. 196 In: 1995–1996 AWWA Officers and Committee Directory. D
e
n
-ver,
CO: AWWA.
AWWA. 2006. Fundamentals and Control of Nitrification in Chloraminated D
r
i
n
k
i
n
g
Water Distribution Systems. AWWA Manual M56. Denver, CO: AWWA.
AWWA and EES, Inc. 2002. Permeation and leaching. h
t
t
p
:
/
/
w
w
w
.
e
p
a
.
g
o
v
/
s
a
f
e
w
a
-
ter/tcr/pdf/permleach.pdf. Washington, DC: EPA.
Barbeau, B., K. Julienne, V. Gauthier, R. Millette, and M. Provost. 1999. D
e
a
d
-
e
n
d
flushing
of a distribution system: short and long-term impacts on water quality. I
n
:
Proceedings of the AWWA Water Quality Technology Conference. Denver, C
O
:
AWWA.

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
Baribeau, H., S. W. Krasner, R. Chinn, and P. C. Singer. 2005a. Impact of biomass o
n
the
stability of HAAs and THMs in a simulated distribution system. J. Amer. Water
Works Assoc. 97:69–81.
Baribeau, H., N. L. Pozos, L. Boulos, G. F. Crozes, G. A. Gagnon, S. Rutledge, D. S
k
i
n
-ner,
Z. Hu, R. Hofmann, R. C. Andrews, L. Wojcicka, Z. Alam, C. Chauret, S. A
.
Andrews, R. Dumancis, and E. Warn. 2005b. Impact of Distribution System Water
Quality on Disinfection Efficacy. Denver, CO:AwwaRF.
Baribeau, H., P. C. Singer, R. W. Gullick, S. L. Williams, R. L. Williams, S. A. Andrews,
L. Boulos, H. Haileselassie, C. Nichols, S. A. Schlesinger, L. Fountleroy, E. Moffat,
and G. F. Crozes. 2006. Formation and Decay of Disinfection By-Products in the
Distribution System. Denver, CO: AWWA and AwwaRF.
Baribeau, H. 2006. Chapter 6: Growth and inactivation of nitrifying bacteria. In: F
u
n
-
damentals and Control of Nitrification in Chloraminated Drinking Water
Distribu- tion Systems. AWWA Manual M56. Denver, CO: AWWA.
Bartram, J., J. Cotruvo, M. Exner, C. Fricker and A. Glasmacher. 2003. H
e
t
e
r
o
t
r
o
p
h
i
cPlate Counts
and Drinking-Water Safety. London: IWA Publishing.
Bellen, G. E., S. H. Abrishami, P. M. Colucci, and C. Tremel. 1993. Methods for a
s
s
e
s
s
-ing the
biological growth support potential of water contact materials. Denver, CO:
AwwaRF.
Benjamin, M. M., H. Sontheimer, and P. Leroy. 1996. Corrosion of iron and steel. P
p
.
29–77 In: Internal Corrosion of Water Distribution Systems. Denver, CO: Aw-
waRF/DVGW-Technologiezentrum Wasser.
Berend, K., G. Van Der Voet, and W. H. Boer. 2001. Acute aluminum encephalopathy in a
dialysis center caused by a cement mortar water distribution pipe. Kidney I
n
-
ternational 59(2):746–753.
Besner, M.-C., V. Gauthier, P. Servais, and A. Camper. 2002. Explaining the o
c
c
u
r
- rence
of coliforms in distribution systems. J. Amer. Water Works Assoc. 9
4
(
8
)
:
9
5
–109.
Bitton, G. 1994. Role of microorganisms in biogeochemical cycles. Pp. 51
–73In:
Wastewater Microbiology. New York: John Wiley.
Block, J. C. 1992. Biofilms in drinking water distribution systems. In: Biofilms—
Science and Technology. L. F. Melo, T. R. Bott, M. Fletcher, and B.
Capdeville (eds.). Dordrecht, Netherlands: KluwerPublishers.
Bouchard, D. C., M. K. Williams, and R. Y. Surampalli. 1992. Nitrate contamination
o
f
groundwater: sources and potential health effects. J. Amer. Water Works
Assoc. 84(9):84−90.
Bremer, P. J., and G. G. Geesey. 1991. Laboratory-based model of microbially induced
corrosion of copper. Appl. Environ. Microbiol. 57:1956–1962.
Bremer, P. J., G. G. Geesey, and B. Drake. 1992. Atomic force microscopy examination
of the topography of a hydrated bacterial biofilm on a copper surface. Current Mi-
crobiol. 24:223–230.
Bruchet, A. 1999. Solved and unsolved cases of taste and odor episodes in the files of
Inspector Cluzeau. Water Science and Technology 40(6):15–21.
Bryant, E. A., G. P. Fulton, and G. C. Budd. 1992. Chloramination. Pp. 128–
170 I
n
:
Disinfection Alternatives for Safe Drinking Water. New York: Van Nostrand
Rein- hold.
Burlingame, G., J.Choi, M. Fadel, L. Gammie, J. Rahman, and J. Paran. 1994. Sniff n
e
w
mains...beforecustomerscomplain. Opflow20(10):3.

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
Burman, N. P. 1965. Taste and odor due to stagnation and local warming in long l
e
n
g
t
h
sof
piping. In: Proceedings of the Society for Water Treatment and Examination
14:125–131.
Burman, N. P. 1973. The occurrence and significance of actinomycetes in water supply.Pp.
219–230 In: Actinomycetaeles; Characteristics and Practical Importance. G
.Sykes and
F. A. Skinner (eds.). London and New York: Academic Press.
Calles, J., R. Gottler, M. Evans, and J. Syage. 2005. Early warning surveillance
of drinking water by photoionization/mass spectrometry. J. Amer. Water
Works Assoc. 97:62–73.
Camper, A. K., K. Brastrup, A. Sandvig, J. Clement, C. Spencer and A. J. Capuzzi.
2003. Impact of distribution system materials on bacterial regrowth. J. Amer. Wa-
ter Works Assoc. 95(7):107–121.
Chadderton, R. A., G. L. Christensen, and P. Henry-Unrath. 1992. Implementation a
n
d
Optimization of Distribution System Flushing Programs. Denver, CO: AWWA and
AwwaRF.
Chapelle, F. H., and D. R. Lovely. 1992. Competitive exclusion of sulfate reduction b
y
Fe(III)-reducing bacteria: a mechanism for producing discrete zones of high-iron
ground water. Groundwater 30:29–34.
Characklis, W. G. 1988. Bacterial Regrowth in Distribution Systems. Denver, C
O
:AWWA
and AwwaRF.
Characklis, W. G., M. H. Turakhia, and N. Zelver. 1990. Transport and I
n
t
e
r
f
a
c
i
a
lTransport
Phenomena. Chapter 9 In: Biofilms. W. G. Characklis and K. C. M
a
r
- shall (eds.).
New York: John Wiley & Sons.
Choi, J., and R. L. Valentine. 2002a. A kinetic model of N-nitrosodimethylamine (NDMA)
formation during water chlorination/chloramination. Wat. Sci. T
e
c
h
.6(3):65–71.
Choi, J. H., and R. L. Valentine. 2002b. Formation of N-nitrosodimethylamine (NDMA)
from reaction of monochloramine: a new disinfection by-product. Water Res.
36(4):817–824.
Chun, C. L, R. M. Hozalski, and T. A. Arnold. 2005. Degradation of drinking w
a
t
e
r
disinfection byproducts by synthetic goethite and magnetite. Environ. Sci. Technol.
39:8525–8532.
Clark, R. M., B. W. Lykins, Jr., J. C. Block, L. J. Wymer, and D. J. Reasoner. 1
9
9
4
. Water
Quality Changes in a Simulated Distribution System. Journal of Water S
u
p
- ply
Research and Technology—Aqua 43(6):263–277.
Clark, R. M. 1998. Chlorine demand and TTHM formation kinetics: a second- o
r
d
e
r
model. J. Environ. Eng. 124:16–24.
Clark, R. M., and M. Sivaganesan. 2002. Predicting chlorine residuals in drinking w
a
t
e
r
:a
second order model. J. Water Resources Planning and Management 128:1–10.
Clark, R. M., R. Thurnau, M. Sivaganesan, and P. Ringhand. 2001. Predicting the f
o
r
-
mation of chlorinated and brominated by-products. J. Environ. Eng. 127:493–501.
Clark, R. M., and R. A. Haught. 2005. Characterizing pipe wall demand: i
m
p
l
i
c
a
t
i
o
n
sfor water
quality. J. Water Resources Planning and Management 131:208–217.
Clark, R. M., and M. Sivaganesan. 1999. Characterizing the effect of chorine and c
h
lo
r
-
amines on the formation of biofilm in a simulated drinking water distribution sys-
tem. EPA/600/R-01/024. Washington, DC: EPA.
Cohen, Y. K. 1998. Forming chloramine and maintaining residual. Opflow 24(9):1.
Coss, A., K. P. Cantor, J. S. Reif, C. F. Lynch, and M. H. Ward. 2004. Pancreatic cancer
and drinking water and dietary sources of nitrate and nitrite. Amer. J. Epidemiol.
159(7):693−701.

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
bution systemdeficiencies. J. Amer. Water Works Assoc. 93(11):65–75
DeAraujo-Jorge, T. C., C. M. L. Melo Coutinho, and E. V. De Aguiar. 1992. Sulfate -
reducing bacteria associated with biocorrosion - a review. Memorias Instituto
Oswaldo Cruz. 87:329.
Denisov, G. V., B. G. Kovrov, and T. F. Kovaleva. 1981. Effect of the pH and t
e
m
p
e
r
a
-ture of
the medium on the rate of oxidation of Fe+2
to Fe+3
by a culture of Thiobacil- lus
ferrooxidans and the coefficient of efficiency of biosynthesis (translated from
Russian). Mikrobiologiya 50:696.
De Roos, A. J., M. H. Ward, C. F. Lynch, and K. P. Cantor. 2003. Nitratein public w
a
-ter
supplies and therisk of colon and rectum cancers. Epidemiology 14(6):640–649.
Dietrich, A. M., and R. C. Hoehn. 1991. Taste-and-odor problems associated with c
h
l
o
-
rine dioxide. Denver, CO: AwwaRF.
Dietrich, A. M., R. C. Hoehn, G. A. Burlingame, and T.Gittelman. 2004. Practical t
a
s
t
e
-and-
odor methods for routine operations:decision tree. Denver, CO: AwwaRF.
Dodge, D. J., A. J. Francis, J. B. Gillow, G. P. Halada, and C. R. Clayton. 2002. A
s
s
o
- ciation
of Uranium with Iron Oxides Typically Formed on Corroding Steel S
u
r
f
a
c
e
s
.Environ. Sci.
Technol. 36:3504.
Doshi, P. E., W. M. Grayman, and D. Guastella. 2003. Field testing the chlorine w
a
l
l
demand in distribution mains. Pp. 1–10 In: Proceedings of the Annual Conference
of the American Water Works Association. Denver, CO: AWWA.
Duirk, S. and R. L. Valentine. 2002. Reaction of monochloramine with humic acid. J
.
Environ. Monitoring4:85–89.
Edwards, M., and A. Dudi. 2004. Role of chlorine and chloramine in corrosion of l
e
a
d
-
bearing plumbing materials. J. Amer. Water Works Assoc. 96(10):69–81.
Emde, K. M., K. Oberoi, and D. Smith. 1995. Evaluation of various methods for d
i
s
t
r
i
- bution
system biofilm control. In: Proceedings of the AWWA Annual Conference. Denver,
CO: AWWA.
Emde, K. M., K. Oberoi, and D. Smith. 1997. Maximizing distribution system m
a
i
n
t
e
-nance
activities. In: Proceedings of the AWWA Annual Conference. Denver, C
O
:AWWA.
EPA (Environmental Protection Agency). 1993. Total Coliform Rule. 40 CFR, Code o
f
Federal Regulations. Washington, DC: National Archives and Records Administra-
tion.
EPA. 2003. Economic Analysis for Proposed Stage 2 DBPR. EPA 815-D-03-001.
Washington, DC: EPA.
Friedman, M. J., K. Martel, A. Hill, D. Holt, S. Smith, T. Ta, C. Sherwin, D. Hiltebrand,
P. Pommerenk, Z. Hinedi, and A. Camper. 2003. Establishing Site Specific Flush-
ing Velocities. Denver, CO: AWWA and AwwaRF.
Gang, D. D., R. L. Segar, T. E. Clevenger, and S. K. Banerji. 2002. Using chlorine d
e
-
mand
to predict TTHM and HAA9 formation. J. Amer. Water Works Assoc.
94(10):76–86.
Gauthier, V., C. Rosin, L. Mathieu, J. M. Portal, J. C. Block, P. Chaix, and D. G
a
t
e
l
.
1997. Characterization of the loose deposits in drinking water distribution systems.
In: Proceedings of the AWWA Water Quality Technology Conference.
Denver, CO: AWWA.
Geesey, G. G., P. J. Bremer, W. R. Fischer, D. Wagner, C. W. Keevil, J. Walker, A. H. L.
Chamberlain, and P. Angell. 1993. Unusual types of pitting corrosion of copper
tubes in potablewater systems. In:Biocorrosion. London: CRC Press.

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
Geldreich, E. E. 1996. Creating microbial quality in drinking water. Pp. 39–101 In:
Microbial Quality of Water Supply in Distribution Systems. Boca Raton, FL: Lewis
Publishers.
Grayman, W. M., L. A. Rossman, M. A. Gill, Y. Li, and D. E. Guastella. 2002. M
e
a
s
u
r
-ing and
modeling disinfectant wall demand in metallic pipes. In: Proceedings of the EWRI
Conference on Water Resources Planning and Management. Reston VA: ASCE.
Grayman,W. M., L. A. Rossman, R. A. Deininger, C. D. Smith, C. N. Arnold, and J.
F
.
Smith. 2004. Mixing and aging of water in distribution system storage facilities. J.
Guistino, J. 2003. Are you getting a real distribution system sample? Source 16(4):10–
11.
Gulis, G., M. Czompolyova, and J. R. Cerhan. 2002. An ecologic study of n
i
t
r
a
t
ein municipal
drinking water and cancer incidence in Trnava District, Slovakia. Envi-
ronmental Research 88(3):182–187.
Haas, C. N., M. Gupta, R. Chitlluru, and G. Burlingame. 2002. Chlorine demand
in disinfecting water mains. J. Amer. Water Works Assoc. 94(1):97–102.
Harrington, G. W., Z. K. Chowdhury, and D. M. Owen. 1992. Developing a c
o
m
p
u
t
e
r model
to simulate DBP formation during water treatment. J. Amer. Water Works
Assoc.84(11):78–87.
Hatcher, M., W. Grayman, C. Smith, and M. Mann. 2004. Monitoring and modeling o
f
the
Sweetwater Authority distribution system to assess water quality. In: Proceed-
ings of the AWWA Annual Conference. Denver, CO: AWWA.
Haudidier, K., J. L. Paquin, T. Francais, P. Hartemann, G. Grapin, F. Colin, M. J. Jour-
dain, J. C. Block, J. Cheron, O. Pascal, Y. Levi, and J. Miazga. 1988. Biofilm
growth in drinking water networks: a preliminary industrial pilot plant experiment.
Water Sci. Technol. 20:109–115.
Ireland, C., and M. Knudson. 1998. Portland’s experience with chloramine
residual management. In: Proceedings of the Conference on Protecting Water
Quality in the Distribution System: What is the Role of Disinfection Residuals?
April 26–28, Philadelphia, PA.
Jafvert, C. T., and R. L. Valentine. 1992. Reaction scheme for the chlorination of a
m
-
moniacal water. Environ. Sci. Technol. 26:577–586.
Joret, J. C., C. Volk, R. Javadpoour, G. Randon, and P. Cote. 1994. Control of b
i
o
d
e
- gradable
organic matter during drinking water treatment. In: Proceedings of the In-ternational
Seminar on Biodegradable Organic Matter. Montreal, Quebec.
Karim, M. R., M. Abbaszadegan, and M. W. LeChevallier. 2003. Potential for pathogen
intrusion during pressuretransients. J. Amer. Water Works Assoc. 95(5):134–146.
Keinanen, M. M., L. K. Korhohen, M. J. Lehtola, and I .T. Miettinen. 2002. The m
i
c
r
o
-bial
community structure of drinking water biofilms can be affected by p
h
o
s
p
h
o
r
u
s availability.
Appl. Environ. Microbiol. 68:434–439.
Khiari, D., S. Barrett, R. Chinn, A. Bruchet, P Piriou, L. Matia, F. Ventura, I. Suffet,
T
.
Gittelman, and P. Leutweiler. 2002. Distribution generated taste-and-odor
phenomena. Denver, CO: AwwaRF.
Kirmeyer, G. J., G. W. Foust, G. L. Pierson, J. J. Simmler, and M. W. Lechevallier.
1993. Optimizing Chloramines Treatment. Denver, CO: AWWA and AwwaRF.
Kirmeyer, G., M. Friedman, J. Clement, A. Sandvig, P. F. Noran, K. D. Martel, D. Smith,
M. LeChevallier, C. Volk, J. Dyksen, and R. Cushing. 2000. Guidance Manual f
o
r
Maintaining Distribution SystemWater Quality. Denver, CO: AwwaRF.

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
Klein, H., and R. Forster. 1998. Network operation without safety chlorination. Z
u
r
i
c
hWater
Supply 16(3/4):165–174.
Kovalenko, T. V., G. I. Karavaiko, and V. P. Piskunov. 1982. Effect of Fe3+
ions on t
h
e
oxidation of ferrous iron by Thiobacillus ferroxidans at various temperatures (trans-
lated from Russian). Mikrobiologiya 51:42.
Krasner, S. W., M. J. McGuire, and V. B. Ferguson. 1985. Tastes and odors: the f
l
a
v
o
r profile
method. J. Amer. Water Works Assoc. 77:34–39.
LeChevallier M. W., B. H. Olson, and G. A. McFeters. 1990. Assessing and Controlling
Bacterial Regrowth in Distribution Systems. Denver, CO: AWWA.
LeChevallier, M. W., W. Schulz, and R. G. Lee. 1991. Bacterial nutrients in drinking
water. Appl. Environ. Microbiol. 57:857–862.
LeChevallier, M. W., C. D. Lowry, R. G. Lee, and D. L. Gibbon. 1993. Examining t
h
e
relationship between iron corrosion and the disinfection of biofilm bacteria. J
.
LeChevallier, M. W., N. J. Welch, and D. B. Smith. 1996. Full-scale studies of f
a
c
t
o
r
srelated
to coliform regrowth in drinking water. Appl. Environ. Microbiol. 62:2201– 2211.
Leclerc, H. 2003. Relationships between common water bacteria and pathogens in
drinking-water. Pp. 80–118 In: Heterotrophic Plate Counts and Drinking-Water
Safety. J. Bartram, J. Cotruvo, M. Exner, C. Fricker, and A. Glasmacher
(eds.). London: IWA Publishing.
Lee, S. H., S. O'Connor, and B. K. Banerji. 1980. Biologically mediated corrosion a
n
d
its
effects on water quality in distribution systems. J. Amer. Water Works Assoc.
72:636–645.
Lehtola, M. J., and I. T. Miettinen. 2001. Microbially available organic carbon, p
h
o
s
- phorus,
and microbial growth in ozonated drinking water. Water Res. 3
5
:
1
6
3
5
– 1640.
Lehtola, M. J., I. T. Miettinen, and P. J. Martikaninen. 2002a. Biofilm f
o
r
m
a
t
i
o
n in
drinking water affected by low concentrations of phosphorus. Can. J. Microbiol.
48:494–499.
Lehtola, M. J., I. T. Miettinen, T. Vartianen, and P. J. Martikaninen. 2002b.
Changes i
n
content of microbially available phosphorus, assimilable organic carbon
and microbial growth potential during drinking water treatment processes.
Water Res. 36:3681–3690.
Lehtola, M. J., I. T. Miettinen, M. M. Keinanen, T. K. Kekki, O. Laine, A. Hirvonen,
T
.
Vartianen, and P. J. Martikaninen. 2004. Microbiology, chemistry and biofilm de-
velopment in a drinking water distribution system with copper and plastic
pipes. Water Res.38:3769–3779.
Levy, D. A., M. S. Bens, G. F. Craun, R. L. Calderon, and B. L. Herwaldt. 1998. S
u
r
-
veillance for waterborne-disease outbreaks—United States, 1995–1996. M
M
W
R
47(SS-
5):1–34.
Lu, C., P. Biswas, and R. M. Clark. 1995. Simultaneous transport of substrates, disinfec-
tants, and microorganisms in water pipes. Water Res. 29(3):881–894.
Lytle, D. A., and V. L. Snoeyink. 2002. Effect of ortho- and polyphosphates o
n the
properties of iron particles and suspensions. J. Amer. Water Works Assoc.
94(10):87–99.
Lytle, D. A., T. J. Sorg, and C. Frietch. 2002. The significance of arsenic-bound s
o
l
i
d
sin
drinking water distribution systems. In: Proceedings of the AWWA Water Qual- ity
Conference. Seattle, WA.

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
Lytle, D. A, T. Sorg, and C. Frietch. 2004. The accumulation of arsenic in water d
i
s
t
r
i
- bution
systems. Environ. Sci. Technol. 38:5365–5372.
Martin, P., and E. Cummings. 1993. Rechloramination in the distribution system. In:
Proceedings of theAWWA Annual Conference. Denver, CO: AWWA.
McClellan, J. N., D. A. Reckhow, J. E. Tobiason, J. K. Edzwald, and D. B. Smith. 2
0
0
0
.A
comprehensive kinetic model for chlorine decay and chlorination byproduct for-
mation. Pp. 223–246 In: Natural Organic Matter and Disinfection By-Products:
Characterization and Controlin Drinking Water. S. W. Krasner, S. E. Barrett, and
G. L. Amy (eds.). Washington, DC: American Chemical Society.
McFeters, G. A., B. H. Pyle, J. T. Lisle, and S. C. Broadaway. 1999. Rapid direct m
e
t
h
-ods
for enumeration of specific, active bacteria in water and biofilms. J. Appl. M
i
-crobiol.
SymposiumSuppl. 85:193S–200S.
McGuire, M. J. 1995. Off-flavor as the consumer's measure of drinking water safety. W
a
t
e
rSci.
Technol. 31(11):1–8.
McGuire, M. J., N. and Graziano, N. 2002. Trihalomethanes in U.S. Drinking Water: N
O
R
S
to ICR.
In: Information Collection Rule Data Analysis. M. J. McGuire, J. L. McLain, and
A. Obolensky (eds.). Denver, CO: AwwaRF.
McGuire, M., N. Graziano, L. Sullivan, R. Hund, and G. Burlingame. 2004. Water Util-ity
Self-Assessment for the Management of Aesthetic Issues. Denver, CO: AWWA.
McNeill, L. S., and M. Edwards. 2001. Iron pipecorrosion in distribution systems. J.
Miettinen, L. T., T. Vartianinen, and P .J. Martikainen. 1997. Phosphorus and bacterial
growth in drinking water. Appl. Environ. Microbiol. 63:3242–3245.
Mitch, W. A., and D. L. Sedlak. 2002. Formation of N-nitrosodimethylamine (NDMA)
from dimethylamine during chlorination. Environ. Sci. Technol. 36(4):588–595.
Montiel, A., S. Rigal, and B. Welte. 1999. Study of the origin of musty taste i
n
the drinking
water supply. Water Science and Technology 40(6):171–177.
Muylwyk, Q., J. MacDonald, and M. Klawunn. 2001. Success! Switching from c
h
lo
r
-amines
to chlorine in the distribution system: results from a one year full-scale trial. In:
Proceedings of the Water Quality Technology Conference. Denver, C
O
:AWWA.
National Research Council (NRC). 2004. Indicators for Waterborne Pathogens. W
a
s
h
-ington,
DC: National Academies Press.
NRC. 2005. Public Water Supply Distribution Systems:Assessingand Reducing Risks.
Washington, DC: National Academies Press.
Nemati, M., and C. Webb. 1997. A kinetic model for biological oxidation of ferrous i
r
o
n
by
Thiobacillus ferroxidans. Biotechnol. Bioeng. 53:478.
North Carolina Department of Environment and Natural Resources (NCDENR). 2
0
0
4
. Title
15A, Subchapter 18C of the North Carolina Administrative Codes. S
e
c
t
i
o
n
2000—Filtration
and Disinfection. Raleigh, NC: NCDENR.
Norton, C. D., and M. W. LeChevallier. 1997. Chloramination: its effect on d
i
s
t
r
ib
u
t
i
o
nwater
quality. J. Amer. Water Works Assoc. 89(7):66–77.
Okereke, A., and S. E. Stevens, Jr. 1991. Kinetics of iron oxidation by
Thiobacillus ferroxidans. Appl. Environ. Microbiol. 57:1052–1056.
Ollos, P. J., R. M. Slawson, and P. M. Huck. 1998. Bench scale investigations of b
a
c
t
e
-rial
regrowth in drinking water distribution systems. Water Sci. Technol. 38:275–282.
Olson, B. H. 1982. Assessment and implications of bacterial regrowth in water d
i
s
t
r
i
b
u
-tion
systems. EPA-600/S2-82-072. Washington, DC: EPA.

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
Palacios, B., and C. D. Smith. 2002. The use of decay studies to plan San Francisco's
conversion to chloramines. In: Proceedings of the Distribution and Plant Operators
Pattison, P. L. 1980. Conducting a regular main flushing program. J. Amer. W
a
t
e
r Works
Assoc. 72(2):88–90.
Payment, P., and E. Franco. 1993. Clostridium perfringens and somatic coliphages
a
s
indicators of the efficiency of drinking water treatment for viruses and
protozoan cysts. Appl. Environ. Microbiol. 59:2418–2424.
Pepper, I. L., P. Rusin, D. R. Quintanar, C. Haney, K. L. Josephson, and C. P. G
e
r
b
a
. 2004.
Tracking the concentration of heterotrophic plate count bacteria from th
esource to
the consumer’s tap. Int. J. Food Microbiol. 92:289–295.
Pereira, V. J., H. S. Weinberg, and P. C. Singer. 2004. Temporal and spatial variability
of DBPs in a chloraminated distribution system. J. Amer. Water Works Assoc.
96:91–102.
Potts, D. E., W. G. Richards, and C. G. Hitz. 2001. A satellite chloramine booster s
t
a
- tion:
design and water chemistry. In: Proceedings of the AWWA Distribution S
ystem
Symposium. Denver, CO: AWWA.
Powell, J. C. 1998. Modeling Chlorine in Water Distribution Networks. Ph.D. Thesis.
School of Civil Engineering. Faculty of Engineering. The University of Birming-
ham.
Powell, J. C., J. R. West, N. B. Hallam, C. F. Forster, and J. Simms. 2000. Performanceof
various kinetic models for chlorine decay. J. Water Resources Planning and
Management ASCE126:3–20.
Propato, M., and J. G. Uber. 2004. Vulnerability of water distribution systems to p
a
t
h
o
-gen
intrusion: how effective is a disinfectant residual? Environ. Sci. Tech
no
l
.
38:3713–3722.
Pryor, M., S. Springthorpe, S. Riffard, T. Brooks, Y. Huo, G. Davis, and S. A. Satter.
2004. Investigation of opportunistic pathogens in municipal drinking water
under different supply and treatment regimes. Water Sci. Technol. 50:83–90.
Regan, J. M., G. W. Harrington, H. Baribeau, R. DeLeon, and D. R. Noguera. 2
0
0
3
.
Diversity of nitrifying bacteria in full-scale chloraminated distribution systems. W
a
-
ter
Res. 37:197–205.
Richardson, S. D. 1998. Drinking Water Disinfection By-Products, Vol. 3. Pp. 1
3
9
8
–
1421
In: Encyclopedia of Environmental Analysis and Remediation. R. A. Meyers (ed.).
New York: John Wiley & Sons.
Richardson, S. D., and S. W. Krasner. 2003. Disinfection by-products and other e
m
e
r
g
-ing
contaminants in drinking water. Trac-Trends in Analytical Chemistry 22(10):666–
684.
Richardson, S. D., H. S. Weinberg, S. W. Krasner, and A. D. Thruston. 2004. N
a
t
i
o
n
- wide
DBP Occurrence Study. EPA-600-R02-068. Athens, GA: EPA NERL.
Ridgway, H. F., E. G. Means, and B. H. Olson. 1981. Iron bacteria in drinking-water
distribution systems: elemental analysis of Gallionella stalks using X-ray energy-
dispersive microanalysis. Appl. Environ. Microbiol. 41:288–292.
Rittmann, B. E., and V. L. Snoeyink. 1984. Achieving microbially stable d
r
i
n
k
i
n
g
water.
J. Amer. Water Works Assoc. 76(10):106–114.
Robertson, W., and R. Brooks. 2003. The role of HPC in managing treatment and d
i
s
t
r
i
-bution
of drinking-water. Pp. 233–244 In: Heterotrophic Plate Counts and D
r
i
n
k
-ing-Water
Safety. J. Bartram, J. Cotruvo, M. Exner, C. Fricker, and A. G
l
a
s
m
a
c
h
e
r(eds.). London: IWA
Publishing.

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
Rossman, L. A., R. M. Clark, and W. M. Grayman. 1994. Modeling chlorine r
e
s
i
d
u
a
l
sin
drinking-water distribution systems. J. Environmental Engineering 120(4):803– 820.
Rossman, L. A., R. A. Brown, P. C. Singer, and J. R. Nuckols. 2001. DBP formation
kinetics in a simulated distribution system. Water Res. 35(14):3483–3489.
Sandor, J., I. Kiss, O. Farkas, and I. Ember. 2001. Association between gastric cancer
mortality and nitrate content of drinking water: ecological study on small area ine-
qualities. European Journal of Epidemiology 17(5):443−447.
Sarin, P., V. L. Snoeyink, D. A. Lytle, and W. M. Kriven. 2004. Iron corrosion s
c
a
le
s
:model
for scale growth, iron release and colored water formation. J. E
n
v
i
r
o
n
m
e
n
ta
l Engineering
130(4):364–373.
Sarin, P., J. Clement, V. L. Snoeyink and W. M. Kriven. 2003. Iron release from c
o
r
- roded,
unlined cast-iron pipe. J. Amer. Water Works Assoc. 95(11):85–96.
Sarin, P., V. L. Snoeyink, J. Bebee, W. M. Kriven, and J. A. Clement. 2001. Physico-
chemical characteristics of corrosion scales in old iron pipes. Water Res.
35(12):2961–2969.
Sathasivan, A., and S. Ohgaki. 1999. Application of new bacterial regrowth
potential method for water distribution system – a clear evidence of
phosphorus limitation. Water Res. 33:137–144.
Sathasivan, A., S. Ohgaki, K. Yamamoto, and N. Kamiko. 1997. Role of i
n
o
r
g
a
n
i
c
phosphorus
in controlling regrowth in distribution systems. Water Res. Technol. 5:37–44.
Schoenen, D. 1986. Microbial growth due to materials used in drinking water systems.
In: Biotechnology, Vol. 8. H. J. Rehm and G. Reed (eds.). Weinheim: VCH Ver-
lagsgesellschaft.
Sen, K., and M. Rogers. 2004. Distribution of six virulence factors in Aeromonas s
pe
-
cies isolated from U.S. drinking water utilities: a PCR identification. J. Appl.
Microbiol. 97:1077–1086.
Servais, P., G. Billen, C. Ventresque and G. P. Bablon. 1991. Microbial activity in G
A
C
filters
at theChoisy-le-Roi treatment plant. J. Amer. Water Works Assoc. 83:62–68.
Shair, S. 1975. Iron bacteria and red water. Industrial Water Eng. 12:16.
Skadsen, J., and Y. K. Cohen. 2006. Operational and Treatment Practices to P
r
e
v
e
n
t
Nitrification. Chapter 8 In: Fundamentals and Control of Nitrification in Chlorami-
nated Drinking Water Distribution Systems. AWWA Manual M56. Denver, CO:
AWWA.
Skjevrak, I., V. Lund, K. Ormerod, A. Due and H. Herikstad. 2004. Biofilm in w
a
t
e
r
pipelines; a potential source for off-flavours in the drinking water. Water Sci. T
e
c
h
-nol.
49:211–217.
Smith, C. 2005. Nitrification: problems and solutions. AWWA Cal/Nevada Conference
Presentation. Available at: www.CharlotteSmith.us.
Smith, C. 2006. Monitoring for nitrification prevention and control. Chapter 7 In: F
u
n
-
damentals and Control of Nitrification in Chloraminated Drinking Water
Distribu- tion Systems. AWWA Manual M56. Denver, CO: AWWA.
Smith, D. B., A. F. Hess, and S. A. Hubbs. 1990. Survey of distribution system c
o
l
i
f
o
r
m
occurrences in the United States. In: Proceedings of the Water Quality Technolog
y
Smith, C. D., J. F. Smith, and B. Milosky. 1996. Excessive loss of chloramine r
e
s
i
d
u
a
l
:a case
study. In: Proceedings of the AWWA Annual Conference. Denver, CO:
AWWA.

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
Snead, M. C., V. P. Olivieri, K. Kawata, and C. W. Kruse. 1980. The effectiveness of
chlorine residuals in inactivation of bacteria and viruses introduced by post-
treatment contamination. Water Res. 14:403–408.
Speight, V. L., and P. C. Singer. 2005. Association between residual chlorine loss a
n
dHAA
reduction in distribution systems. J. Amer. Water Works Assoc. 97(2):82–91.
Suffet, I. H., J. Ho, D. Chou, D. Khiari, and J. Mallevialle. 1995. Taste-and-Odor P
r
o
b
-lems
Observed during Drinking Water Treatment. Pp. 1–21 In: Advances in Taste- and-
Odor Treatment and Control. I. H. Suffet, J. Mallevialle, and E. Kawczynski
(eds.). Denver, CO: AwwaRF.
Szewzyk, U., R. Szewzyk, W. Manz, and K. H. Schleifer. 2000. Microbiological s
a
f
e
t
yof
drinking water. Annu. Rev. Microbiol. 54:81–127.
Tallon, P., B. Magajna, C. Lofranco, and K.-T. Leung. 2005. Microbial indicators o
f
faecal
contamination in water: a current perspective. Water, Air, Soil Poll. 166:139– 166.
Texas Commission on Environmental Quality (TECQ). 2005. 30 TAC290.110(b)(4)
§290.110. Disinfectant Residuals.
Tuovinen, O. H., K. S. Button, A. Vuorinen, L. Carlson, D. Mair, and L. A. Yut. 1
9
8
0
.
Bacterial, chemical, and mineralogical characteristics of tubercles in distribution
pipelines. J. Amer. Water Works Assoc.72:626–635.
Tuovinen, O. H., D. M. Mair, and J. Banovic. 1984. Chlorine demand and t
r
i
h
a
-
lomethane
formation by tubercles from cast iron water mains. Environmental Tech- nology
Letters 5:97–108.
Uber, J. G. 2003. Maintaining Distribution System Residuals through Booster C
h
l
o
r
i
n
a
-tion.
Denver, CO: AwwaRF.
Valentine, R. L., and S. W. Stearns. 1994. Radon Release from Water Distribution S
ys
-
tem Deposits. Environ. Sci. Technol. 28(3):534–537.
Valentine R. L., K. Ozekin, and P. R. Vikesland. 1998. Chloramine
Decomposition i
n
Distribution Systemand Model Waters. Denver, CO: AwwaRF.
van der Kooij, D. 1992. Assimilable organic carbon as an indicator of b
a
c
t
e
r
i
a
l regrowth.
J. Amer. Water Works Assoc. 84:57–63.
van der Kooij, D., and W. A. M. Hijnen. 1990. Criteria for defining the biological s
t
a
b
i
l
-ity
of drinking water as determined with AOC measurements. Pp. 1281–1333 In:
Proceedings of theWater Quality Technology Conference. Denver, CO: AWWA.
van der Kooij, D., W. A. M. Hijnen, and J. C. Kruithof. 1989. The effects of ozonation,
biological filtration, and distribution on the concentration of easily assimilable or-
ganic carbon in drinking water. Ozone Sci. Engrg. 11:297.
van der Wende E., W. G. Characklis, and D. B. Smith. 1989. Biofilm and bacterial w
a
- ter
quality. Water Res. 23:1313–1322.
van der Leeden, F., F. L. Troise, and D. K. Todd. 1990. Water quality. Pp. 417–493 I
n
:
The Water Encyclopedia, Second Edition. Chelsea, MI:Lewis Publishers.
van Lieverloo, J. H. M., R. Buuren, G. Veenendaal, and D. van der Kooij. 1998. C
o
n
- trolling
invertebrates in distribution systems with zero or low disinfectant residual.
Water Supply 16(3/4):199–204.
Vasconcelos, J. J., P. F. Boulos, W. M. Grayman, L. Kiene, O. Wable, P. Biswas, A.
Bhari, L. A. Rossman, R. M. Clark, and J. A. Goodrich. 1996. Characterization and
modeling of chlorine decay in distribution systems. Denver, CO: AwwaRF.
Vasconcelos, J. J., L. A. Rossman, W. M. Grayman, P. F. Boulos, and R. M. Clark.
1997. Kinetics of chlorine decay. J. Amer. Water Works Assoc. 89(7):54–65.

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
Venczel, L. V., M. Arrowood, M. Hurd, and M. D. Sobsey. 1997. Inactivation of
Cryptosporidium parvum oocysts and Clostridium perfringens spores by a
mixed- oxidant disinfectant and by free chlorine. Appl. Environ. Microbiol.
63:1598–1601.
Victoreen, H. T. 1974. Control of water quality in transmission and distribution mains.
J. Amer. Water Works Assoc. 66:369–370.
Vikesland, P. J., K. Ozekin, and R. Valentine. 2001. Monochloramine decay in m
o
d
e
l and
distribution systemwaters. Water Res. 35(7):1766–1776.
Wagner, D., A. H. L. Chamberlain, W. R. Fischer, J. N. Wardell and C. A. C. Sequeira.
1997. Microbiologically influenced corrosion of copper in potable water i
n
s
t
a
l
l
a
-
tions—a European project review. Mater. Corrosion. 48:311–321.
Walker, J. T., K. Hanson, D. Caldwell, and C. W. Keevil. 1998. Scanning confocal m
i
-
croscopy study of biofilm induced corrosion on copper plumbing tubes. B
i
o
f
o
u
l
i
n
g12:333–
344.
Watson, S. 2004. Aquatic taste and odor: a primary signal of drinking-water integrity. J
.
Toxicol. Environ. Health. 67:1779–1795.
Watson, S. B., and J. Ridal. 2004. Periphyton: a primary source of widespread and s
e
-
vere
tasteand odour. Water Sci. Technol. 49:33–39.
Webster, B. J., S. E. Werner, and P. J. Bremer. 2000. Microbially influenced corrosion
of copper in potablewater systems—pH effects. Corrosion 56:942–950.
Wilczak, A. 2001. Chloramine decay rate: factors and research needs. In: Proceedings of
the AWWA Annual Conference. Denver, CO: AWWA.
Wilczak, A., C. D. Smith, Y. K. Cohen, and P. B. Martin. 2003. Strategies for combin-
ing free ammonia and boosting chloramines in distribution systems—survey of util-
ity practice. In: Proceedings of the AWWA Annual Conference and Exposition.
Denver, CO: AWWA.
Wolfe, R. L., N. I. Lieu, G. Izaguirre, and E. G. Means. 1990. Ammonia oxidizing b
a
c
-teria
in a chloraminated distribution system: seasonal occurrence, distribution, a
n
d
disinfection resistance. Appl. Environ. Microbiol. 56(2):451−462.
Wolfe, R. L., E. G. Means, M. K. Davis, and S. E. Barrett. 1988. Biological nitrification in
covered reservoirs containing chloraminated water. J. Amer. Water Works
Assoc. 80(9):109−114.
Woolschlager, J., B. Rittmann, P. Piriou, and B. Schwartz. 2001. Using a comprehen-
sive model to identify the major mechanisms of chloramine decay in distribution
systems. Water Sci. Technol.: Water Supply 1(4):103–110.
Zhang, L., W. A. Arnold, and R. M. Hozalski. 2004. Kinetics of haloacetic acid reac-
tions with Fe(0). Environ. Sci. Technol.38:6881–6889.
Zhang, X. R., and R. A. Minear. 2002. Decomposition of trihaloacetic acids and forma-
tion of the corresponding trihalomethanes in drinking water. Water Res.
36(14):3665–3673.

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
7
Integrating Approaches to R
e
d
u
c
i
n
gRisk
from Distribution Systems
The few regulations that govern water quality in distribution systems are t
h
e
result of years of research leading to the demonstration of a risk to the w
a
t
e
r
-
consuming public from specific contaminants. The development of
regulations is a complex process that includes cost analysis (EPA, 2003) and,
more recently, stakeholder input as described in the Federal Advisory
Committee Act. Many state regulatory agencies are either reluctant to or
prohibited by statute to require measures to protect drinking water beyond
those mandated by federal statute. However, drinking water utilities may
independently choose to conform to industry standards to design and
operate their systems beyond regulatory requirements.
Standards are useful to water suppliers that have adopted such a precaution-
ary stance. Recommended Standards for Water Works: Ten State
Standards (The Great Lakes-Upper Mississippi River Board of State Public
Health and Environmental Managers, 2003), NSF International, and the
American National Standards Institute (ANSI) are third party producers of
standards that are widely used in the drinking water industry. Voluntary
adoption of standards by a utility requires reallocation of resources. Nevertheless
adoption of certain standards is almost universal for community water systems,
such as ANSI/NSF 60 governing components that come in contact with drinking
water, ANSI/NSF 61 governing additives to water, and many American Water
Works Association (AWWA) standards related to design of infrastructure
such as D100-96—Welded Steel Tanks for Water Storage. Other widely used
AWWA standards related to distribution system integrity include the C651—
Disinfecting Water Mains, C652— Disinfection of Water-Storage Facilities, and
D101-53 (R86)—Inspecting and Repairing Water Tanks, Standpipes, Reservoirs,
and Elevated Tanks for Water Storage. In addition to industry standards,
AWWA “Manuals of Water Supply Practices,” such as M6 Water Audits and
Leak Detection, are commonly used by drinking water utilities to enhance their
operations and service to the public.
In 1999 a technical workgroup was organized to develop a Drinking Water
Distribution System Assessment Workbook, which began the process that cul-
minated in the G200 Standard. The purpose of the G200 standard is to “define
the critical requirements for the operation and management of water distribution
systems, including maintenance of facilities” (AWWA/ANSI, 2004). Several
components of the G200 standard relate directly to issues highlighted in the U.S.
Environmental Protection Agency (EPA) Distribution System White Papers (see
Chapter 1) and characterized as high priority by this committee (see Appendix
269

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
A). These include Section 4.1.1: Compliance with regulations, 4.1.3: Disinfec-
tant residual maintenance, 4.2.1 System pressure monitoring and requirements,
4.2.2 Backflow prevention, and 4.3.1 Storage facilities. As listed in Table 7
-
1
,
G200 includes requirements related to water quality, distribution system m
a
n-
agement, and facility operation and maintenance. The standard references sev-
eral existing standards such as those cited above.
TABLE 7-1 G200 Requirements
Section Title Requirement
4.1 Water Quality
4.1.1 Compliance with regulatory re-
quirements
Meet or exceed r
e
g
u
l
a
t
o
r
y
requirements.
4.1.2 Monitoringand control
4.1.2.1 Sampling plan Establish plan, review annual
l
y,
analyze/trend data, h
a
v
e
action plan
to respond to changes.
4.1.2.2 Sample sites Include all types of l
o
c
a
t
i
o
n
s
including
dead ends and storage. Past
problem areas require more sam-
pling.
4.1.2.3 Sample collection Use Standard Methods, s
tan
da
rd
-
ized labels and chain of c
u
s
t
o
d
y
forms.
4.1.2.4 Sample taps Protect fromcontamination.Inspect
annually.
4.1.3 Disinfectantresidualmaintenance
4.1.3.1 Disinfectant residual Maintain detectable or H
P
C
< 500
CFU/mL.
4.1.3.2 Nitrification control Monitor free ammonia, c
o
n
t
r
o
lchlo- rine-
to-ammonia ratio.
4.1.3.2.2 Nitrification monitoring Monitor nitrification i
n
d
i
c
a
t
o
rpa-
rameters.
4.1.3.3 Booster disinfection
4.1.3.3.1 Document residualgoals. Monitor
compliance w ith goals.
4.1.3.3.2 Maintain operating procedures that
take into account s
e
a
s
o
n
a
l variation,
quality, flow, and s
y
stem op-
erations.
4.1.3.3.3 Written Plan show ing response to
variation between goa
l
s and ob-
served values.
4.1.3.4 Disinfection byproduct monitoring
and control
4.1.3.4.1 Monitor and controlDBPs. Set
goals for DBPs at criticalpoints.
4.1.3.4.2 Have action plan to respond to lev-
els that exceed goals.
4.1.4 Requirements for utilities notutil-
izing a disinfectantresidual
Monitor and record HPC.
4.1.4.1 Response program Have action plan to r
e
s
p
o
n
d
w hen HPC
levels are above g
o
a
ls
.
continues

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
INTEGRATING APPROACHES TO REDUCING RISK FROM DISTRIBUTION SYSTEMS 271
TABLE 7-1 Continued
4.1.5
4.1.5.1
Internal corrosion monitoringand
control
Prevention and response program Have action plan to respond t
o
inter-
nal corrosion and d
e
p
o
s
i
t
i
o
n
.
4.1.6 Aesthetic water quality parameters
4.1.6.1 Color and staining Have action plan to address c
o
l
o
r
and staining.
4.1.6.2 Taste and odor Have action plan to address t
a
s
t
e
and odor.
4.1.7 Customer relations
4.1.7.1 Customer inquiries Have systemto document c
u
s
t
o
m
e
r
inquires.
4.1.7.2 Service interruptions Have systemto document p
l
a
n
n
e
d
and
unplanned service interrup-
tions.
4.1.8 System flushing Develop and implement a s
y
s
t
e
m
atic
flushing program.
4.2.1 System pressure
4.2.1.1 Minimum residual pressure Minimum pressure >20 psi.
4.2.1.2 Pressure monitoring Monitor pressure. P
r
e
s
s
u
r
e
alarms may be
used.
4.2.2 Backflow prevention Have program at least as stringent as
AWWA M14.
4.2.3 Permeationprevention Address in utility operation p
l
a
n
.
4.2.4 Water losses
4.2.4.1 Water loss Have goal for the amount of w
a
ter
loss. Document calculation.
4.2.4.2 Response program Have action plan to respondif goal is
not met.
4.2.4.3 Leakage Quantify leakage on annual basis.
4.2.5 Valve exercising andreplacement
4.2.5.1 Valve exercising program Have valve exercising program.
4.2.6 Fire hydrant maintenance and
testing
4.2.6.1 Maintenance and testing Comply w ith AWWA M17.
4.2.7 Materials in contact with potable
water
4.2.7.1 Approved coatings or linings Specify in accordance to AWWA
standards, NSF 61, or other.
4.2.8 Metering
4.2.8.1 Metering requirements Determine daily peak f
l
o
ws
and
maximum day peak flow s.
4.1.8.2 Metering devices Meters shallmeet AWWA r
e
q
u
i
r
e
-ments or
other applicable s
t
a
n
d
a
r
d
.
4.2.8.3 Testing Test as recommended in AWWA M6.
4.2.8.4 Repair and replacement programs Have program that includes records
to verify conformance with AWWA
M6.
continues
4.2 Distribution System M anagement Programs

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
TABLE 7-1 Continued
4.2.9 Flow
4.2.9.1 Flowrequirements Be capable of delivering m
a
x
i
m
u
m
day
demand and fire flow .
4.2.10 External corrosion
4.2.10.1 Leaks/breaks Have a standardized systemfor re-
cording and reporting leaks and
breaks.
4.2.10.2 Monitoring program Have externalcorrosion m
o
n
i
t
o
r
i
n
gplan.
4.2.11 Design review for water quality
4.2.11.1 Policies and procedures Have standardized design p
r
o
c
e
-
dures
that review c
o
n
s
t
r
u
c
t
i
o
n
pro-
jects to reduce potential f
o
r
w ater
quality degradation.
4.2.11.2 Records Prepare as-built draw ings.
4.2.12 Energy management
4.2.12.1 Energy management program Review and optimize e
l
e
c
t
r
i
c
a
len-
ergy usage.
4.3 Facility OperationandMaintenance
4.3.1 Treated water storage facilities
4.3.1.1 Storage capacity Establish minimum operating l
e
v
e
l
s
in storage facilities.
4.3.1.2 Operating procedures Write Standard Operating P
r
o
c
e
-
dures
for turning over facilities and
minimizing w ater age.
4.3.1.3 Inspections Write Standard O
p
e
r
a
t
i
n
g
Proce- dures for
facility i
n
s
p
e
c
t
i
o
n
.
4.3.1.4 Maintenance Have a maintenance p
r
o
g
r
a
m
for facilities.
4.3.1.5 Disinfection Facilities shall be di
si
nfected accord-
ing to ANSI/AWWA C652.
4.3.1.6 Additional requirements All facilities shall be covered.
4.3.2 Pump station operationand main-
tenance
4.3.2.1 Operating procedures Write Standard O
p
e
r
a
t
i
n
g
Proce- dures
describing the o
p
e
r
a
t
i
o
nof
each pump station.
4.3.2.2 Maintenance program Write Standard Operating P
r
o
c
e
-
dures
describing the m
a
i
n
t
e
n
a
n
c
eof
the equipment in each pump s
t
a
t
i
o
n
.
4.3.3 Pipeline rehabilitation and re-
placement
4.3.3.1 Rehabilitation and replacement pro-
gram
Have a programfor e
v
a
l
u
a
t
i
n
g
and
upgrading the d
i
s
t
r
i
b
u
t
i
o
n
system.
4.3.4 Disinfectionof new or repaired
pipes
4.3.4.1 Disinfection of new or repaired pipes Disinfect according to A
N
S
I
/
A
W
W
A
C651
requirements.
4.3.4.2 Bacteriologicaltesting Testing shall be performed a
c
c
o
r
d
ing
to ANSI/AWWA C651.
4.3.4.3 Disposal of chlorinated water Disposal shall follow l
o
c
a
l
,
state, and
federalregulations.
continues

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
TABLE 7-1 Continued
5.1.1 General Include statements of policy and
quality objectives, s
t
a
n
d
a
r
d
operating
procedures etc.
5.1.2 Examples ofdocumentation Document to include requirements of
Section 4.
5.1.3 Control ofdocuments Establish procedures to review and
approve and maintain documents.
5.1.4 Control ofrecords Maintain evidence of c
o
n
f
o
r
m
i
t
yto
requirements of this standard.
5.2.1 General Personnel performing workon the
DS w illbe competent on the basis
of appropriate education,
training, skills, test requirements,
and experience.
5.2.2 Competence, awareness,and
training
The utility shall provide t
r
aining and
determine competence.
SOURCE: Excerpted, with permission, f rom AWWA/ANSI G200 (2004). © 2004 by A
m
e
r
i
c
a
n
Water
Works Association.
As discussed in Chapter 2, the use of the standards such as ANSI/NSF 6
0
,
ANSI/NSF 61, and AWWA G200 and Manuals of Practice have advantages
over programs such as Hazard Analysis and Critical Control Points (HACCP) in
that they are more easily adapted to the dynamic nature of drinking water distri-
bution systems. Use of a standard such as G200 that is intended to assess
whether the system can be managed under all conditions is appropriate for utili-
ties that desire to operate beyond regulatory requirements. To minimize the
public health risks of distribution systems, it is recommended that drinking wa-
ter utilities adopt G200 or an equivalent program in order to develop distribution
system management plans that combine their regulatory requirements and avail-
able voluntary standards.
The purpose of this chapter is to discuss certain elements of G200 that d
e
-
serve more thoughtful consideration because emerging science and t
e
c
h
n
o
l
o
g
yare
altering whether and how these elements are implemented by a typical water
utility. Much of the current scientific thrust is in the development of new moni-
toring methods, models, and methods to integrate data, all to better inform deci-
sion making.
MONITORING
Drinking water of “acceptable quality” is defined by the Safe Drinking Wa
-ter
Act (SDWA) and its amendments and is framed in terms of the Maximum
Contaminant Levels (MCLs), treatment techniques, rules, and regulations prom-
ulgated under the Act. The regulations contain significant monitoring requi
re-
5.2 Human Resources
5.1 DocumentationRequired

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
ments that prescribe the sampling frequency (minimum monitoring frequencies),
sampling locations, testing procedures, record keeping, and the water quality p
a
-
rameters to be monitored, and are classified according to systemsize and v
u
l
n
e
r
-
ability.
The regulations also cover specific reporting procedures to be followed
if a
contaminant exceeds an MCL. Failure to have the proper water quality
analyses performed or to report the results to the state primacy agency can result in
the water systemhaving to provide public notification.
Under the SDWA, monitoring or treatment techniques are required for al
l
contaminants regulated under the Act, both at the entry point to a water distribu-
tion system and, in some cases, at various locations within the system. R
u
l
e
s and
regulations that explicitly require monitoring in the distribution system i
n-
clude the Total Coliform Rule (TCR), the Surface Water Treatment Rule
(SWTR) and Long-Term Enhanced Surface Water Treatment Rule
(LTESWTR), Lead and Copper Rule (LCR), and the Stage 2 Disinfec -
tants/Disinfection By-Products Rule (Stage 2 D/DBPR). These requirements are
summarized in Table 7-2. Routine compliance monitoring is a useful tool for
detecting and assessing some common water quality problems throughout a sys -
tem if the event is large enough and long enough in duration to be detected
(Byer and Carlson, 2005). Note that pressure monitoring is not required by any
of the existing rules, which is unfortunate.
The compliance monitoring required by the SDWA is limited in its ability
to protect public health because the end-point or customer tap monitoring re-
quired under the regulations is typically (1) not sufficient to provide early warn-
ing of contamination, (2) not indicative of what could have gone wrong between
the treatment plant and the consumer’s tap so as to effectively guide remedia -
tion, and (3) too limited across space (too few sampling locations) and time (dis-
crete small volume samples are collected too infrequently) to provide informa -
tion that applies to every potential user. The realities of financial and personnel
resources in most cases preclude expanding monitoring programs to cover vastly
larger areas and periods of time.
Rather, it is more useful for utilities to consider how to control the proc-
esses taking place within the distribution system, as well as activities to maintain
the processes, such that the risk of the customer being exposed to contaminated
drinking water is minimized. This concept hinges on viewing a water distribu-
tion system as a linkage of processes working together to maintain flow, pres-
sure, and water quality. These processes include pumping, valving, metering,
transmission, distribution, service, storage, and corrosion control, to name a few.
Though each individual distribution system is a unique linkage of processes, the
processes have common characteristics that allow generalizations to be made
about their control. For example, the number of storage tanks from one system
to another may be different, but there are common problems with hydraulic
retention time and chlorine loss in all storage tanks. The variety of pipes
used (materials and sizes) will differ from one systemto another, but cast iron
dis- plays a common corrosion problem in all systems.

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
TABLE 7-2 Federal Distribution SystemWater Quality Monitoring Requirements
Regulation MonitoringRequirement
Total Coliform Rule  Samples must be collected at sites that are repre-
sentative of the w ater throughout the d
i
s
t
r
i
b
u
t
i
o
n
systembased
on a sample siting plan that is subject to review by
the primacy regulatory agency.
 The minimum number of samples that m
u
s
tbe col-
lected per month depends on the population
served by the system.
 For each positive total coliform s
a
m
p
l
e
,there are various
repeat sampling requirements.
Surface Water T
r
e
a
t
m
e
n
tRule (SWTR)
and multiple L
o
n
g
-
Term Enhanced
S
u
r
fa
c
e
Wa- ter Treatment Rules
(LTESWTRs)
 Disinfectant residuals must be m
e
a
s
u
r
e
dat TCR
monitoring sites.
 Disinfectant residualmust be m
o
n
i
t
o
r
e
dat the entry to the
distribution system. Larger systems (>3,300
population) must provide continuous monitoring.
Systems serving less than 3,300 population can take
grab samples.
Lead and Copper Rule (LCR)  All systems serving a population > 50,000people
must do w ater quality parameter (WQP) monitoring.
 Samples must be collected for Pb/Cu a
t
Tier I sites.
The number of sample sites for Pb/Cu and w ater
quality monitoring is based on systemsize.
Stage 2 Disinfectants/
Disinfection By-Products
Rule (DBPR)
 Standard Monitoring Program requiresone year of
data on THMs and HAAs. Number of sampling
locations based on utility size and source character-
istics. Modeling can reduce sampling requirement.
SOURCE: Ow ens (2001) and Lansey and Boulos (2005).
In addition to being a linkage of processes, the distribution system is
also a
reactor, in that treated drinking water begins to change physically (e.g., iron
and manganese particles settle out), chemically (e.g., chlorine begins to
decompose) and biologically (e.g., bacterial cells begin to adhere to pipe
surfaces and form biofilms) as soon as water leaves the treatment plant. Each of
the processes display common tendencies to promote these changes
irrespective of how they are linked within a distribution system.
Real-time feedback on whether a utility’s distribution system processes a
r
ein
or out of control goes beyond the regulatory requirements for water quality
monitoring mentioned above. The following sections discuss monitoring for
process control; they are intended to build upon discussions of detection meth-
ods and tools, such as such Geographic Information System(GIS) and hydraulic
modeling, found earlier in the report. A systematic strategy for distribution sys -
tem monitoring to detect water quality alterations is comprised of the following
actions: (1) develop a list of parameters to be monitored, (2) assess appropriate
temporal and spatial scales for monitoring, (3) develop a response plan for
monitored parameters, and (4) implement. Each of these activities is discussed

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
below, focusing on recent scientific developments that should lead to i
m
p
r
o
v
e
-ment in
how the activities are conducted.
Parameters to be Monitored
The parameters that are useful for monitoring distribution system processes
may include those that are required from a regulatory point of view (e.g., turbi
d-
ity, chlorine residual), but likely would include others. The key requirements are
that the monitoring parameters can be measured relatively quickly, inexpen-
sively, and (ideally) continuously at multiple locations in the system. The p
a
-
rameters should be selected with consideration for the potential mechanisms that
may induce adverse changes in water quality. For example, in corros ive waters
passing through ductile iron pipe, conductivity, pH, and oxidation reduction
potential (ORP) may be useful. In waters passing through polymeric pipe or
vulnerable to intrusion in contaminated overlying soils, UV254 or TOC may be
useful.
Table 7-3 lists sentinel parameters that could be used to indicate changes i
n
distribution system integrity. These parameters include indicators of physical
deterioration (pressure changes, main breaks, water loss, or corrosion), hydraulic
failure (turbidity, complaints of low flow or pressure)or a water quality failure
TABLE 7-3 Sentinel Parameters for Distribution SystemIntegrity
Parameter Physical Hydraulic Water Quality
Routine (Primary)
Pressure X X
Turbidity X X (flow r
e
v
e
r
s
a
l
s
) X
Disinfectantresidual X (w ater age) X
Main breaks X
Water loss X
Color X (corrosion) X
Coliforms X (sanitary, main b
r
e
a
k
) X (biofilms)
Flow velocity and
direction
X (pipes, tanks)
pH, Temperature X
Chemical parameters X X X
Secondary
TOC X
UV Adsorption X
T&O X (permeation) X (w ater age) X (biofilms)
Metals X (corrosion) X
Nitrite X (ni
tri
fi
cati
on)
HPC X (biofilms??)
Tank level/volume X
Note: Bold entries indicate those parameters f or which on-line real-time sensors are av ailable.

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
(particulates, tastes, odors, or color). For some of the parameters (listed in
bold), on-line monitoring equipment is available to provide real-time control of
distribution system operations. Most methods for monitoring physical charac-
teristics of water (e.g., flow, velocity, water level in a storage tank) tend to be
relatively inexpensive, quite durable, and able to generate continuous, real-time,
on-line data (Grayman et al., 2004; Panguluri et al., 2005a). Less available in
on-line, real-time versions are methods for detecting inorganic chemicals, syn-
thetic organic chemicals, volatile organic chemicals, and radionuclides. The
direct real time detection of biological changes within distribution systems re -
mains beyond current technology (Bernosky, 2005).
Pressure. One of the most important parameters for utilities to consider
monitoring for is transient pressure change using high-speed, electronic pressure
data loggers. Recent research has documented the frequency and magnitude of
pressure transient events (Friedman et al., 2004; Gullick et al., 2005). High-
speed data loggers are required for monitoring distribution systempressure tran-
sients because such transients may last for only seconds and may not be ob-
served by conventional pressure monitoring. High-speed pressure data loggers
can measure pressures at a rate of up to 20 samples per second, allowing meas -
urement of sudden changes in pressure. The units can be programmed with pre-
set alarm levels to notify operators when specific thresholds have been ex-
ceeded. Additionally, some units can be programmed to capture and store spe-
cific data surrounding a pressure transient event, permitting the episode to
be analyzed and corrective actions to be determined.
Turbidity. Turbidity in distribution systems, which can be can be
caused by suspended sediments, oxidized iron or manganese, or other
corrosion products, is another critical parameter for which on-line, real-time
methods are available. Various models exist but in the finished water
distribution system, turbidity probes need to be sensitive at low ranges (i.e., < 1
NTU). Measurement accuracy may be improved further by employing wiper or
shutter mechanisms that are activated immediately prior to measurements to
avoid interferences from particulates or air bubbles. In general, turbidity units
from different manufacturers behave similarly, and calibration frequencies
vary from weekly up to three monthly intervals, but require a good level of
operator skill. On-line turbidime- ters are being used successfully under the
Partnership for Safe Water Program to monitor low level (< 0.3 NTU) turbidity,
and therefore should prove valuable for low level turbidity in distribution system
monitoring.
Disinfectant Residual. Disinfectant residual monitors can measure f
r
e
e
chlorine, chloramines, or ORP. The principle of detection for residual o
n
-
l
i
n
esensors
relies on either polarographic, voltametric, or colorimetric methods which can
influence their sensitivity, calibration, and interferences from other water
quality parameters. Operation of an ORP sensor is similar to that of the pH
sensorwhere a two-electrode systemis used to make potentiometric meas-

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
Continuous Monitoring of Total Chlorine Residual
urements. The calibration frequency for these monitors is usually on a monthl
y
basis. Typical data from continuous monitoring of total chlorine residual
is shown in Figure 7-1.
Flow. In-line meters are available to measure flow in the distribution sys-
tem but are typically used only to monitor flows into distribution system
sub- districts. Monitoring flows by sub-district can be compared to customer
meter data to indicate the amount of leakage in specific areas of the distribution
system. Flows can be influenced by pumping regimes, storage tank operations,
and manipulations of hydrants or blow-off valves. Use of a well-calibrated
distribution system hydraulic model along with pump, tank, and flow data is
required to generate detailed descriptions of distribution system water velocities
and flow reversals.
pH. Measurements of pH are made with a pH meter using a glass i
ndi
cator
electrode. These measurements are reliable, but the meter requires regular c
a
l
i
-
bration to avoid drift.
Temperature. Temperature thermistors typically work over a r
e
l
a
t
i
v
e
l
y small
temperature range and can be very accurate within that range. The meas-
urements are very reliable and typically do not require routine calibrations.
FIGURE 7-1 Data from a continuous, on-line chlorine analyzer, showing how a total chlo-
rine residual can vary through a day and the need to relate this to system
operations. SOURCE: Data from Philadelphia Water Department, Bureau of Laboratory
Services.
Continuous Monitoring of Total Chlorine Residual
0.5
0.45
0.4
0.35
0.3
0.25
ppm
0.2
0.15
0.1
0.05
0
08/25/04 08/27/04 08/29/04 08/31/04 09/02/04 09/04/04 09/07/04 09/09/04 09/11/04 09/13/04
00:00 04:47 09:35 14:23 19:11 23:59 04:47 09:35 14:23 19:11
Date/Time

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
- -
Chemical Parameters. For chemical parameters, EPA has been examining
the reliability of on-line sensors under the Environmental Technology Verifica -
tion (ETV) Program (EPA, 2004a,b). Currently this program has examined 40
monitoring and treatment technologies and plans to conduct additional testing
under the newly formed Technology Testing and Evaluation Program (TTEP)—
an off shoot of the ETV program, which is not dependant upon voluntary vendor
involvement. This independent testing is providing a valuable database on the
reliability of on-line monitors (http://www.epa.gov/etv/). The sensors being de-
veloped are not specific for the chemical contaminants themselves. Rather, the
premise of the research is that a chemical contaminant in a distribution system
would elicit a pattern of changes in other, primary parameters that can be easily
measured in real time, such that changes in their detection would indicate the
presence of the contaminant. Table 7-4 shows the responsiveness of various
water quality parameters to a range of contaminants in controlled experimental
tests.
It should be noted that the actual ability of the on-line sensors shown in T
a
-ble
7-4 to detect a target contaminant in field situations has not been ascertained.
Hence, whether a particular pattern of shifts in a battery of on-line analysis re-
sults can be reliably associated with a particular type of contaminant (e.g.,
malathion) is uncertain. Another National Research Council committee is in the
process of examining research needs in the area of drinking water
homeland security, and further discussion of this issue may be found in its
report.
TABLE 7-4 Responsiveness of parameters that can be easily measured on-line to various
contaminants
Contaminant
Compound Free/total
chlorine
Ferricyanide NC (F+ w/
DPD t
e
s
t
)
Water Quality Parameter
ORP TOC SC Turbidity NH3 N2 NO3 Cl
+ ++a
+ + F- F- F-
Malathion
(pesticide)
Glyophosphate
(herbicide)
Nicotine
(organic)
Arsenic
trioxide
+ + + NC + +
+ + + NC NC +
++ ++ NC
++ ++ NC + ++ ++
Aldicarb ++ ++
Groundwater + + NC + NC +
Wastew ater + + + + + +
Key : ++ = v ery responsiv e, + = responsiv e, F+ = f alse positiv e, F- = f alse negativ e, NC = no c
h
a
n
g
e
Abbrev iations: ORP, oxidation/reduction potential; TOC, total organic carbon; SC, specif ic c
o
n
d
u
c
t
a
n
c
e
.
a
May be due to bound carbon in the cy anide complex, f rom Hall et al. (2005).
Note: For the pesticides and herbicides, commercial products were used that had dif f erent concentra-
tions of organic compounds.
SOURCE: Adapted f rom Hall et al. (2005).

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
Secondary Parameters. In addition to the primary on-line monitoring
tools, a range of other monitoring approaches can be used to enhance the meas -
urement of distribution system integrity. Both ultraviolet absorbance and trans-
mittance monitors operate on the optical principle where light of known wave-
length (typically 254 nm) and intensity is passed through a sample cell of
a known path length. A photo-detector on the opposite side to the light source
measures the degree of light attenuation by the sample. The percent of UV light
passing through the water determines the UV transmittance (UVT) or alterna-
tively can be translated into absorbance. Double bonds and ring structures
strongly absorb light at 254 nm and therefore absorbance (or transmittance) pro-
vides useful relative measures of the amount of organic matter, which can con-
tribute to color in water. Because the detection procedure utilizes optics, small
particles or other materials (such as dissolved iron) that can deposit on the cell
windows can lead to interference.
Measurement of specific conductance, color, TOC, chemical ions, and m
e
t
-als
can be used to measure changes in water quality baseline values. Conductiv-ity
is directly affected by the number of dissolved ions, and when adjusted for a
given temperature (usually 25o
C), it is referred to as specific conductance (Sie-
mens per cm) and can be used for approximating the total dissolved solids
con-tent. Ion selective electrodes for Cl-
, NO -,
NH +
and others analytes are
avail- 3 4
able but are not entirely ion-specific and can lead to problems of ionic interfer-
ence. Several on-line TOC monitors are commercially available but the routine
maintenance and calibration are cumbersome and require oxidants, carrier gas,
and UV lamp replacements. Some units have simplified this process by using
pre-packaged chemical packs that are easily replaced. Some units use high tem-
perature catalysis for the oxidation step, and thus eliminate the need for oxidant
chemicals.
***
Various manufacturers have either single or multi-parameter sensors t
ha
t
can monitor distribution systemwater quality and directly communicate the data
to the Supervisory Control and Data Acquisition (SCADA) system, as shown in
Table 7-5. There are a multitude of options currently available that have been
recently reviewed (Hasan, 2005). Multi-parameter sensors are available in a
panel format that uses a side-stream to draw a sub-sample of water from
the distribution system or alternatively exist as sondes that can be installed
directly or indirectly within distribution system pipes. The former are large
conspicuous units, and, to avoid being tampered with, their use in the
distribution system would have to be limited to secure locations. Also, the
side-stream of water drawn from the distribution system into the multi-
parameter sensors requires

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
appropriate disposal (i.e., into a sanitary sewer) under normal operation. In con-
trast, the in-line sondes have the potential to be installed discretely within
the distribution system, which reduces their vulnerability to tampering but
may complicate routinecalibration.
It should be noted that the on-line monitoring technologies discussed a
b
o
v
e
are
currently cost-prohibitive and too complex for premise plumbing and service
lines. However, this may change with technological advances.
Consumer Complaints
A final type of monitoring that utilities may want to consider is
consumercomplaints monitoring. Consumers can detect off odors, changes in
taste or flavor, color, turbidity, and particulates resulting from system failures
(e.g., water main breaks, cross connections) as well as from system operations
(e.g., hydrant flushing, valve operations). While these untrained assessors are
subjective and unreliable from a laboratory testing point of view, they are
everywhere at all times in a distribution system and thereby serve as valuable
sources of information on potential water quality problems (Burlingame,
1999a,b; Laurer, 2005).
Collection and mapping of customer complaints should be done in the con-
text of a GIS-linked database to monitor conditions in the distribution system,
track operational issues, and determine the boundary of water quality events.
Critical to the functionality of using customer data as a monitoring tool is the
seamless integration and transfer of complaint data into operational databases so
that all functional departments (production, network, water quality, manage-
ment, communications, etc.) are instantly aware of any disturbances. Use of
mobile computers can effectively communicate and coordinate workforce re -
sources in the field to respond to and mitigate any events. Unfortunately, a sur-
vey in North America showed that while 84 percent of the responding utilities
have formal procedures in place for investigating customer complaints, only 61
percent had a customer complaint database (Deb et al., 2000). This was the case
even though these utilities often relied on customer notification for early detec-
tion of problems.
A three-year study demonstrated the benefits of certain practices at r
e
d
u
c
i
n
g
customer complaints related to water quality from a distribution
system i
n
Southern California (Wen et al., 2005). Problems with manganese, old
cast iron pipes, and rusty water were addressed by keeping good records of
customer complaints and developing a database to sort and track the complaints.
These data were then used to show the improvements made by the chosen
controls.
One drawback to customer complaints is that they are end-user in origin a
n
d
so
cannot distinguish between contamination originating within the customer’s
premise, the service line and local water main, a regional storage facility, or all
the way back to the treatment plant and the source water (Burlingame, 1999b).
Nonetheless, customer complaints may be the first line of detection of water
quality problem short of having an exhaustive and expensive monitoring plan.

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
Emerson
Model 1055
Comp II A
n
a
-
lyzer
Multi-
parameter,
customizable,
on-line
15,885
0–10
0–15
-1.4–1.4
0–200
0–20 ppm
0.001–200
0–14
Depolox 3 p
l
u
s
Free or totalchlo-
rine (or c
h
l
o
r
i
n
e
dioxide or ozone)
0–20
3,500
Wallace &
Tiernon
TABLE 7-5 Performance Specifications of Commercially A
v
a
i
l
a
b
l
e
Sensors
Manufacturer Dascore YSI
Hydrolab-
Hach
Analytical Technology,
Inc
Model Sixcense 6-series
(DS5X,
DS5,
MS5)
Series Q
4
5
Model A
15/B-2-1
Brief
description
Multi-
parameter,
on-line, free
chlorine or
chloramines
Multi- Multi-
parameter, param
eter,on-
line on-line
Multi-
paramet er,
customizable
on-line
Free c
h
l
o
-
rine or
chloramines
Cost (US $
)
Free c
h
lo
r
in
e(mg
per L)
Chloramines
(mg per L)
TOC
(mg per L)
ORP (
V
o
l
t
s
)
SC (mS p
e
r
cm)
DO (%)
9,700 15,000 15,000
0–5
0–20
< 10,000
0–2, 0–20 o
r
0–200
0–2, 0–20 o
r
0–200
3,000
0–2, 0
–
2
0
or 0
–
2
0
0
0–2, 0
–
2
0
or 0
–
2
0
0
-1.4–1.4
0.1–10
-0.999–
0.999
0–100
-0.999–
0.999
0–100
0–200 0–500 0–200*
Turbidity
(NTU)
0–3000
PH 2–12 0–14 0–14
-0.999–2.0
0–0.2 and
0-40
0–40 ppm
0.001–4 p
l
u
s
other wi
d
e
r
ranges
0–14
Temp (°C) 0–50 -5–45 -5–50
*LDO: luminescent DO measurements. For the bottom nine row s, the v
a
l
u
e
s
given are the
detection ranges of the sensors. SOURCE: Reprinted, w ith permission, fromBukhari and
LeChevallier (2006). © 2006 by American Water.

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
cal Technology, Inc
Model A 1
5
/
B
-
2
-
1
Free chlorine or chlor-
amines
3,000
0–2, 0–20 or 0
–
2
0
0
0–2, 0–20 or 0
–
2
0
0
Wallace &
Tiernon
Emerson Hach ProM inent In-Situ
Depolox 3
plus
Model 1
0
5
5
Comp II
Analyzer
WDM Panel o
r
Pipe S
o
n
d
e
™
TOC
Process
Analyzer
D1C &D2C Troll 9000
Free or t
o
t
a
l
chlorine (
o
r
chlorine d
i
o
x
-
ide
or o
z
o
n
e
)
Multi-
parameter,
customizable,
on-line
Multi-
paramet er,
on-line
Used w ith
WDM
Multi-
paramet er,
on-line
Multi-
paramet er,
on-line
3,500 15,885 12,000 18,000 7,000 11,200
0–20 0–10 0–4 (DPD)
0–0.5, 0–2,
0–10
0–15 0–5 (total)
0–0.5, 0–2,
0–10
< 5–20,000
-1.4–1.4 -1.5–1.5 -1.0–1.0 -1.4–1.4
0–200 0–100 0–200 0–200
0–20 ppm 0–20 ppm
0.1–10 or
0.1–20 ppm
0–20 ppm
0.001–200 1–100 1–2000
0–14 0–14 0–12 0–12
-5–50 0–100 -5–50

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
Temporal and Spatial Scales for Monitoring
Two important aspects of distribution system monitoring need to be recon-
sidered by many utilities. The first involves the timing of sampling,
which i
s
dictated largely by the method by which samples are taken. Currently
most routine water quality monitoring in a distribution system is carried out
through manual grab samples followed by analysis in the field or in the
laboratory. A grab sample is a single water sample collected at a specific point
in time. Essen- tially all distribution system regulatory monitoring uses this
method. For example, samples required under the SWTR are manually
collected at sites within the distribution system and manually tested for
disinfectant levels in the field. Samples taken to satisfy the requirements of
the TCR are also manually collected in the field and subsequently analyzed
in the laboratory. Manual sampling is labor intensive, and the number of
samples that can be collected is limited by personnel and analysis costs. In
addition, grab sampling can only show the water characteristics at the time the
samples were taken. Important events (e.g., night-time events) that occur
between samples are lost or unusual results may be dismissed (Premazzi and
Hargesheimer, 2002). Thus, grab sampling is of limited use as an alert
system to warn against potential contaminants that might pose a threat to
publichealth.
Moving from grab-sampling to real-time, on-line monitoring is essential for
more expeditious and accurate water quality assessment. This trend is b
e
i
n
g
reinforced because of increased emphasis among water utilities in consolidating
and automating data processing and control functions (Premazzi and Harge-
sheimer, 2002). On-line monitoring has the benefit of providing—in real time—
early warning of intentional or accidental contamination, and when fully de-
velop and deployed it could help water utilities take the appropriate actions to
safeguard public health. On-line monitoring requires a mechanism for moving
the sample water from the distribution system to an instrument, instrumentation
for analyzing the water, a mechanism for communicating the results, and a
means of assessing the results of the monitoring. As discussed above, relatively
inexpensive on-line water quality monitoring instruments are becoming
more prevalent (Byer and Carlson, 2005). Additionally, the instrumentation
must be periodically calibrated and maintained for quality control/quality
assurance to guarantee the reliability of generated data (i.e., minimize false
positives and negatives). The issue of minimizing false positives cannot be too
strongly em- phasized. Unless the individual analyzer false positive rate is kept
extremely low (e.g., < 1/1000 analyses), when a large number of analyzers are
deployed in a single systemthere is a high likelihood that most of the “hits” will
be the result of false positives (if the occurrence of actual true “hits” is rare).
Furthermore, for any given analyzer there is likely to be a trade-off between its
specificity (ability to detect specific contaminants) and its sensitivity (ability to
detect lower levels of contaminants). The statistics of deploying systems of
analyzers with specific false positive and negative rates must be considered
during the design of the monitoring program.

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
A second critical consideration is the location of sampling points. The T
C
R
requires the development of an approved distribution system sampling pl
an.
Unfortunately, there is a requirement for routine access for sample collection to
both the primary monitoring site and locations within five service
connections up-stream and down-stream of the primary site. This results in
decreased monitoring in residential areas of the system where access can be
limited. Although use of dedicated sampling stations can be used to overcome
accessibility problems in these areas, installation of sampling stations require
extra cost and can be prone to vandalism, freezing, and contamination.
Current practices for on-line monitors typically locate these devices on util-
ity-owned property where power, sewer, and telemetry to the SCADA s ystem
are available so that the results can be instantly communicated to a central op-
erations office for improved system management. These requirements typically
restrict monitoring locations to pump stations, storage tanks, well stations, and
perhaps government-owned buildings (all of which, coincidentally, may be high
priority areas for monitoring for security purposes because they represent points
of easy access). Rather than relying on such “convenience monitoring,” utilities
should consider employing more “risk-based monitoring” where sensors are
strategically located based on hydraulic flow, the population at risk, and sensi-
tive locations (e.g., hospitals, government installations, etc.). In particular, wa -
ter utilities should strive to sample areas where water quality may be more prone
to intensive deterioration. These areas may include (depending on the system),
areas of low flow, areas subject to frequent flow reversal, areas achieving vari-
able blends of waters from different plants, and areas of old and/or deteriorating
pipe. Utilities need the technology, regulatory support, and public understand-
ing to customize their routine water quality monitoring programs to accomplish
these more risk-based goals.
Recent research has examined algorithms for placing sensors
based o
n
population exposed and time and flow for contaminant detection
(Berry et al., 2004). Often, there are trade-offs for one approach versus
another, such that optimization programs are needed to choose a best overall
strategy. EPA’s Threat Ensemble Vulnerability Assessment (TEVA)
program is developing an add-on tool in EPANET to allow water utilities to
select the optimal number of sensors and identify strategic locations for
installation of on-line sensors to maximize public health protection
(http://www.epa.gov/NHSRC/news
/news111505b.htm). Conceptually, given a spectrum of potential threats or vul-
nerabilities to a distribution system (from either unintentional or i
ntenti
onalevents)
it is possible to determine the optimal locations to site a given number o
fdetectors
such that the likelihood of detecting such events is maximized. T
he
computational framework uses Monte Carlo simulations to vary parameters,
such as the quantity or concentration of contaminant, location of injection, dura-
tion (or rate) of injection and the probability of ingesting an infectious or toxic
dose of these selected contaminants, to generate threat ensembles (collections of
many threat scenarios). These threat ensembles are collectively analyzed to es -
timate health impact statistics,including mean infections or mean fatalities. The

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
public health benefits of no sensors in the distribution systemhave been com-
pared with both utility convenience monitoring and TEVA designs (Table 7-6).
These computer simulations support the use of on-line sensors in providing an
early indication of drinking water contamination events. However, achieving the
maximum benefits from on-line installation of sensors requires optimization of
the number of sensors in the distribution system. Clearly this number is
likely to be system specific and will vary depending upon the distribution sys -
tem network, the number of service connections, the type of service connections
(i.e., primarily residential or commercial), the size of the population being
served, and the length of the distribution system pipes. Modeling tools
like those being developed by the TEVA program and others (Lee et al.,
1991; Murray et al., 2004; Ostfeld, 2004; Ostfeld and Salomons, 2004; Uber
et al., 2004a,b) are still under evaluation and will likely undergo significant
refinement and validation before finalization. However, until these have been
adequately tested, sensor deployment at locations serving the highest population
densities may be an appropriate initial strategy. Of course, practical
considerations, such as access to power, communication lines, waste disposal
(from samplers), and equipment security may limit where sampling can be
located. Nonetheless, TEVA analyses could be used to delineate which
locations, amongst those identified as feasible, would offer the highest
protection.
Data Analysis and Reporting
Although on-line monitors can provide a continuous stream of information,
the data needs to be analyzed, reported, and stored at some prescribed f
r
e
-
quency. For example, a single on-line multi-parameter sensor measuring six
water quality parameters every 15 minutes on a 24-hour basis will lead to the
TABLE 7-6 Public health benefits provided by various sensor location strategies
Sensor Design Health Impacts(Fatalities)
Biological Attack
Health Impacts (Fatalities)
Chemical Attack
Median Mean Max Median Mean Max
No sensors 980 1,544 22,287 158 139 284
Utility c
o
n
v
e
n
i
e
n
c
e 671 1,015 5,107 110 113 284
monitoring (31.5%) (34.3%) (77.1%) (30.0%) (18.70%) (0%)
TEVA 227 350 2,730 67 78 229
(76.8%) (77.3%) (87.6%) (57.6%) (43.0%) (19.0%)
*Values in parenthesis are % public health protection relativ e to the sy stem with no sensors.
SOURCE: Based on analy sis by R. Janke and R. Murray , EPA National Homeland S
e
c
u
r
i
tyResearch
Center, as cited in Bukhari and LeChev allier (2006). Reprinted, with permission, f rom Bukhari and
LeChev allier (2006). © 2006 by American Water.

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
generation of 4,032 data points per week. Thus, timely management and i
nter-
pretation of large quantities of data are imperative to the efficient utilization of
an on-line monitoring system. It can quickly become an onerous task without
the aid of interpretative software. It would be desirable for an automated data
analysis package to be capable of not only capturing data from on-line monitor-
ing devices, but also able to (1) perform automated trend analysis that would
compare real-time data with baseline historic data to define and characterize
anomalies and (2) allow user-defined and programmable triggers with auto-
mated notification by means of alarms (on cell phones, pagers, or via e-mail).
Presently there are only a limited number of options for predicative data man -
agement tools,making this an area ripe for research and innovation.
Advanced Monitoring for Contaminant Identification
As discussed above, real-time monitoring is currently not useful for identi-
fying specific contaminants in distribution systems; rather, they determine base-
line water quality conditions and look for deviations from historical trends
(Hrudey and Rizak, 2004; Watson et al., 2004). On -line sensors that could de-
tect a range of chemical or biological parameters are a number of years away
from commercial development or utility utilization. Additionally, there is a
need to advance the technology for parameters that cannot be measured in real-
time and on-line. Advances in microfluidics, robotics, and miniaturized compo-
nents are lowering costs and may have the potential to perform analyses for
chemical and microbiological contaminants that just a few years ago
required sophisticated and expensive laboratory equipment. Ultimately, a
multi-tiered monitoring system is envisioned where on-line water quality
monitoring sensors would detect a deviation in baseline water quality and draw
a side-stream sample that would be automatically analyzed using an advanced
“lab-on-a-chip” that can detect multiple contaminants. On-site microprocessors
would analyze the results and send an alarm to the centralized SCADA
system. While a mobile analyst is dispatched to verify the on-line
monitoring results, the centralized event management software is checking
other on-line monitors, customer service, and operational databases for any
other anomalies to determine potential causes and a range of corrective actions.
Although the above description of distribution system monitoring is
years away from implementation, some water utilities have begun developing
elements of what will likely evolve into the envisioned comprehensive
monitoring program. For example, the Arizona Department of Environmental
Quality has partnered with the Tucson Water Department, the University of
Arizona, and several Pima County agencies, businesses, and organizations to
provide citizens with on-line information about drinking water quality. The
effort was made possible by an Environmental Monitoring for Public Access
and Community Tracking (EMPACT) grant from EPA. System monitoring
focuses on three components: water quality parameters that are common to
all water systems;

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
specific water quality parameters that focus on public health in water a
n
d
wastewater
treatment; and the volume of water flowing through the cycle. The monitored
parameters include pH, conductivity, temperature, hardness, sodium, and total
dissolved solids. Additional parameters important for public health include
coliform bacteria, disinfectant residuals, total trihalomethanes, fluoride, and
nitrate. The overall objectives of the project are (http://www.ci.tucson.az.us/
water/water_quality.htm):
 Increase the amount of water quality testing by continuous on-line sam
-
pling
 Improve the access to water quality data in the potable distribution sys-
tem
 Provide information for customers by identifying specific constituen-
cies and methods to individualize data by location
 Create a context for understanding water resources data, thus r
em
ovi
ng
misperceptions
 Serve as a source of reliable, authoritative information on fast-breaking
water quality issues
The automatic monitoring stations are currently running and continuously updat-
ing water quality data on a map-based website; this program is a model for other
utilities.
How to Interpret Data and Respond to Monitoring Data
Given a stream of data from a monitoring program, a critical task is to d
e
-
termine whether the results indicate an “event” or “problem” and if so, h
o
w
utilities should respond. The occurrence of alterations may be ascertained b
y
formal
statistical tests (Ortiz-Estarelles et al., 2001), of which there are several types.
The presence of outliers (from historical past behavior) can be ascertained
using statistical quality control methods (Egan and Morgan, 1998; Lalor and
Zhang, 2001). There can be tests of trends to determine if a systematic drift in
water quality has occurred. The underlying concept is to assess water quality
using a statistical process control concept. To do this, a utility needs to assess
what the “normal” water quality, and its fluctuations, might be. Furthermore,
water quality data streams require site-specific “tuning” of software in order to
detect unusual deviations, and there has been reluctance on the part of some
vendors to disclose their tuning and detection algorithms.
There are other fields in which similar problems to the one outlined above
have been experienced; it is possible that the methods used in their
solution could be adapted. These include thefollowing:
 Identification of financial enterprises that are on the verge of diffi
cul
ty
(Booth et al., 1989)

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
 Assessment of machine malfunctioning (Javadpoourand Knapp,2003)
 Detection of outliers in chemical (Egan and Morgan, 1998) and g
e
o
-
chemical (Lalor and Zhang,2001) data
 Structural health monitoring (Omenzetter et al., 2004)
 Detection of computer intrusion or other unwarranted use of
computerresources (Lazarevic et al., 2003)
There are several broad approaches to the problem of identifying
unusual observations in time series of multivariate data that have been
outlined in the literature. These include (1) distance and generalized distance
approaches, (2) regression approaches, and (3) neural network approaches. In
distance and generalized distance approaches, an observation is regarded as
unusual if it is away from the typical population of observations. There are a
number of design alternatives for this strategy including the following:
 Euclidean versus otherdistance scales
 Transformation of variables prior to evaluation (including rank t
r
a
ns-
formations and extraction of principal components)
 Use of direct versus cross-validation distance
 Choice of criteria to call an observation unusual(false positive a
n
d
negative rates that are deemed acceptable)
 Use of “de-trending” or otherpreprocessing steps to eliminate n
o
n
-
stationary components of the data.
These methods have been reviewed and discussed by Egan and Morgan (1998)
in the context of interpreting analytical chemistry data. These authors discuss
use of trimming, distance measurements, use of sub-sampling and test various
approaches against sample data from the literature, and specifically recommend
against the use of ordinary distance measurements and Mahalanobis distance
measurements (i.e., distance scaled by the sample variance/covariance matrix) in
detecting outliers. Future research is needed to assess the applicability, sensitiv-
ity, and selectivity of the various numerical approaches applied to various com-
binations of measurements which might be taken in a distribution system.
Utilities deploying advanced monitoring in their distribution systems should
do so with a specific response plan developed in advance. There are limited
options for responses at the disposal of utilities, including boil-water, do-not-
consume, or do-not-use notices, that can be applied for particular sections of a
system or system-wide. The response to a detection event froma monitoring
network carries risks associated with both false positives and false negatives. If
an alarm signal is triggered when an actual systemdeterioration has not occurred
(a false positive) and results in an action such as a “do-not-use” or a “boil-
water” notice, there may be consequences associated with unavailability of sup-
ply for fire fighting and economic impacts on individuals and businesses. If an
alarm signal fails to result when an actual deterioration has occurred (a false
negative), then there is a failure to detect and respond to an event and thereby

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
reduce its impact. The adverse consequences from such false negatives a
n
d
false
positives need to be considered when determining the action levels at
which management will respond to monitoring data.
Implementation of an Enhanced, Process-Oriented Monitoring Program
The monitoring program discussed above, which includes pressure a
n
d
chlorine
residual monitoring as well as other parameters as needed, all continuously
monitored and with deviations followed up on in a timely manner, repre - sents a
considerable step up from what many water utilities may already be doing.
This will require an increase in training, supervision, maintenance, docu-
mentation, and management. Training will be needed for using new on-line
technology and for its data management and interpretation; for using GIS, hy-
draulic modeling, and other data integration tools; and for identifying deviations
in monitoring data and in responding to the deviations to determine if the asso-
ciated processes are being adequately controlled. New tools, such as on-line
water quality analyzers, will need quality assurance and quality control (calibra -
tion, maintenance, data approval) in order to provide reliable data.
Software will be needed for the data management and integration.
Documentation is critical to providing feedback for the whole program; not only
must data be recorded but associated information is needed (on maintenance,
accuracy, and quality control) to provide an appropriate level of certainty.
The monitoring program outlined above will require comprehensive
management, with lines of responsibility clearly outlined and funding and
staffing adequately provided for. Furthermore, the monitoring program will have
to evolve as the distribution system is adjusted and expanded to meet changing
demands overtime.
At present, a program of real-time monitoring and the use of the
advanced technologies discussed in this section is likely to be feasible only for
the largest and most sophisticated utilities and not to smaller (and even non-
community) systems, which unfortunately are where a large fraction of disease
outbreaks are reported. This mismatch highlights the need for further
technical development of sensors alongside alternative strategies for protecting
water quality (discussed extensively in Chapter 4, 5, and 6) among smaller
utilities.
DISTRIBUTION SYSTEM MODELING
Water distribution network (mathematical) models have become
increasingly accepted within the water industry as a viable mechanism for
simulating the behavior of water distribution systems. They are intended to
replicate the behavior of an actual or proposed system under various
demand loading and operating conditions. Their purpose is to support the
decision-making processes in various utility management applications
including planning, design, operation, and water quality improvement of
water distribution systems:

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
 Planning applications include capital investment decisions to i
denti
fy and
prioritize capital improvements to meet projected growth or to replace aging
infrastructure; development of water system master plans to schedule, stage,
locate, and size new facilities to support projected growth as well as to analyze
the interconnection of separate systems for emergencies; infrastructure rehabili-
tation and replacement to identify and prioritize water mains that need to
cleaned, lined, paralleled (duplicated), or replaced; and water conservation stud-
ies to maximize the use of existing supply sources and evaluate sound conserva-
tion measures to reduce overall water consumption and capital improvement
costs.
 Design applications include estimation of fire protection capacity (
e
.
g.,
available flow at 20 psi) to verify compliance with fire protection
standards; pressure zone management to keep supply pressures within
acceptable ranges in regions with significant differences in elevation;
determination of the location and size (or capacity) of new water mains, storage
facilities, and pump stations to keep pace with projected growth; and hydraulic
transient analysis to identify weak spots and select the optimal combination of
surge protection or suppres-sion devices to ensure safe systemoperation.
 Systems operations include energy management applications to o
p
t
i
-
mize storage-pumping trade-off and minimize energy costs; emergency planning
to develop an effective emergency response program to reduce or eliminate the
damage or impact of unplanned outages at wells, pump stations, pipes,
storage tanks, and treatment plants; and daily operational and management
decisions to optimize use of existing facilities and train systemoperators.
 Water quality improvement applications include calculation of w
a
t
e
r
retention time for tanks, travel time in pipes, and the spatial and temporal distri-
bution of water quality throughout the systemto predict locations of poor water
quality and evaluate improvement measures such as installation of rechlorina-
tion facilities and improving reservoir turnover; locating permanent water qual-
ity monitoring stations for compliance with federal regulations; and design and
implementation of unidirectional flushing programs. Other applications include
area isolation during repairs, water loss calculation, leakage minimization, statis -
tical and probabilistic analyses,and, more recently, water security assessment.
Early models simulated hydraulic behavior only and were steady stat
e
(static) in nature. But with the advent of more powerful computers and numeri-
cal algorithms, extended period simulation (dynamic) models were developed
(e.g., Wood, 1980) to simulate behavior under time varying demand and opera-
tional conditions, which is necessary because systemdemands and consequently
the flows in the network vary over the course of a day. These models have be-
come ubiquitous within the water industry and are an integral part of most water
system design, master planning, and fire flow analyses. In the early 1980s in -
vestigators began introducing the concept of water quality modeling (Clark and
Males, 1986; Grayman, et al., 1988; Clark and Coyle, 1990), and now most wa-
ter distribution systemmodels routinely incorporate sophisticated water quality

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
simulation capability. In addition to hydraulic and water quality simulation,
many distribution network models are capable of analyzing water hammer
(surge/transient) and tank and reservoir mixing characteristics. Currently avail-
able water distribution network models have become very sophisticated and
many incorporate Computer Aided Drafting and Design (CADD) and GIS capa-
bility as well as interfacing with SCADA and Asset Management Systems
(AMS).
This section discusses the basic principles underlying routine hydraulic a
n
d
water quality modeling in drinking water distribution systems and presents new
developments. In addition, integration of network modeling and optimization
with a range of information management systems into an effective decision sup-
port and utility management and protection systemis presented.
Hydraulic Modeling
Hydraulic models simulate flows and pressures throughout the water distri-
bution system and can be divided into four broad categories (Wood et al.,
2005a):
(1) Steady State Theory: The basic network hydraulic approach, applica-
ble to time-invariant conditions, solves the conservation of mass (at each n
o
d
e
) and
energy (around each loop) equilibrium expressions using an i
terati
ve
scheme (e.g., Newton-Raphson) based on known (static) demand loading and
operatingconditions.
(2) Extended Period Simulation (EPS): The second approach, applicable
to very slow transients, is called extended period simulation (EPS) or quas
i-steady
theory, and involves solving a sequence of steady-state solutions linked by an
integration scheme for the differential equation describing the storage tank
dynamics. Both inertial and elastic effects are neglected. These models have
become ubiquitous within the water industry and are an integral part of
most water system design, master planning, and fire -flow assessment studies.
They also provide flow information used in distribution system water quality
models.
(3) Rigid Water Column Theory: Another category of unsteady flow i
s
suitable for faster (but still relatively slow) transients and is called rigid water
column theory (lumped parameter approach). It considers gradually varied flow
and slow moving transients under the assumption that water acts as a rigid-
column and elastic properties of the pipe walls are of no consequence. In this
approach, the inertia of the fluid in a particular pipe is treated as lumped instead
of continuously distributed.
(4) Waterhammer (Surge) Theory: The last category of unsteady f
l
o
w
applicable to rapid transients is called elastic or waterhammer theory
(distrib-

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
uted parameter approach) and takes into account the elasticity of both the fl
ui
d
and the pipe walls in the calculations. It represents situations with more rapi
d and
sudden changes in flow velocity (e.g., rapid valve closure, pump trip) that
require consideration of liquid compressibility and pipe wall elasticity.
The last three hydraulic modeling categories are known as unsteady (or d
y
-
namic) flow analysis. These models can be effectively used to estimate intru-
sion potential, identify susceptible regions in the distribution system that are of
greatest concern for vulnerability to objectionable (low or negative) pressure
surges, and evaluate how they may be avoided and/or controlled (Boulos et al.,
2005).
Rigorous optimization approaches have been developed and applied to a
full range of problems associated with water distribution systems (Boulos et al.,
2006). Applications include optimizing network model calibration, satellite
treatment (booster disinfection station location and operation), data collection
and sampling/monitoring, as well as pump and storage tank operations to mini-
mize energy cost, and valve operation for pressure management and leakage
reduction. The optimization methods applied are common between problems
and include linear, nonlinear and dynamic programming, and stochastic search
procedures.
Water Quality Modeling
Water quality models utilize the flow and velocity information generated b
y
the hydraulic models to predict the temporal and spatial variability of w
a
t
e
r
quality within the distribution system. They can be used to simulate water qual
-ity
concentrations and water age, and they can perform source tracing t
h
r
o
u
g
hthe
distribution system. As with hydraulic models, water quality modeling has
evolved from the initial development of steady-state models (Wood, 1980;
Males et al., 1985; Clark and Males, 1986; Males et al., 1988) to more dynamic
models (Liou and Kroon, 1986; Hart et al., 1986; Clark et al., 1988; Grayman et
al., 1988). Dynamic water quality models are predicated on extended period
simulation quasi-steady network hydraulics, and they solve the equations for
nodal mixing and advective transport in pipes to compute the spatial and tempo-
ral variation in water quality parameters. Solution methods for dynamic models
can be classified as either Eulerian or Lagrangian (Rossman and Boulos, 1996;
Clark and Grayman, 1998; Panguluri et al., 2005b). Eulerian methods consider
fixed grids or cells and move water to the grid locations or through the cells to
represent the movement of a constituent in a pipe. Chemical reactions are in-
cluded during transport. Lagrangian methods track locations of discrete changes
in water quality known as fronts. Front locations are updated at a fixed
time step or when a front reaches a junction node. Longitudinal dispersion
is neglected and complete mixing at the junction nodes is assumed.

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
In a distribution system water quality analysis, the constituents of
concern may be conservative (e.g., inert) or non-conservative (e.g., reactive).
Conserva- tive substances are useful as tracers that are purposely injected,
monitored, and modeled to improve model calibration with measured travel
times and dilution effects. Non-conservative constituents, such as bacteria,
disinfection byproducts, or chlorine, undergo reactions that are dealt with
in most water quality models via simple reaction kinetics (Panguluri et al.,
2005b).
During the 1990s, advancements focused more on making existing model
s
user friendly and less on improving predictive capabilities. For example, t
h
e
EPA
public domain EPANET model (Rossman et al., 1994; Rossman, 2000)
greatly facilitated the easy application of existing water quality models to mu -
nicipal drinking water distribution systems. In early 2000s, research initiated by
Zierolf et al. (1998) and Shang et al. (2002) focused on development and appli-
cation of control theory-based methods. While distribution system water quality
models are forward methods (i.e., they begin at a source and track forward in
time and space to determine where a constituent is going), control theory begins
at a location of interest and tracks backward in time and space to identify the
constituent source. This approach has at least two important uses. The first is
developing injection policies; if the disinfection level is unacceptable at a certain
location, control theory will determine the relative contributions fromalternative
booster stations in one analysis rather than performing a series of forward tracer
runs. Another application is identifying the potential sources of contamination
detected at a downstream monitor. A contaminant that is detected at a given
location may be supplied by a number of inlet points and times, and control the-
ory can identify the range of locations and the pipes and nodes that contribute to
flow at the monitor. Because of uncertainty due to changing demands and pump
and tank operations, these analyses generate a significant amount of data. For-
tunately, control theory model analysis can be done on-line, which facilitates
data collection and information storage. After the potential sources have been
identified, a forward model can be applied to determine the extent of contamina-
tion for containment purposes and theneed for flushing.
With the recent concern over water distribution systems as part of the n
a
-
tion’s critical infrastructure, research into water quality modeling has be
c
om
emore
active, with the intent of developing greater predictive capabilities. F
orexample,
EPA has extended EPANET to allow general multi-species reactions. However,
the code lacks a user interface at present, it is intended for research purposes,
and it is in the beta-testing phase (http://www.epa.gov/nhsrc/pubs/
tbEPANet051106.pdf). Other recent developments include the application of
transient analysis software (Boulos et al., 2005) and optimization tools for cali-
bration, design, and operational purposes (Berry et al., 2004; Uber et al., 2004a;
Murray et al., 2004; Ostfeld, 2004; Ostfeld and Salomons, 2004). An emerging
area of research is the incorporation of stochastic analysis to water quality mod -
eling (Buchberger et al., 2003). More complex kinetics can be used to describe
multi-component interactions (relating transformation rates to the concentration
of other constituents)and are known as multi-component or multi-species mod-

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
els. Clark (1998), Clark et al. (2001), and Clark and Sivaganesan (1998, 2002)
have furthered the development of multi-species and competitive reaction mod -
els that could be included in general purpose algorithms such as EPANET. Ex-
amples of two promising areas for future research are illustrated by Uber et al.
(2004a), which has extended EPANET to allow for modeling the fate and trans -
port of multiple interacting chemical and biological components, and Uber et al.
(2004b), which has developed algorithms for optimizing the location of water
quality sensors in drinking water distribution systems.
Flushing Models
The last two years have seen the introduction of computerized u
n
i
d
i
r
e
c
-
tional
flushing models of water distribution systems (Boulos et al., 2006). Uni-
directional flushing models utilize the flow and velocity information generated
by the hydraulic models and make use of graph-theoretical algorithms to deter-
mine the sequences of fire hydrants and water main valves that should be ma -
nipulated to create a one way flow in the water mains while avoiding excessive
pressure drops (e.g., below 20 psi) and maintaining the desired level of hydrau-
lic performance in the distribution system. These models also compute the
minimum flushing time, total flushing volume and pipe length, and the flushing
velocity of every pipe in the sequence.
Modeling of Storage Facilities
The hydraulics and mixing of waters within storage facilities must be prop-
erly understood to accurately represent the constituent reactions and the effect of
tanks on system water quality. In addition, understanding the mixing character-
istics in storage facilities is useful in assessing the likely impacts of an injected
contaminant. Tank models simulate both aging and mixing phenomena within
distribution system tanks and reservoirs. The most complex is
computational fluid dynamic or hydrodynamic modeling that includes a detailed
physical tank description and divides a tank into a mesh of small discrete
volumes known as finite elements (Grayman and Arnold, 2003). The basic
governing laws of conservation of mass, energy, and momentum are written in
partial differential form for each element to describe the flow patterns and the
distribution of substances through the tank. The remaining approaches are
known as systems models. Systems models are classified by the spatial
representation in the tank. The sim- plest model, a continuous stirred tank
reactor, considers the tank as a single unit and assumes complete mixing of
water within the tank. The next level of detail represents the water in the tank in
layers assuming plug flow. The tank is parti- tioned into several compartments
to represent the flow patterns and mixing zones. These are known as multi-
compartment tank models (Mau et al., 1995; Clark et al., 1996).

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
***
All network models are approximate representations of distribution systems,
and many sources of error exist that can hinder their ability to accurately sim
u-
late actual system behavior. Sources of error can range from measurement and
typographical errors to errors derived from system maps or introduced by the
skeletal representation of the network as well as uncertainties in some system
parameters and boundary (e.g., loading and operating) conditions. Network
models should be properly calibrated and validated so that a level of confidence
in their predictive capabilities can be established. Box 7-1 discusses these ac-
tivities in greater detail.
BOX 7-1
Calibrating and Validating Network Models
A water distribution netw ork model must be properly c
a
l
i
b
r
a
t
e
d before it can be used to
support planning, design, operation, or water quality improvement decisions. Calibration
establishes the accuracy and credibility of the network model so that its predictions can be
interpreted w ith confidence. It is the process of fine-tuning (adjusting) network model
parameters so that the simulated hydraulic andwater quality results sufficiently mirror field
observations. If the field data and model results are reasonably close, the model is consid-
ered calibrated. The objective is to reduce the uncertainty in the model parameters to ac-
curately reproduce actual“real-world” systembehavior.
To be calibrated, the network model must accurately simulate p
r
e
s
s
u
r
e
, flow, tank level, and
chlorine residual values within an acceptable tolerance for a range of specified time
horizons. Hydraulic parameters that are typically adjusted include pipe roughness factors,
minor (local) loss coefficients, isolation valve status, control valve settings, pump curves,
base demands, and demand patterns. For w ater quality models, the parameters include
reaction rate coefficients, source quality, and initial conditions. The calibration tolerance
refers to the difference between model simulated and actual field values. The smaller
the tolerance the greater the accuracy of model predictions. Calibration can be
performed to a single time frame such as maximum hour or dynamically such as maximum
day for an extended period simulation (EPS). The more calibration time frames, the more
accurate the model predictions will be. Common practice is to calibrate the network model
first for maximum-hour and minimum-hour static conditions and then in an EPS mode for
maximum day. The netw ork model is first calibrated for hydraulic parameters and the water
quality parameters are subsequently adjusted. Thus, if the hydraulic model is not properly
calibrated, resulting in inaccurate flow and velocity estimates, the water quality model
will not perform correctly. Water quality simulations require a dynamically calibrated (EPS)
model. Netw ork model parameters can be adjusted manually using an iterative trial-and-
evaluation approach or automatically using optimization techniques until the desired degree
of accuracy is attained (Panguluri et al., 2005b; Boulos et al., 2006).
Although automated calibration methods are becoming more r
e
a
d
i
l
yavailable, manual
calibration still remains the predominant methodology. How ever, since there is a vast
number of combinations of parameter values that can be considered for adjustment, man-
ual evaluation of all options through trial-and-error is unlikely to be practically feasible
or manageable, and even know ledgeable modelers often fail to obtain good results. As
a result, model calibration has generally been neglected or done haphazardly.
continues

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
SOURCE: Adapted fromAWWA ECAC, (1999).
BOX 7-1 Continued
There are currently no universally accepted standards for c
a
l
i
b
r
a
t
i
n
g
w ater distribution network
models. The extent of calibration w ill normally depend on the intended use of the model. A
greater degree of calibration will be required for models that are used for detailed
analyses, such as design, operations, and water quality modeling, than for models used for
more general planning purposes (e.g., master panning). The AWWA Engineering Com-
puter Applications Committee (AWWA ECAC, 1999) has proposed a draft set of calibration
guidelines for modeling based on intended use. These performance criteria were not in-
tended as true calibration standards, but can serve as a good starting point for illustrating
the extent of calibration needed for various modeling applications. These calibration crite-
ria are summarized in Table 7-7.
Netw ork model validation follows the calibration process and m
a
k
e
s use of an inde-
pendent field data set for use in verifying that the model is w ell calibrated. The model must
first be calibrated using one or more sets of field data and then validated w ith an independ-
ent set of field data. The degree of confidence in the model increases with the number of
independent data sets w ith which it is validated. Tracer studies can also be used to vali-
date network models. These studies consist basically of measuring the concentration of a
tracer over time (e.g., using on-line monitors and grab samples) at various locations
throughout the distribution system and comparing observed values w ith model predictions.
The most commonly used tracers are fluoride, calcium chloride, and sodium chloride. The
use of tracer studies greatly enhances the ability of network models to accurately estimate
w ater age and traveltimes in the system.
TABLE 7-7 Draft Calibration Criteria for Modeling
Intended
Use
Level of
Detail
Type of
Simula-
tion
Number of
Pressure
Readings
Accuracy of
Pressure
Readings
Number of
Flow Readings
Accuracy of
Flow Readings
Planning Low Steady o
r
EPS
10% o
f
Nodes
±5 psi f
o
r
100% Read-
ings
1% of Pipes ± 10%
Design M
o
d
e
r
a
te to
High
Operations Low t
o
High
Steady o
r
EPS
Steady o
r
EPS
5% – 2% o
f
Nodes
10% –2%
of Nodes
±2 psi for 9
0
%
Readings
±2 psi for 9
0
%
Readings
3% of Pipes ± 5%
2% of Pipes ± 5%
Water
Quality
High EPS 2% o
f
Nodes
±3 psi for 7
0
%
Readings
5% of Pipes ± 2%
The efficacy of calibration and calibration techniques is highly d
e
p
e
ndent on the quality of
the calibration data available and the quality of the constructed network model (e.g.,
skeletal representation of the network, node elevation, geometric anomalies). Poorly col-
lected field data (e.g., from poorly calibrated measuring equipment) and poorly defined
network models will result in inadequate calibrations and unreliable model predictions, and
w ould defeat the w hole purpose of the calibration process.

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
DATA INTEGRATION
Data and information on the additional parameters of distribution system i
n
-
tegrity discussed above and in previous chapters—water pressure and f
l
ow,
valve operations, main breaks, customer complaints, condition assessment,
inventory, and new water quality data—must be integrated in order to make
informed decisions, preferably using models and GIS tools. For example, if
a contamination event were to occur in a water distribution system, depending
on the hydraulic design and operational conditions, much of the distribution
system could be impacted, potentially affecting a large percentage of the
population served. It would be very difficult to track the spread of the
contaminant, find its originating source, and understand its impact based on
monitoring information alone. Rather, an understanding of the system’s
hydraulic behavior as well as a proper visualization of all system facilities (e.g.,
pumps, tanks, isolation valves) and rapid access to customer data and
SCADA information are required. GIS and modeling can be used to predict the
movement of contaminated water in the system, locate the appropriate facilities
that need to be closed manually or via the SCADA system for event
containment, identify populations at risk and report customer notification
information, compute affected water volumes that need to be purged, and help
develop an effective flushing program (e.g., which sequence of hydrants to open
and how long to keep them open).
In the past few years, advances in infrastructure management technology
have been occurring at an accelerated pace, with potential significant
benefits for the water works industry. The development of GIS is greatly
expanding the applications of water distribution network models. Because of the
spatial nature of water distribution systems, many aspects of managing these
systems consist of using, analyzing, and displaying geospatial data, which
includes the geographic location and characteristics of various water system
facilities, including pipes, pumps, storage tanks, reservoirs, and valves. Using
GIS, a water valve can be identified by its geographical location in the system
and its characteristics, such as valve type, size, manufacturer, pressure or flow
setting, loss coefficient, year of installation, condition, maintenance records,
opening direction, and location and distance to operating nut. Similarly, a pipe
can be described by its route, length, diameter, material, installation date, lining,
wall thickness, pressure class, service connections, ground surface type, street
identifier, parallel pipe indicator, cost of installation, condition, leakage and
burst records, fire service capacity, physical samples (e.g., observations of
tuberculation or corrosion exhibited by the pipe samples after a breakage
event), and taste and odor complaints. Indeed, a GIS can encompass all of the
data collected during asset management (see Chapter 4) such as the
condition of pertinent pipes, pumps, and valves as well as other water system
facility characteristics that are used for model construction and emergency
response (e.g., how many turns are required to close a valve and in which
direction).
A GIS is able to store and maintain (keep up-to-date) these spatial d
a
t
a
while
allowing easier access to the data, flexibility in data sharing and modeling,

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
and a substantial decrease in data storage and redundancy requirements. G
I
S
enables the model input and output data to be developed and displayed in
graphical form, the generation of accurate and detailed facility data, verification
of data integrity, and the visualization and cartographic analysis provided
by system maps. It can be effectively used to calculate and allocate
consumption data (e.g., from land use/population data or geocoding), help
identify (representative) water quality sampling sites, and quantify customer
exposure to a specific event. A GIS can also assemble data on the physical
characteristics of service to the customer as well as customer billing data,
including customer name, street address, contact information, service line
diameter, location, installation date, material, tap number, and meter number
(Cesario, 1995). Customers’ contact information can prove very useful for
alert notification during a water quality emergency event. In addition, the
GIS can store pump information and all pertinent electricity charges as well as
demand and energy consumption data. This information can feed an
optimization model to assist in developing improved daily pump scheduling
policies that meet desired hydraulic and water quality (minimize water age)
objectives while maximizing energy savings, as well as sound operational
strategies for using alternative water supplies during a contamination event.
The SCADA system, which compiles real-time and historical operational
data for all remote facility sites, is critical to making informed decisions during a
contamination event. SCADA information is useful in defining boundary (e.g.,
tank and reservoir water levels, pump status, valve settings) and loading (e.g.,
total zone and system demands) conditions for the network model as well
as real-time measurement data (e.g., pressure and flow measurements) for
network model calibration and for identifying water losses during main breaks
(e.g., un- explained low pressure reading, excessive pump flow). The SCADA
system can effectively control the contamination spread by isolating the
contaminated areas and associated facilities by shutting off all critical in-line
isolation valves. The isolation can be carried out automatically either locally
(i.e., at the valves) or remotely (i.e., the control room).
The Internet and the World Wide Web are also rapidly evolving to the bene-
fit of water supply operations. Among their many improvements are easier a
n
d
greater accessibility, efficient distribution, effective administration, and c
r
o
s
s
-
platform flexibility (Molenaar and Songer, 2001). A web-based interface,which is
becoming common in standard GIS platforms, can enable the rapid depl
oy- ment
of critical GIS data and modeling results over the Internet. This could f
a
-cilitate
the sharing of critical information with federal, state, and local em
er- gency
response and regulatory agencies during contamination events. Such data may
include valve and hydrant locations for firefighting as well as information
needed to isolate and flush accidentally released or intentionally introduced con-
taminants in the distribution systemand identify all affected individuals.
These various infrastructure management systems are highly complimentary
applications that, taken together with various distribution system models, consti-
tute a decision support systemfor a water utility. The ability to seamlessly ex-

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
change data between these systems and models is critical to improving w
a
t
e
r
distribution system operation, management, protection, and emergency planning
activities. Table 7-8 lists the key components of the decision support s
y
s
t
e
malong
with their response roles.
Managing and protecting distribution systems from contamination threat
s and
a wide range of emergency situations will require the use of a comprehen-
sive decision support systemthat integrates modeling applications with the vari-
ous infrastructure management systems. The term “integration” can refer to
a
TABLE 7-8 Systems and their Response Roles
System Response Role
On-line monitoring  Identify event (location andtype)
 Alarmnotification
SCADA  Alarmnotification
 Event isolation (operational control)
 Event reporting and archiving
 Resumption of normal operation after removalof threat
Hydraulic Model  Provide system-wide flows at moment of event
 Locate valves to isolate event
 Compute required purge volumes
 Compute available fire fighting capacity
 Check w hether normaloperations can continue
 Determine w ater rerouting scheme (use of a
l
t
e
r
n
a
t
i
v
e
watersupplies)
Unidirectional
Flushing Model
 Determine flushing sequences (opening of h
y
d
r
a
n
tsand closing
of valves) for proper decontamination
Water Quality Model  Determine extent of contamination
 Provide system-wide water quality at moment of event
 Narrow event region
 Identify grab sample locations
 Track contaminant to originating source(s)
 Develop re-chlorination plan
 Predict future event region (as contaminant moves)
Grab Sample  Verify event (degree of confidence)and associated emer-
gency level
GIS  Visualize all systemfacilities
 Map modeling results
 Coordinate response units
 Identify population at risk
 Report customer alert /notification information
Web Portal  Instant access and sharing of critical information
 Monitor event response progressin real-time
 Provide means of centralmessaging exchange

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
wide range of capabilities related to use of network models in conjunction w
i
t
h
various infrastructure management systems. With respect to linking a GIS
to a
network model, integration refers specifically to a method that combines the
GIS and modeling functions into one complete seamless package, such that the
network model operates within the GIS using the same spatial database. This
approach is becoming more commonly used in the water and wastewater
industries (www.nagcs.com), and it is facilitating the rapid development of more
detailed and accurate distribution system models. Information on facilities and
demands can be routinely updated. The results of a modeling application can be
rapidly displayed and analyzed along with other spatial data. The potential for
real-time monitoring and application of models to confirm normal system
performance and assist in system operation under routine and emergency
conditions can also be made possible with the additional integration of the
SCADA systemthrough the common database.
An integrated decision support system should give water utilities r
e
a
l
-
t
i
m
e
surveillance and control on finished water quality in the distribution system. I
t
can
greatly assist water utilities in reducing infrastructure vulnerability and en-
hancing their ability to prepare for and respond to natural and/or man-made dis-
asters, terrorist attacks, and other emergencies, providing the public with added
security and peace of mind. The system can be used not only as an early warn-
ing system to detect potential contamination threats but also as an effective
planning tool to identify viable solutions before an incident or disaster occurs
(e.g., evaluating the potential impact of unforeseen facility breakdown, ass essing
the effect of water treatment on contaminants, as well as using surge modeling
to predict and eliminate potential weak spots), or to assist in responding should
it occur (e.g., increasing the chlorine dose at the treatment plant). These en-
hanced capabilities create significant management advantages for water utilities,
including greater operational efficiency and emergency preparedness,
reduced systemvulnerability, improved public notification, shortened response
time, more informed decision making, and stronger customer ties. In evaluating
the potential implementation of an integrated decision support system, a water
utility will need to balance the reduction of risks with the costs associated with
implementing the system.
Although there is not currently a decision support system that satisfies all o
f
the above requirements simultaneously, significant progress has been made t
o
date in both GIS and modeling technologies that meet many of these needs. The
availability of robust and reliable on-line, real-time sensors that can rapidly and
accurately detect and report all potentially detrimental chemical and biological
contaminants that will be both affordable and useful for most utilities is
still years away. Water utilities should monitor the development and maturity of
this technology and determine when such a systemis practical for their use. Box
7-2 presents an example of integrating GIS with water quality modeling,
master planning, and operational decision-making within the Las Vegas Valley
Water District.

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
BOX 7-2
Data Integration at the Las Vegas Valley Water District
Chapter 1 mentioned an AwwaRF project that w ill test the c
o
n
c
e
p
tof decentralized
treatment and its application to the Las Vegas Valley Water District (LVVWD). Various
technologies w ill be evaluated for their use in the decentralized treatment, including air
stripping, granular activated carbon, and biological activated carbon. A key feature of this
study is the use of distribution system modeling interfaced w ith a GIS. Indeed, the LVVWD
is linking its master planning, operational planning, and development review functions by
integrating its GIS database w ith distribution system modeling, SCADA, and enterprise
data. Figure 7-2 presents the conceptual relationship model of these functions and poten-
tial integration benefits (Jacobsen et al., 2005).
FIGURE 7-2 Conceptual relationship model for integration.
During the process of integration, LVVWD developed a one-to-one relationship be-
tw een the GIS spatial data and its network model (Jacobsen and Kamojjala, 2005). The
advantages of taking this approach include ease of search and retrieval with other
data/applications, and ease of importation, development, and maintenance of data. For
a large network model, the disadvantages include an increase in the run-time of the
network model due to the addition of detailed components and relatively slow water quality
simulations. To minimize this problem, LVVWD has taken an “all-pipes capable”
approach where the distribution system is divided according to existing pressure zones
and attached to an operational backbone network (skeletonized). Each of the zone models
can be attached seamlessly to the backbone network for detailed hydraulic and water
quality modeling. LVVWD uses GIS data on a day-to-day basis for pressure complaint
resolution and main break analysis.

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
FEASIBILITY OF ADOPTING G200 FOR SMALL SYSTEMS
While it is important to design, promote, and employ voluntary
improvement programs that will allow utilities to be proactive in maintaining
high water quality in distribution systems, it is equally critical to scale such
programs to accommodate small water systems. Water systems serving small
populations have limited monetary and personnel resources, whose expertise
and technical knowledge are often limited. For many small water systems,
simply meeting the requirements of the existing federal and local regulations
presents a continuous struggle, leaving little or no time and money for quality
improvement programs described in the previous sections.
Monitoring of water quantity and quality beyond compliance requirements
should be based on site-specific characteristics and priorities. For example, a
l
l
small systems should seriously consider monitoring of water pressure, because i
tis
critical to avoiding contamination via cross connections, which is likely to b
ea
ubiquitous problem. On the other hand, monitoring for non-regulated c
o
n-
taminants and the maintenance of an extended water quality database could pro-
vide important information regarding changes in water quality and allow for
optimization of water treatment prior to its distribution, but it may be cost-
prohibitive. Additional water quality monitoring beyond compliance monitoring
should (1) be targeted to contaminants of local concern and help to
overcome site-specific challenges, (2) be associated with source
characteristics and treatment, and (3) be conducted at critical points in the
distribution system during critical times (changes in weather, flow, system
maintenance, etc.).
In general, adherence to the G200 standard by small water systems, whi
ch
will invariable exceed current regulatory requirements, should be implemented
using the following guidelines: (1) implement new activities using a step-
wise approach; (2) provide technical assistance, education, and training; and (3)
develop regulatory, financial, and social incentives. Such a tiered approach to
the implementation of G200 activities should allow for the prioritization of
needs, for the planning of resources, and for the implementation of additional
monitoring practices and maintenance activities over a long period of time.
The consolidation or cooperation of small water systems may also make
adoption of more advanced monitoring and modeling techniques more feasible,
as discussed in NRC (1996).
Training materials, scaled for small-size systems, are essential for o
p
e
r
a
t
o
r
sand
maintenance crew. For example, the EPA provides guidance to small water
systems in asset management—i.e., developing an inventory of assets and de-
termining when repair should give way to rehabilitation or replacement (EPA,
2004c). This type of guidance would be helpful for the other important elements
of a distribution system management plan mentioned above and found in G200.
Technical assistance could also be provided to small systems by larger
water utilities in the area. Finally, public education could result in an
increased awareness and emphasis on the significance of implementing
proactive voluntary efforts.

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
HOW TO PROVIDE INCENTIVES TO ADOPT G200
For a utility to engage in all of the activities mentioned above in addition t
o
compliance monitoring is a considerable challenge that may require the creation
of incentives, perhaps through existing regulations and associated policies.
There are several instruments already in place that could be modified to better
implement the monitoring, modeling, and other approaches for reducing risk
from public distribution systems discussed above.
Federal Regulatory Approach
A first option is that federal regulations could require adherence to a p
r
e
-
scribed list of activities deemed necessary for reducing the risk of contaminated
distribution systems. This list could partly or fully parallel the G200 standard.
Given the accreditation atmosphere in which G200 was created and its history to
date, making the standard a federal requirement seems unlikely. Indeed, G200 is
currently viewed as one of many available industry programs that are voluntary
and include best practices, such as the Partnership for Safe Water,
QualServe, and other accreditation standards. In the opinion of the committee,
G200 is the most comprehensive voluntary programand should be central to a
utility’s development of a distribution systemmanagement plan.
MCL vs. TTR. In lieu of making G200 a federal requirement, portions o
fthe
standard could be explicitly made part of other federal requirements. F
orexample,
new maximum contaminant levels (MCLs) or treatment technique
requirements (TTR) could be created that would capture essential elements of
G200. There is some precedence for this line of action. For example, in the late
1980s, the turbidity requirement in the SDWA went from being an MCL to
being a TTR. The regulation specified treatment techniques that would provide log
removals for Giardia and viruses by requiring treatmentplants to achieve at least
99.9 percent (3 log) and 99.99 percent (4 log) removal/inactivation, respectively. I
t
also specified the performance criteria for treatment based on turbidity. Somethi
ng
similar could be done for many of the distribution system practices mentioned i
n
this chapter. For example, with respect to water pressure monitoring, a
minimum water pressure requirement could be established, not unlike
the measurable chlorine residual that is now required.
Sanitary Surveys. Another mechanism to capture elements of G200 within
existing federal regulations would be via the sanitary surveys conducted by the
state and required for some systems under federal regulations. Under 40 CFR
142.10(b)(2), states are required to have in place a program for conducting sani-
tary surveys of public water systems, especially those who are out of regulatory
compliance (EPA, 1999). The purpose of such surveys is to “evaluate and
document the capabilities of the water system’s sources,treatment, storage, dis-

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
tribution network, operation and maintenance, and overall management to con-
tinually provide safe drinking water and to identify any deficiencies that m
a
y
adversely impact a public water system’s ability to provide a safe, reliable water
supply.” The TCR has a requirement for periodic sanitary surveys for all small
systems that collect less than five samples per month. The SWTR requires an
annual on-site inspection for surface water systems that do not filter. The
IESWTR now requires a survey for all surface water and groundwater-
under- the-direct-influence systems and requires that each of eight elements be
addressed (source; treatment; distribution system; finished water storage;
pumps, pump facilities, and controls; monitoring and reporting and data
verification; system management and operation; and operator compliance with
state requirements) as well as the correction of significant deficiencies. These
surveys are required every three to five years.
A sanitary survey might reveal an absence of training, use of standards, rou-
tine inspections, or certifications that could be predictive of a loss of distribution
system integrity. The distribution system components of the sanitary survey
include:
 Distribution system maps and records, field sampling and measure-
ments, systemdesign and maintenance
 Finished water storage location, capacity, design, painting, cl
eani
ngand
maintenance, security
 Pumps and pump facilities and controls capacity, condition, pum
pi
ngstation
 Water system management and operation administrative records, w
a
t
e
r
quality goals, water system management, staffing, operations and maintenance
manuals and procedures, funding
 Operator compliance with state requirements such as certification a
n
d
competency.
The Drinking Water Academy developed software for use by state sanitary i
n-
spectors in accomplishing all aspects of a sanitary survey with some level
of uniformity. This software can be used during field inspections with a PDA
or Tablet PC (http://www.epa.gov/safewater/dwa/e-sansurvey.html).
A benefit to using this mechanism for promoting G200 is that the sanitary
surveys encompass a wide variety of activities and would likely capture
those felt to be of highest priority for reducing risk (e.g., cross-connection
control and water storage facility inspections). Indeed, the EPA’s Sanitary
Survey program could be reviewed and compared to G200 to see whether the
former might be expanded. However, this approach is likely to succeed only if
sufficient funds are provided to support more comprehensive sanitary surveys.
In addition, current regulations for the sanitary survey exempt or avoid a large
number of water supply systems, effectively limiting the reach of this
mechanism.

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
GASB Accounting Requirements. Another approach for encompassing
some of the elements of G200 and the committee’s high priority activities under
existing federal regulations is through the Government Accounting Standards
Board (GASB) Statement 34 on Basic Financial Statements and Management’s
Discussion and Analysis for State and Local Governments (GASB, 1999). This
regulation requires that all capital assets be documented and reported in finan-
cial statements by looking at the long-term health of government
institutions throughout the United States, including municipally owned water
utilities (Ro- mer et al., 2004). The reporting includes the valuation of
infrastructure and related disclosure of deferred maintenance costs on
treatment plants, pump stations, storage facilities, and distribution systems
(Donahue, 2002), and thus is well positioned to provide the asset management
functions called for in G200 and in Chapter 4. Indeed, GASB 34 has
encouraged the application of asset management in order to meet
requirements (Cagle, 2005). With respect to the specific problems that cause
water quality deterioration in distribution system, GASB 34 may “inadvertently
become the regulatory mandate for corrosion control since uncontrolled asset
deterioration can negatively impact financial statements and, therefore, limit
or degrade the ability of a utility to raise money for capital improvement using
bonds. Utilities that have good corrosion control programs will have better
financial statements and bond ratings” (Romer et al., 2004).
State Regulatory Approach
In lieu of federal regulations, state regulations could require adherence t
o
G200 or the committee’s list of preferred activities for reducing risk in distribu-
tion systems. This approach is limited primarily by the fact that some
states would legally be unable to make such modifications, while others could.
State and Local Building and Plumbing Codes. A logical avenue would
be to consider enhancing state building and plumbing codes to cover more issues
or simply to make enforcement of current codes more uniform. Tables 2-3
through 2-5 show that states vary in their enforcement of state and local codes
for plumbing, health, building, real estate, etc. Clearly, more stringent
details within these codes could be applied. These codes are, however, unlikely
to be able to cover all activities considered to be of high priority for reducing
distribution systemrisk.
Similarly, design and construction requirements at the state level could b
e
modified to capture important elements of G200. Tables 2-3 through 2-5 show
that the SDWA and states already have in place design and construction
standards, enforced largely through the permitting process. Permits are
required when a new system is built or when a significant change to an existing
systemis made. State building codes could be expanded, for example, to require
inspections for cross connections prior to granting building permits on existing
proper-

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
ties or prior to closing of a sale (as is the case for radon inspections in some j
u
-
risdictions). The permit process could also address the design of service lines
and premise plumbing for water quality maintenance (e.g., existence of
dead ends, oversized lines, compatible materials), the extent of lead and copper
corrosion, hot water system maintenance, the level of disinfectant residual at
taps,and the presence of scale and sediment.
State Revolving Fund. Another possibility is that to qualify for a l
o
a
n
from
the State Revolving Fund a utility would have to demonstrate that it is a
d-hering
to G200 or an equivalent list of activities. The 1996 Amendments (Pub- lic Law
104-182) to the SDWA established the Drinking Water State Revolving Fund
(DWSRF), intended to facilitate compliance with applicable national drinking
water regulations or significantly further the health protection objectives of
SDWA. States operate their respective DWSRF programs using annual
capitalization grants from EPA and a 20 percent matching contribution from the
state. Up to 30 percent of the federal grant can be used to assist public water
systems serving disadvantaged communities through subsidized loans or loan
forgiveness. However, under SDWA section 1452(a)(3) states are prohibited
from providing DWSRF assistance to a public water supply that does not have
the technical, managerial, and financial capability to ensure compliance with the
requirements of the SDWA. EPA could clarify that any public water systemthat
does not have a program for managing the distribution system such as
G200 should be viewed as lacking suchcapability.
The SDWA does allow a public water systemto receive DWSRF funding i
f
the owner or operator of the system that lacks capacity agrees to undertake fea-
sible and appropriate changes in operations (including ownership, management,
accounting, rules, maintenance, consolidation, alternative water supply, or other
procedures) that the state determines would ensure the system’s technical,
managerial, and financial capacity. This provision could be used to promote the
use of G200 by requiring that a public water system, as a condition of receiving
DWSRF funding, agree to develop a plan for the implementation of those ele-
ments of G200 that are feasible given the size, complexity and resources of the
system.
It should be noted that the State Revolving Fund is generally used for capi
-tal
investment, but it might be used to comply with G200 if it had to do w
i
t
h
construction practices in some way. This might, however, dilute the objective o
f
the Fund, which is to bring water supplies into compliance.
Bond Ratings. In addition to facilitating the acquisition of DWSRF fund-ing
by small, disadvantaged communities, implementation of G200 could a
l
s
o
improve a drinking water utility’s access to capital in other ways,
particularly for municipally owned water systems. Drinking water utilities in
the United States have historically depended on the municipal bond market to
finance both the development of public water supplies and their expansion into
surrounding

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
areas (Cutler and Miller, 2005). The same practices that are encouraged by the
implementation of G200 may also help improve a municipality’s bond rating.
The two main categories of long-term bonds available to municipalities a
r
e
general obligation bonds, secured by a pledge of the government’s taxing power,
and revenue bonds, secured by the exclusive (in most cases) pledge of a
project’s revenues. The following five factors are generally used by bond
rating authorities for general obligation bonds: (1) general economy, (2) debt
structure,
(3) financial condition, (4) demographic factors, and (5) management practices
of the governing body and administration. Because of this last factor (manage-
ment and administrative practice), a utility that follows the elements of G200 as
outlined in Table 7-1 along with a sound pricing policy should be in a position to
receive a better bond rating, as well as to obtain funding under the DWSRF, than
a utility that does not adhere to G200.
Federal Guidance
In lieu of a regulatory incentive for adopting G200, EPA could advocate a
list
of preferred activities as a way of meeting federal regulations for distribution
systems. This might appear in updated versions of guidance manuals. The EPA
already provides extensive guidance to help water utilities achieve and maintain
compliance, including a capacity development program to assist water systems
in achieving SDWA compliance (Stubbart, 2005). The program addresses
managerial, technical, and financial capacities involving all aspects of the sys -
tem from source water through treatment to distribution. Technical aspects in-
clude how to provide certified operators and reliable infrastructure. Especially
with small systems, the program can help identify weaknesses and in turn iden-
tify avenues for support to eliminate those weaknesses. It also discusses the
various support that is needed to fund and maintain an adequate distribution
systemmaintenance and replacement program.
This chapter has discussed the limitations of compliance monitoring to de-
tect, respond to, and protect against an internal or external contamination event
in the distribution systemthat might jeopardize public health. To affect real risk
reduction from contaminated distribution systems, efforts beyond compliance
monitoring are required. The AWWA G200 standard outlines voluntary activi-
ties that if implemented would provide substantial risk reduction from distribu-
tion systems. Many elements of G200 are critical to maintaining distribution
system integrity, although they do not necessarily suffer from scientific or tech-
nological limitations. The reader is referred to previous chapters for conclusions
and recommendations on these activities, which include cross-connection con-
trol, maintenance of storage facilities, asset management, and training and certi-

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
fication of system operators, inspectors, foremen, and managers. The following
conclusions and recommendations pertain to those elements of G-200 for which
emerging science and technology are altering whether and how these
elements are implemented by a typical water utility. The committee recognizes
that because of cost and personnel limitations these recommendations are
probably not feasible for many medium- and small-sized utilities at the present
time. None- theless, the monitoring and modeling activities discussed represent
an endpoint toward which utilities should be striving. It is hoped that the gap
between what is needed to affect water quality improvement and what utilities
are capable of will shrink in the near future.
Distribution system integrity is best evaluated using on-line, real-time
methods to provide warning against any potential breaches in sufficient
time to effectively respond and minimize public exposure. This will require
the development of new, remotely operated sensors and data collection systems
for continuous public health surveillance monitoring. These types of syst
em
sshould
be capable of accurately (with sufficient precision) determining the na-ture, type,
and location/origin of all potential threats to distribution system integrity. The
availability, reliability, and performance of on-line monitors are improving,
with tools now available for detecting pressure, turbidity, disinfectant residual,
flow, pH, temperature, and certain chemical parameters. These devices have
reached the point for greater full-scale implementation. Additional re-
search is needed to optimize the placement and number of monitors.
Research is needed to better understand how to analyze data from on-
line, real-time monitors in a distribution system. This should focus on algo-
rithms that can integrate real-time hydrological conditions, water quality inputs,
and operational data to evaluate and interpret on-line monitor signals, establish
alarm triggers, and suggest remedial actions. A number of companies are selling
(and utilities are deploying) multi-parameter analyzers. These companies, as
well as EPA, are assessing numerical approaches to convert such data into a
specific signal (or alarm) of a contamination event—efforts which warrant fur-
ther investigation. Some of the data analysis approaches are proprietary, and
there has been limited testing reported in “real world” situations. Furthermore,
when multiple analyzers are installed in a given distribution system, the pattern
of response of these analyzers in space provides additional information on sys -
tem performance, but such spatially distributed information has not been fully
utilized. To the greatest degree possible, this research should be
conducted openly (and not in confidential or proprietary environments).
A rigorous standardized set of network model development and cali-
bration protocols should be developed. While there is a general agreement i
n
the modeling profession that the extent of development and calibration required
for a water distribution network model depends largely upon its intended use,
there are no universally accepted standards and there is currently no
apparent

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
movement toward establishing such standards. Poorly defined and cali
brated
models can lead to management decisions being made based on false or errone-
ous data, and recommendations that may not even work. Continued research i
s
needed to improve network model development and calibration
methodologies (including optimization techniques) and in standardization of
calibration. In addition, improved monitoring technology, such as more
affordable meters that can be inserted into distribution pipes and automated
monitoring for use in conjunction with tracer studies, will greatly improve
calibration of distribution sys-tem models.
Additional research, development, and experimental applications in
data integration are needed so that distribution system models can be used
in real-time operation. Real-time monitoring and modeling of water distribu-
tion systems to assist water utilities in making informed operational deci
si
ons
under routine and emergency conditions requires the integration of network
models with SCADA systems, which has yet to be accomplished at most utili-
ties. The SCADA system can be used to update the boundary conditions in the
network model such as tank water levels, pump on/off status, isolation valve
status, control valve settings, and systemdemands, and the model can in turn be
used to identify the “best” operational strategy for the selected facilities and pass
their control logic back to the SCADA systemfor implementation.
The ability to integrate network models with SCADA systems offers a
number of benefits to the water industry including at a minimum: confirmation
of normal system performance; real-time calibration; system trouble shooting;
projection of operating scenarios; evaluating “what-if” scenarios; training for
and responding to emergencies; and improvement of overall operations. These
benefits can only be realized if both systems can communicate quickly
and properly with one another. Continued work is needed to develop data
integration standards that will allow seamless data exchange for monitoring
and controlling system operations and make them available to the water
industry. Fur- ther development is also needed to expand the ability of GIS to
enable time- series data (e.g., historical or obtained from real-time
measurements) to be asso-ciated with geospatialattributes.
REFERENCES
American Water Works Association/American National Standards I
n
s
t
i
t
u
te(AWWA/ANSI).
2004. G-200: Distribution Systems Operation and Management. Denver, CO:
AWWA.
AWWA Engineering Computer Applications Committee. 1999. Calibration
guidelines for water distribution system modeling. In: Proceedings of the 1999
AWWA In- formation Management and Technology Conference, New Orleans,
Louisiana, April 1999.

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
Berry, J., W. Hart, C. Phillips, and J. Uber. 2004. A General Integer-
Programming- Based Framework for Sensor Placement in Municipal Water
Networks. World Wa- ter & Environmental Resources Congress, EWRI-ASCE.
Bernosky, J. J. 2005. Distribution system security. Pp. 155–181 In: Distribution System
Water Quality Challenges in the 21st
Century: A Strategic Guide. M. J. MacPhee
(ed.). Denver, CO: AWWA.
Booth, D. E., P. Alam, S. N. Ahkam, and B. Osyk. 1989. A robust multivariate p
r
o
c
e
- dure
for the identification of problem savings and loan institutions. Decision S
c
i
- ences
Journal 20(2):320–333.
Boulos, P. F., B. W. Karney, D. J. Wood, and S. Lingireddy. 2005. Hydraulic t
r
a
n
s
ie
n
tguidelines
for protecting water distribution systems. J. Amer Water Works A
s
s
o
c
. 97(5):111–124.
Boulos, P. F., K. E. Lansey, and B. W. Karney. 2006. Comprehensive Water D
i
s
t
r
i
b
u
- tion
Systems Analysis for Engineers and Planners. Pasadena, CA: MWH Soft Pub-
lisher.
Buchberger, S. G., J. T. Carter, Y. H. Lee, and T. G. Schade. 2003. Random demands,
travel times and water quality in deadends. Denver, CO: AWWA.
Bukhari, Z., and M. W. LeChevallier. 2006. Early warning systems to protect d
i
s
t
r
i
b
u
-tion
system water quality. A report submitted to American Water, Voorhees, NJ.
Burlingame, G. A. 1999a. Solving customers’ taste and odor complaints—Part 1: t
h
e
importance of the first response. Opflow 25(10):10–11.
Burlingame, G. A. 1999b. Solving customers’ taste and odor complaints—Part 2: tracking
odors to their source. Opflow 25(11):6–7.
Byer, D., and K. H. Carlson. 2005. Real-time detection of intentional chemical contami-
nation in the distribution system. J. Amer Water Works Assoc. 97(7):130–133.
Cagle, R. F. 2005. Daddy, are we there yet? Underground Infrastructure Management
Jan/Feb:43–46.
Cesario, L. 1995. Modeling, analysis, and design of water distribution systems. D
e
n
v
e
r
, CO:
AWWA.
Clark, R. M., and J. A. Coyle. 1990. Measuring and modeling variations in distribution
systemwater quality. J. Amer. Water Works Assoc. 82(8):46–53.
Clark, R. M., and R. M. Males. 1986. Developing and applying the water supply s
i
m
u
l
a
-tion
model. J. Amer. Water Works Assoc. 78(8):6l–65.
Clark, R. 1998. Chlorine demand and TTHM formation kinetics: a second-order model.
J. Environmental Engineering 124(1):16–24.
a
t
e
r
distribution
Clark, R. M., W. M. Grayman, and R. M. Males. 1988. Contaminant propagation
in distribution systems. Journal of Environmental Engineering, ASCE
114(4):929– 943.
Clark, R. M., and M. Sivaganesan. 1998. Predicting chlorine residuals and the f
o
r
m
a
t
i
o
nof
TTHMS in drinking water. Journal of Environmental Engineering 124(12):1203–
1210.
Clark, R. M., and M. Sivaganesan. 2002. Predicting chlorine residuals in drinking w
a
t
e
r
:a
second order model. Journal of Water Resources Planning and Management
128(2):1–10.
Clark, R. M., R. Thurnau, M. Sivaganesan, and P. Ringhand. 2001. Predicting the f
o
r
-
mation of chlorinated and brominated by-products. Journal of Environmental Engi-
neering 127(6):493–501.

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
Clark, R. M. 1998. Chlorine demand and TTHM formation kinetics: a second-
order model. Journal of Environmental Engineering 124(1):16–24.
Clark, R. M., F. Abdesaken, P. F. Boulos, and R. Mau. 1996. Mixing in d
i
s
t
r
i
b
u
t
io
nsystem storage
tanks: its effect on water quality. Journal of Environmental Engi- neering
122(9):814–821.
Cutler, D., and G. Miller. 2005. Water water everywhere: municipal finance and w
a
t
e
rsupply
in American cities. In: Proceedings of the National Bureau of Economics
Corruption and Reform Conference held in Salem, MA. Grant No. T32
A00186. Washington, DC: National Bureau of Economics Research (NBER).
Deb, A. K., K. A. Momberger, Y. J. Hasit, and F. M. Grablutz. 2000. Guidance
for t
h
e
management of distribution system operation and maintenance. Denver,
CO: AWWA and AwwaRF.
Donahue III, E. J. 2002. GASB 34 and water utilities: deferred maintenance and c
o
n
-
tributed capital. In: Assessing the Future: Water Utility Infrastructure Management.
D. M. Hughes (ed). Denver, CO: AWWA.
Egan, W. J., and S. L. Morgan. 1998. Outlier detection in multivariate analytical chemi-
cal data. Analytical Chemistry70:2372–2379.
Environmental Protection Agency (EPA). 1999. Guidance manual for conducting s
a
n
i
-tary
surveys of public water systems; surface water and ground water under the d
i
- rect
influence (GWUDI). EPA-815-R-99-016. Washington, DC: EPA.
EPA. 2003. National Primary Drinking Water Regulations; Announcement of C
o
m
p
l
e
-tion of
EPA's Review of Existing Drinking Water Standards. Federal Register, 42907–
42929.
EPA. 2004a. Draft Report on Evaluation of Water Quality Sensors in Distribution S
ys-
tems. Washington, DC: EPA Office of Research and Development and Shaw Envi-
ronmental.
EPA. 2004b. Draft Report on Water Quality Sensor Responses to Chemical and B
i
o
-
logical Warfare Agent Simulants in Water Distribution Systems. Washington, D
C:
EPA Office of Research and Development and Shaw Environmental.
EPA. 2004c. Taking stock of your water system—a simple asset inventory for v
e
r
y
small drinking water systems. EPA 816-K-03-002. Washington, DC:EPA.
Friedman, M., G. Kirmeyer, G. Pierson, S. Harrison, K. Martel, A. Sandvig, and A. H
a
n
-son.
2004. Development of distribution system water quality optimization plans.
AWWA Research Foundation, Tailored Collaboration Project Final Report. Denver,
CO: AwwaRF.
GASB (Government Accounting Standards Board Financial Accounting Foundation). 1999.
Statement No. 34 of the Governmental Accounting Standards Board: Basic
Financial Statements—and Management’s Discussion and Analysis—for State and
Local Governments. No. 171-A. Washington, DC: GASB.
Grayman, W. M., and C. N. Arnold. 2003. Overview of CFD Methods in Analysis
o
f
Distribution System Tanks and Reservoirs. In: Proceedings of the AWWA Annual
Conference, Denver, CO.
Grayman, W. M., R. M. Clark, and R. M. Males. 1988. Modeling distribution system
water quality: dynamic approach. Journal of Water Resources Planning and Man-
agement, ASCE 114(3):295–312.
Grayman, W. M., R. A. Deininger, R. M. Males, and R. W. Gullick. 2004. Source w
a
t
e
rearly
warning systems. Pp. 11.1–11.33. In: Water Supply Systems Security. L. W
.Mays
(ed.). New York: McGraw-Hill.

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
Gullick, R. W., M. W. LeChevallier, R. C. Svindland, and M. J. Friedman. 2004. O
c
c
u
r
-
rence
of transient low and negative pressures in distribution systems. J. Amer. W
a
-ter Works
Assoc. 96(11):52–66.
Hall, J., A. Zaffiro, R. B. Marx, P. Kefauver, R. Krishnan, R. Haught, J. G. and
Herrmann. 2005. Parameters for Rapid Contamination Detection in a Water Distri-
bution System. AWWA-Water Security Congress, Oklahoma City.
Hart, F. L., J. L. Meader, and S. N. Chiang. 1986. CLNET a simulation model for t
r
a
c
- ing
chlorine residuals in a potable water distribution network. In: Proceedings of the
AWWA Distribution SystemSymposium, Minneapolis,MN.
Hasan, J. 2005. Technologies and techniques for early warning systems to monitor a
n
d
evaluate drinking water quality: state-of-the-art review. Available
a
t
:
http://www.epa.gov/ordnhsrc/news/news120105.htm. Accessed May 13, 2006.
Hrudey, S. E., and S. Rizak. 2004. Discussion of rapid analytical techniques for d
r
i
n
k
i
n
g
water
security investigations. J. Amer. Water Works Assoc. 96(9):110–113.
Jacobsen, L. 2005. Las Vegas Valley Water District. Presented to the NRC Committee
on Public Water Distribution Systems. April18, 2005. Washington DC.
Jacobsen, L., and S. Kamojjala. 2005. Full System Models and GIS Integration.
AWWA Annual Conference and Exposition, San Francisco, June 2005.
Javadpoour, R., and G. M. Knapp. 2003. A fuzzy neural network approach to m
a
c
h
i
n
econdition
monitoring. Computers and Industrial Engineering 45:323–30.
Lalor, G. C., and C. Zhang. 2001. Multivariate outlier detection and remediation in g
e
o
-
chemical databases. The Science of the TotalEnvironment 281:99–109.
Lansey, K. E., and P. F. Boulos. 2005. Comprehensive Handbook on Water Q
u
a
l
i
t
y
Analysis in Distribution Systems. Pasadena, CA: MWH Soft.
Lauer, W. C. 2005. Water Quality Complaint Investigator’s Field Guide. Denver, C
O
:
AWWA.
Lazarevic, A., L. Ertoz, A. Ozgur, J. Srivastava, and V. Kumar. 2003. A c
o
m
p
a
r
a
t
i
v
estudy of
anomaly detection schemes in network intrusion detection. In: SIAM Con- ference
on Data Mining. San Francisco, CA. May, 2003.
Lee, B., R. Deininger, and R. Clark. 1991. Locating monitoring stations in water distri-
bution systems. J. Amer. Water Works Assoc. 83(7):60–66.
Liou, C. P., and J. R. Kroon. 1986. Propagation and distribution of waterborne s
u
b
-
stances in
Networks. In: Proceedings of the AWWA Distribution System Sympo- sium,
Minneapolis, MN.
Males, R. M., R. M. Clark, P. J. Wehrman, and W. E. Gates. 1985. Algorithm for m
i
x
- ing
problems in water systems. Journalof the Hydraulics Division, A
S
C
E III(2):206–211.
Males, R. M., W. M. Grayman, and R. M. Clark. 1988. Modeling water quality in d
i
s
t
r
i
-bution
systems. Journalof Water Resources Planning and Management, A
S
C
E
ll4(2):l97–209.
Mau, R., P. Boulos, R. Clark, W. Grayman, R. Tekippe, and R. Trussell. 1995. E
x
p
l
i
c
i
t
mathematical models of distribution system storage water quality. J. Hydraulic E
n
-
gineering 121(10):699–709.
Molenaar, K. R., and A. D. Songer. 2001. Web-based decision support systems: c
a
s
estudy
in project delivery. J. Comput. Civil Engineering 15(4):259–267.
Murray, R., R. Janke, and J. Uber. 2004. The Threat Ensemble Vulnerability A
s
s
e
s
s
-
ment Program for Drinking Water Distribution System Security. World W
a
te
r &
Environmental Resources Congress, EWRI-ASCE.
National Research Council (NRC). 1996. Safe Water from Every Tap: Improving W
a
t
e
r
Service to Small Communities. Washington, DC: National Academies Press.

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
Omenzetter, P., J. M. W. Brownjohn, and P. Moyo. 2004. Identification of u
n
u
s
u
a
levents
in multi-channel bridge monitoring data. Mechanical Systems and Sig
nal
Processing 18:409–430.
Ortiz-Estarelles, O., Y. Martín-Bioscaa, M. J. Medina-Hernández, S. Sagrado and E
.
Bonet-
Domingob. 2001. Multivariate data analysis of quality parameters in drinking
water. TheAnalyst 126:91–96.
Ostfeld, A. 2004. Optimal monitoring stations allocations for water distribution system
security. In: Water Supply Systems Security. L. Mays (ed.). New York: McGraw-
Hill.
Ostfeld, A., and E. Salomons. 2004. Optimal layout of early warning detection stations
for water distribution systems security. J. Wat. Res. Plan. Manag. 130(5):377–385.
Owens, J. 2001. A review of federal drinking water regulations in the U.S. Pp. 2
-
1
–
2
-14 In:
Controlling Disinfection By-products and Microbial Contaminants in Drink- ing
Water. EPA/600/R-01/110. Washington, DC: EPA.
Panguluri, S., R. Krishnan, L. Garner, C. Patterson, Y. Lee, D. Hartman, W. Grayman, R
.
Clark, and Piao. 2000a. Using continuous monitors for conducting tracer studies in
water distribution systems. In: Proceedings of the Environmental and Water
Re- sources Institute(ASCE). Anchorage, Alaska.
Panguluri, S., W. M. Grayman, and R. M. Clark. 2005b. Distribution system water q
u
a
l
-ity
quality i
n
Development.
Powell, J., J. Clement, M. Brandt, R. Casey, D. Holt, W. Grayman, and M. LeChevallier.
2004. Predictive Models for Water Quality in Distribution Systems. Denver, CO:
AwwaRF.
Premazzi, G., and E. Hargesheimer. 2002. Introduction. Pp. 1–6 In: Online M
o
n
i
t
o
r
i
n
gfor
Drinking Water Utilities: Comparative Research Report. E. Hargestheimer, O.
Conio, and J. Popovicova (eds.). Denver, CO: AwwaRF.
Romer, A. E., G. E. C. Bell, S. J. Duranceau, and S. Foreman. 2004. External corrosion
and corrosion control of buried water mains. Denver, CO: AwwaRF.
Rossman, L A. 2000. EPANET Version 2 Users Manual. Cincinnati, OH: EPA D
r
i
n
k
-ing
Water Research Division.
Rossman, L. A., and P. F. Boulos. 1996. Numerical methods for modeling water quality in
distribution systems: a comparison. J. Water Resources Planning and M
a
n
a
g
e
- ment
122(2):137–146.
Rossman, L. A., R. M. Clark, and W. M. Grayman. 1994. Modeling chlorine r
e
s
i
d
u
a
l
sin drinking
water distribution systems. J. Environmental Engineering 120(4):803– 820.
Shang, F, J. Uber, and M. Polycarpou. 2002. A particle backtracking algorithm for w
a
t
e
r
distribution systemanalysis. ASCE J. Environ. Eng. 128(5):441–450.
Stubbart, J. 2005. What is a Governing Board’s Role? Opflow Feb.:8–9.
The Great Lakes - Upper Mississippi River Board of State Public Health and E
n
v
i
r
o
n
- mental
Managers. 2003. Ten-States Standards. Albany, NY: Health EducationStandards.
Uber, J., R. Janke, R. Murray, and P. Meyer. 2004a. Greedy Heuristic Methods for L
o
-
cating Water Quality Sensors in Distribution Systems. World Water & Environ-
mental Resources Congress, EWRI-ASCE. Reston, VA: Environmental &
Water Resources Instituteof the American Society of Civil Engineers.
Uber, J., F. Shang, and L. Rossman. 2004b. Extensions to EPANET for Fate and T
r
a
n
s
-port
of MultipleInteracting Chemical or Biological Components. World Water &

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
Environmental Resources Congress, EWRI-ASCE. Reston, VA: Environmental &
Water Resources Instituteof the American Society of Civil Engineers.
Watson, J., H. J. Greenberg, and W. E. Hart. 2004. A Multiple-Objective Analysis o
f
Sensor
Placement Optimization in Water Networks. World Water & Environmental
Resources Congress, EWRI-ASCE. Reston, VA: Environmental & Water Re-
sources Instituteof the American Society of Civil Engineers.
Wen, J., E. Busuego, II, A. Bleemers, and J. Long. 2005. Reduce customer complaints
through control of water quality in distribution systems. In: Water Quality in the
Distribution System. W. C. Lauer (ed.). Denver, CO: AWWA.
Wood, D. J. 1980. Slurry flow in pipenetworks. Journal of Hydraulics, A
S
C
E
106(1):55–70.
Wood, D. J., S. Lingireddy, and P. F. Boulos. 2005a. Pressure Wave Analysis of T
r
a
n
- sient
Flow in PipeDistribution Systems. Pasadena, CA: MWH Soft.
Wood, D. J., S. Lingireddy, P. F. Boulos, B. W. Karney, and D. L. McPherson. 2
0
0
5
b
.
Numerical methods for modeling transient flow in distribution systems. J. A
m
e
r
.
Water Works Assoc. 97(7):104–115.
Zierolf, M. L., M. M. Polycarpou, and J. G. Uber. 1998. Development and a
u
t
o
-
calibration of
an input-output model of chlorine transport in drinking water distribution systems
I.E.E.E. Trans. on Control Systems Technology 6(4):543–553.

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
8
Alternatives for Premise Plumbing
Premise plumbing includes that portion of the potable water distribution
systemassociated with schools, hospitals, public and private housing, and other
buildings. It is connected to the main distribution system via the service
line. The quality of potable water in premise plumbing is not ensured or
monitored by
U.S. Environmental Protection Agency (EPA) regulation. Indeed, the only S
a
f
e
Drinking Water Act (SDWA) rule in which drinking water quality is
purpose- fully measured within premise plumbing is the Lead and Copper Rule
(LCR) for which samples are collected at the tap after the water has been
allowed to remain stagnant.
Virtually every problem previously identified in the main water
transmission system can also occur in premise plumbing. However, unique
characteristics of premise plumbing can magnify the potential public health
risk relative to the main distribution system and complicate formulation of
coherent strategies to deal with problems. This chapter discusses these
characteristics and then considers both technical issues such as the need for
monitoring of premise plumbing condition and policy alternatives for controlling
public health issues related to premise plumbing.
KEY CHARACTERISTICS OF PREMISE PLUMBING
Premise plumbing systems have noteworthy differences from the main di
s-
tribution system that are often under-appreciated by scientists and regulators
with respect to public health goals. These are summarized in Table 8-1 and dis-
cussed more comprehensively below.
High Surface Area to Volume Ratio. Premise plumbing is characterized
by relatively lengthy sections of small-diameter tubing. The total pipe length o
fthe
main distribution system has been estimated at about 1 million miles
(Brongers et al., 2002; Grigg, 2005), whereas 5.3 million miles of copper tubing
alone were installed in buildings between 1963 and 1999 (CDA, 2005). Premise
plumbing has about ten times more surface area per unit length than in the main
distribution system. One study of a distribution system in Columbia, Missouri
determined that household plumbing and service connections had 82 percent of
the total pipe length, 24 percent of the total surface area in the distribution sys-
tem, and held just 1.6 percent of the total volume of water in the system (Brazos
et al., 1985). Another 10 percent of the total distribution system volume was in
premise plumbing if toilets and water heaters were considered (Brazos et al.,
1985).
316

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
ALTERNATIVES FOR PREMISE PLUMBING 317
TABLE 8-1 Characteristics of U.S. Public and Private Transmission Systems
Characteristic Public Infrastructure Private Infrastructure
Approx. Pipe S
u
r
f
a
c
e
per
Volume W
a
t
e
r*
0.26 cm2
/mL* 2.1 cm2
/mL*
Total Pipe Length (U.S.) 0.97 million miles > 6 m
i
l
l
i
o
nmiles Replacement
Value $0.6 trillion Much greater than $0.6
trillion
Prediction of F
a
i
l
u
r
e
Events
Property D
a
m
a
g
e
($/consum er)
Statistically predictable Unpredictable for i
n
d
i
v
i
d
u
a
lhomeow ner
Relatively low Potentially very high
Common Pipe Material Cement, ductile i
r
o
n
,
plastic, cast i
r
o
n
Stagnation Relatively rare e
x
cept
dead ends
Copper, plastics, g
a
l
v
a
n
i
z
e
d
iron, stainless steel,
brass
Frequent and of variable length
Disinfectant Residual Almost alw ays present Frequently absent
Regrow th Potential Rarely realized (
p
a
r
t
l
y
because r
a
r
e
l
y
measured)
Frequently realized
Pipe Wall Thickness > 6.6 mm 0.71–1.7 mm for copper tube
Velocity 2 to 6 ft/sec Can be > 33 ft/sec, on/off or continuous
Infiltration Abrupt changes inflow
are relatively control-
lable (e.g., via
scheduled flushing,
proper distribution
systemdesign)
The service line can be the p
o
i
n
t
of
minimum pressure and experience
frequent w aterhammer, the highest
velocities, and the most leaks
Temperature 0–30 C 0–100 C (at thesurface o
fheating
elements)
Control of Water Quality Utility t
r
e
a
t
m
e
n
t
sand
operation
No controlover w aterc
o
m
i
n
g
into home, but
home treatment d
e
vicesand selection
of plumbing materials can
influence w ater quality
Ow nership Utility End user
Maximum Cost over30
Years per Consumer
$500–$7,000 US As much as $25,000 per homeow ner,
frequency determined by l
i
f
e
t
i
m
e
of
plumbing
Financial Responsibility Distributed b
u
r
d
e
nover
time and large c
u
s-
tomer base
Individual consumer
Cross Connections Relatively rare Widely prevalent
Frequency of S
a
mple
Collection and
Evaluation of WQ
Degradation
Regular s
a
m
p
l
i
n
gre- quired
by r
e
g
u
l
a
t
i
o
nand industry
best standards
Often sampled only in r
e
a
c
t
i
v
e
mode to
consumer complaints
*Based on a 15.2-cm diameter f or mains and 1.9-cm diameter f or home plumbing.
SOURCE: Reprinted, with permission, f rom Edwards et al. (2003). © 2003 by Marc Edwards.

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
Water Age. Utilities and consumers have little control over water age i
n
consumer plumbing. Water can sit stagnant in some buildings for extended p
e
-
riods if they are irregularly occupied, as exemplified by schools in s
um
m
e
rmonths,
vacation homes, or residences whose occupants have work that requires frequent
prolonged travel. Even under full-time occupancy, some sections of plumbing
within a given building are rarely used, and flow patterns can be highly
variable dependent on water use patterns of the occupants. The upshot is that
premise plumbing adds a layer of complexity to the hydraulics of distribution
systems (see Chapter 5). That is, water residing in a given premise will have a
wider distribution of water age than water at the entrance to the premise,
resulting in greater variation in disinfectant residual levels, bacterial regrowth,
and otherissues than occurs in the main distribution system.
It should be noted that the negative effects of water age are exacerbated i
f
the
biological stability of the finished water is poor. Viable strategies to prevent
problematic regrowth include local codes mandating premise plumbing materi-
als that do not quickly react with disinfectants, removal of nutrients fromthe
water to minimize regrowth potential when the disinfectant does disappear, rec-
ommendations that consumers flush unused premise plumbing lines, installation
of booster stations (e.g., as is sometimes done in hospital plumbing systems) to
ensure that residuals are supplied to all points of the distribution system, and use
of on-demand water heaters to minimize storage volumes in premises.
Presence of Different Materials. Premise plumbing systems are com-prised
of a wide range of materials including copper, plastics, brass, lead, galvanized
iron, and occasionally stainless steel. Many of these materials are not
typically present in the main distribution system. The impact of water quality
changes on the performance of materials within premises, and the effects of ma-
terials on water quality within premises, are often overlooked by water utilities.
For instance, Brazos et al. (1985) show that the majority of chlorine demand in
water systems often arises from pipe surfaces. Extensive work has been done
investigating the reactions between chlorine and materials used in the main dis -
tribution system including polyethylene, PVC, iron, and cement (e.g., Clark et
al., 1994), and routine samples collected in distribution systems reflect disinfec-
tant loss from reaction with these materials. In general, reaction rates for
chloramine with materials in the main transmission system are very low com-
pared to free chlorine. But recent research has demonstrated that under at least
some circumstances, chlorine and monochloramine decay very rapidly via reac -
tions with copper and brass in premise plumbing (Powers, 2000; Nguyen, 2005;
Nguyen and Edwards, 2005). Domestic water heaters also have very
reactive aluminum and magnesium anodes that can contribute to rapid
chlorine and chloramine decay in buildings.
Extreme Temperatures. Water sitting in premise plumbing is subject t
o
greater extremes of temperature than in the main distribution system (Rushing and
Edwards, 2004). In summer months, even the cold water line in prem
i
se

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
plumbing can be 10–15° C warmer than for the mains. In addition, there i
s a
hot-water distribution system with storage in most buildings, and often water
chillers or refrigerated lines. The sampling of the main distribution system can-
not capture effects of these variations on water quality in premise plumbing,
particularly in relation to microbial type and concentrations. This is especially
true for moderate thermophiles such as Legionella in water heaters.
Low or No Disinfectant Residual. Due to the high surface area to v
o
l
u
m
eratio,
presence of reactive materials such as copper, long storage times, a
n
dwarmer
temperatures in premise plumbing, it is not possible to continuously maintain
residual disinfectant throughout premise plumbing systems. Continu- ous
contact with the water heater and copper pipe in hot water recirculation sys-
tems may be especially problematic with respect to maintaining chlorine residu-
als. Furthermore, water treatment devices are often installed by homeowners to
remove tastes and odors—devices that also remove the disinfectant from the
water. Figure 8-1 shows how the residual detected in hot water in Philadelphia
residences during random sampling was well below the average disinfectant
residual found in the main distribution system, even when chloramine, which is
more persistent than chlorine, is used. The observed variability in disinfectant
residual in homes would not be detected by a routine monitoring program
for regulatory compliance; it is due to factors such as variability in water
temperature, retention time in water heaters, condition of internal materials,
type of heaters and pipes,etc.
Total Chlorine Residual in Hot Water
5
4
3
2
1
0
FIGURE 8-1 Water quality test results for hot water in 27 customer h
o
m
e
sin Philadelphia. The
average chloramine residual throughout the distribution system w as 1.73 mg/L during
December, 2003. At this time of year, chloramine decay rates in this distribution
system are very low, such that the observed decreases in residual occurred primarily in
premise plumbing.
1
.5
-
1
.59
1
.4
-
1
.49
1
.3
-
1
.39
1
.2
-
1
.29
1
.1
-
1
.19
1
.0
-
1
.09
0
.9
-
0
.99
0
.8
-
0
.89
0
.7
-
0
.79
0
.6
-
0
.69
0
.5
-
0
.59
0
.4
-
0
.49
0
.3
-
0
.39
0
.2
-
0
.29
0
.1
-
0
.19
0
-
0
.0
9
Frequency
(number
of
households)

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
Rapid chloramine decay within cold water copper pipe in consumers’
homes has also been observed (Murphy et al., 1997a,b), and the resulting growth
of nitrifiers can influence lead and copper leaching to water due to lowered pHs
and other impacts (Garret, 1891; AWWA, 2003; Edwards and Dudi, 2004). In
other instances that the committee is familiar with, chloramine has not been
found to decay rapidly in premise plumbing, so additional research is needed to
determine the prevalence and specifics of the problem.
Bacterial Levels and Potential for Regrowth. The lack of persistent dis-
infectant residuals, high surface area, long stagnation times, and warmer t
e
m
-
peratures can make premise plumbing very suitable for microbial regrowth in at
least some circumstances. Typical distribution system monitoring stipulates
thoroughly flushing water through premise plumbing when sampled; conse-
quently, problems with regrowth in premise plumbing systems can be missed.
Brazos et al. (1985) noted a two- to three-order of magnitude increase in bacteria
after water was held stagnant in home plumbing versus levels obtained in the
same water after flushing. Using the same basic protocol in two systems experi-
encing difficulties with microbial control, Edwards et al. (2005) found a five-log
increase in bacteria during stagnation in premise plumbing systems in Maui,
Hawaii, and a three-log increase in Washington, DC. It is undoubtedly the case
that high levels of bacteria in first draw samples are sometimes due to regrowth
of bacteria in the faucet aerator (LeChevallier and Seidler, 1980). However, the
Edwards et al. (2005) study was supplemented by bench-scale results that repro-
duced the problem without a faucet present. On the basis of their monitoring
results, Brazos et al. (1985) recommended monitoring for bacteria in first draw
samples in addition to routine monitoring of bacteria in the main transmission
system.
To date there is little direct evidence that high levels of heterotrophic bacte-
ria in premise plumbing systems have adverse health effects. However,
opportunistic pathogens such as Legionella spp. and nontuberculous
Mycobacterium spp. have been found in the biofilms of premise plumbing
systems (Pryor et al., 2004; Tobin -D’Angelo et al., 2004; Vacrewijck et al.,
2005; Flannery et al., 2006; Tho mas et al., 2006; Tsitko et al., 2006).
Hot-water storage tanks and showerheads may permit the amplification of
these bacteria. As discussed in Chapter 3, outbreaks in healthcare facilities of
Legionnaire’s disease have been attributed to Legionella pneumophila in hot
water tanks and showerheads. There is some evidence that nontuberculous
mycobacteria may colonize biofilms, and the species found in treated
drinking water have been linked to infections in immunocompromised
individuals.
Highly Variable Velocities. Premise plumbing is characterized by
start– stop flow patterns that can scour scale and biofilms from pipe surfaces.
Flows up to 10 meters per second can occur. This makes premise plumbing
more susceptible to the dislodgement of biofilms and associated negative
health effects (see Chapter 6) than the main distribution system.

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
Exposure through Vapor and Bioaerosols. Stripping and formation o
f
bioaerosols in relatively confined spaces such as home showers can be an impor-
tant exposure pathway. This is relevant to waterborne disease caused by Myco-
bacterium avium and Legionella, as well as to overall exposure to volatile con-
taminants such as THMs and inhalation exposure to endotoxin (e.g., Little,
1992; Anderson et al., 2002; Mayo,2004).
Proximity to Service Lines. As discussed in Chapter 1, service lines c
a
r
r
ywater
from the distribution main to the premise plumbing in the building or
property being served such that service line contamination can be a source
of degraded water quality in premise plumbing. The majority of water leaks in
a distribution system occur in service lines, service fittings, and connections
(ferrules, corporation stops, valves, and meters) (AWWA, 2005). These
locations therefore provide the greatest number of potential entry points for
intrusion. The lower total chlorine residuals, lack of dilution, and short detention
time before potential consumption might increase the potential health threat to
individual consumers if intrusion were to occur at service lines. Little is
known about the factors that might cause intrusion into service lines. Negative
pressure transients could be responsible, but lower pressures and high
velocities in service lines can cause a venturi effect (e.g., suction) and negative
pressure waves due to water hammer that might also be significant.
Compared to the main distribution system, much less is known about t
h
e
type
and cause of service line failures. Possibilities include internal and external
corrosion, poor installation such as improper backfilling techniques and materi-
als, damage during handling, and improper tapping. In general, the collection of
data documenting the occurrence of such failures is poor.
There is wide variation across the United States regarding ownership of ser-
vice lines, which ultimately affects who takes responsibility for their mainte-
nance. This can greatly complicate the extent to which service lines are in-
spected, replaced, and repaired in a timely manner when leaking. In most cases
a drinking water utility, and thus most regulatory bodies, only takes responsibil-
ity for the quality of water delivered to the corporation stop, curb stop, or water
meter. For that portion of the service line owned by consumers, the responsibil-
ity and cost of repairs fall on consumers, and the speed and effectiveness of re-
pairs can therefore be even less efficient (AWWA,2005).
Prevalence of Cross Connections. In contrast to the main transmission
system, it is relatively common for untrained and unlicensed individuals to do
repair work in premise plumbing. Furthermore, as discussed in Chapter 2, there
is tremendous variability in state cross-connection control programs, both with
respect to the breadth of the programs and the extent to which these programs
are routinely enforced at the local level. As a result of these factors, premise
plumbing is more likely to have cross-connections and potential backflow
events than the main transmission system. For example, repairs by consumers as
simple as replacing a ballcock anti-siphon valve in the toilet tank can create a
direct

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
cross connection if the line to the tank is not air-gapped. Hazardous chemicals
added to the tank could then backsiphon or backflow under some circumstances.
In a study in Davenport, Iowa, 9.6 percent of homes were found to have direct
cross connections to a health hazard, most frequently due to failure to air gap the
line in the toilet tank. Only 4.3 percent of homes investigated did not have a
direct or indirect connection to a health hazard (USC, 2002).
It should be noted that for individual residences, backsiphonage is t
he
greatest
risk. However, it does not occur frequently, and when it does it would likely
only affect a small population (usually only the population utilizing the
building). Thus, these events are likely to be underreported. Backflow events are
more likely to be reported when they occur in institutional settings, potentially
affect a larger population, and are more likely to propagate back into the main
distribution system.
The EPA white paper on cross connections (EPA, 2002a) makes it clear that
the majority of backflow events occur in premise plumbing. As shown in Table
8-2, the portion of the distribution system controlled by the utility accounted for
only 18 of 459 reported backflow events.
Responsible Party. There is lack of clarity over who is responsible f
o
r
maintaining water quality in premise plumbing. Many consumers mi
stakenly
believe that EPA regulations and their water utility guarantee that tap water i
s
always safe to drink. Some public advertisements and educational m
a
t
e
r
i
a
l
s
reinforce the perception that EPA regulations and utility responsibility extend t
o
the tap. Historically, however, in the United States the property line demarcat-
ing the public from the private system has not been crossed for regulatory pur-
poses. The notable exception is the LCR, which has successfully reduced
the general corrosivity of public water supplies in relation to lead and copper
leaching. But ultimately individual homeowners and building supervisors bear
final responsibility for protecting themselves from excessive lead or copper
exposure and other degradation to water quality occurring beyond the property
line.
Economic Considerations. The net present replacement value of
premise plumbing and the corresponding cost of corrosion far exceed those for
the main distribution system (Ryder, 1980; Edwards, 2004). Moreover, costs
associated with premise plumbing failures are unpredictable and fall directly on
the consumer. Leaks occurring in premises also have implications for
insurance renewal and mold growth.
Leaching and Permeation. Leaching and permeation mechanisms are t
h
e
same in premise plumbing as in the main distribution system. However, t
hehigher
pipe surface area to water volume ratio, very long stagnation times, and
lessened potential for dilution increase the potential severity of the problem in
premise plumbing. If permeation were to occur through a consumer’s service

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
TABLE 8-2 Numbers of Documented Backflow Incidents from1970 to 2001.
Location of cross connection Number of reportedbackflow events
Homes 55
Apartments 27
Mobile homes 1
Neighborhoods 3
Public Water Supply 15
Medical buildings 27
Schools 31
Other government buildings 24
Restaurants 28
Office buildings 18
Other commercial buildings 66
Agricultural, recreational, and industrial 56
sites
Unknown or other miscellaneous sites 108
SOURCE: Adapted f rom EPA (2002a).
line or premise plumbing, it would not be detected by routine distribution
sys-temmonitoring.
Scaling/Energy. At present about eight percent of U.S. energy demand i
s
attributable to costs of pumping, treating, and heating water, and water
heating accounts for 19 percent of home energy use (EPA, 2005). Hot
water systems and small diameter tubes in premises are more sensitive to build
up of scale, which can increase head loss and decrease water heater efficiency.
The implications and costs of scaling in buildings tend to constrain the
range of feasible water chemistries that might be considered to protect public
health. For example, higher pH values that might be desirable to reduce
nitrification and protect public infrastructure from internal corrosion could cause
unacceptable scaling.
GAPS IN RESEARCH AND MONITORING
The preceding section highlights some of the unique challenges
posed b
y
premise plumbing relative to the main water distribution system. Even
more so than with the main distribution system (see Chapter 3), very few
studies have been done to assess the magnitude of the public health threat posed
by premise plumbing. This is partly due to a lack of water quality monitoring
in premise plumbing. Normal distribution system monitoring under EPA
regulations often utilizes taps located in buildings, but water is thoroughly
flushed from the pipes before sampling with the exception of samples for
lead and copper. Thus, if there were problems related to water quality in a given
premise plumbing system it would not necessarily be detected. No drinking
water maximum contaminant levels (MCLs) protect consumers against water
quality degradation resulting from premise plumbing.

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
While solid evidence is not available, water quality degradation occurring
within premise plumbing may have public health implications. For instance,
recent trends in the United States to decrease water heater temperature to mini-
mize scalding and save energy could be increasing the growth of opportunistic
pathogens.Increased use of phosphate inhibitors, chloramine disinfectants,and
point-of-use devices can also benefit or worsen the ultimate quality of water
after it is held stagnant in premise plumbing. However, the lack of monitoring
and isolated nature of problems that are discovered hinder rigorous risk analysis.
Considering the emergence of Legionella and Mycobacteriumas waterborne
pathogens,and recognizing the threat from these microbes arising from
regrowth in premise plumbing systems,more decisive action is necessary.For
instance, existing EPA regulations are likely to produce water with a low level
of Legionella in water leaving the treatment plant, but the effective Legionella
levels in premises may still result in adverse health effects. There are 8,000–
18,000 estimated Legionella cases in the United States each year with a fatality
rate between 10 and 15 percent (http://www.cdc.gov/ncidod/dbmd/diseaseinfo/
legionellosis_t.htm). Drinking water was judged responsible for 12 percent of
Legionella cases in one case study mentioned in the United Kingdom (VROM,
2005a), but the methodology and certainty of that analysis is open to question.
If a similar percentage of Legionella cases in the United States were caused by
drinking water in premise plumbing systems,the health threat from Legionella
alone would be very high relative to all other reasonably quantified risks from
waterborne disease.
Despite this relatively well established and high health risk, only a few s
t
u
d
-ies
have been conducted into possible broad community interventions that might
reduce risk in buildings. Those studies have consistently found that chloramine
was more effective than free chlorine in reducing Legionella levels (Kool et al.,
1999a,b; Pryor et al., 2004; Stevens et al., 2004). A recent study in San Fran-
cisco demonstrated that the change from free chlorine to chloramine reduce the
percentage of buildings with detectable Legionella from 60 percent to 4 percent,
respectively (Flannery et al., 2006). However, one of the studies found higher
levels of mycobacteria after chloramination (Pryor et al., 2004), and the possible
impact of free ammonia as a nutrient on Legionella growth (if chloramine were
to completely decay) has not yet been assessed. Nor have studies correlated
Legionella occurrence and concentrations in drinking water with actual out-
breaks of legionellosis.
Targeted research to improve understanding of water quality degradation
within premise plumbing is recommended and must overcome several chal-
lenges. All three approaches discussed in Chapter 3 for relating distribution
system contamination events to public health risk (pathogen occurrence
measurements, outbreak surveillance, and epidemiology studies) have
unique challenges that increase the difficulty of their execution when applied
to premises. Legionella has only recently (since 2001) been added to the CDC
outbreak surveillance system. Unfortunately, existing CDC outbreak data
would rarely implicate premise plumbing because backflow and regrowth
events likely would

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
not be reported unless an institutional building with large numbers of people w
a
s
affected. Furthermore, there are minimal data on exposure routes other t
han
ingestion, and this is yet another reason why so few data exist on the health ef-
fects of Legionella in tap water. The CDC has recently changed it
reporting requirements for the outbreak surveillance system so that outbreaks
that arise from events in premise plumbing are more clearly identified (see
Chapter 3). Box 8-1 presents one of the few outbreaks clearly linked to
contamination of premise plumbing.
The little epidemiological research done to date has attempted to track the
impacts of premise plumbing components on gastrointestinal upset (e.g.,
Pay- ment et al., 1997; Colford et al., 2005), but not health problems
arising from exposure to bio-aerosols as would be necessary for Legionella and
Mycobacte- ria. The Davenport study (LeChevallier et al., 2003, 2004; Colford
et al., 2005)
BOX 8-1
Waterborne Disease Outbreak Associatedwith
Premise PlumbingContamination:North Dakota, USA, April 1987
Ethylene glycol is a solvent w ith a sweet, acrid taste that is u
s
e
d
in antifreeze solution and
in heating and cooling systems in buildings. Ingestion of ethylene glycol causes acute
poisoning w ith central nervous system depression, vomiting, hypotension, respiratory fail-
ure, coma, convulsions, and renal damage, depending on the dose. The fatal dose for
adults is approximately 100 g. Several incidents of ethylene glycol ingestion have been
reported to the CDC w aterborne disease surveillance system. All these incidents have
involved public buildings and have been linked to contamination of premise plumbing
through backflow via cross-connection with an air conditioning or heating system.
In April 1987, two children in rural North Dakota w ere admitted t
oa local hospital w ith
acute onset of somnolence, vomiting, and ataxia. Toxicologic analysis of their urine indi-
cated the presence of ethylene glycol. Further investigation revealed that both children had
been to a picnic earlier in the day at a fire hall in rural North Dakota. Approximately
400 persons had attended the picnic, and telephone interviewswith about 91 percent of the
attendees identified 29 additional cases of apparent ethylene glycol poisoning w ith 66 per-
cent of the cases occurring in children under ten years of age. The most frequently re-
ported symptoms w ere excessive fatigue and sleepiness, unsteadiness when walking, and
dizziness. Data collected during the telephone interview about food and beverages con-
sumed during the picnic indicated that one beverage, a noncarbonated soft drink, was
strongly associated w ith illness (relative risk = 31.0). A clear dose-response was also ob-
served among children, w ith no cases occurring among children who did not drink the im-
plicated beverage, two cases among children who drank less than or equal to half a cup,
five cases among children who drank one-half to one and a half cups and 12 cases among
those w ho drankmore than one and a half cups.
The implicated beverage had been prepared on-site using a powdermix and w ater
drawn from a spigot near the fire hall heating system that used a mixture of water and anti-
freeze and was cross-connected to the potable water supply. There was a valve on the
cross-connection but no information on whether the valve had been closed before
collecting water to prepare the beverage. Other foods and beverages had been
prepared in the fire hall kitchen, and the kitchen sink was about 30 feet from the
spigot w ith the cross- connection. A water sample collected from the spigot on the evening
of the picnic w as determined to have an ethylene glycol concentration of 9 percent.
SOURCE: MMWR September 18, 1987/36(36):611-4.

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
is the only known example of an epidemiology study where premise
plumbing was investigated as a source of contamination contributing to
gastrointestinal upset, and no impact was observed. Payment et al. (1997) cited
a lower incidence of gastrointestinal upset after water contacted premise
plumbing, and speculated it was due to disinfecting properties of copper. Of
course very high levels of soluble copper in water leached from premise
plumbing can also cause gastrointestinal upset (Craun et al., 2001). Thus, a
range of health impacts from premise plumbing issues can be expected.
With respect to pathogen occurrence measurements in premise plumbing,
there is also no regulation or even voluntary standards recommending such sam-
pling, and as a result background data are not being collected. Guidelines from
the CDC (CDC, 2003, 2004) exist for Legionella in high risk buildings such as
hospitals where an infection control officer is often responsible for monitoring
and mitigating risk, but such monitoring and control measures are not routinely
followed in other situations or in individual residences. Indeed, the current EPA
guidelines on scalding prevention run counter to common control measures
for Legionella (see section below under Policy Alternatives). Other
opportunistic pathogens such as Mycobacteria are emerging concerns, for
which there is a weaker link to disease and therefore even less incentive for
monitoring. Moni- toring samples could be collected by utilities fro m public
buildings or from consumers’ homes during Lead and Copper Rule monitoring.
Box 8-2 discusses the routine monitoring conducted on tap water in Seoul, South
Korea.
WHY HOME TREATMENT DEVICES
ARE NOT ALWAYS THE ANSWER
Home treatment devices have become increasingly popular as a means
to further treat drinking water supplied by public water systems, and they are
considered to be a potential technical solution to some problems
associated with premise plumbing. There are a myriad of available devices
designed to remove organic and inorganic chemicals, radionuclides, and
microbiological agents from tap water. Common home treatment devices
include point-of-use (POU) devices that are mounted at the end of the
faucet, canister type devices that are plumbed in-line under the sink, stand-
alone pitchers in which water is gravity fed through a filter, and refrigerated
filtered-water systems. Home treatment devices used to treat the entire flow
into the premise are called point-of-entry (POE) devices. POE devices can be
as simple as a water softener to more complicated devices that combine
sediment filters, activated carbon filters, and ultraviolet (UV) disinfection.
Home treatment devices can range in cost from tens of dollars for a pitcher-
type filter to thousands of dollars for a whole house treatment system.
Most devices have components that need to be changed at a regular interval or
after a specified volume of water has been treated. Membranes for reverse
osmosis treatment systems are changed at a given frequency or when there is a
reduction

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
in total dissolved solids removal efficiency. UV disinfection systems must b
e
inspected periodically to prevent scale from forming on the lamps, which w
i
l
l
reduce the light intensity.
Most manufacturers of home treatment devices test their devices
according to ANSI/NSF standard test protocols. However, there is no national
certification program that requires such testing. Only three states (California,
Iowa, and Wisconsin) mandate that, before making a health claimand selling
devices in their state, home treatment devices must be tested and certified as
meeting ANSI/NSF standards.
For many reasons, POU and POE devices are not a panacea to p
r
e
m
i
s
e
plumbing
issues. First, although home treatment devices are effective in removing the
contaminants for which they are designed, they cannot work past their point of
application. For example, if a POE device is used and there are cross -
connections within the home “downstream” from the device, contaminants
re-
BOX 8-2
City of Seoul Water Works’ Customer Tap Water Quality CertificationProgram
In South Korea, the City of Seoul Water Works (SWW) has conducted a customer tap
water quality certification program since November 2001. The program is part of its
water quality management system to enhance reliability and meet customer satisfaction of
its municipal drinking w ater supply. Under this program more than 50,000 drinking w ater
taps are checked each year. A total of 344,600 taps have been covered by SWW since
2001. The targeted sites include apartment complexes, schools, households, public parks,
and shopping malls.
Seoul has over 10 million inhabitants, w hich are provided for wi
thsix drinking
water treatment plants w ith a total daily production capacity of 5.4 million cubic meters of
finished water. The total length of distribution system pipe in the city is approximately
15,870 km. While SWW’s treated water meets the national water quality standards set
forth by the Korea Ministry of Environment, it is well documented that city water can
deteriorate upon standing in customers’ water storage systems that have been poorly
maintained. Many residential and commercial facilities in the city have indoor water storage
systems.
The customer tap water quality certification team consists of S
W
W
employees and
representatives from environmental or citizen groups. At each targeted site the team col-
lects a cold water sample from a kitchen faucet and examines the conditions and integrity
of the water pipes and storage systems of the customer. Each water sample is field tested
for chlorine residual, turbidity, pH, iron, and copper at the site. If the water meets the na-
tional drinking w ater quality standards, the team issues and attaches a water quality certifi-
cation on the faucet. If it does not meet the standards, the following secondary parameters
are tested back in the laboratory: heterotrophic plate count, total coliforms, E. coli, ammo-
nium nitrogen, zinc, and manganese. If the team finds inadequate plumbing or poor sani-
tary conditions that could createwater quality problems, they provide guidance to the occu-
pant to correct the problems.
SOURCE: Communicated to Gary A. Burlingame, Philadelphia Water D
e
p
a
r
tm
e
nt via e- m ail on July
20, 2005 by Jung J. Choi, Philadelphia W ater Department and Dr. Lee Suw on and Lee Gyu
Sub, Waterw orksResearch Institute, SeoulMetropolitan Government.

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
sulting from the cross connection will not be removed. Second, the media
o
r
membranes used in POU and POE treatment devices may be susceptible to mi-
crobial colonization. Higher levels of bacteria have been found in the finished
water produced by some POU and POE treatment devices, particularly those that
incorporate an activated carbon element (Rollinger and Dott, 1981; Camper et
al., 1985; Calderon and Mood, 1987; EPA, 2002b). Granular activated carbon in
point-of-use treatment devices can accumulate nutrients and neutralize disin-
fectant residuals, thereby providing an ideal environment for microbial
growth (Tobin et al., 1981; Geldreich et al., 1985; Reasoner et al., 1987;
LeChevallier and McFeters, 1988). Several coliform bacteria (Klebsiella,
Enterobacter, and Citrobacter) have been found to colonize granular activated
carbon filters, regrow during warm-water periods, and discharge into the
process effluent (Camper et al., 1985). The presence of a silver bacteriostatic
agent did not prevent the colonization and growth of HPC bacteria in granular
activated carbon filters (Tobin et al., 1981; Reasoner et al., 1987). Rogers et al.
(1999) reported the growth of Mycobacterium avium in point-of-use filters in the
presence of 1,000 µg silver/mL filter medium. The health implications of this
regrowth are uncertain. Third, although POU disinfection devices are
available, including UV and distillation systems, these devices are not designed
to treat water used for showering and bathing. Some POE devices include UV
disinfection, which can potentially be effective in reducing the levels of
Legionella and other microorganisms entering the premise (Gilpin, 1985; EPA,
1999), but they would not stop regrowth of opportunistic pathogens in the
premise plumbing system.
POLICY ALTERNATIVES
Although the magnitude of the public health threat from bacterial regrowth,
cross connections, intrusion, leaching, and permeation in premise plumbing is
not clearly defined, improved control should be a high priority based on existing
data and best professional judgment regarding the potential for problems. It is
possible to address these problems through legislation and regulation, the
plumbing code, voluntary standards, and public education. Examples of each
approach, including their use in other countries, are provided in the sections that
follow.
Problems Addressed Through Plumbing or Building Codes:
Scalding and Regrowth in Water Heaters
Several countries are addressing the complicated issue of simultaneously
controlling scalding problems and preventing Legionella growth in water heat-
ers. A conflict arises because the hotter temperatures that control Legionella
also increase the likelihood of scalding (see Figure 8-2). The consumer product
safety commission estimates that scalding from tap water results in 3,800 injuries

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
FIGURE 8-2 Higher temperature decreases the time required for L
e
g
i
o
n
e
l
l
ainactivation but also
decreases the time to acquire burns from scalding. SOURCES: Illustrative data derived
fromCPSC (2005) and Armstrong (2003).
and 34 deaths annually in homes (CPSC, 2005), and children are especially a
trisk
from scalding (NSKC, 2004).
To ensure that water storage is hot enough to prevent microbial regrowth
but that delivered water temperatures are not high enough to cause scalding, one
solution is to install a plumbing device that physically limits the percentage of
hot water flowing to the tap based on target delivery temperatures. For instance,
Australian standards require maintenance of hot water systems at a minimum of
60° C to control Legionella and installation of a mixing valve at points of deliv-
ery to prevent scalding (Spinks et al., 2003). The Australian work recognized
that growth of many pathogenic microbes in hot water systems other than
Le- gionella would also be controlled by this change in the plumbing code.
Canada is in the process of finalizing a plumbing code that requires installa-
tion of valves that prevent water outlet temperatures exceeding 49°C at shower-
heads, bathtubs, or lavatories, while also requiring temperatures above 55°C in
hot water recirculation or 60°C in water service heaters to prevent
Legionella (C. R. Taraschuk, personal communication, Standing Committee on
Building and Plumbing Services, final proposed wording for code, 2005). A
cost/benefit analysis of the mixing valve requirement in Canada indicated a
benefit of $0.7–
$4.2 million in reduced scalding versus a cost of $48–$119 million per year (
K
.
Newbert, personal communication to B. E. Clemmensen, August 26, 2005).
However, the estimated benefit did not include costs of reduced Legionella
death rates or reduced outpatient care, nor did it consider higher energy
costs incurred in maintaining higher water temperatures. The code change in
Canada would be relevant only for new dwellings, since retrofit cost/benefit
analyses
Scalding
Legionella D
e
ath
30
25
20
15
10
5
0
300
250
200
150
100
50
0
53 55 57 59 61 63 65
Temperature (° C)
Time
to
Adult
3rd
Degree
Burns
(seconds)
Time
to
Legionella
Death
(minutes)

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
were less favorable. However, specific localities could choose to require
retro-fits.
The American Society of Plumbing Engineers has recommended a similar
approach in the United States (George, 2001), but the official
recommendation from EPA is that consumers reduce their water heater
temperature to 48° C to save energy, prevent scalding, and reduce scaling (EPA,
2004). Many U.S. water utilities highlight the EPA advice to reduce water
heater temperatures on their web page.
Problems Addressed By Regulation:
Control of Regrowth in Premise Plumbing Systems
Consistent with the current U.S. approach, English water companies ha
ve
met their obligations if a failure to meet standards at the tap can be attributed to
degradation occurring in privately owned premise plumbing (Colburne, 2004;
Jackson, 2004; WHO, 2005). But in public buildings, including schools, hospi-
tals, and restaurants, water quality must meet all regulations for potable water at
the tap. While details are still under discussion, guidelines suggest that water
must be sampled at taps in 10 percent of public buildings each year. “First
draw” samples for bacteria must also be collected, and disinfection of
sample taps is not allowed before collecting samples (Colburne, 2004). A similar
regulatory approach could be considered of U.S. utilities to detect microbially
unsta-ble water and rapid disinfectant loss in premise plumbing.
Legislation and regulation has also targeted operators of premise plumbing
systems. In the Netherlands, the owners of collective water systems
including hotels, camp sites, and sports facilities have been required to
complete a risk analysis for microbial regrowth. The focus was mostly on
Legionella, but a new Drinking Water Directive 98/83/EC also will eventually
consider other microbial parameters at the tap (Regal et al., 2003). If a high
risk is identified, the owner must indicate measures to protect against
Legionella (VROM, 2005a,b). A recent survey of European approaches to
controlling Legionella found that some countries directly addressed premise
plumbing issues (VROM, 2005b).
Problems Addressed By Voluntary Compliance: Hong Kong
A survey in 1999 that revealed 48 percent of Hong Kong respondents rated
their water quality at the tap fair to poor, but also indicated that less than 0.1
percent of the customers made complaints to the water company. Beginning in
2001 the Advisory Committee on the Quality of Water Supplies (ACQWS) be-
gan meeting to discuss strategies that would protect water to the tap. The key
concerns were turbidity and discolored water from older galvanized plumbing.
Various strategies were initially considered including:

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
1) encourage designers of new buildings to design plumbing with w
a
t
e
r
quality
at the tap in mind
2) educate the public to increase confidence and encourage drinking of
water from taps and to maintain plumbing systems
3) encourage renovation of plumbing systems as part of routine m
a
i
nt
e
-
nance
4) inspection programs for older buildings to determine if they need m
ai
n-
tenance,with potential issuance of orders requiring repair
5) require building owners to inspect internal plumbing using li
cense
d
plumbers and submit a report, with possible fines for non-compliance
6) empower utilities to make repairs or remediation for consumers when
problems are persistent
A staged plan was considered for implementing some of the above strate-
gies, in which the first three years of effort would focus on education of c
on
-
sumers, required implementation at government or quasi-governmental buil
d-
ings, and voluntary compliance. Thereafter, if progress was unsatisfactory, laws
would be considered. Loans were already available to customers from the build -
ing department for maintenance ofplumbing.
Consideration eventually gave rise to a Fresh Water Plumbing Quality
Maintenance Recognition Scheme in buildings. The general idea is to create
market forces that would make compliance desirable for participants. Voluntary
successful applicants are awarded a certificate that can be used as a symbol of
effective premise plumbing maintenance to the consumers’ taps. To qualify, the
plumbing system must (1) be inspected at least once every three months by
qualified personnel, (2) have all defects quickly repaired, (3) have water
tanks cleaned every three months, and (4) have water samples collected at least
once a year for analysis. The program is overseen by the water supply
department, and the program is confidential. Checklists are provided for tank
cleaning and water quality analysis and inspection. The program was started
in July 2002; 32 months later, 2,807 certificates had been issued for residential
buildings, hotels, and restaurants. Logos are provided to place on taps, and
the time period in which certified compliance is valid is indicated. About 34
percent of residential households were covered by the program. Other progress
included issuance of a plumbing maintenance guide, which might eventually
become mandatory. A system was in development to track problem premises for
which frequent complaints occurred.
A survey was conducted of approaches for controlling premise pl
um
bi
ng
problems in other Asian cities, and the results are included in Table 8-3. In gen-
eral, the survey revealed that consumers in many of the Asian cities do not drink
water from the tap (< 0.5 percent drink tap water directly in Hong Kong).

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
Problems Addressed by Public Education
Because water utilities have very limited control beyond the water meter o
f
the customer, much of our existing information on water quality is not relevant to
premises. Thus, there is a need to engage customers in identifying problems and
coming up with solutions through publiceducation.
A multifaceted and long-term approach to providing safe drinking water
fromthe treatment plant to the water meter is already used by the drinking water
industry—one that involves compliance with the Safe Drinking Water Act,
use
TABLE 8-3 World-w ide Perspectives on Responsible Party to Prevent Degradation o
fWa-
ter Within Premise Plumbing.
Country Approach
U.S.A. Explicit requirements for Lead and Copper only. Utility h
a
s responsibility to
“Optimize” corrosion control to minimize Pb/Cu at the tap of select homes.
Regulated by “action levels” for lead and copper. Lead pipe and solder banned
in new construction. Guidelines for lead in schools but no regulation.
U.K. By-laws in some instances requires draw off point for p
o
table water directly from
utility services, thereby completely avoiding home plumbing and allow ing direct
access to drinking water. Compliance w ith all regulations required at the tap in
public buildings.
Hong Kong Utility publishes free books and TV ads to encourage u
p
grades to plum bing and to
clean storage tanks. Inspection for dirt and testing for bacteria (utility
inspects based on complaints).
Singapore Code of practice for consumers and their agents r
e
c
o
m
m
e
n
d
s
that samples from
various premise plumbing locations be examined periodically by water analysis.
Chemical examination is beneficial in showing if corrosion is taking place, and
bacterial contamination can be determined by sampling. Storage should be
inspected at least once a year and cleaned. For “housing estates” and
government buildings the recommendations are follow ed, but for “private
estates” recommendations are voluntary. Reports are made to the water de-
partment. Making the recommendations into law w asbeing considered.
Shenzen,
China
At least every half year, water tanks must be cleaned and s
t
e
r
i
l
i
z
e
d
,w ith testing of water
quality at the inlet and outlet by labs. The w ater company has responsibility
for this task for low-rise buildings whereas the building ow ner has re-
sponsibility in high rises. The building management bears the cost, and a
financial penalty can be given to those not complying. Reports are required
to the w ater utility and department of health.
Taipei,
Kuala
Lumpur,
Malaysia
Consumer generally has complete responsibility. However, Kuala Lumpur
requires sufficient residual chlorine, and the desirability of regularly cleaning
cisterns is publicized in new spapers and on television in Taipei.
SOURCE: Adapted f rom ACQWS (2005), except the entry f or the United States.

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
of AWWA and ANSI standards for specifications and best practices, and e
n-
rollment in certification programs such as the Partnership for Safe Water and
AWWA’s QualServe. This broad approach involves regulations, best practices,
and peer review. However, regulation might prove to be the most
expensive way and least efficient way to reduce risk and achieve control when
one considers the customers’ premises as an integral component of the
distribution system. Rather, regulation is only part of an overall approach to
minimizing therisk from everyday use of tap water.
Public education is needed to spur the public to incorporate new actions i
nt
o
everyday life. Similar changes are needed within the water and plumbing indus -
tries, such as the sanitary handling and storage of materials that come in contact
with drinking water. Concepts such as the value of water, the need to conserve
water (which has already taken place in some areas of the United States), and the
need for good materials in guaranteeing good water quality are basic to
bringing about solutions to problems the drinking water industry is faced with.
These concepts must become part of the public psyche, as natural as washing
ones hands after handling raw meat. Altering public behavior with respect to
water will requires a multifaceted approach, broad-based support, and long-term
commitment. It will require numerous efforts directed at premise plumbing such
as:
 Basic education of concepts in elementary schools and higher education
 Education of trades such as plumbing contractors and building super
vi-sors
in health effects of premise plumbing, and the need for standards in products
and design
 Available, easy-to-understand information in public libraries
 Government officials, politicians, consumers, and advocacy g
r
o
u
p
swho
are properly educated and can represent the best interests of the public a
t
large
 Health officials, doctors, and nurses who educate their patients and t
h
e
public on how to minimize risks in practical and achievable ways.
Some progress has been made in the above areas for control of
Legionella in institutional settings through published voluntary guidelines
(ASHRAE, 2000; CDC, 2003). For the analogous problem of indoor air
pollution and radon control, EPA has developed “A Guide to Indoor Air
Quality” (EPA, 1995) that is easy to understand and which highlights the
nature of the health threat and mitigation strategies that can be implemented
to reduce the magnitude of the risk. A similar manual with an accompanying
website would be highly desirable relative to premise plumbing systems. At a
minimum the manual should include consideration of:
 Taste, odor, and aesthetic issues that can arise from premise plumbing
 Maintenance, including flushing of water heaters

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
 Issues related to energy conservation, scalding, and microbial regrowt
h in
water heaters
 Trade-offs with different types of water heaters
 Benefits, limitations, and appropriate uses for various POU and P
OE
devices
 The need to prevent cross connections
 Risks of untrained repair
 Recognizing obvious repairs or plumbing designs that could be pr
ob
-
lematic
 Troubleshooting premise plumbing problems, with information on w
hoto
contact for additional information, investigation, and repair.
Premise plumbing should be recognized as a contributor to the loss of di
s-
tribution system integrity, particularly due to microbial regrowth, backflow
events, and contaminant intrusion via holes in service lines. Imp roper design or
operation of premise plumbing systems can pose a substantial health threat to
consumers, although additional research is needed to better understand its mag-
nitude. In particular, more extensive sampling of water quality within premise
plumbing by utilities or targeted sampling via research is required. The follow-
ing detailed conclusions and recommendations are made.
Communities should squarely address the problem of Legionella, both
via changes to the plumbing code and new technologies. Changes in
theplumbing code such as those considered in Canada and Australia that involve
mandated mixing valves would seem logical as a compromise that would pre-
vent both scalding and microbial regrowth in premise plumbing water systems.
On-demand water heating systems may have benefits worthy of
consideration versus traditional large hot water storage tanks in the United
States. It may be desirable for building owners to conduct risk analysis for
Legionella on their properties as per the Netherlands, and to develop a plan to
address obvious deficiencies. The possible effects of chloramination and other
treatments on Le- gionella control should be quantified to a higher degree of
certainty.
To better assess cross connections in the premise plumbing of privately
owned buildings, inspections for cross connections and other code violations
at the time of property sale could be required. Such inspection of privately
owned plumbing for obvious defects could be conducted during inspection upon
sale of buildings, thereby alerting future occupants to existing hazards and hi
gh-
lighting the need for repair. These rules, if adopted by individual states, mi
ght
also provide incentives to consumers and building owners to follow code and
have repairs conducted by qualified personnel, because disclosure of sub-
standard repair could affect subsequent transferof the property.

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
EPA shouldcreate a homeowner’s guide and website that highlights the
nature of the health threat associated with premise plumbing and mitiga-
tion strategies that can be implemented to reduce the magnitude of the risk.
As part of this guide, it should be made clear that water quality is regulated onl
y
to the property line, and beyond that point responsibility falls mainly on
consumers. Whether problems in service lines are considered to be the
homeowner’s responsibility or the water utility’s varies from systemto system.
Research projects are needed that specifically address potential prob-
lems arising from premise plumbing. Because no organized party has ha
dclear
responsibility for this problem, research has been under-funded. T
h
r
e
e
lines of
research are needed, each of which would help to improve future understanding
of the public health risks from distribution systems:
 Collection of data quantifying water quality degradation in represen-
tative premise plumbing systems in geographically diverse regions and cli-
mates. Some of the needed data include those routinely collected in the main
distribution system, including water residence time, disinfectant residuals, and
microbial monitoring. In addition, greater attention should be focused on under-
standing the role of plumbing materials. Furthermore, the role of nutrients in
distributed water in controlling regrowth should be assessed for premises be-
cause their longer holding times, chronic lower disinfection residuals, warmer
temperatures, and most importantly their colonization by opportunistic patho-
gens such as Legionella and Mycobacterium avium make the biological stability
of the water even more important than in the main distribution system. Special-
ized sampling is needed to quantify regrowth of opportunistic pathogens such as
Legionella and Mycobacteria as a function of consumer water use
patterns, plumbing system layout, and water heater operation. Finally, the
potential impacts of representative POU and POE devices need to be
quantified.
 Practical insights should be developed regarding exposure routes
other than ingestion, including inhalation of bioaerosols from water. Effects
of climate, consumer behavior in bathing and showering, and the specifics
of plumbing system design and operation are likely to be key contributing
factors in disease transmission from premise plumbing contamination. With
respect to contracting disease such as legionellosis, such information would
make it possible to develop steps that might reduce risk or explain why disease
is contracted in some cases and not in others.
 An epidemiological study to assess the health risks of contaminated
premise plumbing should be undertaken in high risk communities. Without
information from the two bullets above, it would be very difficult to
identify such groups with confidence.

e
d
u
c
i
n
g
Risks
i
g
h
t
s
reserved.
 Environmental assessments of outbreaks should begin to incorporate
new insights and allow possible cause-and-effect relationships to be estab-
lished. Such assessments have traditionally focused on documenting outcomes of
waterborne disease and not on key factors related to human exposure. Chap-ter 3
has documented that the reporting of outbreaks is being revised to include more
explicit consideration of distribution system and premise plumbing defi-
ciencies that might contribute to waterborne disease. Much greater emphasis
must also be placed on dose reconciliation in outbreaks, which would require
specialized sampling techniques for bioaerosols in the case of premise
plumbing, in order to develop basic practical data on dose-response
relationships. It is possible to genetically li

Drinking_Water_Distribution_Systems_NoRestriction.docx

More Related Content

Similar to Drinking_Water_Distribution_Systems_NoRestriction.docx (20)

More from HuynhNgocNghia (6)

Recently uploaded (20)

Drinking_Water_Distribution_Systems_NoRestriction.docx