United States
Environmental Protection
Agency
Office of Research and
Development
Washington DC 20460
EPA/625/R-92/001
June 1992
v>EPA Seminar Publication:
Control of Biofilm
Growth in Drinking Water
Distribution Systems
-------
-------
EPA/625/R-92/001
June 1992
SEMINAR PUBLICATION:
CONTROL OF BIOFILM GROWTH IN
DRINKING WATER DISTRIBUTION SYSTEMS
OFFICE OF RESEARCH AND DEVELOPMENT
WASHINGTON, DC 20460
Printed on Recycled Paper
-------
Notice
The information in this document has been funded wholly or in part by the United States Environmental
Protection Agency under Contract 68-C8-0011 to Eastern Research Group, Inc. It has been subject to
the Agency's peer and administrative review, and it has been approved for publication as an EPA
document. Mention of trade names or commercial products does not constitute endorsement or rec-
ommendation for use.
-------
Acknowledgments
In November 1990, the U.S. Environmental Protection Agency (EPA) called together a wide
range of experts to discuss the issue of biofilms in drinking water distribution systems. Workshop
participants represented EPA, states, public water systems, academia, trade organizations, and a
public interest group. They assisted EPA in defining nationally applicable criteria for the issuance
of variances to the maximum contaminant level for total conforms when distribution system
biofilms are present. This seminar publication is based on their discussion of the issues central to
biofilm occurrence and control.
Individuals who made a major contribution to writing and reviewing this document include:
Paul S. Berger U.S. Environmental Protection Agency
Mark W. LeChevallier American Water Works Association
Donald J. Reasoner U.S. Environmental Protection Agency
Other individuals who contributed to this document include:
Steven R. Clark
Edwin E. Geldreich
Steven A. Hubbs
Alexis M. Milea
Wesley O. Pipes
T. Jay Ray
Darrell B. Smith
John H. Sullivan
Roy L. Wolfe
U.S. Environmental Protection Agency
U.S. Environmental Protection Agency
Louisville Water Company
California Department of Health Services
Environmental Studies Institute, Drexel University
State of Louisiana
South Central Connecticut Regional Water Authority
American Water Works Association
Metropolitan Water District of Southern California
The document was drafted and edited by Pamela DiBona and Jennifer Helmick of Eastern
Research Group, Inc. James E. Smith, Jr. of EPA's Center for Environmental Research
Information provided substantive guidance and review.
This document was reviewed and approved as an EPA publication. Mention of trade names, prod-
ucts, or services does not convey, and should not be interpreted as conveying, official EPA ap-
proval, endorsement, or recommendation.
HI
-------
Finding Answers in the Guidance Manual
IS MY
DISTRIBUTION
SYSTEM IN
DANGER OF
DEVELOPING A
BIOFILM?
WHY SHOULD I
CARE ABOUT
BIOFILMS?
HOW DO I KNOW IF
MY DISTRIBUTION
SYSTEM
HARBORS
BIOFILMS?
HOW CAN I MAKE
SURE I WON'T
HAVE A BIOFILM
PROBLEM?
WHAT IF I
DISCOVER A
BIOFILM
PROBLEM?
HOW CAN I
ELIMINATE THE
BIOFILM?
WHERE CAN I
GET MORE
INFORMATION?
It is likely that all distribution
systems contain biofilms to
some degree because of the
difficulty in controlling the
factors that support their
growth. See Chapters 1 and 3.
The organisms that live in
biofilms inhibit the system's
ability to detect fecal
contamination. See Chapter 1
and Appendix A.
Many clues point to a biofilm
problem. See Chapter 4.
Knowing the factors that
contribute to biofilm growth and
ways to control those
parameters is the best
prevention. See Chapters 3
and 5.
Implement a biofilm control
plan. See Chapters 4 and 5.
It is nearly impossible to
eliminate biofilms, but they can
be controlled. See Chapter 5.
The references listed in
Chapter 7. provide more
detailed information; Appendix
C lists additional resources.
A biofilm could cause your
system to exceed EPA limit
for total conforms even if
your treatment system is
performing properly.
See Chapter 1.
1
Microorganisms that
are potentially harmful
to human health can
exist in biofilms.
See Chapter 2.
If you can prove, using Table
4-1, that your distribution
system has a biofilm that
does not pose an
unreasonable risk to health,
you may qualify for a variance
from the MCL for total
conforms. See Appendix A.
IV
-------
Table of Contents
Page
Chapter 1
Introduction -j
Biofilms and the Total Coliform Rule ; ^" 1
How This Document Is Organized 2
Chapter 2
Biofilms 3
What IsaBiofilm? """""I""!"!!!""!!!!!""""3
What Kinds of Microorganisms Make Up the Biofilm? 4
Bacteria 4
Fungi '"[ 7
Protozoa and Other Invertebrates """' 7
Potable Does Not Mean Sterile 8
Chapters
Factors That Favor Biofilm Growth g
Environmental Factors '.'.".'.'.'.[[".'.'.'.'.'.'. Q
Hydraulic Effects 10
Nutrient Availability -\-\
Carbon n
Nitrogen and Phosphorus '".'.'.".'.'.""".. 12
Other Sources of Nutrients 12
Disinfection Residual Concentrations '.12
Corrosion ["[ 13
Sediment Accumulation 13
Chapter4 ;
How to Recognize a Biofilm Occurrence '. 17
Detection of Breakthrough Contamination : [ 17
Detection of Biofilms """!!!!!!!""!!" 18
Characteristics of Biofilm Problems """!l9
Examination of Pipe Surfaces ' 21
Measurement of Nutrient Levels. 22
Corrosion ............" 22
Examination of Hydrodynamics 23
Chapters
Biofilm Control Strategies 25
The Biofilm Control Plan .25
Comprehensive Maintenance Program...: 25
Maintenance of Reservoirs [ 27
Corrosion Control 27
Appropriate Disinfection Practices '.'.'.'.'.'.28
Controlling Nutrient Levels : 28
Other Issues Related to Biofilm Control ........30
Training/Upgrading Personnel .30
Applying Best Available Technology "39
Consideration of Financial Burden 30
-------
Establishing Timetables for Carrying Out a Biofilm Control Plan 31
Public Notification --SI
Chapter 6
Summary 33
Chapter 7
References 35
Appendix A
Drinking Water Regulations for Microorganisms • 41
Appendix B
Glossary 47
Appendix C
Resources 49
VI
-------
CHAPTER 1
Introduction
The occurrence of bacterial growths in finished drinking
water is not new. In 1930, the American Water Works
Association (AWWA) Committee on Water Supply re-
ported on the problem of regrowth of "B. coif (Committee
on Water Supply, 1930). Adams and Kingsbury (1937)
described "bacterial growths" in distribution water that
seemed to come from nowhere: bacteria could not be
detected in the finished water at the point of entry; how-
ever, microorganisms were apparently multiplying in the
distribution pipelines (Schoenen and Scholer, 1985).
Later, scanning electron microscope (SEM) photographs
of water distribution pipes showed complex communities
of microorganisms on pipe surfaces and in pipeline tu-
bercles (knob-like mounds of corrosion on pipe surfaces)
(Allen et al., 1979; Tuovinen et al., 1980; Ridgway and
Olson, 1981; Ridgway et al., 1981). Researchers realized
that water treatment and disinfection systems were
merely inactivating bacteria in the raw water; the microor-
ganisms were surviving the treatment process. Once
sent into the distribution system, the microorganisms
could adapt to the distribution system environment.
Scanning electron micrograph of a distribution pipe
surface
This problem persists today. While water treatment and
disinfection systems can remove most of the bacteria
found in raw water, the water produced is not sterile, and
low levels of bacteria do persist even in properly treated
supplies. Bacterial growth in the drinking water distribu-
tion system makes monitoring for bacterial quality in the
distribution system difficult, hiding significant bacterial
contamination introduced after treatment via cross con-
nections, pipe breaks, or backsiphonage. Growths of
bacteria on pipe walls, called biofilms, also can provide a
haven for potentially pathogenic (disease-causing) bacte-
ria (van der Kooij, 1992).
The current definition of a biofilm is an organic or inor-
ganic surface deposit consisting of microorganisms, mi-
crobial products, and detritus (Marshall, 1976;
Characklis, 1981; Characklis and Marshall, 1990). It is
likely that biofilms exist in all distribution system pipe-
lines, and they are now recognized as part of the normal
aquatic system. Factors that influence the types and
numbers of microorganisms found in finished drinking
water are poorly understood, but probably include the
type and quality of source water; the effectiveness of
treatment and disinfection; physicochemical parameters
(i.e., temperature, degree of corrosion); and the engi-
neered system (Geldreich, 1988; LeChevallier et al
1991 a).
Biofilms and the Total Coliform Rule
The U.S. Environmental Protection Agency's (EPA's)
goal in developing regulations regarding microbial water
quality is to reduce the threat to public health from micro-
organisms found in source waters through adequate
treatment and disinfection (U.S. EPA, 1989a, 1989b,
1991). The Total Coliform Rule states that coliform bacte-
ria (a class of microorganisms used as indicators of the
presence of disease-causing microorganisms) should not
be detected in more than a certain percentage of sam-
ples taken from finished drinking water. The presence of
coliforms in drinking water represents a potential threat to
public health because it may indicate that disinfection has
been inadequate to kill all pathogenic organisms associ-
ated with human and animal waste. It is recognized, how-
ever, that biofilms can harbor coliform organisms.
Although these coliforms usually are not of fecal origin,
they can cause violations of the Total Coliform Rule when
-------
released into the distribution system. For this reason, the
Total Coliform Rule allows states to grant a variance if
the system can prove that biofilms are the sole cause of
positive coliform results and the contamination does not
pose an unreasonable risk to health.
How This Document Is Organized
This document describes the types of organisms often
present in drinking water distribution system biofilms,
how biofilms are established and grow, the public health
problems associated with having biofilms in the distribu-
tion system, and tools that water treatment personnel
can use to help control-biofilm growth. Chapter 2 de-
scribes the formation and composition of biofilms in
drinking water systems. Knowledge of the types of or-
ganisms usually present in biofilms and their require-
ments for survival will aid the water treatment facility in
anticipating biofilm problems and preventing their occur-
rence. Chapter 3 provides information on the factors that
influence biofilm growth. It is these factors that water
utilities need to know to control biofilm problems. Chap-
ter 4 explains how to recognize a biofilm occurrence and
describes how various utilities have worked to pinpoint
and solve biofilm problems. Chapter 5 provides guidance
for controlling biofilm growth, using EPA's outline for an
acceptable biofilm control plan.
Finally, Appendix A outlines the federal regulations per-
taining to microorganisms in drinking water and their im-
plications for systems with biofilm problems. Appendix A
also includes a reprint of the January 1991 Federal Reg-
ister. Appendix B is a glossary containing often-used
terms, and Appendix C lists resources for additional in-
formation on the topics covered here.
-------
CHAPTER 2
Bio films
Knowing the types of organisms that can grow in a distri-
bution system biofilm and their requirements for survival
can help facility operators provide safe drinking water by
anticipating biofilms and taking precautions to prevent
their occurrence. Once you have read this chapter, you
will have a general overview of the organisms present in
drinking water, how they can survive treatment to colo-
nize the distribution system, and how they can pose a
public health threat.
What Is a Biofilm?
Biofilms are formed in distribution system pipelines when
microbial cells attach to pipe surfaces and multiply to
form a film or slime layer on the pipe (Figure 2-1). Prob-
ably within seconds of entering the piping system, large
particles, including microorganisms, adsorb to the clean
pipe surface. Some microorganisms can adhere directly
to the pipe surface via appendages that extend from the
cell membrane; other bacteria form a capsular material of
extracellular polysaccharides (EPS), sometimes called a
glycocalyx, that anchors the bacteria to the pipe surface
(Geldreich, 1988; Costeron et al., 1978; Bitton and Mar-
out
Cells
"a
I Detachment a ri h^V
r~3 U ij PL*—
, Transformation ^
Biofilm Phase a B
/ / /
Substratum.
Figure 2-1. A composite of all processes contributing to
biofilm accumulation: (1) transport and adsorption of mac-
romolecules to form a film, called a substratum; (2) trans-
port of cells to the substratum; (3) adsorption/desorption
of cells at the substratum; (4) growth, product and spore
formation, and death within the biofilm; and (5) attach-
ment/detachment at the biofilm-water interface (Characklis
and Marshall, 1990).
shall, 1980, 1990a). The organisms take advantage of
the macromolecules attached to the pipe surface for pro-
tection and nourishment. The water flowing past carries
nutrients (carbon-containing molecules, as well as other
elements) that are essential for the organisms' survival
and growth.
Biofilms are dynamic microenvironments, encompassing
processes such as metabolism, growth, and product for-
mation, and finally detachment, erosion, or "sloughing" of
the biofilm from the surface (Characklis, 1981; Safe
Drinking Water Committee, 1982; Characklis and Mar-
shall, 1990). The rate of biofilm formation depends on the
physicochemical (chemical, thermodynamic) properties
Cut-away of biofilm on a pipe surface
-------
of the interface, the physical roughness of the surface,
and physiological factors of the attached microorganisms
(Fletcher and Marshall, 1982). Sheer forces generated by
fluid velocity and possible effects of disinfectants on EPS
may be important in the release of biofilms from surfaces
(Characklis, 1981; Safe Drinking Water Committee,
1982). The biofilm may grow until the surface layers be-
gin to slough off into the water (Geldreich and Rice,
1987). The pieces of biofilm released into the water may
continue to provide protection for the organisms until they
can colonize a new section of the distribution system.
The ability of bacteria to attach to surfaces in flowing,
generally nutrient-deficient environments (such as drink-
ing water) demonstrates several important ecological ob-
servations (Fletcher and Marshall, 1982):
• Macromoleculefc tend to accumulate at solid-liquid in-
terfaces, creating a favorable environment in an other-
wise nutrient-deficient situation.
• A high flow rate in the system can transport tremen-
dous quantities of nutrients to fixed microorganisms,
even when the nutrient concentration in the water is
low.
• Production of EPS helps to anchor attached bacteria;
EPS also may be a factor in nutrient capture.
• Bacteria embedded in EPS matrices are protected
from disinfectants by a combination of physical and
transport phenomena.
These factors and others have led microbiologists to con-
clude that most bacteria in aquatic environments can ex-
ist at solid-liquid interfaces, as long as sufficient nutrients
are available.
Scanning electron photomicrographs of pipe "coupons"
(small pieces of pipe material) that have been submerged
in distribution water flow provide a picture of the biofilm
microenvironment. The photomicrographs reveal a hard
but porous surface, a complex of crystals beneath the
surface, and microcolonies of similarly shaped organ-
isms, suggesting growth, at the biofilm surface (Allen et
at., 1979). They also show that microcolonies of cells
tend to form at rough surfaces, such as cracks, crevices,
and pits in old and corroding pipes. Such corrosion pro-
vides an increased surface area and greater protection
from the shear force of the flowing water.
What Kinds of Microorganisms Make Up the
Biofilm?
Knowing the types of organisms likely to survive in the
distribution system and their requirements for growth will
aid in controlling biofilm organisms or preventing them
from becoming established. In situ Studies of biofilm com-
munities in the pipe are difficult to perform, but analyses
of samples scraped from the pipe walls and growth on
pipe coupons have revealed large variations in the num-
ber and types of microorganisms.
Few organisms living in distribution system biofilms pose
a threat to the average consumer. The following survey
of the organisms found in biofilms shows that, although
water treatment is intended to remove all pathogenic
(disease-causing) bacteria, systems should be aware
that treatment does not produce water free of all microor-
ganisms (that is, it is not sterile). In fact, some otherwise
harmless organisms may survive the treatment process
and cause disease in children, the elderly, or others with
weakened resistance to infection. (These types of organ-
isms are called opportunistic pathogens.)
Bacteria
Bacteria comprise the largest portion of the biofilm popu-
lation. Heterotrophic bacteria (those requiring organic
compounds as sources of carbon and energy) are often
measured by the Heterotrophic Plate Count (HPC)
method. These bacteria are the most common, and their
source normally is not known. These organisms may sur-
vive the disinfection process to colonize the distribution
system at the time of installation, or they may be intro-
duced through cross connections, backflow events, line
breaks, or repair operations. The public health risk from
these organisms is not known (Geldreich, 1990a), al-
though a study by Payment et al. (1991) describes a cor-
relation between heterotrophic bacteria growing in home
water filtration devices and gastrointestinal illness.
Among the heterotrophic bacteria are a group of closely
related microorganisms, the total coliforms. Conforms are
usually present at high densities in water contaminated
with human and/or animal feces, but may also grow in
nonfecal environments such as water, soil, and vegeta-
tion. Although they do not cause disease as a group
(Geldreich 1986, 1988), they are usually present when
enteric pathogens are present. This is one reason coli-
forms are used as the primary microbial indicator of
drinking water quality.
Coliforms are used to determine the efficiency of
treatment, integrity of the water distribution system,
and as a screen for fecal contamination, even in the
absence of fecally contaminated samples at the
times and locations of sample collection.
Fecal coliforms are a subgroup of the total coliform
group. The predominant fecal coliform is Escherichia
co//, a bacterium closely associated with the gut of warm-
blooded animals. Because E. coli usually do not survive
long in the aquatic environment, their presence in drink-
ing water indicates that fresh fecal contamination is pre-
sent and, consequently, that an urgent public health
problem probably exists, since human pathogens usually
coexist with fecal coliforms.
The types of coliform bacteria found in distribution sys-
tem biofilms may vary according to location and the pro-
4
-------
Scanning electron micrograph of the bacteria E. coli
cedures used to analyze samples, but the predominant
coliform species (spp.) generally include Enterobacter
cloacae, Klebsiella spp., Citrobacter freundii, and Entero-
bacter agglomerans (Geldreich, 1986). E. coli, most often
There are many problems associated with trying to
detect specific enteric pathogens. Because of these
problems, bacteria that are not themselves patho-
genic are measured as surrogates for the more
harmful bacteria. The ideal indicator for drinking
water contamination is:
• Suitable for all types of drinking water.
• Present in polluted water at higher concentrations
than harmful bacteria.
• Able to survive in water at least as long as patho-
gens and is at least as resistant to disinfection.
• Easy and inexpensive to measure in drinking
water samples.
• Generally not present unless harmful contamina-
tion is also present.
used as an indicator of fecal contamination, has been
found in distribution system biofilms, but only rarely (Ol-
son, 1982; LeChevallier et al., 1990a). More often, when
E. coli is found it is evidence of recent fecal contamina-
tion (Geldreich, 1986).
Coliforms of both fecal and nonfecal origin may enter the
drinking water distribution systems and grow in biofilms
even in the presence of excess chlorine remaining after
treatment (called the chlorine residual) (LeChevallier et
al., 1987; Earnhardt, 1980; Lowther and Moser, 1984;
Olivieri et al., 1985; Smith et al., 1989; Wierenga, 1985;
Hudson et al., 1983; LeChevallier et al., 1990b). Al-
though biofilms may represent the greatest concentration
of biological material (biomass) in the distribution sys-
tem, health surveys conducted in systems experiencing
biofilm growth problems (New Haven, Connecticut;
Springfield, Illinois; and Muncie, Indiana) have revealed
no increase in illnesses due to contaminated drinking
water (Geldreich, 1988). However, coliform bacteria that
do not themselves necessarily pose a health threat can
interfere with the system's ability to detect the presence
of bacteria that do cause diseases (those that enter the
water system because of loss of integrity of the treatment
or distribution systems).
Opportunistic Pathogens
An opportunistic pathogen is an organism that can cause
disease in individuals with compromised immune sys-
tems, but that a healthy person's immune system can re-
sist. Elderly people, infants, cancer patients receiving
chemotherapy or radiation, people with AIDS, and burn
or transplant patients in hospitals are especially suscepti-
ble to infection by opportunistic pathogens (Jarvis, 1990).
Opportunistic bacteria include some species of mycobac-
teria, Pseudomonas aeruginosa, Legionella spp., Aero-
monas spp., Flavobacterium spp., and some species of
Klebsiella and Serratia (Geldreich, 1988; Jarvis, 1990).
Klebsiella spp. have been widely studied as opportunistic
pathogens. Different strains of Klebsiella may originate
from environmental sources such as fruits and vegeta-
bles, wood and bark, other plants (Geldreich and Rice,
1987), and soil. They also inhabit the intestinal tracts of
30 to 40 percent of all warm-blooded animals. Some
Klebsiella spp. can produce capsular material that sur-
rounds the cell and helps protect the organism from dis-
infection (LeChevallier et al., 1988a). When grown in a
biofilm, these capsule-producing organisms may form a
slime layer that provides protection against disinfection
for many bacteria (Geldreich, 1988; LeChevallier et al.,
1988a,b, 1990b).
Nosocornial (hospital) infections,have been attributed to
several strains of Klebsiella (Jarvis et al., 1985; Jarvis,
1990; Highsmith and Jarvis, 1985); however, these
cases have not been attributed to drinking water. In fact,
Klebsiella have been isolated from many sources in hos-
pitals, including carpeting, sinks, flowers, and other sur-
-------
faces. Hospital workers' hands are often contaminated
with Klebsiella (Bagley, 1985). The bacteria isolated from
these types of environments are probably less able to
cause disease and therefore not considered a public
health threat (Duncan, 1988; Bagley, 1985).
Antibiotic-Resistant Bacteria
Some bacteria have developed or acquired resistance to
antibiotics as a result of previous exposure to the antibi-
otics (for example, in farm animals treated with drugs),
heavy metals, or genetic transfer. This may create a pub-
lic health problem if the resistant bacteria are also patho-
gens (Armstrong et al., 1981). Armstrong et al. (1981)
showed that water treatment actually may increase the
percentage of bacteria present in treated water that are
resistant to multiple antibiotics. As a result, a large per-
centage of the heterotrophic bacteria in distribution sys-
tem biofilms, and therefore in the water throughout the
distribution system, may be resistant to antibiotics (Arm-
strong etal., 1981).
Disinfectant-Resistant Bacteria
Most bacteria survive in disinfected drinking water by
finding or creating environments where they are pro-
tected from the disinfectant residual. Factors related to
increased survival of bacteria in chlorinated water include
attachment to surfaces, encapsulation, aggregation, low-
nutrient growth conditions, and strain variation.
Surfaces. Extensive research has shown that bacteria
are more resistant to disinfection when they are attached
to or associated with various surfaces, such as turbidity
particles, macroinvertebrates, algae, pieces of carbon
from treatment filters, and pipe surfaces (Tracy et al.,
1966; Levy et al., 1984; Hoff, 1978; Hejkal et al., 1979;
Silverman et al., 1983; Ridgway and Olson, 1982; Herson
et al., 1987; LeChevallier et al., 1980,1988a,b). Ridgway
and Olson (1982) showed that the majority of viable bac-
teria recovered from chlorinated drinking water were at-
tached to particles. Presumably, microbes entrapped in
particles or adsorbed to surfaces are shielded from disin-
fection and are not inactivated. LeChevallier et al.
(1988a,b) showed that bacterial resistance to disinfection
increased more than 600 fold when the organisms were
grown on pipe surfaces.
Biofilms may not only protect bacteria from disinfection,
but also provide an environment where disinfectant-
injured cells can repair cellular damage and grow. Wal-
ters and McFeters (1990) examined the ability of Entero-
bacter cloacae and Klebsiella pneumoniae to recover
after being injured by exposure to monochloramine.
Chloramine is a combination of chlorine and ammonia.
They found that reducing agents, such as sodium sulfite,
can reverse the chemical oxidation caused by mono-
chtoramine. They also found that biofilms can provide
that same reducing environment, aiding in repair-injured
cells.
Encapsulation. Production of an extracellular capsule
provides protection from disinfection when the cells are
grown under low-nutrient conditions (as in drinking
water). In fact, several investigators have reported isolat-
ing encapsulated bacteria from chlorinated drinking water
(Reilly and Kippen, 1983; Clark, 1984). LeChevallier et
al. (1988b) showed that encapsulated Klebsiella pneu-
moniae were three times as resistant to inactivation as
unencapsulated cells when each was exposed to the free
chlorine remaining in the treated water after disinfection.
Aggregation. Sloughing of clumps of cells from treatment
filters or pipe walls has been suggested as a possible
mechanism by which coliform bacteria enter drinking
water supplies. This clumping, or aggregation, may af-
ford bacteria defense against disinfectants. For example,
Stewart and Olson (1986) reported that aggregation of
Acinetobacter strain EB22 increased resistance to free
hypochlorous acid (a form of chlorine) over 100 fold,
while aggregation increased resistance to mono-
chloramine only 2.3 fold. The researchers found that
treatment of the strain with Tween® 80 (a surfactant) pre-
vented aggregation and eliminated the increased disin-
fection resistance.
Scanning electron micrograph of aggregated cells
-------
Growth Conditions. The environment that microorgan-
isms grow in and become accustomed to plays a large
role in determining their sensitivity to disinfection. Carson
et al. (1972) reported that Pseudomonas aeruginosa
growing in distilled water were'markedly more resistant to
acetic acid, glutaraldehyde, chlorine dioxide, and a qua-
ternary ammonium compound than cells cultured on
Tryptic Soy Agar, a growth medium with a high concen-
tration of nutrients. Similar work by Berg et al. (1983) and
Harakeh et al. (1985) has shown that bacteria grown in
low-nutrient and low-temperature conditions, conditions
similar to the natural aquatic environment, were resistant
to several disinfectants. Legionella pneumoniae grown in
a low-nutrient "natural" environment have been reported
to be six to nine times more resistant than agar-grown
cells (Kuchta et al., 1985).
Strain Variation. Wolfe et al. (1985) found that a number
of types of bacteria recovered from chlorinated water
demonstrated a variety of disinfection resistance patterns
to free chlorine and monochloramine. Ward et al. (1985)
reported a strain of Flavobacterium that was more sensi-
tive to monochloramine than to free chlorine, unlike other
species of Flavobacterium. Conversely, Olson and Milner
(1990) indicated that certain strains may develop resis-
tance to monochloramine with repeated exposure. The
fact that disinfection itself can select for a variety of bac-
teria is demonstrated by the work of several researchers
(LeClerc and Mizon, 1978; Armstrong et al., 1982; Mur-
ray et al., 1984) who presented evidence that chlorination
selects for survivors that are resistant to multiple antibiot-
ics. The results of these studies indicated that the selec-
tive pressures of different aspects of water treatment can
produce microorganisms with resistance - mechanisms
that favor survival in an otherwise restrictive environment.
Interaction of Resistance Mechanisms. When LeCheval-
lier et al. (1988a) examined the interaction of disinfection
resistance mechanisms, they found that resistance
mechanisms were multiplicative (that is, the resistance
conferred by one mechanism was multiplied by the resis-
tance factor of a second). For example, the resistance
conferred by attachment of bacteria to a glass surface (a
150-fold increase) and the resistance gained by produc-
tion of extracellular polysaccharides (a 3-fold increase)
made attached encapsulated bacteria 450 times (3 x
150) more resistant to free chlorine than were unat-
tached, unencapsulated bacteria. The researchers con-
cluded that, given the scenario of encapsulated bacteria
growing under low-nutrient conditions attached to pipe
surfaces for long periods of time, it is easy to understand
how bacteria can survive in biofilrns within chlorinated
distribution systems.
Pigmented Bacteria and Actinomycetes
Some heterotrophic bacteria that live in biofilrns may
cause esthetic problems with water quality, including off-
tastes, odors, and colored water problems. Biofilm organ-
isms that fall into this nuisance category include
Actinomyces, Streptomyces, Nocardia, and Arthrobacter
(Geldreich, 1990a; LeChevallier et al., 1987). Complaints
about taste and odor have resulted from Streptomyces
and Nocardia spp. at concentrations greater than 10 or-
ganisms per 100 mL of water. For pigmented bacteria,
the degree of pigment formation observed in cultured
cells will depend on the media used for isolating the bac-
teria in the water sample. Many HPC bacteria isolated
from distribution system biofilrns will produce yellow, or-
ange, or pink colonies when grown on RaA agar (Her-
man, 1978; Geldreich, 1990ai). These organisms may
occur at high levels, coloring the treated water.
Fungi
Fungi, which include yeasts (single-celled spherical
fungi) and molds (multibranched, filamentous fungi)
(Boyd, 1984), can be found in finished water and can
colonize and multiply in the pipe system (Jan/is, 1990;
Geldreich, 1990a; Hinzelin and Block, 1985). Fungi have
been found on pipe surfaces in densities ranging from
0.0 to 5.6 x 104 cells/100 cm2 for yeast and 0.0 to 2.0 x
10 colony-forming units (cfu)/100 cm2 for filamentous
fungi (Nagy and Olson, 1985). Yeast are more resistant
to disinfection than bacteria, probably due to their thick
cell walls (Geldreich, 1990a).
The primary concerns for fungi in drinking water are taste
and odor complaints, although some strains may cause
allergies and toxic reactions when inhaled in vapors or
through contact while bathing (Geldreich, 1990a). Drink-
ing water, however, is not a major source of fungal infec-
tion. Food, soil, and even air contain far more fungi and
are probably more important factors in human infection
(Jarvis, 1990).
Protozoa and Other Invertebrates
Biofilrns in potable water systems may contain a variety
of nonpathogenic protozoa and other invertebrates in-
cluding amoebae, nematodes, amphipods, copepods,
and fly larvae (Levy, 1985). There is no evidence that
these organisms present any health risk themselves, al-
Photo of copepods, magnified
-------
though recent research has shown that Legionella may
grow and survive inside certain amoebae (Smith-
Somerville et al., 1991).
Potable Does Not Mean Sterile
Biofiims provide protection for microorganisms, including
disinfectant-resistant microorganisms and opportunistic
pathogens. These microorganisms may be present in
water obtained at the tap. Therefore, it is possible to drink
disinfected water and still become ill.
Problems arise when health care facilities view disin-
fected water as sterile (U.S. EPA, 1990e). Hospitals and
other clinical facilities (including home health care agen-
cies) need to be aware of the presence of microorgan-
isms in finished water. EPA's pamphlet, Protecting Our
Drinking Water from Microbes (U.S. EPA, 1989c), de-
scribes drinking water treatment and related federal
regulations in simple language. The pamphlet is useful
for educating facility managers about this issue. More-
over, health care facilities should be aware of guidelines
for operation and maintenance of clinical water systems
(hot water heaters, plumbing systems, faucets, showers,
and condenser systems).
-------
CHAPTER 3
Factors That Favor Bio film Growth
For years researchers have investigated the factors that
lead to biofilm growth. Geldreich et al. (1972, 1977) and
Hutchinson and Ridgway (1977) concluded that in gen-
eral, growth occurs when organic materials and sediment
accumulate in distribution pipes, disinfectant residuals
dissipate, and water temperatures increase. Environ-
mental factors (e.g., pH, temperature, and rainfall); nutri-
ent availability; the presence and effectiveness of
disinfectant residuals; internal corrosion and sediment
accumulation; and hydraulic effects have been related to
growth of coliform bacteria in drinking water (LeCheval-
lier et al., 1990a; Smith et al., 1990). The results of these
studies are summarized below. These results can help
you develop an investigative protocol to determine
whether and when your system is susceptible to biofilm
growth. They also can suggest ways to manipulate the
environmental variables to control bacterial growth in the
system.
Environmental Factors
Water temperature is perhaps the most important rate-
controlling factor regulating microbial growth (LeCheval-
lier, 1989). Directly or indirectly, temperature affects all of
the factors that govern microbial growth. Temperature in-
fluences treatment plant efficiency, microbial growth rate,
disinfection efficiency, dissipation of disinfectant residu-
als, corrosion rates, and distribution system hydraulics
and water velocity through customer demand (i.e., water-
ing lawns, filling swimming pools, washing cars). Unfortu-
nately, most water utilities can do little to change water
temperature. Therefore, efforts should focus on control-
ling the parameters that contribute to temperature's influ-
ence. For example, if changes in temperature affect the
effectiveness of disinfection residuals, the system should
monitor the temperature and adjust the residual concen-
tration accordingly.
Most investigators have observed significant microbial
activity in water at temperatures of 15°C or higher
(Howard, 1940; Rizet et al., 1982; Fransolet et al., 1985;
Donlan and Pipes, 1988; LeChevallier et al., 1990a). E.
coll and other enteric bacteria (bacteria that normally live
in animals' intestines) are known as mesophiles, growing
in temperatures ranging from 5° to 45°C. Fransolet et al.
(1985) found that growth of E. coll and Enterobacter
aerogenes was very slow (growth rates divisions per
hour) at temperatures lower than 20°C.
In temperate climates, seasonal phases of coliform
growth often are observed in distribution systems
(Geldreich, 1986; Smith et al., 1989). Smith et al. (1990)
observed seasonal coliform occurrence trends in 81
water distribution systems, with highest coliform levels
occurring during summer months (Figure 3-1). The re-
searchers also found that the species of coliform bacteria
present varied with water temperature (Figure 3-2). How-
ever, in warm climates, or in large buildings where the
plumbing is kept at room temperature, seasonal vari-
ations in temperature are less pronounced, and therefore
seasonal variations in coliform presence will be less dra-
matic as well.
In a careful study by Fransolet et al. (1985), the investi-
gators found that water temperature influenced not only
the growth rate, but the lag time (the length of time after
entering the system before cell division starts) and cell
yield as well. The length of the lag time was found to be
quite important to the organisms' survival in the distribu-
Season
Spring (Mar, Apr, May)
Summer (Jun, Jul, Aug)
Fall (Sep, Oct, Nov)
Winter (Dec, Jan, Feb)
Mean Temperature (°C)
10.2+/-2.9
19.2+/-3.1
1 6.0 +/- 3.2
7.1 +/-1.8
Mean percent of samples
containing conforms
5.2 +/- 4.9
1 2.3+7- 8.3
9.4 +/- 6.0
2.0+/-2.1
Mean cell count
(cells/100 mL)
0.57 +/- 1 .07
1. 98+7- 1.89
1.15+/-1.41
0.14+/-0.28
Figure 3-1. Seasonal distribution system temperature (°C), mean percent of samples containing coliforms, and mean
coliform count in the New Haven, Connecticut, water system from 1986 to 1988 (Smith et al., 1989).
-------
% Colilorm Positive
Jtn-Mir
C. Inundll
E. cloaca*
Apr-Jun
Jul-Sap
Oct-Dac
K. pneumonia*
1. agalomarana
K. oxytooi
Figure 3-2. Distribution system coliform species oc-
curence by season In the New Haven, Connecticut, water
system, 1986 to 1988 (Smith et al., 1989).
Ropiintod from Technology Conference Proceedings - Advances in
Water Analysis and Treatment, by permission. Copyright 1990, Ameri-
can Water Works Association
tion system. For Pseudomonas putida the lag in the
growth phase was about 3 days at 7.5°C, but only 10
hours at 17.5°C. These results show that at low tempera-
tures, cells are washed out of the distribution system be-
fore significant growth is achieved.
Rainfall is another environmental factor that influences
the bacterial quality of drinking water. Some investigators
have suggested that rainfall is a catalyst for coliform
growth (Lowther and Moser, 1984; LeChevallier et al.,
1990a). Lowther and Moser (1984) found that raw water
organic nutrient levels were highest when turbidity in-
creased after rainfall events. LeChevallier et al. (1990a)
observed that coliform bacteria routinely appeared in dis-
tribution system waters 7 days after rainfall events (Fig-
ure 3-3). The authors speculated that rainfall washed
nutrients into the watershed resulting in increased bacte-
rial densities after a transit period and growth lag. For
some systems, however, rainfall events can lead to
breakthrough of bacteria from the treatment system di-
rectly into the distribution system. The increased turbidity
caused by runoff may provide bacteria with particles for
attachment and protect the organisms from disinfection
(LeChevallier et al., 1980; Baker, 1984), and the high
load of bacteria and particles can overwhelm the treat-
ment system capacity. For example, in Rochester, New
York, coliform occurrences were preceded by heavy rain-
falls that increased turbidity in the system's open surface
water reservoirs. Several New England water systems
have experienced increased coliform densities after
heavy rainfall as well (Geldreich, 1986).
Hydraulic Effects
Flow velocity may regulate microbial growth on pipe sur-
faces in several ways (Safe Drinking Water Committee,
1980). Increased velocities cause greater flux of nutrients
Jun Jul
— Coliform
Aug Sep Oct Nov
— • Rainfall
Figure 3-3. Relationship between rainfall and daily coli-
form levels in the New Jersey American distribution sys-
tem. Coliform data have been offset by 7 days
(LeChevallier et al., 1991 a).
Reprinted from Applied and Environmental Microbiology, by permis-
sion. Copyright 1991, American Society for Microbiology
to the pipe surface, greater transport of disinfectants, and
greater shearing of biofilms from the pipe surface.
Changes in water velocity can be due to seasonal condi-
tions. As seasonal consumer demand changes, corre-
sponding changes occur in the hydraulics and water
pressure throughout the distribution system. Changes in
water velocity also can be changed by flow to fire hy-
drants, pipe network design and pipe size, water main
breaks, or distribution maintenance practices such as
flushing (Smith et al., 1989; Geldreich, 1988).
Reversal of water flows can shear biofilms, and the
"hammer" effect that occurs upon sudden opening or
closing of the lines (e.g., when firefighters open and
close hydrants) can dislodge tubercles from pipe sur-
faces. Opheim et al. (1988) found that bacterial levels in
an experimental pipe system increased 10 fold when
flows were started and stopped. Larger releases of bac-
teria were noted when the system was exposed to physi-
cal and vibrational forces.
Distribution system hydraulics also can affect corrosion
and sediment accumulation. Stagnation of water in the
distribution system can result in loss of disinfectant resid-
ual and accumulation of sediment and debris, leading to
microbial growth. Donlan and Pipes (1988) showed that
water velocity had an inverse relationship with biofilm
counts. Dead-end lines often show significant deteriora-
tion in microbial water quality (Smith et al., 1989;
Geldreich, 1986; Hanson et al., 1987; Opheim et al.,
1988; LeChevallier et al., 1987; Geldreich, 1980; Rae,
1981). Stagnation of water in service lines also can result
in high bacterial counts at the customer's tap (Brazos et
al., 1985; LeChevallier et al., 1987).
Researchers have developed hydraulic models to moni-
tor the fate of chlorine residuals and their reaction prod-
ucts in distribution systems (Characklis et al., 1988; Clark
10
-------
et al., 1988). Applying these hydraulic models to better
understand microbial growth in distribution systems could
be useful.
Nutrient Availability
To grow, organisms must derive from the environment all
the substances that they require to synthesize cell mate-
rial and generate energy. For coliform and heterotrophic
bacteria, the principal nutrient sources are phosphorus,
nitrogen, and organic carbon. Trace nutrients also are re-
quired, but these compounds have not been investigated
in drinking water. " ,
Carbon
Organic carbon is utilized by heterotrophic bacteria for
production of new cellular material (assimilation) and as
an energy source (dissimilation). Because heterotrophic
bacteria require carbon, nitrogen, and phosphorus in a
ratio'of approximately 100:10:1 (C:N:P), organic carbon
is often a growth-limiting nutrient. Most organic carbon
compounds in water supplies are natural in origin, de-
rived from living and decaying vegetation. These com-
pounds may include humic and fulvic acids, polymeric
carbohydrates, proteins, and carboxylic acids.
Carbon in drinking water is measured in three ways, as
total organic carbon (TOG), which is the total amount of
soluble and insoluble organic carbon compounds present
in the water; dissolved organic carbon (DQC), which is
the soluble fraction of TOG; and assimilable organic
carbon (AOC), which is the fraction of DOC that can be:
readily digested and used for growth by aquatic organ-
isms. The U.S. EPA National Organic Reconnaissance
Survey found that the nonpurgeable total organic carbon
(NPTOC, see Standard Methods, section 5310 [AWWA,
1989]) concentration of finished drinking water in 80 loca-
tions ranged from 0.05 mg/L to 12.2 mg/L, with a median
concentration of 1.5 mg/L (Symons et al., 1975). Often,
AOC comprises just a fraction (0.1 to 9.0 percent) of the .
total (van der Kooij et al., 1982b).
AOC is measured using a bioassay first proposed by van
der Kooij in 1978. The method employs inoculation of a
water sample with a variety of microorganisms (Pseudo-
monas fluorescence strain PI 7, Spirillum sp. strain NOX,
Flavobacterium sp. strain S12 or Klebsiella pneumonia
strain CF17) (van der Kooij et al., 1982a,b; van der Kooij
and Hijnen, 1985, 1988). The organisms' growth is moni-
tored and the maximum growth yield is determined. Based
on known yield coefficients, the equivalent amount of carb-
on (usually expressed in |ig of acetate-carbon/L) is calcu-
lated (van der Kooij et al., 1982b). The method is labor- and
materials-intensive, and care is needed to properly handle
water samples to avoid contamination with extraneous
organic material. Using this method, drinking water sup-
plies in North America have been found to contain be-
tween 1 and 2,000 u.g acetate carbon equivaients/L
(Characklis et al., 1988; Gaidish et al., 1987; LeGheval-
lieretal., 1987,1988a,1990a, 1992).
Application of the AOC Test
HPC Bacteria. The AOC test has been used in the Neth-
erlands for the past 10 years to help determine treatment
strategies to limit bacterial growth in water (van der Kooij,
1987, 1990; van der Kooij et al., 1989). The European
guideline for safe water limits heterotrophic plate count
(HPC) bacteria in water to less than 100 bacteria/ml. In
their work to help water suppliers achieve the guideline,
researchers in the Netherlands have found that growth of
HPC bacteria is limited at AOC levels of less than 10
u.g/L. In ranges of 20 to 50 u.g/L, problems with excessive
plate counts occasionally occur. At AOC levels greater
than 50 u.g/L, bacterial growth always occurs! Control of
AOC levels has so effectively limited bacterial survival
and growth that secondary disinfection has been discon-
tinued in some systems (Schellart, 1986; van der Kooij
1987). >• •
Coliform Bacteria. Most of the information related to the
growth of coliform bacteria in drinking water has been ob-
tained from the Swimming River Treatment Plant of the
New Jersey American Water Company. It is uncertain,
however, how the results from that plant will apply to
other distribution systems; research is currently under
way to examine other systems across North America.
Based on other research at the New Jersey American
Water Company, however, (see pp 14 to 15), LeCheval-
lier et al. (1992) recommended that systems trying to re-
duce coliform levels in drinking water supplies limit total
AOC levels to less than 100 u.g/L (median AOC concen-
trations can range from 1.5 to 135 M.g/L). The level of sec-
ondary disinfection (e.g., the residual concentrations
maintained in the distribution isystem) could be lowered if
AOC levels were reduced to very low levels (van der
Kooij, 1987; Schellart, 1986). This approach will be valu-
able to the water industry as it tries to limit bacterial levels
while simultaneously reducing disinfection by-products (the
potentially harmful compounds formed when free chlorine
reacts with organic compounds in the water).
BDOC Analyses. Researchers at the Compagnie Gener-
ate des Eaux in France developed a method to measure
biodegradable dissolved organic carbon (BDOC) (Pascal
et al., 1986; Hascoet et al., 1986; Servais et al., 1987),
which is essentially the same portion of dissolved carbon
measured by the AOC procedure. In the BDOC test, in-
digenous bacteria are allowed to grow for a specified
time in a water sample, and then are removed by filtration
through prewashed 0.22 u.m membrane filters. Finally,
the DOC remaining in the filtered water is measured. If
the bacteria are incubated in the water samples for 10 to
30 days, the test allows measurement of slowly degrad-
able organic materials (Pascal et al., 1986). This proce-
dure has some disadvantages, including insensitivity at
low DOC levels and the relatively high cost of a TOC
11
-------
analyzer. Recently, a rapid (3- to 5-day) procedure has
been developed to measure biodegradable organic carb-
on using aerated biofilms on sand particles (Joret etal.,
1988). There are no operational data to relate specific
BDOC levels to HPC or coliform problems; however, a
level of less than 0.1 BDOC mg/L is thought to produce
biologically stable water (i.e., water that is unable to sup-
port bacterial growth).
Nitrogen and Phosphorus
Nitrogen is used by microorganisms to build amino acids
and genetic material. The exact role of nitrogen in growth
of coliform bacteria is unclear, especially because some
strains of Klebsiella can fix molecular nitrogen (Orskov,
1984). Nitrogen is often present in raw water supplies
due to vegetation decay, runoff containing agricultural
fertilizers, leachate from landfills, or wastewater dis-'
charges. An Indiana University study examining biofilm
problems in the Eastern and Midwestern United States
found that many bacterial occurrences corresponded
with applied agricultural fertilizer that entered source
water in stormwater runoff (Geldreich, 1986).
Ammonia, a reduced form of nitrogen, can promote bac-
terial growth in distribution systems. Rittmann and
Snoeyink (1984) found that ammonia!concentrations in
ground-water supplies were frequently high enough to al-
low bacterial survival and growth. Bacteria that can use
ammonia for growth and need only carbon dioxide as a
carbon source (autotrophic nitrifiers) sometime prove to
be a problem when water utilities use chloramines (chlo-
rine plus ammonia) as a disinfectant in the distribution
system. Because autotrophic bacteria grow slowly, long
retention times and warm water temperatures also con-
tribute to their growth, which leads to more problems. For
example, the proliferation of ammonia-oxidizing bacteria
in large, covered finished-water reservoirs in Southern
California was found to eliminate total chlorine residuals,
increase nitrite levels, and stimulate the growth of HPC
bacteria (Wolfe et al., 1988). To a lesser extent, het-
erotrophic nitrifiers are known to contribute to nitrite and
nitrate levels in ambient waters (Verstraete and Alexan-
der, 1973), and also may increase levels in drinking
water.
Phosphorus in the environment occurs almost exclu-
sively as orthophosphate (PO43~). Phosphates are some-
times added to the water supply to control corrosion.
Rosenzweig (1987) found that phosphate-based corro-
sion inhibitors did not significantly influence the growth of
several strains of coliform bacteria. High levels of
Virchem 932, a zinc orthophosphate, showed inhibitory
effects for certain coliform species.
Other Sources of Nutrients
Certain construction materials, including rubber, silicon,
polyvinyl chloride (PVC), polyethylene, and bituminous
coatings, have been reported to stimulate bacterial
growth (Schoenen and Scholer, 1985; Frensch et al., •
1987; Schoenen and Wehse, 1988). Ashworth and Col-
bourne (1986) reported that a substantial proportion of
customer water quality complaints was due to microbial
growths on polymeric materials used in the construction
of storage tanks, fittings, and pipework for buildings. Af-
ter exposure of a bituminous coating to water, 48 organic
compounds could be detected in the water by gas chroma-
tography/mass spectrometry (GC/MS) analysis (Frensch et
al., 1987). A dose of 30 mg/L chlorine was necessary to re-
duce bacterial counts on the coatings.
Disinfection Residual Concentrations
An inability to maintain a disinfectant residual may allow
bacterial growth in drinking water supplies. If disinfectant
levels are too low (e.g., if more than 5 percent of monitor-
ing samples do not contain a detectable disinfectant re-
sidual), then the utility should increase disinfectant
doses, install "booster" stations that add disinfectant at
various points in the distribution system, or use a more
stable disinfectant (e.g., chloramines).
Experience has shown.that maintenance of a chlorine re-
sidual alone cannot be relied on to prevent bacterial oc-
currences, however. Several researchers (Reilly and
Kippen, 1983; Goshko et al., 1983; Olivieri et al., 1985,
Ludwig, 1985; LeChevallier et al., 1987) have indicated
that maintenance of a free chlorine residual did not corre-
late with reduced bacterial counts in the water. Reilly and
Kippen (1983) found that 63 percent of the coliform bac-
teria in two water systems in Massachusetts were iso-
lated from, drinking water that contained greater than 0.2
mg/L chlorine. Nagy et al. (1982) reported that a'l- to 2-
mg/L chlorine residual reduced bacterial levels iri the Los
Angeles aqueduct biofilms by 2 logs, but the bacteria
were still present at 104 cfu/cm2. Maintenance of a 3- to
5-mg/L chlorine residual was necessary to reduce bacte-
rial biofilms by 3 logs, to 103 cfu/cm2. These investiga-
tors, however, found no correlation between free chlorine
residuals (0.15 to 0.94 mg/L chlorine) and the densities
of HPC bacteria in the distribution system biofilms (Nagy
and Olson, 1985). Even direct contact with chlorine does
not stop biofilm growth for long: Seidler et al. (1977) re-
covered conforms in a potable water supply 1 week after
scrubbing redwood tank biofilms with a 200-ppm chlorine
solution.
Ridgway et al. (1984) found that a 15- to 20-mg/L chlo-
rine residual was necessary to control biofilm growth on
reverse osmosis membranes. Characklis et al. (1979) re-
ported that application of 12.5 mg/L free chlorine for 60
minutes contact time was required to reduce the thick-
ness of experimental biofilms 29 percent in an annular
fouling reactor (Rototorque system). The authors pre-
dicted that disinfection using 5 mg/L chlorine would result
in continued biofilm development. None of the experi-
ments, however, resulted in complete biofilm removal.
12
-------
During coliform episodes at Muncie and Seymour, Indi-
ana, disinfectant residuals were boosted as high as 15
mg/L, because coliform occurrences could not be reliably
controlled with free chlorine residuals less than 6 mg/L
(Lowther and Moser, 1984; Olivieri et al., 1985). In most
cases, this course of action is not acceptable to water
utility operators since the use of high chlorine levels to
control biofilms causes other problems, including exces-
sive trihalomethane (THM, a disinfection by-product) for-
mation; customer complaints about chlorinous tastes and
odors, and increased corrosion rates.
LeChevallier et al. (1988a,b) indicated that various disin-
fectants may interact differently at biofilm interfaces. In a
companion study, LeChevallier et al. (1990b) found that
low levels (1 mg/L) of either free chlorine or mono-
chloramine could reduce viable counts by greater than
100 fold (2 logs) for biofilms grown on galvanized, cop-
per, or PVC (plastic) pipe surfaces. However, free chlo-
rine residuals ranging from 3 to 4 mg/L were ineffective
for biofilm control when the microorganisms were grown
on iron pipes. In this situation, only monochloramine re-
siduals greater than 2.0 mg/L were successful for reduc-
ing biofilm viable counts. Haas et al. (1991) have
modeled the interaction of free chlorine and mono-
chloramine with biofilm surfaces and suggested that free
chlorine, because of its high reaction rate, was largely
consumed before it penetrated the biofilm. Because
monochloramine is more limited in the types of com-
pounds with which it will react (Jacangelo et al., 1987), it
is better able to penetrate the biofilm layer and inactivate
attached organisms.
The inability of the disinfectant to penetrate distribution
system biofilms can account for the occurrence of coli-
form bacteria in highly chlorinated waters. A better un-
derstanding of the interaction of disinfectants with
distribution system interfaces is necessary to formulate
appropriate strategies for biofilm control. The remaining
sections consider these interfaces, including the pipe
surface and the particles adsorbed to it.
Corrosion
Corrosion provides a protective surface for microorgan-
isms, slows water flow, and contributes to backflow oc-
currences where iron pipe walls corrode. Corrosion of
distribution system pipes can be due to chemical, physi-
cal, or biological action (O'Connor and Banerji, 1984). In
iron pipes, electrochemical reactions at the pipe surface
dissolve the metal to form pits (releasing free ferrous
ions) at one point while building a tubercle or nodule
(composed of ferric hydroxide) at a remote spot. The pits
and nodules formed may catch and concentrate nutrients
and provide the organisms with protection from water
shear (Allen and Geldreich, 1977; Victoreen, 1977,1980,
1984). In scrapings from inside several distribution sys-
tems, high levels of coliforms were found to be associ-
ated with iron tubercles (LeChevallier et al., 1987;
Photo of corroded pipe
Opheim et al., 1988). LeChevallier et al. (1987) found
that corrosion of the iron pipe surface could protect HPC
and coliform bacteria from disinfection by free chlorine.
Free chlorine itself promotes the pitting type of corrosion
by reacting with the ferrous ions and precipitating ferric
hydroxide. This not only accelerates corrosion but also
represents another demand on the free chlorine residual
(U.S. EPA, 1984). Some corrosion also may be caused
by iron or sulfur bacteria (Jarvis, 1990).
Victoreen (1977, 1980, 1984) indicated that iron may be
an important nutrient for microbial growth. He found that
substantial coliform growth was stimulated by iron oxides
found in distribution system tubercles. Under these con-
ditions, coliforms could increase to 2 x 108 bacteria/100
mL within 90 hours at 20°C. Armstrong et al. (1981)
found that increases in copper levels due to corrosion of
household plumbing can increase the proportion of multi-
ply antibiotic-resistant bacteria, At the same time, copper
ions also can cause injury to coliform bacteria, making
detection of these organisms difficult by conventional
media (Domek et al., 1985).
Sediment Accumulation
Sediments and debris in pipe systems can provide habi-
tats for microbial growth and protection from disinfection.
Carryover of aluminum floe from primary treatment or im-
proper formation of calcium carbonate scale (used in the
distribution system to protect pipes against corrosion)
may form uneven deposits on pipe walls, increasing the
concentration of organic compounds available for assimi-
lation and protecting bacteria from disinfection (Dixon et
al., 1988).
Organic and inorganic sediments can transport microor-
ganisms into the distribution system and provide protec-
tion from disinfection. Carbon fines from application of
powdered activated carbon and granular activated car-
bon filters in the treatment system can break through the
treatment process and enter the distribution system
(Camper et al., 1986; Stewart et al., 1990). Because
carbon particles are black, they may not be detected by
13
-------
AOC Levels and Coliform Growth: Experience at the New Jersey American Water Company
The first indication of the relationship between growth of
coliform bacteria and AOC levels was observed in 1986
(LeChevallier et a!., 1987). The New Jersey American
Water Company (then named Monmouth Consolidated)
had been experiencing episodes of elevated coliform lev-
els since 1984. AOC determinations at various sites in
the distribution system showed high levels in the plant ef-
fluent and lower levels as the water flowed through the
pipe network (Figure 3-4). AOC levels declined rapidly
over distance and time: At Site 1 (0.7 miles [1.13 km]
from the treatment plant, about 1 hour flow time) the
AOC level had declined 37 percent, while at the dead-
end site a short 2,000 ft beyond, AOC had declined to 40
percent of the original starting level. When an E. coll iso-
late from another portion of the distribution system was
inoculated into dechlorinated water samples, cells grew
in the plant effluent and Site 1 water, but not in water
from the dead-end site (Figure 3-5). The amount of
growth in each sample was proportional to the amount of
AOC and suggested that growth of coliform bacteria
could be limited at AOC levels less than 50 u.g/L.
These studies were followed by a survey of factors that
could contribute to growth of coliform bacteria in water
(LeChevallier et al., 1991 a). The research found that
growth of coliform bacteria was related to a complex set
of nutritional and physicochemical parameters. Of the
nutritional parameters examined, only AOC levels de-
clined as the water moved through the distribution sys-
tem (Table 3-1). Overall, AOC levels in the dead-end
site were 73 u.g/L lower than levels in the plant effluent. It
was calculated that this amount of carbon could provide
sufficient nutrients to support the growth of 7 x 104 bacte-
ria/mL. The decrease in AOC levels as the water flowed
through the distribution system was greatest during the
summer months when microbial activity was the highest.
During August 1987, when coliform levels were the high-
est (averaging up to 5 coliforms/100 mL), the concentra-
tion in the dead-end site, the test site furthest from the
entry point, averaged almost 300 u,g/L lower than that in
the plant effluent.
AOC ug
acetate
carbon/I
Plant
Effluent
Site 1
Site
Figure 3-4. Changes in AOC in a New Jersey distribu-
tion system, August 23, 1986. T-Bars represent stand-
ard deviations; AOC values at each site were
significantly (p < 0.05) different (LeChevallier et al.,
1987)
Reprinted from Applied and Environmental Microbiology, by permis-
sion. Copyright 1987, American Society for Microbiology.
E. coli
bacteria
/ml .
500
400
300
200
100
Maximum
Count
Inoculated
Count
Plant
Effluent
Site 1
Site 4
Figure 3-5. Growth of Escherichia coli in the New Jersey
distribution system samples, August 23, 1987 (LeCheval-
lier et al., 1987).
Reprinted from Applied and Environmental Microbiology, by permis-
sion. Copyright 1987, American Society for Microbiology.
Table 3-1. Changes in bacterial nutrients at various points in the New Jersey American Distribution System
Site
1
2
3
4
Nitrate-N
(mg/L)
0.64
0.64
0.64
0.66
Nitrite-N
(mg/L)
0.05
0.05
0.04
0.04
Ammonia-N
(mg/L)
0.01
0.01
0.02
0.03
Ortho-P
(mg/L)
0.13
0.12
0.14
0.14
Total-P
(mg/L)
0.15
0.15
0.18
0.14
TOC
(mg/L)
2.31
2:31
2.32
2.31
AOC
(mg/L)
214
145
134
141
Abbreviations: Ortho-P, ortho-phosphate; Total-P, total phosphates; TOC, total organic carbon; AOC, assimilable organic carbon
(LeChevallier etal., 1991 a)
14
-------
Most of the coliform occurrences in the New Jersey
American system could be related to AOC levels be-
tween 100 and 2;000 u,g/L An experiment conducted
during the first 12 days of August 1988 showed fluctua-
tions in AOC levels (ranging as high as 170 to 900 u.g/L)
in treated drinking water. When these data were com-
pared to coliform densities in the drinking water 7 days
later, the peaks in AOC concentration corresponded to
peaks in coliform density (Figure 3-6). The time delay be-
tween AOC and coliform peaks was thought to be due to
transport of the water through the distribution system and
growth of the coliform bacteria. (
Monitoring of the New Jersey American system through
August 1990 showed a continued relationship between
AOC levels and occurrences of coliform bacteria
(LeChevallier et al., 1992). Peak coliform levels were re-
lated to total AOC levels averaging 180 to 260 u.g/L At
AOC levels less than 100 u.g/L, coliform densities were
generally 0.1 bacteria/100 ml_ or less.
Under the new Total Coliform Rule (U.S. EPA, 1989b),
which took effect in 1991, coliform occurrences greater
than 5.0 percent per month trigger a violation of the MCL.
The results shown in Figure 3-7 indicate that the New
Jersey American Water Company exceeded this MCL in
June 1989, when AOC levels averaged 260 u.g/L Gener-
ally, when total AOC levels were below 100 u.g/L, coli-
form occurrences were less than 1 percent per month.
1000
100
o>
o
o
2 3 4 5 6 7 8 9 10 11 12
Days (August 1988)
-•- Coliform --»- AOC
Figure 3-6. Relationship between daily fluctuations in AOC
levels and distribution system coliform levels, August 1 to
12, 1988. Coliform data have been offset by 7 days (Au-
gust 8 to 19) (LeChevallier et al., 1991 a).
Reprinted from Applied and Environmental Microbiology, by permis-
sion. Copyright 1987, American Society for Microbiology
4)
'55
o
Q.
E
0
<*•
"o
o
a?
o.vw
6.40-
4.8O-
3.2O-
1.6O-
O.OO
j '
A
//V
V +'/ \V~\
/ ' \
V | t V . x**"
I ( + ^v. ^*"
\ ' \ -*x r*
' *- -•. **
"*~"*-v / \ ""* ***
** V /"
— i — i — : — i — : — i — i — i — r-^H1 — r V •! — ^i — i w'.-.,,^tL — —I-
300
o
o
e
DJFMAMJJASONDJFMAMJJA
Month, 1988-1990
—»- Totat AOC
Figure 3-7. Relationship between percent coliform-positive
samples and total AOC levels, New Jersey American
Water Company, 1988 to 1990 (LeChevallier et al., 1992).
To Plant
-0.7 mi "7/^3o inch
New Jersey American distribution system study area
Reprinted from Applied and Environmental Microbiology, by permis-
sion. Copyright 1987, American Sociely for Microbiology
15
-------
turbidimeters, which would otherwise alert the system
operator to a breakthrough. As a result, bacteria on the
surfaces of these particles are carried into the distribution
system protected from disinfection by their attachment
(LeChevallier et al., 1984). If these organic and inorganic
sediments accumulate in dead-end and low-flow areas of
the distribution system, they can provide protection and
nutrients for significant microbial activity.
16
-------
CHAPTER 4
How to Recognize a Biofilm Occurrence
Pathogens may occur in drinking water supplies due to
breakthrough of contamination into the distribution sys-
tem from the treatment facility; disruption of the integrity
of the distribution system (e.g., cross connections or pipe
breaks); or growth of bacteria in distribution system
biofilms. It is usually difficult to distinguish with reason-
able certainty between coliforms associated with biofilms
and those from other sources. For example, low-level
breakthrough contamination may subsequently result in
growth in the distribution system. It is important, there-
fore, to thoroughly examine treatment practices and dis-
tribution system maintenance procedures (which can
detect and control contamination events) before deciding
that growth of biofilm bacteria is the cause of excess total
bacteria and/or coliform levels. This decision is no trivial
matter because contamination may be intermittent or not
detectable by traditional monitoring methods.
The conclusion that bacteria are growing in the distribu-
tion system often is based on negative findings, i.e, an in-
ability to find an alternative cause, such as a problem in
the treatment system. However, once the available infor-
mation supports the consistent reliability of the treatment
procedures and integrity of the distribution system, the
water system should turn its attention to locating and
controlling the growth of biofilm bacteria, particularly fecal
coliform bacteria.
Detection of Breakthrough Contamination
Characteristic of breakthrough and regrowth events is a
large initial episode of coliform organism occurrence fol-
lowed by a gradual decline in bacterial levels over time,
possibly as long as several months. If a system experi-
ences occurrences of coliform bacteria, the first priority is
to determine whether fecal contamination has occurred.
The criteria listed in Table14-1 are intended to help rule
out a treatment failure or cross connection. Criteria #1a
and #4 address the detection of E coll in treated efflu-
ents and distribution system samples. Although E coll is
generally harmless and may be found in distribution sys-
tem biofilms not associated with pathogens, its presence
may be an indication of recent fecal contamination, and
immediate steps should be taken to protect public health.
Criteria #1 and #5 in Table 4-1 demonstrate adequate
treatment plant performance. If coliform bacteria, spikes
of turbidity, or periods of low chlorine residual are de-
tected, then treatment efficiency is suspect. Other indica-
tors of treatment deficiency may include increases in
particle counts, heterotrophic bacteria, or changes in the
number of non-coliform background bacteria in the mem-
brane filter (MF) total coliform test. Treatment plant moni-
toring should include not only the treatment plant effluent
but also individual filter effluents. It is possible for the
faulty performance of one filter in a series (or in parallel)
to be masked, or averaged, by the good performance of
the other filters. Particulates and microorganisms from
the faulty filter can enter the distribution system and be
responsible for bacterial problems. Wierenga (1985)
identified several sources of contamination in the Grand
Rapids system including turbid discharge from a treat-
ment filter, seepage of rainwater into the filter beds, cross
connections, and leaks in the clean/veil, where the treated
water is held after disinfection. Only by intensive exami-
nation were operators at Grand Rapids able'to locate and
correct these problems. Each contamination event, how-
ever, resulted in a prolonged occurrence of coliform or-
ganisms in the distribution system.
Breakthrough of coliform organisms in treatment plants
may occur even when effluents are apparently of good
microbiological quality. Incomplete disinfection may only
injure bacteria, which may not be detected using stand-
ard coliform media (LeChevallier and McFeters, 1985;
McFeters et al., 1986). Obseivations made by McFeters
et al. (1986), McFeters (1989), and Kippen (1986) indi-
cate that injured coliform bacteria in treatment plant efflu-
ents may be recoverable on conventional media after
spending some time in the distribution system. Walters
and McFeters (1990) showed thatJnjured bacteria can re-
pair cellular lesion and resuscitate in biofilms. A medium
to recover injured coliform bacteria (m-T7 agar) is com-
mercially available. For chloraminated systems, m-T7
agar should be supplemented with 0.1 percent, sodium
sulfite (Walters et al., 1989). Several reports describe
.situations in which the detection of injured coliforms in
treatment plant effluents has helped plant operators de-
tect and correct microbiological problems (McFeters et
al., 1986; Clark, 1988; Bucklin et al., 1991).
Bacteria also may break through treatment barriers by at-
tachment to organic or inorganic particles. Because tur-
17
-------
Tablo 4-1. Criteria for Obtaining a Variance to the Total Coliform Rule
The following criteria serve as a guidance for states in identifying systems that could operate under a variance without posing an
unreasonable risk to health:
1) Overthe past 30 days, water entering the distribution is shown to: ,
a) be free from fecal coliform or E. coli based on at least daily sampling
b) contain less than 1 total coliform/100 mL of influent water in at least 95 percent of al! samples based on at least daily sam-
pling.
c) comply with the total turbidity requirements under the Surface Water Treatment Rule.
d) contain a continuous disinfection residual of at least 0.2 mg/L.
• \
2) The system has had no waterborne disease outbreak while operating in its present configuration.
3) The system maintains biweekly contact with the state and local health departments to assess illness possibly attributable to mi-
crobial occurrence in the public drinking water system.
4) The system has evaluated, on a monthly basis, at least the number of samples specified in the Total Coliform Rule and has not
had an E co/Apositive compliance sample within the last 6 months, unless the system demonstrates to the state that the oc-
currence is not due to contamination entering the distribution system.
5) The system has undergone a sanitary survey conducted by a party approved by the state within the past 12 months.
6) The system has a cross connection control program acceptable to the state and performs an audit of the effectiveness pro-
gram.
7) The system agrees to submit a biofilm control plan to the state within 12 months of the granting of the first request for a vari-
ance.
8) The system monitors general distribution system bacterial quality by conducting heterotrophic bacteria plate counts on at least
a weekly basis at a minimum of 10 percent of the number of total coliform sites specified for that system size in the Total Coli-
form Rule (preferably using RaA medium and the procedure outlined in Standard Methods [AWWA, 1989]).
9) The system conducts daily monitoring at distribution system sites approved by the state and maintains a detectable disinfec-
tant residual at a minimum of 95 percent of those points and a heterotrophic plate count of less than 500 colonies/mL at sites
measured without a disinfectant residual. v t-
Source: U.S. EPA, 1991.
bidity interferes with detection of coliform bacteria by the
membrane filter method (LeChevallier et al., 1981), parti-
cle-associated bacteria may not be detected in plant ef-
fluent samples.
The solution to solving breakthrough and subsequent
growth problems is to eliminate the source of contamina-
tion (Geldreich et al., 1972). A thorough sanitary survey
can help detect treatment deficiencies and distribution
system problems (e.g., cross connections, breaks in
pipes, backsiphonage). Application of microbiological
media that will support and allow identification of injured
coliforms (e.g., m-T7 agar), an intensive sampling re-
gime, large volume analysis, or desorption of particle-
associated bacteria, however, may be necessary to iden-
tify sources of contamination.
Some utilities have used an automatic sampler to aid in
the investigation of microbiological episodes. Some sam-
plers can be programmed to collect 250 mL into an indi-
vidual bottle every hour. The sample line is flushed first,
then thiosulfate is automatically added as the sample is
collected. The 24 bottles are stored in a refrigerated
compartment until analysis. In several investigations, us-
ing an automatic sampler has provided the necessary
information to adequately address the contamination
problem. Pipes and Minnigh (1990) evaluated the use of
a composite sampler to improve coliform detection in fin-
ished drinking water. The researchers found that the
probability of detecting contamination events increased
through the use of an automatic sampler.
Another important criterion listed in Table 4-1 is mainte-
nance of an effective disinfectant residual throughout the
distribution system (Criterion #9). Coliform and HPC bac-
teria may grow in distribution system sections that are
unable to maintain an effective disinfectant residual
(McCabe et al., 1970; Geldreich et al., 1972). McCabe et
al. (1970) showed that free chlorine levels of 0.2 mg/L or
more were associated with HPC levels (standard pour
plate technique) of less than 500 cfu/mb in 98 percent of
the water samples.
Detection of Biofilms
In systems with biofilm problems, the phenomenon of
bacterial growth is best characterized by the persistent
occurrence of coliforms in the treated drinking water.
18
-------
Several factors distinguish chronic coliform growth in dis-
tribution systems (LeChevallier et al., 1987, 1990b,
Geldreich, 1990b):
• No coliform organisms (or extremely low counts) are
detected in treatment plant effluents even when sens!-'
tive methodologies (m-T7 agar, high-volume sample
analysis) are employed.
• High densities of coliform bacteria are routinely de-
tected in distribution system samples.
• Coliform bacteria persist in distribution system samples
despite the maintenance of a disinfectant residual.
• The duration of the coliform episode is prolonged
(several years).
• Proper operations and maintenance practices have
been carried out, including:
• Consistently maintaining positive pressure in the
distribution system.
• Implementing an aggressive cross connection con-
trol program.
• Thoroughly flushing pipes after repairs and new
construction. •
Lab technician performing a test to detect E. colt in a
drinking water sample
When coliform growth occurs, the increased bacterial
levels typically occur as randomized patterns in different
types of pipes, valves, and fittings throughout the distri-
bution system. In severe cases, the occurrence can be
nearly continuous, even though coliform counts for water
entering the distribution system are below 1 per 100 ml_.
In less severe cases, coliform occurrence may be spo-
radic, random throughout the system, and last for short
periods of time (although these short episodes may oc-
cur repeatedly over several years). Such occurrences
often are not associated with treatment disturbances.
Often, no other water quality parameters (e.g., HPC and
chlorine levels, water temperature) indicate any deterio-
ration in water quality; the only deviation from normal
water quality is the coliform level in the sample (Hubbs,
1991).
Coliform occurrence due to the presence of a biofilm first
may appear to be the result of laboratory contamination,
especially if identification of the coliform isolates is not
performed. Any utility experiencing possible biofilm prob-
lems must take extra measures to establish quality con-
trol in the bacteriological laboratory. The laboratory must
record the order of analysis of the samples and the
equipment used (e.g., numbering the funnels) to verify
that randomness of the coliform occurrence is not due to
contamination of laboratory equipment (Hubbs, 1991).
Characteristics of Biofilm Problems
Several characteristics of the bacterial population in the
distribution system may point to the development of a
biofilm: seasonality, density, types and diversity of bacte-
ria, and the persistence of coliforms in spite of a disinfec-
tant residual.
Seasonality
Smith et al. (1989) reported that coliform occurrences fre-
quently show a seasonal distribution that may be charac-
teristic of biofilm problems for systems in temperate
climates. The typical pattern was:
• Increased recovery of distribution system coliforms
usually began in March or April.
• The greatest percentage of coliform-positive samples
usually occurred in July or August.
• More than 50 percent of the samples were coliform-
positive for several sampling periods.
• Coliform occurrences usually began to subside by
mid-October.
An investigation of biofilm problems at the South Central
Connecticut Regional Water Authority in New Haven re-
vealed that coliform occurrences began in the spring and
peaked in August. (Figure 3-1 shows the mean coliform
densities during each season.) A statistical analysis of
variance (ANOVA) showed that coliform P/A was signifi-
cantly different between the seasons. Coliforms were
found throughout the system, where water from two dif-
ferent sources mixed, and in two hydraulically isolated
systems (Smith et al., 1989).
A seasonal pattern to coliform occurrence does not nec-
essarily point to a biofilm problem, however, because at
warm temperatures the concentration of organisms in the
source water will increase as well.
Density Pattern
During warm weather, or when there is a release of some
biofilm material, it is not unusual to detect coliform densi-
ties ranging from 1 to 12J5 organisms per 100 mL
(Geldreich, 1988).
In a study of a New Jersey distribution system, LeCheval-
lier et al. (1987) noted that a 20-fold increase in coliform
densities occurred in a 0.7 mi-long (1.13 km-long) seg-
ment of transmission line near the treatment plant. This
19
-------
increase could not be explained by bacterial growth in the
water. Computer modeling of the distribution system
showed that the flow time between the plant effluent and
the sampling point was between 97 and 102 minutes.
The researchers calculated that the bacterial growth rate
in the water would have had to exceed one division every
0.5 hr, a rate not possible given the low nutrient levels,
low temperature, and high chlorine residuals in the trunk
lines. There was only a low level of injured coliforms rela-
tive to the total number recovered, suggesting that the
bacteria had not passed through the treatment system.
Finally, because cross connection and back siphonage
had been ruled out as causes of coliform occurrence,
biofilms were considered the most likely explanation.
The finding of bacterial growth close to the plant, despite
the high disinfectant residuals, is not unexpected. This is
the point where nutrients enter the distribution system
and provide the first opportunity for growth. Characklis et
al. (1988) described growth of biofilms in a series of reac-
tors following treatment. The greatest growth occurred in
the first reactor where nutrients were first available. As
chlorine residuals were increased, biofilm growth was
jjushed" farther back into the system.
Coliform Species Diversity
Geldreich (1988) noted that Klebsiella or Enterobacter
coliform species usually predominate in systems experi-
encing biofilm problems. Camper et al. (1991) examined
the growth rates of clinical and environmental coliform
isolates, on the theory that the higher growth rate of a
certain strain can help it become established in the
biofilm. However, higher diversity of coliform species in
the distribution system also may indicate that a biofilm is
PERCENT
OF
ISOLATES
SITE
Plant
123456
BACTERIAL STRAIN
Figure 4-1. Coliform diversity at various sites in a New
Jersey distribution area. Bacterial strains: 1, Ł. Co//; 2,
Klebstella pneumonias; 3, K. oxytoca; 4, E. agglomerans;
5, Ł. cloacae; 6, others (LeChevallier et al., 1987).
Reprinted from Applied and Environmental Microbiology, by permis-
sion. Copyright 1991, American Society for Microbiology.
providing a favorable environment for bacterial growth. In
the New Jersey distribution system studied by LeCheval-
lier et al. (1987), only 3 species were identified in the first
part of the distribution system, while 6 to 10 were iso-
lated from the ends of the study area (Figure 4-1).
Coliform Persistence in the Presence of a
Disinfectant Residual
If a biofilm is present, coliforms may be recovered in dis-
tribution system samples even in the presence of a free
chlorine residual. In 1984, Klebsiella pneumonia isolates
were recovered from the distribution system of the South
Central Connecticut Regional Water Authority, frequently
when free chlorine residuals exceeded 5 mg/L (Ludwig,
1985). Similarly, coliform levels averaging 19 cfu/100 rnL
were recovered from the New Jersey American Water
Company during the month it averaged free chlorine re-*
siduals of 4.2 mg/L (LeChevallier et al., 1987). Earnhardt
(1980) reported recovering 51 coliform bacteria/100 m'L
in samples containing between 10 and 12 mg free
chlorine/L.
The following protocol can help determine if coliform
bacteria in the water may have come from a biofilm
(Geldreich, 1986; LeChevallier et al., 1987):
Select one of the distribution locations where there
are coliforms as well as adequate disinfection, and
collect two samples: one in a sterile sample bottle
containing no sodium thiosulfate, the other in a bottle
with sodium thiosulfate. Hold the sample without so-
dium thiosulfate for 10 min (out of the light) and then
add 0.01 percent sodium thiosulfate. Repeat this ex-
periment a number of times. If, upon examination,
the two samples contain an equivalent number of
coliforms, the implication is that the bacteria are ag-
gregated or on particles and may have come from
biofilms near the point of sampling.
Increases in the Concentration ofHPC Bacteria
Graphing trends in HPC levels may help determine the
presence of a biofilm. Growth of heterotrophic bacteria
frequently occurs before coliforms are detected in tap
samples (Geldreich, 1986). Although background levels
of HPC bacteria will vary among systems, levels of more
than 1,000 cfu/mL may indicate a growth problem
(Geldreich, 1990a). Table 4-2 summarizes the distribu-
tion and species of HPC bacteria found in the New Jer-'
sey American distribution system (LeChevallier et al.,
1987).
The method used to monitor the level of heterotrophic
bacteria can have a significant effect on the number of
bacteria recovered. LeChevallier et al. (1987) reported
that the standard pour plate method produced counts as
much as 2.5 x 10 -fold less than spread plate counts in-
cubated on RaA agar at 20°C for 7 days. Determinations
20
-------
Table 4-2. Occurrence of HPC Bacteria In Distribution System Biofilms
Bacteria
Location*
Water
Zinc
Floe
Flushed
Sediment
Iron
Tubercle
Pipe
Surface
Pseudomonas vesicularis
Flavobacterium spp.
Pseudomonas dimlnuta
Pseudomonas cepacia
Pseudomonas picketti
Pseudomonas stutzeri
Pseudomonas fluorescens
Pseudomonas putida
Pseudomonas paucimobilis
Pseudomonas maltophilia
Alcaligenes spp.
Acinetobacter spp.
Moraxella spp.
Agrobacterium radiobacter
Arthrobacterspp.
Corynebacterium spp.
Bacillus spp.
Yeasts
CDC group IIJ
Enterobacter agglomerans
Micrococcus spp.
*,++ indicates predominant organisms in that location.
Source: LeChevallier et al., 1987.
of HPC levels using RaA agar may indicate fluctuations
in bacterial populations that are undetected by less sen-
sitive methods. It is recommended that HPC monitoring
suggested in Criterion #8 (Table 4-1) be performed using
RaA agar.
Examination of Pipe Surfaces
A direct method of biofilm analysis .is examination of the
pipe surface itself. This approach is complex and should
be considered a long-term research approach. Because
coliform bacteria occur at specific, discrete locations
within the distribution system (LeChevallier et al., 1987),
a random sampling of pipe surfaces may not detect
these organisms; coliform bacteria have not been iso-
lated by all researchers examining distribution pipe sur-
faces. Tuovinen and Hsu (1982) failed to recover
coliform organisms in 24 tubercle samples they exam-
ined. Coliform bacteria initially were not detected in 15
tubercle samples collected from the New Haven distribu-
tion system (Characklis et al., 1988); however, followup
experiments (Opheim et al., 1988) detected coliform bac-
teria in tubercles flushed from the distribution system.
Coliform isolates including Enterobacter cloacae, E. ag-
glomerans, E. alvei, E. sakazakii, Citrobacter freundii,
Klebsiella pneumonia, and K. oxytocia were similar to
isolates recovered from the water with respect to bio-
chemical tests, antibiotic resistance, and plasmid
composition.
LeChevallier et al. (1987) isolated only one coliform or-
ganism from 20 pipe coupon samples collected .from six
distribution systems. This inability to detect coliform bac-
teria may be more related to the sampling methodology
than to the lack of biofilm organisms, since the same re-
searchers detected coliform bacteria in pipeline tubercles
by scraping, or pigging, a 2,000-ft (610-m) section of dis-
tribution main. Nylon netting over the end of the pipe col-
lected all the debris scraped from the pipe surface by the
polypropylene pig.
Dpnlan and Pipes (1988) developed a corporation sam-
pling device that could be used to insert pipe coupons
into pressurized water mains (Figure 4-2). Using this
method, various pipe materials could be inserted into dif-
ferent sections of the distribution system to examine the
effects of flow velocity, water temperature, and chlorine
residuals on biofilm development relative to the type of
pipe material. In Vancouver, British "Columbia, this
method detected the growth of biofilm bacteria in the
presence of water with a chlorine'residual of 3 to 4 mg/L
(Geldreich, 1986).
21
-------
Sampling cylinders
(10 mm x 15 mm, cast iron)
1-in. Corporation stop
with Mueller thread
1 1/4-in. Thread coupling,
PVC Schedule 80
1 1/4-in. nipple,
PVC Schedule 80
Gaskets
t
Sampling rod
(rotated 90 degrees
for Illustration)
Water main
(6-in. diameter
minimum)
Safety chain attachment
• Adaptor
Safety ring and chain
Figure 4-2. Exploded view of a corporation sampling device (Donlan and Pipes, 1988).
Some utilities have installed vaults at various points in
the distribution system where sections of the active pipe
network can be removed for analysis. During analysis,
water can be diverted to a bypass line.
Measurement of Nutrient Levels
Nutrient levels play an important role in the growth of
biofilm organisms. LeChevallier et al. (1987, 1990a,
1992) showed that AOC declined as the water flowed
through the distribution system (see Chapter 3). The de-
cline in AOC levels is consistent with bacterial growth in
the distribution system. Measurement of AOC or BDOC
levels at different points in the distribution system may
help to determine the activity of pipeline biofilms. The
supplement to the 17th edition of Standard Methods
(AWWA, 1989) contains the methodology for the AOC
procedure.
Corrosion
As described in Chapter 3, accumulation of corrosion
products may provide a protective habitat for growth of
neterotrophic and coliform bacteria. Therefore, if coliform
occurrence decreases in response to enhanced corro-
sion control, biofilms may be the source of the bacterial
contamination.
In New Haven, Connecticut, the South Central Regional
Water Authority modified its corrosion control program in
September 1988 by increasing the concentration of cor-
rosion inhibitor (SHAN-NO-CORR®, a zinc metaphos-
phate mixture) from 1 mg/L to 2 mg/L (Smith et al.,
1989). Total phosphorus (P) concentrations in the distri-
bution system increased from an average of 0.31 mg P/L
to an average of 0.43 mg P/L. While no immediate effect
was observed, long-term occurrence of coliform bacteria
decreased. Comparison of weekly distribution system
coliform occurrences and turbidity averages for compara-
ble 30-week periods before and after undertaking in-
creased corrosion treatment showed a statistically
significant decrease for both variables (Figure 4-3). Mean
coliform occurrence was 12.9 percent before the in-
crease and 5.1 percent after (t = 3.88, p = 0.001). Mean
turbidity in distribution system samples was 0.19 NTU
before the increase and 0.14 NTU after (t = 5.92, p =
0.000). For 1990 and 1991, coliform bacteria have been
virtually eliminated from the New Haven system.
JFMAMJJA SONDJ FMAMJ JASOND
1988 1989
Figure 4-3. Effect of corrosion inhibitor concentration on
(A) percent coliform positive and (B) distribution system
turbidity (Smith et al., 1989).
Reprinted from Technology Conference Proceedings - Advances in
Water Analysis and Treatment, by permission. Copyright 1990, Ameri-
can Water Works Association
22
-------
Examination of Hydrodynamics
Hydrodynamic changes in the distribution system can in-
fluence the transport and detachment of materials at the
pipe-water interface. Decreased flow can lead to stagna-
tion and loss of disinfectant residuals, while sudden in-
creases in flow can result in increased biofilm detachment.
Increased coliform levels in the water may occur after
changes in distribution system hydrodynamics due to in-
creased flow resulting from fire-fighting, flow reversals, or
distribution system flushing. Costello (1984) emphasizes
periodic sampling of the distribution system to assess
problems caused by rainfall as well.
Dead-end locations may show increased coliform levels
due to loss of disinfectant residual and accumulation of
bacteria dislodged from upstream biofilms. LeChevallier
et al. (1987) found that the farther the sample site was
from the plant, the lower the free and total chlorine re-
siduals and the higher the pH and heterotrophic bacteria
levels. Statistical modeling showed that a 1.2-mg/L resid-
ual chlorine level would be required to maintain HPC lev-
els below 100 cfu/mL (LeChevallier et al., 1990b). Smith
et al. (1989) found the highest coliform levels in sections
of the distribution system with low flow (Figure 4-4). Sam-
pling sites with long service tines or long flushing times,
while below the average percentage of total coliform-
positive samples for all sites, had nearly twice the aver-
age of sites located in high-flew areas.
Sites
Average percent of
samples containing
conforms
Low-flow/dead-end areas
Areas with low internal flow rates
Areas with extended flush times,
High-flow areas
9.91
9.37
6:27
3^20
Average for all sampling sites
7.42
Figure 4-4. Average percent of samples containing coli-
forms in the New Haven, Connecticut, water system (Smith
etal.,1989).
23
-------
-------
CHAPTER 5
Biofilm Control Strategies
After determining that a biofilm problem exists, the sys-
tem manager should not assume that all coliforms are
due to the biofilm. Coliforms in the water could indicate
an important treatment or distribution system deficiency,
and therefore a potential public health threat. If potable
water samples are positive for total coliforms, particularly if
the system has had no problems in the past, the possibility
of a treatment breakthrough or cross connection should be
investigated. The system should take precautions to en-
sure that public health is protected at all times through
careful monitoring and followup testing. Figure 5-1 de-
scribes steps that can be taken to control biofilms.
J
When a biofilm problem occurs, drinking water systems
should take immediate steps to limit the factors that favor
bacterial growth. Sometimes a task force can be formed
Comprehensive
Distribution System
Maintenance Program
e.g.,
-Regular flushing
-Pigging
-Pipe replacement
Reservoir
Maintenance
-Rinse prior to use
-Limit retention times
-Maintain adequate residuals
-Keep covered
Corrosion Control
e.g.,
-Use chemical inhibitors
-Adjust pH
Appropriate Disinfection
Pratices
e.g.,
-Increase free chlorine
residual
-Use alternate disinfectant
Reduced Nutrient
Levels
e.g., using
-Activated carbon filters
-Mixed carbon/sand filters
-Biologically activated filters
Best Available Technologies to Meet
the Total Coliform Rule
-Wellhead protection program
-Maintenance of disinfectant residual
in the distribution system
-Proper distribution system
maintenance
-Filtration and/or disinfection
Personnel
Training
Program
Figure 5-1. Biofilm control measures.
to deal with coliform occurrences. For example, Massa-
chusetts established a group of EPA, state health depart-
ment, and local water treatment officials to address water
quality problems. Water systems have found this ap-
proach helpful because the problems of different treat-
ment systems are not always alike, and group members'
pooled knowledge can yield a solution more quickly
(Geldreich, 1986).
The best way to avoid coliform biofilm problems is to an-
ticipate their occurrence. Knowing the factors that con-
tribute to biofilm growth gives the system a head start in
ensuring that biofilm growth is limited in the distribution
system. The system may consider instituting a coliform
biofilm control plan' before positive tests for coliforms
appear.
The Biofilm Control Plan
EPA has developed national criteria for granting vari-
ances to the Total Coliform Rule when coliforms are pre-
sent in distribution system biofilms (U.S. EPA, 1991).
One criterion for obtaining a variance to the rule is to de-
velop and implement a biofilm control plan (Criterion #7,
Table 4-1). The following section describes the strategies
that should be included in a plan to prevent and control
biofilms (U.S. EPA, 1990d). EBiofilm control measures are
described in the order of ease of implementation.
Comprehensive Maintenance Program
A maintenance program for the distribution system is
central to controlling and preventing biofilm growth. How- -
ever, routine systematic flushing, a primary component of
distribution system maintenance, is frequently neglected
due to a need to cut costs or lack of personnel. Regular
flushing helps to distribute the disinfectant residual to all
portions of the system and scour existing biofilms. Proce-
dures for designing and conducting a flushing program
have been outlined by the American Water Works Asso-
ciation (1986). More aggressive cleaning, using cable-
drawn or water-propelled devices (pigging), may be
necessary when corrosion tuberculation is severe
(AWWA, 1987).
Rae (1981) described a survey of the Colorado Springs
water treatment system to assess biofilm growth in 17
dead ends, each equipped with a hydrant. Researchers
25
-------
took samples at each dead end, two from the hydrant
placed at the dead end of the pipe, and one from the
household tap nearest the dead end. The sample from
the house represented the water that the household near
the dead end routinely received. The first hydrant sam-
ple, taken after 1 min of flow, provided an overview of the
types of organisms present in the stagnant mains. The
last sample, taken from the hydrant after the water had
cleared, revealed the quality of water supplied in active
portions of the distribution system. Bacteria, fungi, and
algae were cultured and identified in each sample. The
researchers also recorded the type of pipe, type of water
treatment, amount of chlorine residual, pH, temperature,
and turbidity at each sampling location.
They found that distance from the treatment plant was
the most important determinant of microorganism den-
sity. Dead-end sites sampled 5 mi from the point of treat-
ment showed bacterial densities more than two times
greater than those in dead ends within 1 mi of treatment.
The amount of free chlorine in the treatment effluent did
not seem to affect bacterial densities in the dead ends,
nor did the type of pipe seem to determine the bacterial
densities. PVC (plastic) and cast iron dead-end pipes
showed little difference in bacterial densities. Flushing
the dead ends via the hydrants was sufficient for restor-
ing water quality in this case (Rae, 1981).
Flushing and mechanically cleaning distribution system
lines can be effective preventive procedures, but may not
be sufficient to resolve biological growth once the prob-
lem has become severe. Increased chlorination and
flushing of the New Haven distribution system actually in-
creased coliform levels in drinking water, presumably by
releasing biofilms from the pipe surface through changes
in shear forces or oxidative processes (Centers for Dis-
ease Control, 1985). Three days after systematic flushing
of the distribution system in Muncie, Indiana, ^.con-
forms/I 00 mL were recovered just a few blocks away
from the treatment plant (Earnhardt, 1980).
In practice, it is difficult to apply flushing and pigging pro-
cedures effectively to transmission mains and trunk lines
without extreme effort, high costs, and, usually, disrup-
tion of service to customers. Flushing only sections of the
distribution system, however, has not proven effective. In
Seymour, Indiana, flushing sections of the system did not
eliminate coliform occurrences (Lowther and Moser,
1984). Flushing and mechanically cleaning a section of
the New Jersey American system did not eliminate coli-
form bacteria because the organisms were later found to
originate in other parts of the system.
In some cases, these procedures can rupture older water
pipes. It may be more economical'to replace or rehabili-
tate pipe sections than to continue to apply more tempo-
rary solutions such as flushing and pigging. The AWWA
Research Foundation (1990) has assessed various tradi-
tional and innovative water main rehabilitation practices.
When new portions are added or used to replace old
sections of the distribution system, the following AWWA
standard procedures should be used (AWWA, 1986):
» Choose pipe-joining materials that are nonporous
(e.g., plastic, rubber, or treated paper) and use non-
nutritive lubricants such as food-grade oils (White and
LeChevallier, 1991). Microorganisms can colonize
joint-packing materials, drawing nutrients from lubri-
cants used in seals. Biofilms also can derive nutrients
from surfaces of nonporous materials that release sol-
vents (Schoenen and Scholer, 1985).
• Keep new pipe sections, fittings, and valves covered
while in storage, and guard against habitation by ani-
mals. Protect the materials from contamination by soil,
runoff, and water or sewer line leaks.
• Before using new materials:
• Flush all pipes with clean water to remove visible
debris and soil. A water system in Halifax, Nova
Scotia, experienced coliform problems when a
piece of wood left in a new pipe section provided
nutrients for bacteria (Geldreich, 1986).
• Fill pipes with water containing 50 mg/L free chlo-
rine and hold for 24 to 48 hr. Chlorine levels should
not fall below 25 mg/L during the holding period.
Flushing the distribution line
Demonstration of pigging
26
-------
• Test for'total conforms and heterotrophic bacteria.
Repeat the disinfection step until no coliforms are
- detected and HPC bacteria are below 500/mL
Maintenance pf Reservoirs
Reservoirs used to store finished water are constructed
below ground, above ground, or at ground level. All are
subject to bacterial growth from a variety of factors.
Above-ground tanks, usually constructed of steel with a
corrosion-resistant liner, can suffer from bacterial growth
on the liners. These problems may be severe within the
first several months of service as organic compounds
leach from the liner surface (Frensch et al., 1987; Alben,
1989). These reservoirs should be rinsed prior to use, re-
tention times should be limited, and adequate disinfec-
tion residuals should be maintained. Frequent monitoring
of the reservoirs, especially immediately after placement
into service, can help detect problems before they be-
come severe.
Below-ground basins and ground-level tanks are prone
to bird, animal, and human contamination. Reservoirs of
treated water always should be covered to guard against
contamination by animals, birds, insects, air pollution, ac-
cidental spills, and surface water runoff. Several water
systems have found coliform contamination resulting
from contamination of uncovered reservoirs (Wierenga,
1985; Geldreich, 1988). Even covered reservoirs, how-
ever, can become contaminated when air is drawn
through air vents to replace exiting water; installing air fil-
ters can help guard against pollution entering the system.
Many smaller storage tanks in the western United States
are constructed of redwood, which, especially when un-
lined, supports microbial growth (Seidler et al., 1977).
The tanks should be lined, and a free chlorine residual of
at least 0.2 to 0.4 mg/L should be maintained (Geldreich,
1990a;Talbotetal., 1979).
Hospitals and health care facilities have complex plumb-
ing systems and may store water in tanks that are prone
to biofilm growth. These facilities should be aware of the
problems that can occur in these systems and apply ap-
propriate control measures (Highsmith, 1988).
Corrosion Control •
Limiting corrosion in distribution system pipes inherently
limits biofilm growth by reducing the numbers of places
available for attachment by microorganisms (AWWA,
1990). Corrosion can be monitored by direct or indirect
methods. Direct methods involve sampling scale from
the inside of the pipes or immersing coupons in the distri-
bution water for a period of time and determining the
amount of weight loss. Electropotentiaf devices (e.g., the
Corrater by Rohrback Cosasco Systems in Santa Fe
Springs, California) can provide immediate readouts of
water corrosivity. These instruments are useful because
they can provide immediate information on changes in
treatment practices. In this way, utilities can operate at a
target corrosion level. If resources and equipment are
available, x-ray diffraction and Raman (infrared) spec-
troscopy may be used to examine the pipes.
Indirect monitoring may draw information from customer
complaint logs; calculated'corrosion indices that predict
problems for the treated water, such as the Langelier
Saturation or Aggressive Indices; or analysis of water
samples from various points in the distribution system
(U.S. EPA, 1984; AWWA, 1990). Customer complaints,
when plotted on a map of the distribution system, can
help correlate frequent complaints in a distribution area
with the type of pipe materials used in that area (U.S.
EPA, 1984). The presence of iron and sulfur bacteria in
the water samples also may provide an indication that
corrosion is taking place. Because corrosion can be very
slow, indirect monitoring methods require a data base
gathered over a long time period.
Since corrosion occurs at the pipe surface and involves
chemical reactions between the pipe material and the
water, the primary methods for controlling corrosion
serve to separate the water and pipe, or change the cor-
rosive characteristics of either one. These methods in-
clude:
• Modifying the water quality (e.g., changing the pH
and/or reducing oxygen content) to make it less corro-
sive.
• Providing a protective barrier between the water and
pipe, such as corrosion-resistant linings, coatings, or
paints.
• Using corrosion inhibitors (e.g., sodium silicate or
phosphate-based inhibitors) that form a molecular
layer on the pipe surface, protecting it from the water.
Corrosion control measures such as chemical inhibitors
and pH adjustments have been shown to increase the ef-
fectiveness of free chlorine for disinfection of biofilms on
iron pipes (LeChevallier et al., 1990b; Lowther and
Moser, 1984; Martin et al., 1982). LeChevallier et al.
(1990b) showed that application of polyphosphate, zinc
orthophosphate, and pH and alkalinity adjustment re-
sulted in improved (10 to 100 fold) biofilm disinfection by
free chlorine. There is some danger, however, that the
biofilm could simply slough off and cause a coliform oc-
currence if corrosion control chemicals are improperly ap-
plied.
Lowther and Moser (1984) suggested that corrosion may
have contributed to the occurrence of coliform bacteria in
the Seymour, Indiana* distribution system. They reported
that levels of coliform bacteria'decreased within a few
weeks following the application of zinc orthophosphate.
Zinc orthophosphate also has been used successfully at
other Indiana operations to control coliform occurrences
(unpublished data). Martin et al. (1982) repprted that ad-
dition of lime to treated water supplies was an effective
method of pH and bacterial control. The authors pre-
27
-------
sented data suggesting that the high pH levels killed the
bacteria. Hudson et al. (1983) increased distribution sys-
tem pH to 10.2 and free chlorine residuals to 3 to 5 mg/L
to control coliform occurrences successfully in the
Springfield, Illinois, network. In both of these situations,
reduced corrosivhy of the water could have resulted in
improved free chlorine disinfection of biofilm organisms.
As illustrated by the experience of the South Central
Connecticut Regional Water Authority (Figure 3-5), appli-
cation of corrosion control will not have immediate ef-
fects. LeChevallier et al. (1987) also applied zinc
orthophosphate and elevated pH levels at the New Jer-
sey American Water Company. Corrosion chemicals
were applied for short periods of time, and no immediate
effects were observed. If applied over time, microbial
habitats are eliminated and the system is able to main-
tain disinfectant residuals more easily.
Appropriate Disinfection Practices
One of the first steps that utilities usually take to control
bacterial problems is to increase disinfectant residuals.
The disinfectant chosen to control bacteria originating
from distribution system pipe surfaces must be evaluated
carefully, however. The problem requires a disinfectant
capable of penetrating the biofilm and inactivating at-
tached microorganisms. The disinfectant also must be
relatively stable to be able to persist in the distribution
system. In addition, it must be potable and not produce
hazardous by-products. Removal of compounds that use
up the disinfectant through selection of appropriate treat-
ment practices, pipe relining, or main replacement—or all
three—can help maintain a disinfectant residual. In the
end, a more stable alternative to free chlorine residual
(e.g., chloramines) may be needed to help control bacte-
rial growths. Note, however, that for first-stage disinfec-
tion of pathogenic organisms in the treatment system,
chloramines are not recommended unless the utility can
demonstrate adequate inactivation of Giardia and viruses
(U.S. EPA, 1989a).
Resolving bacterial problems in situations where disin-
fection residuals are low is straightforward: The system
is flushed and disinfectant applied so that a residual is
maintained in all parts of the distribution system. In some
cases, rechlorination facilities may help boost disinfec-
tant residuals. In many cases, this can resolve the prob-
lem. It has been the practice of utilities experiencing
coliform regrowth problems to maintain high free chlorine
residuals in the distribution system in an effort to control
bacterial occurrence (Earnhardt, 1980; Martin et al.,
1982; Reilly and Kippen, 1983; Hudson et al., 1983; Cen-
ters for Disease Control, 1985; Ludwig, 1985; Olivier! et
al., 1985; LeChevallier et al., 1987). In general, free chlo-
rine residuals of 3 to 6 mg/L have been necessary to
control coliform regrowth events. However, Earnhardt
(1980) reported recovering 51 coliform bacteria/100 ml
in samples containing between 10 and 12 mg/L free chlo-
rine. If higher residuals are not effective, or if effective re-
siduals cannot be maintained throughout the distribution
system, the utility should consider using an alternative
disinfectant. "
Many in the water industry have found that applying a
second disinfectant such as a combined chlorine residual
just before the water enters the distribution system can
effectively control bacterial'levels in the distribution sys-
tem (Brodtmann and Russo, 1979; Norman et al., 1980;
Shull, 1981; Mitcham et al., 1983; Kreft et al., 1985; Dice,
1985; Means et al., 1986; MacLeod and Zimmerman,
1986). Although there is no perfect disinfectant, recent
research has suggested that monochloramine may be
more effective for biofilm control than free chlorine
(LeChevallier et al., 1988b, 1990b). As reviewed by Kreft •
et al. (1985), more than 70 utilities in the United States
effectively use chloramines for disinfection of distribution
water supplies. MacLeod and Zimmerman (1986) re-
ported that before conversion to chloramines, 56.1 per-
cent of the distribution system water samples were
positive for coliform bacteria and that, after conversion,
only 18.2 percent of the samples contained coliform or-
ganisms. Although there may be many reasons for the
reduced coliform counts, the system has remained coli-
form-free since February 1984 (MacLeod, 1989):
Hackensack Water Company converted to chloramine in
their distribution system in 1982 (Fung, 1989). Inifially the
system was dosed with a 2 mg/L chloramine residual,
but because of sporadic occurrences of what were
thought to be coliforms and evidence of nitrification in the
distribution system, the company increased the
chloramine dose to 3 mg/L in 1986. In that year, only a
few coliform bacteria were recovered during the summer
months. In August 1986, chloramine doses were in-
creased to 4 mg/L (average distribution system residuals
ranged from 2 to 3 mg/L) for the remainder of the sum-
mer. Since November 1986, ho coliform bacteria or nitrifi-
cation problems have been found in the distribution
system. The results of more recent studies (LeChevallier
et al., 1990b) suggest that biofilm control can be
achieved using chloramine levels ranging from 2 to 4
mg/L.
LeChevallier et al. (1990b) showed that there was a
"threshold" concentration at which monochloramine was
effective for biofilms on iron pipes. The authors indicated
that a 2 mg/L monochloramine residual was necessary to
inactivate attached bacteria. In actual practice, the
threshold level will likely vary depending on water quality
and pipe characteristics.
Controlling Nutrient Levels
Controlling the levels of nutrients available for bacterial
growth is the most direct means of resolving biofilm prob-
lems. Unfortunately, it is also the most difficult. To control
bacterial nutrients, utilities must adopt new monitoring
and treatment techniques.
28
-------
Data presented in Figures 3-4 through 3-7 show the rela-
tionship between AOC and growth of coliform bacteria in
the New Jersey American distribution system. Based on
these data, LeChevallier et al. (1987, 1990a,b) recom-
mended total AOC levels (measured using both P17 and
NOX strains of bacteria) of less than 100 \ig/L for utilities
trying to control growth of coliform bacteria in distribution
system biofilms. The advantage of reducing AOC in the
finished water is two fold: Not only is bacterial growth
limited, but less residual chlorine is consumed in side re-
actions with organic compounds.
One way to reduce AOC levels in water is through the
use of activated carbon filters. Granulated activated carb-
on (GAC) and powdered, activated carbon (PAC) are po-
rousparticles that adsorb.and hold organic contaminants.
They" are commonly used Jo remove contaminants that
cause taste and odor problemsJridrinking water. In the
context of biofilm control, they are agents for removing
dissolved organics from the source water, discouraging
bacterial growth. In fact, LeChevallier et al. (1991 b) noted
that many of the published descriptions of biofilm prob-
lems involve systems that did not practice GAC filtration.
Mixed filters of GAC and sand can be more effective for
reducing AOC levels than are filters made of sand alone
(monomedia filters). This is probably because GAC has a
greater surface area to support biological growth and ad-
sorb organic substrates. Bablon et al. (1988) found that
biologically active GAC/sand filters (16 in. [41 cm] of
GAC over 24 in. [61 cm] of sand) performed better than
monomedia sand filters. The GAC/sand filters had better
turbidity removal, lasted longer, used less .energy,
showed greater biological activity, and were less affected
by changes in water temperature. Although the dual me-
dia filter was not as effective for AOC removal as sys-
tems using sand filters followed by a GAC filter, the
authors concluded that mixed GAC/sand filters are an
economical and practical alternative to two filters in se-
ries.
LeChevallier et al. (1991b) observed that addition of PAC
as a sludge blanket reactor (a reactor in which coagula-
tion, flocculation, and sedimentation are combined) corre-
lated to reduced AOC levels. These results support
Hamann et al.'s (1986) findings that addition of PAC in a
sludge blanket reactor (Superpulsator) reduced instanta-
neous concentrations of disinfection by-products by 50 to
75 percent.
LeChevallier et al. (1991b) suggested that allowing mi-
croorganisms to attach to the PAC, producing a "biologi-
cally activated" sludge blanket, could further reduce AOC
levels. This type of biological treatment is a widely used
method of reducing the concentration of organic com-
pounds in source water. This treatment process encour-
ages microbial activity within the treatment system,
removing nutrients before the water passes to the distri-
bution system, and therefore limiting bacterial growth in
the distribution system. When microbial activity is suffi-
cient, biologically stable water results, because all nutri-
ents that might support significant bacterial growth in
treated effluents already have been removed. In addition,
Bablon et al. (1987) reported that rechlorination stations
in the distribution network were not needed due to im-
proved chlorine stability of biologically treated waters.
Application of biological processes may take many forms.
Biological removal of organic material has been reported
for aerated submerged media reactors (Sibony, 1982);
fluidized-bed filters (Foster, 1972; Short, 1975; Jeris et
al., 1977); slow sand filters (Eberhardt et al., 1977;
Schmidt, 1979); rapid sand filters (Eberhardt et al., 1977;
Sontheimer et al., 1978; Bourbigot et al., 1982; van der
Kooij and Hijnen, 1985); and GAC filters (Miller and Rice,
1978; Committee Report AWWA, 1981; Bablon et al.,
1986; van der Kooij, 1987). A number of recent reviews
of biological treatment are available (Sontheimer and
Hubele, 1988; Crowe and Bouwer, 1987; Faust and Aly,
1987; Rittmann and Huck, 1989).
The choice of disinfectant for primary treatment will influ-
ence biological activity in the treatment system. For ex-
ample, ozonation converts complex, long-chain, non-easily
degradabje compounds in the raw water to more readily
biodegradable substances that can be adsorbed to GAC or
consumed by microbes (Janssens et al., 1984; Baozhen
et al., 1985; Maloney et al., 1985; Bablon et al., 1986).
The increased oxygen content of water by ozonation also
stimulates microbial activity.
The increased biodegradability of ozonated compounds
may overwhelm treatment capabilities, however. Janssens
et al. (1984) reported that ozonation increased biodegrad-
able organic carbon levels to the extent that an early break-
through of organics was observed in treatment plant
effluents. Van der Kooij (1987) has indicated that in cer-
tain instances ozonation actually increased AOC levels in
treated effluents. In addition, high ozone doses may also
produce low molecular weight and polar oxidation prod-
ucts that are adsorbed less readily onto activated carbon
(Janssens et al., 1984).
LeChevallier et al. (1991b) compared the effects of using
ozone, free chlorine, chlorine residual neutralized using
sodium thiosujfate, monochloramine, and no predisinfec-
tant on a mixed GAC/sand filter's effectiveness for AOC
removal. The results suggest that many conventionally
operated GAC filters already may be achieving good
AOC removals. Preozonation increased AOC levels in
the water an average of 2.3 fold, and filter effluent AOC
levels were always increased relative to nonozonated
water. Because GAC can rapidly neutralize free chlorine,
application of free chlorine to GAC filters did not inhibit
biological activity. Appjication of chloramines to GAC fil-
ters showed an inhibitory effect relative to free chlorine,
and the stability of the chloramine residuals allowed re-
sidual disinfectant to appear in the filter effluent. HPC
bacteria levels in the prechloraminated filters were 10
29
-------
times lower than in the prechlorinated or preozonated fil-
ters.
Empty bed contact time, or the length of time each vol-
ume of water remains in contact with the adsorbent, is
another variable to consider when using biologically acti-
vated carbon. Prevost et al. (1990) reported that 62 to 90
percent of the AOC was removed within 2 min of contact
time in biologically active filters. LeChevallier et al. (1991)
showed that 5 to 10 min of contact time was sufficient to
reduce the AOC concentration in preozonated raw water
from over 450 jxg/L to below 100 jig/L.
Other Issues Related to Biofilm Control
Training/Upgrading Personnel
The technical ability and level of understanding of the
water treatment plant operator is crucial to the success of
the treatment and monitoring required under the Safe
Drinking Water Act (SDWA). The Surface Water Treat-
ment Rule, for example, states that the operator must
meet qualifications set by the primacy state. Most states
have operator license certification programs. Not only do
system operators need to have an understanding of the
regulations and requirements under SDWA, but they
must be familiar with the day-to-day operations of the
plant. System operators must be responsible for:
• Protecting public health.
• Maintaining the distribution system and source water
pump.
• Maintaining water quality according to federal- and
state-required monitoring and testing.
• Determining sources of contamination and developing
methods for managing those sources.
• Obtaining and understanding power sources used in
the treatment plant.
• Carrying out emergency procedures.
• Performing detailed recordkeeping and reporting ac-
cording to federal and state regulations.
Operators also should participate in continuing education
programs to learn new treatment and distribution system
maintenance techniques (U.S. EPA, 1990a).'Various pro-
grams are available through state trade associations, na-
tional meetings sponsored by EPA or AWWA, and
AWWA Research Foundation technology transfer confer-
ences. With experienced and knowledgeable operations
staff, utilities have greater latitude in choosing treatment
options, maintenance and analytical procedures, and
equipment.
Applying Best Available Technology
Best available technologies (BATs) are determined by
EPA based on effectiveness for removing or treating con-
taminants and availability to the regulated community.
BATs are proposed with the Maximum Contaminant
Level (MCL) for any contaminant or pollutant. Under the
Total Coliform Rule (U.S. EPA, 1989b), the BATs for con-
trolling conforms in drinking water include:
• Protection of wells from contamination by conforms
through proper well placement and construction. This
comes under the State Wellhead Protection Programs
described in the SDWA (Section 1428) and adminis-
tered by state governments.
• Maintenance of a disinfection residual of at least 0.2
• mg/L throughout the distribution system.
• Proper maintenance of the distribution system includ-
ing pipe replacement and repair, operation of water
main flushing programs, proper operation and mainte-
nance of storage tanks and reservoirs, and mainte-
nance of continual positive water pressure in all parts
of the system.
• Filtration and/or disinfection of surface water according
to the SWTR, or disinfection of ground water.
These techniques were chosen by EPA as feasible for
meeting the MCL for total coliforms, but water systems
are not required to use these particular methods. Other
state-approved methods that maintain the water quality
may be used as well (U.S. EPA, 1989b).
The BATs listed above may not be adequate for resolving
coliform occurrences in systems with biofilm problems.
Once all BATs are in place and the criteria in Table 3-1
are met, the utility may apply for a variance to the MCL
for total coliforms. The biofilm control plan, in effect, be-
comes the BAT for that system.
Consideration of Financial Burden
Both the SWTR and the Total Coliform Rule provide esti-
mates of the costs and benefits of implementing the regu-
lations (see Appendix A). As with most regulations, small
systems will pay the highest incremental cost of imple-
menting the SWTR and Total Coliform Rule. It may be
possible, however, to obtain federal or state grants or
loans for improvements. Appendix C lists several infor-
mation resources for small systems, including information
about loans and financial management assistance.
In some cases a small community can share resources
with other small communities or a larger community
through a cooperative or regional water supply authority.
Multi-community cooperative arrangements can improve
cost effectiveness, upgrade water quality* and result in
more efficient operation and management. Cooperative
approaches include:
• Centralizing functions such as purchasing, mainte-
nance, laboratory services, engineering services, and
billing. •
• Pooling resources to hire highly skilled personnel who
travel within a region.
30
-------
• Physically connecting existing systems to achieve
economy of scale.
• Creating a satellite utility that taps into the resources
of an existing larger facility without being physically
connected to, or owned by, the larger facility.
• Creating water districts that combine resources and/or
physically connect systems and provide for public
ownership of the utilities, making the facility eligible for
public grants and loans.
• Creating county or state utilities under jurisdiction of
the county or state government.
Establishing Timetables for Carrying Out a Biofilm
Control Plan
If a variance is granted, the state will prescribe a timeta-
ble for carrying out the components of the biofilm control
plan. The schedule may include progress reports that
show the state that the system is on its way to meeting
the MCL (U.S. EPA, 1990c).
Public Notification
In October 1987, EPA set forth new regulations (U.S.
EPA, 1987) for notifying the public when a water system
is found in violation of EPA regulations regarding drinking
water treatment and monitoring (53 FR 41534; 40 Code
of Federal Regulations [CFR] 141.32). These require-
ments remain in place even if a variance is granted by
the state or EPA, although notification is not necessarily
required as often.
System representatives should use the following guide-
lines regarding public notification when the water system
exceeds the limit for total coliforms (Geldreich, 1986;
U.S. EPA, 1987):
• Use the language spelled out in the Total Coliform
Rule to notify customers of the exceedance.
• Place one person in charge of answering questions
and providing information to the public and the press.
That person should be both technically knowledgeable
and able to communicate effectively with the public.
• Indicate to the public the steps under way to find and
eliminate the cause of the contamination.
• Ask state and federal regulators to assure the public
that there is no immediate health problem posed by
the exceedance, if indeed this is true.
•• Inform local health directors and hospital officials of
the MCL exceedance and describe the actions being
taken.
• Hold frequent press conferences and public meetings
to field questions. (The New Haven, Connecticut,
water supply, when faced with a chronic biofilm prob-
lem, sent a public affairs representative to a local ra-
dio talk show to provide information and answer
questions.)
• Work with reporters; reporters want to provide accu-
rate information to the public, and the system,has that
information.
In addition, larger water supply systems that provide
water to smaller distribution-only systems should help the
smaller system communicate with the public and pinpoint
and remediate problems.
31
-------
-------
CHAPTER 6
Summary
Distribution system monitoring and biofilm control strate-
gies require a thorough understanding of many aspects
of water supply and distribution, as well as information
about water chemistry and microbiology. Bacterial growth
has been found in many water systems, but the condi-
tions favoring biofilm growth have not been completely
characterized. Armed with the knowledge of the condi-
tions that allow microbes to pass into the distribution sys-
tem and the factors that favor microbial growth, the water
system can develop a comprehensive monitoring strat-
egy to identify trouble spots before they cause full-blown
problems. This strategy includes monitoring of not only
easy-to-reach outlets but also peripheral portions of the
distribution system. Consistent, thorough monitoring pro-
vides a historical data base from which to detect changes
in bacterial quality problems and determine the sources
of the contamination, whether biofilms, cross connec-
tions, or treatment breakthrough.
The biofilm control plan is not only a remediation plan but
a prevention program as well. Systems that maintain an
adequate treatment residual, flush the distribution lines
regularly, and practice good pipe maintenance will have
a lower risk of developing a biofilm problem. Figure 6-1
summarizes the steps to be taken when increased coli-
form levels are detected.
Finally, if a potential public health problem is identified,
the system should take quick action to resolve the issue
and protect the health of consumers. Providing the public
with accurate information on the problem and its status
will ensure that consumers understand the problem and
its implications without undue alarm.
Increased coliform occurrences detected
Review operations
procedures
r
Check distribution system
maintenance practices
Filtering Disinfection
system practices
(CT adequate?)
Regular Disinfection
flushing of new
pipe sections
increase
disinfection
dosage
check and/or
backwashing contact
procedures, time
filter
stabilization;
decrease nutrient
(assimilable organic carbon)
concentrations
flush entire
system, apply
adequate
disinfectant
to achieve
5 mg/L
free chlorine
Corrosion Adequate
control disinfection
system and covers for
storage tanks
and reservoirs
Proper flow
and circulation
scrape
pipe
system,
apply
corrosion
control
chemicals
cover
all reservoirs,
check
air vents;
filter if
necessary
check for
cross
connections,
modify
dead ends/
low-flow
areas
Figure 6-1. Steps for controlling biofilm growth.
33
-------
-------
CHAPTER 7
References
Adams, G.O. and F.H. Kingsbury. 1937. Experiences with chlorinating
new water mains. J. New England Water Works Assn. 60-68. As cited
in: Schoenen and Scholer, 1985.
Alben, K.T. 1989. Leachate from organic coating materials used in po-
table water distribution system. American Water Works Association,
Denver, CO.
Allen, M.J., E.E. Geldreich, and R.H. Taylor. 1979. The occurrence of
microorganisms on water main encrustaceans. Proceedings of the
Water Quality Technology Conference, Philadelphia, PA. American
Water Works Association, Denver, CO.
Allen, M.J. and E.E. Geldreich. 1977. Distribution line sediments and
bacterial regrowth. Proc. Water Quality Tech. Conf., Kansas City, MO.
American Public Health Association. 1989. Standard Methods for the
Examination of Water and Wastewater, 17th Edition. L.S. Clesceri, A.F.
Greenberg, and R.R. Trussell (eds.) American Public Health Associa-
tion, Washington, DC.
Armstrong, J.L., J.J. Calomiris, and R J. Seidler. 1982. Selection of an-
tibiotic-resistant standard plate count bacteria during water treatment.
Appl. Environ. Microbiol. 44:308-316.
Armstrong, J.L., J.J. Calomiris, D.S. Shigeno, and R.J. Seidler. 1981.
Drug resistant bacteria in drinking water. Proceedings of the AWWA
Water Quality Technology Conference, Seattle, WA. American Water
Works Association, Denver, CO.
Ashworth, J. and J.S. Colbourne. 1986. Microbial alterations of drinking
water by building services materials—field observations and the United
Kingdom water fittings testing scheme. Proc. Biodeterioration Society
Meeting, Delft, Holland.
AWWA (American Water Works Association). 1990. Recommended
practice for backflow prevention and cross-connection .control. Ameri-
can Water Works Association, Denver, CO.
AWWA (American Water Works Association). 1987. Cleaning and lin-
ing water mains. American Water Works Association, Denver, CO.
AWWA (American Water Works Association). 1986. AWWA standard
for disinfecting water mains. AWWA C651 -86. American Water Works
Association, Denver, CO. As cited in: Geldreich, 1990a.
AWWA (American Water Works Association) Research Foundation.
1990. Assessment of existing and developing water main rehabilitation
practices. American Water Works Association, Denver, CO.
Bablon, G., C. Ventresque, and R.B. Aim. 1988. Developing a sand-
GAC filter to achieve high-rate biological filtration. J. AWWA 80(12): 47-
53.
Bablon, G., C. Ventresque, F. Damez, and M.C. Hascoet. 1986. Re-
moval of organic matter by means of combined ozonation/BAC filtra-
tion, a reality on an industrial scale at the Choisy-Le-Roi treatment
plant. Rroc. AWWA Annual Conf., June 22-26, Denver, CO. pp. 1705-
1713.
Bablon, G., C. Ventresque, and F. Roy. 1987. Evolution of organics in a
potable water treatment system. Aqua 2:110-113.
Bagiey, S.J. 1985. Habitat association of Klebsiella species. Infect.
Control 6(2):52-58.
Baker, K.H. 1984. Protective effect of turbidity on Escherichia coli dur-
ing disinfection. Proc. Water Quality and Public Health Conf. Worcester
Consortium for Higher Education, Worcester, MA.
Baozhen, W., Y. Jun, T. Jinzhi, F. Qixiang, L. Renfen, and H. Junli.
1985. A preliminary study of the efficiency and mechanism of THM re-
moval in the ozonation and BAG process. Ozone Sci. Eng. 6:261-273.
Berg, J.D., A. Matin, and P.V. Roberts. 1983. Growth of disinfection-
resistant bacteria and simulation of natural aquatic environments in the
chemostat. In: R.L. Jolley et al. (eds.), Water Chlorination Environ-
mental Impact and Health Effects. Ann Arbor Science, Ann Arbor, Ml.
4:219-243.
Bitton, G. and K.c' Marshall. 1980. Adsorption of microorganisms to
surfaces. Wiley-lnterscience, New York.
Bourbigot, M.M., A. Dodin.and R. L'herritier. 1982. Limiting bacterial af-
tergrowth in distribution systems by removing biodegradable organics.
Proc. AWWA Annual Conf., Miami Beach, FL
Boyd, R.F. 1984. General Microbiology. Times Mirror/Mosby College
Publishing, St. Louis, MO.
Brazos, B.J., J.T. O'Connor, and S. Abcouwer. 1985. Kinetics of chlo-
rine depletion and microbial growth in household plumbing systems.
Proc. AWWA Water Quality Tech. Conf., Houston, TX.
Brodtmann, N.V., Jr. and P.J. Russo. 1979. The use of chloramines for
reduction of trihalomethanes and disinfection of drinking water. J.
AWWA7VAQ-42.
Bucklin, K.E., G.A. McFeters, and A. Amirtharajah. 1991. Penetration of
conforms through municipal drinking water filters. Water '• Res
25(8):1013-1017.
Camper, A.K., G.A. McFeters, W.G..Characklis, and W.L Jones. 1991.
Growth kinetics of coliform bacteria under conditions relevant to drink-
ing water distribution systems. Appl. Environ. Microbiol. 57:2233-2239.
Camper, A.K., M.W. LeChevalliec, S.C. Broadaway, and G.A.
McFeters. 1986. Bacteria associated with granular activated carbon
particles in drinking water. Appl. Environ. Microbiol. 52:434-438.
Carson, L.A., M.S. Favero, W.W. Bond,-and N.J. Peterson. 1972. Fac-
tors affecting comparative resistance of naturally occurring and subcul-
tured Pseudomonas aeruginosa to disinfectants. Appl. Environ.
Microbiol. 23:863-869.
Centers for Disease Control. 1985. Detection of elevated levels of coli-
form bacteria in a public water supply. Morbidity Mortality Weekly Rep.
34:142-144.
Characklis, W.G. 1981. Fouling biofilm development: a process analy-
sis. Biotechnol. Bioengin. 23: 1923-1960.
Characklis, W.G: and K.C. Marshall. 1990. Biofilms. John Wiley &
Sons, Inc., New York.
Characklis, W.G., D. Goodman, W.A. Hunt, and G.A. McFeters. 1988.
Bacterial regrowth in distribution systems'. American Water Works As-
sociation Research Foundation, Denver, CO.
Characklis, W.G., M.G. Trulear, N. Stathopoulis, and L.C. Chang. 1979.
Oxidation and destruction of microbial biofilms. In: R.L. Jolley et al.
(eds). Water Chlorination Environmental Impact and Health Effects.
Ann Arbor Science, Ann Arbor, Ml. 3:349-368.
35
-------
Ctatk, T.F. 1988. New culture medium detects stressed conforms. Op-
/fow 11(3)3-6.
Clark, T.F. 1984. Chlorine-tolerant bacteria in water distribution system.
Public Works 115:65-67.
Clark, R.M., J.A. Coyle, W.M. Grayman, and R.M. Males. 1988. Devel-
opment, application, and calibration of models for predicting water qual-
ity in distribution systems. Proc. AWWA Water Quality Tech. Conf., St.
Louis, MO, pp. 247-286.
Committee Report, AWWA. 1981. An assessment of microbial activity
on GAC. J. AWWA 73:447-454.
Committee on Water Supply. 1930. Bacterial aftergrowths in distribu-
tion systems. Am. J. Pub. Health 20:485-491.
Costolo, J.J. 1984. Postpredpitation in distribution systems. J. AWWA
76(11):46-49.
Costoron, J.W., G.C3. Geesey, and K.J. Cheng. 1978. How bacteria
stick. Scl. Amor. 238:86.
Crowe, P.B. and E.J. Bouwer. 1987. Assessment of biological proc-
esses In drinking water. AWWA Research Foundation, Denver, CO.
Dice, J. 1985. Denver's seven decades of experience with chloramina-
tfoo. J. AWWA77(\ ):34-37.
Dixon, K.L., R.G. Lee, and R.H. Moser. 1988. Residual aluminum in
drinking water. Report to the American Water Works Service Company,
Voofhoes, NJ.
Domok, M.J., M.W. LeChevallier, S.C. Cameron, and G.A. McFeters.
1985. Evidence for the rote of copper in the injury process of conforms
In drinking water. Appl. Environ. Microbiol. 48:289-293.
Donlan, R.M. and W.O. Pipes. 1988. Selected drinking water charac-
teristics and attached mtoroblal population density. J. AWWA 80:70-75.
Duncan, LB.R. 1988. Waterbome Klebsielta and human disease. Int. J.
Tbx./*ss. 3:581-598.
Earnhardt, K.B., Jr. 1980. Chlorine resistant conforms— the Muncie, In-
dfana, experience. Proc. AWWA Water Quality Tech. Conf., Miami
Beach, FL. American Waterworks Association, Denver, CO.
Ebothardt, M., S. Madsen, and H. Sontheimer. 1977. Investigations of
tho uso of biologically effective activated carbon filters in the process-
Ing of drinking water. EPA-TR-77-503, U.S. Environmental Protection
Agoncy, Cincinnati, OH.
Faust, S.D. and O.M. Aly. 1987. Absorption processes for water treat-
ment Butterworth's Publishers, Boston, MA.
Fletcher, M. and K.C. Marshall. 1982. Are solid surfaces of ecological
significance to aquatic bacteria? In: K.C. Marshall (ed.), Advances in
Microbial Ecology, Vol. 6. Plenum Press, New York, pp. 199-236.
Foster, J.D. 1972. Biological treatment of R. Trent water. Water Treat-
ment Exam. 21(4):327-333.
Fransotet, G., G. Villars, and W.J. Masschelein. 1985. Influence of tem-
perature on bacterial development in waters. J. Int. Ozone Ass. 7:205-
227.
Fronsch, K., J.U. Hahn, K. Levsen, J. Nieben, H.F. Scholer, and D.
Schooner). 1987. Solvents from the coating of a storage tank as a rea-
son of colony Increase in drinking water. Vom Wasser68:101-109.
Fung, L. 1989. Personal communication.
Gakflsh, T.J., R.L Calderon, and J.G. Grochowski. 1987. Assimilable
organic carbon determinations in different source waters and water
supply treatments, N-31. Abst. Ann. Meet. Amer. Soc. Microbiol., p.
249.
Goldrelch, E.E. 1990a. Microbial quality control in distribution systems.
In: F.W. Pontius, (ed), Water Quality and Treatment, 4th ed. AWWA
(American Waterworks Association), McGraw-Hill, Inc., New York.
GeWrelch, E.E. 1990b. Flexibility in bacteriological monitoring. Pre-
sented at the AWWA Water Quality Technology Conference, San Di-
ego, California, November 11-15.
Gatofolch, E.E. 1988. Coliform noncompliance nightmares in water
supply distribution systems, Chapter 3. In: Water Quality: A Realistic
Perspective. U. of Michigan, College of Engineering; Michigan Section,
American Water Works Association; Michigan Water Pollution Control
Association; Michigan Department of Public Health, Lansing, Ml.
Geldreich, E.E. 1986. Edited transcript of the workshop on coliform
noncompliance nightmares: scenarios and action plans. Proc. 1986
Water Quality Tech. Conf. American Water Works Association, Denver,
CO.
Geldreich, E.E. 1980. Microbial processes in water supply distribution.
Seminar presentation at the American Association for Microbiology, Mi-
ami Beach, FL. May 14. As cited in: Jarvis, 1990.
Geldreich, E.E. and E.W. Rice. 1987. Occurrence, Significance, and
detection of Klebslella in water systems. J. A WWA 79(5):74-80.
Geldreich, E.E., H.D. Nash, and D. Spino. 1977. Characterizing bacte-
rial populations in treated water supplies: a progress report. Proc.
Water Quality Tech. Conf., Kansas City, MO.
Geldreich, E.E., H.D. Nash, D.J. Reasoner, and R.H. Taylor. 1972. The
necessity of controlling bacterial populations in potable waters: commu-
nity water supply. J. AWWA 75:568-571.
Goshko, M.A., H.A. Minnigh, W.O. Pipes, and R.R. Christian. 1983.
Relationship between standard plate counts and other parameters in
distribution systems. J. AWWA 75:568- 571.
Haas, C.N., M.W. LeChevallier, and M. Geoffry. 1991. Modeling of
chlorine inactivation of drinking water biofilms. Water Research. Sub-
mitted.
Hamann, C., R.C. Hoehn, E.R. Hoffmann, and E.G. Snyder. 1986.
Evaluation of organic removal options at Newport News, Virginia. In:
P.M. Huck and P. Toft (eds.), Treatment of Drinking Water for Organic
Contaminants, PergamOn Press, New York, pp. 299-315.
Hanson, H.F., L.M. Mueller, S.S. Hasted, and D.R. Goff. 1987. Deterio-
ration of water quality in distribution systems. American Water Works
Association, Denver, CO.
Harakeh, M.S., J.D. Berg, J.C. Hoff, and A. Matin. 1985. Susceptibility
of chemostat-grown Yers/nia enterocolitica and Klebsiella pneumoniae
to chlorine dioxide. Appl. Environ. Microbiol. 49:69-72. -
Hargesheimer, E.E., G.A. Irvine, A. Badakhshan, and R.T. Seidner.
1986. Pilot studies into effects of disinfection strategies on drinking
water quality. In: P.M. Huck and P. Toft (eds.), Treatment of Drinking
Water for Organic Contaminants, Pergamon Press, New York, pp. 135-
149.
Hascoet, M.C., P. Servais, and G. Billen. 1986. Use of biological ana-
lytical methods to optimize ozonation and GAC filtration in surface
water treatment. Proc. AWWA Annual Conf., June 22-26, Denver, CO,
pp. 205-222.
Hejkal, T.W., F.M. Wellings, M.A. LaRock, and A.L. Lewis. 1979. Sur-
vival of Poliovirus within organic solids during chlorination. Appl. Envi-
ron. Microbiol. 38:114-118.
Herman, L.G. 1978. The slow-growing pigmented water bacteria: prob-
lems and sources. Adv. Appl. Microbiol. 23:155-171.
Herson, D.S., B. McGonigle, M.A. Payer, and K.H. Baker. 1987. At-
tachment as a factor in the protection of Enierobacter cloacae from
chlorination. Appl. Environ: Microbiol. 53:1178-1180.
Highsmith, A.K. 1988. Water in health care facilities. In: Architectural
Design and Indoor Microbial Pollution. Oxford University Press, Lon-
don, pp. 81-102.
Highsmith, A.K and W.R. Jarvis. 1985. Klebsiella pneumoniae:, se-
lected virulence factors that contribute to pathogenicity. Infect. Control
6(2):75-77.
Hinzelin, F. and J.C.' Block. 1985. Yeasts and filamentous fungi in
drinking water. Environ. Tech. Lett. 6:101.
Hoff, J.C. 1978. The relationship of turbidity to disinfection of potable
water. In: C.W. Hendricks (ed.), Evaluation of the Microbiology Stand-
ards for Drinking Water. EPA-570/9-78-OOC. U.S. Environmental Pro-
tection Agency, Washington D.C., pp. 103-107.
Howard, N. 1940. Bacterial depreciation of water quality in distribution
systems. J. AWWA32:1501-1506.
36
-------
Hubbs, S., 1991. Personal communication. Louisville (Kentucky) Water
Company.
Hudson, L.D., J.W. Hankins, and M. Battaglia. 1983. Coliforms in a
water distribution system: a remedial approach. J. AWWA 75(11):564-
568.
Hutchinson, M. and J.W. Ridgway. 1977. Microbiological aspects of
drinking water supplies. In: F.A. Skinner, and J.M. Shewan (eds.),
Aquatic Microbiology. Academic Press, London, United Kingdom.
Jacangelo, J.G., V.P. Olivieri, and K. Kawata. 1987. Mechanism of in-
activation of microorganisms by combined chlorine. American Water
Works Association, Denver, CO.
Janssens, J.G., J. Meheus, and J. Dirickx. 1984. Ozone-enhanced bio-
logical activated carbon filtration and its effect on organic matter re-
moval, and in particular on AOC reduction. Water Scl. Tech.
17:1055-1068.
Jarvis, W.R. 1990. Opportunistic Pathogenic Microorganisms in
Biofilms. Centers for Disease Control, Washington, DC,
Jarvis, W.R., V.P. Munn, A.K. Highsmith, D.H. Culver, and J.M.
Hughes. 1985. The epidemiology of nosocomial infections caused by
Klebsiellapneumonias. Infect. Control 6(2):68-74.
Jeris, J.S., R.W. Owens, R. Hickey, and F. Flood. 1977. Biological
fluidized-bed treatment for BOD and nitrogen removal. J. Water Pollut.
Control Fed. 49(5):816-831.
Joret, J.C., Y. Levi, T. Dupin, and M. Gilbert. 1988. Rapid method for
estimating bioeliminable organic carbon in water. Proc. AWWA Annual
Conf., Orlando, FL.
Kippen, J.S. 1986. Personal communication.
Kreft, P., M. Umphres, J-M. Hand, C. Tate, M.J. McGuire.'and R.R.
Trussell. 1985. Converting from chlorine to chioramines: a case study.
J. 4WV,477(1):38-45.
Kuchta, J.M., S.J. States, J.E. McGlaughlin, J.H. Overmeyer, R.M.
Wadowsky, A.M. McNamara, R.S. Wolford, and R.B. Yee. 1985. En-
hanced chlorine resistance of tap water-adapted Legionella pneumo-
phila as compared with agar-medium passed strains. Appl. Environ.
Microbiol. 50:21 -26.
LeChevallier, M.W. 1990. Biocides and the current status of biofouling
in water systems. In: H.-C. Flemming and G.G. Geesey (eds.), Biofoul-
ing and Biocorrosion in Industrial Water Systems, Springer-Verlag, Ber-
lin, Germany, pp. 113-137.
LeChevallier, M.W. 1989. Treatment to meet the microbiological MCL
in the face of a coliform regrowth problem. Proc. AWWA Water Quality
Tech. Conf., Philadelphia, PA. American Water Works Association,
Denver, CO.
LeChevallier, M.W. and G.A. McFeters. 1985. Enumerating injured coli-
forms in drinking water. J. AWWA 77(6):81 -87.
LeChevallier, M.W., W. Becker, P. Schorr, and R.G. Lee. 1992. Appli-
cation of biological processes to current water treatment practices. J.
AWWAS4(4). In press.
LeChevallier, M.W., W. Schulz, and R.G. Lee. 1991a. Bacterial nutri-
ents in drinking water. Appl. Environ. Microbiol. 57:857-862.
LeChevallier, M.W., R.G. Lee, and J.S. Young. 1991 b. Application of
biological processes to current water treatment practices. A report to
the American Water Works Service Company, Inc., Belleville, IL.
LeChevallier, M.W., W.H Schulz, and R.G. Lee. 1990a. Bacterial nutri-
ents in drinking water. In: Assessing and Controlling Bacterial Re-
growth in Distribution Systems. American Water Works Association
and American Water Works Association Research Foundation, Denver,
CO.
LeChevallier, M.W., C.D. Lowry, and R.G. Lee. 1990b. Disinfection of
biofilms in a model distribution system. J. AWWA82(7): 87-99.
LeChevallier, M.W., C.D. Cawthorn, and R.G. Lee. 1988a. Factors pro-
moting survival of bacteria in contaminated water supplies. Appl. Envi-
ron. Microbiol. 54(3):649-654.
LeChevallier, M.W., C.D. Cawthon, and R.G. Lee. 1988b. Inactivation
of biofilm bacteria. Appl. Environ. Microbiol. 54(10)':2492-2499.
LeChevallier, M.W., T.S. Babcock, and R.G. Lee. 1987. Examination
and characterization of distribution system biofilms. Appl. Environ. Mi-
crobiol. 53(12):2714-2724.
LeChevallier, M.W., T.S. Hassenauer, A.K. Camper, and G.A.
McFeters. 1984. Disinfection of bacteria attached to granular activated
carbon. Appl. Environ. Microbiol. 48:918-923.
LeChevallier, M.W., R.J. Seidler, and T.M. Evans. 1981. Effect of tur-
bidity on chlorination efficiency and bacterial persistence in drinking
water. Appl. Environ. Microbiol. 42:159.
LeChevallier, M.W., R.J. Seidler, and T.M. Evans. 1980. Enumeration
and characterization of standard plate count bacteria in raw and chlorin-
ated water supplies. Appl. Environ. Microbiol. 40:922-930.
LeCIerc, H. and F. Mizon. 1978. Eaux d'alimentation et bacteries resis-
tantes aux antibiotiques. Incidences sur les normes. Rev. Epidemiol.
Med. Soc. Sante Publique 26:137-1,46. (French)
Levy, R.V. 1985. Invertebrate protection of coliforms and heterotrophic
bacteria. Proc. Seminar on Current Status of Drinking Water Microbiol-
ogy. AWWA Annual Conf., Washington, DC.
Levy, R.V., R.D. Cheetham, J. Davis, G. Winer, and F.L. Hart. 1984.
Novel method for studying the public health significance of macroinver-
tebrates occurring in potable water. Appl. Environ. Microbiol. 47: 889-
894.
-Lowther, E.D. and R.H. Moser. 1984. Detecting and eliminating coliform
regrowth. Proceedings of the AWWA Water Quality Technology Confer-
ence, Denver, CO. American Water Works Association, Denver, CO.
Ludwig, F. 1985. The occurrence of coliforms in the Regional Water
Authority water supply system. A report submitted by the South Central
Connecticut Regional Water Authority, New Haven, CT.
MacLeod, B.W. 1989. Personal communication. Manatee County Pub-
lic Works Department, Bradenton, FL.
MacLeod, B.W. and J.A. Zimmermann. 1986. Selected effects on distri-
bution system water quality as a result of conversion to chioramines.
AWWA Water Quality Technology Conference, Portland, OR..
Maloney, S.W., I.H. Suffet, K. Bancroft, and H.M. Neukrug. 1985.
Ozone-GAC following conventidnal U.S. drinking water treatment. J.
AWWA 77:8:66-73.
Marshall, K.C. 1976. Interfaces in Microbial Ecology. Harvard University
Press, Cambridge, MA, and London, England.
Martin, R.S., W.H. Gates, R.S. Tobin, D. Grantham, P. Wolfe, and P.
Forestall. 1982. Factors affecting coliform bacteria growth in distribution
systems. J. AWWA 74:34.
McCabe, L., J.M. Symons, R.D. Lee, and G.G. Robeck. 1970. Study of
community water supply systems. J. AWWA 62(11 ):670-687.
McFeters, G.A. 1989. Detection and significance of injured indicator
and pathogenic bacteria in water. In: G.A. McFeters (ed.), Advances in
Aquatic Microbiology, Science Tech Publishers, Madison, Wl.
McFeters, G.A., J.S. Kippin, and M.W. LeChevallier. 1986. Injured coli-
forms in drinking water. Appl. Environ. Microbiol. 51:1-5.
Means, E.G., T.S. Tanaka, D.J. Otsuka, and M.J. McGuire. 1986. Ef-
fects of chlorine and ammonia application points on bactericidal effi-
ciency. J. A WWA 78(1 ):62-69.
Miller, G.W. and R.G. Rice. 1978. European water treatment prac-
tices—the promise of biological activated carbon. Civil Eng. 48(2):81.
Mitcham, R.P., M.W.- Shelley, and C.M. Wheadon. 1983. Free chlorine
versus ammonia-chlorine: disinfection, trihalomethane formation and
zooplankton removal. J. AWWA 75:1536-200.
Murray, G.E., R.S. Tobin, B. Junkins, and D.J. Kushner. 1984. Effect of
chlorination on antibiotic resistance profiles of sewage-related bacteria.
Appl. Environ. Microbiol. 48:73-77.'
Nagy, L.A. and B.H. Olson. 1985. Occurrence and significance of bac-
teria, fungi, and yeasts associated with distribution pipe surfaces. Proc.
Water Quality Tech. Conf., Houston, TX
37
-------
Nagy, L.A.. A.J. Kelly. M.A. Thun. and B.H. Olson. 1982. Biofilm com-
position, formation, and control in the Los Angeles aqueduct system.
Proc. Water Quality Tech. Conf., Nashville, TN.
Norman, T.S., LL Harms, and R.W. Looyenga. 1980. The use of
chloraminas to prevent trihatomethane formation. J. AWWA 72:176-
180.
O'Connor, J.T. and S.K. Banerji. 1984. Biologically mediated corrosion
and water quality deterioration in distribution systems. EPA-6 00/52-84-
056. U.S. Environmental Protection Agency, Cincinnati, OH.
Oliviori. V.P.. A.E. Bakalian, K.W. Bossung, and E.D. Lowther. 1985.
Recurrent coliforms In water distribution systems in the presence of free
residual chlorine. In: R.L Joltey etal. (eds), Water Chlorination, Chemis-
try, Environmental Impact and Health Effects. Lewis Publishers, Inc.,
Chelsea. Ml. pp. 651-666. ,
Olson, B.H. 1982. Assessment and implications of bacterial regrowth in
water distribution systems. EPA-600/52-82-072. U.S. Environmental
Protection Agency.
Olson, B.H. and B.B. Miner. 1990. Development of disinfectant-resis-
tant organisms. In: Assessing and controlling bacterial regrowth in dis-
tribution systems. American Water Works Association and American
Waterworks Association Research Foundation, Denver, CO.
Ophokn, 0., J. Grochowski, and D. Smith. 1988. Isolation of conforms
from water main tubercles, N-6. Abst. Annual Meet. Amer. Soc. Micro-
btol..p.245.
Orskov. I. 1984. Klabsielta. In: N.R. Krieg and J.G. Holt (eds), Bergy's
Manual of Systematic Bacteriology, Vol. 1. Williams and Wilkens, Balti-
more, MD, pp. 461-465.
Pascal, O.. J.C. Joret, Y. Levi, and T. Dupin. 1986. Bacterial after- •
growth in drinking water networks measuring biodegradable organic
carbon (BDOC). Proc. Ministere de I'Environnement/U.S. Environ-
mental Protection Agency Franco-American Seminar, Oct. 13-17, Cin-
cinnati, OH.
Payment, P., E. Franco. L. Richardson, and J. Siemiatycki. 1991. Gas-
trointestinal health effects associated with the consumption of drinking
water produced by point-of-use domestic reverse-osmosis filtration
units. Appl. Environ. Microbiol. 57(4):945-948.
Pipes, W.O. and H.A. Minnigh. 1990. Composite sampling for detection
of coliform bacteria in water supply. EPA/600/52-90/014 U.S. Environ-
mental Protection Agency, Cincinnati, OH.
Prevost, M., R. Desjardins, J. Coallier, D. Duchesne, and J. Mailly.
1990. Comparison of biodegradable organic (BOC) techniques for proc-
ess control. Proc. AWWA Water Quality Tech. Conf., San Diego, CA.
Rae, J.F. 1981. Algae and bacteria: dead end hazard. Proc. AWWA
Water Quality Tech. Conf., Seattle, WA. American Water Works Asso-
ciation, Denver, CO.
Roilly, J.K. and J.S. Kippen. 1983. Relationship of bacterial counts with
turbidity and free chlorine in two distribution systems. J. AWWA 75:309-
312.
RIdgway, H.F. and B.H. Olson. 1982. Chlorine resistance patterns of
bacteria from two drinking water distribution systems. Appl. Environ. Mi-
crobiol. 44:972-987.
RIdgway, H.F. and B.H. Olson. 1981. Scanning electron microscope
evidence for bacterial colonization of a drinking water distribution sys-
tem. Appl. Environ. Microbiol. 41274-287.
Rklgway, H.F.. C.A. Justice, C. Whittaker. D.G. Argo, and B.H. Olson.
1984. Biofilm fouling of RO membranes—its nature and effect on treat-
ment of water for reuse. J. AWWA 77S4-102.
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. Mi-
crobiol. 41:288-297.
Rittmann, B.E. and P.M. Huck. 1989. Biological treatment of public
water supplies. Crit. Rev. Environ. Contri. 19(2):119-184.
Rittmann. B.E. and V.L. Snoeyink. 1984. Achieving biologically stable
drinking water. J. AWWA 76(10):106-114.
Rizet, M., F. Fiessinger, and N. Houel. 1982. Bacterial regrowth in a
distribution system and its relationship with the qaulity of the feed
water: case studies. Proc. AWWA Annual Conf., Miami Beach, FL, May
16-20.
Rosenzweig, W.D. 1987. Influence of phosphate corrosion control
compounds on bacterial regrowth. EPA CR-811613-01-0, U. S. Envi-
ronmental Protection Agency, Cincinnati, OH.
Safe Drinking Water Committee. 1982. Biological quality of water in the
distribution system. In: Drinking Water and Health, Vol. 4. National
Academy Press, Washington, DC.
Schellart, J.A. 1986. Disinfection and bacterial regrowth: some experi-
ences of the Amsterdam water works before and after stopping the
safety Chlorination. Wat. Supply4:217-225.
Schmidt, K. 1979. Experience with the removal of micro-impurities in
slow sand filters. In: W. Kukn and H. Sontheimer (eds), Oxidation
Techniques in Drinking Water Treatment. EPA-570/9-79- 020, U.S. En-
vironmental Protection Agency, Cincinnati, OH, pp. 620-646.
Schoenen, D. and H.F. Scholer. 1985. Drinking Water Materials: Field
Observations and Methods of Investigation. John Wiley & Sons, New
York.
Schoenen, D. and A. Wehse. 1988. Microbial colonization of water by
the materials of pipes and hoses: changes in colony counts. Zbl. Bakt.
Hyg. 8186:108-117.
Seidier, R.J., J.E. Morrow, and ST. Bagley. 1977. Klebsielleae in drink-
ing water emanating from redwood tanks. Appl. Environ. Microbiol.
33:893-900.
Servais, P., G. Billen, and M.C. Hascoet. 1987. Determination of the
biodegradable fraction of dissolved organic matter in waters. Water
Res. 21:445-450.
Short, C.S. 1975. Removal of ammonia from river water. 2. Technical
Report TR3, Water Research Center, Medmenham, England.
Shull, K.E. 1981. Experiences with chloramines as primary disinfec-
tants. J. AWWA 73:104.
Sibony, J. 1982. Development of aerated biological filters for the treat-
ment of waste and potable water. Proc. 14th Congress A.I.D.E., Sept.
6-10, Zurich, Switzerland.
Silverman, G.S., L.A. Nagy, and B.H. Olson. 1983. Variations in par-
ticulate matter, algae, and bacteria in an uncovered, finished drinking
water reservoir. J. AWWA 75:191-195. ,
Smith, D.B., A. Hess, and S. Hubbs. 1990. Survey of distribution sys-
tem coliform occurrence in the United States. Proc. AWWA Water
Quality Tech. Conf., San Diego, CA. American Water Works Associa-
tion, Denver, CO.
Smith, D.B., A.F. Hess, and D. Opheim. 1989. Control of distribution
system coliform regrowth. Proc. AWWA Water Quality Tech. Conf.,
Philadelphia, PA. American Water Works Association, Denver, CO.
Smith-Somerville, H.E., V.B. Huryn, C. Walker, and A.L Winters. 1991.
Survival of Legionella pneumophila in the cold water ciliate Tetra-
hymena vorax. Appl. Environ. Microbiol. 57:\2742-2749.
Sontheimer, H. and C. Hubele. 1988. The use of ozone and granular
activated carbon in drinking water treatment. In: P.M. Huck and P. Toft
(eds.), Treatment of Drinking Water for Organic Contaminants, Per-
gamon Press, New York, pp. 45-66.
Sontheimer, H., E. Heilk'er, M.R. Jekel, H. Nolle, and F.H. Vollmer.
1978. The Mulheim process. J. AWWA 70(7):393-396.
Stewart, M.H. and B.H. Olson. 1986. Mechanisms of bacterial resis-
tance to inorganic chloramines. In: Proc. AWWA Water Quality Tech.
Conf., Portland, OR. American Water Works Association, Denver, CO.
Stewart, M.H., R.L Wolfe, and E.G. Means. 1990. Assessment of the
bacteriological activity associated with granular activated carbon treat-
ment of drinking water. Appl. Environ. Microbiol. 56: 3822-3829.
Symons, J.M., T.A. Bellar, J.K. Carswell,.J. DeMarco, K.L Kropp, G.G.
Rqbeck, D.R. Seeger, C.J. Slocum, B.L Smith, and A.A. Stevens.
1975. Natural organics reconnaissance survey for halogenated or-
ganics. J. AWWA 67:11:634-647.
38
-------
Talbot, H.W., Jr., J.E. Morrow, and R.J. Seidler. 1979. Control of coli-
form bacteria in finished drinking water in redwood tanks J AWWA
71(6):349.
Tracy, H.W., V.M. Camarena, and F. Wing. 1966. Coliform persistence
in highly chlorinated waters. J. AWWA58:1151-1159.
Tuovinen, O.H. and J.C. Hsu. 1982. Aerobic and anaerobic microor-
ganisms in tubercles of the Columbus, Ohio, water distribution system.
Appl. Environ. Microbiol. 44:761 -764.
Tuovinen, O.H., K.S. Button, A. Vuorinen, L. Carlson, D.M. Mair, and
L.A. Yut. 1980. Bacterial, chemical, and mineralogical characteristics of
tubercles in distribution pipelines. J. AWWA 72:626-635.
U.S. EPA (U.S. Environmental Protection Agency). 1991. Drinking
Water; National Primary Drinking Water Regulations; total coliforms
Fed. Reg. 56(10):1556-1557. January 15.
U.S. EPA (U.S. Environmental Protection Agency). 1990a. Draft guid-
ance in developing health criteria for determining unreasonable risks to
health. Office of Drinking Water, U.S. EPA, Washington, DC.
U.S. EPA (U.S. Environmental Protection Agency). 1990b. Environ-
mental pollution control alternatives: drinking water treatment for small
communities. EPA/625/5-90/025. Center for Environmental Research
Information, U.S. EPA, Cincinnati, OH.
U.S. EPA (U.S. Environmental Protection Agency). 1990c. Fact sheet:
drinking water regulations under the Safe Drinking Water Act. Criteria
and Standards Division, Office of Drinking Water, U.S. EPA, Washing-
ton, DC.
U.S. EPA (U.S. Environmental Protection Agency). 1990d. Manual for
the certification of laboratories analyzing drinking water: criteria and
procedures, quality assurance. EPA 570/9-90/008. Office of Water
U.S. EPA, Washington, DC.
U.S. EPA (U.S. Environmental Protection Agency). 1990e. Summary
report: workshop on coliform biofilms in water distribution systems, No-
vember 15-16. Office of Drinking Water, Washington, DC, Center for
Environmental Research Information, U.S. EPA, Cincinnati, OH.
U.S. ,EPA (U.S. Environmental Protection Agency). 1989a. Drinking
Water; National Primary Drinking Water Regulations; total coliforms
(including fecal coliforms and E. coll); final rule Fed Reg
54(124):27544-27568. June 29.
U.S. EPA (U.S. Environmental Protection Agency). 1989b. National Pri-
mary Drinking Water Regulations: filtration and disinfection; turbidity,
Giardia lamblia, viruses, Legionella, and heterotrophic bacteria; final
rule. Fed. Reg. 54(124):27486-27543. June 29.
U.S. EPA (U.S. Environmental Protection Agency). 1987. Drinking
water regulations; public notification; final rule Fed Reg
52(208):41534-41550. October 28.
U.S. EPA (U.S. Environmental Protection Agency). 1984. Corrosion
manual for internal corrosion of water distribution systems. EPA/570/9-
84/001. Office of Drinking Water, U.S. EPA, Washington, DC.
Van der Kooij, D. 1992. Assimilable organic carbon as an indicator of
bacterial regrowth. J. AWWA 84:57-65.
Van der Kooij, D. 1990. Assimilable organic carbon (AOC) in drinking
water. In: G.A. McFeters (ed.), Drinking Water Microbiology Springer-
Verlag, New York, pp. 57-87.
Van der Kooij, D. 1987. The effect of treatment on assimilable organic
carbon in drinking water. In: P.M. Huck and P. Toft (eds.), Proc. Sec-
ond National Conf. on Drinking Water, Edmonton, Canada, April 7-8,
1986. Pergarnon Press, London, pp. 317-328.
Van der Kooij, D. and W.A.M. Hijnen. 1988. Multiplication of a
Klebsiella pneumoniae strain in water at low concentrations of sub-
strates. Proc. International Conf. Water Wastewater Microbiol. Newport
Beach, CA.
Van der Kooij, D. and W.A.M. Hijnen. 1985. Measuring the concentra-
tion of easily assimilable organic carbon (AOC) treatment as a tool for
limiting regrowth of bacteria in distribution systems. Proc. AWWA Water
Quality Tech. Conf., Houston, TX. American Water Works Association,
Denver, CO.
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 organic carbon (AOC) in drinking water. Ozone Sci
Engineer. 11:297-311.
Van der Kooij, D., A. Visser, and J.P. Oranje. 1982. Multiplication of
fluorescent pseudomonads at low substrate concentrations in tap
water. Antonie van Leeuwenhoek 48:229-243.
Van der Kooij, D., A. Visser, and W.A.M. Hijnen. 1982. Determining the
concentration of easily assimilable organic carbon in drinking water J
x*WV,474(10):540-545.
Verstraete, W. and M. Alexander. 1973. Heterotrophic nitrification in
samples of natural ecosystems. Environ. Sci. Technol. 7(1):39-42.
Victoreen, H.T. 1984. The role of rust in coliform regrowth. Proc.
AWWA Water Quality Tech. Conf., Denver, CO.
Victoreen, H.T. 1980. The stimulation of coliform growth by hard and
soft water main deposits. Proc. AWWA Water Quality Tech. Conf., Mi-
ami Beach, FL
Victoreen, H.T. 1977. Water quality deterioration in pipelines, Proc.
Water Quality Tech. Conf., Kansas City, MO.
Ward, N.R., R.L. Wolfe, E.G. Means, and B.H. Olson. 1985. The inacti-
vation of total and selected Gram-negative bacteria by inorganic mono-
chloramines and dichloramines. Proc. AWWA Water Quality Tech
Conf., Nashville, TN, December 5-8.
Walters, S.K. and G.A. McFeters. 1990. Reactivation of injured bacte-
ria. In: Assessing and controlling bacterial regrowth in distribution sys-
tems. American Water Works Association and American Water Works
Research Foundation, Denver, CO.
Walters, S., G.A. McFeters, and M.W. LeChevallier. 1989. Reactivation
of injured bacteria. Appl. Environ. Microbiol. 55(12):3226-3228.
Wierenga, J.T. 1985. Recovery of coliforms in the presence of a free
chlorine residual. J. AWWA 77(11):83-88.
Wolfe, R.L, E.G. Means, M.K. Davis, and S. Barrett. 1988. Biological
nitrification in covered reservoirs containing chloraminated water J
4MV/I80(9):109-114.
Wolfe, R.L, N.R. Ward, and B.H. Olson, 1985. Inactivation of heterotro-
phic bacterial populations in finished drinking water by chlorine'and
chloramines. Water Res. 19:1393-1403.
39
-------
-------
APPENDIX A
Drinking Water Regulations for Microorganisms
Under the 1986 Amendments to the Safe Drinking Water
Act, EPA must set Maximum Contaminant Level Goals
(MCLGs) for any drinking water contaminant that is
known to, or may, occur in public water systems and that
may have an adverse effect on human health. An MCLG
is a nonenforceable health goal that is set at a level (with
an adequate margin of safety) at which no known or an-
ticipated adverse health effects occur. A second regula-
tory level, the Maximum Contaminant Level (MCL), is set
as close to the MCLG as possible, but takes economic
and technical feasibility into account as well as public
health considerations (represented by the MCLG). MCLs
are usually specific drinking water levels, but in some
cases required treatment technologies are set forth in-
stead. MCLs are enforceable levels, enforced either by
the state or EPA. States are allowed to enforce MCLs if
they have primary enforcement responsibility (i.e., the
state enforcement program is approved by EPA).
In June 1989, EPA promulgated final MCLs and MCLGs
for surface water treatment under the Surface Water
Treatment Rule (SWTR) (54 Federal Register [FR]
27486). EPA also set an MCL for total coliforms as part
of the TotalJDoliform Rule (54 FR 27544). The Total Coli-
form Rule applies to all drinking water treatment sys-
tems. These rules specify treatment and monitoring
requirements that must be met by all public water suppli-
ers, whether community or noncommunity systems. Non-
community systems include those serving nonresidential
populations, such as restaurants, schools, office build-
ings, and campgrounds. Public water systems (either
community or noncommunity) have at least 15 service
connections to year-round residences or serve 25 or
more persons for at least 60 days per-year. The SWTR
applies to facilities that use surface water (or ground
water that is directly influenced by surface water) as raw
water, while the Total Coliform Rule applies to all ground
and surface water.
Surface Water Treatment Rule
The MCLGs set under the SWTR are zero (0) for Giardia
lamblia, viruses, and Legionella. These organisms are
regulated via treatment requirements rather than specific
concentration levels (i.e., MCLs), because EPA believes
that monitoring for these organisms is not technically or
economically feasible, especially for small systems (U.S.
EPA, 1989a). The treatment is based on the removal or
inactivation of 99.9 percent of Giardia, and inactivation of
99.99 percent of any enteric (human intestinal) viruses as
determined by CT values. The state (or EPA) must be
sure that the system meets this requirement by enforcing
Best Available Technology (BAT) treatment options
based on certain site-specific and surface water quality
criteria.
Although EPA has not published MCLGs for turbidity or
heterotrophic bacteria, low turbidity and heterotrophic
bacteria counts suggest that adequate treatment is in
place. Therefore these measures are used under the
SWTR as indicators of treatment effectiveness. One of
the recommended variance criteria for systems using sur-
face water (see Table 4-1) is that they comply with the
SWTR, including those requirements for turbidity and het-
erotrophic bacteria.
Total Coliform Rule
The presence of coliforms in drinking water above the
MCL is cause for concern, because it may indicate that
disinfection has been inadequate to kill all organisms as-
sociated with human and animal wastes. Therefore, the
MCLG for total coliforms is zero, and the MCL limits the
percentage of samples per month that may have any to-
tal coliforms present. Total coliform measurements in-
clude E. coll, which come from the human intestinal tract,
as well as other coliform bacteria that are not normally
associated with human disease. Therefore, each sample
that is positive for total coliform (total coliform-positive) is
tested for fecal coliforms or E. coli.
The regulations specify that for systems collecting 40 or
more samples per month, the MCL permits no more than
5.0 percent total coliform-positive samples per month.
For systems that collect fewer than 40 samples per
month, no more than one sample may be total coliform-
positive in each month. The number of monthly samples
ranges from less than 1 to 480 samples, depending on
the population size served (U.S. EPA, 1989b). In addi-
tion, the system must develop a sample siting plan indi-
cating where the samples will be taken; this plan will be
reviewed and approved by the state. Other monitoring re-
quirements state that:
41
-------
• If total coliforms are detected in any sample, the sys-
tem must collect repeat samples within 24 hours and
have them analyzed for total coliforms (the 24-hour
limit may be waived by the state in extenuating cir-
cumstances).
• If total coliforms are detected, in any repeat sample,
then more repeat samples must be taken, unless the
MCL has already been exceeded and the system re-
ports this to the state.
• If a system collecting fewer than 5 samples per month
detects total coliforms in any sample, that system
must collect at least 5 routine samples the next month.
• Systems that use unfiltered surface water, or ground
water under the influence of surface water, must ana-
lyze one sample for total coliforms every day that the
turbidity of the raw water exceeds 1 NTU
(nephelometric turbidity unit).
Some public water systems have persistent coliform
biofilm problems that may not pose risks to human health
but may cause the system to violate the MCL for total
coliforms. Therefore, the Total Coliform Rule allows pri-
macy states to grant a variance (an exception to the
MCL) if the system can prove that distribution system
blofilms are the sole cause of the positive coliform re-
sults, and the contamination does not pose an unreason-
able risk to health (U.S. EPA, 1990c).
The variance summary in the Federal Register (U.S.
EPA, 1991; Table 4-1; also included in full as an Attach-
ment to this Appendix) summarizes the steps that states
may require when a system applies for a variance to the
rule because of coliform biofilms. The criteria are an at-
tempt to ensure that biofilms are the sole cause of the
occurrence and that human health will not be compro-
mised by the granting of a variance. The system must
prove that there have been no treatment lapses or defi-
ciencies, there is no measurable fecal or pathogenic con-
tamination present, and proper operation and
maintenance procedures have been carried out (U.S.
EPA, 1991). The state will examine the evidence care-
fully, possibly taking into account the criteria listed in the
variance notice. If the system is able to meet all the re-
quirements listed under Criteria #1 and #2, then the pri-
mary treatment is assumed adequate to protect human
health.
Criterion #3 directs the system to maintain biweekly con-
tact with state and local health departments so that
health officials can assess any illnesses that may be at-
tributable to microbial contamination in the finished
water. It is important that any disease associated with
drinking water contamination be recognized immediately
so that immediate action can be taken to prevent further
infection.
If Criteria #4 through #6 are met, then a cross-connection
problem can probably be ruled out as the source of con-
tamination. The remaining criteria prove that the system
is carrying out maintenance and monitoring 'actions. To
resolve distribution system growth problems, Criterion #7
indicates that the system should put in place a biofilm
control plan.
When a variance is granted, the state must set up a com-
pliance schedule for the system. Biofilm control meas-
ures should be undertaken immediately when a variance
is granted, if not before. The best way to ensure that a
variance will not be needed at all is to establish a biofilm
control plan now.
42
-------
Attachment
Tuesday
January 15, 1991
Part II
Environmental
Protection Agency
40 CFR Parts 141 and 142
Drinking Water; National Primary Drinking
Water Regulations; Total Coliforms;
Partial Stay of Certain Provisions of Final
Rule
43
-------
Attachment
1556 Federal Register / Vol. 56. No. 10 / Tuesday. January 15, 1991 / Rules and Regulations
ENVIRONMENTAL PROTECTION
AGENCY
40 CFR Parts 141 end 142
[WH-FRL-3B58-8]
Drinking Water; National Primary
Drinking Water Regulations; Total
Conforms
AGENCY: Environmental Protection
Agency (EPA).
ACTION: Partial stay of certain
provisions of final rule.
SUMMARY: On June 19.1989. EPA
promulgated revised National Primary
Drinking Water Regulations (NPDWRs)
for total coliforms (54 FR 27544, June 29.
1989) pursuant to section 1412 of the
Safe Drinking Water Act (SDWA).
Sections 141.4 and 142.63 of the rule
prohibit States from granting variances
and exemptions to violators of the total
coliform maximum contaminant level
(MCL) of § 141.63(a). Today's action
stays the no variance provisions of
§5 141.4 and 142.63. thus allowing States
to issue variances to the requirements of
5141.03(a) under limited conditions.
EFFECTIVE DATE: January 15,1991.
FOR FURTHER INFORMATION CONTACT:
Paul S. Berger. Ph.D.. Office of Drinking
Water (WH-550D), Environmental
Protection Agency. 401M Street. SW.,
Washington. DC 20460. telephone (202)
382-3039; or the Safe Drinking Water
Hotline, telephone (800) 426-4791;
callers in the Washington. DC area and
Alaska may reach the Hotline at (202)
382-5533. The Safe Drinking Water
Hotline is open Monday through Friday,
excluding Federal holidays, from 8:30
a.m. to 4 pja. Eastern Time.
SUPP1EMENTARY1NFORMATION: On June
19,1989, EPA promulgated revised
regulations for total coliforms (54 FR
27544, June 29.198S). with an effective
date of December 31,1990. Sections
141.4 and 142.63 of the rule prohibit
States from granting variances and
exemptions to the total coliform rule.
Pursuant to section 1412 of the Safe
Drinking Water Act and section 705 of
the Administrative Procedures Act, 5
U.S.C. 705. as required by justice.
today's action stays parts of §§ 141.4
and 142.63 and allows States to grant
variances to the total coliform MCL of
§ 141.63(a) of the rule under certain
conditions.
Section 142.03 of the revised total
coliform rule does not permit variances
to the rule. In the preamble to the final
rule, the Agency explained that total
coliforms are the primary indicator of
the microbiological quality of water. To
the extent a variance would permit the
continued presence of coliforms, the
potential for pathogens to be present
also would remain. As stated in the
preamble, EPA believed that States
would be unable to make the statutorily
required determination that no
unreasonable risk to health would result
from a variance or exemption, since A
variance or exemption would permit the
continued presence of total coliforms in
drinking water above the MCL (see 54
FR 27557, June 29,1989).
The Agency also stated that we were
aware of systems with persistent
coliform problems in distribution
systems not associated with fecal or
pathogenic contamination or with
waterborne disease (54 FR 27557-8). The
source of these coliforms are often
biofilms, which are accumulations of
bacteria which line the walls of some
water distribution pipes. Coliform
bacteria which are released from
biofilms can indicate a violation of the
total coliform MCL when an
unreasonable risk to health does not .
exist. The Agency did not allow
variances to the total coliform MCL in
the rule promulgated on June 29,1989
because of difficulty in distinguishing
these types of total coliform
exceedances from those resulting from
sources of contamination which are an
actual threat to health.
The American Water Works
Association (AWWA) has petitioned the
U.S. Court of Appeals for the District of
Columbia to review EPA's decision to
prohibit variances and exemptions
under the total coliform rule. AWWA
believes that a number of systems have
-a persistent biofilm problem that does
not pose a risk to public health but will
nonetheless cause the system to violate
the rule. They request that States be
permitted to review the particular
circumstances of each such system's
violation and therefore request that EPA
suspend the prohibition against
variances for these systems.
More specific data are now in the
docket that were made available to the
Agency by AWWA and as the result of
a recently held workshop. These data
indicate that some water systems will
experience repealed total coliform
violations due to biofilmn that do not
appear to be associated with fecal or
pathogenic contamination or with
waterborne disease.
The Agency does not believe it is in
the public interest to have continuous
monthly public notification where the
exceedance of the total coliform MCL is
not associated with fecal or pathogenic
contamination of the drinking water.
Besides the adverse impact on the
public confidence in the quality of water
being delivered, repeated public
notification would diminish the efficacy
of the public notice where there in fact .
is a threat to public health.
The Agency has stated that variances
to the total coliform presence/absence
MCL of § 141.63(a) might be appropriate
if a finding of no unreasonable risk to
health could be established (see 54 FR
27557). The difficulty has been in
developing nationally applicable criteria
for variances which would assure
continued protection of public health
while there is positive coliform
occurrence and violation of the MCL.
Until the Agency finalizes criteria for
•distinguishing whether a system
categorized as above is not at risk,
today's action will provide assurance of
no unreasonable risk. This will be done
.by limiting variances to a small number
of systems that can demonstrate to the
Stale protection of public health is at
least equivalent to that provided by the
total coliform MCL.
Specifically, variances shall apply
only to systems not at risk of fecal or
pathogenic contamination because there
is no evidence of treatment lapses or
deficiencies, measured fecal or
pathogenic contamination, or improper
operation or maintenance of the
distribution system. Such systems can
.demonstrate compliance with section
1415 requirements by operating in
conformance with the BAT requirements
identified under 141.63(d).
The following criteria are guidance to
States seeking to identify systems that
could operate under a variance without
posing an unreasonable risk to health:
•(!•) Over the past thirty days, water
entering the distribution system is
shown to:
(a) Be free from fecal coliform or E.
Coli occurrence based on at least daily
sampling,
(b) Contain less than one total
coliform per hundred milliliters of
.influent water in at least ninety-five per
cent of ail samples based on at least
daily sampling,
(c) Comply with the total turbidity
requirements of § 141.13, except that
surface water sources presently filtering
should comply with § 141.73, and
(d) Contain a continuous disinfection
residual of at least 0.2 mg/1;
.(2) The system has had no walerbome
disease outbreak while operated in its
present configuration;
(3) The system maintains biweekly
contact with the State and local health
departments to assess illness possibly
attributable to microbial occurrence in
fee public drinking water system;
{4) The system has evaluated, on a
monthly basis, at least the number of
samples specified in § I41.2l(a)(2) and
44
-------
Attachment
Federal Register / Vol. 56. No. 10 / Tuesday, January 15. 1991 / Rules and Regulations 1557
has not had an E. co//-positiye
compliance sample within the last six
months, unless the system demonstrates
to the State that the occurrence is not
due to contamination entering the
distribution system;
(5) The system has undergone a
sanitary survey conducted by a party
approved by the State within the past
twelve months;
(6) The system has a cross connection
control program acceptable to the State
and performs an audit of the
effectiveness program:
(7) The system agrees to submit a
biofilm control plan to the State within
twelve months of the granting of the first
request for a variance:
(8) The system monitors general
distribution system bacterial quality by
conducting heterotrophic bacteria plate
counts on at least a weekly basis at a
minimum of ten percent of the number of
total coliform sites specified for that
system size in § 141.21(a)(2) [preferably
using the R2A medium in method 907A.
907B, or 907C, as set forth in the 16th
edition of Standard Methods for the
Examination of Water and Wastewater,
1985, American Public Health
Association, et. al.]; and
(9) The system conducts daily
monitoring at distribution system sites
approved by the State and maintains a
detectable disinfectant residual
(measured as specified in § 141.74(a)(5))
at a minimum of ninety-five percent of
those points and a heterotrophic plate
count of less than 500 colonies per ml
(measured as specified in § 141.74(a)(3))
at sites without a disinfectant residual.
The Agency believes the above
priteria identify a set of conditions that
insure equivalent protection to the
current total coliform MCL. When the
Agency ultimately proposes nationally
applicable variance criteria, it is likely
that these requirements or a subset
thereof will be included.
A workshop was held in November
1990 to assist the Agency in refining
nationally applicable criteria for
issuance of variances to the total
coliform MCL. The workshop was
attended by a wide range of experts
familiar with biofilm problems. A copy
of the workshop proceedings is included
in the docket for the total coliform rule.
This stay is issued in order to allow
the Agency time to consider the
recommendations of the workshop and
determine what additional factors may
need to be considered in order to issue
nationally applicable variance criteria.
Pursuant to section 705 of the
Administrative Procedures Act (APA), 5
U.S.C. 705, "when an Agency finds that
justice so requires, it may postpone the
effective date of action taken by it,
pending judicial review." In addition
pursuant to section 553 of the APA, 5
U.S.C. 553, "when the Agency finds good
cause exists, it may issue a rule without
first providing notice and comment and
make the rule immediately effective."
This Notice defers a pending legal
challenge to the total coliform rule while
the Agency reviews the issue of
variance criteria. Since it is in the public
interest to avoid unnecessary litigation
and since this action provides relief for
certain systems, the Agency finds there
is good cause not to solicit comment and
to have the stay immediately effective.
List of Subjects
40 CFR Part 141
Chemicals, Microorganisms. Indians—
land. Intergovernmental relations,
Radiation protection. Reporting and
recordkeeping requirements, Water
supply.
40 CFR Part 142
Chemicals, Microorganisms, Indians—
land, Intergovernmental relations,
Radiation protection, Reporting and
recordkeeping requirements, Water
supply, Administrative practice and
procedure.
Dated: December 31,1990.
F. Henry Habicht,
Administrator.
Parts 141 and 142 of title 40 of the
Code of Federal Regulations are
amended as follows.
1. The authority citation for part 141
continues to read as follows:
Authority: 42 U.S.C. 300f. 3(Mg-l, 300g-2,
300g-3, 300g-4, 300g-5, 300g-a, -IOOj-4 and
300J-9.
2. Section 141.4 is amended by
designating the existing text as
paragraph (a) and by adding paragraph
(b) to read as follows:
§ 141.4 Variances and exemptions.
*****
(b) EPA has stayed the effective date
of this section relating to the total
coliform MCL of § 141.63(a) 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.
3. The authority citation for part 142
continues to read as follows:
Authority: 42 U.S.C. 300f, 300g-l, 300g-2,
300g-3, 300g-J, 300g-5, 300g-6, 300J-4 and
300J-9.
4. Section 142.63 is amended fay
designating the existing text as
paragraph (a) and by adding paragraph
(b) to read as follows:
§ 142.63 Variances and exemptions from
the maximum contaminant level for tota.1.
coliforms.
*****
{b) EPA has stayed the effective date
of this section relating to the total
coliform MCL of § 141.63(a) of this
chapter 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.
[FR Doc. 91-925 Filed 1-14-91; 8:45 amj
BILLING CODE 656O-50-M
45
-------
-------
APPENDIX B
Glossary
activated carbon. Carbon that has been exposed to
very high temperatures and steam to form pores that will
adsorb and hold organic substances.
Aggressive Index. A criterion for establishing the corro-
sive tendency of the water relative to asbestos-cement
pipe.
ANOVA (analysis of variance). A statistical method for
determining the precision of numerous individual meas-
urements.
AOC (assimilable organic carbon). The portion of
DOC that is easily used by microorganisms as a carbon
source.
autotrophic bacteria. Bacteria that can use carbon di-
oxide as their sole source of carbon.
backflow. Reversal of flow in the water distribution sys-
tem resulting in contamination due to a cross connection.
backpressure. Increased pressure, for example, due to
a pump or elevated tank, that causes backflow in the dis-
tribution system.
backsiphonage. Backflow caused by reduced pressure
in the water distribution system.
BDOC (biodegradable dissolved organic carbon).
The portion of dissolved TOC that is easily degraded by
microorganisms. (See AOC; BDOC is the same dis-
solved carbon, measured with a different method.)
biofilm. Organic or inorganic surface deposit consisting
of microorganisms, microbial products, and detritus.
biologically stable. Water from which almost all nutri-
ents have been removed and that therefore will not sup-
port bacterial (or other) growth.
breakthrough. An increase in the numbers of bacteria in
the distribution system that passed through or avoided
disinfection.
cfu (colony-forming units). When spread on a plate of
medium, each bacterium will divide to form a visible col-
ony. These colonies are counted and the count used to
determine the number of bacteria originally present in the
original water sample.
chlorine demand. The amount of chlorine that will com-
bine with impurities and therefore not be available to act
as a disinfectant.
coliform bacteria. A group of microorganisms, charac-
terized by their rod-like shape and fermentation of lac-
tose. These bacteria are usually found in the intestinal
tract of mammals.
community water system. A PWS serving at least 25
people or at least 15 service connections year round.
coupon test. A method for determining the rate of corro-
sion. It involves inserting sample strips (coupons) of the
pipe material into the distribution system.
cross connection. A junction between a potable water
system and contaminated air, water, or solids.
CT value. The disinfection effectiveness, as determined
by multiplying the concentration of residual disinfectant
(C, mg/L) by the disinfectant contact time (T, in minutes).
disinfection. A process that inactivates or kills patho-
gens using chemicals or ultraviolet light.
DOC (dissolved organic carbon). The fraction of TOC
that is dissolved in the water sample.
empty bed contact time (EEJCT). The volume of the
tank holding a bed of activated carbon divided by the flow
rate of the water. The result, in minutes, represents the
length of time each volume of water spends in contact
with the bed.
enteric bacteria. Bacteria that normally reside in the in-
testines of animals.
heterotrophic bacteria. Bacteria that require preformed
organic compounds as carbon and energy sources. Al-
most all pathogenic bacteria are heterotrophic bacteria.
HPC (heterotrophic plate count). A method for enu-
merating heterotrophic bacteria.
lag phase. The length of time from a microorganism's
entry into the system until cell division begins.
Langelier Index. A calculated saturation index for cal-
cium carbonate, useful in predicting the scaling behavior
of water.
MCL (Maximum Contaminant Level). The highest con-
centration of a contaminant permitted in drinking water
under the Safe Drinking Water Act. The MCL takes into
account public health and economic and technical feasi-
bility.
47
-------
nitrification. Oxidation of reduced nitrogen (e.g., ammo-
nia) by microorganisms to form nitrite and nitrate.
noncommunity water system. A PWS that is not a
community water system. These systems serve transient
or nonresident populations, such as a campground,
school, factory, or restaurant.
opportunistic pathogen. A microbe that can cause dis-
ease in immunocompromised individuals (e.g., the eld-
erly, the very young, or ill persons), but usually not in
healthy individuals.
pathogen. A microbe that causes disease.
public water system (PWS). A system that distributes
potable water to at least 25 people or has at least 15
service connections.
regrowth. Growth of microorganisms in the distribution
system.
residual. Disinfectant remaining in the finished water af-
ter primary treatment has been carried out.
surface water. According to the Surface Water Treat-
ment Rule, water that is 1) open to the atmosphere and
subject to surface runoff; or 2) directly influenced by sur-
face water, such as springs or wells.
Surface Water Treatment Rule (SWTR). Regulations
published by EPA in 1989 requiring disinfection for all
surface waters, and filtration if necessary (54 FR 27486,
June 29, 1989).
TOC (total organic carbon). The total amount of or-
ganic compounds, both soluble and insoluble, present in
the water.
Total Coliform Rule. Regulations published by EPA in
1989 specifying limits on coliforms in drinking water (54
FR 27544 June 29,1989).
trihalomethanes (THMs). A type of disinfectant by-prod-
uct (e.g., chloroform) formed when chlorine reacts with
organic compounds in water.
tubercles. Knob-like mounds of corrosion on pipe sur-
faces.
waterborne disease outbreak. When two or more peo-
ple become ill within a short time due to contaminated
drinking water.
48
-------
APPENDIX C
Resources
Safe Drinking Water Hotline
1-800-426-4791
This hotline, directed by the U.S. Environmental Protec-
tion Agency, provides information on drinking water regu-
lations, policies, and documents to the public, state and
local government, public water systems, and consultants.
Up-to-date EPA publication lists are also available. The
Safe Drinking Water Hotline's hours are 8:30 a.m. to 4:30
p.m. Eastern Standard Time, Monday through Friday, ex-
cluding holidays.
U.S. Environmental Protection Agency
Regional Offices
Regional offices of the U.S. Environmental Protection
Agency are listed in Table C-1.
State Drinking Water Agencies
State agencies responsible for public water supervision
are listed in Table C-2.
Organizations Assisting Small Systems
American Water Works Association (AWWA) Small
Systems Program
This program provides information, training, and techni-
cal assistance to small systems, in coordination with
state regulatory agencies and other organizations assist-
ing small systems. Contact the AWWA at 6666 W.
Quincy Avenue, Denver, CO 80235 (303-794-7711) for
the name of a contact for the small systems program in
your area.
National Rural Water Association (NRWA)
This organization provides training and technical assis-
tance to small systems. Contact the NRWA office at P.O.
Box 1428, Duncan, OK 73534 (405-252-0629) for na-
tional information and the name of your local contact.
Rural Community Assistance Program (RCAP)
This program consists of six regional agencies formed to
develop the capacity of rural community officials to solve
local water problems. It provides onsite technical assis-
tance, training, and publications, and works to improve
federal and state government responsiveness to the
needs of rural communities. Table C-3 lists the six RCAP
regional agencies.
Farmers Home Administration (FmHA)
The Farmers Home Administration provides grants and
loans for rural water systems and communities with
populations less than 25,000. Contact FmHA at:
USDA/FmHA, 14th and Independence Avenue, SW,
Washington, DC 20250; 202-447-4323.
Publications
A wide variety of publications on specific topics of con-
cern to water systems is available from the American
Water Works Association and the National Rural Water
Association. EPA's Center for Environmental Research
Information (CERI) Forms and Publications Distribution
Center (513-569-7562) distributes research reports from
the Office of Research and Development. Call for a cata-
logue and to be added to their mailing list. The Safe
Drinking Water Hotline will also send a list of EPA Office
of Drinking Water publications on request.
General
American Water Works Association. Basic Management
Principles for Small Water Systems. 1982.
American Water Works Association. Design and Con-
struction of Small Water Systems - A Guide for Manag-
ers. 1984.
National Rural Water Association. Water System Deci-
sion Makers: An Introduction to Water System Operation
and Maintenance. Duncan, OK, 1988.
Opflow. (A monthly publication of the American Water
Works Association focusing on the "nuts and bolts" con-
cerns of treatment plant operators.)
Schautz, Jane W. The Self-Help Handbook. This manual
gives specific guidelines and techniques for establishing
self-help projects (projects where the community does
some of the work itself to save money). Focus is on im-
proving or creating water and wastewater systems in
small rural communities. For ordering information, con-
tact: Rensselaerville Institute, Rensselaerville, NY
12147, (518)797-3783,
U.S. Environmental Protection Agency, Office of Ground
Water and Drinking Water. Guidance Manual for Compli-
49
-------
Table C-1. U.S. Environmental Protection Agency Regional Offices
EPA Region 1
JFK Federal Building
Boston, MA 02203
617-565-3424
Connecticut, Massachusetts, Maine, New Hampshire,
Rhode Island, Vermont
EPA Region 2
26 Federal Plaza
New York, NY 10278
212-264-2515
New Jersey, New York, Puerto Rico, Virgin Islands
EPA Region 3
841 Chestnut Street
Philadelphia, PA 19107
215-597-9370
Delaware, Maryland, Pennsylvania,
Virginia, West Virginia, District of Columbia
EPA Region 4
345 Courtland Street, NE
Atlanta, GA 30365
404-347-3004
Alabama, Florida, Georgia, Kentucky, Mississippi,
North Carolina, South Carolina, Tennessee
EPA Region 5
230 South Dearborn Street
Chicago, IL 60604
312-353-2000
Illinois, Indiana, Ohio, Michigan, Minnesota, Wisconsin
EPA Region 6
1445 Ross Avenue
Dallas, TX 75202
214-655-2200
Arkansas, Louisiana, New Mexico, Oklahoma, Texas
EPA Region 7
726 Minnesota Avenue
Kansas City, KS 66101
913-236-2803
Iowa, Kansas, Missouri, Nebraska
EPA Region 8
One Denver Place
999 18th Street, Suite 1300
Denver, CO 80202
303-293-1692
Colorado, Montana, North Dakota, South Dakota,
Utah, Wyoming
EPA Region 9
215 Fremont Street
San Francisco, CA 94105
415-974-8083
Arizona, California, Hawaii, Nevada, American Samoa, Guam,
Trust Territories of the Pacific
EPA Region 10
1200 Sixth Avenue
Seattle, WA 98101
206-442-1465
Alaska, Idaho, Oregon, Washington
EPA Headquarters
401 M Street, SW
Washington, DC 20460
202-382-5043
ance with the Filtration and Disinfection Requirements for
Public Water Systems Using Surface Water Sources.
October 1989.
U.S. Environmental Protection Agency, Office of Ground
Water and Drinking Water. Manual of Individual Water
Supply Systems. EPA-570/9-82-004. October 1982.
U.S. Environmental Protection Agency, Office of Water.
Self Assessment for Small Investor-Owned Water Sys-
tems. EPA-570/9-89-011. September 1989.
U.S. Environmental Protection Agency, Office of Water.
Self Assessment for Small Publicly-Owned Water Sys-
tems. EPA-570/9-89-014. September 1989.
Sampling
U.S. Environmental Protection Agency, Office of Ground
Water and Drinking Water. Pocket Sampling Guide for
Operators of Small Water Systems. EPA-814/B-92-001.
April 1992.
U.S. Environmental Protection Agency, Office of Re-
search and Development. Handbook for Sampling and
Sample Preservation of Water and Wastewater. EPA-
600/4-82-029. September 1982.
Filtration
Huisman, L. and Wood, W.E. Slow Sand Filtration. World
Health Organization, Geneva. 1974.
Slezak, L.A. and Sims, R.C. The Application and Effec-
tiveness of Slow Sand Filtration in the United States.
Journal AWWA 76:1238-43.1984.
Visscher, J.T., Paramasivam, R., Raman, A., and Hei-
jnen, H.A. Slow Sand Filtration for Community Water
Supply. Technical Paper 24. International Reference
Centre for Community Water Supply and Sanitation, The
Hague, The Netherlands. 1987.
50
-------
Disinfection
American Water Works Association. AWWA Standard for
Disinfecting Water Mains. AWWA C651-86. Denver CO
1986.
American Water Works Association. Water Chlorination
Principles and Practices (M20). 1973.
American Water Works Association. Water Quality and
Treatment, 4th ed. McGraw-Hill, Inc., New York. 1990.
SMC Martin, Inc. Microorganism Removal for Small
Water Systems. EPA-570/9-83-012. Valley Forge, PA.
June 1983.
U.S. Environmental Protection Agency, Office of Water.
Protecting Our Drinking Water from Microbes. EPA-
570/9-89-008. September 1989.
Corrosion Control
American Water Works Association. Cleaning and Lining
Water Mains. Denver, CO. 1987.
American Water Works Association Research Founda-
tion. Economic and Engineering Services. Lead Control
Strategies. Denver, CO. 1989.
U.S. Environmental Protection Agency, Office of Ground
Water and Drinking Water. Corrosion Manual for Internal
Corrosion of Water Distribution Systems. EPA 570/9-84-
001. September 1984.
U.S. Environmental Protection Agency, Office of Water.
Cross-Connection Control Manual. EPA-570/9-89-007.
September 1989.
Costs/Financial Management
American Water Works Association. Water Utility Capital
Financing (M 29). 1988.
Gumerman, R.C., Burris, B.E., Hansen, S.P., Culp/Wes-
ner/Culp. Estimation of Small System Water Treatment
Costs. Final Report. Municipal Environmental Research
Lab, Cincinnati, OH. 1984.
U.S. Environmental Protection Agency, Office of Water.
A Water and Wastewater Manager's Guide for Staying
Financially Healthy. EPA 430-09-89-004. July 1989.
Consultants
Directory—Professional Engineers in Private Practice.
Published by the National Society of Professional Engi-
neers. Contact SPE Order Department, 1420 King
Street, Alexandria, VA 22314. Who's Who in Environ-
mental Engineering. Published by the American Acad-
emy of Environmental Engineers. Contact the American
Academy of Environmental Engineers, 132 Holiday
Court, Suite 206, Annapolis, MD 21401.
The Federal Register
The Federal Register is published daily to make avail-
able to the public regulations and legal notices issued by
federal agencies. It is distributed by the U.S. Government
Printing Office, Washington, DC 20402. To order copies
call 1-202-783-3238.
51
-------
Table C-2. State Drinking Water Agencies
Region 1
Connecticut Department of Health Services
Water Supplies Section
150 Washington Street
Hartford, CT
203-566-1251
Division of Water Supply
Department of Environmental Protection
One Winter Street, 9th Floor
Boston, MA 02108
617-292-5529
Drinking Water Program
Division of Health Engineering
Maine Department of Human Services
State House (STA10)
Augusta, ME 04333
207-289-3826
Water Supply Engineering Bureau
Department of Environmental Services
P.O. Box 95, Hazen Drive
Concord, NH 03302-0095
603-271-3503
Division of Drinking Water Quality
Rhode Island Department of Health
75 Davfs Street, Cannon Building
Providence, Rl 02908
401-277-6867
Water Supply Program
Vermont Department of Health
60 Main Street
P.O. Box 70
Burlington, VT 05402
802-863-7220
Region 2
Bureau of Safe Drinking Water
Division of Water Resources
New Jersey Department of
Environmental Protection
P.O. Box CN-029
Trenton, NJ 06825
609-984-7945
Bureau of Public Water Supply Protection
New York State Department of Health
2 University Place
Western Avenue, Room 406
Albany, NY 12203-3313
518-458-6731
Water Supply Supervision Program
Puerto Rico Department of Health
P.O. BOX70184
San Juan, PR 00936
809-766-1616
Planning and Natural Resources
Government of Virgin Islands
Nilky Center, Suite 231
St. Thomas, Virgin Islands 00802
Region 3
Office of Sanitary Engineering
Delaware Division of Public Health
Cooper Building
P.O. Box 637
Dover, DE 19903
302-736-4731
Water Supply Program
Maryland Department of the Environment
Point Breeze Building 40, Room 8L
2500 Broening Highway
Dundalk, MD 27224
301-631-3702
Water Hygiene Branch
Department of Consumer and Regulatory Affairs
5010 Overlook Avenue, SW
Washington, DC 20032
202-767-7370
Division of Water Supplies
Pennsylvania Department of Environmental Resources
P.O. Box 2357
Harrisburg, PA 17105-2357
717-787-9035
Environmental Engineering Division
Office of Environmental Health Services
State Department of Health
Capital Complex Building 3, Room 550
1900 Kanawha Blvd., East
Charleston, WV 25305
304-348-2981
Division of Water Supply Engineering
Virginia Department of Health
James Madison Building
109 Governor Street
Richmond, VA 23219
804-786-1766
Region 4
Water Supply Branch
Department of Environmental Management
1751 Congressman W.L. Dickinson Drive
Montgomery, AL 36130
205-271-7773
Drinking Water
Department of Environmental Regulation
Twin Towers Office Building
2600 Blair Stone Road
Tallahassee, FL 32399-2400
904-487-1779
Drinking Water Program
Georgia Environmental Protection Division
Floyd Towers East, Room 1066
205 Butler Street, S.E.
Atlanta, GA 30334
Drinking Water Branch
Division of Water
Department of Environmental Protection
18 Reilly Road, Frankfort Office Park
Frankfort, KY 40601
502-564-3410
52
-------
Table C-2. State Drinking Water Agencies (continued)
Division of Water Supply
State Board of Health
P.O. Box 1700
Jackson, MS 39215-1700
601-354-6616/490-4211
Public Water Supply Section
Division of Environmental Health
Department of Environment, Health, and
Natural Resources
P.O. Box 27687
Raleigh, NC 27611-7687
919-733-2321
Bureau of Drinking Water Protection
Department of Health and Environmental Control
2600 Bull Street
Columbia, SC 29201
803-734-5310
Division of Water Supply
Tennessee Department of Health and Environment
150 Ninth Avenue, North
Terra Building, 1st Floor
Nashville, TN 37219-5404
615-741-6636
Region 5
Division of Public Water Supplies
Illinois Environmental Protection Agency
2200 Churchill Road
P.O. Box 19276
Springfield, IL 62794-9276
217-785-8653
Public Water Supply Section
Office of Water Management
Indiana Department of Environmental Management
105 South Meridian
P.O. Box 6015
Indianapolis, IN 46206
317-633-0174
Division of Water Supply
Michigan Department of Public Health
P.O. Box30195
Lansing, Ml 48909
517-335-8318
Section of Water Supply and Well Management
Division of Environmental Health
Minnesota Department of Health
925 S.E. Delaware Street
P.O. Box 59040
Minneapolis, MN 55459-0040
612-627-5170
Division of Public Drinking Water
Ohio Environmental Protection Agency
1800 WaterMark Drive
P.O. Box 1049
Columbus, OH 43266-0149
614-644-2752
Bureau of Water Supply
Department of Natural Resources
P.O. Box 7921
Madison, Wl 53707
608-267-7651
Region 6
Division of Engineering
Arkansas Department of Health
4815 West Markham Street - Mail Slot 37
Little Rock, AR 72205-3867
501-661-2000
Office of Public Health
Louisiana Department of Health and Hospitals
P.O. Box 60630
New Orleans, LA 70160
504-568-5105
Drinking Water Section
New Mexico Department of Health and
Environment Department
1190 St. Francis Drive
Room South 2058
Santa Fe, NM 87503
505-827-2778
Water Quality Service
Oklahoma State Department of Health
P.O. Box 53551
Oklahoma City, OK 73152
405-271 -5204
Bureau of Environmental Health
Texas Department of Health
• 1100 W. 49th Street
Austin, TX 78756-3199
512-458-7533
Region 7
Surface and Ground-Water Protection Bureau
Environmental Protection Division
Iowa Department of Natural Resources
Wallace State Office Building
900 East Grand Street
Des Moines, IA 50319
515-281-8998
Public Water Supply Section
Bureau of Water
Kansas Department of Health and Environment
Forbes Field, Building 740
Topeka, KS 66620
913-296-1500
Public Drinking Water Program
Division of Environmental Quality
Missouri Department of Natural Resources
P.O. Box 176
Jefferson City, MO 65102
314-751-5331
Division of Drinking Water and
Environmental Sanitation
Nebraska Department of Health
301 Sentenial Mall South
P.O. Box 95007, 3rd Floor
Lincoln, NE 68509
402-471-2541
53
-------
Table C-2. State Drinking Water Agencies (continued)
Region 8
Drinking Water Program
Colorado Department of Health
4210 East 11th Avenue
Denver, CO 80220
303-320-8333
Water Quality Bureau
Department of Health and Environmental Sciences
Cogswell Building, Room A206
Helena, MT 59620
406-444-2406
Division of Water Supply and Pollution Control
NO State Department of Health and
Consolidated Laboratories
1200 Missouri Avenue
P.O. Box 5520
Blsmark, ND 58502-5520
702-224-2370
Office of Drinking Water
Department of Water and Natural Resources
Joe Foss Building
523 East Capital Avenue
Pierre, SD 57501
605-773-3151
Bureau of Drinking Water/Sanitation
Utah Department of Health
P.O. Box 16690
Salt Lake City. UT 84116-0690
801-538-6159
Water Quality Division
Wyoming Department of Environmental Quality
Herschter Building, 4 West
122 West 25th Street
Cheyenne, WY 82002
307-777-7781
Region 9
Field Services Section
Office of Water Quality
2655 East Magnolia Street
Phoenix, AZ 85034
602-257-2305
Office of Drinking Water
California Department of Health Services
714 P Street, Room 692
Sacramento, CA 95814
916-323-6111
Safe Drinking Water Branch
Environmental Management Division
P.O. Box 3378
Honolulu. HI 96801-9984
808-548-4682
Public Health Engineering
Nevada Department of Human Resources
Consumer Health Protection Services
505 East King Street, Room 103
Carson City, NV 89710
702-885-4750
Guam Environmental Protection Agency
Government of Guam
Harmon Plaza Complex Unit D-107
130 Rojas Street
Harmon, Guam 96911
Division of Environmental Quality
Commonwealth of the Northern Mariana Islands
P.O. Box1304
Saipan, CM 96950
670-322-9355
Marshall Islands Environmental Protection Authority
P.O. Box 1322
Majuro, Marshall Islands 96960
VIA HONOLULU
Government of the Federated States of Micronesia
Department of Human Resources
Kolonia, Pohnpei 96941
Palau Environmental Quality Protection
Board
Hospital
Koror, Palau 96940
Region 10
Alaska Drinking Water Program
Wastewater and Water Treatment Section
Department of Environmental Conservation
P.O. Box O
Juneau.AK 99811-1800
907-465-2653
Bureau of Water Quality
Division of Environmental Quality
Idaho Department of Health and Welfare
Statehouse Mail
Boise, ID 83720
208-334-5867
Drinking Water Program
Department of Human Resources
Health Division
1400 S.W. 5th Avenue, Room 608
Portland, OR 97201
503-229-6310
Drinking Water Section
Department of Health
Mail Stop LD-11, Buildings
Airdustrial Park
Olympia, WA 98504
206-753-5954
54
-------
Table C-3. Rural Community Assistance Program (RCAP) Agencies
Community Resources Group, Inc.
2705 Chapman
Springdale, AR 72764
501-756-2900
Great Lakes Rural Network
109 South Front Street
Freemont, OH 43420
419-334-8911
Midwest Assistance Program, Inc.
P.O. Box 81
New Prague, MN 56071
612-758-4334
Rural Community Assistance Corporation
2125 19th Street, Suite 203
Sacramento, CA 95818
916-447-2854
Rural Housing Improvement, Inc.
218 Central Street, Box 429
Winchendon, MA 01475-0429
617-297-1376
Virginia Water Project, Inc.
Southeastern Rural Community
Assistance Program
702 Shenandoah Avenue, NW
P.O. Box 2868
Roanoke, VA 24001
703-345-6781
55
-------
-------
-------
-------
-------
m
-a
$
to
en
?
(O
s
o
o§ =5
o oi 2.
•? —
:*• n>
o c
"• !2.
—• 0)
ss
CO
CD
o o
5'
m c
O S1
ft I
8i
00 CD
I
I
CD
D)
I |