United States
Environmental Protection
Agency
Office of Water (4601M)
Office of Ground Water and Drinking Water
Distribution System Issue Paper
The Potential for Health Risks from Intrusion of
Contaminants into the Distribution System from
Pressure Transients
-------
PREPARED FOR:
U.S. Environmental Protection Agency
Office of Ground Water and Drinking Water
Standards and Risk Management Division
1200 Pennsylvania Ave., NW
Washington DC 20004
Prepared by:
Mark W. LeChevallier, Richard W. Gullick, Mohammad Karim
American Waterworks Service Company, Inc., Voorhees, NJ
Background and Disclaimer
The USEPA is revising the Total Coliform Rule (TCR) and is considering new possible
distribution system requirements as part of these revisions. As part of this process, the
USEPA is publishing a series of issue papers to present available information on topics
relevant to possible TCR revisions. This paper was developed as part of that effort.
The objectives of the issue papers are to review the available data, information and
research regarding the potential public health risks associated with the distribution
system issues, and where relevant identify areas in which additional research may be
warranted. The issue papers will serve as background material for EPA, expert and
stakeholder discussions. The papers only present available information and do not
represent Agency policy. Some of the papers were prepared by parties outside of EPA;
EPA does not endorse those papers, but is providing them for information and review.
Additional Information
The paper is available at the TCR web site at:
http://www.epa.gov/safewater/disinfection/tcr/regulation revisions.html
Questions or comments regarding this paper may be directed to TCR@epa.gov.
-------
Distribution System White Paper
The Potential for Health Risks from Intrusion of Contaminants into the
Distribution System from Pressure Transients
Mark W. LeChevallier, Richard W. Gullick, Mohammad Karim
American Water Works Service Company, Inc., Voorhees, NJ
Issue Statement
This paper examines the potential for public health risks associated with intrusion of
contamination into water supply distribution systems resulting from transient low or negative
pressures, as well as methods for preventing intrusion of contaminants that may lead to increased
health risks, and mitigation of existing contaminant intrusion problems. This problem is defined
as a specialized backflow situation that occurs in an otherwise pressurized system, and therefore
the reader is referred to the cross connection white paper for a broader consideration of cross
connection issues, health risks, and mitigation techniques.
Definition of the Problem
A pressure transient in a drinking water pipeline is caused by an abrupt change in the velocity of
water. This event is sometimes termed "surge" or "water hammer." The energy at any point in
the pipeline is composed of kinetic and potential energy. Water will move through a pipe from
points of higher energy to points of lower energy regardless of its position. Any change in flow
in a pipe (due to valve closure, pipe fracture, or pump stoppage) will result in an exchange of
energy between flow and pressure. The change in pressure can be defined by the Joukowsky
equation (Thorley 1991):
H = 4660 * (Vi - Vf) where:
(1 + Mw *ID)0J*g
Mp th
H = pressure increase (ft)
Mw = bulk modulus of water (psi)
Mp = bulk modulus of pipe materials (psi)
ID = inside diameter of the pipe (in)
th = wall thickness of the pipe (in)
g = acceleration due to gravity (ft/sec2)
Vi = initial water velocity (ft/sec)
Vf = final water velocity (ft/sec)
The magnitude of the pressure change is influenced by the materials of construction, pipe
characteristics, and the water velocity. Operational characteristics can further affect the
significance of pressure transients, including: non-networked and dead-end pipelines, a lack of
elevated distribution system storage tanks, undulating topography, entrained air, valve
characteristics, and frequent power failures of pumping stations (AWWSC 2002).
1
-------
For example, consider a pipeline on which an open valve is located at a distance downstream
from a reservoir. If the valve is closed instantaneously, water will decelerate to zero velocity and
the kinetic energy will be converted into pressure. The transient wave will travel upstream and
downstream from the valve and ultimately reach the ends of the pipe. If the pressure wave in
the pipe is not relieved (as in a surge tank), it will travel in the reverse direction back to the
valve. Because the valve is closed and there is no relief for this flow, a negative pressure wave
(suction) will be created at the valve (Simon and Korom, 1997). This wave will travel back and
forth until the kinetic energy is dissipated by friction. The process will occur both upstream and
downstream from the valve. However, the initial pressure will be positive on the upstream side
and negative on the downstream side (Simon and Korom, 1997).
The analysis of transient flow in large ~
distribution systems or other incompressible
fluids requires the solution of the wave A psi
equations coupled to the boundary (high)
conditions of the flow. A widely used ^ pSj
technique is the so-called method of (low)
characteristics (Streeter and Wylie 1967) or
the wave plan method (Wood et al. 1966).
Pressure transients can be described as
waves (Figure 1), having both a positive
and negative amplitude (Simon and Korom -
1997, Funk et al. 1999). Because these Figure 1. A Pressure Transient
waves travel through the distribution system,
the resulting low or negative pressures may occur in many different locations. The
circumstances that produce these pressure waves may commonly occur in every water system.
Pressure transients can be caused by main breaks, sudden changes in demand, uncontrolled pump
starting or stopping, opening and closing of fire hydrants, power failures, air valve slam, flushing
operations, fire flow, feed tank draining and other conditions including venturi effects (Funk et
al. 1992). As a general rule of thumb, for every 1 ft/sec of velocity forced to a sudden stop,
water pressures increase 50 to 60 psi (depending on the pipe materials, topography, etc.). The
opposite is true for a sudden velocity increase, resulting in an instantaneous low or negative
pressure (Kirmeyer et al. 2001).
The production of negative pressure transients creates the opportunity for backsiphonage or
backpressure of non-potable water from domestic, industrial or institutional piping into the
distribution system (USC FCCCHR 1993). These conditions of backflow are more thoroughly
addressed in the cross connection white paper. Intrusion refers to the flow of non-potable water
into mains though leakage points, submerged air valves, faulty seals, or other openings.
Magnitude of the Risk
The public health significance of intrusion from a pressure transient depends on the number and
effective size of orifices (leaks), the type and amount of contaminant external to the distribution
system, the frequency, duration, and magnitude of the pressure transient event, and the
population exposed.
2
-------
Pipe Leakage, Orifices, and Location
In the American Water Works Association Research Foundation (AwwaRF) report Pathogen
Intrusion into the Distribution System (Kirmeyer et al. 2001), 77% of 26 utilities surveyed had a
leak detection program that used a variety of different leak detection techniques (e.g., leakage
correlator, comparison of metered sales, electronic noise detection). The percent of leakage
(unaccounted for water) for these utilities ranged from less than 10 percent to as high as 32
percent. It is not uncommon for water systems to lose more than 10 percent of the total water
production through leaks in the pipelines (AWWA and AwwaRF 1992). In reality it is very
difficult to precisely know how
much of the unaccounted water is
due to leakage unless a significant
effort is exerted to track all losses.
Hydraulic modeling can be used to
estimate the impact of orifice
diameter on the volume of water
that could intrude during a negative
pressure event (Funk et al. 1992,
Funk et al. 1999, Kirmeyer et al.
2001). Depending on the effective
size of the orifice, the external
pressure, and the nature of the
transient event, the volume of
intrusion can range from milliliters
to hundreds of gallons (Table 1).
Pipes located below the water table are subject to pressure from the exterior water (depending on
the height of the water table above the pipe) and thus an opportunity exists where water exterior
to pipe could intrude into the pipe under low or negative pressure conditions within the pipe.
Utilities were surveyed as to the percentage of mains that are submerged, and the results showed
that at least 20% of the systems had pipes below the water table (42% had no information)
(Kirmeyer et al. 2001). It is assumed that all systems have some pipe below the water table for
at least some time of the year.
Water may also intrude into a distribution
system by means other than pipelines. It has
been speculated that faulty joints seals may
leak under certain circumstances when
exposed to negative pressures (Grigory,
2002). A survey of the percentage of flooded
vaults or meter boxes showed that although
the rate changed seasonally, approximately
20% of the systems reported between
twenty-five and seventy five percent of meter
boxes flooded, with about half of the systems
Table 1. Determination of the Intrusion Volume (in gallons)
During a 30 Second Negative Pressure Event
Orifice
Diameter
fin.)
Power Loss
Main Break
Fire Flow
1 ft
10 ft
1 ft
10 ft
1 ft
10 ft
1/32
0.01
0.08
0.04
0.12
0.04
0.12
1/8
0.2
1.2
0.6
1.8
0.6
1.8
1/2
3
18
8
27
8
26
1
8
58
23
96
24
87
2
13
185
55
335
46
244
From Kirmeyer et al. 2001.
lft and 10 ft refers to the height of the external water table above the pipe.
Figure 2. Submerged air release valve
3
-------
not knowing how much flooding had occurred (Kirmeyer et al. 2001). One utility provided
pictures of an air valve vault that was flooded with an oily film and a second picture from a short
while later when the vault was drained (Figure 2). It is presumed that a pressure transient caused
the air valve to open and allowed the water to enter into the distribution system. Engineering
standards (Recommended Standards for Water Works 1997) specify that all air release valves
(and similar appurtenances) be designed with above-grade-venting (this venting should be
tamper-proof to prevent deliberate contamination of the system), or be modified in a way to
prevent the flooding of the vault (e.g., via drainage or a pump).
Presence of Contaminants External to the Distribution System
Any contaminant exterior to the distribution system may enter potable water supplies during a
negative pressure event. Chemical contaminants could include pesticides, petroleum products,
fertilizers, solvents, detergents, pharmaceuticals, and other compounds. Predominant pesticides
in urban areas include atrazine, simazine, prometon, and diazinon (Patterson and Focazio 2001).
Other studies have detected insect repellants, fire retardants, and other industrial chemicals
(Koplin et al. 2002). If chemical compounds intrude in sufficient concentration or volume, they
might result in acute toxicity. Microbial contaminants are a concern because even with dilution,
some microbes (e.g., viruses) could cause an infection with a single organism.
Karim et al. (2001) reported on a study that examined 66 soil and water samples collected from 8
utilities in 6 states. The samples were collected immediately adjacent to the drinking water
pipelines. The purpose of the study was to determine the presence of microbial contaminants in
the soil immediately external to the distribution system. Whenever a main was excavated,
samples were collected of either the water or the undisturbed soil next to the pipe. Total
coliform and fecal coliform bacteria were detected in water and soil in about half of the samples,
indicating the presence of fecal contamination (Figure 3). Bacillus was found in almost all the
samples, which is not a surprise since it is a normal soil organism. Viruses were detected using
culturable methods in 12 percent of the soil and water samples, and by molecular methods in 19
percent of the soil samples and 47 percent of the water samples. When these data are combined,
56 percent of the samples were positive for viruses either in the water or the soil. Sequence
analysis showed that these viruses were predominantly enteroviruses (the vaccine strain of
Poliovirus), but Norwalk and Hepatitis A viruses were also detected, providing clear evidence of
human fecal contamination immediately exterior to the pipe.
In the same study an analysis of the levels of organisms detected showed that they could be quite
high; for example, total fecal coliform levels were as high as 104 bacteria per 100 grams of soil
(Table 2). This may not be surprising considering that sewer lines are often located only a few
feet away (Figure 4). Engineering standards call for a minimum separation of 10 ft between
drinking water and sewer pipelines, although separations can be as little as 18 inches if the
drinking water pipe is located at a higher elevation than the sewer pipe (Recommended
Standards for Water Works 1997). In saturated soil conditions, microbes can move several
meters in short periods of time (Abu-Ashour et al, 1994). This transport could be aided by water
flowing out of the sewer (exfiltration).
4
-------
0)
CJ
c
£
s
CJ
U
o
/ / ^
~ Water
~ Soil
Figure 3. Summary of Microbial Occurrence in Water and Soil Samples
The soil and water samples in the study (Karim et al. 2001) were randomly collected from urban
environments and the location of adjacent sewer lines is not known. More detailed studies could
develop better guidelines for the separation of water and sewer mains. The concentration of
...8
Bacillus spores in soil was as high as 10 colony-forming units (CFU) per 100 grams of soil, with
some of the highest levels associated with samples containing human enteric viruses. It is
possible that seepage of sewage stimulated the growth of the soil flora in these locations.
Table 2. Microbe Concentration in Water and Soil
Organism
Water
CFU orPFU/100 ml
Soil
CFU or PFU/100 gm
Total Colifonns
< 2 -1.6 x 103
< 2 - 1.6 x 10 4
Fecal Colifonns
< 2 -1.6 x 103
< 2 - 1.6 x 10 4
Clostridium
0- 2.5 xlO3
0 - 1 x 105
Bacillus
0- 4.6x 10 6
0- 1.2 x 108
Phage
0 - 1 x 104
0
CFU, colony-forming units; PFU, plaque-forming units
5
-------
Figure 4. Leaky water pipe laid next to a sewer pipe
(Source: Opflow 1999)
Frequency and Magnitude of Pressure Transient Events
Problems with low or negative pressure transients have been reported in the literature (Walski
and Lutes 1994, Qaqish et al. 1995). Recent research efforts have focused on documenting the
frequency and magnitude of pressure transient events to determine whether negative pressure
events occur during normal distribution system operations. A high-speed pressure logger (RDL
1071L/3 Pressure Transient Logger, Radcom Technologies, Inc.; Woburn, MA), with a
monitoring rate of 1-20 measurements per second and a range from 0 to 300a psi, was used to
detect negative pressure events. Other manufacturers offer similar equipment.
A comparison of a high-speed
electronic data logger to a
conventional strip chart recorder
showed a good correspondence
between the measurements of
sudden high pressures, but the
Radcom monitor was much more
sensitive for capturing the low
pressure events, on average showing
values 10 psi lower than those
recorded by the conventional
recorder (Figure 5).
Q. OJ
c P>
-10
Conventional
Radcom
-40
Different transient pressure events
Figure 5. Comparison of the Radcom and Conventional
Pressure Recorders for Measurement of Low Pressure Events
Application of high-speed pressure
loggers to routine operations in
approximately 10 systems has shown substantial variability in pressure values, however,
negative values have only rarely been observed. Various attempts to examine hydrant flushing
with different rates of valve opening demonstrated the production of pressure transients, but none
6
-------
of the events produced negative distribution system pressures (Kirmeyer et al. 2001). Additional
investigation of hydrant operation is warranted because hydraulic modeling has suggested that
negative distribution system pressures could be produced under certain hydrant flushing
circumstances.
Examination of a household tap showed large fluctuations with pressures as low as 4.3 psi (data
not shown). These fluctuations may be due to domestic water use patterns. If there was an
external water table of 10 ft over the pipe (as in a stream crossing or low land area), there could
be enough external water pressure to cause intrusion. The point is that it is not necessary to have
a negative pressure - a low pressure can cause intrusion under certain circumstances.
Pressures were analyzed in one system while conducting a routine draw down test in Spring
2001 (an annual test to verify accuracy of the venturi meters at the water treatment plant).
During this test, the main service pumps were shut down at the treatment plant clearwell and
restarted with all flow going through one venturi meter. Two Radcom monitors were installed at
high elevation points on a 30-inch main (one was -2.5 miles from the plant, and the other was
-4.5 miles from the plant), and a third monitor was located about 80 feet from the treatment
plant's high-service pumps. Pressure readings both near the treatment plant and within the
distribution system showed large pressure fluctuations. While the static pressures near the plant
ranged between 125 and 150 psi, the pressure transients caused by the pump shutdowns resulted
in pressures as low as 18 psi in the plant effluent. However, several miles away in the
distribution system these fluctuations resulted in a pressure of minus 10 psi lasting for 16
seconds (Figure 6). The valve closure speed for the main service pumps was 20 seconds, which
may have been too fast, and thus contributing to the pressure transient. A second test was
conducted with the valve closure speed slowed to 30 seconds, but negative pressures resulted
from this second test as well.
Routine pressure monitoring of another distribution system in December 2000 showed a negative
pressure event during a power outage at a pumping station that lasted for 24 seconds and
produced a negative 4.4-psi (Figure 7). Similarly, a power outage at the treatment plant of
another system in July 2001 produced zero pressure for 51 seconds in a section of the
distribution system (Figure 8).
Based on the above information, it is concluded that transient pressure events occur in
distribution systems; that these events can result in negative pressures; that negative pressures
provide a potential portal for entry of non-potable water into potable water distribution pipelines;
and that fecal indicators and culturable human viruses are present in the soil and groundwater
exterior to the distribution system. However, the characteristics of distribution systems that
contribute to producing negative pressure transients have not been examined. These
characteristics may include the presence of storage tanks, valve closure speed, placement of air
relief and other surge control devices, pump operation, and shut down procedures. To date, all
observed negative pressure events have been related to power outages or other pump shutdowns.
More research is needed to better characterize the types of systems (e.g., those without
distribution storage, without air or vacuum relief valves, etc.) most prone to negative pressure
transient events.
7
-------
r
near WTP
'""Vn
">>
T**
Dist. System
Ll
—Jt
i-" o> m «x« :o jc'SOj
*-ft mtty m «. >«x«**. * »t#o>«»•»¦:
e> e.¦ w« e^ftfiA O v«
..;mPfa"-nttao* -xJ«W» 'WK->n
jjm«> 16 sec;
• as low as -10.1 psi
HSJAaVd LI—X'P* IBM*
r? CftrM* UkSM •• •*.»• .y»^i
di
l l*;- #.!:. h'nA H ¦
iHSH «•»«•
il-kfeXOfl «#•!«<
ij&r^p 6a«i*i
J
IWEP««»
• !W»0
•-i««4f5K li-MUCOe
I"1*1 CfrMAP
Jj
Figure 6. Pressure Recording During a Pump Draw Down Test.
A) Pressures near the water treatment plant (WTP) and in the distribution
system. B) Enlargement of one of the distribution system negative pressure
events shown in A. illustrating duration of the negative pressure event. A
second recorder in the distribution system showed similar results.
8
-------
¦* ra nvt mmm ma _uij* (mi
_rn_^
1 1
I-]
l~t. ir
t
'ovver outage due
r\ lirrh+ninrr ctriVp
I
W
JULB ' Ct/yil A'ltt M T4
Jn lmj ' « ;¦ ;*
ail* ' Hii
MftLi LUJjjijj ' -fc-vjjiE ' itttJiiXi '
ar» Miit
i.iimm.' ijjfrHO
i.n.gdji 11* sta
B
Negative for -24 sec;
as low as -4.4 psi
I
ji
r ^ n ¦-
i-inca
Figure 7. Distribution System Pressure Monitoring Following a Pump Station
Power Outage. A) Daily monitoring data, B) Enlargement of the negative pressure
event.
9
-------
A
I,
hjM»
^ Power outage due
to lightning strike
y> www
ET KV«*1
"flWWW
D7.f4.QI
•*WWW
E-JKfll
"5 WWW
C'.WXJI
leramw
CPCCC*
•i ro WW
(Ft***
iirauw
•f www i&wcini
CPH UH 'ir.11131
I -li
B
-0 psi for
~51 sec
•ii|
id? r-ww
g«r rt-nirfv
dl 1
fl m *."¦
1W
311 vino
PW!
m yi hi
(f II. JIT!
rn Hfrw,w
•pwn
ri ir> an i ¦
rrQMf
ti« n.n
srrans
11 »iin
mwi
*1
Figure 8. Distribution System Pressure Monitoring Following a Pump Station Power Outage
due to a Lightening Strike. A) Daily monitoring data, B) Enlargement of the negative
pressure event.
10
-------
Public Health Impact
Payment et al. conducted two epidemiology studies (Payment et al, 1991; Payment et al, 1997),
each suggesting that the distribution system was at least partially responsible for increased levels
of gastrointestinal illnesses. The studies examined the health of people who drank tap water and
compared the group to people receiving water treated by reverse osmosis to determine which
group had higher levels of gastrointestinal illness. Both studies pointed to the fact that people
who drank tap water had increased cases of gastroenteritis. Analysis of Payment's data shows
that people who lived in zones far away from the treatment plant had the highest risk of
gastroenteritis. Transient pressure modeling (Kirmeyer et al. 2001) found that the distribution
system studied by Payment was extremely prone to negative pressures, with more than 90
percent of the nodes within the system drawing negative pressures under certain modeling
scenarios (e.g., power outages). The system is located in the Montreal area, and reported many
pipe breaks, particularly during the Fall and Winter when temperature changes place added
stresses on the distribution system. Although the system employed state-of-the-art treatment, the
distribution network maintained low disinfectant residuals, particularly at the ends of the system.
Low disinfectant residuals and a vulnerability of the distribution system to pressure transients
could account for the viral-like etiology of the illnesses observed.
A double-blinded, randomized, trial was recently completed in Melbourne, Australia, to
determine the contribution of drinking water to gastroenteritis (Hellard et al. 2001). Melbourne
draws its drinking water from a protected forest watershed and has an unfiltered surface water
supply using only free chlorine treatment. Free chlorine levels in the distribution system ranged
from 0 to 0.94 mg/L, with a median of 0.05 mg/L, and 90% of samples had < 0.20 mg/L. Total
coliform bacteria were detected in 18.9% of 1,167 routine 100-mL water samples, but fecal
coliform bacteria were not detected. Distribution system samples were positive for Aeromonas
spp. (50%) of 68 weekly samples), Campylobacter (1 occasion) and Giardia (2 viable samples by
reverse transcriptase-polymerase chain reaction). Six hundred families were randomly assigned
to receive either a real or placebo water treatment unit installed on the kitchen faucet. Real units
were designed to remove viruses, bacteria, and protozoa using microfiltration and ultraviolet
light treatment. Study participants completed a weekly health diary reporting gastrointestinal
symptoms during the 68-week observation period. The study found that the water was not a
source of measurable gastrointestinal disease (the ratio of illness between the group drinking
treated water compared to the normal tap water was 0.99, with a 95% confidence interval of
0.85-1.15; p = 0.85). Analysis of 795 fecal specimens from participants with gastroenteritis did
not reveal any difference in pathogen detection between the two groups. Pressure transient
modeling of the Melbourne system has not been done and specialized pressure monitoring was
not performed during the study.
The 1996 amendments to the Safe Drinking Water Act required the U.S. Centers for Disease
Control and Prevention (CDC) and the U.S. Environmental Protection Agency (EPA) to conduct
epidemiology studies to determine the occurrence of waterborne disease in the U.S. Dr. Jack
Colford of the University of California at Berkeley School of Public Health is conducting one of
these epidemiology studies in collaboration with the Iowa-American Water Company in
Davenport, Iowa. The study began in November 2000, and will be completed in June 2002. The
study is a randomized, triple-blinded, placebo-controlled, crossover intervention study. The
11
-------
intervention to be tested is household-level treatment of drinking water. The water is treated
using a kitchen countertop device that treats tap water with ultraviolet light and microfiltration.
Participating households have been randomly assigned to two different groups. One group
received the active device and the other received an identical-looking placebo device. Half way
through the study, "cross-over" will take place: active devices will be replaced with inactive
devices, and inactive devices will be replaced with active devices. The participants, the study
staff, and the data analysis team will be blinded to (unaware of) which group each household has
been assigned throughout the study. A total of 456 households residing in Davenport,
Bettendorf, Panorama Park, and Riverdale have been enrolled.
The American Water Works Association Research Foundation has funded the American Water
Works Service Company to conduct a water quality study in the Davenport area in parallel to the
epidemiology study. The study is conducting extensive analysis of the raw water, treatment
plant performance, distribution system and household water quality. Seven pressure data loggers
(one in each pressure zone) are being used to monitor distribution system pressures to determine
if pressure transients are associated with any health impacts that may be observed during the
epidemiology study. To date, although fluctuations in pressures have been noted, no negative
pressure events have been recorded in the distribution system. Modeling of the distribution
system is underway to extrapolate the pressure data to the whole pipe network.
In summary, although there are data to demonstrate that negative pressure events do occur, there
are insufficient data to indicate whether these events result in substantial risk to water quality in
the distribution system. Direct monitoring of drinking water would be impractical due to the
transient nature of the pressure effect, the relatively small volume of intrusion water (compared
to the total volume within the pipe network), and the plug flow nature (e.g., limited dispersion)
of water within distribution systems. In addition, a source of microbial contamination (e.g.,
leaky sewer lines) must be relatively near the pipe system, and the soil must be saturated to allow
for microbial transport. These factors may be important variables explaining the disparate
epidemiology results and should be factored into any future epidemiological studies.
Risk Mitigation
The first step in risk mitigation for the issue of transient negative pressures in the distribution
system is simply the recognition that the phenomenon does exist. Some have dismissed the issue
as being not significant, too brief, or too small of a volume to be an important source of
contamination. On-going studies are beginning to document the occurrence of negative transient
pressure events within distribution systems, but additional research is necessary. The frequency
of negative pressure transients need to be determined, as well as the characteristics of the
distribution system that contribute to these events. Studies need to be conducted for ground
water systems, particularly in non-disinfected systems.
Engineering standards require consideration of pressure transients for pipeline and pump design,
distribution system network analysis, and valve selection and installation (Table 3). Information
on transient analysis and control can be found in standard engineering texts on pump design,
pipeline flow, and fluid dynamics (Karassik et al. 1976, Larock et al. 2000, Thorley 1991, Simon
and Korom, 1997). Surge control, particularly control of high-pressure events, has typically
12
-------
Table 3. Available Standards and Guidelines for Surge and Intrusion Mitigation
Existing Standards and Guidelines
• ANSI/AWWA C510 (Double Check Valve Backflow-Prevention Assembly)
• ANSI/AWWA C511 (Reduced-Pressure Principle Backflow-Prevention Assembly)
• ANSI/AWWA C512 (Standard for Air Release, Air/Vacuum. And Combination Air Valves for
Waterworks Services)
• Recommended Standards for Water Works (10 State Standards)
• AWWA Manual M14 Recommended Practice for Backflow Prevention and Cross-Connection Control
• AWWA Manual M32 Distribution Network Analysis for Water Utilities
• AWWA Manual M36 Water Audits and Leak Detection
• AWWA Manual M44 Distribution Valves: Selection, Installation, Field Testing, and Maintenance
• AWWA Manual M51 Air-Release, AirA^acuum, and Combination Air Valves
been thought of in terms of preventing pipe bursts and efforts have been directed at reducing the
maximum pressures. Concerns regarding negative pressure transients and their public health
implications have not received similar attention. However, mitigation measures are well
described and include slow valve closure times, avoiding check valve slam, minimized
resonance, air vessels, surge tanks, pressure relief valves, surge anticipation valves, air release
valves, combination two-way air valves, vacuum break valves, check valves, surge suppressors,
and by-pass lines with check valves. A surge tank or standpipe provides water when system
pressure decreases and can also absorb pressure increases. Four common types of surge tanks
include: pneumatic or closed tank, open standpipe, a feed tank with a check valve, and a bladder
tank. If water is stored in the tank for long periods of time the water quality may degrade and
proper operation and maintenance is required to avoid poor quality water from entering the
distribution system.
Air relief valves and similar appurtenances should be designed to have above-grade venting (at
least 1-ft [0.3 m] and be designed to be tamper-proof to avoid deliberate contamination of the
system). All below-grade vacuum or air relief valves should be retrofitted to above-grade
venting, or modified in a way to prevent the flooding of the vault (e.g., drainage or pump).
The results of these studies emphasize the need to maintain an effective disinfectant residual in
all parts of the distribution system. Although the effectiveness of a residual disinfectant has been
debated (Trussell 1999), critics typically question the effectiveness of a disinfectant residual to
inactivate volumes of sewage mixed with drinking water (Snead et al. 1980, Payment 1999). For
distribution system negative pressure events, the volume of intruded water is a fraction (much
less than 1%) of the water within the pipe network, so the opportunity for effective disinfection
exists. Unknown is the effect of turbidity, compounds causing a chlorine demand, and limited
mixing (in a relatively plug flow condition) on the disinfection efficacy of the residual
disinfectant. Chloramine residuals will be particularly ineffective for viruses that intrude into the
distribution system, as the CT (disinfectant concentration multiplied by the contact time) for
preformed chloramines would not be effective for enteric viruses. Studies examining the
microbial risk-risk tradeoffs (e.g., disinfection effectiveness for intrusion contaminants compared
to biofilms) are needed as many U.S. water suppliers continue to convert from free chlorine to
chloramines due to disinfectant by-product regulations.
13
-------
Efforts to reduce distribution system pipeline leakage are beneficial not only from a water
conservation standpoint, but also to minimize the potential for microbial intrusion into potable
water supplies. Leaks are not simply a loss of revenue for a water utility, but the leak is a
potential pathway for contamination. The public health benefits of leak control should be
recognized and encouraged. Repair of leaking sewer lines should similarly be a top priority, not
only to minimize the occurrence of pathogens near drinking water pipelines, but to reduce these
sources of contamination being transported to groundwater supplies and receiving streams,
particularly under wet weather conditions.
High-speed pressure data loggers would probably benefit distribution system monitoring, as they
appear to be more sensitive, particularly for low-pressure events. Additional studies are needed
to examine the accuracy of the pressure transducers and determine the appropriate placement of
the recorders within the distribution system. Installation of the monitors at high elevation points
within the distribution system would seem reasonable, but additional work is needed to identify
other useful monitoring locations. The generation of high-quality pressure data would help
determine the effect of routine operational practices on distribution system pressures. This
monitoring data could evaluate the impact of hydrant operations, pump start-up and shut down
procedures, and valve closing speed, among others. This information should be compiled to
develop standard operating procedures to minimize low-pressure surges.
Surge modeling can be used to determine the potential vulnerability of a system to negative
pressures under a number of worst-case scenarios (e.g., power failure, main break, flushing, etc.).
This modeling would be useful especially after addition of new pipelines, interconnections, or
changes in distribution system storage or consumption patterns that may have changed original
design parameters. Modeling may be able to identify zones of the distribution system most
prone to negative pressure events. These areas would then be prioritized for maintenance of a
disinfectant residual, leak detection and control, main replacement, and rehabilitation of nearby
sewer systems. This engineering analysis can apply surge control techniques, like installation of
air relief valves (above grade), surge tanks, and other activities to mitigate negative pressure
events.
Personnel training with respect to hydrant and valve operations, and prevention of unauthorized
or inappropriate use of hydrants or blow-offs, would be useful so that maintenance and repair
crews understand the concerns regarding the potential for intrusion.
Indicators
Many States have requirements to maintain minimum distribution system pressures based on
conventional pressure recorder data. It would be inappropriate, and possibly impractical to apply
the same guidelines to data collected by electronic pressure loggers. Additional research is
needed to evaluate new guidelines based on the frequency and duration of the event, the
concentration and type of residual disinfectant, the proximity of the drinking water main to sewer
lines, soil conditions and the level of the water table, and other data that still need to be collected
to assess the public health significance of such events.
14
-------
Additional research is needed to develop guidelines for proper placement of pressure monitors.
Distribution system modeling of a power outage suggested that negative pressures may have
occurred in locations other than those selected for pressure monitoring. Monitoring locations are
often selected based on the availability of land, access, and electrical power or communications;
not necessarily because the location is most prone to negative pressures.
Increased microbiological monitoring, particularly using existing methodologies, is not
recommended because of the low probabilities of actually detecting an intrusion event. Use of
continuous chlorine residual monitors may have some application, but the effectiveness of such
an approach needs to be evaluated. Development of new on-line microbial monitoring
techniques may have some future application, particularly those related to fiber optic or real-time
analysis.
Current or Planned Research
AwwaRF has completed one project related to distribution system intrusion (Kirmeyer et al.
2001) and has another project in progress (Field-Testing of Surge Modeling Predictions to
Verify Occurrence of Distribution System Intrusion, #2686). The project is anticipated to 1)
verify by field and pilot measurements surge model results and illustrate how operating
conditions affect the production of low or negative pressures, 2) conduct pilot test studies
comparing intrusion volume estimates for various operating conditions, and 3) develop
guidelines for surge modeling, pressure monitoring, and other design and operation and
maintenance practices to prevent intrusion. Drafts of this report should be available in 2003.
The Microbial/Disinfection By-Product Research Council organized a workshop in 2001 to
identify the research gaps that were highlighted during development of the Stage 2 M/DBP Rules
(M/DBP Research Council 2002). One workgroup dealt with distribution system issues, and the
committee developed several projects that addressed intrusion. The project, "Characterizing the
Importance of Distribution System Intrusion Events," would define the importance of
distribution system intrusion events with respect to the frequency and level of contamination.
Another project, "Distribution System Operations Assessment and Guidance Manual," would
assess distribution system operational practices and goals (including intrusion) to develop a
guidance manual outlining best operational practices. These recommendations will be forwarded
to the USEPA and AwwaRF for consideration, but funds for these have not yet been allocated.
A report developed for the National Drinking Water Advisory Committee on recommendations
for the USEPA drinking water research strategy identified a number of distribution system
issues, including research on intrusion, as areas requiring future research (Working Group on
Drinking Water Research 2002). The report concluded that research on the frequency, causes,
mitigation, and health effects of intrusion events was one of the top research needs.
Summary
In summary, it is concluded that transient pressure events occur in distribution systems; that
during these negative pressure events pipeline leaks provide a potential portal for entry of
groundwater into treated drinking water; and that fecal indicators and culturable human viruses
15
-------
are present in the soil and water exterior to the distribution system. To date, all observed
negative pressure events have been related to power outages or other pump shutdowns, although
more research is needed to better characterize the types of systems most prone to these events.
There is insufficient data to indicate whether pressure transients are a substantial source of risk to
water quality in the distribution system. Nevertheless, mitigation techniques can be
implemented, principally the maintenance of an effective disinfectant residual throughout the
distribution system, leak control, redesign of air relief venting, and more rigorous application of
existing engineering standards. Use of high-speed pressure data loggers and surge modeling may
have some merit, but understanding the effectiveness of these tools requires additional research.
More research is needed and this topic should become a priority for both the USEPA and
industry-funded programs.
Acknowledgement
This report was funded by the American Water Works Service Company, Inc. Much of the data
provided is from American Water Works Association Research Foundation project #2686, and is
used with their permission. Project members include Melinda Friedman, EES, Inc; James Funk
and Don Wood, University of Kentucky; Glen Boyd, Tulane University; and AWWSC, Inc. The
AwwaRF project manager is Stephanie Morales, and the Project Advisory Committee is Kevin
Laptos, Tom Walski, Peter Gaewski, and Don Reasoner. The comments and suggestions of
Richard Moser, John Young, Richard Hubel, Stephen Schmidt, Dave Reeves, James Funk, Don
Wood, and Daniel Kelleher are appreciated.
REFERENCES
Abu-Ashour, J. et al, 1994. Transport of microorganisms through soil. Water, Air and Soil
Pollution. 75:141-158.
American Water Works Association and American Water Works Association Research
Foundation. 1992. Water Industry Database: Utility Profiles. AWWA, Denver, CO.
American Waterworks Service Company, Inc. 2002. American Waterworks Service Company
Buried Infrastructure Management Plan. White paper prepared for the US Environmental
Protection Agency workshop on Distribution System Issues, March 7-8, Voorhees, NJ.
Funk, J.E., S.J. VanVuuren, D.J. Wood, M. LeChevallier, and M. Friedman. 1999. Pathogen
Intrusion into Water Distribution Systems Due to Transients. Proceedings of the 3rd
ASME/JSME Joint Fluids Engineering Conference, July 18-22, San Francisco, CA.
Funk, J.E., DJ. Wood, L.S. Reddy, and D.C. Denger. 1992. Pressure surges due to rapid
expulsion of air. In: Bettess & Watts (eds.), Unsteady Flow and Fluid Transients, A.A.
Balkema, Rotterdam, Holland.
Geldreich, E.E., 1996. Microbial Quality of Water Supply in Distribution Systems. Lewis
Publishers, Boca Raton, Florida.
16
-------
Grigory, S. 2002. Time for Leak Free Water Systems.
www.qrigory.com/WaterpipeSeal.htm.
Hellard, M.E., M. I. Sinclair, A. B. Forbes, and C. K. Fairley. 2001. A Randomized, Blinded,
Controlled Trial Investigating the Gastrointestinal Health Effects of Drinking Water Quality.
Environ Health Perspect 109:773-778.
Karassik, I. J., W. C. Krutzsch, W. H. Fraser, and J. P. Messina (eds.). 1976. Pump Handbook.
McGraw-Hill Book Co., New York.
Karim, M, M. Abbaszadegan, and M.W. LeChevallier. 2001. Potential for pathogen intrusion
during pressure transients. JAWWA submitted.
Kirmeyer, G.J., M. Friedman, K. Martel, D. Howie, M. LeChevallier, M. Abbaszadegan, M.
Karim, J. Funk, J. Harbour. 2001. Pathogen Intrusion into the Distribution System. AWWA
Research Foundation and the American Water Works Association, Denver, CO.
Kolpin, D. W., E. T. Furlong, M. T. Meyer, E. M. Thurman, S. D. Zuagg, L. B. Barber, and H. T.
Buxton. 2002. Pharmaceuticals, hormones, and other organic wastewater contaminants in US
streams, 1999-2000: a national reconnaissance. Environ. Sci. Technol. 36: 1202-2111.
Larock, B. E., R. W. Jeppson, and G. Z. Watters. 2000. Hydraulics of Pipeline Systems. CRC
Press, New York.
LeChevallier, M.W., 1999. The Case for Maintaining a Disinfectant Residual. JAWWA. 91:
86-94.
M/DBP (Microbial Disinfection By-Product) Research Council. 2002. Assessment of Microbial
and Disinfection Byproduct Research Needs: Final Report. AWWA Research Foundation,
Denver, CO.
Working Group on Drinking Water Research. 2002. Cross-Cutting Issue Background
Document, Final Draft. Water Distribution System Research Status and Needs. A report to the
USEPA National Drinking Water Advisory Committee, Washington, DC.
Patterson, G.G. and M. J. Focazio. 2001. Contaminants and drinking water sources in 2001:
recent findings of the U.S. Geological Survey. Open File Report 00-510. USGS, Denver, CO.
Payment, P., L. Richardson, J. Siemiatycki, R. Dewar, M. Edwardes, and E. Franco. 1991.
Randomized trial to evaluate the risk of gastrointestinal disease due to consumption of drinking
water meeting microbiological standards. Am. J. Pub. Health 81(6): 703-708.
Payment, P., J. Siemiatycki, L. Richardson, G. Renaud, E. Franco, and M. Prevost. 1997. A
prospective epidemiological study of gastrointestinal health effects due to the consumption of
drinking water. Intern. J. Environ. Health Res. 7:5-31.
17
-------
Payment, P. 1999. Poor efficacy of residual chlorine disinfectant in drinking water to inactivate
waterborne pathogens in distribution systems. Can. J. Microbiol. 45(8): 709-715.
Qaqish, A., D.E. Guastella, J.H. Dillingham, and D.V. Chase. 1995. Control of Hydraulic
Transients in Large Water Transmission Mains, p 507-526, AWWA Annual Conference
Proceedings - Engineering and Operations, Anaheim, CA, June 18-22, 1995 American Water
Works Association, Denver, CO.
Recommended Standards for Water Works, 1997. Great Lakes Upper Mississippi River Board of
State Public Health & Environmental Managers. Health Education Services, Albany, New York.
Simon, A. L. and Korom, S. C, 1997. Hydraulics. 4th edition. Upper Saddle River, NJ, Prentice
Hall.
Snead, M. C., V. P. Olivieri, K. Kawata, and C. W. Kruse. 1980. The effectiveness of chlorine
residuals in inactivation of bacteria and viruses introduced by post-treatment contamination.
Wat. Res. 14:403-408.
Streeter, V.L. and E.B. Wylie. 1967. Hydraulic Transients. McGraw Hill Inc., new York.
Thorley, A.R.D. 1991. Fluid Transients in Pipeline Systems. D. & L. George, Ltd., Herts,
England.
Trussell, R.R. 1999. Safeguarding distribution system integrity. JAWWA 91(1): 46-54.
USC FCCCHR. 1993. Manual of Cross Connection Control, 9th ed. University of Southern
California Foundation for Cross Connection Control and Hydraulic Research. Los Angeles, CA.
Walski, T.M. and T.L. Lutes. 1994. Hydraulic Transients Cause Low-Pressure Problems.
JAWWA 86(12): 24-32.
Wood, D. J., R.G. Dorsch, and C. Lightner. 1966. Wave Plan Analysis of Unsteady Flow in
Closed Conducts. Journal of the Hydraulic Division, Proc. Paper 4716, ASCE, 92( HY2): 83-
110.
18
------- |