United States
Environmental Protection
Agency
Office of Water (4601M)
Office of Ground Water and Drinking Water
Distribution System Issue Paper
Effects of Water Age on Distribution
System Water Quality
August 15, 2002
-------
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:
AWWA
With assistance from
Economic and Engineering Services, Inc
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/requlation revisions.html
Questions or comments regarding this paper may be directed to TCR@epa.gov.
-------
Effects of Water Age on Distribution
System Water Quality
1.0 Introduction
Water age is a major factor in water quality deterioration within the distribution system. The two
main mechanisms for water quality deterioration are interactions between the pipe wall and the
water, and reactions within the bulk water itself. As the bulk water travels through the
distribution system, it undergoes various chemical, physical and aesthetic transformations,
impacting water quality. Depending on the water flow rate, finished water quality, pipe materials
and deposited materials (i.e., sand, iron, manganese), these transformations will proceed to a
greater or lesser extent. The goal of this document is to review existing literature, research and
information on the potential public health implications associated with the decay of water quality
in distribution systems piping networks with time.
2.0 General Description of Topic
2.1 Factors Contributing to Increased Water Age
In addition to meeting current demands, many water systems are designed to maintain pressures
and quantities needed to meet future demands or to provide extra reserves for fire fighting, power
outages and other emergencies. The impacts of these design practices on water age are discussed
below.
2.1.1 Demand Planning
Capital planning necessitates installation of facilities that have excess capacity for water storage
and distribution. It is normal practice to size pipelines for water demands that will occur 20 years
or more into the future. Building distribution facilities that are large enough to accommodate
future demand can in the near term increase water age as the storage volume in the constructed
facility may be large relative to the present day demand. Changes in water demands or use
patterns, such as those caused by the relocation of an industrial water user, annexation of a
neighboring system, or consolidation of multiple systems, can have a significant impact on water
age.
Water demand variations also occur on a daily basis. Daily demand variations can be shown on
a diurnal demand curve, which plots the percentage of daily demand versus time. Figure 1
shows a diurnal curve for a utility that serves approximately 100,000 people, and illustrates how
maximum water use varies over a 24-hour period, based on maximum day demand (MDD)
conditions. The figure also shows that the peaking factor for residential users is slightly larger
than the commercial factor, and that peak demands occur at different times of the day for the two
user groups. Review of the composite usage pattern suggests that typically, water age due to
storage in the distribution system is highest in the early morning hours and lowest in the late
evening (see Figure 2).
Prepared byAWWA with assistance from Economic and Engineering Services, Inc.
1
-------
2.5
Comrnfl!fc*Ol Customer:
Residentiol Customers
2,0
Comp'd&ile i
o
cr
0.5
2400 0300 0600 0900 1200 1500 1800 2100 2400
Time of Day
Figure 1
Diurnal Curve Peaking Data
Figure 2 shows the standard diurnal demand curve developed by A WW A based on average day
flows (AWWA Manual M32, 1989).
Figure 2.
AWWA Average Day Flow Diurnal Curve
Figure 2
AWWA Average Day Flow Diurnal Curve
(Source: AWWA Manual M32)
Prepared by AWWA with assistance from Economic and Engineering Sendees, Inc.
2
-------
2.1.2 Fire Flow Requirements
The effect of fire flow requirements on drinking water reservoir and distributions system
capacity must be quantified on a system-specific basis. The American Water Works Association
Manual M31 Distribution System Requirements for Fire Protection (1998) states the following:
"The decision of whether or not to size distribution system components, including water lines,
appurtenances, and storage facilities for fire protection must be made by the governing body of the
community. This decision is made in conjunction with the water utility if the utility is privately
owned. ..."
Most States require consideration of needed fire flow and may direct the designer to a local fire
official and a particular technical method. Three such methods are presented in AWWA Manual
M31 including:
¦ Insurance Services Office Method,
¦ Iowa State University Method, and
¦ Illinois Institute of Technology Research Institute Method.
The Fire Suppression Rating Schedule is the manual the Insurance Services Office (ISO) uses in
reviewing the fire-fighting capabilities of individual communities. Forty percent of the grading
ISO gives is based on the community's water supply. This part of the ISO survey focuses on
whether the community has sufficient water supply for fire suppression beyond daily maximum
consumption. ISO surveys all components of the water supply system, including pumps, storage,
and filtration. (ISO, 2000). Fire flow requirements for buildings are also given in the Uniform
Fire Code (1997) but the specific technical method is not identified. Each method analyzes a
specific building and is not based on system-wide considerations. According to AWWA Manual
M31, comparisons between the various techniques for computing fire flow are not easily made,
because each situation to which the fire flow calculation is applied varies greatly.
While each method may produce different design flow rates for a given building, once a flow
rate is calculated, an appropriate duration of time over which that flow rate should be applied
must be determined. In Table 1, fire flow rates and durations provided by the Uniform Fire Code
(1997) were used to calculate a fire flow volume.
Table 1
Fire Flow Rates, Durations and Volumes.1
Fire Flow Rate (GPM; at a
minimum pressure of 20 psi.)
Fire Flow Duration
(Hours)
Calculated Volume (Gallons)2
Up to 2875
2
345,000
From 2875 to 3875
3
697,000
Above 3875
4
1,920,000
1. Adapted from the Uniform Fire Code (1997)
2. Calculated Volume = Maximum fire flow rate multiplied by duration.
It is important to note that only a portion of the calculated fire flow volume is provided by
storage created specifically for fire flow. The Water Distribution System Handbook (Mays,
Prepared by AWWA with assistance from Economic and Engineering Services, Inc.
3
-------
2000) provides the following equation (Equation 1) which is based on the Fire Suppression
Rating Schedule (Insurances Services Office, 1980), and other information regarding what
portion of the fire flow rate must be provided from storage (all quantities are in flow units, i.e.,
volume per time):
SSR = NFF + MDC - PC - ES - SS - FDS (Eq. 1)
Where:
SSR = Storage Supply Required,
NFF = Needed Fire Flow.
MDC = Maximum Daily Consumption,
PC = Production Capacity, which is based on the capacity of (the) treatment plant, the well capacity
or the pump capacity, depending on the system
ES = Emergency Supply, or the water that can be brought into the system from connections with
other systems
SS = Suction Supply, or the supply that can be taken from nearby lakes and canals during the fire,
and
FDS = Fire Department Supply, or water that can be brought to the fire by tracks.
Equation 1 shows that fire flows should be achievable in addition to and simultaneous with,
flows associated with maximum daily consumption. It also shows that several supply variables
can affect the need for storage related to fire flow. Figure 3 is an example of the reservoir
storage components required by the Washington State Department of Health Water System
Design Manual (1999). The figure illustrates the relatively minor portion of "operational
storage" compared to the "equalizing" and "emergency" components.
OVERFLOW ElEVATCM
Tom
VOLUME.
OFF
PUWP
HU' - -i
OPERATIONAL
STORAGE (OS)
EQUALIZING
STORAGE
-------
According to AWWA Manual M31, one of the most significant distribution system impacts from
fire flow requirements includes providing adequate storage capacity and meeting requirements
for minimum pipe sizes (e.g., 6-in. [150-mm] pipes in loops and 8-in [200-mm] dead ends) in
neighborhood distribution mains when much smaller pipes would suffice for delivery of potable
water only. Recommended Standards for Water Works (Ten State Standards, 1997) specify a
minimum pipe size of six inches at all locations for providing fire protection. Table 2 shows the
volumetric effect of increased pipe diameter on a per-mile basis.
Table 2
Pipe diameter vs. Pipe Volume (per mile)
Pipe Diameter
2"
4"
6"
8"
10"
12"
18"
Gallons per mile
862
3,466
7,755
13,786
21,540
31,019
69,792
Thus, for every mile of 4-inch pipe that is replaced with 8-inch pipe, the effective volume of the
distribution system increases by greater than 10,000 gallons.
In summary, the effects of fire-flow considerations on system volume and water age vary greatly
from system to system. Few generalizations can be drawn, but AWWA Manual M31 does offer
this:
"In larger systems fire protection has a marginal effect on sizing decisions, but in
smaller systems these requirements can correspond to a significant increase in the
size of many components. In general, the impact of providing water for fire
protection ranges from being minimal in large components of major urban systems
to being very significant in smaller distribution system pipes and smaller
distribution systems."
AWWA Manual M31 also states that most communities are willing to incur the higher cost for
sizing distribution systems for fire flow requirements because of the reduction in property loss
that is possible by using the water system for fire protection. Local planning and zoning
ordinances require specific fire flows for various developments (i.e., single family, multiple-
family, commercial, industrial, etc.), thus mandating the upsizing of installed distribution system
piping. The AwwaRF study "Impacts of Fire Flow on Distribution System Water Quality,
Design, and Operation" (Snyder et al., In Press) is scheduled for publication in 2002.
2.2 Determination of Water Age
The Water Industry Database (AWWA and AwwaRF 1992) indicates an average distribution
system retention time of 1.3 days and a maximum retention time of 3.0 days based on a survey of
more than 800 U.S. utilities. The literature cites examples of both "short" (i.e., less than 3 days)
and "long" (i.e., greater than 3 days) water ages. Several water age estimations published in the
literature are summarized in Table 3 and are described below.
Prepared by AWWA with assistance from Economic and Engineering Services, Inc.
5
-------
Table 3
Summary of water age evaluations
Population Served
Miles of Water
Mains
Range of Water Ages within
System (Days)
Method of Determination
750,000*
1,100
<1-3
Fluoride Tracer
800,000
2,750
3-7+
Hydraulic Model
87,900*
358
> 16
Chloramine Conversion
24,000
86
12-24
Hydraulic Model
*Estimated by using 2.5 multiplier on number of customers served.
As discussed previously, water age is a function primarily of water demand, system operation,
and system design. As water demand increases, the amount of time any given liter of water is
resident in the distribution system decreases. Demand is related to land use patterns, types of
commercial-industrial activity present in a community, the weather (i.e., lawn watering), and
water use habits of the community (i.e., conservation practices, reuse practices). Conservation,
particularly use of reclaimed water on-site or through separate distribution systems, will tend to
lead toward greater water age when all other factors are held constant. The following four
examples illustrate how water age varies from community-to-community as a result of these
factors.
o A utility in North Carolina serving 300,000 customers with 1,100 miles of main
calculated water ages ranging from 2 to 75 hours throughout the distribution system
using a fluoride tracer study (DiGiano, Travaglia and Zhang 2000).
o A Midwest utility with a service population of 800,000 and 2,750 miles of main
recently found based on a hydraulic model that the water age in the distribution
system was typically less than 80 hours while several sites exhibited a water age up to
150 hours (Vandermeyden and Hartman 2001).
o One California utility found water ages exceeding 400 hours in certain areas of the
system, particularly dead end areas, under minimum day and average day demand
conditions (Acker and Kraska 2001).
o A Canadian utility serving 24,000 people with 86 miles of main estimated water age
using a hydraulic model and found that dead-end nodes had a water age ranging from
300 to 600 hours under average day demand conditions (Prentice 2001).
Consequently the importance of water age as a significant driver for water quality conditions in
these distribution systems is variable from system-to-system and even within each system.
The objective of a new AwwaRF Study (#2769) entitled "Evaluating Retention Time to Manage
Distribution System Water Quality" is to evaluate the feasibility and effectiveness of using
distribution system retention time (or water age) as a tool for managing distribution system water
quality. This study will provide case and field study examples of actual system detention times,
and will document impacts of water age on distribution system water quality.
Prepared byAWWA with assistance from Economic and Engineering Services, Inc.
-------
2.3 Water Quality Problems Associated With Increased Water Age
Table 4 lists water quality problems that can be caused or worsened by increased detention time
in the distribution system. Those items marked with an asterisk were identified as having direct
potential health impacts, and are discussed further in this White Paper or in other White Papers.
Other items may impact water quality, but direct health impacts have not been identified.
Table 4
Summary of water quality problems associated with water age
Chemical issues
Biological issues
Physical issues
*Disinfection by-product
Formation
*Disinfection by-product
Biodegradation
Temperature increases
Disinfectant decay
*Nitrification
Sediment Deposition
*Corrosion control effectiveness
*Microbial regrowth / recovery /
shielding
Color
Taste and odor
Taste and odor
* Denotes water quality problem with direct potential public health impact.
3.0 Potential Health Impacts
Various potential health impacts have been associated with the chemical and biological issues
identified in Table 4. The Chemical Health Effects Tables (U.S. Environmental Protection
Agency, 2002a) provides a summary of potential adverse health effects from high/long-term
exposure to hazardous chemicals in drinking water. The Microbial Health Effects Tables (U.S.
Environmental Protection Agency, 2002b) provides a summary of potential health effects from
exposure to waterborne pathogens.
3.1 Disinfection By-Product Formation
Disinfectants can react with naturally occurring materials in drinking water to form organic and
inorganic disinfection by-products (DBPs). With over 200 million people served by public water
systems that apply a disinfectant, there is a very large population potentially exposed to DBPs
through drinking water in the U.S. (USEPA 1998).
The DBP formation potential for each system's water is a function of several chemical and
physical characteristics including type and level of organic matter, type and level of specific
inorganic parameters, pH, temperature, type and level of disinfectant residual, and contact time.
As water ages, there is a greater potential for DBP formation. Higher water temperatures during
summer seasons can increase DBPs as the chemical reactions proceed faster and go further at
higher temperatures. Also, higher water temperatures often cause a higher chlorine demand,
requiring an increased disinfectant dose and resulting in higher DBP formation potential.
Decreases in HAA5 concentrations in some distribution systems are attributed to microbial
activity.
Prepared byAWWA with assistance from Economic and Engineering Services, Inc.
7
-------
The USEPA has identified the following potential adverse health effects associated with HAA5
and TTHMs:
"Some people who drink water containing haloacetic acids in excess of the MCL
over many years may have an increased risk of getting cancer. Some people who
drink water containing trihalomethanes in excess of the MCL over many years may
experience problems with their liver, kidneys, or central nervous system, and may
have an increased risk of getting cancer."
The forthcoming Stage 2 Disinfectants and Disinfection Byproducts Rule will change existing
National Primary Drinking Water Regulations which consist of maximum contaminant levels
and monitoring, and reporting requirements for total trihalomethanes, and haloacetic acids. EPA
believes the implementation of the proposed Stage 2 Disinfectants and Disinfection Byproducts
Rule will reduce peak and average levels of disinfection byproducts in drinking water supplies
which will result in a further reduction in risk from cancer.
The forthcoming Stage 2 Disinfectants and Disinfection By-Products Rule is expected to include
a new monitoring and reporting approach. Compliance with TTHMs and HAA5 standards will
be based on a locational running annual average using monitoring data gathered at new
monitoring locations selected to capture representative high levels of occurrence. MCL
violations could potentially occur at single locations such as a finished water storage facility due
to site-specific situations, including excessive water age or chlorine addition at the storage
facility.
3.1.2 Biodegradation of Disinfection Byproducts
Biodegradation of some HAA species has been observed over several years in a southeastern
U.S. utility and at other utilities (Williams, Williams and Gordon 1996). An evaluation of the
water quality in the southeastern U.S. utility did not reveal any unusual conditions other than
relatively low residual chlorine levels and a high heterotrophic bacteria count (Williams,
Williams and Gordon 1996). A certain bacillus species (X. autotrophicus) is reported to be
capable of degrading dichloroacetic acid and dibromoacetic acid, but not trichloroacetic acid
(Williams, Rindfleisch and Williams 1994; Baribeau et al. 2000).
The proposed Stage 2 DBPR monitoring strategy requiring systems to identify representative
high TTHMs and HAA5 formation locations (not necessarily maximum residence times) will
address this issue. According to the July 27-28, 2001 Federal Advisory Committee (FACA)
meeting summary, Initial Distribution System Evaluation (IDSE) studies can be based on various
data sources including historical TTHM and HAA5 data, calibrated network hydraulic models,
and tracer studies, in place of, or in combination with, an intensive monitoring program.
Associated guidance is anticipated to identify biodegradation of HAAs as an issue to consider in
selecting Stage 2 DBPR monitoring locations.
3.2 Nitrification and Microbial Regrowth
Nitrification is a microbial process by which reduced nitrogen compounds (primarily ammonia)
are sequentially oxidized to nitrite and nitrate. Nitrifying bacteria are slow growing organisms,
Prepared byAWWA with assistance from Economic and Engineering Services, Inc.
8
-------
and nitrification problems usually occur in large reservoirs or low-flow sections of the
distribution system. According to Kirmeyer et al (1995), operational practices that ensure short
residence time and circulation within the distribution system can minimize nitrification
problems. Low circulation areas of the distribution system are the types of locations where
nitrification is more likely to occur since detention time and sediment buildup can be much
greater than in other parts of the system.
Under the Safe Drinking Water Act (SDWA), primary MCLs have been established for nitrite-N,
nitrate-N, and the sum of nitrite-N plus nitrate-N. The MCLs are 1 mg/L for nitrite-N, 10 mg/L
for nitrate-N, and 10 mg/L for nitrite + nitrate (as N). The nitrite and nitrate MCLs are
applicable at the point-of-entry to the distribution system. Review of nitrification episodes and
information gathered from the literature indicates that an MCL exceedence within the
distribution system due to nitrification is unlikely, unless source water nitrate-N or nitrite-N
levels are close to their applicable MCLs. Additional information related to nitrate and nitrite
formation is provided in the Nitrification White Paper.
Much research has been conducted related to the growth of microorganisms within the
distribution system. There are numerous factors that impact microbial growth rate, many of
which are related to increased water age. These include temperature, system hydraulics, nutrient
availability, and disinfectant efficacy, among others. A detailed discussion of health implications
associated with the survival and/or growth of pathogens within the distribution system is
provided in a separate White Paper.
3.3 Corrosion Control Effectiveness
Corrosion control effectiveness can be related to water age. With increased detention time there
are impacts on the effectiveness of phosphate inhibitors and pH management in poorly-buffered
waters.
The use of orthophosphates has been a successful practice for minimizing corrosion of piping
and materials containing lead and copper. For utilities with hard water and high levels of
dissolved inorganic carbonate (DIC), blended ortho- and poly-phosphates have been used. Many
polyphosphates, particularly those of the linear chain variety, tend to hydrolyze, i.e., convert to
orthophosphate to some degree with time and passage through the distribution system.
(Economic and Engineering Services, Inc; Illinois State Water Survey; 1990). Reversion to
orthophosphate can limit the effectiveness of blended ortho- and poly-phosphates as corrosion
control inhibitors.
pH stability can be difficult to achieve for utilities with soft, poorly-buffered waters, especially in
portions of the system with increased water age. Interaction with cement linings can
significantly increase pH (as discussed in the Permeation and Leaching White Paper), and
unstable, soft, low-mineralized waters can revert to untreated water pH conditions by the time
the water reaches the customer's tap. These waters are common in surface water supplies of the
Pacific slope, New England, and the southeastern United States (Economic and Engineering
Services, Inc,; Illinois State Water Survey; 1990).
Prepared byAWWA with assistance from Economic and Engineering Services, Inc.
9
-------
Requirements for installing and maintaining effective corrosion control treatment are addressed
under the USEPA Lead and Copper Rule. Utilities are required to maintain optimal water
quality parameters at the point of entry to the distribution system and at several locations within
the distribution system. Problems associated with increased water age and corrosion control
effectiveness may impact compliance with water quality parameter requirements or with lead and
copper action level compliance.
4.0 Prevention/Mitigation Methods
4.1 Tools for Determining Water Age
Tracer Studies
Tracer studies have been performed to calculate water age throughout a distribution system,
calibrate water quality and hydraulic models, and to enhance the study of water age in relation to
water quality parameters such as chlorine residual or trihalomethanes. Tracer studies can utilize
injected chemicals such as fluoride, or calcium chloride. Alternatively, in systems with multiple
sources with varying water quality characteristics (such as differences in water hardness or
conductivity); these natural constituents can be used as the tracer. Finally, during transitional
periods in system operation, such as changeovers from chlorination to chloramination, the
resulting constituents can be traced. Other utilities have taken advantage of a fluoride system
shut down, use of alternative water sources, or a switch in coagulant to trace water through the
distribution system.
Consideration of the tracer chemical stability, continued regulatory compliance, and customer
perceptions must be included in the planning and implementation of a tracer study. For example,
if fluoride is used as a tracer, it may be preferential to discontinue fluoride feed so that
interference with fluoride uptake on pipe walls for newly-fluoridated systems can be avoided.
Conversely, State health departments may not allow for the purposeful discontinuation of
fluoridation. Although lithium chloride is used as a tracer in the United Kingdom, customer
acceptance in the United States has prevented its wide-spread use. Table 5 (Smith, Grayman,
and Friedman, N. P.) provides a summary of example tracers and associated research and
development needs.
Table 5
Example Tracer Summary Matrix
Tracer
Research and
Development Needs
Regulatory Issues
Problems/Comments
Fluoride
Continuous on-line
monitor for use in
distribution system
Difficulties in adding
or shutting off fluoride
in some places
May be non-conservative due to pipe
wall uptake in systems that do not
normally use fluoride.
Sodium chloride
NA
NA
Requires relatively large volumes of
tracer
Calcium chloride
NA
NA
NA
Lithium chloride
NA
NA
Popular in the UK
Coagulants
Possible post-precipitation
NA
Utilities may be reluctant to vary
coagulant feed (type and quantity)
Prepared byAWWA with assistance from Economic and Engineering Services, Inc.
10
-------
Chlorine (pulsed)
NA
Limits on upper and
Non-conservative and affected by wall
lower chlorine
demand
residuals
Note: NA indicates that research needs have not yet been identified.
Models
Mathematical models that represent the hydraulic behavior of the movement of water have been
used to estimate water age in distribution systems (Clark and Grayman, 1998). Steady-state
travel time models were first introduced in the mid-1980s (Males et al., 1985). These models
were subsequently extended to dynamic representations that determined varying water age
throughout the distribution system (Grayman et al., 1988). Water quality models can be used in
conjunction with hydraulic models to predict concentrations of chlorine, DBPs, and other
constituents in a distribution system (Vasconcelos et al., 1996). The effect of interaction of the
flowing water with the pipe wall can be estimated using relationships proposed by Rossman et al.
(1994). For each non-conservative water quality parameter, rates and mechanisms of reaction
(decay and growth) must be identified and measured for a particular system.
Careful calibration of both hydraulic and water quality models is needed to generate an accurate
prediction of water age and water quality conditions under varying demand scenarios.
Typically, hydraulic and water quality conditions must be measured simultaneously since the
changes in water quality are directly related to hydraulic and system operation conditions.
Where feasible, hydraulic and facility operational information may be provided by the SCADA
system. In addition, a calibration program may use existing chlorine monitor outputs. Where
automated data is unavailable at facilities, manual data collection may be necessary.
Models may be limited in their ability to accurately predict water age for the following reasons:
¦ Skeletonization: Skeletonization may be needed if the water system contains more pipe
segments than the model can handle. While the physical distance between two nodes is
maintained in a skeletonized model, smaller diameter pipes ( 8-inch) may be excluded.
The impact of skeletonization on the accuracy of water age predictions will vary from
system to system, depending on the proportion of smaller diameter pipes compared to
the overall number of pipes in the system and the piping configuration.
¦ Insufficient calibration: Typical calibration just looks at a single time step and does not
look at extended period analysis. If there is a small error in the single step or if the
pump operation is not exactly right, the flows throughout the system may not be
accurately represented.
¦ Water Storage Tanks: Tanks are modeled as completely mixed reactors in most
models. This typically leads to an underestimation of the water age. Reservoir mixing
is addressed in more detail in the Finished Water Storage Facilities White Paper.
¦ Inaccurate total demand or demand allocation: If the overall demand is miscalculated it
could result in more/less source and reservoir operation than actually occurs. It could
also lead to more/less water crossing a system than actually occurs.
Prepared byAWWA with assistance from Economic and Engineering Services, Inc.
11
-------
¦ Errors in Model Development: Inaccurate settings on PRVs or pump curves would
limit the accuracy of water age analysis.
Because water age calculations require extended period simulations and include more complex
analysis of the water system than do typical modeling exercises, the chances of inaccuracy
increase.
A simplified model of water age in tanks and reservoirs was developed in the early 1990s
(Grayman and Clark, 1993) and later refined as part of an AwwaRF study entitled "Water
Quality Modeling of Distribution System Storage Facilities" (Grayman et al., 2000). In that
study, more complex computational fluid dynamics (CFD) models that represent the
hydrodynamic behavior were also applied to tanks and reservoirs. Methods were designed to
provide estimates of the mixing characteristics, and the distribution and concentration of
conservative constituents (or substances that decay according to a first-order decay function) to
predict water age within a reservoir.
4.2 Design
Standard design guidelines for hydraulic considerations in the planning and design of distribution
system piping networks are available in:
• Recommended Standards for Water Works (Ten State Standards, 1997)
• Guidance Manual for Maintaining Distribution System Water Quality (AwwaRF,
2000)
• Sizing Water Service Lines and Meters (AWWA M22, 1975)
• Distribution System Requirements for Fire Protection (AWWA M31, 1998)
• State regulations
Design guidelines for sizing storage facilities are discussed in the Finished Water Storage
Facility White Paper.
4.3 Operations
Operations practices can impact flow direction, flow velocity, and water age. The water's
hydraulic path and resulting detention time are affected by distribution system valve settings and
pump station operations. Closing distribution system valves (purposefully or accidentally) can
result in dead-ends with out proper blow-offs. In an effort to reduce hydraulic retention times,
the utility may modify pressure zone boundaries or pressure set points at pumping stations. With
these changes, it is important to assure that a minimum pressure of 20 psi is maintained at all
locations at all times, and that service dependability is maintained.
Additionally, reservoir operations can significantly impact water age and associated water
quality decay as discussed in the White Paper on Finished Water Storage. Increased water age
within reservoirs is usually attributed to under utilization and/or poor mixing. According to
Grayman et al. (2000), a measure of the mixing in a reservoir is the time it takes for the contents
of the reservoir to become relatively well mixed following a change in the inflow concentration
Prepared by AWWA with assistance from Economic and Engineering Services, Inc.
12
-------
of a constituent of tracer. Based on empirical work for utilizing a scale model, the relationship
shown in Equation 2 was developed for cylindrical tanks under fill and draw operation:
Mixing time (in seconds) = 9 V2/3 (d/Q) (Eq. 2)
Where:
V = volume of water in tank at start of fill (cubic feet)
Q = inflow rate (cubic feet per second)
d = inlet diameter (feet)
Mixing time is highly dependent on the inflow rate and quantity of water flowing into the
reservoir. Often, mixing can be increased by operating a reservoir so that the inflow rate is
increased. Additionally, daily turnover rates (inflow/outflow quantity) can be increased to
reduce overall water age.
4.4 Maintenance
Water system flushing is an important tool for helping to keep the water system clean and free of
sediment, and to remove stagnant water.
Utilities can also clean or replace deteriorated pipelines that are known to be contributing to bulk
water quality decay. A variety of pipeline cleaning techniques are utilized including mechanical
scraping, pigging, swabbing, chemical cleaning, and flow jetting. Each technique has it benefits
and drawbacks and should be tailored to the specific site. Depending on the cleaning technique
used, relining of the pipeline may be necessary to prevent accelerated corrosion.
4.5 Source Water Treatment
Source water treatment can help mitigate many of the water quality problems associated with
increased water age by improving the biochemical stability of finished water. With enhanced
stability, the rates and extent of many chemical and biological reactions are suppressed.
Biochemical stability is closely related to the amount and speciation of organic matter in water.
Over time, natural organic matter reacts with disinfectants to produce DBPs. Biodegradable and
assimilable organic carbon provide energy and biomass which support the proliferation of
microbial communities. The removal of these precursors during treatment precludes the ability
of many of these reactions to occur, regardless of water age. Common treatment practices
include enhanced coagulation, biological filtration, ultra- or nanofiltration, and granular activated
carbon filtration. Ozone may be used as a primary disinfectant to convert organic carbon to a
form that is more readily biodegraded by microbes in biological filters. Enhanced coagulation is
specified as a best available technology in the D/DBP Rule for removing natural organic matter.
Treatment to remove organics, inorganics, and turbidity will also curb the rate of chlorine decay,
thus allowing a higher residual to reach further into the system and persist longer.
Prepared byAWWA with assistance from Economic and Engineering Services, Inc.
13
-------
pH stability is important in minimizing DBP formation. Without the appropriate level of source
water treatment, the pH in some systems may vary by up to two pH units. pH variability may be
reduced by increasing source water alkalinity and buffer intensity.
Iron corrosion in unlined cast iron pipe can contribute to the water's chlorine demand. Corrosion
may be controlled with pH and alkalinity adjustment or addition of a corrosion inhibitor such as
polyphosphate or addition of calcium.
Treatment to remove iron and manganese can help to reduce the sediment load in pipelines and
storage facilities. Treatment alternatives for iron and manganese removal include
oxidation/filtration processes, oxidation/adsorption process, lime softening, and caustic soda
softening.
The relationship between water age and temperature cannot be altered with source water
treatment.
4.6 Indicators of High Water Age
There are several indicators that may suggest high water age. These include aesthetic
considerations that may be identified by consumers, as well as the results of distribution system
monitoring efforts. It should be noted that indicators can be triggered by factors other than water
age, such as insufficient source water treatment, pipe materials, and condition/age of distribution
system.
Aesthetic Indicators
The following indicators may be identified during water consumption:
• Poor taste and odor - Aged, stale water provides an environment conducive to the
growth and formation of taste and odor causing microorganisms and substances.
• Discoloration - Water in low flow areas and dead-ends often accumulate settled
deposits over time. During a demand period, these deposits are entrained and
degrade the clarity and color of the water.
• Water temperature - Stagnant water will approach the ambient temperature.
Monitoring Indicators
The following indicators require sample collection and analysis:
• Depressed disinfectant residual - Chlorine and chloramines undergo decay over
time.
• Elevated DBP levels - The reaction between disinfectants and organic precursors
occur over long periods.
• Elevated bacterial counts (i.e., heterotrophic plate count).
• Elevated nitrite/nitrate levels (nitrification) for chloraminating systems.
Prepared byAWWA with assistance from Economic and Engineering Services, Inc.
14
-------
5.0 Summary
Water age is a major factor contributing to water quality deterioration within the distribution
system. Water age is primarily controlled by system design and system demands. Thus, water
age can vary significantly within a given system. Increased temperatures typically associated
with increased water age can cause reactions to proceed faster and go further. Water quality
problems that can be exacerbated by increased water age include DBP formation, decreased
corrosion control effectiveness, nitrification, and microbial growth/regrowth. The Stage 1 and
proposed Stage 2 DBP Rules recognize the relationship between DBP occurrence and water age,
and have established monitoring requirements based on this link. The Lead and Copper Rule
requires utilities to maintain optimal water quality parameters within the distribution system to
ensure effective corrosion control treatment. For poorly buffered waters, increased water age
can impact compliance with water quality parameter requirements. Potential health issues
associated with Nitrification and Microbial Regrowth are addressed in separate White Papers.
Tools for evaluating water age include hydraulic models, tracer studies, and monitoring
programs. Existing AwwaRF and AWWA manuals provide guidelines for considering water age
during design, operation, and maintenance of distribution system facilities. The objective of a
new AwwaRF Study (#2769) entitled "Evaluating Retention Time to Manage Distribution
System Water Quality" is to evaluate the feasibility and effectiveness of using distribution
system retention time (or water age) as a tool for managing distribution system water quality.
Bibliography
Acker, C. and D. Kraska. 2001. Operation and Maintenance of Water Distribution Systems
During and After Conversion to Chloramine Disinfection. In Proc. of AWWA Annual
Conference. Denver, Colo.: AWWA.
AWWA (American Water Works Association). 1975. AWWA Manual M22—Sizing Water
Service lines and Meters. Denver, Colo.: AWWA.
AWWA (American Water Works Association). 1989. AWWA ManualM32-Distribution Network
Analysis for Water Utilities. Denver, Colo.: AWWA.
AWWA (American Water Works Association). 1998. AWWA ManualM31—Distribution System
Requirements for Fire Protection. Denver, Colo.: AWWA.
American Water Works Association and American Water Works Association Research
Foundation. 1992. Water Industry Database: Utility Profiles. Denver, Colo.: AWWA.
Baribeau, H., S.W. Krasner, R. Chinn, and P.C. Singer. 2000. Impact of Biomass on the
Stability of Haloacetic Acids and Trihalomethanes in a Stimulated Distribution System.
In Proc. of the AWWA Water Quality Technology Conference. Denver, Colo.: AWWA.
Clark, R.M. and W.M. Grayman. 1998. "Modeling Water Quality in Drinking Water Systems",
Denver, Colo.: AWWA.
Prepared by AWWA with assistance from Economic and Engineering Services, Inc.
15
-------
DiGiano, F.A., A. Travaglia, and W. Zhang. 2000. Water Age (EPANET) and Mean Residence
Time of Fluoride Tracers in Distribution Systems. In Proc. ofAWWA Water Quality
Technology Conference. Denver, Colo.: AWWA.
Economic and Engineering Services, Inc., Illinois State Water Survey. 1990. Lead Control
Strategies. Denver, Colo.: AWWA and AwwaRF.
Grayman, W.M., R.M. Clark, R.M. Males. 1988. Modeling Distribution-System Water Quality:
Dynamic Approach., J. WRPMD, ASCE, 114(3) 295-312.
Grayman, W.M., and R.M. Clark. 1993. Using Computer Models to Determine the Effect of
Storage on Water Quality, Jour. AWWA, 85(7):67-77.
Insurance Services Office. 2000. http://www.iso.com/catalog/docs/prod037.htm
Insurance Services Office. 1980. "Fire Suppression Rating Schedule" ISO Commercial Risk
Services, Duluth, Ga.
Kirmeyer, G.J., M. Friedman, J. Clement, A. Sandvig, P. Noran, K. Martel, D. Smith, M.
LeChevallier, C. Volk, E. Antoun, D. Hiltebrand, J. Dyksen, R. Cushing. 2000.
Guidance Manual for Maintaining Distribution System Water Quality. Denver, Colo.:
AWWA and AwwaRF.
Lee, Y. and S.G. Buchberger. 1999. Migration of Pollutants in Dead-End Zones of Water
Distribution Systems. In Proc. of the 1999 AWWA Water Quality Technology
Conference. Denver, Colo.: AWWA.
Males, R.M., R.M. Clark, P.J. Wehrman, W.E. Gates. 1985. Algorithm for Mixing Problems in
Water Systems. J. HY, ASCE, 111(2): 206-219.
Mays, Larry W. 2000. Water Distribution System Handbook. McGraw-Hill Companies Inc.
Moore, A.C., B.L. Herwaldt, G.F. Craun, R.L. Calderon, A.K. Highsmith, and D.D. Juranek.
1994. Waterborne Disease in the United States, 1991 and 1992. J AWWA, 86(2): 87-99.
Prentice, M. 2001. Disinfection Modeling Study for the Owen Sound Water Distribution System.
In Proc. of AWWA Annual Conference. Denver, Colo.: AWWA.
Pontius, F.W. 2000. Chloroform: Science, Policy, and Politics. JAWWA. 92(5): 12, 14, 16, 18.
Recommended Standards for Water Works (Ten State Standards). 1997. Great Lakes Upper
Mississippi River Board of State Public Health and Environmental Managers. Health
Education Services. Albany, N.Y.
Prepared by AWWA with assistance from Economic and Engineering Services, Inc.
16
-------
Rossman, L.A., R.M. Clark, and W.M. Grayman. 1994. "Modeling Chlorine Residuals in
Drinking-Water Distribution Systems", J.EE, ASCE, 120(4): 803-820.
Smith, C., W. Grayman, M. Friedman. 2001. "Evaluating Retention Time to Manage
Distribution System Water Quality". Proposal for RFP 2769. AwwaRF and KIWA.
Snyder, J.K., A.K. Deb, F.M. Grablutz, S.B. McCammon, W.M. Grayman, R.M. Clark, D.A.
Okun, and D. Savic. In Press. "Impacts of Fire Flow on Distribution System Water
Quality, Design, and Operation". Denver, Colo.: AWWA and AwwaRF.
Uniform Fire Code (Volume 1). 1997. International Fire Code Institute. 10th Edition. Whittier,
California.
U.S. Environmental Protection Agency. 2002a. Chemical Health Effects Tables.
http://www.epa. gov/safewater/tcrdsr. html. May 7, 2002.
U.S. Environmental Protection Agency. 2002b. Microbial Health Effects Tables.
http://www.epa. gov/safewater/tcrdsr. html. May 7, 2002.
U.S. Environmental Protection Agency. 2001. National Primary Drinking Water Regulations:
Draft Stage 2 Disinfectants and Disinfection By-Products Preamble.
U.S. Environmental Protection Agency. 1999. EPA Guidance Manual Alternative Disinfectants
and Oxidants.
U.S. Environmental Protection Agency. 1998. National Primary Drinking Water Regulations:
Disinfectants and Disinfection By-Products; Final Rule. Federal Register, 63(241):
69389-69476.
Vandermeyden, C. and D. Hartman. 2001. Water Quality Modeling of a Complete Distribution
System Development, Results, and Practical Applications. In Proc. of AWWA Annual
Conference. Denver, Colo.: AWWA.
Vasconcelos, J., P. Boulos, W. Grayman, L. Kiene, O. Wable, P. Biswas, A. Bhari, L. Rossman,
R. Clark, and J. Goodrich. 1996. "Characterization and Modeling of Chlorine Decay in
Distribution Systems", Denver, Colo.: AWWA and AwwaRF.
Washington State Department of Health Water System Design Manual (1999)
Williams, S.L., D.F. Rindfleisch, and R.L. Williams. 1994. Deadend on Haloacetic Acid. In
Proc. of the AWWA Water Quality Technology Conference. Denver, Colo.: AWWA.
Williams, S.L., R.L. Williams and A.S. Gordon. 1996. The Impact of Bacterial Degradation of
Haloacetic Acids in the Distribution System. In Proc. of the AWWA Water Quality
Technology Conference. Denver, Colo.: AWWA.
Prepared by AWWA with assistance from Economic and Engineering Services, Inc.
17
------- |