Technical and Economic Capacity
of States and Public Water Systems
to Implement Drinking Water Regulations
REPORT TO CONGRESS
U.S. ENVIRONMENTAL PROTECTION AGENCY
SEPTEMBER 1993
-------
•
-------
TABLE OF CONTENTS
Page
List of Exhibits
Executive Summary
Overview of SDWA, Public Water Systems, and National Surveys 1976-86
Section 1 Identification, Selection, and Regulation of Contaminants
1.0 Identification of Contaminants for the Drinking Water Priority List
1.1 Selection of Contaminants for Regulation
1.2 Basis for Regulatory Decisions
1.3 Regulatory Development Process
Section 2 History, Benefits, and Costs of Current Regulations
2.0 History of Drinking Water Regulations
2.1 Adverse Health Effects of Contaminants
2.2 Benefits of Contaminant Regulation
2.3 Cumulative Costs of Treatment
2.4 Total National Cost of Drinking Water Regulations
Section 3 Financial and Technical Capacity of Systems to Monitor
3.0 Cumulative Costs of SDWA Monitoring Requirements
3.1 Laboratory Capacity
Section 4 Capacity of Systems to Treat
4.0 Cumulative Cost of Compliance Requirements
i-v
10
10
14
20
22
26
26
32
32
40
41
44
44
50
57
57
-------
4.1
4.2
4.3
4.4
Section 5
5.0
5.1
5.2
Financial Capacity to Afford SDWA Treatment and Other Compliance Costs 63
Existing Financing Mechanisms 70
Technical Capacity of Systems to Treat 75
Options Available to Improve the Financial and Technical Capacity of
Small Water Systems
Capacity of States to Implement Drinking Water Regulations
Past and Current Funding Levels
Ways to Address the Shortfall
States That Have Successfully Addressed Budget Shortfalls
Section 6 Compliance with Federal Regulations
Section 7 Public Water System Supervision (PWSS) Information Management
79
103
103
106
109
113
122
APPENDICES
-------
LIST OF EXHIBITS
Overview
Exhibit 0.1
Exhibit 0.2
Exhibit 0.3
Community Water Systems: 12 Size Categories
Public Water Systems: Five Size Categories
Distribution of Community Water Systems by Size
Section 1
Exhibit 1.1 Risk Assessment and Risk Management Relationship
Section 2
Exhibit 2.1
Exhibit 2.2
Exhibit 2.3
Exhibit 2.4
First 83 Contaminants
Contaminants Regulated Under the SDWA
Known Health and Cost Benefits of Regulated Contaminants
Total National Cost Impact of SDWA Regulations
Section 3
Exhibit 3.1
Exhibit 3.2
Exhibit 3.3
National Annual Monitoring Costs Under the SDWA
Average Annual Monitoring Costs Per Household Under the SDWA
Assumptions used in Estimating Monitoring Costs
Section 4
Exhibit 4.1
Exhibit 4.2
Exhibit 4.3
Exhibit 4.4
Exhibit 4.5
Exhibit 4.6
Exhibit 4.7
Exhibit 4.8
Exhibit 4.9
Exhibit 4.10
Exhibit 4.11
Exhibit 4.12
Exhibit 4.13
Exhibit 4.14
Profile of SDWA Treatment Requirements - Ground Water Systems
Profile of SDWA Treatment Requirements - Surface Water Systems
Annual Household Costs for Treatment - Ground Water Systems
Annual Household Costs for Treatment - Surface Water Systems
Change in Average Household Costs for Drinking Water
Distribution of Household Annual Costs for SDWA Treatment — Systems
Serving Fewer Than 3,301 Persons
Household Expenditures on Selected Utilities
Summary of Technology Applications
Illustration of Small System Ownership as a Function of Size in Pennsylvania
Ownership of Small Community Water Systems
Restructuring in Response to Environmental Change
Common Techniques Used in Restructuring Non-Viable Systems
Estimate of Restructuring Potential for Small Community Water Systems
Small System Restructuring Experience in Selected States
-------
Section 6
Exhibit 6.1
Exhibit 6.2
Exhibit 6.3
Exhibit 6.4
Exhibit 6.5
Community Water Systems in Violation in FY 1992
Percentages by System Size of CWSs in Violation
Community Water Systems in Violation by Contaminant Group
CWSs in Violation
PWSS State Activity FY's 1988-1992
Section 7
Exhibit 7.1
Exhibit 7.2
PWSS Two-Tiered Information Management Approach
Status of State Information Systems
-------
EXECUTIVE SUMMARY
The institutions and organizations responsible for providing safe drinking water in the U.S.
are undergoing fundamental change due to Congress' 1986 amendments to the Safe Drinking Water
Act (SDWA). These changes are bringing about impressive gains in public health protection. EPA
estimates that full implementation of the SDWA Lead and Copper Rule will reduce the exposure of
156 million people to lead. Another 600,000 children will be protected from unsafe levels of lead in
their blood. Compliance with the Surface Water Treatment Rule is expected to prevent at least
80,000-90,000 cases of gastro-intestinal illness. Other EPA rules protect against cancer and a range
of chronic diseases. The passage of the 1986 amendments, and the subsequent efforts of States and
water suppliers, signal a revitalized national commitment to safe drinking water and public health
protection.
Despite progress so far, threats of waterborne disease and chemical contamination remain a
national concern. EPA solidly supports the SDWA's basic premise that safe drinking water is a right
of all Americans. However, resource constraints, and some statutory obstacles, could hinder further
progress. The funding shortfall, particularly for State programs, is well documented. Between 1991
and today, the General Accounting Office (GAO) has prepared five reports that identify funding
shortages as a serious problem in the Public Water Supply Supervision (PWSS) program and the
ground water program.1
SDWA's requirement that EPA regulate an additional 25 contaminants every 3 years is a
statutory obstacle to realizing the full health protection benefits of existing standards. Eighty-four
contaminants are currently regulated under SDWA, and an estimated total of 112 will be regulated by
1995. At that time, the most serious public health threats will be addressed by standards. However,
the prospect of fully realizing the health benefits of these regulations is diminished by the "25 every 3
years" mandate. The continuing stream of regulations will add considerably to the regulatory burden
on States and drinking water systems (especially small systems), and detract from implementation of
priority contaminants among the first 112 standards. In short, fundamental reform of the SDWA — to
focus on priority public health threats - is as important as adequate funding for attaining the full
potential of the Act.
State Programs and Primacy
In this report, EPA estimates that the current annual State funding shortfall for implementing
Federal drinking water requirements is approximately $162 million. EPA estimates that 1993 State
funding needs total $304 million, yet only $142 million is available from State and Federal sources.
The five GAO reports are: (1) Drinking Water: Stronger Efforts are Needed to Protect Areas Around Public Wells
From Contamination, April 1993; (2) Drinking Water: Key Quality Assurance Program is Flawed and Underfunded, April
1993; (3) Drinking Water: Widening Gap Between Needs and Available Resources Threatens Vital EPA Program, July 1992;
(4) Water Pollution: More Emphasis Needed on Prevention in EPA's Efforts to Protect Ground Water, December 1991; (5)
Environmental Protection: Meeting Public Expectations With Limited Resources, June 1991.
-------
Because of this shortfall, several States have been unable to implement and enforce all the SDWA-
mandated regulations, or cannot adequately fund key functions. Moreover, funding problems
limit the ability of regulatory agencies to tailor programs to recognize local conditions or to
effectively target priority concerns. Addressing the State shortfall is important because funding
problems lead to inefficient regulation and, ultimately, greater costs to water systems, ratepayers, and
taxpayers.
EPA has tried to address the State capacity problem head-on. In 1976 Federal grants to States
for the Public Water Supply Supervision program were $7.5 million. By 1986, when SDWA
reauthorization took place, Federal grants to States totalled $29.5 million. Federal grants this year
(FY 1993) total $58.9 million. Such increases, however, have not matched the pace o| program
growth. To address the shortfall EPA has worked closely with the States to find other means of
building State capacity. With on-site technical support, EPA has been a partner with State agencies
and other organizations in forming coalitions to obtain increased State revenues and to enact
alternative financing mechanisms to support State primacy programs. States have been successful in
increasing their funding of drinking water programs from $63 million in 1988 to $83 million in 1993.
EPA has undertaken a number of other efforts to address the problem of State resource
shortages. EPA has worked with the States to develop PWSS Program Priority Guidance (issued in
1992) that identifies the "baseline" requirements and ranks the discretionary components of a State
primacy program. This guidance is designed to encourage efficient use of State resources by
focussing on priority public health risks. The guidance also specifies activities that must be carried
out to maintain primacy. As a follow-up to the guidance, EPA developed a "national resource
model" that States can use to gauge program funding needs. To help communicate drinking water
program needs to citizens and decisionmakers, EPA has begun an initiative called the "State
Measures" project that identifies the strengths and weaknesses of individual State programs. These
efforts were designed not only to help States find solutions, but to send the clear message that failure
on the part of a State to implement basic SDWA requirements would result in primacy withdrawal.
Despite EPA's efforts, many States have been unable to adequately fund or carry out drinking
water regulations. This past year, EPA began the process of withdrawing primacy from the State of
Maine because of inadequate program funding. Two other States, Washington and Alaska, were
formally notified of the potential withdrawal of primacy because of a failure to adopt Federal
regulations. Washington and Alaska have now taken steps to avoid the loss of primacy, and the
process to withdraw primacy from Maine is on "hold" due to recent actions by the State to address its
resource shortage.
Unless the situation changes, however, other States are likely to have trouble meeting primacy
requirements. Unfortunately, public health protection is the first victim of failed State primacy, since
EPA is not staffed to run effective programs at the State level. States provide a range of services and
functions, as part of their overall drinking water programs, that simply cannot be duplicated
effectively by the Federal government. EPA would need substantial staff increases to properly
administer more than a couple State programs. Replacing State primacy with an EPA-run program is
not a realistic national strategy to protect drinking water and public health.
-------
Drinking Water Systems
This report contains detailed cost estimates for all of EPA's regulations. Compliance with the
standards for 84 contaminants regulated so far is expected by 1995 to cost Public Water Systems
about $1.4 billion (in 1991 dollars) per year. Costs to individual households to comply with Federal
mandates range from a few dollars per year in metropolitan areas, to several hundred dollars per year
in small communities that have contamination problems. Because of the high costs of meeting
Federal requirements in small communities, ensuring the safety of drinking water in these areas is one
of the program's greatest challenges.
EPA has long recognized that small communities face difficulties complying with drinking
water regulations. In recent years, small communities have become the focal point of EPA technical
assistance. In the past year, EPA's Small System Technology Initiative, a cooperative effort
involving the private sector, began to show promise in demonstrating the effectiveness of affordable
treatment technologies. EPA has worked with the States and other organizations to form the National
Training Coalition to improve the focus and delivery of technical assistance. EPA has developed
handbooks, guidances, personal computer software, and other materials geared toward helping small
systems. In addition, the Agency has worked closely with non-profit technical assistance providers to
improve field-level service. In early 1993, EPA appointed a new Small System Coordinator for all of
EPA's small systems activities and to work with other Federal agencies to help small systems.
Technical assistance alone, however, will never solve compliance problems for small systems
that cannot afford treatment. Financial assistance is also needed. In February, 1993, President
Clinton proposed a Drinking Water State Revolving Fund program to provide financial assistance to
systems for infrastructure investments needed for SDWA compliance. The President has asked
Congress to provide $599 million in 1994 and $1.0 billion each year from 1995 through 1998 for the
proposed program. In addition, the President's plan calls for increasing the U.S. Department of
Agriculture/Rural Development Administration's water and wastewater loan authority by $230 million
and its grant authority by $140 million in 1994, with additional increases for 1995 through 1998.
While the drinking water SRF will surely help systems meet rising compliance costs, financial
assistance needs to be coupled with a new regulatory paradigm (discussed below), as well as other
programs specifically targeted to small systems. These programs include: efforts to promote lower-
cost small scale technologies, opportunities to comply with regulations by practicing pollution
prevention, improved operator training and technical assistance, and programs for physical or
institutional restructuring to obtain economies of scale.
Prevention Efforts
Although Congress did not ask EPA to report on the effectiveness of prevention approaches as
a means of achieving the goals of the SDWA, EPA believes prevention should have a prominent role.
The Wellhead Protection Program, created by the 1986 amendments to SDWA, as well as EPA's new
National Guidance for Comprehensive State Ground Water Protection Programs, provide a good
foundation for building prevention principles into the SDWA's regulatory program. The wellhead
protection prdvision of the SDWA requires States to develop programs to protect ground water in
111
-------
areas where contamination may enter into drinking water supplies. Currently, 27 States and
Territories have adopted wellhead programs which have been approved by EPA. EPA is conducting
preliminary reviews of wellhead program submittals for 8 additional States. These State programs are
intended to create locally controlled wellhead protection activities, including delineation of wellhead
areas, inventories of potential contamination sources, and institution of management measures to
protect ground water supplies. Although many communities have undertaken efforts to protect their
wellhead areas, overall progress in ensuring local implementation on a nationwide basis has been
limited, primarily due to funding limitations. EPA estimates that there are about 2000 ongoing
community ground water protection efforts, including local wellhead protection programs, throughout
the country.
The wellhead protection program is a central component of EPA's efforts to encourage States
to develop Comprehensive State Ground Water Protection Programs (CSGWPP). Under EPA's
guidance, States are working to integrate a range of prevention activities to protect ground water,
particularly those high priority ground waters that serve as sources of drinking water. The Agency is
working particularly closely with 11 pilot States throughout the country to demonstrate accelerated
development of CSGWPPs which satisfy EPA's criteria.
Drinking Water Standards
Congress' mandate to regulate 83 specific contaminants identified in the SDWA's 1986
amendments, plus an additional 25 contaminants every 3 years, limits the Agency's ability to
concentrate on establishing and implementing national standards for only the highest priority
contaminants. While Congress' mandate has successfully brought about many protective standards,
this approach has yielded some unfortunate results. In some cases, contaminants have been forced
onto regulatory schedules that out pace EPA's ability to develop needed technical information. Some
regulations have unqualified benefits, yet impose significant costs. New approaches for selecting
contaminants and developing regulatory responses need to be a central component of SDWA reform.
The complexity of EPA drinking water rules is another issue often cited as a major
implementation problem for States and systems. EPA has already taken a number of steps to expand
State flexibility and to more actively involve the regulated community and States in the regulatory
development process. This summer, EPA successfully completed regulatory negotiations, involving a
wide array of participants, to develop the disinfectant and disinfection byproducts regulation. This
process is designed to lead to regulations that are agreed upon in advance by parties with often
conflicting points of view. EPA also commissioned a State and Regional task force to identify
implementation problems and to recommend solutions. Tailoring solutions to each State's regulatory
program requires building considerable flexibility into Federal regulations. Unfortunately, many
opportunities for increasing flexibility also add complexity and increase State administrative costs,
since multiple decisions at the State and system level take the place of one-time decisions at the
national level. While EPA has statutory authority to address a number of implementation problems,
some SDWA changes are necessary.
IV
-------
OVERVIEW OF SDWA, PUBLIC WATER SYSTEMS, AND NATIONAL SURVEYS 1976-86
Section 519(a) of Public Law 102-389 (EPA's 1993 Appropriations Act) requires the
Administrator of the United States Environmental Protection Agency to report to Congress with
recommendations concerning the reauthorization of the Safe Drinking Water Act. Section 519,
known as the Chafee-Lautenberg Amendment, also requires EPA to provide information on seven
general topics:
• the adverse health effects associated with contaminants in drinking water and public
health and the other benefits that may be realized by removing such contaminants;
• the process for identifying contaminants in drinking water and selecting contaminants
for control;
• schedules for the development of regulations and compliance with drinking water
standards;
• the financial and technical capacity of drinking water systems to implement monitoring
requirements associated with regulated and unregulated contaminants and options to
facilitate implementation of such requirements, with special emphasis on small
communities;
• the financial and technical capacity of drinking water systems to install treatment or
take other action needed to ensure compliance with drinking water standards and
options to facilitate compliance with such standards, with special emphasis on small
communities;
• the financial and technical capacity of States to implement the drinking water
program, including options for increasing funding of State programs; and
« innovative and alternative methods to increase the financial and technical capacity of
drinking water systems and the States to assure effective implementation of the Act.
This report provides insight into each of these areas. The first two sections provide
background on the status of drinking water regulations, including the Agency's contaminant selection
and regulatory processes, health effects, health benefits, and costs of current regulations; and
timetables for implementation.
The next five sections focus on the capabilities of systems and States to implement drinking
water regulations, including monitoring, installation and operation of treatment systems, reporting,
enforcement, and data management. An emphasis is given to small systems, since they typically
experience a greater relative impact from the regulations than larger systems. The sections also
include options for improving capacity.
Due to the short time frame of this project, only data currently available were used in this
report. Major sources of information included the Regulatory Impact Assessments (RIAs) for each
regulation and national drinking water surveys. Given the variety and date of information sources,
-------
• establish filtration requirements for nearly all surface water systems and disinfection
requirements for all public water systems;
• implement a new ban on lead-based solder, pipe, and flux materials;
• implement enhanced enforcement powers; and
• develop additional programs to protect ground water supplies (i.e., wellhead
protection and sole source aquifer protection programs).
In 1988 Congress passed a new provision to the SDWA, the Lead Contamination Control Act,
requiring EPA to maintain an updated accounting of water coolers with lead-based components and
develop guidance for controlling lead contamination in school drinking water supplies.
Background: Public Water Systems
Presently there are approximately 200,000 Public Water Systems (PWSs) regulated under the
Safe Drinking Water Act serving 243 million Americans (the remainder obtain their drinking water
from private wells).
A Public Water System (PWS) provides piped water for human consumption to at least 15
service connections (such as households, businesses or schools), or serves an average of at least 25
people at least 60 days a year. PWSs can be community, non-transient non-community, or transient
non-community systems. Approximately 60,000 of the almost 200,000 PWSs are community water
systems (CWSs), about 25,000 are non-transient non-community (NTNC) water systems, and
approximately 115,000 are transient non-community (TNC) water systems. Each type of PWS is
defined as follows:
A Community Water System (CWS) is a PWS that provides water to the same population
year-round. Thirty percent of all PWSs are CWSs. Even though CWSs collectively serve a lot of
people, most CWSs are small, serving less than 3,300 people. Many of these systems are privately-
owned and operated. (Privately owned systems include those owned by single individuals or by a
group of investors.) Some of the smallest systems are in trailer parks or housing subdivisions.
Approximately 80 percent of all CWSs obtain their water primarily from a ground water source. The
proportion using ground water is greater for small systems. The remaining systems are served
primarily by surface sources such as lakes, rivers, and reservoirs.
A Non-transient Non-community Water System (NTNC) is a PWS that regularly serves at
least 25 of the same people at least six months of the year. Approximately 13 percent of all PWSs
are NTNCs. Examples of .these systems include schools, factories, and hospitals that have their own
water supplies. Historically, NTNC water systems were required to meet only those standards
designed to prevent short-term health problems such as bacteria, nitrate, and turbidity. Since the
1986 SDWA Amendments, however, EPA requires NTNCs to meet the same standards as CWSs.
A Transient Non-community Water System (TNC) caters to transitory customers in non-
residential areas such as campgrounds, motels and gas stations. Approximately 57 percent of all
-------
Twr* Unlike CWSs and NTNCs, EPA requires TNCs to meet only those standards
-
standards for filtration and disinfection.
Size Categories Used in Estimates of Benefits and Costs
When analyzing PWS data, EPA frequently divides systems into either 12 or five size
wnen J^J*' * . n ' . Q 2 Based on the best information available, EPA has
raise capital.
Exhibit 0 3 shows the distribution of Community Water Systems by size. Systems serving
their ability to comply with drinking water regulations is limited.
Exhibit 0 3 and many of the exhibits and analyses presented later in this report, are based
on data fof CWSs raC than d'ata for all PWS, EPA has better data on CWSs than for the other
types of systems.
Nearly 90 percent of the total number of CWSs are small and very small
and reporting requirements. More
be provided in the later sections of this report.
information on these costs
-------
Background: National Surveys 1975-86
to asses the nation's water s
coS from 969
^^ of conditions in the noon's
a risibility for drinking water was
^rs^
rr^S^^
proWem. These monitoring efforts include: ; ;
National Organics Reconnaissance Survey (1975);
National Organics Monitoring Survey (1976-77); ; ,
. National Screening Program for Organics (1977-1981);
Community Water Supply Survey (1979);
Ground Water Supply Survey (1981); .
. National Inorganics and Radionuclide Survey (1984-86*
National Pesticide Survey (1987-1990).
Below is a
current regulations:
of *e
„«..„;,.
^, j » A :« 107*; this survey measured organic
from 80 public water
^
-- nonitortag requirem
March 1976 and January 1977
of to survey supported the
-------
Community Water Supply Survey. This survey, from 1979, examined samples from about
450 surface water and ground water systems serving populations from 25 to 100,000 persons.
Samples were examined for the presence of 10 volatile organic chemicals, trihalomethanes, and 15
inorganic contaminants. Total organic carbon levels were also determined to provide an indication of
general organic contamination. This survey was the first national survey to indicate widespread
occurrence of industrial solvents in drinking water.
'Ground Water Supply Survey; Conducted in 1981, this survey was initiated to provide a
clearer picture Of the extent of contamination of ground water systems by volatile organic compounds.
A total of 945 systems were sampled, of which 466 were chosen at random. The other 479 systems
were selected from areas near potential sources of contamination. Analyses were made for 29 volatile
organic compounds, five trihalomethanes, and total organic carbon. The results from this and the
earlier EPA surveys were used to regulate volatile organic chemicals (VOCs) in 1987. The data are
also being used currently to develop the disinfection by-products rule.
National Inorganics and Radionuclide Survey. Conducted from July 1984 through May 1986,
this survey provided information on 990 ground water public water systems. For one of the
radionuclides, Radium 228, a special geologically based design was implemented to assure that highly
vulnerable areas of the county were adequately represented in the study. The survey included five
other radionuclides and 36 inorganic chemicals. The results were used to support development of the
Phase II and V rules and the radionuclides rule.
National Pesticide Survey. The National Pesticide Survey was a statistically designed
monitoring survey undertaken to assess the occurrence and concentration of 126 pesticides, pesticide
degradation products, and nitrate in 566 public water systems using ground water and 783 private
rural drinking water wells before any treatment. The study was conducted from March 1987 to
February 1990. Its results are being used to develop current regulations for pesticides.
The EPA surveys described above provided only "snapshot" measurements of contamination,
which may not have been representative of average water quality over time at the system studied.
The number of systems sampled ranged from 0.1 % to 3% of the total number of systems. Most
systems were sampled only once; a few were sampled for four consecutive quarters. The occurrence
results cannot be used to completely determine whether a detection was a false (or atypical) positive,
and whether a non-detection was a false (or atypical) negative result. .The same contaminants were
not tested for in each survey; therefore, it is difficult to compare testing results from survey to
survey. In addition, since some of EPA's surveys targeted vulnerable systems, the results may
overstate the number of systems and their degree of contamination for the chemicals examined. This
bias towards contaminated systems, and the small system size as a portion of the total number of
systems, may have limited EPA's ability to estimate national occurrence trends for these chemicals
and for others as well.
Despite these shortcomings, the results from these surveys led to a greater understanding of
the extent of organic chemical contamination in drinking water. It also became evident that inorganic
chemical compounds, chemicals introduced into the aquatic environment, the byproducts of the
chemicals used to make the water safe, and* the pipes used to convey the drinking water were all
contributors to drinking water contamination. The results helped the Agency design the criteria for
selecting and regulating contaminants described in Section 1, and assessing health benefits and costs
discussed in Section 2.
-------
SECTION 1
IDENTIFICATION, SELECTION, AND REGULATION OF CONTAMINANTS
Section 1412 of the 1986 amendments to the Safe Drinking Water Act required the Agency to
issue National Primary Drinking Water Regulations for 83 specified contaminants. Since 1986, most
of the Agency's effort to develop drinking water regulations has been directed at those 83
contaminants. Seventy-six of those 83 contaminants are now regulated and regulations for the
remaining seven contaminants are underway. (The regulatory status of those 83 contaminants is
discussed in Section 2.)
Section 1412 of the Act also directs the Administrator to publish a triennial list of
contaminants which may require regulation (the Drinking Water Priority List, or DWPL). Within 24
months of publication, EPA must propose regulations for not less than 25 contaminants from the list.
To satisfy Congress' mandate to regulate a specific number of contaminants, EPA is working to
develop a process for selecting contaminants that pose the greatest remaining public health risks.
EPA has identified contaminants for the DWPL based on a preliminary determination that a
compound does or could possibly pose a threat to contaminate drinking water supplies. EPA's second
step is to decide whether the health risk of the listed contaminant is sufficient to warrant national
regulation. EPA has gone through the first step (placing contaminants on the DWPL) twice: first in
1988 and again in 1991. EPA has not yet completed the second stage (selecting 25 contaminants for
regulation).
This section characterizes the approach EPA used in 1988 and 1991 for placing a contaminant
on the priority list, and then describes a new approach the Agency is considering for the next DWPL,
due to be published in 1994. After that, the section describes the criteria EPA is considering for
selecting the first 25 contaminants from the DWPL for regulation. This section also discusses the
decision criteria EPA uses to establish Maximum Contaminant Level Goals (MCLGs) and Maximum
Contaminant Levels (MCLs) or treatment techniques. This section closes with a discussion of the
Agency's process for issuing drinking water regulations.
1.0 Identification of Contaminants tor the Drinking Water Priority List
There are over 77,000 chemicals in production or use in the United States, all released into
the environment to some extent. Most of these chemicals do not occur in drinking water at levels that
pose a health risk. On the other hand, some chemicals may occur at levels of health concern that are
not even measurable using routine, broad-spectrum analytical technologies (e.g., dioxin and arsenic).
1.01 Historical Criteria for Placement on Priority List
In January 1988, EPA published the first Drinking Water Priority List and in January 1991,
the Agency expanded the DWPL to seventy-seven contaminants. EPA used three general criteria to
select candidates for the list:
• Occurrence of the substance in public water systems; or
physical/chemical/environmental characteristics and use patterns of the substance that
indicate the potential for occurrence in public water systems at levels of concern;
10
-------
• Documented or suspected adverse health effects of the contaminant; and
• Availability of sufficient information on the substance, including health effects data,
analytical methods, and treatability studies, to suggest that a regulation could be
developed before the court-ordered deadline of proposal (which, at that point, was
June, 1993). ;,
EPA believed these criteria would meet the Congressional mandate to identify "important or
potentially important" drinking water contaminants.
The major contaminant data used to develop the 1991 priority list were Superfund site
information and the National Pesticide Survey. The list also included disinfectants and disinfection
by-products known to occur in drinking water on a widespread basis, substances identified in other
national surveys besides the National Pesticide Survey, and chemicals specifically requested by States.
Numerous other chemicals were included based only upon toxicity and a subjective perception that
they potentially could occur and persist in drinking water. For most of the chemicals under
consideration, data on occurrence in drinking water were not available. (At the time, EPA had used
its authority under Section 1445 of the Act to require monitoring for a number of unregulated volatile
organic chemicals on the list, but the data were not yet available.) Compounds identified through
Toxics Release Inventory reports under SARA Section 313 were not included. (Appendix A contains
a discussion of past experience in identifying contaminants for the priority list.)
The Agency's goal in developing the 1991 priority list was to create a sufficiently large
working list of chemicals to ensure the availability of 25 candidates for regulation as required by the
statute. While the Agency focussed on identifying candidates that had the potential to occur in
drinking water, availability of sufficient information to develop a standard within the specified time
frame also was important. This approach led to the inclusion of some chemicals which the Agency
subsequently determined to be unlikely to occur in drinking water.
1.02 New Algorithm for the Identification of Chemicals for the Priority List
The complexities of identifying drinking water contaminants for possible regulation is
illustrated by a survey of 263 organic chemicals in drinking water wells conducted by the California
Department of Health Services. This targeted survey found that individual chemicals generally did
not occur in more than one percent of the sampled wells. However, some contaminants were found
at a higher proportion of wells where all contaminants were sampled in every well. Taken as a
whole, almost eight percent of all wells were found to be contaminated by at least one chemical.
Similar results have Been observed in other state surveys. To the extent these surveys are
representative of national conditions, occurrence at less than one percent of the sites corresponds to
exposure for hundreds of thousands of individuals.
Assessing chemical occurrence and exposure is further complicated by the highly variable
nature of occurrence. Chemical levels in a given system vary over time as a function of rate and
volume of pollutant release, environmental conditions and even the rate of water utilization by a
community. For example, data obtained by the US Geological Survey in the midwest demonstrated
ten- to hundred-fold variation in the water-based concentrations of pesticides over the course of a
single growing season.
11
-------
The Agency is now developing a new scheme to identify the chemicals most likely to occur or
actually occurring in the largest number of areas at levels of health concern. The preliminary
approach discussed below would provide the "biggest bang for the buck." While it might also leave
unattended problems that occur in more limited geographic areas, the scheme needed to recognize the
limited resources the Agency could apply to the identification effort and the limited data available on
many candidates. EPA hopes to make considerable enhancements to this scheme over the next year
or two. A scheme like this will be used to identify contaminants for regulation or further study as
described in EPA's recommendations to Congress for legislative changes.
For the near term, EPA will focus on chemicals with Agency-quantified health effect
estimates, and chemicals with sufficient health data for the risks to be quantified. These chemicals
may be identified by other EPA offices, other Agencies, or the public as posing a potential drinking
water health risk. Several hundred chemicals have quantified health effects estimates. These health
effects include short term (acute) health effects like nervous system dysfunction and gastro-intestinal
disturbances. They also include long term (chronic) health effects like cancer, organ damage (kidney,
liver, heart, lung), and reproductive abnormalities. ,
Whenever possible, EPA will use information on contaminant occurrence, whether from the
Agency's or other Federal and State surveys, or from monitoring data supplied by public water
systems. The Agency will supplement occurrence information with data oh chemical production, use
and release. While these data are indirect indicators, predictions could be compared to existing and
new monitoring data on occurrence of both regulated and unregulated contaminants in raw and
finished water. Over time, comparison of these indicators to actual monitoring data would allow EPA
to improve its ability to predict occurrence.
Actual Occurrence Data -
As part of the new scheme, EPA plans to examine the drinking water surveys described above
and other data bases. Whenever possible, both intake water and treated water data will be
considered. The Storage and Retrieval Database (STORET) is likely to be one of the principal
sources of surface-water information. STORET data from a recent year identified 159 chemicals as
occurring in the nation's waters.
Characterizing ground water occurrence is more difficult. For pesticides, the principle tool
available to the Agency is the file associated with the recently updated Pesticides in Ground Water
Database. This file contains information on as many as 302 pesticides and related compounds in
more than 60,000 public and private drinking water and monitoring wells nationwide. The Pesticides
in Ground Water Database, which is a compilation of many State surveys, is designed to assess
pesticide mobility. Considerable analysis will be required to develop approaches for its use in
prioritizing regulatory development.
Even less ground water data are available on organic chemicals other than pesticides. For
these chemicals, the Agency hopes to use other sources of information, including the unregulated
contaminant data collected under Section 1445 of SDWA. The National Inorganic and Radionuclides
Survey data will continue to be used for estimating inorganic chemical and radioriuclide occurrence.
Regional, State and local information will also be used to the greatest extent possible, but generalized
watershed and aquifer characteristics must always be evaluated to ensure that Regional/State results
12
-------
can be used in a national context. New, targeted surveys might also be undertaken to develop
national estimates.
Release/Discharge ,
The Agency will also consider in the new scheme release and discharge data, weighted for the
relative toxicity and persistence of a chemical in ground and surface water, to project the
pervasiveness of a chemical. The rates of dispersal and degradation of chemicals are dependent upon
physical and chemical properties which are adequately definable for most chemicals. Sediment and
fish tissue analyses also provide an indication of environmental fate and persistence.
To date, the Agency has identified four sources of information on releases:
• The Toxics Release Inventory (TRI) contains information on large volume chemical
releases by industrial facilities to all media, including surface water and ground water.
Chemicals currently regulated under SDWA account for only 7.5 percent of the
tonnage which industries report as being discharged.
• The Permit Compliance System (PCS) contains information on discharges to surface
water from the largest twenty percent of the more than 63,000 facilities with National
Pollution Discharge.Elimination System (NPDES) permits under the Clean Water Act.
- As with the TRI data, chemicals currently regulated under the SDWA account for
only ten percent of the total volume of chemicals permitted under NPDES to be
released, into surface water. ,
• The "Summary of Data on Municipal Solid Waste Landfills" contains a limited
amount of information on leachates from municipal landfills. These data were
assembled as part of the effort to develop Subtitle D landfill regulations under the
Resource Conservation and Recovery Act. Efforts to identify additional State sources
of data from State enforcement and implementation files are planned for the coming
year.
• The Agency for Toxic Substances and Disease Registry's HAZDAT contains narrative
information on releases from Comprehensive Environmental Response, Compensation,
and Liability Act (CERCLA, or Superfund) National Priority List sites and other
emergency events. Use and periodic review of this data base will help to address the .
priorities of the, Superfund emergency response program.
In general, there is more release and discharge information for organic chemicals (other than
pesticides) than for pesticides and for many inorganic chemicals.
Production/Use Assessment
Occurrence data and release/discharge information tend to be biased toward past
contamination and chemicals which have undergone significant regulatory scrutiny. Occurrence data
generally are available only for chemicals which are measurable in water using multi-analyte, broad-
spectrum laboratory methods. Thus, the Agency will utilize production and chemical-use information
to provide a more forward-looking and possibly preventive component to the priority-listing effort.
13
-------
For minor contaminants and those which are not directly produced (e.g., dioxin), the Agency will
attempt to estimate occurrence using data on chemicals with similar properties and uses.
1.1 Selection of Contaminants for Regulation
Most of the 83 contaminants listed in the 1986 SDWA amendments have now been regulated
and EPA is now developing its first set of regulations under the requirement to regulate 25
contaminants every three years. EPA is considering regulating four types of contaminants (microbes,
inorganic and organic chemicals, disinfectants and their byproducts) in this rule. A proposal is now
scheduled for March 1994.
SDWA requires EPA to issue standards for any contaminant which "may have an adverse
effect on the health of persons and which is known or anticipated to occur in public water systems."
The law authorizes the Agency to require public water systems to install a particular treatment
technique to control a contaminant if there is no way for the PWS to ascertain whether the
contaminant is present at levels of concern.
The Agency is participating in a negotiated rulemaking with representatives of key
stakeholders to address pathogens, disinfectants, and disinfectant byproducts in this rule. In June
1993, this group reached consensus on the contents of these portions of the rule, including which
disinfectants and disinfection byproducts would be regulated. The Agency has committed to publish
the product of this consensus in its proposed rule. With the exception of Cryptosporidium, the group
believes there are insufficient data at this time to add to the list of regulated pathogens. Instead, the
group agreed to propose an information collection rule so that an appropriate decision can be made in
a few years as to whether to regulate more pathogens in an Enhanced Surface Water Treatment rule.
With regard to organic and inorganic chemicals, EPA has developed a strategy for selecting
these chemicals from the priority list which follows the procedures described below.
1.1.1 Risk Assessment and Determination of the MCLG
The first step in deciding whether to regulate a contaminant is to conduct a risk assessment.
During the risk assessment, EPA combines health information with data on exposure of persons to the
contaminant in drinking water to characterize the health risk to the population and establish a
maximum contaminant level goal (MCLG) "at the level at which no known or anticipated adverse
effects on the health of persons occur and which allows an adequate margin of safety." The MCLG
is not an enforceable limit.
EPA risk assessments, including those of the drinking water program, generally follow the
process described by the 1983 National Academy of Sciences publication, Risk Assessment in the
Federal Government: Managing the Process, as modified by Agency-specific guidance adopted by the
Agency's Risk Assessment Forum. There are four steps to the process:
• Hazard Identification
• Dose-Response Analysis
• Exposure Assessment
• Risk Characterization
14
-------
In the hazard identification step, the Agency determines whether the available scientific data
.indicate a causal relationship between the contaminant and an adverse human health effect. Data can
come from either animal studies or epidemiological studies of humans. EPA performs a critical
review of available data on both acute and chronic (i.e., short and long term) effects related to oral,
dermal, inhalation, and other intake pathways.
In the dose-response analysis, the Agency quantifies the relationship between the exposure
(dose) and the adverse health effects. Since animal study data are often the only data available, the
Agency applies uncertainty factors to account for the possibly higher sensitivity of humans and other
uncertainties related to human toxicity.
For the exposure assessment, EPA considers whether and how much of the contaminant
transfers from drinking water supplies to humans. For most contaminants the major route of
exposure is ingestion of tap water. EPA assumes that about half of one's ingestion of tap water
occurs directly. The other half occurs indirectly through water used in cooking and through
consumption of beverages such as coffee, tea, or reconstituted orange juice. For most contaminants,
EPA assumes the consumption of two liters of water daily as an average over a lifetime. For volatile
contaminants, the exposure pathway is different. Directly ingested tap water will contain some of the
contaminant, but the contaminant will probably volatilize from water which is cooked. Exposure
from volatile contaminants also results from inhalation of gases released from water during showers,
washing dishes, and other activities. The exposure assessment also considers other sources of
exposure to the contaminant (such as food).
Results from the hazard identification, dose-response analysis and exposure assessment are
brought together in the risk characterization step, which describes the overall risk - the nature and
significance of any adverse health effects - to the potentially exposed individual or population. The
end result of the risk assessment process is a proposed MCLG.
EPA evaluates many types of information in conducting a risk assessment, including data
from animal toxicity studies, epidemiology studies and exposure information. The risk
characterization identifies uncertainties in the assessment, qualitative and quantitative, as part of the
overall confidence in the assessment.
If the basis for the MCLG is carcinogenicity, information on the weight of evidence and
quantitative estimate is presented, including a discussion of:
• the type of data (human or animal)
• weight of evidence of all data considered as a whole
• potency and route of exposure in the studies
• whether tumors appeared in multiple sites, species and sexes
• the time it takes tumors to develop
• correlation between the amount of exposure and the number and type of tumors
formed,
• survival or confounding effects in the animals,
• historical background incidence of tumors,
• mechanism of carcinogenicity if known,
• supporting mutagenicity data, and
• cancer risk associated with consumption of drinking water.
15
-------
For contaminants which are regulated as carcinogens, EPA's longstanding policy to set the
MCLG at zero. Currently, scientists do not fully understand what causes cancer, and whether even
very low levels of exposure might result in tumor formation. Therefore, the health goal is no, or
zero, exposure.
If the basis for the MCLG was non-cancer health effects, the following information is
discussed in the risk characterization:
reference dose, or level at which no adverse effects are likely to occur,
basis for calculating the reference dose (human or animal data),
critical effect/endpoint,
route of exposure and doses,
duration of exposure,
no and/or lowest adverse effect level,
uncertainty factors,
data gaps and qualitative uncertainties,
sensitive subpopulations,
amount of exposure coming from drinking water compared to other sources like food,
and
• (in some cases) use of a safety factor to account for possible carcinogenicity
Risk assessments are ultimately reviewed by EPA's Reference Dose Work Group (RfD Work
Group) for non-carcinogenic health effects or EPA's Cancer Risk Assessment Verification Endeavor
group (CRAVE) for carcinogenic health effects, as well as the EPA Science Advisory Board.
Departures from the process described above may occur because of the absence of sufficient
data, which necessitates omitting certain analyses or using default assumptions. Also, because of the
significant differences between microbial and chemical contaminants, portions of the process for
microbial contaminants differ.
1.1.2 Known or Anticipated Occurrence
In addition to health effects, the Agency considers potential for occurrence in drinking water
when selecting chemicals for regulation. This analysis is based on the same type of information used
to identify and select contaminants for the priority list but is more rigorous. In this effort, the
Agency attempts to go beyond the prioritization question of "how likely is it that the contaminant will
appear in drinking water," to the question of "how frequently will it occur in drinking water."
Ideally, the Agency would be able to consider the magnitude, extent, frequency, environmental fate,
and environmental transport of individual chemicals in soil and drinking water when making this
selection. Although data and science are steadily improving, a database that would allow
consideration of this full range of factors is rarely available.
• Known Occurrence
Numerous studies by Federal, State, university, or trade groups provide a variety of
monitoring data that may be helpful in assessing the likelihood of a chemical occurring in public
drinking water. The monitoring data EPA required for unregulated volatile organic chemicals has
16
-------
also provided valuable information, as has the Toxic Release Inventory data base and effluent
discharge information collected under the Clean Water Act. When such monitoring data are
available, EPA considers data base quality. EPA also examines information on the level and extent of
occurrence, including the frequency of detections, number of sampling points, and to the extent
known, the conditions at sampling sites where a chemical was monitored. In some situation, State-
specific studies are difficult to use when there are only a few of them and their results conflict with
national surveys.
EPA believes that a chemical's presence must be estimated to occur at a level of potential
significance from a health perspective (based on the MCLG) and in enough locations to warrant
Federal regulation. At the same time, EPA requires that a satisfactory margin of safety be present so
that the public can be assured of the safety of its water supply. In the past, EPA has not always had
sufficient occurrence data to make this determination for large numbers of compounds, and has had to
make many judgments based on anticipated occurrence.
As part of the effort to improve on the process for selecting contaminants for regulation, EPA
is considering several markers to assess whether occurrence may be sufficient to warrant a possible
health concern. For chemicals with known occurrence, the marker may be the relationship of the
occurrence levels to the MCLG. EPA is considering whether a contaminant found at levels equivalent
to a percentage of the MCLG in a raw drinking water source (e.g., a reservoir or river near a water
intake, an aquifer used as a drinking water source), or in finished drinking water is an appropriate
trigger for further consideration. Another factor may be the number of areas in which contamination
is believed to occur at that level. EPA may also use results of ambient water monitoring as a
predictor of a contaminant's likelihood to occur in public water supplies. For example, the presence
of a contaminant in ambient ground or surface water at levels within a specified range below the
MCLG in a certain number of locations would trigger further consideration. For carcinogens, where
the MCLG is zero, EPA would first calculate what level of occurrence would equate to a lO^'to 1O6
individual lifetime risk level. EPA would then look at the relationship between occurrence data and
levels representing a 10'6 risk level.
• Anticipated Occurrence
For many chemicals, limited occurrence information is available, whether for lack of an
analytical method or because limited resources or a targeted purpose prevented the inclusion of these
chemicals in occurrence surveys. For these chemicals, EPA is considering assessing the physical-
chemical properties of the chemical in question, such as solubility; half life in water; tendency to
leach or run off; and whether it is discharged directly to, or used in, the water. EPA may also utilize
production and use information.
For pesticides, those contaminants with use of 100,000 pounds or more of active ingredient a
medium to high leaching potential or runoff potential, plus a reasonably long half life might be
deemed "anticipated" in drinking water. If a pesticide is used for crops directly in the water such as
tor rice, it might have potential to contaminate drinking water. In some parts of the country the
potential for ground or surface water contamination by such pesticides is being addressed through
preventive measures. However, EPA believes an enforceable standard may still be necessary as a
back-stop to these efforts. For other synthetic organic chemicals, discharge to surface waters or to
the air in substantial quantity and in numerous locations may qualify the contaminant for consideration
based on anticipated occurrence, provided the contaminant remains stable in water.
17
19
-------
selected.
Notwithstanding this new approach to selecting contaminants for regulation EPA is required
the current statute to regulate an additional 25 contaminants every 3 years. The specific criteria
SiS colSts couW well be determined - under the current SDWA mandate - by the
TSSSSLts rather than on the basis of public health threat. A V^*™"**
would be to gather additional occurrence data and other information (as necessary) to determme the
need for regulation.
1.2 Basis for Regulatory Decisions
Under section 1412 of the Safe Drinking Water Act (SDWA), 42U.S.C 300g-l, EPA
establishes Maximum Contaminant Level Goals (MCLGs) and National *«™%Q^f™*"
for drinking water contaminants. As discussed in Section 1 .1 of this report,
eaMe health-based goals which are to be set at the level at which no known or
effects on the health of persons occur and which allows an adequate margm of
safety.
NPDWRs are enforceable requirements generally based upon the highest allowable>
concentration of a contaminant in drinking water, called a Maximum Contaminant Level (MCL).
Under SDWA EPA must "specify a maximum contaminant level for such contaminant which is as
Sofe toTmaxImum contaminant level goal as is feasible." Where it is not feasible t« ««£m *e
level of a contaminant in drinking water, EPA is authorized to specify a treatment technique, in heu
of an MCL, that will prevent adverse health effects to the extent feasible.
EPA determines the appropriate level of the MCL through risk management. The Agency
considers the findings of the risk assessment, monitoring and treatment feasibility, analytical
meiurement performance, and costs. The legislative history to the 1986 Amendments guided EPA to
consider costs to large metropolitan and regional utilities at various contaminant levels when
considering feasibility. Congressional discussion pointed to the availability of exemptions for small
systems^which could not afford to install treatment. Once established, the MCL is an enforceable
limit.
In issuing a drinking water regulation, EPA not only sets the MCLG and MCL, but also
determines the best available technology, monitoring requirements, analytical methods, and acceptable
levels of uncertainty in measurements.
1.2.1 Analytical Method and Acceptable Levels of Uncertainty
EPA identifies candidate analytical methods from within the Agency or from other sources^
including academia, industry, U.S. Geological Survey, U.S. Department of Agriculture, U.S. Food
and Drug Administration, and the American Society for Testing and Materials. The extent towh.ch
methods are developed depends on program resources, the relative importance or novelty of the
method, and other factors.
20
-------
Prior to publication, a method is tested in reagent or drinking water samples to determine
method detection limits, accuracy, precision, and other factors. The data are often produced by one
analyst, but care is taken to ensure that the method is not specific to one instrument.
If lead time is sufficient, the published method may include results from several analysts and
laboratories. Ideally, methods are tested in one or more of the following ways:
• The method is field-tested in an occurrence survey (e.g., the National Pesticide Survey);
• EPA solicits several laboratories to validate a method in several types of finished drinking
water samples;
• EPA conducts a multi-laboratory validation study among laboratories that volunteer (e.g.,
EPA's joint studies with volunteer labs from the Association of Official Analytical Chemists);
• Accuracy and reproducibility data are obtained from EPA-sponsored laboratory performance
evaluation studies. These studies are conducted twice a year as part of the drinking water
laboratory certification program. They are primarily designed to certify laboratories for
regulated contaminants only. Recently, though, the studies were expanded to include early
testing of new analytes for possible regulation in the future.
Ideally, performance evaluation, method-validation and other studies test the performance of a method
over a range of concentrations that include the MCLG (provided it is not zero).
The published method specifies the statistically-based method detection limit (MDL), which is
the minimum concentration of a substance that can be measured and reported with 99% confidence
that the true value is greater than zero. The method also includes quality assurance criteria and
single-lab accuracy and reproducibility results for spikes into tap or reagent water. Quantitation limits
with error estimates (i.e., acceptance limits) are specified in the drinking water regulation, not in the
published method.
EPA considers it important to be able to analyze a contaminant consistently and accurately at
the MCL level in order to enforce the regulatory standard. Historically, EPA has used a measure
called the practical quantitation level (PQL) to describe the lowest concentration that can be reliably
achieved within specified limits of precision and accuracy during routine laboratory operating
conditions. The PQL thus represents a level consistently achievable by good laboratories within
specified limits during routine operating conditions. EPA is currently considering an alternative
approach to determining reliable measurement levels that is more statistically rigorous. The new
approach would set Reliable Quantitation Limits (RQLs) and is based on American Chemical Society
methodology. It is expected to be proposed for public comment shortly.
1.2.2 Feasible Treatment Technologies
The SDWA directs EPA to set the maximum contaminant level (MCL) as close to the MCLG
as is "feasible with the use of the best technology, treatment techniques and other means, which the
Administrator finds, after examination for efficacy under field conditions and not solely under
21
-------
laboratory conditions, are available, taking costs into consideration." The legislative history to
SDWA directs the Agency to consider feasibility in terms of costs to metropolitan and regional water
systems with relatively clean source water. Based on this directive, the MCL decision relies
extensively on evaluations of the availability of technology, the performance of technologies in
removing the subject contaminant, and the costs of applying those technologies.
EPA first makes an engineering assessment of technologies that are capable of removing a
contaminant from drinking water. EPA considers contaminant treatability, technology performance,
design considerations, engineering and construction costs, and operation and maintenance costs and
issues. From these data, EPA determines which technologies are the best in terms of having the
highest removal efficiencies that are affordable to large systems. Best available technologies (BATs)
must also be widely available, not limited to a particular geographic region, and compatible with
other water treatment processes.
EPA also determines the total national compliance costs for monitoring and treatment of
contaminated water to meet the various MCL options. The resulting national costs are directly related
to the MCL selected (the more stringent the MCL, the greater number of public systems that would
have to install treatment to achieve compliance).
1.2.3 Final Selection of the MCL
EPA may set the MCL at the MCLG if neither the analytical chemistry nor the treatment and
cost feasibility are limiting factors. If practical removal levels achieved by available technology in the
field are not sufficient to achieve the MCLG, then the higher MCL options will be considered.
EPA's risk assessment and risk management decisions are based on science, but inevitably include
some application of judgement, including the use of safety factors when data are limited or
unavailable. As a result, many have questioned whether costs to water systems should receive more
consideration when EPA is determining the MCL. As discussed above, however, the legislative
history to the SDWA directs EPA to consider only the costs to metropolitan and regional systems
when selecting best available technologies (large systems typically have low per-household costs of
compliance as compared to small systems; see Section 4 below).
1.3 Regulatory Development Process
EPA's drinking water regulations undergo thorough internal agency review before publication
in the Federal Register. The process described below is the process designed for rule development
and issuance. However, the SDWA requirement to publish rules for 83 contaminants within three
years has often required the Agency to truncate the process described below, which has sometimes led
to unintended adverse results. When the Agency did not complete rules for all 83 contaminants on
schedule, a citizens coalition filed a series of lawsuits and the Agency came under a number of court
orders to complete rulemaking on particular contaminants by specified dates. To meet these
deadlines, data collection and analyses have not always been as thorough as desired. Document
drafting and management review has had to occur simultaneously and documents have needed to be
rewritten and rereviewed. Short review periods have resulted in oversights and the need to publish
correction notices. Regulations covering multiple contaminants have often been lengthy and complex.
Thus, the public had difficulty providing thoughtful comments and the Agency had limited resources
22
-------
for gathering and analyzing additional data in response to comments. In some cases, unrealistic
deadlines have contributed to the Agency's difficulty in addressing the unique technical and economic
capacity problems of very small systems.
Within the Agency, a workgroup has primary responsibility for analyzing data and developing
regulatory options. The workgroup also provides for full consultation and coordination on a
rulemaking package among the staff of all affected EPA offices. The Headquarters Offices typically
participating in workgroups that develop drinking water regulations are the lead office (the Office of
Water, or OW), EPA Regional offices, the Office of Policy, Planning and Evaluation (OPPE), the
Office of Research and Development (ORD), the Office of General Counsel (OGC), the Office of
Enforcement (OE), and other concerned media offices (such as the Office of Solid Waste and
Emergency Response and the Office of Prevention, Pesticides and Toxics).
"Option selection" is the process by which senior Agency managers decide the major issues
associated with a particular rule, such as the MCL, Best Available Technology designations, and
monitoring frequencies. Options are first developed by the workgroup, and then presented to
successive levels of management up to the Deputy Administrator and the Administrator. For option
selection, the workgroup prepares detailed briefing documents that explain, among other things, the
results of the health effects assessments, analytical methods, estimates of occurrence in public water
systems, treatment technology evaluations, cost estimates and economic impact analyses. The briefing
documents also summarize any major issues that may have arisen.
OW prepares a rulemaking package for proposal, including draft preamble and regulatory
language, and circulates several drafts of the preamble and regulatory language for comment by the
participating offices. Drafts are first circulated at the staff level to workgroup members, and then
later to senior Agency management, including Assistant Administrators, Regional Administrators, the
Deputy Administrator, and the Administrator. This final Agency review phase, known as "Red '
Border" review, includes time for Assistant Administrators and Regional Administrators to raise any
outstanding issues for resolution to the Assistant Administrator for Water prior to a decision meeting
with the Administrator. Unresolved issues and consensus recommendations are raised to the Deputy
Administrator or Administrator for decision or concurrence.
The final Agency review phase also includes sufficient time for completing revisions to the
rulemaking package in response to Red Border comments and final review of key technical and
economic support documents. Government-wide directives mandate the development of Information
Collection Requests to cover any reporting requirements in the rule. Regulatory Impact Analyses
must be completed which examine the costs and benefits of particularly costly rules (including
analysis by size of public water systems and how alternative options for regulation were considered).
The Agency must also prepare Regulatory Flexibility Analyses (which review impacts on small
systems and how these impacts were minimized). As needed, the rulemaking package is revised
based on the Administrator's decisions. This final phase also includes compilation of the
Administrative Record which includes all the key documents supporting the rule.
EPA consults with the Agency's Science Advisory Board (SAB) during development of the
proposal (the SAB is an independent group of scientists who, by statute, comment on the Agency's
risk assessment supporting the rule and other technical aspects of the proposal, including occurrence
and technology data). To ensure that interagency review, including review by the Office of
Management and Budget (OMB), is focused on the Agency's preferred approach, Executive Branch
23
-------
review generally takes place after Red Border review. OMB/interagency review includes not only the
preamble and regulatory language but also the Information Collection Request, Regulatory Flexibility
Analysis, and Regulatory Impact Analysis. .
Once the Administrator approves and signs the rulemaking package, the proposal is published
in the Federal Register, opening the public comment period. During this period, the Agency may
also hold public hearings on the proposed regulations. The Agency provides a sufficient period of
time, typically 60 to 90 days, for the public to review the proposal and submit comments. Drinking
water regulations typically present complex technical, economic and policy issues that take
considerable time to evaluate. Interested persons must have the opportunity to review the proposed
regulations as well as the extensive record which supports the regulations. Moreover, interested
persons frequently will conduct independent data gathering and analyses as part of their evaluation of
the Agency's proposal.
At the close of the comment period on the proposed rule, the comments are compiled,
organized reviewed and summarized to identify significant issues. The Agency often receives
hundreds of comments. A careful analysis of each comment is essential since it may raise important
issues the Agency must address. The Administrative Procedures Act requires the Agency to respond
to all significant comments received during the comment period.
When EPA proposes a regulation, the Agency seeks to ensure that it has a sound database and
that it has considered all the appropriate factors. Nevertheless, commenters frequently challenge the
adequacy and validity of the data on which the Agency has based the proposed regulation. To
evaluate this type of comment, the Agency must review the data and analyses supplied by the
commenter and may need to gather additional data and perform additional scientific, engineering or
economic analyses. New data on which final decisions will be based must be published in the Federal
Register for comment. Responding to comments submitted in drinking water rulemaking is often an
enormous task because of the variety of pollutants typically covered by the proposal (for example, 38
contaminants in the Phase II rule and 23 contaminants in the Phase V rule), the range of data that
may be available on their health effects, the range of treatment technologies that may be required, the
array of laboratory methods that the rulemaking proposed to approve for measuring contaminants, and
the number and diversity of parties and citizens affected by the rule. Commenters may also question
the Agency's policy decisions and raise arguments that support alternative positions.
The Agency must decide whether to modify the proposed rule in response to public comments
(including data submitted by the commenters), new data developed by EPA after proposal, or revised
analyses performed by EPA. If the Agency decides to change the package following proposal in ways
that commenters could not reasonably anticipate, or if significant new data or analyses become
available, the Agency may re-propose all or parts of a rule in the Federal Register or publish a
supplemental notice of data availability. Any notices issued between publication of the proposal and
promulgation of the final rule undergo internal review following many of the steps outlined above.
The Agency's final rulemaking package must reflect appropriate resolution of comments
received and issues raised since proposal. The workgroup is reconvened to develop final regulatory
options for options selection up through the Assistant Administrator for Water and into another Red
Border review. OW staff complete the final rulemaking package including a detailed document
responding to the public comments and supporting documents (the Information Collection Request,
Regulatory Flexibility Analysis, and Regulatory Impact Analysis). These documents and the preamble
24
-------
and regulation may build on the proposal package, but important portions addressing major issues
often cannot be completed until after senior managers are briefed and make key decisions. Following
Red Border review, the rulemaking package is again submitted to the Office of Management and
Budget for Executive Branch consultation. The Administrator is briefed in a decision meeting and the
appropriate changes made to the package. The final regulation becomes effective and is enforceable
18 months after it is signed by the Administrator and published in the Federal Register.
25
-------
S3S=3=====£2S=3==S33==SS=SS=S===SS=======
CONTAMINANT
— ^^^=^=
Nitrate*
Nitrite
PCBs
Pentachlorophenol
Selenium*
Styrene
Toluene
Toxaphene
Xylenes (total)
Lead and Copper - , , ^ - \ ',«•.
Lead*
Copper
-
Phase V
(di(2-ethylhexyl)adipate
Antimony
Beryllium
Cyanide
1 Dalapon
Dichloromethane
^-
1,1 . 2-Trichloroethane
L . —
Dinoseb
|| 2,3,7,8-TCDD (Dioxin)
| Diquat
=====
MCLG
(mg/1)
HL—~—-^ZSSSSSSS£
10
1.0
zero
zero
0.05
0.1
1
zero
10
,''
zero
1.3
0.4
0.006
0.004
0.2
0.2
zero
0.003
0.007
zero
0.02
======= —
MCL
lmg/1)
============
10
1.0
0.0005
0.001
0.05
0.1
1
0.003
10
S f, detections limit of
1 count/100 ml.
* * Not on list of 83.
* * * Regulation currently not in effect.
-------
CONTAMINANT
Endothall
Endrin
Glyphosate
Hexachlorobenzene *
Hexachlorocyclopentadiene
PAHs (benzo(a)pyrene)
Diethylhexyl phthalate
Picloram
Nickel
Oxamyl (Vydate)
Simazine
Thallium
(1 ,2,4-) Trichlorobenzene
Arsenic (interim)
Arsenic*
Disinfection By-Products {interim)
Total Trihalomethanes
MCLG
(mg/IJ
0.1
0.002
0.7
zero
0.05
zero
zero
0.5
0.1
0.2
0.004
0.0005
0.07
none
MCL
(mg/U
0.1
0.002
0.7
0.001
0.05
0.0002
0.006
0.5
0.1
0.2
0.004
0.002
0.07 .
0.05
POTENTIAL HEALTH EFFECTS
Liver, kidney, gastro-intestinal effects
Liver, kidney, heart effects
Liver, kidney effects
Cancer {Group B2)
Kidney, stomach effects
Cancer (Group B2)
Cancer (Group B2)
Kidney, liver effects
Liver effects
Kidney effects
Body weight and blood effects,
possible carcinogen (Group C)
Kidney, liver, brain, intestine effects
Liver, kidney effects
Dermal, nervous system effects
' •,'-,-.
none
0.10
••!!•!!!!!!!!!!( ' " n •!
Cancer (Group B2)
•••-••••••^^•••i . . .... . ,--
* Indicates original contaminants with interim standards
which have or will be revised.
TT Treatment technique requirement.
+ + Action level = 1.3 mg/L.
37
+ Less than 5% positive or > detections limit of
1 count/100 ml.
** Not on list of 83.
* * * Regulation currently not in effect.
-------
n
N
I-
5
1
K
H
<
I
0
o
o
S
U.
"•
09
1
I
O
'1
1
i
Jt by contami
5)
OS
•si
Ii
.E
i
o
u
E
5
S
1
S
1
Aldicarb**
Cadmium*
S
s
0>
1
s
1
.5S «
a>
r? S •£ -5
ii
I
38
-------
1 -
Annual Cases Avoided
Reduced Exposure
(rounded to 1000s
.H E "
"» 2 "5
§W "O
>*
< 5>
is ^.
03 £
s
Total Popu-
lation Affected
by Treatment
(rounded to
1000s)
•5 S S
|||||
u "w
l-
z
S
H
O
-
w
O
•p
m
2
(0
in
reduced exposure to 1
million people
o
CO
o
oo
m
8
o
o"
o
Ifl
to"
Ifl
I—
0
o
o
CM
Ifl
1
t
*
•a
so
reduced exposure to
87,000 people
o
T—
V
8
o
00
8
V
£
1
Q.
(3
>
m
S
Q.
reduced exposure to
304,000 people
0
CO
8
o
S"
CO
8
CM
(O
8
d
>•
1
0
reduced exposure to 4,
people
V
o
1
o
o
V
IX
o
o
d
n
o
Q
j
u
g
<
1 0 cancer cases
c
V
o
o
o
o
o
o
CO
o
o
CM
in
o
d
Arsenic
•
1 t
O »
Q. £
111
•£75.5
» c 3.
m 3 oa
O O O
-J O £C
T2
<0
1
o
I
o
39
-------
various levels of concern. Once installed, treatment processes often reduce contaminant levels to below
the drinking water standards, thus providing additional benefits beyond those achieved by simply meeting
the standards. Treatment also provides, a-margin of safety against short term "spikes" in contaminant
levels that are unlikely to be identified through normal monitoring. EPA has not attempted to quantity
these benefits.
2.3 Cumulative Costs Of Treatment
Even though the Safe Drinking Water Act does not require the Agency to develop national
cost estimates for its regulations, the Agency calculates these costs in accordance with Executive
Order 12291, which requires Regulatory Impact Analyses for major regulations.
Since 1986, analyses were developed for the following regulations:
Fluoride
Phase I (Volatile Organic Chemicals)
Surface Water Treatment Rule
Total Coliform Rule
Phase II Inorganic, Volatile Organic and Synthetic Organic Chemicals
Lead and Copper Rule
Phase V Inorganic, Volatile Organic and Synthetic Organic Chemicals
Radionuclides (based on proposal)
The public water system costs calculated by EPA include monitoring, installation of treatment
technology, operation and maintenance of the treatment equipment, and disposal of waste generated
by treatment. In calculating treatment costs, the Agency considers what treatment technologies are
likely to be used by various size systems. The Agency does not assume all systems will adopt the
same technology. In making its estimates, the Agency considers varying source water quality, degree
and sophistication of operation and maintenance, and impact of other drinking water and
environmental regulations. (State costs are addressed in Section 5.)
Before proceeding with the discussion of national costs, it is helpful to define the cost terms
used in this section. These terms are as follows:
• Capital costs: Expenditures associated with the purchase and installation of equipment
for water treatment and waste disposal.
• Annualized capital costs: The annual payment on a loan for capital equipment over a
20 year period using a seven percent interest rate.
• Operations and Maintenance costs CO&Nf): The annual expenditures for chemicals,
materials, energy, repairs, and other inputs, including labor costs, associated with the
operation of the treatment, and cost of waste disposal resulting from treatment.
• Total annualized treatment and waste disposal costs: The sum of the annualized
capital cost plus the annual O&M cost.
• Annual monitoring cost: The cost associated with the collection and analysis of
compliance monitoring samples calculated as the annual average over two
Standardized Monitoring Framework cycles (18 years) of initial and repeat monitoring
requirements. An 18 year period was used to provide an accurate, long term annual
40
-------
average cost, since initial and repeat monitoring frequencies will reach a normalized
level during this period. It should be noted that the cost of monitoring in the early
years will likely be higher than average. Also, the potential monitoring cost savings
from State waiver programs is an area of uncertainty at this time.
• Total annualized compliance costs: The sum of amortized capital costs for treatment,
operation and maintenance including waste disposal costs, and annualized monitoring
costs.
Cost data presented in this section are taken from the Regulatory Impact Analyses (RIAs)
supporting final or proposed regulations, or from other EPA Office of Ground Water and Drinking
Water documents such as the 1990 Total Costs and Benefits study and are expressed in 1991 constant
dollars. In addition to public water system costs, RIAs also include costs of implementation of each
rule to the States. Implementation costs include obtaining and reviewing system compliance reports,
taking enforcement actions, providing technical assistance and guidance on implementing the rules,'
reviewing requests for variances and exemptions, and processing requests for monitoring waivers.
The costs of SDWA regulations provided in this report, such as the Phase II rule, assume that
many systems will receive waivers from monitoring requirements. In some cases the State or the
systems had already sampled for the contaminant and could not detect it. In other cases, the
contaminant is not used in the area or the water source is not susceptible to contamination by the
chemical. In these situations, EPA regulations allow States the discretion to issue monitoring waivers
to a system for one or more contaminants. As shown in Exhibit 3.3 in Section 3, depending on the
contaminant, as few as one percent of all systems might actually end up monitoring for a particular
contaminant. Monitoring costs are discussed further in Section 3.
It is often assumed that all public water systems which need to treat for a particular
contaminant will choose from among the equipment or "treatment technique" cited as Best Available
Technology in the regulation. However, under the law and regulations, systems are free to use any
technology which allows them to come into compliance with the regulation. In calculating costs, EPA
uses professional engineering judgment to determine how many systems are likely to use different
technologies or approaches to meet the standard, and the cost of each. Systems will likely use
whatever approach is least expensive which meets the standard. The level of contamination in their
source water and the configuration of their treatment plant will affect a system's approach to
compliance. EPA's national costs reflect the total of each of those component costs. Some persons
outside the Agency who have criticized the Agency for underestimating the costs of drinking water
rules have assumed that all systems needing to come into compliance will use the Best Available
Technology. Because Best Available Technologies are often a more expensive compliance option,
these persons will likely estimate the cost of the regulation to be higher than EPA's estimate.
2.4 Total National Cost of SDWA Regulations
The total national cost of compliance with currently promulgated drinking water regulations
mandated by the SDWA is estimated to be $1.4 billion annually for public water systems, as shown in
Exhibit 2.4. This estimate is derived by summing the individual point estimates of compliance cost at
the relevant MCL as reported in the RIAs and impact summaries for these contaminants. Costs do
not include monitoring for unregulated contaminants as required by some of these rules. Exhibit 2.4
41
-------
shows that the largest capital requirements are associated with the Surface Water Treatment, and Lead
and Copper rules. The replacement of lead service lines serving households that exceed the action
level is a major cost component of the Lead and Copper rule.
By statute, EPA drinking water regulations become effective 18 months after promulgation.
Typically, after the effective date, monitoring begins. Only after an initial or subsequent round of
monitoring has demonstrated clearly that a system is out of compliance does the system decide on
what actions to take to come into compliance. Treatment expenditures are assumed to lag behind
initial monitoring expenditures by approximately two years due to the lead time needed for such
things as follow-up monitoring to confirm the source of the problem, determination of the type of
equipment needed for treatment, and for installation of equipment. In the six months between July
1992 and January 1993, costs for the Phase II and V and Lead and Copper rules all began to be
incurred. Thus, there was a sharp rise in the national cost of compliance with drinking water
regulations in this period. Costs in 1993 are estimated at $250 million and in 1994 at $300 million.
The $1.4 billion average annual compliance cost for current promulgated regulations, which is
expressed in constant 1991 dollars, is expected to begin in 1995.
Exhibit 2.4 describes the magnitude of the cost impact nationally across all public water
supplies. For each of these regulations, a particular water system's actual compliance costs will vary
depending on the results of monitoring and the site specific conditions within each facility. The
following sections attempt'to examine the cumulative treatment and monitoring costs in order to
establish a perspective as to the likely compliance requirements incurred by individual drinking water
systems.
42
-------
I
§
cv
**
S
1
til
•o
:=!'
(0
a M
£! «
a o
go
o -o-
pi
o S
£ W
3 05
Z
\q
vo vo •*
5j si S
W i-! !=<
69 69 69
oq
s
69
oq
?
69
\q
f!
O O »•« O
S 8 2.
69 V!
a g
Si
•* 00 «-;
-: P
ft
eo
1 S
00 ^-t VO
S
o
w
O
O
>
£ O
s en
r*
"3 3
JS
OH
U CA
Q. U
O* O
(J <*>
i{ >
^ O JB 43 § JC
on H o< EL, j CL,
II ill
i)i
.a
a
S a
il
Si
I
JS-
3
n
s
re
.U
3
s.
; <« s- »
rl§ 1
2
'8-
S.
43
-------
SECTION 3
FINANCIAL AND TECHNICAL CAPACITY OF SYSTEMS TO MONITOR
Section 3 deals with the financial and technical capacity of public water systems to monitor
for currently regulated contaminants. The technical aspects of monitoring discussed in this section
deal with the issue of laboratory capacity, and the availability of EPA-certified laboratories to conduct
analysis. Issues regarding capacity of State laboratories are addressed in Section 5.
3.0 Cumulative Costs of Monitoring Requirements
Exhibits 3.1 and 3.2 show the total national cost of SDWA regulations to public water
systems, including the capital, operation and maintenance, and monitoring costs. Costs for
contaminant monitoring under the Safe Drinking Water Act were obtained from data presented in the
Regulatory Impact Analyses (RIAs), Information Collection Requests (ICR), or Economic Impact
Analyses (EIA) developed for each final or proposed rule. Exhibit 3.1 shows that annual monitoring
costs for all currently promulgated rules total approximately $250 million, or about 18 percent of the
total national compliance costs. However, because this number represents an average cost that has
been annualized over 18 years, actual annual costs in some years could be three to five times higher,
particularly in the first several years. Individual water systems must pay for these costs as they are
incurred.
The cost estimates presented in Exhibit 3.1 can be interpreted to represent the national
monitoring costs of complying with the NPDWRs. National monitoring costs are organized by the
population categories of water systems (i.e., large systems, small systems). Therefore it would be
expected that the highest costs occur in the population categories with the greatest number of water
systems. Exhibit 3.1 indicates that small and very small systems, those serving fewer than 3,300
persons, will incur 80 percent of the total national monitoring costs.
Because of economies of scale, EPA expects per household costs to be highest where there are
the fewest number of houses to share system-wide costs. Exhibit 3.2 shows that the average annual
cost per household in each population size category varies significantly. The actual calculation is
based on the median population within each population category. This estimate was calculated as the
simple average over an 18-year period (two complete nine year compliance cycles) of total monitoring
costs in each size category. ,
For most regulations the number of systems was the total of community and non-transient
non-community systems. For the Surface Water Treatment and Total Coliform Regulations, the
number of systems also includes transient non-community systems. The average cost was calculated
in order to illustrate the cost of monitoring over time for each rule and for the total across all
currently regulated contaminants.
The monitoring cost to any system depends on: (1) the frequency of required monitoring
which is typically driven by the level of contamination and sometimes the system size; (2) the unit
cost of laboratory analysis which varies widely for many of the recently regulated contaminants; (3)
the availability of historical monitoring data that may be 'grandfathered' in substitution for the some
monitoring requirements in the 1993-1995 compliance period, (4) the availability of .sample
compositing (the blending of two to five samples before analysis), and (5) the availability of waivers
44
•
-------
which allow reduced monitoring at sampling sites found unlikely to become contaminated by specific
chemicals. The last two conditions, availability of sample compositing and monitoring waivers, are
subject to state discretion.
EPA regulations allow States to grant PWSs a waiver of monitoring requirements if certain
conditions are met. Many of the regulated contaminants are pesticides or products of industrial
processes. If a pesticide is not used, or a particular industry is not located, in the area where it could
contaminate a water supply, the PWS is not likely to find those contaminants. In other situations, the
geologic nature of an area may make it non-vulnerable to a particular contaminant. Systems can
petition and be granted waivers under such circumstances. Waivers are granted contaminant by
contaminant, not rule by rule.
In developing monitoring cost estimates for the regulations, EPA assumed a certain percentage
of systems would be granted waivers and not incur monitoring costs. Exhibit 3.3 lists the
assumptions used in the various cost estimates for the percentage of systems assumed to be granted
monitoring waivers for each contaminants. The cost estimates assume that between 70 percent to 98
percent of monitoring requirements will be waived, depending on the contaminant.
The monitoring cost estimates also assume each water system will composite to the maximum extent
possible among its own sampling points, but that there will be no compositing among separate water
systems. These assumptions translate to an average compositing ratio of 2.5 samples for each
laboratory analysis.
It is unclear how many States will allow compositing or will issue waivers before the end of
1995, when the initial sampling requirements must be completed. Compositing creates additional
compliance tracking overhead for State agencies. Reviewing vulnerability assessments to determine
eligibility for sampling waivers also involves significant State workload that many States are finding
difficult to manage.
If more realistic assumptions about monitoring waivers and sample compositing had been used
in the cost analysis, EPA believes the cost estimates cpuld be substantially higher than those presented
in the exhibits. Due to the time limitations in completing this report and the absence of reasonably
complete, current and well documented data to replace that used in the older RIAs, EPA is unable to
provide updated estimates of aggregate sampling costs for the nation.
There are scenarios for individual water systems in which the annual sampling cost for just
the chemicals regulated under Phases II and V could approach $12,000 to $15,000. For example,
many ground water systems serving less than 150 service connections are likely to conduct full rounds
of initial monitoring for all Volatile Organic Compounds (VOCs) in 1993, 1994, or 1995, because
they were exempt from monitoring VOCs as unregulated contaminants and thus do not have prior
sampling data to grandfather for the initial sampling requirements. These systems will also have to
monitor for some inorganic chemicals (lOCs) and some synthetic organic compounds (SOCs). This
would cost $2,000 to $5,000 for each sampling point in the water system, depending mainly on how
many and which SOCs are waived by the state.
For example, Wisconsin has invested almost $450,000 in designing a sampling waiver
program for organic compounds. The State expects the program to save its water systems about
$10.5 million in reduced sampling costs. Approximately 70 percent of Wisconsin's sampling points
will qualify for waivers from pesticide (SOCs) sampling and about 30 percent will qualify for waivers
45
-------
involving solvents (VOCs) like trichloroethylene. The EPA national cost estimates shown assume the
percentages of sampling points qualifying for waivers to be 80 percent for SOCs and 70 percent for
VOCs.
Some States have voiced strong disagreement with the EPA national estimates, and EPA now
agrees that some of its assumptions about the extent of sample compositing and the availability of
sampling waivers may have been optimistic - resulting in low estimates of total sampling costs. In
order to reconcile the perceived disparity between earlier EPA estimates of sampling costs and the
actual costs, the Association of State Drinking Water Administrators (ASDWA) is conducting a State
by State survey to acquire the best current estimates of monitoring costs and laboratory capacity to
handle the increase in monitoring generated by the new requirements. 'The results of this endeavor
are expected to be available in the autumn of this year. If no waivers or compositing were allowed,
monitoring costs could be as high as $4000 per treatment site during initial sampling and $1200 per
year per treatment site if a number of contaminants are detected above levels of concern.
46
-------
a
o
S
C
R o
1 °
O)
(0
3
C
C
"5
c
o
'i
s.
&
i
S
a
CO
+^t
I
CM
o ^
9
K
CO *
u
i
7
K-50K
CO*
egul
(O
C4 ••— T- r-- t-. CO C
O O CM (O O) CM CM
5P O) O>
co m
CVI
0
*~ CM CM
eo
O CM O CO i- in CM
O CO i-
R8S5
o in co in CD CD
SCO CO i- i- O
o> in in in ^
to
^ to' CM CM"
co to co CM at in in
V O CM CD O T*
co o> in v? *f
*~ S eo co
COO »^
ffi j- O) O K CO
CJ> ^- ^ *" ^ CO
«- CM eo i
^ in^ cc^ *9£ o^ co CA
co CD eo e> co »-
S^ *~ O •-
G Q CO 0>
^ _ o * ^TO
i S co CJ o a> °8
a> ° co
•£
S
•a
i
•i
S
S
|
i
m
I
a
£
%
a.
£
TB
S
eo _
I
a
fr
2
!
o
a
8 .
^
c *
ll
<0 ±f |Q
«
"i
a
i ji i
ml i
5"S i
CM CO
47
-------
<;
I
to
I
1
u
g
§
i
Ed
O
co*
in
8
in
CD*
S
evi
SSS58SS ^
O C3 O C) C3 O O
d o d o o o o
O O CM CO i- (O O
p o cq o p p p
d d d o o o o
o CM *r in T- co o
O O V i- O O O
d d d d d d o
§CM eo to CM en o
p
-------
EXHIBIT 3.3
ASSUMPTIONS USED IN ESTIMATING MONITORING COSTS
SDWA
Contaminant
Percentage of Systems
which obtain a Wavier
Based on1
Acceptable
Recent Data
Base
Monitoring
Percentage of
Systems which
Do not obtain
a Waiver2
Phase II:
Asbestos
Nitrate
Nitrite
Other lOCs
SOCs
VOCs
Phase V:
lOCs
Adipates,
Phthalates
Diquat,Endothall
and Glyphosate3
Dioxin
PAHs
Other SOCs and
VOCs4
Lead and Copper:
Source Water
Corrosion
Microbiologicals5
SWTR
Coliforms
0%
5%
5%
99%
0%
70%
0%
80%
75%
80%
80%
100%
27%
0%
N/A
N/A
99%
90%
94%
1%
80%
8%
98%
19%
24%
19%
19%
0%
73%
80%
N/A
N/A
1%
5%
1%
<1%
20%
22%
2%
1%
1%
1%
1%
0%
<1%
20%
N/A
N/A
1.
2.
3.
4.
5.
These systems are permitted to conduct minimum repeat monitoring.
These systems must conduct increased repeat monitoring.
Because these contaminants occur in surface water, percentages apply to those systems.
Monitoring results for these contaminants will be generated simultaneously with results for
Phase II contaminants.
Monitoring waivers are not available. All systems must continue to monitor though recent
data are generally available.
49
-------
3.1 Laboratory Capacity .
In 1992; drinking water systems were monitoring for trihalomethanes, arsenic, VOCs,
coliforms, lead and copper. Under the standardized monitoring framework, the monitoring for Phase
II and Phase V contaminants became effective beginning January 1993.
? 'All the: states, except Wyoming, have assumed "primacy" or primary enforcement
responsibility for enforcing drinking water regulations. To obtain and maintain "primacy", a state
must comply with EPA regulations which require the "establishment and maintenance of a state
program for; the certification..of.laboratories conducting analytical measurements of drinking water
contaminants," or analyze all drinking water samples in a State laboratory certified by EPA.
The laboratory capacity issues for a state where all compliance samples are analyzed in a state
laboratory are somewhat different from the laboratory capacity issues where the state has alaboratory
certification program and'the compliance .samples are mostly analyzed by local or private laboratories.
The issue of .laboratory capacity, in reality, is an issue of resources available to the state, either for:
conducting the .analyses in the state laboratory or for having a program for certifying laboratories to
conduct the analyses. At its discretion, the State can employ technicians and purchase equipment ^
necessary'for conducting analyses. Alternately, if the state does not want to conduct the analyses in a
state laboratory, the state needs to hire the necessary personnel to manage the program for certifying
commercial laboratories to conduct analyses for drinking water samples.
v t '. ' -. i - " ' '
Gradual Increase iii the Laboratory Capacity for Chemical Analysis
To be certified for chemical contaminants, the laboratories must successfully analyze the EPA
Performance Evaluation (PE) samples or equivalent samples provided by the state and successfully
pass an on:site audit every three years. In some states the audit interval is shorter. For states with a
commercial laboratory certification program, the number of laboratories participating in PE studies
provides a good indicator of the number of laboratories interested in being certified for analyses of
drinking water samples. EPA sends PE samples to laboratories twice every year. Not all laboratories
participate in every PE study, but the following table provides an approximate number of laboratories
participating in the chemistry PE program:
50
-------
Year
Laboratories Participating
in Chemistry PET Program
1987
1988
1989
1990
1991
1992
1100
1500
1700
2200
2300
2500
It can be seen that there has been a gradual increase in the number of laboratories
participating in the PE sample program as the demand for drinking water analysis has increased. It
must be noted that states like New York do not participate in EPA's PE program. New York siate
has its own PE program.
The PE data also indicate that more and more laboratories are analyzing for more chemical
contaminants as the demand in the drinking water program has increased. Not only the number of
laboratories participating has increased but also the percentage of laboratories analyzing within
acceptable limits has increased over a period of time.
The total laboratory capacity appears to have kept pace with the increasing demand for
laboratory analyses. However, some laboratories and state agencies have expressed concern about
what they see as the relatively short time they have to develop competence in a given method
Nonetheless, the total number of laboratories certified for analysis of drinking water for all types of
drinking water contaminants has shown a steady increase-from 1990 to 1992, years for which data are
currently available.
Year
1990
1991
1992
Total Number nf
Certified Laboratories
3255
3269
3454
Laboratory Capacity for Inorganic Contaminants
The laboratory capacity to conduct analysis of inorganic contaminants in drinking water seems
to have kept pace with the increasing demand for laboratory analyses. The total number of
laboratories certified for analysis of inorganic contaminants in drinking water has shown a steady
increase from 1990 to 1992, years for which data are currently available
Year
1990
1991
1992
Number of Laboratories Certified
For Inorganic
1079
1199
1447
51
-------
laboratory fnpncitv for Microbiological Contaminants
The laboratory capacity to conduct analysis for microbiological contaminants in drinking
waterha?3ioSS»Swime increasing demand for laboratory analyses..The total number of
Sratorif ceSfor analysis of microbiological contaminants in drinking water has shown a
steady increase from 1990 to 1992, years for which data are currently available.
Year
1990
1991
1992
Number of T-ahnratories Certified
Fnr Microhiftlngical Analysis
2426
2430
2522
The laboratory certification process for microbiology is different from tint for chemistry.
Currently EPA provides PE samples for microbiological contaminants only to me Principal State
s PE samples are at present not available to the local commercial and utility laboratories
t states certify microbiology labs without the use of PE samples. EPA ,s currently
PE samples to a"laboratones who request
them.
EPA is also developing monitoring requirements for two pathogenic protozoa (Giardia and
Crvptosporidium) and a group of viruses (enteroviruses). This requirement may go into effect as
eariy SftSyWW as an outgrowth of the disinfection byproduct negotiated rule. PE Samples and
Sfi^ «Sa «ai ^ed to be developed for these three pathogens, and the full impact of having
to develop these methods is yet to be seen.
Laboratory certification criteria have also been developed for heterotrophic bacteria, Although
PE samples have riot been developed. Heterotrophic bacteria are not a regulated contaminant and do
not have to be monitored. The heterotrophic bacteria population ,s quite diverse, and systemsmay
have a different heterotrophic bacterial population than one developed as a PE sample Thus a PE
sample is less meaningful for these bacteria and can be deferred, at least for the time being.
laboratory C"r";*Y fnf Tntal THMs and VOCs Analvsis
The VOC regulations were published on July 8, 1987 and the monitoring for VOCs became
effective, beginning with large systems, as of January 1, 1988. The states were given the authority to
give conditional approval to me laboratories who were certified for THMs and had successfully
analvzed for VOC PE samples within acceptable limits without any additional requirements. Even
tiiough the monitoring requirement for VOCs did not begin until January 1988 the Agency provided
PE samples to states and commercial laboratories at least two years earlier so that the laboratories had
the option of being certified as soon as the regulations were promulgated Mid much before the
monitoring became effective. Thus the Agency was assured of having sufficient laboratories to
conduct the analyses for VOCs when the monitoring requirements became effective.
52
-------
Since 1990, the Agency has compiled a list of laboratories certified by the states to perform
analysis of drinking water samples. This database indicates that the number of laboratories certified
for THMs and VOCs is as follows:
1990
1991
1992
No. of Laboratories
Certified for THMs
584
589
752
No. of Laboratories
Certified for VQCs
443
514
641
These data indicate that, as the demand for drinking water analyses has increased, the number
of laboratories being certified to do analysis of TTHMs and VOCs in drinking water has also
increased. Thus, the laboratory capacity has kept pace with the increased demand for meeting
monitoring requirement for VOCs.
Laboratory Capacity for Phase II and Phase V Contaminant Analysis
For Phase II and Phase V contaminants, the monitoring requirements became effective
January 1993. Although EPA has seen an increase in the number of states which have increased their
capacity to conduct drinking water analysis, no data are currently available to indicate whether this
increase is sufficient to meet the expected increase in demand for Phase II and Phase V analysis.
Similarly, while there has been an increase in the number of states which have developed a program
for certifying commercial laboratories, there are no data available to indicate if there are a sufficient
number of certified laboratories to conduct Phase II and Phase V analysis. What can be seen is that
there has been a gradual increase in the number of commercial laboratories that are now certified to
analyze for Phase II and Phase V contaminants.
Number of Laboratories
Certified for SOC Analvsis
494
540
680
1990
1991
1992
Although no definite data are currently available, there is a general feeling among the
certification officers that there has been a gradual improvement in the performance of laboratories
participating in the program.
Laboratory Capacity for Asbestos
The regulations for asbestos are part of the Phase II regulations published January 30, 1991.
Monitoring for asbestos began in January 1993. The Agency currently does not have a viable
asbestos PE sample program. Until the PE samples become available, EPA Regions and the states
53
-------
are permitted to grant provisional certification to laboratories for analysis of asbestos in drinking
water EPA estimates there are less than 50 laboratories in the country which have been certified to
perform analysis of asbestos in drinking water. The number is expected to increase during the next
few months Because many systems are likely to receive a waiver from the asbestos monitoring
requirements and the laboratory demand will be somewhat limited, a number of states are forming
syndicates to have one laboratory serve several states.
^Laboratory Capacity for Dioxin
The regulations for dioxin are part of the Phase V regulations published on July 17, 1992.
Monitoring for dioxin started in January 1993. The MCL for dioxin is 3 X 10"8 mg/L (30 ppq) and
analysis may cost over one thousand dollars per sample. The Agency currently does not have a viable
dioxin PE sample program. Until the PE samples become available, EPA Regions and the states are
permitted to grant provisional certification to laboratories for analysis of dioxin in drinking water.
EPA estimates there are less than 25 laboratories in the country which have been certified to perform
analysis of dioxin in drinking water. The number is expected to increase in the near future. Because
many systems are likely to receive a waiver from the dioxin monitoring requirements and the
laboratory demand will be somewhat limited, a number of states are forming syndicates to have one
laboratory serve several states.
Laboratory Capacity for Radionuclides
Currently, many states and EPA regional laboratories do not have the capability to analyze for
all radionuclide contaminants. The Agency currently does not offer a radionuclide laboratory auditor
course similar to one^orfered for chemistry and microbiology. EPA may not be able to,meet the
increased demand for radionuclide PE samples necessary for certification of laboratories. Options are
currently being developed to offer a radionuclides lab certification officers/auditors course and for
increasing the laboratory capacity of the EPA regional, state and commercial laboratories to meet the
increased demand for radiological analysis.
The laboratory capacity to conduct analysis for radiological contaminants in drinking water is
somewhat limited. There are only about one hundred laboratories in the country which are capable of
analyzing for radiological contaminants in drinking water.
Year
1990
1991
1992
Number of Laboratories Certified
For Radiological Analysis
81
114
104
There are concerns that there may not be sufficient laboratories for radiological analysis of
drinking water. Additionally the states may not have sufficient trained staff either to conduct
radiological analysis of water or to certify local commercial laboratories for radiological
contaminants. This issue of laboratory capacity for radionuclides regulation is currently being
studied. The radionuclide regulations are expected to require monitoring beginning in January 1996.
54
-------
Additional Concerns Regarding Laboratory Certification
The current drinking water laboratory certification program determines how well the
laboratories are able to perform under the best operating conditions and when the laboratories know
they are being audited or checked to see if the laboratories are indeed performing well. Insufficient
resources currently exist to be able to determine how well the laboratories perform day to day on
routine drinking water samples.
Options for Improving Capacity
Monitoring is one of the most important ways PWSs can undertake to ensure that their water
systems remain protective of public health. However, the current monitoring requirements do pose
some managerial, technical and financial constraints. Water systems are finding it difficult to keep
pace with the increasing monitoring workload, especially because of the frequency and timing for
monitoring which may vary with different contaminants, and because a number of states are unable to
handle the volume of waiver applications. The standardized monitoring framework coordinated the
frequency and timing for a vast array of drinking water contaminants and thus reduced this difficulty
to some extent. The monitoring requirements also pose significant financial burdens on water
systems, specially those with poor quality of source water and those that do not possess the economies
of scale, in terms of their customer base, to easily absorb the increased cost of monitoring.
EPA has just completed an intensive review of Phase I, II, and V monitoring requirements in
a work group which involved State officials as well as EPA headquarters and regional personnel. The
work group recommendations include a significant modification to the standardized monitoring
framework. The work group proposes that Federal regulations require only one sample per site every
three years. State regulations would then target more frequent monitoring in areas where
contamination is likely to occur. The work group recommendations are currently under review.
Another more long term possibility, is tiered monitoring under which PWSs would be
required to monitor for a certain fixed number of indicator or surrogate contaminants. By reviewing
data collected by EPA, U.S. Geological Survey, the states and other organizations and supplementing
it with additional information on their locale (geography, geology, land use, proximity to
manufacturing facilities and waste sites, and so on), it may be possible to say a lot more about the
quality of water in these supplies prior to the collection of samples. It may be possible to determine
only a few contaminants which are usually of some concern at most of the supplies. It may be
possible to explain the occurrence of these contaminants in almost all cases.
Based on this information it is likely that a tiered monitoring strategy can be developed which
could reduce costs significantly. All localities would monitor for certain contaminants and, depending
on the results of monitoring, some systems would monitor for additional contaminants. EPA may be
able to determine contaminant indicators or surrogates whose presence or absence would determine
the need and type of additional monitoring. This approach could reduce the monitoring cost to a
great extent and also protect the public health at the same time. This option is still under
55
-------
development. Data developed under the occurrence projects related to standard setting would prove
useful here as well.
Another option is to develop quick, low cost analytical methods which can screen for the
presence of contaminants. These methods would generate information that would enable the analysts
to determine whether further analytes should be investigated and/or whether more sophisticated
analytical methods should be employed. These methods could be used for compliance monitoring, or
for granting waivers or variances.
56
-------
SECTION 4
CAPACITY OF SYSTEMS TO TREAT
This section discusses the technical and financial capacity of systems to install and operate
treatment for compliance with SDWA requirements. In addition, this section will discuss options
available to public water systems to increase both financial and technical capabilities, with a special
emphasis on small water system issues.
4.0 Cumulative Cost of SDWA Compliance Requirements
As mentioned in Section 2, the total costs to PWSs of complying with current drinking water
regulations is estimated to be 1.4 billion dollars annually. These cost include monitoring, treatment,
operation and maintenance, and reporting expenses.
A first step in evaluating a water system's financial capacity to meet the treatment
requirements of the SDWA is to define the possible set of regulations or treatments that a given
system is likely to confront. This section of the report describes the cumulative regulatory burden by
estimating the number of systems having to install multiple treatments. The treatments which are
discussed in this section are those which EPA estimates are the results of drinking water regulations.
EPA assumes that some treatment is already in place, at least for most surface water systems, because
of the historic role of public water systems to protect against microbiological contamination. Chlorine
has been the disinfectant of choice for most systems.
The analysis in this section is an update to a previous analysis contained in the 1991 Total
Costs and Benefits Report. The 1991 report indicated that estimates of the number of systems having
to install more than one type of treatment should be derived from data on the co-occurrence of
various contaminants in drinking water systems. However, only two data sources contain data
suitable for making this determination: the National Inorganics and Radionuclides Survey and the
Ground Water Supply Survey.
EPA is presently engaged in an effort to develop a methodology for estimating co-occurrence
of certain organics and radionuclides. The net effect of higher co-occurrence would be to decrease
the number of systems impacted and reduce the costs associated with compliance where treatment was
already in place. Unfortunately, the results of the co-occurrence analysis are not yet available for
inclusion in this report. Therefore, an alternative approach to estimating the number of systems
required to install multiple treatment was employed.
The approach used to examine cumulative treatment requirements is based on the joint
probability that a water system will have to install more than one type of technology to comply with
the various rules examined. The individual RIAs contain estimates of the number of systems affected
and the type of treatment required based on the contaminant occurrence analyses and the treatment
decision matrixes. These estimates from each RIA were entered into a three-dimensional spreadsheet
model (treatment technology by system size by rule) in order to normalize and manipulate the data.
After the data for each rule were entered into the model, they were adjusted to eliminate possible
double-counting of systems that theoretically will choose the same treatment for more than one rule.
The adjusted numbers of systems ware then converted into percentages of systems installing a
particular treatment by dividing the number of systems installing this treatment by the total number of
57
-------
systems in the size and source category. This adjustment and conversion yields the probability of a
given system installing a particular treatment. The adjusted probabilities were then entered into a
statistical program to calculate the probability of multiple treatments being installed by a given water
system. This analysis assumes that co-occurrence is random, that is that the probability of occurrence
of each contaminant is independent of the probability of occurrence of every other contaminant.
Exhibits 4.1 and 4.2 summarize the results of this probability analysis for ground water and
surface water systems, respectively. The analysis considers every possible treatment combination for
ground water and surface water community and non-transient, non-community systems; however,
only those with signSficant'probabilities are displayed. According to this analysis 40 percent of
ground water systems are expected to need no treatment, 43 percent to 'need 1 treatment, and 26
percent to need 2 treatments. The most common treatments (and combinations of treatments) for
ground water systems involve corrosion control, packed tower aeration, and disinfection. For surface
water systems, less than one-half of systems are estimated to require multiple treatment. The
common treatment requirements involve the installation of some form of filtration (i.e.,
coagulation/filtration, direct filtration), modifying a coagulation/filtration unit, or corrosion control.
After defining the set of treatment scenarios a given system may confront, unit treatment and
waste disposal cost estimates were developed. These unit costs were estimated by taking either a
representative cost for a given technology covering several contaminants (e.g., for packed tower
aeration, or PTA, costs for radon removal are assumed to be the same as for VOC removal)) or a
weighted average of a category of treatment technology (e.g., coagulation/ filtration and direct
filtration). The number of systems installing a particular treatment (or combination of treatments) for
each of the size categories was multiplied by the expected cost of installing the treatment. Exhibits
4.3 and 4.4 show the annual household costs for each treatment combination for ground water and
surface water systems, respectively. Annual household costs were calculated from unit production
costs assuming flows of 100,000 gallons per household per year. All unit costs were updated to 1991
dollars. Waste disposal costs for specific treatments were added if appropriate and were calculated as
a weighted average across the disposal technologies for a given treatment technology.
This entire analysis is based upon currently promulgated rules. The costs associated with
rules currently under development are yet to be estimated. Many groundwater systems may require
an inorganic removal technology such as reverse osmosis or ion exchange for sulfate or arsenic. The
Groundwater Disinfection Rule and the Disinfection/Disinfection Byproduct Rule may also have a cost
impact on a large number of systems. In some cases, ftiture rules may cause systems to make
further, unanticipated investments in equipment/facilities in order to comply with both the existing
(old) rules as well as the new requirements. These potential costs are not reflected in this analysis.
In summary, systems are estimated to incur additional treatment costs across a wide range of
treatment scenarios and configurations. For many systems, these increased treatment costs may be
small in relation to existing production costs. For a smaller proportion of systems these increased
costs could have significant implications. Annual household costs for the smallest system size
categories are much greater than for the larger system size categories. These analyses attempted to
describe the variability in estimated impacts depending on system-specific contaminant occurrences
and physical and operational conditions. Finally, based on the methodology used to examine multiple
compliance requirements, it appears that the majority of water supplies will require two or fewer
treatments to comply with currently promulgated regulations.
58
-------
I
I I
£
i
w
•s
-8
u
i
OOOOOOOjfOOOOO
CM O O O O
U>
5
n o n n o
^
.45
td § c
"iS
o o o o o
o o o oo
oo oo o
o o o o o
cna n o o
I .jp CO
IJs
M CH CM CM CM CM
ilil
59
g
g
g
1
I
£
15
£
ii
IS
o
•6
§
§
te
g'
5* r
i I J:
-------
3 I
i
a
en
-------
I
to
I
i
co
I
5
• \3
**o
.- % ' O
•: -1
«*«!
sssssss:
ii$SSS§«!S
• CJ Ol CM CM 01 CM TO
I
I
y 5 r* «s \
**'*
5 8
i
e
i i
#s
& ^
w I
«>
1
1
*•* -r-
I I II
_f _jp *T «rf
£
*
i « I
i is
•3
$
*
I i
« 6
I 1
61
-------
I
1 I
fQ jp|
cJ ff
I
vt
3
i
I
55
I" i
i
5 5 «»
<»*» «»«»*»«» 4»«»«»«»«
l-SsCT
•SUOI
62
,1
S-
ft
ft
LI
t.
i
t if
% fw-
3 -S-3
1
Tat*
£§&
«3
2'
l! *•,
•jl S""
I -1-
s °^"o
^ 111
I |||
i I* §
5 1*1
•h. ^» •— =T
s I?*
8 ||S,
I 111
^ !i!
.5 .
-------
4.1 Financial Capacity to Afford SDWA Treatment and Other Compliance Costs
The ability of water systems to pay for SDWA compliance costs will ultimately be based on
the ability and willingness of customers to pay the increased water rates needed to finance monitoring
and treatment.costs. In most cases, households pay the vast majority of a water system's costs.
In this analysis costs to a public water system are based upon the treatment cost per thousand
gallons using a specific average flow for each population size category. Treatment cost takes into
account borrowing needed capital for a 20 year term at seven percent interest and annual O&M costs.
Household costs are then derived from the system costs by assuming that the average household uses
100,000 gallons of water per year. The analysis makes no assumptions about cross subsidization
between household and commercial/industrial users nor does the analysis assume any explicit amount
of commercial/industrial flow. All system users are assumed to pay for the amount of water they use
at the system's unit production cost. In reality, commercial/industrial flow can be a significant factor,
especially in larger systems.
While it is important to consider the affordability of drinking water standards on the public,
affordability is a subjective concept. The perception of what is affordable can vary by income and by
the perception of the problem, among other things. As a result, what is considered affordable to one
community or individual may not seem affordable to another. In the following analysis, EPA takes
the information previously presented on the costs to water systems of promulgated drinking water
rules and estimates those costs on a per household basis and then compares those with other
household costs.
Another important consideration in assessing the impact of drinking water regulations is the
costs of the rules and the benefits derived from the rules. A cost-benefit analysis is less subjective
than an analysis of affordability, however, it also has limitations. One of the chief limitations is that
it can be difficult in some instances to quantify some of the benefits of a rulemaking. It is important
to note that these two types of analyses can yield different conclusions. What is perceived to be
affordable may not be cost-beneficial and what is cost-beneficial may not be perceived to be
affordable. In section 2 of this report, EPA has presented the aggregate costs and benefits of each
rule, but has not explicitly conducted the comparison of them.
In addition to the analysis presented in this report, EPA is undertaking three other analyses
through which the Agency believes it will gain a significantly improved understanding of the financial
impact of SDWA regulations on water systems, communities, and households. First, the Agency is
conducting an affordability analysis of SDWA regulations for water suppliers, communities, and
households. This analysis will use existing data and will assess a number of financial indicators to
evaluate the affordability of SDWA regulations to these groups. Second, the Agency is preparing a
Community Financial Profile. The profile will consist of a statistically significant assessment of
community level financial data derived from the U.S. census. Its objective is to provide a baseline of
financial conditions in communities against which EPA can better measure the impacts of SDWA
regulations. Finally, EPA will update the Survey of Community Water Systems, last updated in
1986. The survey assesses the technical, financial, and operating conditions of water systems.
63
-------
In this report, EPA estimates household baseline and incremental costs due to promulgated
SDWA regulations. These costs are then combined and affordability at the household level is
assessed based on the total water bill.
4.1.2 Impact on Household Water Expenditures
The SDWA impact on households is examined in two ways. The first expresses annual water
costs per household both in dollars and as a percent of household income. The analysis shows
household costs both with and without SDWA costs, at both the national level and the system size
category level.
The second analysis compares the per household costs of providing water service to the costs
of other typical utilities, including electricity, natural gas, telephone and cable television. The
household cost analyses were developed using two simulation models. One model combined the
national distribution of household incomes with a distribution of SDWA and water service costs per
household. The other model combined the same national income distribution with distributions of
SDWA and water service costs for each of twelve system size categories. The purpose of both
simulation models is to quantify total household water bills (including SDWA implementation costs)
as a percentage of household income, where income is included in the analysis as the full range of
U.S. household incomes, rather than based on one measure of central tendency, such as median
annual income.
Using available data on existing household water bills and household income, and the
previously calculated increase in household water bills from SDWA regulations, EPA calculated the
percentage of household income that the total water bill would represent following implementation of
these regulations.
Exhibit 4.5 displays the change in average household costs for drinking water, both nationally
and for each system size category. The annual cost of SDWA compliance for the average household
is $14, ranging from $145 in size category 1 to $3 in size category 12. Exhibit 4.5 shows that
compliance costs per household increase most dramatically for households served by systems in the
smallest size categories. This result again confirms the popular belief that the SDWA burden is
greater for smaller system customers.
Exhibits 4-6a and 4-6b emphasize the impact of the SDWA on smaller systems. Exhibits 4-6a
and 4-6b show the distribution of incremental household costs in the four smallest system size
categories. In Exhibit 4-6a, the four smallest categories are grouped together. In Exhibit 4-6b, each
system size is displayed separately, and ground water and surface water systems are shown on
separate tables. Exhibit 4-6a shows that approximately 40 percent of systems will require no
treatment to comply with existing rules. Another 48 percent (up to the 85-90th percentile) will
primarily need corrosion control, which will increase household costs by less than $200 per year.
The costs move upward sharply for the remaining 12 percent of the systems, which are comprised
mainly of the smallest systems (and some others that need to install multiple treatment).
On Exhibit 4-6b, the 25th and 75th percentiles have been included with the mean to provide
more information about the distribution of cost impacts within each size category. The data have
been analyzed in three ways to assess the impact of treatment on all systems, on those systems
64
-------
installing one or more treatment, and those systems installing one or more treatment excluding
corrosion control treatment. The corrosion control costs have been removed from one analysis
because it is much less expensive than the other treatments and installing corrosion control exclusively
is the predominant treatment scenario for many systems in these size categories. Therefore, the
corrosion control costs can mask the costs that some small systems will be paying when they are
included in the other two cost distributions.
Exhibit 4-6b shows the distribution of annual incremental household costs for the SDWA
Treatment Scenarios discussed in Exhibit 4-1 for smaller groundwater systems and surface water
systems. The effect of the corrosion control costs on the distribution can be seen in Exhibit 4-6b
(ground water table) where the mean is higher than the 75th percentiles for the two analyses where
the corrosion control cost have been included. The higher cost treatments are required in less than 25
percent of the groundwater systems in the first four size categories. In the surface water table, the
75th percentiles for all systems and those systems installing one or more treatment are higher than the
75th percentile for systems installing one or more treatment excluding corrosion control. The
difference between the 75th percentiles for the three analyses for the surface water systems is that the
costs for corrosion control are removed where corrosion control is excluded from the analysis. For
example, the difference between the 75 percentiles in size category 1 (1100 - 999) is attributable
primarily to the cost for corrosion control treatment (97).
Exhibit 4.7 shows a historical comparison of the percent of household income spent on a
variety of utilities, including drinking water. In this analysis water and sewer costs are combined and
do not include projected costs of future SDWA regulations. This chart indicates that water and sewer
costs account for the lowest proportion of household utility expenditures, after cable television.
Electricity, natural gas and telephone service all cost more than water, with electricity the highest, at
nearly four times the combined cost of water and sewer service.
65
-------
EXHIBIT 4.5
Change in Average Household Costs
For Drinking Water
(DOLLARS/YEAR)
• vSysterri
1
2
3
4
5
6
7
8
9
10
11
12
Weighted
Average2
SDWA
Baseline Incremental Total Projected Percent
Costs1 Costs Costs Increase
$264 $145 $409
314 53 367
198 30 228
256 20 276
282 22 304
201 13 214
192 9 201
186 11 197
157 10 167
176 12 188
169 4 173
142 3 145
' • ' •' '•' '"'''•' • ' • "' ;' '<: •' '•••.' '•':
$190 $14 $204
55%
17
15
8
8
6
5
6
6
7
2
2
•"' •;;•• ..• .'. .:-.'-;'V:,
-;• 7% .--.:•
aFor systems serving less than 10,000, costs were derived from the 1986
Survey of Community Water Systems and updated according to the CPI;
•^•^•^•^•^•^•^•^•^•••^•^•^•^•••l
for systems serving over 10,000, costs were derived from the
Water Industry Data Base
2Weighted according to population served
._. ___^^.— ,„ ^^^M,^^,M1^p^pjm»ai^pjai
.^.^.^.^.^.^.••.^.•.^^.•.••^.^.^.^.^.•.••.•.•.••.^.•••.^.^.^.^.^.^••.^•^••••^••i
66
-------
Q)
•»->
(0
0)
So
•*= «
2 ro
« S 5
<0 O JS
"
CO 0)
12
a)
§ (O
< OT
S E
o «
C W
o >•
•S w
3
CO
O
o
0
CO
o
o
*•
•^—
o
o
01
«W^">
o
o
o
o
o
00
j
o
o
CD
i
0
o
j:
CO Q-
>• e
» I
•S k
M» **
CO -a
S I
<1> ^
O. j
CO .o
5 I
i i
3 K
01
o
.5
1
i
£
•8
i
s
J8d P|OLJ9SnOH/$
67
-------
i
I?
t2
1s
55
a
2
"o
•S
o
"ra
a
3.
§ '
'"Ȥ
I
Is
"X
.s
(III
8 '
•Is
1
oo
cs
oo
CM
**^
1
Q
Z
I'
O
z
S
•o
1
= 8
I
s
l£!
aM
gill-
8
Pi
S
,-§
i«
**
«*
II
8
|
S
I
Si
<4
HI
li
1
8
cs
S
81
S
5
VO
Jn
^^ S
V) ^0 IE ^H
K %K II —
S
8
^ •—
a§
-S w
«»£
is
M
1
I
§
£
Jf
m
ft
cs
2
?!
S
§
e»>
B
S
I
s!
11
"3 SK
O
"8
.S
a
o
1
$
S
.2
f.S
1^1
»J I
8 § £
III
I S-8
Si^
JTi
< H
S1
'•3 »
Bi
£ €
1
!.
IIP
i
I..
j
6f!
-------
0)
Q)
ts
_o
o
(Of
O
., "•
K«J
^^^
T3
x
•2
I
i
•a
2
o
o
es
U) i- tO t*.
o oj o t-:
£ §? *. *
8 £ 3 g
o oi o T-:
i-: ea
S *• 3 *
in p o> in
o e\i o" T-"
S 52 S 18
c> c\i o" i-"
OJ in n £
iq o p
-------
4.2 Existing Financing Mechanisms
For purposes of discussing financing in this section, water systems are grouped as follows:
• Large systems (those serving more than 10,000 people)
• Small to Medium (those serving 1,000 to 10,000 people)
• Very Small to Small (those serving less than 1,000 people)
These size division do not match the standard size categories shown in the Overview of this
report, but are logical breaks for discussing the financial characteristics of water systems.
The financing mechanisms available to a water system are a function of its size, ownership
and management discipline. The 58,000 Community Water Systems are highly diverse, reflecting a
broad range of institutional structures. The next section discusses the structure of the industry,
summarizes the main financing mechanisms that exist, and lists the financing issues unique to each
industry segment.
4.2.1 Large Water Systems
For purposes of discussing financing mechanisms in this report, large water systems are
defined as those serving populations of more than 10,000. There are approximately 3,300 large
systems, representing about 80 percent of the total population served by Community Water Systems.
They account for nearly half of the capital demands imposed by SDWA regulations and over 52
percent of total industry revenues. Large systems can be categorized into three groups: (1) investor
owned utilities; (2) municipal water systems (owned by a municipal government); and (3) public
water districts or authorities (special purpose governmental entities which may transcend municipal
boundaries). Each group has unique characteristics, relies on different sources of financing, and has
different infrastructure requirements.
Investor Owned Utilities
Large investor owned water utilities typically finance capital improvements by a mix of 50-60
percent equity and 40-50 percent debt. Equity financing comes either from retained earnings or
through public stock offerings. Debt financing is in the form of bond issues. In general, the large
investor owned utilities have little problem in obtaining access to capital for financing infrastructure
improvements.
The larger investor owned utilities are regulated by state public utility commissions (PUCs)
and resemble their counterparts in other public utility sectors, including electric and gas utilities.
State regulatory commissions vary considerably in policies affecting different sized water suppliers.
Generally, State regulatory commissions audit and approve rates charged for water service, including
all costs of providing service, such as SDWA compliance and infrastructure maintenance. PUCs also
set target ranges for the mix of equity and debt financing and monitor the costs of capital, as reflected
in interest and dividend payments. Though not without imperfections, the regulatory commission
process governing investor owned utilities provides a management discipline which accounts for the
full costs of providing water service. The most significant constraints on the ability of large, investor
owned utilities to finance needed infrastructure improvements results from regulatory lag time (time
elapsed between a utility's expenditure for compliance and the PUC's approval of rate increases) and
recent federal tax law changes (e.g.,elimination of investment tax credit and changes in depreciation
70
-------
schedules). By its nature, economic regulation of monopolies results in a time lag in responding to
changes in business conditions. PUC approval of rate increases may lag behind utility expenditures
during periods when the cost structure of the business is fundamentally changing. SDWA compliance
constitutes such a change, because unlike quantity related improvements, it can require significant
expenditures without a corresponding increase in revenues. A significant lag in approval of rate
increases can result in higher financing costs and lower earnings for the utility.
Many state regulatory commissions are confronting the regulatory lag time issue and devising
means to cope with SDWA-induced changes.2 The State of Connecticut, for example, has instituted
an SDWA Construction Work in Progress concept which allows utilities to raise rates for SDWA
compliance expenses quarterly, without having to go through a full-scale rate hearing.
Municipal Water Systems
Unlike investor owned utilities, cost recovery is not always assured in water systems owned
and operated by municipal governments. Often, the water department is a branch of the public works
department. Financial operations of the water department may be fully commingled with those of the
municipal government at large. As a result, there may be no assurance that revenues generated by
water bills actually finance the costs of water system operations. For example, some cities subsidize
water system operations with transfers from the general fund, while others subsidize the general fund
with water system revenues. Commingling of water system financial and fiscal management has
resulted in deferred maintenance and rehabilitation of water system infrastructure and an absence of
full cost pricing of water users. The investment shortfall has led to inefficient operations and high
levels of water loss through leakage, which increases the requirements for treatment capacity and
exacerbates the cost of SDWA compliance. Some communities with poorly run systems and
commingled budgeting cannot demonstrate sound financial management of their system, and
consequently face significant difficulties in raising the funds needed to finance capital improvements.
Some older municipalities with significant infrastructure deterioration problems have turned
the situation around through the use of full cost pricing and enterprise fund accounting (business type
accounting to insure that the water system is operated on a self sustaining basis, neither providing nor
receiving subsidies from other municipal services). Full cost pricing assures that revenues are
adequate to allow for investment in infrastructure rehabilitation, while enterprise fund accounting
assures that revenues from ratepayers are devoted to the water system.3
Capital financing of municipal water systems is typically accomplished through tax-exempt
bonds such as general obligation bonds and revenue bonds. Historically, the most popular mode of
financing has been general obligation (GO) bonds. Since GO bonds are backed by the full faith and
credit of the municipality, they.have historically been considered less risky .and carried a lower
interest cost. (However, as will be discussed shortly, revenue bonds may now be the less costly
option for well run municipal utilities.) The amount of GO bonds a municipality can issue is limited
Division of Ratepayer Advocates of the California Public Utilities Commission. Response to the
Filed Testimony of the California Water Association. WMA, Inc., September 1992.
3 Goldstein, James. "Full-Cost Water Pricing." AWWAJourna], February 1986.
71
-------
by state law, however, which means that the water system must compete for bonding capacity with
other essential public services. Because water system infrastructure is invisible to the average
community resident, other more visible services, such as road improvements and public buildings
have typically received higher priority for GO bonding capacity. Internal competition for municipal
bonding capacity is also an issue with regard to competition between the environmental compliance
needs of water, wastewater and solid waste, and within the water system portion of the capital budget,
between SDWA compliance needs and infrastructure rehabilitation needs.
Revenue bonds provide an alternative to GO bonds in many communities. Revenue bonds
carry a higher interest rate than GO bonds, because repayment is backed by the revenues of the water
utility, rather than the full faith and credit of the municipality. However, the spread in interest rates
between GO and revenue bonds has been decreasing substantially in recent years, to the point where
the difference is relatively minor (approximately 15 basis points)4. The only limits on the amount of
revenue bonds that can be issued would be local limits on indebtedness and limits imposed by the
bond market. The ability to issue revenue bonds and the cost of financing is determined by the
market's assessment of the ability of the utility to meet its repayment obligations from current and
future revenues. As a result, revenue bond funding is more attractive and less costly for municipal
utilities which are recognized by the market to be efficiently managed and operated, with adequate
and predictable revenues.
Public Water Districts and Authorities
A number of states have enacted laws which allow for the creation of independent districts or
authorities to provide water service. These quasi-governmental bodies typically have the authority to
set water rates and raise financing in the capital markets to fund system improvements. Dis-
tricts/authorities operate as autonomous financial entities. Since they operate independently of the
local government, political considerations tend to play less of a role, allowing them to employ full
cost principles to set rates at a level necessary to fully fund system operations and capital
improvements.
Large districts/authorities, some of which are State chartered, share access to the tax exempt
bond market with municipal water utilities, allowing them to finance capital improvements at
subsidized rates. Creation of districts/authorities is an attractive option for many municipalities,
because it allows the community to eliminate one of the sources of competition for capital funding,
freeing up financing capacity for other public services. Because water usage, as opposed to police
and fire protection or roads, can be metered and user charges easily applied, water systems are a
ready candidate for this type of organization. The success of independent water districts and
authorities in accessing the capital markets suggests that there is no constraint to financing SDWA
related investments in large systems when fiscal autonomy, coupled with management discipline, is
adopted.
4.2.2 Small to Medium Size Water Systems
Water systems which serve populations of 1,000 to 10,000 people face a number of financing
challenges which their large counterparts do not. At present, there are approximately 12,400 systems
in this medium to small size group, representing 16 percent of the population served by Community
4A basis point is 0.01%, so a 100 basis point increase would mean a 1 % rise in interest rates.
72
•
-------
Water Systems. These systems account for over 26 percent of total industry revenues. Ownership of
these systems is roughly proportional to large systems, consisting of privately (investor) owned
utilities, municipalities, and special districts or authorities. Many aspects of financing capacity are the
same in these systems as for the larger systems. However, there are three key differences which
result from their smaller size: 1) they have less access to financial markets, where fee structures favor
large scale transactions; 2) their credit worthiness is more sensitive to local economic conditions; and
3) the high transaction costs (e.g., preparing applications and reports and attending hearings) of
applying for water rate increases have made small systems less able to maintain their infrastructure
than their larger counterparts.
Privately Owned Water Systems
Privately owner systems are either owned by a single individual or a group of investors. The
access of these systems to the financial market is limited by their small size. Few firms in this size
category issue publicly traded stock and direct access to the bond market is limited by lack of
adequate information and the absence of economies of scale in bond issuance costs. As a result, the
primary source of private financing for small to medium size private systems is first mortgage and
revenue bonds, bank loans and retained earnings. Publicly subsidized sources of financing, such as
bond pools and revolving loan funds are generally unavailable to these entities, because most federal
and state laws restrict access to subsidized forms of financing for privately held firms. (However,
privately owned CWSs would be eligible for loans under President Clinton's proposed Drinking
Water State Revolving Fund.) Pennsylvania is an exception in that the State Revolving Loan Fund is
open to private water companies. Participation is still constrained, however, by limitations on the
amount of private activity bonds that a state can issue under the provisions of the 1986 Tax Reform
Act, which limits the annual volume of bonds issued by a state to $150 million, or $50 per capita,
whichever is greater.5 Water supply must compete with all other potential users of private activity
bond financing, including wastewater, solid waste, and other infrastructure projects involving private
entities. Historically, water supply systems have obtained only a small fraction of private activity
bond allocations.
Many small private water companies also suffer from inadequate rate relief from public utility
commissions. Historically, rate approval proceedings have been so expensive and complex that small
systems have not sought needed rate adjustments. Many state public utility commissions have begun
to modify their procedures to better accommodate small system needs, but significant levels of
infrastructure deterioration have already occurred in small private water companies due to inadequate
rates. Thus, unlike the larger investor owned systems, many small private systems face the same
backlogged financing needs as municipal systems that have not practiced full cost pricing.
A number of state public utility commissions have recently adopted measures to help attract
new capital to small private water systems. California has approved a higher rate of return for small
water companies. Other states have passed merger and acquisition adjustment laws intended to
simplify or eliminate barriers to absorption of small systems by larger systems or holding companies.
In general, small systems owned by these larger entities rely on the parent company for financing and
therefore do not suffer from lack of access to capital markets. Some states have passed takeover laws
which empower public utility commissions to compel large investor owned companies to absorb small
5Draft Introductory Text for Profiles of Financing Options for Nongovernmental Community
Water Systems. WMA m, 1QQ7, p T - "-^
73
-------
troubled systems. (The tax law changes discussed earlier result in increased tax costs for the private
acquiring utility which translate to higher rates for households than would have been the case without
the tax law changes).
Municipal and Other Public Water Systems
Medium to small size municipal systems share many common characteristics with their larger
municipal counterparts. The more successful systems are characterized by full cost pricing, enterprise
fund accounting and up-to-date plant and equipment. The troubled systems are characterized by
commingled budgeting, lack of full cost pricing, lack of professional management and deteriorated
infrastructure. Compounding these problems is a more limited set of financing options available to
small and medium size systems. GO bonding is more limited in this group. Due to economies of
scale in information gathering and bond issuance costs, the market for small community bonds is thin.
In order to overcome these problems, many medium to small size communities must engage in a
variety of methods to increase the marketability of their debt issues. Many purchase bond insurance
in order to sell their bond offerings. While insurance does increase the marketability of bond issues,
it also adds substantially to the cost of financing. The use of revenue bonds has become much more
important for medium to small size systems, surpassing GO bonds as the primary source of funding.
State loan programs, bond pools and revolving loan funds for drinking water systems have
been established in 29 states and are effective mechanisms for increasing access to financing for
publicly owned systems in this size range. In a bond pool, the state pools the bond issuance needs of
numerous small communities together into a single issue. Pooling provides economies of scale that
reduce the transaction cost of bond issuance. Some states may group offerings into a single bond
issue, backed by the full faith and credit of the state. Grouped offerings further decrease the cost of
financing, especially in states with good bond ratings.
In a state revolving loan fund (SRF), seed monies may either be made available via state
appropriations or bonds may be sold to provide the initial capital to establish the pool of funds from
which loans are made. In the wastewater program, there are also federal matching funds. Both the
bond pool and the SRF concepts inherently involve modest forms of subsidy from the state.
4.2.3 Small to Very Small Water Systems
Water systems serving fewer than 1000 persons face unique financing challenges. At present,
there are about 42,000 systems of this size, which represents 4 percent of the population served by
Community Water Systems. They account for over 20 percent of total industry revenues. These
small to very small water systems frequently bear little institutional resemblance to public utilities or
to small towns. Many are small clusters of homes connected to the same water source. In terms of
ownership, 90 percent of them are either cooperative homeowners associations, mobile home parks,
or small "mom and pop" water companies.
Systems serving fewer than 1000 persons account for about 73 percent of all systems, and
have the most difficulty complying with drinking water regulations. At the time these systems were
developed, it was never envisioned that the requirements of running a water system would require the
significant capital investments required by SDWA implementation. The vast majority of small to very
small systems has never raised capital in the financial markets and has no credit history. Bank loans,
secured by the personal assets of the system owner, have been the main financing mechanism
available to these systems. However, recent changes in bank reform have nearly eliminated bank
74
-------
financing as a source of revenue. Overall, small to very small systems are severely constrained by
lack of access to financial markets.
As a result of the management and financial weaknesses inherent in the existing institutional
arrangements, there is often a significant degree of infrastructure deterioration in small to very small
systems which adds greatly to total financing needs. Pennsylvania conducted a needs study which
included an on-site assessment of over 2000 small systems. The results indicated that roughly a dollar
of infrastructure rehabilitation investment is needed for every dollar of projected SDWA compliance
expenditures. Washington State also completed a comprehensive needs assessment of over 2000 small
water systems and arrived at similar results indicating a dollar-for-dollar relationship between
infrastructure rehabilitation and SDWA compliance costs. EPA regulatory compliance costs are
understated in the sense that they do not include repair of the distribution and storage system, which
is a necessary first step for many small systems if treatment is to be effective.
4.3 Technical Capacity Of Systems To Treat
According to the 1986 Amendments to the Safe Drinking Water Act, whenever EPA sets an
MCL for a contaminant, the Best Available Technology (BAT) for treating that contaminant must also
be specified. In the case of some contaminants, a treatment technique is specified in place of an
MCL. EPA has analyzed twenty-two different treatment technologies, each of which is considered
Best Available Technology (BAT) for meeting at least one regulation or MCL under the SDWA
Amendments. A summary of all potential contaminant removal applications for each process is
presented in Exhibit 4.8. Technology exists for addressing all of the major drinking water
contaminant groups: pathogens, organic and inorganic chemicals, disinfectants, and disinfection
byproducts.
A brief description of each treatment process is as follows:
Disinfection: a process used to deliberately reduce the number of pathogenic microorganisms.
It is accomplished by adding chlorine compounds, ozone, or chlorine dioxide (or any
combinations) to drinking water or by exposing the drinking water to ultraviolet radiation;
Chlorination: a disinfection process where chlorine is used as the treatment agent;
Potassium Permanganate: an oxidant added to water to precipitate metals and enhance
removal of organic contaminants;
Coagulation/Filtration: a process for removing paniculate matter from water by passage
through porous media consisting of the following steps: coagulation, flocculation,
sedimentation, and filtration;
Direct Filtration: a treatment process very similar to Coagulation/Filtration except that there
is no sedimentation step;
In-Line Filtration: the simplest form of direct filtration, wherein filtration is preceded by the
addition of chemicals and rapid mixing;
75
-------
Packed Tower Aeration: a treatment process in which drinking water is transferred out of a
solution in water to a solution in air. A column of water is run parallel to a column of air,
allowing for the transferral. The extent of removal of contaminants from water is determined
by the length of the column and the volatility of the contaminant;
Ion Exchange: a process by which an ion on the solid phase is exchanged for an ion in the
feed water. Exchange resins are insoluble solids comprising fixed cations or anions capable
of exchanging with similarly-charged, mobile ions in the feed water;
Activated Alumina: a form of ion exchange, in which the charged contaminants in the
drinking water are exchanged with the surface hydroxide ions on the alumina;
Ozonation: the same process as chlorination, except ozone is the treatment agent in place of
a chlorinated compound;
UV Irradiation: a disinfection process in which drinking water is exposed to ultraviolet (UV)
wavelengths of light to destroy pathogenic microorganisms;
Chloramination: process similar to chlorination, except chloramine replaces chlorine as the
treatment agent;
Lime Softening: a treatment process used to reduce the hardness of water caused by the
presence of calcium and magnesium compounds in solution. Hardness is removed by
adjusting the Ph to precipitate calcium carbonate out of solution;
Slow Sand Filtration: a treatment process that uses a deep bed of sand to remove particles
and microorganisms from water;
Greensand Filtration: very similar to slow sand filtration, except a specially coated material
(greensand) is used to remove iron, manganese, taste, and odors from water;
Diatomaceous Earth Filtration: this treatment process, similar to other filtration types, uses a
thin layer of diatomaceous earth (DE) supported by a filter to remove particles and
microorganisms from the water. The DE layer must be continuously replenished to maintain
the needed degree of porosity for the filter cake;
Industrial Cartridge Filters: this treatment process utilizes disposable cartridges to filter
drinking water;
Microfiltration. Ultrafiltration. and Nanofiltration: types of membrane filtration which
remove particulates and microorganisms above a specific size as delineated by the filter used;
Reverse Osmosis: a pressure driven treatment process using a specially prepared membrane
that permits the flow of water through the membrane but acts as a selective barrier to
contaminants. The pressure applied exceeds the pressure that would be produced by osmosis,
which forces pure water through the membrane and leaves salts behind;
76
-------
Granular Activated Carbon: a treatment process using a filter containing activated carbon.
The carbon bonds with specific contaminants (such as SOCs) and traps them inside the filter;
and
Diffused Aeration: a treatment process similar to packed columns, except that water is run on
a bed containing air jets. The contaminant is transferred from the water into the air, where it
is then removed;
Appendix B presents cost information for the individual small system BAT processes at
selected average flows (generally corresponding to the four smallest system size categories). Capital
costs are presented in total dollars and O&M costs are presented in cents per 1,000 gallons treated.
Also presented in these tables are total production costs in cents per 1,000 gallons treated. The total
production cost is the sum of the debt service on the capital cost, amortized over 20 years at 10
percent interest, and the O&M cost. A 10 percent interest rate was used for these tables, in contrast
with other tables in the report which use a 7 percent interest rate. Included with the process costs are
process descriptions, equipment lists, design assumptions, and labor recommendations, if available.
Two of the above technologies, coagulation/filtration and lime softening, may have limited
applicability for small systems (systems serving < 3300 persons) because of their frequent need for
close operator attention.
The O&M costs presented in the appendix are based on average flow and include chemicals,
replacement materials, and power. Labor for operation and equipment maintenance is also a major
O&M cost component, but was not included in the process O&M costs because of the variation in
manpower available for small system treatment processes throughout the United States. Where labor
recommendations are presented the cost of labor may be estimated assuming an hourly wage of
$14.506.
EPA has begun to consider the possibility of designating a SuperBAT for CWSs. The idea
behind the SuperBAT is that it may be feasible to replace a large number of individual regulations by
specifying design, operation, and maintenance requirements for a single technology like aeration or
reverse osmosis and offering it as an optional compliance strategy. Installation of SuperBAT would
be considered equivalent to compliance with an MCL or treatment technique. Where a system installs
SuperBAT, EPA could waive or reduce monitoring and reporting requirements. The savings in
monitoring costs could conceivably meet capital cost needs for the equipment over the years.
Membrane filtration and aeration are currently the most promising technologies for consideration as
Super-BAT.
Under the SuperBAT concept, EPA would by regulation mandate the specifications of the
treatment technology which would be acceptable. Characterizing equipment performance in sufficient
detail to get a minimum standard would require considerable research, cost several million dollars,
and take three to five years to develop. The specifications could preclude an equally effective piece
of equipment
'Derived from data published in the Engineering News Record (June 29, 1992 issue). The fully loaded labor rate of $14.50 per hour
for small systems is based on an average wage of $1.65 per hour plus an average benefit rate of 12.2 percent for non-union workers and 50
percent overhead.
77
-------
CO
!
5
CO
\
1
1
-S
X
X
X
X
X
X
!sli
Filt
G
Alur
Ion
78
-------
from being used. This approach is substantially different from current practice. At present,
regulations designate a BAT by stating a treatment by name. The regulations goes into no further
detail about the treatment. The current practice of loosely defining BAT has advantages and
disadvantages. The principal advantages of the present practice are that it greatly simplifies the
regulations and provides states and systems considerable flexibility. It would also reduce State
SDWA implementation costs. A disadvantage is that the loose definition does not provide guidance
needed by Primacy Agencies, water utilities, or consulting engineers in determining if a proposed
system should be classified as BAT. Accordingly, operating performance criteria would have to be
specified. If a Primacy Agency approves a proposed system as BAT and the system does not meet
the MCL, the water utility is still responsible for meeting the MCLs. The ultimate financial burden is
placed on the water utility. No one wants to see a treatment system installed and then not meet a
treatment objective. Consequently, loosely defining BAT places the Primacy Agency decision makers
and water utility at risk of approving and installing an under-designed system based on
recommendations from sources which are not accountable for the performance of the system.
4.4 Options Available to Improve the Financial and Technical Capacity of Small Water
Systems
Many experts familiar with the issue are concerned about the financial and technical ability of
small water systems to comply with the regulations developed under the 1986 Amendments to
SDWA. Small water systems have long posed a problem for State and Federal regulators. Issues
such as private ownership, minimal/decrepit infrastructure, limited technical and managerial expertise,
and limited resources severely restrict the ability of small systems to comply with the SDWA
requirements. Regulators have long spoken of the small system problem; however there are really
many problems - some common to most small systems, and some unique to certain sub-groups of
systems.
4.4.1 Small System Origins
Both rural and suburban small water systems are a product of the environment in which they
were created. Prior to enactment of the Safe Drinking Water Act and especially its 1986
Amendments, regulatory and treatment requirements were minimal. For the most part, all that was
needed was a well, a pump, a tank, and perhaps a chlorinator. Operation and maintenance require-
ments were minimal. The most significant cost element was the initial cost of constructing a distribu-
tion system, in which there are no economies of scale. The result has been the proliferation of
thousands of small water systems. In some suburban areas, this proliferation continues today. Only
now, with the advent of SDWA-induced treatment costs (capital costs and operation and maintenance
costs), does economies of scale become a consideration.
As a result of the relatively weak forces in the cost environment, the institutional mechanisms
that were developed to manage small systems were also weak. There were no great demands to raise
capital, no tough performance standards, no highly trained staff, and no need to impose significant
fees on users. The need to maintain and replace the initial infrastructure could be easily ignored
(especially by developers) because deterioration was slow and, for all practical purposes, invisible.
Initially, small water system institutions appeared to offer a viable means of providing water
supply infrastructure services. In retrospect, however, the present configuration of water supply
79
-------
Exhibit 4.9
Illustration of Small System Ownership as a Function of Size in Pennsylvania
100 -f
<100
Regional Gov't Utilities
Municipal Gov't Utilities
Private Water Companies
Mobile Home Parks
Homeowners Associations
101-500 501-1000 1001-3500
Population Size Category
2501-3300
82
-------
CO
^
CO
O
03
CO
O
QJ
CO
CD
O
""^
o
CO
CL
CO
E
'x
o
Q.
Q.
83
-------
The Sole Source Aquifer (SSA) Program authorizes EPA to designate aquifers as sole or
principle sources of drinking water for an area. These designations allow the Agency to review
federal financially assisted projects in SSA areas. Commitments for federal financial assistance for a
project are to be withheld if EPA finds that the project may contaminate an SSA so as to create a
significant hazard to public healdi.
EPA has designated 58 SSAs nationwide since 1975. About 21 million people live within
these SSA areas. The Agency conducted 366 project reviews in FY 1991-92. Of these reviews, 31
resulted in project modifications to prevent contamination and 5 resulted in negative recommendations
due to their potential for contaminating an SSA.
4.4.4 Small System Viability and Restructuring
In nature animals and plants respond to critical changes in their environment through a
process of behavioral and structural adaptation. They adapt in order to survive. As illustrated in
Exhibit 4 12, the same is true of economic institutions. When there are fundamental changes in the
business environment, businesses must adapt in order to survive. Economists call this process
"restructuring." Companies restructure to meet changes in their environment in order to remain
"viable" enterprises. Small systems must now undergo this process, but will need help. Unlike
business enterprises that are accustomed to responding to the pressures of the market, small water
systems have not had to face major change before. Their progress in adapting to changing conditions
will not be substantially market-driven, and will have to be assisted.
Exhibit 4.11 shows there are many different strategies that can be adopted in restructuring
water systems. They can be broadly classified into two categories: external and internal. External
strategies involve active collaboration with neighboring systems to attain the advantages of economies
of scale In addition to physical advantages, larger scale operations provide greater access to capital
financing and to skilled management. Internal restructuring strategies seek to provide greater access
to capital financing and operational efficiency through internal fiscal and managerial discipline.
External restructuring strategies may involve physical or institutional integration with
neighboring systems through a variety of approaches to consolidation or cooperation.
• Consolidation involves extending a water main to enable a merger, or a purchased
water arrangement with a nearby system. Another example is formation of a county
or regional public authority to provide central management and operation and
maintenance services.
• Cooperation encompasses an array of strategies for obtaining economies of scale in
management, operations, and finance through sharing arrangements. A popular model
is contract operation and maintenance services on a rotating, circuit-rider basis.
There are also an array of looser strategies, involving equipment sharing and joint
procurement to pool buying power.
Exhibit 4.12 provides information on some common restructuring techniques.
86
-------
Internal restructuring strategies involve changes in management and finance to produce a
"turnaround" in the likely fate of a small system. There are many systems that may be able to handle
the changes ahead if they make the right management and financial adjustments, such as raising rates
and adopting enterprise fund accounting procedures. . r
It is not the primary objective to force small water systems into consolidation schemes; not all
small systems need to be restructured. Going it alone will continue to be a popular option. But the
question remains, what constitutes the most viable option in light of the changes that are afoot in the
water business? A viable water system is one which has a sustainable ability to meet performance
requirements over the long-term. The implementation of SDWA rules has merely exposed and
exacerbated the small system problem. The fundamental problem of infrastructure deterioration pre-
dates SDWA requirements, and would require restructuring of many systems, even in the absence of
the SDWA Amendments.
4.4.4.1
The Potential for Restructuring
As noted earlier, about one-half of all small systems are located within SMS As. A recent
survey conducted in the State of Washington indicated that 40 percent of all systems serving fewer
than 1000 persons are within one-half mile of another system. The experiences of Alabama, South
Dakota, and West Virginia similarly indicate that for approximately one-half of small systems,
physical interconnection or shared management should be obvious considerations.
Interpreting the record of experience in this area presents a paradox. Efforts to promote both
consolidation and collaboration schemes have produced outstanding success stories in some instances,
while in other instances such ideas have generated little enthusiasm. There are two factors that can
help to explain this apparent paradox.
• Incentives. In order for consolidation or collaboration schemes to succeed, there
must be a compelling need to act which causes participants to recognize their common
interests. Some States, such as Alabama, have made restructuring a priority and have
been successful in motivating systems to restructure.
• Barriers. There is an array of legal and institutional barriers to consolidation and
collaboration schemes. Major barriers include: limitations imposed by the 1986 Tax
Reform Act; the presence of regressive takeover, merger, and acquisition adjustment
laws; political pressures; and PUC regulatory procedures that result in high
transaction costs for obtaining rate increases.
Exhibit 4.13 provides an estimate of small system restructuring potential. This estimate is
based upon the experience of the States of Alabama, South Dakota, and West Virginia in promoting
small system restructuring over the last decade or so (these States have promoted restructuring much
more aggressively than most other States in the past). Based upon the judgement of State regulators
all the small systems in each of these States were classified as either: always viable; made viable
through restructuring; having the potential to be restructured; or not having the potential to be
87
-------
CO
t
D)
C
O
Z3
m CO.D a.
alSii 88
•!•• ^^^ ^^^L ••• , ^^. ^^«
cSll-r SJS
3 jQ 0) JK — J3 •;
iy) co a. Si co < :
a b
CO
o
,2 £
,— " o
3 «
o o
co a.
c o
O -m
la
O>'«
52 .= C 3T ^ x CO ?: C. 52
£ C0o2 ° ° C m^E
o ^ o != <• <* ^ c .E 3
*= o
I
o
£
CO 3
C O
^2 S
'•^ £= "S co
c co 8"S
co c 2 o
Q 3 O •{= CD
CO OO £Z
o o • o
.88
-------
Exhibit 4.12
Common techniques used in restructuring non-viable systems
Contract 0 & M
A non-viable system contracts with a larger viable
system for the viable system to provide operations
and maintenance, and sometimes management,
services. Such an arrangement allows the non-viable
system to buy into the economies of scale of the
larger system. Such an arrangement is sometimes
referred to as Satellite Management.
Formation of Service or Utility District
Management Consolidation
Physical Consolidation
A new institutional entity such as a Public Service
District or County Utility District is created to
assume operations or operations and ownership of
non-viable systems.
A non-viable system is acquired by a viable system.
The assets and ownership of the non-viable system
are transferred to the viable system. The systems
are not physically interconnected but are owned and
operated by a single entity. Such an arrangement is
sometimes referred to as Satellite Ownership. It
may be achieved through voluntary agreement of the
systems or it may be ordered by a State Agency with
appropriate authority.
A pipe is constructed to physically interconnect a
non-viable with a viable system. The viable system
may actually acquire the non-viable system,
assuming its ownership and assets, or the non-viable
system may be viable once it can purchase water
wholesale from a larger system. In the latter case
the systems could remain independently owned.
Acquisition of the non-viable system by the viable
system may occur through voluntary agreement
between the systems or it may be ordered by a State
agency with appropriate authority.
89
-------
0 CO
Z CO
DC
o
CO CO
LU
DC
LJJ <
!<
LU
CO O
LU Q_
O
LO
o
z
cc
COCCQ.
CO
CO
O
CO
o
o
o
o"
in
25
111
g
CC
Q.
CL
90
-------
restructured. Exhibit 4.14 presents the summary data from this analysis. Since most States have not
pursued restructuring as aggressively as these three States the number of systems made viable through
restructuring was added to the number of systems having the potential for restructuring and
converted to a percent of all small systems in the three States to provide an estimate of the
restructuring potential in other States. That is, the experience of these three States has been
extrapolated to provide a national estimate of restructuring potential.
' t*
This analysis suggests that approximately 50 percent of small community water systems
(systems serving less than 3,300 persons) appear to have the potential for restructuring.
Approximately 30 percent of small systems appear to be viable and do not need to restructure.
Finally, approximately 20 percent of small systems do not currently appear viable and are unable to
restructure. There are a number of reasons why a system might fall into this last category. For
example, it may be geographically isolated such that even operation as a satellite of a larger system is
not feasible. There may be overriding local political issues that prevent restructuring. Finally, grants
and/or subsidized loans may be unavailable in certain cases where they would be needed to make it
feasible for a viable system to take on a non-viable one. .
4.4.4.2
Providing Incentives to Restructuring
SDWA compliance can provide a strong incentive for small systems to consider restructuring.
However, two forms of small system relief which lessen the pressure to restructure have been
proposed: providing additional time to comply and offering financial assistance. It is therefore
important for financial assistance programs to be designed to encourage restructuring and
consolidation of non-viable systems. It is also important for financial assistance programs to be
designed to avoid "foot-dragging" by systems waiting to gain access to such assistance programs.
Collaborative efforts such as physical or institutional restructuring require that all participants
simultaneously recognize a compelling motivation to act. Recognition of the mutual benefits of
restructuring is needed for a successful regional scheme. Currently, the SDWA regulatory program
results in compliance schedules which are not conducive to simultaneous action of systems within a
region. Many of the current financial assistance programs are not sensitive to the possibilities of
restructuring.
Providing financial assistance is a form of relief which should be considered for small
systems. However, unconditional provision of financial assistance can result in an incentive structure
that potentially discourages restructuring and props up non-viable systems.
EPA encourages State drinking water programs to develop eligibility and prioritization
criteria, and enforcement mechanisms, that give preference to re-structuring solutions, and discourage
non-viable systems from making unwise investment decisions. Programs to aid small systems should
be designed to ensure that financial assistance, technical assistance, and regulatory relief are not used
to prop up fundamentally non-viable systems. These programs could be targeted to improving system
viability and creating systems which will have the capacity to sustain compliance in the long term.
Federal and state regulatory agencies could coordinate with those agencies which provide
assistance to small systems, to ensure that the goals, objectives and programs of each yield a
consistent and clear set of incentives designed to improving long term viability. For example, EPA
91
-------
DC Q
LU
3
•^
CO
00
LU
CO CO
coj
0.
X
< UJ
CO
z
O
CO
CC
UJ
a.
o
o
eo
CO
V
o
z
>
CC
111
CO
CO
2
III
I
CO
_J
<
LU
o
z
UJ
CC
LU
OL
X
LU
O
LU
-I
u.
LU
or
92
-------
will coordinate with the Rural Development Administration (RDA) to ensure that RDA funding
priorities are consistent with SDWA compliance priorities and that they encourage restructuring that
leads to long term system viability and compliance.
Privately owned water utilities are not generally eligible for State or Federal financing
programs. (However, privately-owned Community Water Systems would be eligible under the
Administration's propose Drinking Water State Revolving Fund.) The eligibility for such programs,
along with other considerations affect the interest of privately owned utilities in acquiring neighboring
small non-viable system. The factors that the privately owned utility may consider include: the
acquisition cost for the small system; the investment necessary to bring the system into compliance;
the utility's ability to recover its costs through rates (rates for larger systems are usually governed by
a Public Utilities Commission); and the willingness and ability of the systems customers to pay the
increased rates.
Federal and State taxes are an additional consideration for privately owned utilities. For
example, since the 1986 Tax Reform Act, utilities have been taxed on Contributions In Aid of
Construction (CIAC) as revenue in the year the utility receives the contribution. A utility may have
taxable CIAC when it acquires the assets of another system. For instance, the Pennichuck Water
Works of Nashua, NH has documented a case study involving its acquisition of a small developer
built water system serving the Richardson Estates subdivision in East Derry, NH. The developer was
willing to sell the water system to Pennichuck for less than its depreciated replacement cost.
Pennichuck would be taxed on the difference between the purchase price and the depreciated
replacement cost. As a result of this, average annual household water bills in Richardson Estates
would be $838 rather than $620. ,
4.4.4.3
Removing Barriers to Restructuring
Some of the most significant barriers to consolidation and collaboration are emotional
barriers. There is often a resistance to the loss of autonomous control over something as fundamental
to a community as its water supply. Other emotional barriers to collaborative schemes may result
from issues peripherally related to water supply, such as growth and development policies, and from
totally unrelated factors, such as high school football rivalries.
The pressure of SDWA compliance is the type of compelling need that can overcome
emotional barriers and cause adjacent communities to see their common interest in a collaborative
arrangement. SDWA compliance could be viewed as a credible compelling need, in several respects.
First, the need for expensive improvements in the water supply system would need to be accepted by
the community. Since health risk reductions are difficult to express, the change in quality is much
more subtle than the prospective change in cost. The cost dimension is further exaggerated in
systems that have historically undercharged for water service, particularly when regulation-related
cost increases are accompanied by substantial infrastructure rehabilitation costs that are not separable
in the minds of the ratepayers. The result is a distorted appreciation of the cost/benefit relationship of
SDWA requirements. This distortion can only be corrected through a broader understanding of
underlying infrastructure issues.
93
-------
Second, in order for the state primacy agency to become a credible advocate of small system
restructuring, they and other State agencies may need to shift their focus. State institutions developed
over a period of time in which small water system viability was not recognized as a problem. As a
result, the pattern of incentives presented by state government programs and policies is often
insensitive to restructuring possibilities.
The SDWA primacy agency and state financing agencies have essential roles in determining
the incentives facing small water systems. Engineering conservatism and the mere cost of the review
process have presented a barrier to the introduction of potential small-scale innovative technologies.
This area of policy could be reviewed in light of the overall problem of finding lasting solutions to
the small system problem.
There are two other state agencies that affect barriers to beneficial economic restructuring of
water systems: the state public utility commission (PUC) and the state water resources allocation
agency. PUCs play a significant role in barriers and incentives affecting the feasibility of
regionalization and restructuring options. When a municipal system extends service to a suburban
area outside the city limits, the PUC often intervenes to regulate rates charged to the suburban
customers. In many cases, this practice has been a significant barrier to logical extensions of service
to contiguous suburban areas and to the creation of regional water systems.
In many states, investor-owned water companies that own and operate a number of large and
small systems within the state have sought to regionalize rates. In some cases, PUCs have approved
single tariff rates for such situations, allowing the company to incorporate systems that might not be
economically viable within a regionalized scheme. This approach reduces the burden of rate case
filings to one unified application for the entire regional operation.
Another aspect of PUC involvement is in regulating the transfer of ownership between two
private water companies or between a private and a publicly owned company. There are many
situations, such as the municipal/suburban boundary case, in which publicly and privately owned
systems exist in a contiguous pattern. Historically, PUCs have applied a complicated set of iron-clad
rules to the evaluation of ownership transfers in an effort to protect the public from being charged too
much when depreciated plant and equipment changes hands. PUC policies could be revisited in order
to assess whether the benefits of such regulatory protection outweigh the costs of limiting regionalized
solutions that can provide a more viable long-term approach to providing quality service. Several
states, including Connecticut, Pennsylvania, and Washington, have enacted more liberal merger and
acquisition adjustment laws which allow for increased restructuring.
Water resources agencies may also significantly affect incentive structures. A potential
regional ization scheme that might make economic sense in light of the burden of SDWA compliance
and long-term viability may be preempted due to the potential effects of consolidation in causing the
readjustment of water allocation formulas. Water resource allocation policies could be revisited to
support the broader objective of providing water supply in a long-term, sustainable manner.
94
-------
4.4.4.4
State Viability Initiatives
A number of States are developing or implementing programs to ensure the viability of new
small water systems. In general these States are requiring developers to demonstrate that their
proposed system will be viable over the long-term before allowing the system to be built and
operated. For example, the States of Connecticut, Maryland, and Washington use a permitting
process to ensure that new small systems comply with minimum design, operating, and construction
standards. These States also require financial, operational, and management evaluations before
installation of a proposed new system. An additional approach to new system screening is to require
financially-backed assurances or guarantees of viability. The concepts being considered by States
include: escrow accounts, an irrevocable letter of credit from a bank, reputable co-signers, and a
contract with a reputable contract operations and maintenance organization.
In some States, viability initiatives address existing systems as well as new or future systems.
A common conclusion reached by states that are implementing viability initiatives for existing systems
is that strategies for intervention can be most effective when they are viewed as a coordinated,
interagency effort undertaken on a statewide basis. Two primary components of a State viability
initiative could be to promote long-term planning and facilitate restructuring.
Planning initiatives can consist of development of system-level business plans and
comprehensive water supply planning. The facilitation of restructuring requires removal of barriers,
provision of incentives, and mandatory restructuring of basket cases.
The States of Connecticut, Maryland, Pennsylvania, and Washington are at the forefront of
the viability initiative and have incorporated a majority of the planning initiatives stated above.
Changes in federal legislation could build upon the knowledge developed by the primacy agencies
within these states.
The States of Maryland and Washington both require comprehensive water supply plans.
Counties within Maryland develop comprehensive plans which specify service areas, needs for new
service over the next 10 years, and financing proposals. Washington's program is separated into two
parts: financing and operations. The financial program is intended to facilitate financing of
improvements required to operate the system, including estimating potential future growth,
documenting the availability of adequate capital, and showing the existence of an adequate revenue
stream. The operations program requires water systems to identify all persons responsible for normal
operations, preventive maintenance, troubleshooting, monitoring, budget formulation, complaints, and
emergencies.
California has not yet incorporated financial requirements into its permitting process.
However, the drinking water program within the California Environmental Protection Agency has
legislated authority to establish financial and managerial requirements as part of its operating permit
review for public water systems. Section 4023.3, of Assembly Bill 2158 states that "any system
seeking an initial permit...or a permit due to a change in ownership shall...be required to demonstrate
financial responsibility as a condition of receiving the permit." The State is considering other
provisions, which would require new and existing systems that change ownership to create a master
plan which would include management information and a financial plan.
95
-------
The California Public Utility Commission (PUC) also imposes stringent financial requirements
on systems requesting permits (new systems). It projected that, if gross operating revenues would fall
below $200,000, the PUC may require the system to post a bond of up to $50,000. This requirement
discourages the formation of new investor owned systems which do not have adequate financial
resources.
Some of the other states that are notable for their viability initiatives incorporate a variety of
techniques to address non-viable systems. Maryland as well as Connecticut have strict oversight
policies. A majority of the five states require systems to submit plans that describe their financial
status and structure. Information required includes capital expenditures and O&M requirements, as
well as revenue requirements.
An essential element of a strategy to facilitate restructuring is takeover authority: the ability
to direct restructuring of "basket case" systems that have defaulted under regulatory pressure. The
State of Connecticut has developed a program for preventing the creation of new, potentially non-
viable small systems. Connecticut statutes establish guidelines for ordering a municipality or private
company to take over a failing water company. Specifically, a water system may be subject to
acquisition if the system has repeatedly been found in violation of state drinking water regulations, if
a notice of violation and an administrative order have been issued, and if the system has failed to
comply with the administrative order. The acquiring facility must either extend its water mains to
supply water or establish the system as a satellite. The Connecticut Department of Public Utility
Companies (DPUC) subsequently adjusts the allowable water rates to compensate the acquiring
company for the reasonable costs of acquiring and operating the new system.
One notable small system case that has been subject to the takeover authority of Connecticut
is Lebanon Water Company (LWC). LWC served 53 customers in a residential development of
single-family dwellings. The DPUC determined that the owner had abandoned his company and was
in non-compliance with the DPUC's orders. The Connecticut American Water Company (CAWC)
was appointed as the receiver for the takeover of LWC. After unsuccessful attempts by State agencies
to collect fines levied on the owner, and to gain voluntary transfer of ownership, the State, pursuant
to Connecticut General Statutes sections 16-262n and 16-262o, transferred ownership to CAWC.
CAWC was then ordered to pay all of LWC debts. LWC customers were billed at a flat rate
of $144 per year until water meters were installed. After meters were installed, LWC customers
were billed at their new Mystic Valley District water rate of $470 annually. Because this district had
one of the highest water rates in the state, DPUC subsequently ordered CAWC to devise a rate
equalizing scheme. This scheme eventually lowered the annual bill to $269 per year. According to
the DPUC, if acquisition and upgrading costs were not shared by the Mystic Valley customers, the
bill to each of the LWC customers would have been $1,000 annually.
Takeover authority can be very expensive to exercise. Forced restructuring is also likely to
be much more troublesome than a restructuring process driven by incentives. Under the incentive-
driven approach, the number of systems that ultimately have to be restructured is minimized through a
process of: 1) providing incentives for long-term planning to identify options, 2) removing barriers
and creating incentives to increase the range of options available, and 3) applying firm SDWA
enforcement pressure to drive the process.
96
-------
Under the incentive approach, takeover authority is used to follow through on SDWA
enforcement pressure. When a system defaults, the state needs to be able to keep the pressure on,
while opening as many doors to viable restructuring options as possible. A system that is reluctant to
participate in a regionalized or cooperative scheme may be more willing if they realize that the state
has the authority to ultimately compel their participation if they cannot make it on their own.
Presently, the takeover mechanism to deal with SDWA compliance defaults is unclear in many states.
In order to provide a credible enforcement tool, states could be required, as a condition of
primacy, to implement small system viability programs which include mandatory takeover authority.
EPA could be authorized to promulgate a regulation specifying the minimum requirements for state
viability programs. States such as Pennsylvania and Connecticut have developed small system
viability programs which include elements like new system viability screening, development of system
level business plans, and takeover authority. EPA estimates the cumulative annual costs to States of
implementing programs like those in CT and PA to be approximately five million dollars.
Takeover of basket case systems is likely to involve financial subsidies. In this respect,
takeover authority is a safety net: a reflection of state policy regarding poverty, infrastructure, and
economic development. Development of an effective takeover mechanism would draw on these
broader constituencies. An incentive-based approach to the restructuring process provides a means of
minimizing the total amount of subsidy required and a means of assuring that subsidies are directed to
the basket case situations where this type of assistance is truly needed.
4.4.5 Lower Cost Small System Treatment Technologies
In late 1988, EPA launched an effort known as the "Small Systems Low Cost Technology
Initiative." The initiative was intended to focus on systems in size ranges below that served by
conventional treatment plants and above that served by the point-of-use/point-of-entry industry. The
idea was to encourage the manufacturing community to focus their treatment technology development
efforts on small systems serving 25-1500 persons.
Medium to large water systems (those serving more than 3300 persons) generally utilize
conventionally designed and constructed treatment plants. The systems are designed by a consulting
engineer and feature fabricated concrete and steel structures as opposed to skid-mounted, pre-
assembled equipment. Larger systems have customer bases which can usually absorb the associated
costs of fully engineered systems at a reasonable user fee to the ratepayers. Costs of such systems
can be prohibitive for small systems.
Appropriate "packaged" treatment technologies are pre-engineered to be applicable in a broad
spectrum of treatment applications. Depending on the unit and application, these systems can exceed,
meet, or fall short of the performance of engineered systems.
Package systems are usually shipped to the site preassembled and ready for installation. They
typically require a minimum of on-site assembly, construction or interconnection with the existing
system. Such systems are usually delivered complete with all of the necessary apparatus,
instrumentation and controls needed for operation. They are designed to streamline the treatment
process and simplify operations and maintenance. These units can be sized to accommodate small
communities or individual homes.
97
-------
Small system equipment suppliers/manufacturers can often provide some of the operating
services needed by small systems. Some suppliers have the capability to perform installation, pilot
testing, long-term operations and maintenance (O&M), and other services. The equipment supplier
can work directly with the system, state drinking water officials and professional engineers. This
type of operational and maintenance assistance can help small systems achieve or maintain
compliance. '
The Small Systems Initiative deals with state approval of these small scale technologies. All
states have programs in which engineering plans for water treatment facilities and appurtenances must
be reviewed and approved. The engineering plan review process tends to favor conventional
technologies since such designs are more familiar to the reviewers and are often accompanied by
standardized information on the processes and performance associated with these systems. Thus, new
technologies may have trouble securing approval.
For the past four years, EPA, working through a loose knit coalition of states, manufacturers
and water industry associations, has focussed on two aspects of the small system technology problem.
The objective has been to ensure that: 1) small scale, affordable technologies are available; and, 2)
once available, technologies will be approved by the states.
There are still many barriers that stand in the way of small systems trying to obtain
appropriate cost effective technologies, including:
* • Lack of information and associated uncertainty as to the performance and operating
ranges of package technologies vis-a-vis a variety of water characteristics;
Lack of familiarity of small scale technologies by system owners/operators and
technical assistance providers;
Lack of understanding by equipment suppliers of state design review processes and
information requirements; and
Local government procurement regulations.
Because -most states do not have a comprehensive plan review process in place which is user-
friendly for small systems, equipment suppliers still tend-to have a lack of understanding of the design
review process and the information states require. According to EPA's Report on State Engineering
Practices for Small Water Systems, states want additional information from the supplier/manufacturer
on pilot testing, third party approval, and certification of their products.
Other information that states want from the supplier/manufacturer is supporting technical
documentation as well as more R&D and field monitoring of products. States feel this additional
information is needed to ensure technology performance. However, limited resources are available to
state drinking water officials for evaluating new technologies.
Information on package technologies can be used in state viability programs to assist small
systems in developing strategies to maximize the range of low cost technology choices available to
them. State viability programs should also include procedures for simplified and streamlined review
and approval of lower cost package technologies for small systems.
98
-------
EPA is working with States, manufacturers, and third party certiflers to develop protocols to
promote more uniform requirements among States for alternative technology approval. The goal is to
develop an approval protocol for new or alternative technologies and to ensure that once a State has
approved a technology the approval can transfer to other States without expensive, time consuming
additional requirements that are different among the States. This effort is building upon the Western
States Protocol.
EPA could continue to promote the development and application of lower cost package
treatment technologies through cooperative efforts with states, equipment suppliers, and professional
organizations. EPA could, resources permitting, oversee completion of the demonstration projects
begun under the "Small Systems Low Cost Technology Initiative," document and publicize the
results, gather data on other existing installations and make it available to state engineers and others.
These and other actions would promote revisions to state plan review processes.
The use of bottled water, for example, could potentially provide a lower cost alternative to
conventional treatment options for very small systems serving 25-100 customers. For systems serving
more than 100 customers, bottled water is likely a less cost effective option, and will typically be
more expensive, on average, than packaged systems.
In very small non-viable systems when other options are not feasible, eligibility criteria and
performance standards could be established to assure that these alternatives to treatment provide
reliable solutions that are both technically feasible and institutionally implementable.
4.4.6 Targeted Financial Assistance
EPA has proposed, consistent with President Clinton's A Vision of Change for America, a
Drinking Water State Revolving Fund (DWSRF) that would provide $599 million for Fiscal Year
1994 and $1 billion for each of Fiscal Years 1995-1998. The proposed DWSRF would authorize new
capitalization grants to States to establish revolving loan funds to help public water systems comply
with the Safe Drinking Water Act. The premise of the DWSRF is that a national investment is
needed in infrastructure services that touch every American. The DWSRF would limit the type of
financial assistance to low-interest and no-interest loans to ensure the integrity of each State's fund..
EPA would be required to conduct a drinking water needs survey within two years of enactment so
that loans are used to fund the highest priority needs. Establishment of a new DWSRF will help
States and communities meet the costs of investments needed to comply with SDWA regulations.
A number of State and federal programs have been successful in increasing small system
access to capital and in helping systems correct institutional deficiencies hampering direct financing.
These include loan programs, bond banks, and grant programs.
The unconditional provision of financial assistance can inadvertently discourage restructuring
and prop up fundamentally non-viable systems. Well designed financial assistance programs can be
targeted to improving system viability and creating systems which have the capacity to sustain
compliance in the long term.
Grants are the most substantial and direct form of assistance. They are attractive to recipients
because they are essentially a gift. However, grants have several negative consequences which should
oe oorne in mind when considering their use. By helping to resolve a short-term financing crisis,
99
-------
grants may have the negative effect of prolonging operations of non-viable systems. The "basket
case" systems which do not have the capability of funding even modest system modernization needs
will not be capable of meeting SDWA requirements in their present structures without massive
infusions of capital. Grants may merely serve to postpone needed restructuring for these systems.
Grants also tend to discourage full cost pricing by water systems. Water systems should be self
sustaining, relying solely on system revenues to fund ongoing operations and system improvements.
In some cases, a one time grant may be necessary to help fundamentally viable systems
finance a major capital expenditure to meet SDWA compliance needs. Care should be taken
however, to ascertain that the water system has adequate revenue sources to fund ongoing operations
and maintenance and future capital requirements, so that provision of a grant does not create a long
term dependency.
Loan programs have historically focused on smaller systems, in an attempt to compensate for
their limited access to the financial market. Loan programs can be a highly effective means of
leveraging limited Federal funds, especially when used in combination with eligibility criteria which
require State matching loans. Eligibility criteria requiring enterprise funds, capital planning, and full
cost pricing can also help to guarantee that systems obtaining loans practice sound management and
minimize the need for future assistance.
Some small systems will be unable to afford a loan at any interest rate, because they cannot
support the necessary water rates to repay the loan. Such systems are obvious candidates for
restructuring. Other small systems may find it difficult to acquire loan program assistance because
they must compete with larger systems for the assistance and small systems represent higher credit
risks because of their small revenue basis.
4.4.7 Training And^Technical Assistance
An administrative option available to assist small drinking water systems with SDWA
compliance is training and technical assistance (TA). Although it may not serve as a "stand alone"
solution to compliance problems, it can aid in developing long-term implementation strategies. A
common distinction between training and technical assistance is that training usually refers to
classroom instruction and technical assistance is viewed as on-site support.
Much of the training and TA provided by states and other organizations is reactive, focused
on correcting current problems. A proactive approach is more effective in that it can prevent
compliance problems from occurring. Each state's approach to supplying training and TA varies,
depending on resources and the state's philosophy.
Proactive TA strategies can have a significant impact on water system viability. Any
deficiencies found in visits to sites could be documented, and State officials could recommend ways
for systems to address specific deficiencies before they become compliance problems. Sanitary
surveys, conducted by many states, focus on technical performance of the wateir system. Again, with
the on-site presence, this service could be expanded to encompass preventive maintenance. A
"viability evaluation" could be conducted on other long-term factors such as system management,
adequacy of O&M, and financial status. TA programs could be directed towards supporting long
term sustainability and not to prop up non-viable systems.
100
-------
Operator certification provides another means to support water systems in developing the
capabilities needed for long-term compliance. In an April 1991 report, EPA found that 45 States
have some type of operator certification program; however, most of these States exempt small
systems to some extent. Certification can ensure that system operators have a basic knowledge of
SDWA regulations and water system engineering. In the few States that do require all small systems
to have certified operators, there has been increased use of contract O&M services by systems which
lack professional staff.
States could be required as a condition of primacy to implement an operator certification
program covering all community water systems, including small systems. Such a requirement would
not mean that all systems would need to have their own employees certified. Systems could contract
for the services of a certified operator from another system or private firm. Indeed, a major benefit
of requiring operator certification programs which cover small systems is that the requirement will
stimulate the development of a market for competitively priced operations and maintenance services.
EPA could be authorized to promulgate a regulation specifying the minimum requirements for
operator certification programs.
Training and Technical Assistance Providers
By presenting system owners and operators with different ways of looking at compliance
problems, training and TA providers can help systems find the least expensive, yet effective method
of operation. Providers of technical assistance and training offer valuable services and augment the
capabilities of state primacy agencies. Major providers and some of their services are described as
follows:
National Rural Water Association
One of the primary TA providers under the SDWA, NRWA provides assistance to small
drinking water systems in 48 states. Assistance is provided in the form of circuit riders that visit
facilities to lend hands-on support; workshops, seminars and training sessions; on-site technical
assistance, leak detection, treatment/testing procedures and O&M; and management, networking, and
financial assistance.
Rural Community Assistance Program
The RCAP network includes a national office and six regional offices with multi-State service
areas, and field-based staff and delegate agencies at State and local levels throughout the United
States. RCAP is dedicated to securing clean, safe drinking water for rural communities. The RCAP
concentrates efforts on financial assistance for small systems and provides TA for system
owners/operators and community leaders. Support includes training and assistance on small system
issues (e.g., rate setting, regulatory compliance, O&M); workshops and conferences; financing and
system management; and publications of training material, policy documents, books and field guides.
101
-------
National Drinking Water Clearinghouse
Established with funds from the Rural Development Administration (RDA), the goal of the
National Drinking Water Clearinghouse (NDWC) is to make information on drinking water issues
accessible to communities of less than 10,000 people. NDWC provides support through a toll-free
number for small systems; educational products including brochures, videotapes, and government
publications; a newsletter; computer bulletin boards; and various databases. The clearinghouse is
located at the University of West Virginia and is accessible via a toll-free phone number from
anywhere in the United States.
American Water Works Association
AWWA is a nonprofit, scientific, and educational association of 54,000 members in 43
sections or regional chapters in North America. AWWA's mission is to promote the health and
welfare of the public by improving the quality and quantity of drinking water. TA services and a
network of training strategies emphasize regulatory compliance. Specific services include workshops
and conferences; certification promotion and training; publication of scientific, educational and
technical information; a toll-free phone number for information; and training sessions.
State I09(b) Environmental Training Centers
Currently known as the Coalition of Environmental Training Centers, CETC is a network of
39 State centers which provide training and TA on drinking water system O&M and related topics.
The National Environmental Training Association develops training materials and serves as the major
professional organization for environmental trainers.
State Drinking Water Programs
State drinking water agencies are present in every state and provide numerous services and
information to water systems. These services include training on state regulations, operator training
and certification programs, emergency response assistance, disease outbreak surveillance, and
laboratory certification and referral services.
National Training Coalition
The National Training Coalition was developed by representatives of the Association of State
Drinking Water Administrators, American Water Works Association, National Rural Water
Association, Rural Community Assistance Corporation, Coalition of Environmental Training
Centers/National Environmental Training Association, and EPA. The coalition exists to formulate,
manage and execute a comprehensive national training strategy. The Coalition envisions a four part
strategy: 1) development of statewide training strategies; 2) curriculum development; 3) training the
trainer; and 4) delivery of training.
102
-------
SECTIONS
CAPACITY OF STATES TO IMPLEMENT DRINKING WATER REGULATIONS
The financial and technical ability of states to implement SDWA regulations is critical to the
success of the drinking water program. This section will examine the current status of federal and
state funding for State Public Water Supply Supervision (PWSS) programs. Options for improving
the ability of states to fund these programs, as well as the ability of state-run laboratories to perform
compliance testing, is also discussed.
5.0 Past and Current Funding Levels
There are many specific activities that a State PWSS program must perform. These include
enforcement, staff training, data management, sanitary surveys, and lab certification to name just a
few. State and Federal funding for drinking water programs has steadily increased since FY 1988.
In FY 1993, Federal grants for the Public Water System Supervision (PWSS) program had grown to
$59 million from $33.5 million in FY 1988, a 76 percent increase. State funding has grown to $83
million from $63 million during the same time period, a 32 percent increase. Federal and State
combined funding for State drinking water programs increased 47 percent from 1988 to 1993. This
increase, while substantial, has not kept up with State needs. EPA estimates that State funding needs
totalled $304 million in 1993, almost two and one-half times higher than the 1988 estimate of $123
million. The current shortfall is estimated to be $162 million, more than four times the estimated
1988 shortfall of $34 million.
The 1993 funding shortfall has been estimated using a Resource Needs Model developed by
EPA and ASDWA. The model is a significantly refined and enhanced version of the 1989 resource
needs survey previously used to estimate costs for direct implementation of the PWSS program in a
State. The general logic of this model is to take a comprehensive inventory of each activity
performed by a State drinking water agency, price each activity with assumptions that reflect national
aggregate conditions (States are able to override these assumptions with State-specific information),
and then add the cost of all activities to arrive at a drinking water agency resource needs total. The
model's structure and assumptions have been extensively reviewed by State and EPA personnel. The
ability to provide State specific information to the model is essential since State programs differ
widely; for example, some States perform monitoring for systems while other States do not.
5.0.1 The Significance of the Funding Shortfall
The current estimated State funding shortfall of $162 million is hampering State
implementation of the 1986 SDWA Amendments, as States are unable to meet EPA's schedule for
putting into effect regulations required by the Act. Four States, for example/missed the December
31, 1992 deadline for adopting the Surface Water Treatment and the Total Coliform Rules. (This
deadline included a two-year extension that States could apply for, in addition to the 18 months
allowed by the Act, to adopt regulations.) Delays in implementation are postponing health benefits
for millions of Americans.
103
-------
5.0.2 Reasons for Shortfall
Funds for State drinking water programs have fallen short for several reasons. First, the
State costs of implementing new SDWA regulations are high, and increases in Federal gram funding
of State drinking water programs have not kept pace with the increased cost of these regulations.
State implementation costs are estimated to have increased by more than 140 percent from FY88 to
FY93, while Federal funding of State programs increased 76 percent. Second, drinking water
programs compete for State funds with other high priority programs and issues, such as health care
and education. Finally, State budgets have experienced both anticipated and unexpected shortfalls.
Cqst Restrictions of New Regulations
Many States are finding it very difficult to obtain funds to implement new regulations. The
major implementation costs include training State employees, monitoring compliance by water
systems, taking enforcement actions against violators, and managing and submitting data required for
the Federal Reporting Data System (FRDS).
The cost of implementing the Lead and Copper Rule alone has forced several States to
consider returning to EPA primary authority, also known as primacy, for operating their Public Water
Supply Systems (PWSS) programs. Two States, California and Pennsylvania, formally told EPA that
without increased funding, they could not adopt the Lead and Copper Rule. (The California
Environmental Protection Agency estimates that implementing the rule would cost $5.8 million
annually, which is over $2.3 million more than the State's current Federal PWSS grant).
Although California and Pennsylvania were able to negotiate implementation schedules with
EPA, the fact that such negotiations were necessary is further evidence of the financial difficulties that
States face in implementing the 1986 SDWA Amendments.
Competition for Scarce State Resources
To obtain funds to implement new regulations, State drinking water programs must compete
with other programs and issues. In "Leader's Outlook," a 1992 report published by the National
Conference of State Legislatures (NCSL), State legislative leaders listed their top priorities. They
were a balanced budget, education, and health care. Environmental issues such as drinking water did
not come close to being top priorities with these officials.-
Drinking water programs may be located in environmental or health departments. Those
located in health departments are often at a particular disadvantage. Since key budget debates take
place within departments, funding for drinking water must compete with Medicare, Medicaid, AIDS
research, and primary health care services. Competition before the legislature can be just as fierce.
As one legislator recently explained: "People are dying from AIDS. When there are similar, tangible
public health risks from drinking water contamination, come back and see us."
Public health programs that focus on prevention tend to suffer in budgetary battles, especially
if the consequences of inaction will not appear for a long time and, when they do appear, will affect a
relatively small portion of the population. That does not make them any less important, but it does
make their funding more problematical.
104
j
-------
State Budget Shortfalls
Legislative priorities and State budget problems explain why State drinking water programs
cannot afford to implement all of the new regulations. In FY92, 31 States experienced budget
shortfalls. Since every State except Vermont requires by law a balanced budget, tax increases and
budget cuts were used widely to bring these budgets into line. According to the NCSL report, "State
Fiscal Outlook for 1993," 13 States cut their FY93 budgets in order to balance them. While
significantly healthier than FY 1991, the recent State fiscal problems and their attendant personnel
cuts have made it increasingly difficult for drinking water programs to secure additional funding.
Laboratory Capacity
The laboratory capacity issues where all compliance monitoring samples are analyzed in a
State laboratory are different from the laboratory capacity issues where the State has a laboratory
certification program and the compliance samples are mostly analyzed by local or private labs.
The Agency's observation is that many States which have traditionally conducted all
compliance monitoring may not be staffed sufficiently to carry out the increased monitoring
requirements for Phase II and Phase V contaminants. Compliance monitoring in the pre-1986 time
consisted mainly of analyzing for total coliform, turbidity, nitrates, heavy metals, total
trihalomethanes, and a few pesticides. States had developed the capacity to manage this effort. With
the promulgation of Phase I, States were required to monitor for VOCs. Although Phase I
monitoring required Gas Chromatography equipment similar to that required for total trihalomethanes,
the analysis was somewhat more sophisticated. This additional level of sophistication did not create
much hardship for the States. Phase II and Phase V require States to run at least-four analytical
methods and as many as seven in some cases to analyze for all regulated and unregulated
contaminants. The equipment needed could include Gas Chromatography-Mass Spectrometry, High
Pressure Liquid Chromatography, etc. The Lead/Copper rule and asbestos analyses may require the
purchase, installation, and operation of yet more expensive pieces of instrumentation such as an
Inductively Coupled Plasma/Mass Spectrometer and a Transmission Electron Microscope. Laboratory
capacity concerns include personnel, space, and instrumentation. With most States facing revenue
shortfalls, it is reasonable to expect that many States will have some difficulty in meeting all these
needs for increased monitoring.
States which do not analyze all drinking water compliance monitoring samples in their own
labs are required to have a laboratory certification program for certifying commercial laboratories.
These states may experience different kinds of capacity problems. There would be increased demand
for State laboratory auditors, as well as trained personnel to provide supervision of the State's
laboratory certification program. The States which are allowed to charge for certification would not
face economic hardship, but States which do not charge for certifying local laboratories may face an
economic hardship.
105
-------
5.1 Ways to Address State Resource Needs
EPA has tried to address the State capacity problem head-on. EPA has worked closely with
the States to address the problem of limited resources and to find other means of building State
capacity. With on-site technical support, EPA has been a partner with State agencies and other
organizations in forming coalitions to obtain increased State revenues and to push for enactment of
alternative financing mechanisms to support State primacy programs. EPA has also worked with the
States to develop the PWSS Program Priority Guidance (issued in 1992) that identifies the "baseline"
requirements and ranks the discretionary components of a State primacy program. This guidance is
designed to encourage efficient use of State resources by focussing on priority public health risks.
The guidance also specifies activities that must be carried out to maintain primacy.
Additional steps that could be taken to address the resource problem include:
Increase Federal Funding
Implementing the 1986 SDWA Amendments requires significant funds. EPA has requested
and received increases in the Federal PWSS grant over the past five years. In FY88, for example,
EPA received $33.5 million for grants to States; in FY93, PWSS grant funds totaled $59 million, 76
percent more. EPA will continue to support Federal funding as well as State funding efforts.
Increase State Funding
States can increase funding mainly through increased general fund appropriations and new or
increased user fees. Larger general fund appropriations are considered unlikely, given the current
budget climate in most States. Nevertheless, some States have recently increased general fund
appropriations to their drinking water programs. When budgets allow, States prefer using general
funds for drinking water programs since many legislators perceive safe drinking water as a public
good for all State residents.
The increasingly common method of generating funds for State drinking water programs is to
assess user fees. Several forms of fees are commonly employed. These include fees based on water
usage, number of connections, and population served. States also often charge fees for service and
use combinations of fees.
A water usage fee is based on the quantity of water used by each customer of a community
water system. It is straightforward, equitable, and acceptable to legislatures. Revenue predictions,
however, may be difficult at first if information on water usage is incomplete.
A connection fee is an annual levy on each service connection within a community water
system. This stable source of revenue is straightforward and easily passed through to customers in
their regular bills.
A population-based fee is determined by how many people are served by a community water
system. To reduce the administrative burden, systems are categorized by service population, and
each category is assigned a fee. Because each system in a category pays the same fee, per-capita
costs may vary significantly. Conceptually the simplest of the alternatives, the population-based fee
provides a stable source of revenue.
106
-------
A fee-for-service is based on the cost of services, such as sample analysis, plan reviews,
permits, and inspections, that community water systems receive from State drinking water programs.
It is equitable because every public water supply pays the actual costs of the services it receives.
Therefore, large systems dp not subsidize smaller ones, which generally require more services.
Smaller systems, however, may not be able to afford the fees and so may forego certain services.
Also, administrative costs are high. Finally, some State drinking water program activities cannot be
related to services received by systems.
A combination of fees may meet a variety of objectives. Combining fees may be more
equitable on a per-system and per- capita basis. Combined fees, however, are conceptually more
complex than are single-fee alternatives and are likely to entail higher administrative costs.
Raise the Minimum State Match
Another way to increase funding for State drinking water programs is to raise the matching
requirement for Federal PWSS grants in a reauthorization of the SDWA. Currently, States must
contribute an amount equal to 25 percent of the Federal grants they receive. (Many States, however,
contribute substantially more than 25 percent.) Raising the match to 50 percent would increase
program funding in States that currently provide matches of between 25 percent and 50 percent of the
Federal grant. A 50-percent match requirement would be consistent with the match for State
wastewater programs proposed in Clean Water Act reauthorization legislation.
An "anti-backsliding" provision in the reauthorization of the SDWA also would help to fund
State programs. Such a provision would require that States keep constant the percentage of their
contribution when Federal grants increase. Currently, States are not required to maintain or increase
their contribution when Federal grants increase, thereby reducing the percentage of their contribution.
A constant percentage requirement would ensure that the state contribution kept pace with the federal
increases in funding.
There is one caveat on increasing the State match. If, as explained above, some States cannot
meet their current obligations to their drinking water programs, increasing the State match might
simply decrease the number of States that retain primacy. If some State contribution is better than
none, increasing the State match may be counterproductive.
Funding Through an SDWA-Authorized Fee
An SDWA-authorized fee, borrowing on concepts found in the Clean Air Act permit fee
program, could be an effective approach for closing the State resource shortfall. Under this
approach, States would have flexibility to design their own mechanisms for funding State programs.
For States that need additional resources, the SDWA fee would become effective, and the fees would
be used by the State to fund primacy activities. If a State loses primacy, the fees would be used to
cover EPA's cost of implementing the SDWA that State. This approach provides the States and the
regulated community with an incentive to design their own financing mechanisms. The minimum fee
should be set at a level that covers the full cost of primacy programs, and States should have
flexibility to adjust the fee to meet their specific needs. Such a fee could be a valuable source of
funds. For example, an annual fee of $1.00 per customer, or 3 cents per 1000 gallons of water,
could raise enough (on average) to meet the total State budget shortfall.
107
-------
Reduce Requirements or Delete Parts of the Regulations that Require State Decisions
The funding shortfall also could be addressed by reducing the SDWA requirements and,
perhaps, deleting parts of the regulations that allow greater State flexibility in decision making. One
of the most labor-intensive options States have is to waive monitoring requirements for public water
supplies. To support a waiver, a State must determine that a supply is not vulnerable to
contamination. Such determinations may require hundreds of work-years of effort nationwide.
One should note, however, that eliminating State authority to waive monitoring requirements
will either place the burden for these decisions on the Federal government, or eliminate the waivers
altogether, thereby increasing the monitoring costs of individual public water systems. This tradeoff
between flexibility to tailor requirements in a way which reduces cost but which increases state
workload is one which arises at numerous other times in the implementation of drinking water
regulations (e.g., compositing samples, granting variances and exemptions).
Allow States Additional Time to Adopt Regulations
The difficulty States have in implementing the 1986 SDWA Amendments is evident in the
extent to which States have delayed adoption of regulations. Many States have asked for extensions
beyond the 18-month period to adopt regulations as stringent as the NPDWRs. EPA developed the
priority guidance in 1992 to allow States an additional five years to comply with all aspects of
drinking water regulations. Giving States even more time will not eliminate the cost of state
implementation. It will, however, stretch those costs out over a longer period of time and give State
programs more time to obtain increased funding.
Privatization
States could reduce drinking water budget shortfalls by privatizing some functions of their
functions. "Privatization" refers to the use of personnel other than State government employees to
conduct drinking water program tasks. These tasks or services could be provided through "Public-
Private Partnerships" between the State and other organizations such as investor owned utilities;
municipal utilities; consulting firms; or other organizations, such as health districts or planning
agencies.
Both Primacy and non-Primacy elements of State programs are candidates for public/private
partnerships. Primacy elements which might be privatized include sanitary surveys, water quality
compliance determinations, plan and specification review, training and communications, and
laboratory inspection. Non-primacy elements which might be privatized include source
distribution/protection, planning functions, and private well regulations.
The benefit to privatizing tasks is two-fold. First, it moves resources directly within the
private sector where technical capabilities may already exist or where they can be developed, thus
reducing governmental costs and redundancy. Second, it reduces the need to hire additional State
personnel at additional government costs to address new mandates or expand existing ones.
Privatization can help address the State funding shortfall if one of two conditions applies.
Either the State must delegate functions to entities that can provide services more cheaply than the
State can, or State delegation of functions must require utilities to pay the State for services. In the
latter case, for example, if a utility can hire the engineer of its choice to do a sanitary survey, it can
control the cost of the service. Privatization may make fees for service more palatable.
108
-------
Connecticut has established an advisory committee to look at the potential for privatizing
certain functions of the State primacy agencies. .Functions that are candidates for privatization include
sanitary surveys, water quality compliance determination, plan and specifications review for new or
modified water facilities, and laboratory functions. Privatizing these activities under the general
supervision of the State government might help Connecticut address its funding problem. Preliminary
estimates suggest that Connecticut could avoid over $700,000 per year in additional expenditures
through privatization. The question of whether small systems could afford services offered by
contracted providers is, again, a major concern.
5.2 States That Have Successfully Addressed Budget Shortfalls
In large part because of EPA's State Program Capacity Initiative, many States have increased
funding for their drinking water programs. However, these States have eased, but not solved, their
funding shortfalls.
The EPA Initiative bolsters State efforts to retain primacy and increase funding for drinking
water programs by addressing critical resource issues that affect State programs' ability to
successfully implement the 1986 SDWA Amendments. Programs that obtained additional funds
typically used three techniques to achieve success: they educated their constituency and undertook
public outreach efforts; they engaged in consensus building; and they mastered the political and
budgetary process. In FY92, EPA assisted 14 states to increase their resources by $18 million: eight
States passed user fees, five States passed appropriation increases, and one State reduced the level of
free water sample analysis provided and redistributed this funding to the drinking water program. So
far in 1993, 14 States have gained $15 million in additional program resources: one through an
appropriation increase and the other through user fee legislation. Also in FY93, 8 States have either
undertaken or expressed interest in building State capacity with support and involvement of EPA.
Several other States may initiate capacity initiatives in the near future.
Public education efforts stressing the importance of primacy and the need for additional
resources to retain primacy usually were directed toward utilities, legislators, and public interest
groups. These efforts increased the visibility of drinking water issues and encouraged broad-based
support. Successful programs also developed broad consensus within State government and among
interest groups. In some cases, consensus was achieved by creating an advisory committee comprised
of representatives from the water industry and key interest groups. Finally, the programs included
efforts to understand and accommodate the pressures affecting key actors such as agency heads, the
governor, and legislative leaders.
Advisory committees build consensus among all elements of the water supply industry and
interested organizations that drinking water programs face funding shortfalls and that those shortfalls
constitute a major public health concern. The success of these advisory committees is built on the
premise that, if the public is made aware of the value of their State's drinking water program, they
will support increased funding to allow the State to maintain and improve that program. Usually, the
advisory committee (developed jointly by EPA and ASDWA in 1993), agrees that legislation needs to
be drafted to remedy the situation and actively encourages States legislators to pass it. Through
advisory committees and other means, EPA is encouraging States to leverage their scarce resources
by building partnerships with organizations representing constituencies affected by drinking water
regulations. The objective is to obtain meaningful increases in state drinking water program resources
109
-------
session, the State developed a more aggressive campaign to generate support for a proposal to collect
$1.4 million through user fees. This legislation also failed, but by a narrower margin than in 1992.
Beginning in the fall of 1993, the State decided to establish an advisory committee of stakeholders
who will meet prior to the 1994 legislative session to debate the issue of primacy and how to fund the
program. The State hopes that this committee will facilitate agreement among all major stakeholders
before the issue is presented to the legislature for a third time.
If Idaho's experience is any guide, Maryland seems to be on the right track. One lesson
from the Idaho experience is that success comes to the persistent. A drinking water advisory
committee was created in Idaho in 1989 to advise the State on drinking water policy and funding
issues. In 1991, the committee and the drinking water program recommended funding an additional
16 positions. The governor recommended only funding 4 of the 16, and the legislature declined to
fond any. Undaunted, the advisory committee continued their efforts. In 1992, the legislature finally
approved the 16 positions that the advisory committee felt were essential to retain primacy. The
legislature provided funding for one year, and directed the drinking water program to develop a
proposal for long-term funding. In 1993, the advisory committee went to the legislature again. This
time, the legislature passed a user fee proposal that will fund the entire State share ($900,000 per
year) of the drinking water program's budget.
Many States have urged EPA to weaken the primacy requirements on the grounds that some
State program is better than none at all. EPA has resisted this approach. By adopting a tough stand
on minimum requirements, EPA has helped States justify increased funding in order to build stronger
PWSS programs and maintain the benefits of primacy.
112
-------
SECTION 6
COMPLIANCE WITH FEDERAL REGULATIONS
States generally have primary responsibility (primacy) for enforcement of national primary
drinking water regulations (NPDWRs) within their States. They review monitoring data submitted by
public water systems (or conduct the monitoring themselves) to determine compliance, provide
technical assistance, and are expected to take enforcement actions when necessary. EPA monitors
public water system compliance with NPDWRs by reviewing the violation data submitted by primacy
States (and by EPA for Wyoming; Washington, DC; and Indian Lands) to the Federal Reporting Data
System (FRDS). This information is reported on a quarterly basis. This section contains a brief
overview of the status of compliance with federal regulations and focuses primarily on community
water systems.
Compliance has two major components: submission of timely and complete monitoring data
and maintenance of water quality so that the levels of contaminants found in the drinking water are
either below the maximum contaminant levels (MCLs) set by federal regulation, or there is
compliance with treatment technique and filtration requirements. Since 1986, based on the available
data, the compliance rate for community water systems (CWSs) has remained between 70 and 73
percent.
While this is a good record, there were 16,294 (28 percent) CWSs, serving 63 million
persons, with violations in FY 1992. These systems incurred over 71,000 violations (over 63,000
monitoring/reporting [M/R] and 8,000 maximum contaminant level [MCL] violations.) The
following pages and the associated Exhibits provide some analysis of the characteristics of CWSs
which incurred violations and the types of violations.
Exhibit 6.1 shows the number of CWSs by system size that violated either the MCL or M/R
requirements in FY 1992. (See Section 1, Exhibit 0.2 for a description of system sizes.) As seen
from the chart, 14,510 or about 90 percent of the CWS in violation in FY 1992 were either small or
very small systems. While this may indicate a "small system problem," it is important to remember
that about 90 percent of the CWSs in the national inventory are small or very small; therefore, it is
logical to expect that a large number of the systems would be in violation.
It is also important to evaluate the percent of systems which had violations in each of the
different size categories. This is shown in Exhibit 6.2. As can be seen from the charts, 8 percent of
the small systems had MCL violations; however, 11 percent of the large and 6 percent of the very
large systems had MCL violations. In addition, 17 percent of the small systems had M/R violations,
while 18 percent of the large and 25 percent of the very large systems had M/R violations.
Therefore, the percent of small systems in violation is not dramatically different from the percent of
large systems in violation.
113
-------
CD
iti.
-Q
§
114
-------
CNJ
>
I
o
o
B " v-* ^ 5 R^ ff f& *' •" ^ s- ^ ' ••
• ^ f > tfj ffs$vf i-j-v ^
V)
(0
(O
"H
E
tn
10
0
SSMOIBIOJ.OI
SUOIJBIOIA u/im MUM SSMO }0 eBeiueojad
I
Percentage of Population Served by CWSs
with MCL Violations to Total Population
Served by CWSs by System Size
Ss
^ W
§1
H-i'ip
o jo
sJ
O)^
S
§
S H
0)
SSMO
suoijeioiA iov\i MUM SSMO jo eBeiuaoiad
ofi
li"s-
111
W £ ~
•£• • (0
o>15.S
E E >
0> CO —I
l|i
'&|S
-»r??
^>g
O w ^*
> eg o
c£l
10 >, E
O J3 o)
S>-o o
« g£
E S"5
i "^
^.S «
E.lsf
l|?
E S.i- .
«_^ri w
11-51
«o a -5.12
I-S-I5
§5 lo
•8^25
!«•£«
O <5 ^r *
|1||
5"! 1^
.22 g.-a
£ « w-g
15- Q 9
I -E I i
°8S«
S'^l'
bar
Jc feS 5
W Q. M w
l| S |
= P.s 1
i|2i
.= S^g m
^^*J5>
H|l
= S >..!
$ ~° ~ ~
viol
an
the
115
-------
In addition, when analyzing CWSs with violations, it is important to look at the population
affected by these violations. EPA's CWS inventory information indicates that while about 90 percent
of the CWSs are small or very small, these serve only 10 percent of the population, whereas the
remaining 90 percent of the population is served by about 10 percent of the CWSs. So, while the
majority of systems in violation are very small and small, the greatest number of persons are affected
by the relatively small number of large and very large systems in violation. For example, the small
and very small systems in violation affected approximately 6 million persons; however, the large
systems in violation affected approximately 25 million persons. Exhibit 6.2 also indicates the percent
of population served by category of systems which was affected by violations.
In analyzing compliance information, it is also important to evaluate which regulations are
violated most frequently or by the greatest number of CWSs. Exhibit 6.3 shows that the
microbiological (that is, the Total Coliform rule) monitoring and reporting regulations are by far the
most often violated. This is especially significant as microbiological regulations are the one of the
most important regulations of this program and compliance with these requirements is not expensive.
The frequency of violations of the Total Coliform rule is probably related in part to the monitoring
frequency (monthly) compared to other regulations which require quarterly or annual monitoring.
High rates of noncompliance are particular problems for systems in Alaska and Puerto Rico.
These systems are typically very small and face additional constraints that include cultural and
language barriers, transportation difficulties, and more limited available remedies than are available in
other States. Seventy-three percent of the CWSs in Alaska and 72 percent in Puerto Rico violated the
NPDWRs in FY 1992; of these systems with violations, 96 percent in Alaska and 86 percent in
Puerto Rico were very small or small systems.
EPA also monitors compliance for the nontransient noncommunity water systems
(NTNCWSs). In FY 1992, 82 percent, or 19,451 of the NTNCs were in full compliance with all
applicable NPDWRs. 18 percent or 4,489 NTNCs violated the regulations. As with the CWSs, the
most common violations were for microbiological monitoring and reporting. Also, as with CWSs,
the vast majority of the NTNCs in violation (greater than 99 percent) were very small or small;
however, the small and very small systems comprise 99 percent of the NTNC universe. Only 14
medium and 2 large NTNCs violated the NPDWRs in FY 1992.
Significant Non Compliance
Because there are so many violators of the NPDWRs, EPA and the States prioritize them for
enforcement actions. In general, the highest priority for formal enforcement actions are significant
noncompliers (SNCs). These are systems which have more serious, frequent, or persistent violations.
SNCs are divided into two categories: microbiological/turbidity (MIT) SNCs and
chemical/radiological SNCs.
In FY 1992, 5 percent of all CWSs were SNCs (Exhibit 6.4). However, 77 percent of the
SNCs were very small water systems serving 500 or fewer persons; only 9 percent of the SNCs
served more than 3,300 persons. This may indicate that while the larger systems do incur violations,
they are often able to correct those violations before becoming an SNC; very small systems often lack
the resources or the willingness to do so.
116
-------
Q.
il
O (B
CM
* CO C §
O T"
(suoiuiui ui) pewas uojiBjndod
c
1.1
o
o
c
1
= *•
o^"
5.3
o"
'•
-------
C
O
CO
1*"*
!5
0)
0)
E
O
O
118
-------
While the number of SNCs is relatively small, SNCs do represent the most serious violators
of federal regulations. In FY 1992, EPA and the States were able to address or return to compliance
in a "timely and appropriate manner" only 54 percent of the M/T SNCs and 42 percent of the
chemical and radiological SNGs. This low resolution rate is due to many factors, including the large
number of very small systems which often lack resources to comply and, significantly, other
institutional barriers, such as the lack of effective authorities on the State or federal level to deal with
some of these situations.
As this overview shows, there are many violations of the NPDWRs. Historically, States have
relied on training, technical assistance, and other informal means to bring systems back into
compliance; not formal enforcement. Exhibit 6.5 shows the number of Community Water Systems in
violation and the number of state formal enforcement actions over the past several years.
Federal enforcement activity has been steadily increasing since EPA received administrative
order authority in the 1986 SDWA amendments. This authority enabled EPA to "order" a public
water system to comply with the NPDWRS; it freed the Agency from the requirement to go to the
courts every time it needed to respond to a violation. However, EPA currently lacks the statutory
authority to make its enforcement program more efficient and more effective in dealing with the
compliance problems. The compliance situation is likely to grow worse as new regulations (for
example, the Surface Water Treatment Rule and the Lead and Copper regulation) are implemented.
119
-------
Exhibit 6.5
PWSS State Activity
Fiscal Years 1988 to 1992
^\ *- \, •& 1
or
MA
ME
NH
Rl
VT
-Heat-''
NJ
NY
PR
VI
00
DE
MD
PA
VA
WV
._ ^ ,-,,-;,
AL
FL
GA
KY
MS
NC
SC
TN
IL
IN
Ml
MN
OH
Wl
'0|«
6
7
o
6
o
0
; ^j$ - j
30
133
Q
7
>""179- '
0
8
3
54
9
22
*'^W"*
1
25
10
8
12
42
19
6
1
0
3
0
9
5
?&R
6
29
0
4
0
0
€%£<••
11
172
1
0
""^84"
0
0
2
81
5
5
:-5-^93
4
2
18
7
27
12
6
3
ft'tyQ."
10
0
1
0
3
6
ft'smt
ifert&Acffi
6
19
0
4
0
0
A ; ;-£9 '
11
308
12
0
n -SSI'"
0
0
129
92
10
7
^••"SW
11
76
23
20
35
23
9
3
'"" l&fr
21
0
57
4
2
16
1 ^'idfli*
*#••'
7
21
0
10
4
1
-'/413' -
13
276
27
77
>-',-39&/'
0
13
1
114
50
1
/'^'fTSP
6
95
24
23
20
86
11
32
4 M--
16
0
19
3
18
2
:5«.^.S&
'».»iv
34
0
0
0
7
9
-§Q-' I
34
106
107
213
-460 '
0
5
0
45
8
0
,< $&'-
6
27
0
10
19
92
1
3
-Vtw''
24
0
4
0
10
0
.M.XKv
' ' ' ( Sv
_1Q89
61
68
123
175
7
88
~"522,
179
415
329
35
" 958;''
0
23
55
680
461
163
^1,382;
166
1,318
404
221
487
611
171
114
3,492
523
139
312
179
385
343
'''....ifl&
bmfaMnt^
stems In ^
;-i^ao ;
65
41
84
303
8
95
1 596 -
238
501
335
46
,1,120
0
18
66 ,
733
394
159
1,370^
111
1,691
359
199
435
488
196
76
s'^,955
559
48
273
156
348
222
,,,r;MfiL,
/tolatiofi
•}\q& -
99
60
65
195
14
184
'®t? .
216
699
356
49
1,020 '
0
22
94
1,105
450
176
1U847
161
1,171
563
207
383
802
191
99
i$*S77"
490
214
173
171
368
256
siJSfc.
...!9S8...'
111
94
79
175
16
199
' $74 ^
235
555
319
18
4/C27 ;
1
27
59
786
373 I
178 I
1,434
118
803
216
180
258
700
470
117
&$& '
568
401
118
246
577
170
/*«*-;
120
-------
Exhibit 6.5
PWSS State Enforcement Activity: FY 1988 to 1992 - continued
Page 2 of 2
, State * ,,
AR
LA
NM
OK
TX
Beg 6
IA
KS
MO
NE
Reg 7
CO
MT
ND
SD
UT
WY
,.fteg8
AS
AZ
CA
GU
HI
MP
NV
PW
Begs
AK
ID
OR
WA
..ftefl.10-
- , Toial State Actions* -
'"" 196S ™te& ,'t990 1S91 "* %i992
0
0
7
12
51
-70
89
2
1
0
92
15
12
0
0
7
0
'- 34
0
5
48
1
0
0
6
0
60
8
15
3
19
45 ,
0
1
10
13
69
93
50
3
0
0
63
9
1
1
3
9
0
23\
0
12
70
0
1
0
2
0
85
11
14
27
38
90'
8
0
4
26
86
'-' 124 "
135
0
5
0
14CK
15
2
0
20
8
0
* 45
0
10
147
0
3
0
10
0
170
0
8
10
27
'< 45'
3
0
18
240
117
' 378 -:
150
16
34
21.
' aai -
126
41
0
14
6
0
"t87
0
14
86
0
1
0
1
0
i02-
7
11
35
37
" \ 90f
0
18
11
80
76
!1S5"~
221
2
37
28
288
3
0
2
14
4
0
' 28, ,
0
0
3
0
1
0
1
0
- §•
10
8
0
100
\tl18 '
CoHirnunBy Water , -
•. *• "•"> "• "•
Systems in Violation
1989
117
296
171
369
707
; 1,660
440
230
291
76
1,037
222
401
108
204
143
151
" t,229
29
420
472
0
29
0
128
0
'1,078
328
470
446
1,259
:" 2,503
•' 1S90,
125
311
160
349
587
1,532;
399
243
243
56
941
191
361
94
175
172
139
1^*132
27
381
279
0
25
4
143
0
859
378
703
407
1,357
2,845
1901 %
127
318
162
441
820
1,868
451
156
303
106
1,016
190
331
89
183
158
110
1^061
0
41
641
0
17
3
103
0
'-'80$-
530
401
358
1,126
£4ie
,1992.,
99
210
138
365
711
1,523
246
153
343
159
901
137
314
67
221
227
131
1,Q97,
0
593
542
0
18
0
106
0
1,259
526
414
475
1,662
3,077
15,742 15,556 16,198 16,024**
* State enforcement actions include bilateral compliance agreements (BCAs), State admin'rtrative orders (SAOs),
civil referrals (CRs), and criminal cases filed (CrFs). Prior to FY 1990, numbers of BCAs were not available.
** Note: This total does not include the 270 CWSs with FY 1992 violations on Indian Lands,
121
-------
SECTION?
PUBLIC WATER SYSTEM SUPERVISION (PWSS) INFORMATION MANAGEMENT
Two of EPA's major responsibilities under the SDWA are to set national standards for
drinking water and to ensure that states which have assumed primary enforcement responsibility are
carrying-out their responsibility. Both functions require significant amounts of information to
evaluate and report on progress. Historically standard setting and enforcement activities have been
supported by a two-tiered approach to information management. The first tier consists of data and
information systems handled by Primacy Agencies (States and Region 8 for Wyoming). The second
tier entails data and information systems managed by EPA.
Most Primacy Agencies use their own resources to develop and maintain reporting systems.
Many, but not all, of these reporting systems are automated. Automated data systems typically
contain detailed data to support non-compliance determinations and state-specific initiatives (e.g.,
operator certification programs, financing information, sanitary survey scheduling, operator training,
PWS expansion initiatives,-service line replacement programs, PWS capacity information, schedules
of various types, non-Federal monitoring requirements). Although State information management
systems often have common data elements and features, the type, scope, and quality of the data
maintained varies greatly between states.
States periodically report a subset of their inventories and exceptional events (variances,
exemptions, filtration determinations, violations and enforcement) to EPA (See Exhibit 7.1.). The
EPA regions are responsible for ensuring that all the required data from Primacy Agencies are entered
into EPA's national information system, the Federal Reporting Data System, (FRDS).
EPA's system, FRDS, became operational in federal fiscal year (FY) 1977. At that time,
EPA required Primacy States to submit summary information about each PWS in the nation annually.
Summary information included: PWS identification number, population served, sources of water,
treatments applied,,location of the PWSs, etc. Additionally, primacy agencies were required to
annually report data concerning violations of SDWA regulation, any SDWA-associated variance or
exemptions, and any enforcement actions taken against a PWS related to non-compliance with the
SDWA.
Six new major regulation packages, covering approximately 61 contaminants, have been
promulgated since the enactment of the 1986 amendments to the SDWA. By 1995, EPA expects to
promulgate regulations for a total of 112 contaminants. As each new rule is promulgated, States
attempt to modify their information management systems so that programmatic and compliance
activities associated with the new regulations can be tracked and reported to EPA. In most cases,
States making these system modifications have found the costs to be prohibitive. Many States have
not been able to make system modifications to accommodate new rules, because of significant State
fiscal difficulties (See Exhibit 7.2). As a result, 43 States have fallen behind in their ability to track
new rules and can not provide timely compliance information on newly regulated contaminants.
122
-------
Exhibit 7.1
CURRENT PWSS TWO TIERED INFORMATION MANAGEMENT
APPROACH
Summary
Monitoring/
Reporting
Data
HQ
FRDS
Report
Data
EPA
REGION
Tier 2
Monitoring
Reporting
Data
Analytical
Results
PWS
STATE
Report Data
Analytical Results
Samples
Sample Analytical
Result
LAB
Tier 1
123
-------
CO
0
CO
CO
O
to
D
0
CO
O
(0
OS
+-<
CO
"c
0)
O
TJ
0)
J=
O
CO
.
J= CM
-------
The inability to report this data is problematic, because EPA relies on States to submit current
and timely data to FRDS for use in EPA oversight activities. As States fall further behind in their
ability to track and report the status of new rules, EPA's ability to perform oversight has been equally
diminished. Under current conditions EPA cannot even accurately answer basic questions such as,
"Is the quality of drinking water getting better or worse?"
Recognizing that program management would continue to deteriorate further under this
current information management environment, EPA initiated the PWSS Information Systems
Modernization Project in June, 1992. The purpose of the effort is to develop a new information
strategy for the PWSS Program which will be implemented over the next ten years. A key
component of the strategy is the development of a new national information system which better
accommodates new programmatic requirements. Concepts that need to be incorporated into system
design are:
Eliminating, if possible, the need for primacy agencies to develop their own
data management systems to implement the PWSS program;
Reducing the unnecessary reporting burden on the part of the primacy
agencies;
Developing shared information system(s) which could be used by EPA, States,
laboratories, localities and the public;
Creating incentives for primacy agencies to adopt and participate in the new
system, while allowing non-participating primacy agencies to maintain their
own information systems ("translator" states) and reporting on a periodic basis
into the new PWSS system;
Allowing states to customize their implementation of the new PWSS system while
providing a framework for consistency and control;
Providing a capability for consistent non-compliance determinations;
Providing EPA a better tool to develop new regulations, evaluate existing regulations,
and perform oversight functions; and
Ensuring timely access to complete, representative data.
The success of this effort will require reaching consensus on system requirements by a large
number of disparate users. This includes not only drinking water program managers, but also
environmental and water quality resource managers who rely on drinking water program data to help
them implement program responsibilities such as developing ground-water protection and watershed
plans and targeting multi-media enforcement cases.
EPA is using the Information Engineering Methodology (IBM) to ensure that user needs are
adequately addressed during the system design process. IBM is a highly structured and rigorous
process for determining and analyzing user information needs. A broad yet comprehensive
125
-------
Information Strategy Plan is the product of the IBM. A key component of the Information Strategy
Plan (ISP) is a description of the individual components of the proposed information system. These
individual components are called business systems. Each business system is further delineated into
business areas. These business areas become the basic building blocks of the information system.
EPA held facilitated workshops through out 1992 to collect the information necessary to
develop the Information Strategy Plan (ISP). Participants included over 43 subject matter experts
from 27 organizations with a direct interest in drinking water data. -Participants included
representatives from 16 states and all ten EPA regions. Participants in the IEP concluded that:
EPA should proceed with the development of a new information system(s) to
satisfy national, regional, state and public sector information requirements;
The new system(s) should include eight business systems. These are:
Inventory, Field Surveillance^ Compliance, Water Resource Planning,
Regulation, Management and Budget, Disease Prevention and Assessment, and
Technical Assistance;
The new information system should be designed so that it can operate on
either state and/or national EPA computers;
EPA should continue to support "translator" states who desire to maintain
their own information systems, and report an equivalent set of data
periodically to EPA's new system;
The development of the new system should be done in phases. In the first phase,
EPA should develop a pilot for the PWS Inventory sub-system in a select set of states.
This pilot should be completed by November 1993; and ,
After successfully completing and evaluating the pilot effort, EPA should
commence the development of other systems including Field Surveillance
(which includes sampling) and Compliance (which includes violation
determination and enforcement action tracking).
The recommendations generated through the workshops were the basis for the approach described in
the Information Strategy Plan (ISP) completed in December 1992.
EPA has used the PC-based information engineering tool set named Information Engineering
Facility (IEF) to record the information collected throughout the Information Engineering
Methodology (IEM) process. The IEF enables EPA staff to create a series of information structures
that are used to design and implement an information system. Toward the end of the IEM, the PC-
tool is actually used to create the computer system, database, data entry screens and other analytical
modules. The IEF virtually eliminates the need for a computer programmer to write computer code.
Since the IEF also documents the computer code as it is generated, a separate system documentation
does NOT need to be written by EPA staff. This is critical, because creation of complete, consistent
and sufficiently detailed system documentation is frequently a problem in the development and
126
-------
maintenance of large automated systems. Use of the automated tool will save EPA time and money
as they develop the new system.
It is anticipated that these efforts to build the "core system" (satisfying the essential state and
EPA requirements) can be concluded by the end of 1995, if budgetary resources allow. The
estimated cost for the new system is approximately $4.5 million. When the core of the new system is
complete, the current system, FRDS, could be eliminated. At that point resources used for the
operation and maintenance of FRDS could be dedicated to the PWSS information systems
modernization to complete the remaining, yet important, pieces of the system. Completing this
portion of the new PWSS system would take an additional two to three years, if sufficient funds are
available.
The PWSS system modernization would result in the following benefits:
Significant economies of scale in system design, development, operation,
maintenance, training, because the new EPA/state PWSS information system would be
jointly owned, operated and maintained;
Decrease in the difficulty and costs associated with modifying the system as changes
to the Act or regulations are implemented due to changes in the technical design;
Relief for state PWSS ADP staff from the burden of information system design,
development, and support, enabling them to focus more on information analysis;
Significant improvements in consistency in the type and scope of data across primacy
agencies, and data quality because of common data definitions and formats;
Less difficulty in reporting and accessing data due to changes in system design thus
improving timeliness of State reporting and broadening the base of users;
More informed decision-making (e.g., regulation development) due to the
more consistent, complete, and timely data available;
Improved EPA capability to evaluate existing and proposed regulations due to better
data consistency, timeliness and quality; and
Increased facilitation of multi-media enforcement and cross-program integration,
because the system will reflect Agency and Federal data standards and will include
provisions for linkages to other Water Program and EPA information systems.
127
-------
-------
Appendix A
PROCESS FOR IDENTIFYING CONTAMINANTS IN
DRINKING WATER FOR REGULATION
-------
surface1.1.
The above-mentioned problems had limited impact on the EPA's performance in the past
occurrence characterization.
of fte Agency's current approach and discusses in broad fashion an alternate for
revisions to that approach.
jHIstnrical Bnses for
Selection
Th? original eighty-three contaminants
The 1974 amendments to the SDWA provided the Agency with broad discretion in the
selection of contaminants for drinking water regulation. Contaminants could be selected for
rSSn bS^mply upon the potential for causing an adverse health effect ,f present.
Demonstration of occurrence was not a required factor.
Nevertheless, the 1974 Amendments did require the Agency to consult with the National
Academv of Sciences (NAS) and the National Drinking Water Advisory Council (NDWAC) EPA
md cSfwiS Te NAS on the health effects of contaminants" and with the NDWAC for advice
on an approach to the standards setting process.
%or instance data obtained by the US Geological Survey in the Midwest under the National Ambient Water Quality
K«r=£:s£^
variable on spatial and temporal bases.
"The analogy is appropriate if one considers that most surveys project chemical occurrence in only 0.1 to 1.0 percent of
all systems based on measurement above a proximate detection limit.
"As discussed later in this section, a three tiered approach to contaminant evaluation was originally contemplated by the
Agency (see 48 FR 45506).
"National Academy of Sciences, "Drinking Water and Health," Volume I (1977), 11(1980), 111(1980), IV(1981.
2
I
-------
The consultations with the NDWAC resulted in a three tiered approach to the categorization
of studied contaminants for regulatory purposes:
Tier I- Those which occur with sufficient frequency and which are of sufficient concern to
warrant national regulation (MCLs) and consistent monitoring and reporting. Coliforms were
cited as a specific example of this category.
Tier II- Those which are of sufficient concern to warrant national regulation but which occur
at limited frequency, justifying flexible national minimum monitoring requirements to be
applied by State authorities. This category was envisioned to include radionuclides, certain
pesticides and some inorganics (e.g. barium) whose occurrence was perceived to be more
limited and generally predictable based upon geologic or other conditions. Reducing
unnecessary monitoring was a desired result of this second tier.
Tier III- Those which would not warrant development of a regulation but for which non-
regulatory health guidance could be provided to States or water systems. This tier was
proposed to be developed for those infrequently occurring contaminants associated with
isolated events where there might be a need for a short term "acceptable" level with respect to
consumption that would guide States or water systems vis-a-vis the need for immediate
control. Specific examples of contaminants in mis category were not provided.
This categorization and a proposed MCL development process were discussed in a series of public
meetings in 1982 and widely supported. In 1983, the Agency issued an Advanced Notice of Proposed
Rulemaking (48 FR 45506) requesting comment on the approach and also identifying the balance of
eighty-three contaminants to be evaluated under this process.
Prior to the Agency's completion of these studies, the SDWA was amended in 1986 to require
the development of Tier I type MCL's for all of the contaminants on the list. Authority for
substitution of up to seven contaminants for those on the list of eighty-three was also included in the
statute, but without the flexibility to pursue the Tier II and HI regulatory options.
In evaluating the eighty-three contaminants, the Agency considered four factors to be critical:
Are there sufficient health effects data upon which to base an MCLG?
Are there potential adverse health effects from exposure to the contaminant via
ingestion?
Does the contaminant occur in drinking water? Has the contaminant been detected in
significant frequencies, and in a widespread manner?14
14
Judged based on occurrence in the eight national drinking water surveys discussed earlier in this report.
3
-------
If data are limited on the frequency and nature of contamination, is there a significant
potential for drinking water contamination?15
Upon completion of the analysis, EPA did avail itself of all flexibility afforded the Agency. Seven of
the eighty-three contaminants lacking in a clear health effects basis at the time or for which there
were no adverse health effects at observed levels were replaced by other contaminants.
The Agency further requested comment on the need for regulation of an additional twenty-six
of the eighty-three contaminants (see Table 1). The comments received on the notice generally
favored the removal of many of these analytes, as well as the provision of greater flexibility for the
Agency to make the Tier II and Tier III type decisions for these and other contaminants. The
contaminants were not removed, however, since the statute did not provide discretion in this regard.
TABLE 1
CONTAMINANTS PROPOSED FOR
REMOVAL FROM THE MCL LIST BY
EPA IN THE 1987 FRN
Dichloromethane
Antimony
Endrin
Dalapon
Diquat
Endothall
Glyphosate+
Adipates
Standard plate
count
2,3,7,8-
TCDD(Dioxin)+
Trichlorobenzene
Legionella
Sulfate
Nickel
Thallium+
Beryllium
Cyanide
Vydate
1,1,2-
Trichloroethane
Simazine
PAH's
Atrazine
Phthalate+
Pichloram
Dinoseb+
Hexachlorocyclo-
pentacliene
+ These contaminants were not anticipated in drinking water in 1987 based on monitoring information. Subsequent analysis
has demonstrated that analytical methods problems would have hindered their detection. Special methods were developed to
ensure their detection at health levels.
Considered the following in order of decreasing importance: occurrence in private wells, presence in direct or indirect
additives, or occurrence in ambient surface or ground waters. Presence in solid or liquid wastes, persistence and mobility in
aquatic settings, use patterns, and production volumes also were considered collectively as indicator!! of the potential for
drinking water occurrence. It should be noted that the terms "significant frequency" and "significant potential" have never been
defined, but would need to be in order to implement the three Tier approach.
-------
As is detailed in other sections of this report, some of the regulated contaminants are not
expected to occur to a great extent in drinking water. The principal impact of such regulations will
be monitoring costs associated with demonstrating that the contaminants are not present. Extremely
tight deadlines and limited resources, as well as the great complexity of predicting chemical
occurrence, have prevented the Agency from developing substantive guidance on system vulnerability
to infrequently occurring contaminants. Such guidance would have facilitated the granting of
monitoring waivers, at considerable cost savings to small systems. The States, also overburdened and
resource limited, generally adopted a conservative approach to monitoring and granted few sampling
waivers.
Contaminant selection for the existing Drinking Water Priority List
The above discussion has attempted to characterize the events leading up to the regulation of
the original eighty-three contaminants. Other requirements in the 1986 Amendments to SDWA have
guided the subsequent contaminant selection process. Specifically, Section 1412 (b)(3) required the
Agency to identify additional contaminants with the potential to pose adverse health effects and which
were known or anticipated to occur in public water systems. The Administrator was to place the
contaminants on a Drinking Water Priority List (DWPL). The DWPL was to be sufficiently large to
ensure the development of at least twenty-five new MCL's every three years.
The Statute further influenced the contaminant selection process by specifying certain parties
to be consulted and lists to be evaluated in the development of the DWPL16. The Agency published
the first DWPL in the Federal Register on January 22, 1988 (53 FR 1892) and modified the list to
include additional contaminants on January 14, 1991 (56 FR 1470). The general criteria for inclusion
on the list were threefold:
• Occurrence of the substance in public water systems; or physical/chemical/environmental
characteristics and use patterns of the substance indicate the potential for occurrence in
public water systems at levels of concern.
• Documented or suspected adverse health effects of the contaminant.
• Availability of sufficient information on the substance, including health effects data,
analytical methods, and treatability studies, so that a regulation could be developed before
the statutory deadline.
A key point to note with respect to these criteria is the emphasis on the word "potential" in
the first bullet. The Agency's goal in developing the DWPL's was the creation of a sufficiently large
working list of chemicals to ensure the availability of twenty-five suitable candidates for statutorily
mandated regulation. As a consequence, there was a fairly broad interpretation of potential
*ln particular, substances referred to in section 101(14) of the Comprehensive Environmental Response, Compensation,
and Liability Act (CERCLA) and those registered as pesticides under the Federal Insecticide, Fungicide, and Rodenticide Act
(FIFRA).
-------
occurrence. This approach was necessary since actually demonstrating occurrence in drinking water
supplies could have been an insurmountable barrier within available resource levels17.
.Seventy-seven contaminants were ultimately placed upon the DWPL (see Table 2).
Substances were identified based on their occurrence in one of six groups:
The SARA Section 110 priority list which is composed of those substances EPA and
ATSDR have jointly determined to occur most frequently at hazardous waste sites18.
Volatile organic chemicals previously selected for unregulated contaminant monitoring
under Section 1445 of the Act.
Disinfectants and disinfectant by-products which are known to occur in drinking water
on a widespread basis.
Pesticides registered under FIFRA which were detected in the National Pesticide
Survey or which were design analytes in the survey and whose properties indicated
significant potential for groundwater contamination.
The seven contaminants in the original eighty-three for which substitutions had been
made.
Other substances identified through the National Inorganics and Radionuclides Survey
(NIRS) or those specifically requested by States.
While the above lists generate some strong candidates, they also have limitations vis-a-vis their use
for subsequent drinking water prioritization and standards development19.
17See the earlier discussion relating to the impacts of a lack of existing data, cost of monitoring (for contaminants other
than volatile organics), seasonality, low occurrence frequencies, etc. on the occurrence evaluation process.
"Only the first 100 chemicals ranked were considered in developing the DWPL list. The EPA/ATSDR work group has
subsequently evaluated more chemicals. To date, over 600 chemicals have now been evaluated arid priority rankings have been
established for 275 chemicals known to occur in or around Superfund sites.
"As stated earlier, this section is focussing on the prioritization of chemicals. Microbial prioritization poses entirely
different issues and is best deferred until completion of several key studies within the disinfection and byproducts program.
Throughout the balance of this discussion of chemical prioritization, it will be useful to consider chemicals to be divided into
three classes: inorganics, pesticides and other organics. This division flows naturally from differences in their use, disposal
and fate assessment.
6
-------
TABLE 2
CURRENT DRINKING WATER PRIORITY LIST
Aluminum
Boron
Chloramines
Chlorate
Chlorine
Chlorine Dioxide
Chlorite
Cyanogen Chloride
Hypochlorite ion
Manganese
Molybdenum
Strontium
Vanadium
Zinc
Asulam
Bentazon
Bromacil
Cyanazine
Cyromazine
DCPA
Dicamba
Ethylenethiourea
Formesafen
Lactofen/
Acifluorfen
Metal axyl
Methomyl
Metolachlor
Metribuzin
Parathion
Prometon
2,4,5-T
Thiodicarb
Trifluralin
Acrylonitrile
Bromobenzene
Bromochloro-
acetonitrile
Bromodi-
chloromethane
Bromoform
Bromomethane
Chlorination/Chloramination
byproducts
Chloroethane
Chloroform
Chloromethane
Chloropicin
o-Chlorotoluene
p-Chlorotoluene
Dibromoacetonitrile
Dibromochloro-
methane
Dibromomethane
Dichloroace-
tonitrile
1,3-Dichloro-
benzene
Dichlorodi-
fluoromethane
1,1-Dichloroethane
2,2-Dichloropropane
1,3-Dichloropropane
1, l-Dichloropropene
1,3-D5chloropropene
2,4-Dinitrophenol
2,4-Dinitrotoluene
2,6-Dinitroroluene
1,2-Diphenyl-
hydrazine
Fluorotrichloro-
methane
Hexachlorobutadiene
Hexachloroethane
Isophrone
Methyl ethyl ketone
Methyl isobutyl ketone
Methyl-t-butyl ether
Naphthalene
Nitrobenzene
1,1,1,2-Tetrachloroethane
1,1,2,2-Tetrachloroethane
Tetrahydrofuran
Trichloro-
acetonitrile
1,2,3-Trichloropropane
Cryptosporidium
Dealing first with the organic chemicals, use of the SARA priorities meant that many organics
were evaluated in terms of the recognized problems they posed at sites in the past (which may not
correlate with their potential for future harm). Because SARA deals with problems which are
localized in nature, it is also possible that these releases will have little to do with most public
drinking water intakes.
Contaminants ranked under the SARA program can be from an extremely limited number of
sites20 The problems recognized under the SARA program merely demonstrate that a chemical has
been released to groundwater from at least three sites. Other sites may pose a greater threat to public
sites.
example, the 1992 SARA list had Disulfoton and Hydrazine ranked 42nd and 51st, despite being found at only four
-------
drinking water supplies. The list is also biased toward chemicals which are easily measured in the
laboratory, although this is less true now that the prioritization has been extended to over 600
chemicals.
The unregulated,contaminant monitoring, while more current, suffers from limited chemical
coverage. It primarily provides information on volatile organic chemicals (which includes some of
the disinfection by-products). A major problem with these data is that it has taken years to get
information to the present state21. Data from the initial monitoring effort are just now beginning to
be available for use and analysis. Inability to achieve State consensus on data reporting formats and
the low priority assigned to reporting of the collected data relative to implementation and enforcement
of existing regulations seriously hampered efforts to get the data in usable form..
Nevertheless, the information contained in the unregulated contaminant data base suggest it
could be an extremely powerful tool for occurrence estimation. To illustrate this point, one need only
look at data relating to two chemicals which were ultimately covered under a recent Agency
rulemaking.
The first chemical is dichloromethane. The original regulatory impact estimates in the Phase
V rulemaking were derived from the NOMS and NSP data bases. For dichloromethane, it was
estimated that thirty-nine systems would have the chemical present in excess of the MCL.
Groundwater occurrence data from the unregulated contaminant data base, however, suggest that two
percent of all systems would have dichloromethane present in their supplies. Unfortunately, questions
about the statistical validity of the data base make projections risky.
This is not an anomalous result. The regulatory impact assessment projected that 1,2,4-
trichlorobenzene would not occur at levels of concern based on the Agency's earlier surveys. The
unregulated contaminant monitoring subsequently found it in one-half percent of all samples. This
level corresponds to approximately two hundred and fifty systems nationwide. However, the
statistical problem cited above make such projections questionable. Widespread occurrence of this
chemical has been further suggested by its occurrence in fifty-four percent of all sampled fish in the
National Study of Chemical Residues in Fish.
That both of these chemicals were seriously underestimated by the older surveys is not
surprising given the many variables involved. Many chemical releases to the environment tend to be
localized events. Exposure is a consequence of many variables that relate to the nature of the release,
a chemical's properties and (for ground waters) the vulnerability of the receiving source. The natural
consequence of this complexity is that even well designed surveys can fail to adequately estimate the
frequency of chemical occurrence. The Agency believes that these results clearly illustrate why
validation of perceived non-occurrence with real drinking water samples should always be performed
before making a determination of non-occurrence of any chemical.
As to future use of the unregulated contaminant monitoring, the Agency is examining ways to
improve upon the turnaround in the second phase of unregulated contaminant monitoring presently
underway. The use of these data is anticipated to play a significant role in new prioritization efforts.
21Data are now available on over a thousand drinking water supplies in over half of the States. Many industrialized States
are still missing.
8
-------
The third major data base used in establishing the DWPL arose from the mandate to consider
FIFRA regulated pesticides. The National Pesticide Survey (NFS) was jointly conducted by the
Offices' of Drinking Water and Pesticide Programs. It was utilized as the principal source of
pesticide contaminants for past priority lists. The Survey was considered very important for
characterizing pesticides occurrence since many of the pesticides are not measurable with routine,
broad spectrum analytical methods. Additional expense must be incurred to analyze for many of the
pesticides. As a consequence, most Federal environmental surveys conducted in the past had not even
looked for many pesticides.
That significant pesticide occurrence might exist was confirmed in that time period by the
discovery of major aldicarb contamination problems in private wells in Wisconsin and on Long
Island. To assess the significance of pesticide occurrence on a national basis, the Agency undertook
the NFS and conducted targeted sampling to assess the significance of occurrence in public water
supplies.
It was surprising when the NPS identified only very limited occurrence in community water
supplies. While this occurrence could suggest that pesticide contamination is rather limited, that
result would not be consistent with information collected by other Federal and State groups.
For example, the State of California conducted one time sampling of all public supplies in the
State. California contains approximately six percent of all of the Nation's community ground water
systems. Their survey of small water systems found eleven percent of all wells to be contaminated
with DBCP. Even if there were no other wells in the entire country contaminated with DBCP, NPS
results should have reported more contamination based on these results. Actual unregulated
contaminant monitoring of public drinking water supplies in thirty other States uncovered an
additional one-half percent occurrence. Consequently, in this particular case, the occurrence in
monitored public supplies was approximately three times that estimated by the NPS.
The Pesticides in Ground Water Database (PGWDB), a compilation of many State surveys of
private wells, monitoring wells and public supplies, found considerably more pesticide occurrence
(see Column 2 of Table 3). Unfortunately, the PGWDB studies vary widely in quality, and in some
cases include targeted studies that were assessing pesticide mobility. It is therefore difficult to use
them in national occurrence assessment efforts. At best, they merely serve to illustrate the point that
pesticides can contaminate a significant portion of vulnerable wells. The Agency is exploring ways to
further utilize data from portions of these valuable surveys in future drinking water contaminant
prioritization and regulatory efforts. We hope that the current round of unregulated contaminant
monitoring will shed further light on this subject. Regardless, the disparity in results illustrates the
limitations of relying solely upon any one survey for predicting pesticide occurrence.
The fundamental issue which this section has attempted to address is how the Agency has
prioritized drinking water contaminants in the recent past and the adequacy of those efforts. The
existing DWPL was developed based on an evaluation of the best information available at the time.
The many years typically involved in moving from environmental problem awareness to monitoring,
however, have produced a bias in prioritization and in chemical occurrence data bases towards
chemicals of historical interest.
This approach inevitably contributes to the omission of significant chemicals from lists and to
the inclusion of chemicals which are subsequently determined to be unlikely to occur in drinking
-------
water. The SDWA regulation of pesticides clearly illustrates this point. The PGWDB contains
information on over 60,000 monitoring and drinking water wells nationwide. Twenty-two pesticides
were found in 100 or more wells. Eleven are still unregulated under the SDWA, including four which
are ampng the largest volume pesticides22.
Even if the Agency had the perfect survey, it would not suffice for ensuring drinking water
quality in the future. The regulation of atrazine, for example, may reduce its use, but alternative
pesticides would likely fill the gap. Chemical prioritization must of necessity be a dynamic process in
order to achieve lasting improvement in water quality.
Several shortcomings of the existing prioritization scheme have been identified. Because
some chemicals have been seriously underestimated by existing approaches to occurrence estimation
and others appear to have been assigned much too high of a priority, the Agency has initiated a multi-
year effort to better prioritize chemicals.
TABLE 3
CROSS-SURVEY COMPARISON OF PESTICIDE
OCCURRENCE
(percent of sampled wells contaminated)
Contaminant
DBCP
EDB
Simizine
Atrazine
Alachlor
Bromocil
National
Pest.
Survey
0.4
ND
1.1
1.7
ND
ND
US
PGWDB
Results
9
14
2
5
1.8
1.8
CA
PGWDB
Results
26
2.8
6.5
5.2
0.2
3.2
CA Small
System
Survey
11
0.9
0.5
0.6
0.3
0.3
^Metolachlor, Telone, Cyanazine and Carbaryl-Although there are over four hundred, fifty active pesticide ingredients in
use in this country, the top twenty pesticides account for over half of all pesticide use. One hundred, seventeen pesticides were
found in at least one well (ninety-seven of these are presently unregulated under SDWA).
10
-------
Appendix B
COST INFORMATION FOR INDIVIDUAL SMALL SYSTEM BAT PROCESSES
-------
-------
BASIS FOR COSTS
Costs presented in this memorandum were compiled from the Very Small Systems
BAT Document (USEPA, 1992), Verification of Small System Costs Used for Cost and
Technology Documents (USEPA, 1987), and full-scale data on completed small system
projects. No new costs were developed for this assignment. All costs are presented in early-
1992 dollars. Costs contained in the Small Systems BAT Document were already in early-
1992 dollars. All other costs were updated using cost indices as presented in Table 3.
TABLE 3
COST INDICES FOR UPDATING SMALL SYSTEM COSTS
Cost
Capital
O&M
Index Reference
Engineering News Record
Department of Labor
Index
Construction Cost
Producer Price
Early-1992 Value
453.5
330.3
Capital costs are for equipment as specified in the individual process cost tables.
Equipment for these processes is the minimum required to provide treatment for the design
flow. Capital costs from the noted sources were updated as described above. If only
equipment costs were available, they were first updated, and then converted to capital costs
using cost factors as shown in Table 4.
Other equipment or infrastructure costs may be incurred on a site-specific basis.
These costs may include, but not be limited to wellhead remediation, buildings, land, and
costs to maintain the distribution system. These costs could be significant and will be
further discussed in Task 3 of this Work Assignment (to be completed at a later date).
The O&M costs are based on the average flow and include chemicals, replacement
materials, and power. Labor for operation and equipment maintenance is also a major
O&M cost component, but was not included in the process O&M costs because of the
variance in manpower available for small system treatment processes throughout the United
States. To better address this issue, labor costs were calculated based on three levels of
operator attention. These costs assume an hourly wage of $14.70 and 5 day/week operation
and are presented in Table 5.
0313-776
02/10/93
-------
" ^ . ^'\ . - ;** - TABLE 4 % - --
* w. ^ vv , < :< i *# ^ -: * * CAprrAL COST FACTORS " ,
"
i.
^^m^m——m—™—^**—*
Installation
2. Engineering
3.
4,
5.
Contractors Overhead & Profit
Legal, Fiscal, and Administrative
Sitework and Interconnecting Piping
PERCENTAGE Ok COST
see below
10 percent of total cost
12 percent of total cost
3 percent of total cost
6 percent of total cost
INSTALLATIONi
Lime Softening Systems
40%
Membrane Systems
25%
Ion Exchange Systems
Ozone Systems
50%
KMnO4 Feed Systems
10%
NaOH Feed Systems
30%
Chlorination Systems (both gas and NaOCl)
15%
All Others
30%
(1) Installation cost factor obtained from CWC Water modeL
-
0313-776
02/10/93
-------
; TABLES :
{ LABOR COSTS FOR SMALL SYSTEMS
. Type/ .f ;/_
Operation (hrs/day)
Maintenance (hrs/week)
Total Labor (hrs/year)
Annual Labor Cost
- Loir-
•v •••CwttXQOlB'"'* •-
0.5
1.0
182
$2,675
Medium,
Attention
1.0
2.0
364
$5,351
Hlg|t:$??':f '•••
AttClrtlMt :&•'' ••
2.0
4.0
728
$10,702
Upon determining the amount of manpower for each treatment process, the appropriate
labor cost can be determined from Table 5 and added to the O&M costs presented in the
process cost tables. A manufacturer's specific recommendation for labor are included in the
process cost tables. Labor requirements for coagulation/filtration, lime softening, and GAC
are discussed later in this memorandum.
Costs for electricity are based on an electric rate of $0.086/kwh. Chemical costs are
shown in Table 6.
The chemical costs used are shown in Table 6.
It is assumed that sufficient pressure from existing raw water pumping exists to feed
all processes with the exception of nanofiitration, reverse osmosis, and packed tower air
stripping. Similarly, sufficient pressure exists in the distribution system for backwashing.
Therefore, costs for feed, surface wash, and backwash pumps are not included unless
specified in the individual process equipment list.
COST TABLES
Table 7 through Table 27 present costs for the individual small system processes.
Capital costs are presented in total dollars and O&M costs are presented in cents per 1,000
gallons treated. Also presented in these tables are total production costs in cents per 1,000
gallons treated. The total production cost is the sum of the debt service on the capital cost,
amortized over a period of 20 years at 10 percent interest, and the O&M cost. Included
with the process costs are process descriptions, equipment lists, design assumptions, and
labor recommendations, if available.
0313-776
02/10/93
-------
$500
300
100
150
500
20.
SodaAih
250
540
500
Sodium Hypochlorite
190
Liquid Carbon Dioxide
350
1190
1520
230
410
140
Hydrochloric Acid
171
Powdered Activated Carbon
950
1900
1694
2800
909
400
105
1950
680
650
3200
590
490
Calcium HvDOCttlorite
2700
-------
TABLE 7
SMALL SYSTEM CHLORINATION COSTS
Design
Flow
(Kgpd)
14.4
24.0
87.0
270.0
Average
Flow
(Kgpd)
3.4
5.6
24.0
86.0
Total
Capital
(K$)
4.9
4.9
4.9
4.9
Total
O&M
(c/Kgal)
123.5
87.5
32.9
16.2
Total
Production
(c/Kgal)
170.3
115.9
39.5
18.0
Source
1
1
1
1
Notes:
1 =
2 =
3 -
ND
Very Small Systems BAT Document (USEPA,1992)
Verification of Small System Costs Used for Cost and Technology Documents
(USEPA, 1987)
Contractor's in-house projects
Data Not Available
The liquid sodium hypochlorite feed system was selected as the most appropriate small
system chlorination method because of its relative low cost and minimal safety requirements.
Capital costs are based on the following equipment:
• Manually-controlled diaphragm metering pump
• Fiberglass reinforced polyester (FRP) storage tank
• Pipes and Valves
O&M costs are based on a 5 mg/L dosage of 15 percent sodium hypochlorite, pumping
energy, and necessary maintenance materials.
0313-776
02/10/93
-------
TABLES
SMALL SYSTEM OZONATION COSTS
Ozone Concentration = 5 mg/L
Design
Flow
(Kgpd)
aSMMH
24
54
87
240
270
Average
Flow
(Kgpd)
masemma^ffs
5.6
145
24.0
76.0
86.0
Total
Capital
-------
TABLE 8
SMALL SYSTEM OZONATION COSTS
(Continued)
These costs are for ozonation systems using air as a feedgas. Systems that produce less
than 100 Ibs/day generally use air as a source because of its availability, safety, and cost
effectiveness. Capital costs for these air-generated ozonation systems includes the following
equipment: *
Air filter
Air compressor
Air cooler/dryer
Ozone generator
Ozone contactor
Ozone diffusers
Pipes and valves
Instrumentation and controls
Off-gas destruction unit
Costs were developed for ozone dosages of 5 mg/L and 1 mg/L, which are considered
adequate to provide the required inactivation and oxidation in surface and ground waters,
respectively. Original equipment manufacturers (OEMs) estimate required labor to be 1
hr/day.
0313-776
02/10/93
-------
TABLE 9
SMALL SYSTEM ULTRAVIOLET DISINFECTION COSTS
Design
Flow
(Kgpd)
43
14.4
24
87
270
Average
Flow
(Kgpd)
2.1
3.4
5.6
24.0
86.0
Total
Capital
(KS)
3
4
5
12
27
Total
O&M
(c/Kgal)
16
10
8
4
3
Total!
Production
(c/Kgiil)
62
48
38
19
13
Source
1
1
1
1
1
Notes:
1 « Very Small Systems BAT Document (USEPA4992)
2 s Verification of Small System Costs Used for Cost and Technology Documents
(USEPA, 1987)
3 3 Contractor's in-house projects
ND = Data Not Available
Ultraviolet (UV) disinfection systems consist of mercury lamps covered by a quartz
sleeve equally distributed in baffled tanks. Capital costs for these systems includes the
following equipment:
• UVunit
• UV intensity monitor
• Alarm system for system failure
• Flow control valve
These UV costs are additionally based on 85 percent transmission of radiation to the
microorganisms. The UV units are designed to provide a UV dosage of 30,000 /xwatts-
sec/cmj at a 253.7 nm wavelength after 8,000 hours of continuous operation. OEMs
estimate the required labor to be approximately 6 hours/year.
0313-776
02/10/93
-------
TABLE 10
SMALL SYSTEM POTASSIUM PERMANGANATE COSTS
Design
Flow
(Kgpd)
24
87
270
Avenge
Flow
(Kgpd)
5.6
24.0
86
Total
Capital
(K$)
5.2
5.2
5.2
Total
O&M
(c/Kgal)
7.6
3.1
1.2
Total
Production
(c/Kgal)
37.4
10.0
3.1
Source
1
1
1
Notes:
1 = Very Small Systems BAT Document (USEPA.1992)
2 = Verification of Small System Costs Used for Cost and Technology Documents
(USEPA, 1987)
3 =s Contractor's in-house projects
ND = Data Not Available
Potassium permanganate is available in crystal form and is added to water in a
solution. Capital costs for these potassium permanganate feed systems includes the
following equipment:
• Polyethylene storage tank (15 day)
• Metering pump
• Pipes and valves
• Instrumentation and controls
These potassium permanganate costs are additionally based on a 0.4 mg/L potassium
permanganate dosage. OEMs estimate the required labor for these systems to be 15-30
minutes per day.
0313-776
02/10/93
-------
TABLE 11
SMALL SYSTEM COAGULATION/FILTRATION COSTS
Design
Flow
(Kgpd)
24.0
26.0
68.0
87.0
166.0
270.0
500.0
Average
Flow
(Kgpd)
5.6
13.0
45.0
24.0
133.0
86.0
400.0
Total
Capital
(K$)
42.6
60.1
69.7
68.1
983
7013
294.6
Total
O&M
(c/Kgal)
102.7
ND
ND
50.7
ND
25.7
ND
Total
Production
(c/Kgal)
347.4
ND
ND
142.1
ND
89.4
ND
Source
1
2
2
1
2
1
2
Notes:
1
2
3 -
ND
Very Small Systems BAT Document (USEPA.1992)
Verification of Small System Costs Used for Cost and Technology Documents
(USEPA, 1987)
Contractor's in-house projects
Data Not Available
Small system coagulation/filtration installations are typically packaged-type plants
incorporating all of the required equipment integrated into a single factory-built, aluminum
alloy package. Capital costs for these systems are based on the following equipment:
Alum, polymer, and caustic storage and feed systems
In-line static mixing
Hydraulic flocculation
ciarifier with tube settlers
Dual-media filters
Backwash pumps
Pipes and valves
Instrumentation and controls
These coagulation/filtration costs are additionally based on the following design
assumptions:
• Filtration Rate = 25 gpm/ft2
• Alum dosage = 30 mg/L
• Polymer dosage = 0.4 mg/L
• Caustic dosage = 15.7 mg/L
0313-776
02/10/93
-------
TABLE 12
SMALL SYSTEM DIRECT FILTRATION COSTS
Design
Flow
(Kgpd)
14.4
28.8
72
108
144
216
288
Avenge
Flow
(Kgpd)
4.8
7.0
19.3
31.4
43.5
67.8
92.1
Total
Capital
(K$)
31
43
54
68
82
99
116
Total
O&M
(c/Kgal)
25
25
20
18
18
17
17
Total
Production
(c/Kgal)
231
221
110
88
79
64
57
Source
^SSSBSSKSK
1
1
1
1
1
1
1
Notes:
1
2
3 =
ND
Very Small Systems BAT Document (USEPA, 1992)
Verification of Small System Costs Used for Cost and Technology Documents
(USEPA, 1987)
Contractor's in-house projects
Data Not Available
Direct filtration includes all of the components of a coagulation/filtration except
clarification prior to filtration. Capital costs for these systems include the following
equipment:
Alum and polymer storage and feed systems
In-line static mixing
Hydraulic flocculation
Dual-media filters
Pipes and valves
Instrumentation and controls
These direct filtration costs are additionally based on the following design assumptions:
• Filtration Rate = 2.5 gpm/ft2
• Alum dosage » 10 mg/L
• Polymer dosage = 0.4 mg/L
OEMs estimate labor requirements for direct filtration systems to be 2 hrs/day.
0313-776
02/10/93
-------
TABLE 13
SMALL SYSTEM IN-LINE FILTRATION COSTS
Design
Flow
(Kgpd)
14.4
28.8
72
108
144
216
288
Average
Flow
(Kgpd)
4.8
7.0
193
31.4
435
67.8
92.1
— •••i
Total
Capital
(K$)
SSSS^^SSiSS
20
31
41
48
60
72
85
Total
O&M
(c/Kgal)
ssssssss^asss
22
22
18
17
17
17
16
Total
Production
(c/Kgal)
156
165
87
67
62
51
46
Source
1
1
1
1
1
1
1
Notes:
1
2
Very Small Systems BAT Document (USEPA.1992)
Verification of Small System Costs Used for Cost and Technology Documents
(USEPA, 1987)
3 » Contractor's in-house projects
ND = Data Not Available
In-line filtration is the simplest form of direct filtration, consisting of filters preceded
by chemical feed and rapid mixing. Capital costs for these systems include the following
equipment:
• Alum and polymer feed systems
• In-line static mixing
• Dual-media filters
• Pipes and valves
• Instrumentation and controls
These in-line filtration costs are additionally based on the following design assump-
tions:
• Filtration Rate - 25 gpm/ft*
• Alum dosage - 10 mg/L
• Polymer dosage = 0.4 mg/L
OEMs estimate labor requirements for direct filtration systems to be 2 hrs/day.
0313-776
02/10/93
-------
TABLE 14
SMALL SYSTEM SLOW SAND FILTRATION COSTS
Design
Flow
(Kgpd)
5.6
25.5
87.0
143.0
171.0
200.0
270.0
Average
Flow
(Kgpd)
1.5
6.0
24.0
43.0
51.0
60.0
86.0
Total
Capital
(K$)
SESSSOHBaOi
15.4
49.0
134.2
198.0
228.8
258.5
330.3
Total
O&M
(c/Kgal)
4.2
42
4.2
4.2
4.2
4.2
4.2
Total
Production
(c/Kgal)
334.6
267.0
184.2
1514
148.6
142.9
127.8
Source
1
1
1
1
1
1
1
Notes:
1 a Very Small Systems BAT Document (USEPA.1992)
2 = Verification of Small System Costs Used for Cost and Technology Documents
(USEPA, 1987)
3 = Contractor's in-house projects
ND = Data Not Available
Slow sand filtration uses a deep bed of sand to remove particles and microorganisms
from water. This filtration is improved by a gelatinous biological layer, schmutzdecke, on
top of the sand. Capital costs are based on the following equipment:
• Filter box with cover
• Pipes and Valves
• Turbidimeters and flow controllers
These slow sand filtration costs are additionally based on a filtration rate of 0.075 gpm/ft2
and removal of approximately one-inch of sand every 1.5 months. Sand removal requires
approximately 4.5 hours of labor per 1000 ft2 of filter surface area. This translates to 30
minutes of labor per removal for the smallest listed category to 11.25 hours for the largest
listed category.
0313-776
02/10/93
-------
TABLE 15
SMALL SYSTEM GREENSAND FILTRATION COSTS
Design
Flow
(Kgpd)
24.0
87.0
270.0
Average
Flow
(Kgpd)
5.6
24.0
86.0
Total
Capital
(K$)
15
41
123
Total
O&M
(c/Kgal)
82
60
50
Total
Production
(c/Kgai)
171
115
96
Source
1
1
1
Notes:
1
2
3 «
ND
Very Small Systems BAT Document (USEPA,1992)
Verification of Small System Costs Used for Cost and Technology Documents
(USEPA, 1987)
Contractor's in-house projects
Data Not Available
Greensand filtration uses a bed of manganese greensand to remove iron and
manganese and tastes and odors from water. Capital costs are based on the following
equipment:
• Filter tank
• Greensand media
• Pipes and Valves
• Instrumentation and controls
These greensand filtration costs are additionally based on a filtration rate of approximately
3 gpm/ft1. OEMs estimate labor requirements for this process to be 2 hrs/day.
Potassium permanganate is typically added ahead of greensand filtration for oxidation of
iron and manganese, and for regeneration of the greensand media. Costs for this process
are discussed in Table 10 of this memorandum.
0313-776
02/10/93
-------
TABLE 16
SMALL SYSTEM DIATOMACEOUS EARTH FILTRATION COSTS
Design
Flow
(Kgpd)
2.9
24
87
270
Average
Flow
(Kgpd)
1.5
5.6
24.0
86.0
Total
Capital
(K$)
4.6
13.6
42.6
137.9
Total
O&M
(c/Kgal)
36.2
35.2
34.1
33.9
Total
Production
(c/Kgal)
134.9
113.5
9L2
855
Source
1
1
1
1
Notes:
1
2
3 -
ND
Very Small Systems BAT Document (USEPA,1992)
Verification of Small System Costs Used for Cost and Technology Documents
(USEPA, 1987)
Contractor's in-house projects
Data Not Available
Diatomaceous earth (DE) filtration uses a thin layer, or precoat of DE supported by
a septum or filter element to remove particles and microorganisms from water. This thin
layer must be supplemented by a continuous body feed of DE to maintain the porosity of
the filter cake. Capital costs are based on the following equipment:
• Skid-mounted FRF and stainless steel filter housing
• Goretex teflon membrane
• Pre-coat feed system
• Body feed system
• High pressure sprayer to facilitate filter cleaning
• Pipes and Valves
• Instrumentation and controls
These DE filtration costs are additionally based on a filtration rate of 1 gpm/ft2. OEMs
estimate labor requirements for this process to be 0.5 hrs/day.
0313-776
02/10/93
-------
TABLE 17
SMALL SYSTEM INDUSTRIAL-SIZED CARTRIDGE FILTER COSTS
Design
Flow
(Kgpd)
24
87
270
Average
Flow
(Kgpd)
5.6
24.0
86
Total
Capital
(K$)
5
9
20
Total
'O&M
(c/Kgal)
99
82
72
Total
Production
(c/Kgal)
128
95
80
Source
1
1
1
Notes:
1
2
3 =
ND
Very Small Systems BAT Document (USEPA.1992)
Verification of Small System Costs Used for Cost and Technology Documents
(USEPA, 1987)
Contractor's in-house projects
Data Not Available
Industrial-sized cartridge filters are typically cartridges contained in filter housing.
These cartridges are replaced when a specified headless is reached, typically every 1 to 6
months. Backwashable cartridges are also available, but not used for small-system
applications because of elevated initial costs. Capital costs for small-system cartridge filter
applications are based on the following equipment:
• Filter cartridges and housing
• Pipes and Valves
• Pressure gauges and controls
These cartridge filtration costs are additionally based on a filtration rate of 20 gpm per
cartridge and a cartridge life of 3 months. Labor requirements for this is minimal.
0313-776
02/10/93
-------
TABLE 18
SMALL SYSTEM MICROFILTRATION COSTS
Design
Flow
(Kgpd)
Average
Flow
(Kgpd)
Total
Capital
(K$)
Total
O&M
(c/Kgal)
Total
Production
(c/Kgal)
Source
Notes:
1= Very Small Systems BAT Document (USEPA,1992)
2 = Verification of Small System Costs Used for Cost and Technology Documents
(USEPA, 1987)
3 = Contractor's in-house projects
ND = Data Not Available
Capital costs for small-system microffltration (MF) are based on the following
equipment:
Spiral-wound, polyester MF membranes
Painted FRP membrane housing
Raw water feed pump
Cleaning pumps
Schedule 40 PVC pipes and valves
Instrumentation and controls
These MF costs are additionally based on the following:
• Recovery rate
• Operating Pressure
• Membrane life
98.5 percent
5-20 psi
5 years
OEMs recommend 8 hrs/week labor for systems up to 86,000 god design flow and 40
hrs/week for systems with 270,000 gpd design flow.
Removal of biological contaminants without chemical addition is one of the major
advantages of MF ove? conventional processes. Using MF provides a consistent low
Sty product water without producing a chemical sludge residual However, a coagulant
cTnte added to enhance solids removal while the use of PAC with MF can provide effective
removal of dissolved organics.
0313-776
02/10/93
-------
TABLE 19
SMALL SYSTEM ULTRAFILTRATTON COSTS
Design
Floir
(Kgpd)
Average
Flow
(Kgpd)
Total
Capital
(K$)
Total
O&M
(c/Kgal)
Total
Production
(c/Kgal)
Source
Notes:
1
2
3 «
ND
Very Small Systems BAT Document (USEPAJ992)
Verification of Small System Costs Used for Cost and Technology Documents
(USEPA, 1987) ** ocumems
Contractor's in-house projects
Data Not Available
Capital costs for small-system ultraffltration (UF) are based on the following
equipment: 6
Spiral-wound, polysulfone UF membranes
Fainted FRP membrane housing
Raw water feed pump
Cartridge pre-filters
Cleaning pumps
Schedule 40 PVC pipes and valves
Instrumentation and controls
These UF costs are additionally based on the following:
• Recovery Rate
• Operating Pressure
• Membrane Life
• Molecular Weight Cutoff
85 percent
40 psi
2 years
10,000
OEMs recommends a labor requirement of 30 min/day.
Removal of biological contaminants without chemical addition is one of the major
advantages of UF over conventional processes. Using UF provides a consistent low turbidity
product water without producing a chemical sludge residual. However, a coagulant can be
added to enhance solids removal while the use of PAC with UF can provide effective
removal of dissolved organics.
0313-776
02/10/93
-------
TABLE 20
SMALL SYSTEM NANOFTLTRATION COSTS
Design
Flow
(Kgpd)
24.0
87.0
270.0
Average
Flow
(Kgpd)
6.0
24.0
86.0
Total
Capital
-------
TABLE 20
SMALL SYSTEM NANOFILTRATION COSTS
(Continued)
Some NF applications require pre- or post-treatment of water. Pretreatment processes
may be required to remove turbidity and color in order to maintain membrane capacity.
Availablepretreatmentprocesaes include coagulation/filtration, slow sand filtration, MF.and
UF. Post-treatment may be required to remove excess carbon dioxide from product water.
This can be accomplished with an air-stripping technology. Costs for these processes can
be determined and then added to the above costs.
Costs for concentrate disposal can be high; difficult to estimate - based on site specific
factors. Currently practical techniques for concentrate disposal include land application
(commingling with wastewater effluent in a reuse plan), surface water discharge and deep
well injection. Concentrations are currently classified as industrial waste and can be difficult
to permit for surface water discharges because of toxkaty issues. Deep well injection is only
viable in those parts of the country where there is an acceptable geology aquifer zone to
accept the waste without threat of upcoming.
0313-776
02/10/93
-------
TABLE 21
SMALL SYSTEM REVERSE OSMOSIS COSTS
Design
How
(Kgpd)
9.0
24.0
60.0
87.0
100
170
225
270
300
450
500
550
600
930
Average
Flow
(Kgpd)
2.0
6.0
ND
24.0
ND
95.0
92.0
86.0
220
380
210
300
400
701
Total
Capital
(K$)
54.2
85.9
135.3
1953
205.2
193.9
303.1
383.5
307.0
484.7
522.4
397.6
1,939.1
1,959.6
Total
O&M
(c/Kgal)
407.4
357.0
ND
220.2
ND
376.6
142.4
164.4
170.0
121.5
162.7
287.5
446.1
277.6
Total
Production
(c/Kgal)
1,2795
817.7
ND
482.1
ND
4423
248.4
307.9
214.9
162.6
242.8
330.1
602.1
366.4
Source
1
1
2
1
2
2
2
1
2
2
2
2
2
2
Notes:
1 = Very Small Systems BAT Document (USEPA,1992)
2 - Verification of Small System Costs Used for Cost and Technology Documents
(USEPA 1987)
3 = Contractor's in-house projects
ND = Data Not Available
Reverse Osmosis (RO) concentrates dissolved inorganics that may impose recovery
limitations. Therefore, costing of RO systems is dependent of several site-specific water
quality parameters. Costs provided in this table are based on nationwide median water
quality values and gives a general indication of expected costs.
0313-776
02/10/93
-------
TABLE 21
SMALL SYSTEM REVERSE OSMOSIS COSTS
(Continued)
Capital costs for small-system RO are based on the following equipment:
Thin film composite RO membranes
Painted FRF membrane housing
Raw water feed pumps
Cartridge pre-filters
Scale inhibitor, acid, and caustic feed systems
Schedule 40 PVC low pressure piping and valves
Stainless steel high pressure piping and valves
Instrumentation and controls
These RO costs are additionally based on the following:
Two-stage, reject-concentrating configuration
Recovery Rate • 75 percent
Operating Pressure » 200-400 psi
Membrane Life = 4 years
Molecular Weight Cutoff * <200
OEMs recommend a labor requirement of 1-5 hrs/day.
Some RO applications require pre- or post-treatment of water. Pretreatment processes
may be required to remove turbidity and color in order to maintain membrane capacity.
Available pretreatment processes include coagulation/filtration, slow sand filtration, MF, and
UF. Post-treatment may be required to remove excess carbon dioxide from product water.
This can be accomplished with an air-stripping technology. Costs for these processes caii
be determined and then added to the above costs.
Costs for concentrate disposal can be high; difficult to estimate - based on site specific
factors. Currently practical techniques for concentrate disposal include surface water
discharge and deep well injection. Concentrations are currently classified as industrial waste
and can be difficult to permit for surface water discharges because of tenacity issues. Deep
well injection is only viable in those parts of the Country where there is an acceptable
geology aquifer zone to accept the waste without threat of upcoming.
0313-776
02/10/93
-------
TABLE 22
SMALL SYSTEM GRANULAR ACTIVATED CARBON COSTS
Design
Flow
(Kgpd)
24.0
87.0
270.0
Average
Flow
(Kgpd)
6.0
24.0
86.0
Total
Capital
(KS)
42.6
110.7
170.3
Total
O&M
(c/Kgal)
370.5
295.8
242.6
Total
Production
(c/Kgal)
598.8
444.2
306.3
Source
1
1
1
Notes:
1 = Very Small Systems BAT Document (USEPA.1992)
2 = Verification of Small System Costs Used for Cost and Technology Documents
(USEPA, 1987)
3 = Contractor's in-house projects
ND = Data Not Available
Capital costs for small-system GAC applications are based on the following equipment:
• Carbon steel pressure GAC contactors
• Virgin GAC
• Pipes and Valves
• Instrumentation and controls
These GAC costs are additionally based on the following:
• Empty bed contact
time (EBCT)
• GACbedlife
• GAC replacement
25 minutes
90 days
Sl/lb
OEMs estimate labor requirements to be 1 hr/day.
0313-776
02/10/93
-------
TABLE 23
SMALL SYSTEM PACKED TOWER AIR STRIPPING COSTS
Packed Height = 12 ft, Air/Water Ratio = 20:1
Design
Flow
(Kgpd)
5.0
24.0
87.0
270.0
Average
Flow
(Kgpd)
2.0
6.0
24.0
86.0
Total
Capital
(K$)
22.1
22.1
32.4
61.3
Total
O&M
(c/Kgal)
142.6
50.9
29.7
133
Total
Production
(c/Kgal)
498.8
169.7
73.1
36.2
Source
1
1
1
1
Packed Height = 12 ft, Air/Water Ratio = 300:1
Design
Flow
(Kgpd)
5.0
24.0
87.0
270.0
Average
Flow
(Kgpd)
2.0
6.0
24.0
86.0
Total
Capital
(K$)
22.1
22.1
32.4
613
Total
O&M
(c/Kgal)
219.6
78.4
52.2
25.8
Total
Production
(c/Kgal)
57'5.8
197.2
95.6
48.7
Source
1
1
1
1
Packed Height = 40 ft, Air/Water Ration = 20:1
Design
Flow
(Kgpd)
5.0
24.0
87.0
270.0
Average
Flow
(Kgpd)
2.0
6.0
24.0
86.0
Total
Capital
(K$)
32.4
32.4
74.9
105.6
Total
O&M
(c/Kgal)
142.6
50.9
29.7
13.3
Total
Production
(c/Kgal)
663.2
224.5
130.2
52.8
Source
1
1
1
1
0313-776
02/10/93
-------
TABLE 23
SMALL SYSTEM PACKED TOWER AIR STRIPPING COSTS
(continued)
Packed Height = 40 ft, Air/Water Ratio = 300:1
Design
Flow
(Kgpd)
5.0
24.0
87.0
270.0
Average
Flow
(Kgpd)
2.0
6.0
24.0
86.0
Total
Capital
(K$)
32.4
32.4
74.9
105.6
Total
O&M
(c/Kgal)
219.6
78.4
525
25.8
Total
Production
(c/Kgal)
740.2
252.0
152.6
65.3
Source
1
1
1
1
Notes:
1 = Very Small Systems BAT Document (USEPA,1992)
2 = Verification of Small System Costs Used for Cost and Technology Documents
(USEPA, 1987)
3 = Contractor's in-house projects
ND = Data Not Available
Costs for packed tower air stripping facilities is highly dependent on packed tower
height and air/water ratio. Although a 40 ft packed height and an air/water ratio of up to
300:1 are designated BAT criteria, shorter towers with lower air/water ratios may be
sufficient for treatment needs. Therefore, costs are also included for a packed height of 12
ft and an air/water ratio of 20:1. Together, ail of these costs give a range of expected costs
for this process.
Capital costs for packed tower facilities include the following equipment:
• FRP tower structure
• Packing media
« Air blower
• Feed pump
• Equalization basin (located beneath tower)
• Pipes and valves
• Instrumentation and controls
0313-776
02/10/93
-------
TABLE 23
SMALL SYSTEM PACKED TOWER AIR STRIPPING COSTS
(continued)
OEMs estimate O&M requirements to be 1 hr/day.
Chemical pretreatment with a sequestering agent may be required to prevent iron and
manganese and/or calcium from oxidizing to insoluble forms and precipitating onto the
packing media or in the distribution system. If necessary, costs for this pretreatment should
be determined and added to the costs in this table.
These costs do not include off-gas treatment to meet possible state or local restrictions
on air emissions, nor do they include costs for disinfection.
0313-776
02/10/93
-------
TABLE 24
SMALL SYSTEM DIFFUSED AIR STRIPPING COSTS
VOC Removal
Design
Flow
(Kgpd)
24.0
87.0
270.0
Average
Flow
(Kgpd)
6.0
24.0
86.0
Total
Capital
(K$)
10.7
21.5
43.9
Total
O&M
(c/Kgal)
308.2
208.4
193.8
Total
Production
(c/Kgal)
365.8
237.2
210.2
Source
1
1
1
Radon Removal
Design
Flow
(Kgpd)
24.0
87.0
270.0
Average
Flow
(Kgpd)
6.0
24.0
86.0
Total
Capital
(K$)
10.7
21.1
23.5
Total
O&M
(c/Kgal)
251.8
185.2
107.6
Total
Production
(c/Kgal)
309.2
213.5
116.4
Source
1
1
1
Notes:
1 =
2 a
3 -
ND
Very Small Systems BAT Document (USEPA.1992)
Verification of Small System Costs Used for Cost and Technology Documents
(USEPA, 1987)
Contractor's in-house projects
Data Not Available
Costs contained in this table represent the packaged diffused aeration unit. Costs for
diffused air stripping facilities were developed for removal of VOCs, based upon an
air/water ratio of 25:1, and removal of radon, based upon an air/water ratio of 20:1.
Capital costs for these packaged units include the following equipment:
• High density polyethylene aeration basin
• Blower and diffusers
* PVC pipes and valves
• Instrumentation and controls
OEMs estimate O&M requirements to be 1 hr/day.
0313-776
02/10/93
-------
TABLE 25
SMALL SYSTEM ACTIVATED ALUMINA COSTS
Design
Flow
(Kgpd)
24
87
270
650
Average
Flow
(Kgpd)
5.6
24
86
230
Total
Capital
(K$)
5.9
16.2
46.0
107.6
Total
O&M
(c/Kgal)
227.0
146.2
116.4
102.2
Total
Production
(c/Kgal)
260.8
167.9
133.6
117.3
Source
1
1
1
1
Notes:
1 = Very Small Systems BAT Document (USEPA.1992)
2 = Verification of Small System Costs Used for Cost and Technology Documents
(USEPA, 1987)
3 = Contractor's in-house projects
ND - Data Not Available
Capital costs for small-system activated alumina applications with sulfuric acid/caustic
soda regeneration are based on the following equipment:
Alumina column
Acid feed system
Caustic feed system
Pipes and Valves
Instrumentation and controls
These activated alumina costs are additionally based on the following:
EBCT
Regeneration frequency
Regenerant concentration
5 minutes
4.5 days
2,000 mg/L
OEMs estimate labor requirements to be 4 hr/day.
0313-776
02/10/93
-------
TABLE 26
SMALL SYSTEM ION EXCHANGE COSTS
Anton Exchange
Design
Flow
(Kgpd)
24
87
270
650
Average
Flow
(Kgpd)
5.6
24.0
86.0
230
Total
Capital
-------
TABLE 26
SMALL SYSTEM ION EXCHANGE COSTS
. (Continued)
Costs are developed for anionic and cationic exchange. In each case regeneration of
exchange resin is accomplished with a salt solution. Capital costs for both of these types of
ion exchangers include the following equipment:
Pressure ion exchange unit
Brine dilution tank
Brine pump
Salt Storage tank
Pipes and valves
Instrumentation and controls
Costs are additionally based on an EBCT of 2.5 minutes and daily regeneration of resin.
The costs do not include the cost for disposal of regenerate flow. Possible options for this
disposal include direct discharge, sanitary sewer discharge, deep well injection, and
evaporation ponds. Because regenerate from ion exchange units treating water containing
toxic substances may contain high concentrations of these toxic substances, these
wastestreams may have to be chemically-precipitated and settled before direct or sanitary
sewer discharge. Sludge from the chemical precipitation would then be disposed in sanitary
or hazardous waste landfill, depending on the level of toxicity.
0313-776
02/10/93
-------
MANPOWER NEEDS AND OVERSIGHT
Coagulation/Filtration
Small system coagulation/filtration processes are typically available in packaged-type
plants, incorporating all of the required equipment into one unit These package plants are
designed to operate automatically. Required operator attention is typically 1 hour per day.
Two keys for successful operation of these package plants are maintaining proper
chemical dosages and ensuring proper operation of filter backwashing. Other elements of
proper operation are indigenous to other small system processes and include pump, piping,
and valve maintenance, equipment cleaning, and record keeping.
Proper chemical addition is key for water conditioning in order to maximize the
particle removal effectiveness of the clarification and filtration processes. This element is
also important for operation of direct and in-line filtration processes; however, because
coagulation/filtration is chosen for applications requiring greater particle removals, the
chemical dosages are typically greater, resulting in a larger margin for error. Chemical feed
systems typically consist of long-term chemical storage, batch mixing, and feed pumps.
Maintaining proper chemical dosages include making the required batch solution and setting
the feed pumps to deliver the proper amount of batch solution to the water. These
functions must be carried out by the operator.
Proper chemical dosages are initially set by the vendor after installation of the package
plant. Minor adjustments to these dosages may be necessary as the raw water quality
fluctuates. In a best case scenario, the operator performs a jar test to reestablish the
optimal chemical dosages as the raw water quality fluctuates. However, most systems do
not have the necessary facilities to perform jar tests. Therefore, any "on-line" adjustments
are made based on the filtered water quality.
Another important element for successful operation of coagulation/filtration processes
is filter backwashing. Proper filter backwashing is necessary to clean the filter in order to
maintain filtered water quality. Because washwater used for backwashing typically comes
from finished water storage which may deprive the volume of water needed for the
distribution system, excess backwashing reduces the cost effectiveness of the process.
Coagulation/filtration package plants typically have automatic backwashing either on a
timed cycle, or activated when filter headloss or filtered water turbidity reaches a designated
level. As with chemical dosages, this cycle is initially set by the vendor after installation.
0313-776
02/10/93
-------
An important consideration for determining the feasibility of installing a small GAC
system is the availability of replacement GAC. As mentioned above, GAC replacement
must be contracted through OEMs in order to avoid costly on-site regeneration costs.
However, for this alternative to be cost effective, the OEM must also be able to economical-
ly ship the GAC. For this reason, most OEMs will specify a minimum delivery shipment
of GAC. One OEM contacted would deliver a minimum quantity of one-half truckload, or
10,000 pounds. This amount of GAC could recharge a plant with an operating capacity of
approximately 70,000 gpd. Plants smaller than this would have to consider on-site storage,
which could be costly, or negotiate independently with the OEM.
In order to successfully operate a small GAC system, the operator should have a high
school education with some mechanical aptitude. The chief operator or superintendent for
these systems is typically a contracted operator overseeing several installations, the county
engineer, or a consulting engineer. This person should be knowledgeable of basic adsorption
principles in order to be able to determine when to replace the GAC.
Operator training is typically provided by the OEM. During this period, plant
operating parameters are set and the operator is acquainted with operations of the plant.
The operator is also provided with an O&M manual containing all of the points discussed
during training. Some OEMs also provide annual refresher training as part of GAC
replacement contracts.
Lime Softening
Three keys for successful operation of lime softening plants are maintaining proper
chemical dosages, ensuring proper operation of filter backwashing, successful transfer of
lime to the process stream, and finished water pH adjustment. Other elements of proper
operation are similar to other small system processes and include pump, piping, and valve
maintenance, equipment cleaning, and, record keeping.
Proper chemical addition is key for water conditioning in order to maximize
contaminant removal. This element is also important for operation of direct and in-line
filtration processes; however, because lime dosages used in lime softening are typically
greater than other chemical dosages in these other processes, the margin for error is
greater. In addition, overdosage can result in a layer of lime being deposited on the
surfaces of tanks and pipes. This can result in pipe clogging, clarifier mechanism jamming,
and shorter filter runs. Small lime softening systems typically dissolve previously-slaked
0313-776 02/10/93
-------
(hydrated) lime in water to form a slurry which is then metered into the process stream.
The lime feed system typically consists of a dry lime hopper, slurry tank with high-speed
mixer, and a metering pump. Maintaining proper lime dosage includes emptying bags of
lime into the hopper and setting the hopper feed to deliver the proper amount of lime to
the water. The proper slurry feed rate is initially set by the vendor after installation of the
package plant Minor adjustments to these feed rate may be necessary as the raw water
quality and/or lime slurry feed system fluctuates. These adjustments are typically made in
order to produce a desired pH in the mixing chamber or the contact darifier.
Another important element for successful operation of lime softening is filter
backwashing. Proper filter backwashing is necessary to dean the filter in order to maintain
filtered water quality. Because washwater used for backwashing is typically from filtered
water storage, excess backwashing reduces the cost effectiveness of the process. As with
coagulation/filtration package plants, small lime softening systems typically have automatic
backwashing either on a timed cyde, or activated when filter headless or filtered water
turbidity reaches a designated level As with chemical dosages, this cyde is initially set by
the vendor after installation. It is necessary for the operator to assure this automatic
backwashing cyde maintains the filtered water quality while preserving the length of the
filter run.
Because most small systems use groundwater as a source, raw water quality should not
significantly vary from day to day. However, quality may slightly vary with each well.
Therefore, adjustments to lime feed or filter backwashing should be minimal
The final key element for operating a small lime softening system is the successful
transfer of lime to the feed stream. As mentioned above, lime is typically fed into the
process stream as a slurry. Because high concentrations of lime tend to form deposits of
lime over treatment surfaces, it follows that the make-up, addition, and dispersion of the
lime slurry must be done with care to prevent dogging of the feed equipment or any
downstream process. Operators of lime softening plants have reported breakdowns in slurry
tanks, slurry feed pumps, and darifier mechanisms from excess lime accumulation.
Similarly, filter media can cement together, rendering the filter useless. For this reason,
small lime softening systems typically require more preventative maintenance to wash away
lime accumulations before they become troublesome. Providing this maintenance can be
burdensome to small systems that cannot afford to dedicate an operator solely to the
treatment facility.
0313-776 02/10/93
-------
In order to perform to successfully operate a small lime softening system, an operator
should have the following:
• A working knowledge of pump hydraulics and proper maintenance of pumps, piping,
and valves.
• A basic knowledge of filter operations.
These skills can be typically provided by a high school education and certification as a water
treatment plant operator by the local regulatory agency.
Three days of operator training is typically provided by the package plant OEM.
During this period, chemical feed rates and filter backwashing strategy are set, and the
operator is acquainted with the operations and maintenance of the plant The operator is
also provided with an O&M manual containing all of the points discussed during operator
training. After training, the local representative of the manufacturer is available to answer
periodic questions or may even provide a service contract for routine inspection and/or
operation of the facilities. The cost of this contract would depend on the location of the
plant and the level of service required. Other sources of information for the operator
include the municipal/utility engineer or an operator of a larger lime softening plant in the
area.
REFERENCES
U.S. Environmental Protection Agency (1987). Verification of Small System Costs Used for
Cost and Technology Documents.
U.S. Environmental Protection Agency (1992). Very Small Systems BAT Document.
0313-776
02/10/93
-------
|