Water Pollution: Poor Quality Assurance and Limited Pollutant Coverage
Undermine EPA's Control of Toxic Substances (Chapter Report, 02/17/94,
GAO/PEMD-94-9).
The Environmental Protection Agency (EPA) controls toxic pollutant
discharges into waterways through seven "core" activities spread across
several programs. GAO found that some of the information used to
analytically support these activities may be of doubtful quality,
raising questions about the effectiveness of these activities in
controlling toxic pollutants. GAO also found that the current permit
process does not limit the vast majority of toxics being discharged from
the nation's factories and sewage treatment plants. Although most of
these toxicants are "nonpriority" pollutants, they do pose risks to both
human health and aquatic life. GAO tried to examine the implications of
uncontrolled pollution cases identified in the facility sample
population, but the majority of cases could not be evaluated because of
a lack of criteria for assessing the health risks posed by the
discharges. This is so for most toxicants discharged across the nation.
--------------------------- Indexing Terms -----------------------------
REPORTNUM: PEMD-94-9
TITLE: Water Pollution: Poor Quality Assurance and Limited
Pollutant Coverage Undermine EPA's Control of Toxic
Substances
DATE: 02/17/94
SUBJECT: Quality assurance
Water pollution
Water pollution control
Wastewater management
Toxic substances
Environmental policies
Monitoring
Risk management
Data bases
Health hazards
IDENTIFIER: Superfund Program
EPA National Pollutant Discharge Elimination System
EPA Permit Compliance System
EPA Water Data Storage and Retrieval System
EPA Toxics Release Inventory Program
EPA Great Lakes Water Quality Initiative
NPDES
**************************************************************************
* This file contains an ASCII representation of the text of a GAO *
* report. Delineations within the text indicating chapter titles, *
* headings, and bullets are preserved. Major divisions and subdivisions *
* of the text, such as Chapters, Sections, and Appendixes, are *
* identified by double and single lines. The numbers on the right end *
* of these lines indicate the position of each of the subsections in the *
* document outline. These numbers do NOT correspond with the page *
* numbers of the printed product. *
* *
* No attempt has been made to display graphic images, although figure *
* captions are reproduced. Tables are included, but may not resemble *
* those in the printed version. *
* *
* A printed copy of this report may be obtained from the GAO Document *
* Distribution Facility by calling (202) 512-6000, by faxing your *
* request to (301) 258-4066, or by writing to P.O. Box 6015, *
* Gaithersburg, MD 20884-6015. We are unable to accept electronic orders *
* for printed documents at this time. *
**************************************************************************
Cover
================================================================ COVER
Report to the Chairman, Subcommittee on Superfund, Ocean, and Water
Protection, Committee on Environment and Public Works, U.S. Senate
February 1994
WATER POLLUTION - POOR QUALITY
ASSURANCE AND LIMITED POLLUTANT
COVERAGE UNDERMINE EPA'S CONTROL
OF TOXIC SUBSTANCES
GAO/PEMD-94-9
Water Pollution
Abbreviations
=============================================================== ABBREV
ASTM - American Society of Testing and Materials
BAT - Best available technology
BOD - Biochemical oxygen demand
EPA - Environmental Protection Agency
GAO - General Accounting Office
ICS - Individual control strategy
NPDES - National Pollutant Discharge Elimination System
PCS - Permit Compliance System
POTW - Publicly owned treatment works
QA/QC - Quality assurance/quality control
STORET - STOrage and RETrieval
TMDL - Total maximum daily load
TRI - Toxic Release Inventory
USGS - U.S. Geological Survey
WET - Whole effluent toxicity
Letter
=============================================================== LETTER
B-249852
February 17, 1994
The Honorable Frank R. Lautenberg
Chairman, Subcommittee on Superfund, Ocean,
and Water Protection
Committee on the Environment and Public Works
United States Senate
Dear Mr. Chairman:
In a January 6, 1992, letter, you asked us to evaluate the quality of
the information EPA uses to control the discharge of toxic pollutants
to surface water and the extent to which EPA controls these
substances. This report presents our findings on these issues.
In our evaluation, we identified the EPA activities that have a
direct effect on controlling the release of toxics to surface water
and the information requirements for these activities. We also
evaluated the extent to which EPA takes steps to ensure that the
information used to support these activities is of acceptable
quality. In addition, we addressed the extent to which toxic
pollutants are controlled through the NPDES permit process and
examined the risk implications of discharged pollutants that are
uncontrolled.
If you have any questions or would like additional information,
please call me at (202) 512-2900 or Kwai-Cheung Chan, Director of
Program Evaluation in Physical Systems Areas, at (202) 512-3092.
Other major contributors to this report are listed in appendix III.
Sincerely yours,
Eleanor Chelimsky
Assistant Comptroller General
EXECUTIVE SUMMARY
============================================================ Chapter 0
PURPOSE
---------------------------------------------------------- Chapter 0:1
Toxic pollutants threaten the biological integrity of the nation's
water. State officials say that toxic pollutants affect the quality
of more than 28,000 miles of the nation's rivers. Aquatic life has
been affected and human health risk has increased. Contamination by
toxic and conventional pollutants has resulted in fishing advisories
or bans on more than 1,000 bodies of water. Pollutants discharged
from the nation's factories and sewage treatment plants are a major
cause of this water and fish contamination.
To protect surface water from pollutants discharged from industrial
facilities and sewage treatment plants, the Congress passed
legislation in 1972 commonly referred to as the Clean Water Act. The
Environmental Protection Agency (EPA) is responsible for setting
national standards on the types and amounts of pollutants that
industries and sewage treatment plants may discharge. EPA and the
states enforce the standards by incorporating toxic limits on
facilities' discharge permits. Many factors influence EPA's ability
to implement programs authorized under the Clean Water Act. One
critical variable is the quality of available information. EPA's
program activities are analytical efforts, resulting in decisions
about which toxic pollutants to control and at what levels. If its
information is not of high quality, toxic pollutant control
activities are weakened in consequence.
The Chairman of the Subcommittee on Superfund, Ocean, and Water
Protection of the Senate Committee on the Environment and Public
Works asked GAO to evaluate the quality of the information EPA uses
to control the discharge of toxics to surface water and the extent to
which the EPA process controls toxic substances.
GAO (1) identified EPA activities that directly affect the control of
the toxic discharges and the information requirements of these
activities, (2) determined EPA's steps to ensure that the information
it uses to support these activities is of acceptable quality, (3)
addressed the extent to which permits control toxic pollutants, and
(4) examined the risk implications of uncontrolled discharged
pollutants.
BACKGROUND
---------------------------------------------------------- Chapter 0:2
For a number of years, EPA has been faced with reports by GAO and
others that individual activities it administers to control toxic
substances in surface water are producing information of suspect
quality. Information about three specific types of entities--the
environment, the facilities that discharge toxic pollutants, and the
toxic substances themselves--is at the heart of any system for
controlling toxics. Yet the quality of that information across EPA's
approach as a whole has not been evaluated.
To identify the critical activities that control the release of toxic
substances into waterways, GAO submitted to EPA an extended list from
the literature and agency documents. EPA identified activities that
are directly related to controlling discharges of toxic substances
into surface waters. A panel of experts reviewed and validated EPA's
list. GAO then interviewed managers for each of the principal
activities to identify information requirements as well as the
principal sources of that information. Subsequently, GAO compiled
data quality assurance criteria from the scientific literature and
from EPA's quality assurance guidelines, assigning the relevant,
critical quality assurance criteria to each information requirement
for each EPA activity. (A criterion was considered critical if its
absence, in and of itself, would compromise the quality of the
information required by an activity.) GAO then reinterviewed managers
and examined data to estimate the extent to which the relevant
quality assurance criteria were followed in the design, collection,
and analysis of the information. Finally, GAO obtained discharge
permit files for 236 industrial facilities and compared a list of
their toxic limits with information from three sources about the
facilities' actual discharges.
RESULTS IN BRIEF
---------------------------------------------------------- Chapter 0:3
EPA implements the control of toxic pollutant discharges into
waterways through 7 "core" activities spread across a number of
programs. GAO identified 13 types of information that are required
to analytically support these 7 activities and found that 5 of the 7
fell short in implementing the quality assurance steps that are
needed to produce accurate information. This raises questions not
only about the quality of the information these 5 activities produced
but also about their effectiveness within a strategy for controlling
toxic pollutants.
GAO also found that the current permit process does not limit the
vast majority of toxics being discharged from the nation's
facilities. Although most of these toxicants are "nonpriority"
pollutants, they can and do pose human health and aquatic life risks
(for example, ammonia and xylene).
GAO attempted to examine the risk implications of uncontrolled
pollution cases identified in the facility sample population, but the
majority of cases could not be evaluated because the criteria were
lacking by which to assess whether discharges posed a human health or
aquatic life risk. This is so for most toxicants discharged across
the nation.
PRINCIPAL FINDINGS
---------------------------------------------------------- Chapter 0:4
"CORE" PROGRAMS AND THEIR
CRITICAL INFORMATION
REQUIREMENTS
-------------------------------------------------------- Chapter 0:4.1
GAO initially identified 46 EPA water quality protection activities
and then refined this list to the 7 that are directly related to the
control of toxic pollutants. Taken together, these activities
determine the toxic concentrations that are safe for human health and
aquatic life, establish allowable discharge limits, identify impaired
waterways, and monitor compliance infractions. Thirteen types of
information are required to support these 7 core activities. (See
figure 1.)
Figure 1: Thirteen Types of
Information A-M Required Within
Activities 1-7
(See figure in printed
edition.)
ADHERENCE TO QUALITY
ASSURANCE
-------------------------------------------------------- Chapter 0:4.2
Two of the 7 activities (effluent guidelines development and water
quality criteria development) have been exceptional in meeting
quality assurance requirements, and confidence in the integrity of
the information they have produced is thus relatively high.
Collectively, these 2 activities have met approximately 93 percent of
their related quality assurance requirements. However, the 5 other
activities fell short of the criteria GAO used, meeting only 58
percent of their requirements. These activities (impaired water and
point source identification, total maximum daily load development,
water-quality-based permit issuance, local limits development, and
compliance problem identification) are critical to the success of a
standards-based toxics control program.
EXTENT OF CONTROL OF TOXIC
POLLUTANTS
-------------------------------------------------------- Chapter 0:4.3
GAO reviewed the pollutants discharged from 236 facilities from three
industrial sectors (pulp and paper, pharmaceutical, and pesticides
manufacturing). The vast majority of the toxic pollutant cases GAO
identified (77 percent) were not controlled through the permit
process. Priority pollutants (for example, dichlorobenzene and
carbon tetrachloride) made up 33 percent of these cases while the
remainder were nonpriority pollutants. GAO found that, for the
sample of facilities evaluated, the permit process rarely controlled
nonpriority toxic pollutants (with the exception of acids, ammonia,
and chlorine) and priority pollutants only partially. It should not
be presumed that the former are innocuous: nonpriority pollutants
can pose human health and aquatic life risks. In calculating the
percentage of nonpriority pollutants reported in EPA's Toxic Release
Inventory for 1989, GAO found that that group makes up 98 percent of
the discharges by mass. These pollutants are reported in the
inventory because they are used in commerce and because their release
to the environment could pose human health risks.
THE RISK IMPLICATIONS OF
UNCONTROLLED TOXIC
POLLUTANTS
-------------------------------------------------------- Chapter 0:4.4
GAO used standard modeling methods to estimate the water
concentration of uncontrolled toxic substances from the sample
population of facilities mentioned above to determine what proportion
presented human health or aquatic life risks. For cases that GAO
could evaluate, only 8 percent did pose such risks. However, GAO
could not evaluate many of the cases (62 percent) because the
criteria needed for determining whether their discharge posed a risk
had not been established.
MATTER FOR CONGRESSIONAL
CONSIDERATION
---------------------------------------------------------- Chapter 0:5
The Clean Water Act emphasizes a standards approach for controlling
toxic pollutants. In its reauthorization of the act, the Congress
should consider augmenting this approach with additional authority to
allow EPA to emphasize pollution prevention as a way of managing
toxic pollutant discharges.
RECOMMENDATIONS
---------------------------------------------------------- Chapter 0:6
To improve EPA's efforts to control discharges of toxic water
pollutants, GAO recommends that the Administrator of EPA
initiate immediate efforts to address the information quality
assurance problems GAO identified in the 5 toxic control
activities where these problems occur and
expand the use of the Toxic Release Inventory data base to identify
nonpriority pollutants being discharged to water that should be
considered for control through the permit process.
AGENCY COMMENTS
---------------------------------------------------------- Chapter 0:7
GAO discussed the results of its work, including tentative findings,
conclusions, and recommendations, with responsible EPA officials and
has incorporated their comments where appropriate. Their major
comments and GAO's responses are included in chapter 6.
INTRODUCTION
============================================================ Chapter 1
Toxic pollutants pose a substantial threat to the biological
integrity of the nation's water. Reports prepared by the states
under section 304(l) of the Clean Water Act cite toxic discharges
from 879 factories and sewage treatment plants around the nation as
causing streams to violate water quality standards.\1 State
governments have also identified toxic pollutants as affecting the
quality of more than 28,000 miles of the nations' rivers.\2 Moreover,
state officials have reported that 15 percent of monitored river
miles and 39 percent of lake areas were contaminated by toxic
substances.\3 Contamination by toxic and conventional pollutants
resulted in more than 1,000 bodies of water around the country with
fishing advisories or bans.\4
In an effort to control discharges of toxic pollutants from factories
and sewage treatment plants, EPA, the agency with jurisdiction in
matters of toxic pollution of waterways, has developed an approach
that includes the collection of three different types of information.
Under this approach, information about the environment, the
facilities that dispose of toxic pollutants, and the toxic substances
themselves provides the foundation for controlling toxics. Ensuring
information of adequate quality about these elements, then, is
critical to the approach's success. Adhering to information and data
quality assurance standards is a requisite for attaining such
quality. In short, decisions made within the EPA water quality
programs can be only as good as the data on which they are based. At
minimum, those data should be of high enough quality to appropriately
and accurately characterize the event or object of interest. In the
recent past, concerns have been raised about the quality of
information EPA used in its water protection program.\5
The Chairman of the Subcommittee on Superfund, Ocean, and Water
Protection of the Senate Committee on Environment and Public Works
asked us to evaluate (1) how well EPA ensures information and data
quality in the water toxic control program and (2) the extent to
which the program controls toxic discharges. This report responds to
that request.
--------------------
\1 Bureau of National Affairs, Environment Reporter, June 23, 1989,
p. 466.
\2 Environmental Protection Agency, National Water Quality Inventory:
1990 Report to Congress (Washington, D.C.: 1992), p. 88.
\3 Environmental Protection Agency, p. 86.
\4 Environmental Protection Agency, p. 90.
\5 U.S. General Accounting Office, Environmental Enforcement: EPA
Cannot Ensure the Accuracy of Self-Reported Compliance Monitoring
Data, GAO/RCED-93-21 (Washington, D.C.: March 31, 1993), and Water
Pollution Monitoring: EPA's Permit Compliance System Could Be Used
More Effectively, GAO/IMTEC-92-58BR (Washington, D.C.: June 22,
1992).
BACKGROUND
---------------------------------------------------------- Chapter 1:1
As already noted, dealing with toxic discharges within the framework
of the Clean Water Act requires information on the aquatic
environment, the facilities that discharge the pollutants, and toxic
substances. Collecting this information is not easy, because of both
the complexity and variability of the subject matter and the
technical and methodological challenges associated with measurement,
data collection, and analysis. EPA has created a framework to
collect data on all these elements, as described below.
THE AQUATIC ENVIRONMENT
-------------------------------------------------------- Chapter 1:1.1
The nation's surface-water resources are truly immense. They consist
of more than a million miles of rivers and streams (enough to extend
from the Earth to the Moon four times). There are 61,000 square
miles of inland water bodies and 94,000 square miles of the Great
Lakes. Taken together, this constitutes an area larger than the
combined size of 12 states.
Besides being vast, the aquatic environment is highly complex. It is
composed of different "compartments," including the "water column,"
sediment, and aquatic life. Marshes, wetlands, lakes, rivers,
streams, estuaries, and open oceans constitute greatly diverse
habitats for aquatic life. Within both the water column and the
sediment layers are varied collections of aquatic life, both animals
and plants, with interdependent life cycles. Site-specific
characteristics of a body of water (such as temperature, pH, and
hardness) influence the areas in which different species reside and
their sensitivity to stresses from toxic substances and other
pollutants.
DISCHARGES INTO THE AQUATIC
ENVIRONMENT
-------------------------------------------------------- Chapter 1:1.2
For as long as people have lived in villages and towns there has been
pollution of rivers and streams. Many of the epidemics of the Middle
Ages in Europe were caused by exposure to polluted waters. In the
United States, the industrial revolution of the 18th and 19th
centuries and the accompanying increases in manufacturing increased
discharges, sowing the seeds of a water quality crisis for the
nation's lakes, rivers, and streams. The chemical revolution of the
20th century dramatically increased the complexity of these
discharges.
These bodies of water serve as sources of drinking water for about
half the nation's population and have recreational uses as well.
They are also the repository for wastes from about 64,000 factories
and sewage treatment plants. Of these dischargers, about 7,000 are
considered "major dischargers" under federal legislation. Industrial
processes have developed around the ability to dispose of industrial
wastes into adjacent bodies of water. Disposing of wastes in this
way is integral to modern industry. Billions of pounds of pollutants
are released into the nation's waters every year from these "point
sources" of pollution.\6
--------------------
\6 Discharges from specific locations are called point source
discharges. Pollution that does not enter the water from a pipe is
"nonpoint" pollution. Examples of sources of nonpoint pollution are
farms, fields, and highways.
TOXIC POLLUTANTS
-------------------------------------------------------- Chapter 1:1.3
More than 65,000 chemicals are manufactured or used in the United
States; over 1,000 new substances are introduced each year. A
contrast is often drawn between "toxic" and "conventional"
pollutants.\7 The short list of conventional pollutants, so-named
because they were traditionally the objects of wastewater treatment
and control, consists of the following: biochemical oxygen demand
(BOD), coliform, oil and grease, pH, and total suspended solids. The
toxic effects produced by conventional pollutants are usually
immediately apparent. Too high a concentration of nitrogen or
phosphorus in a lake can turn it green with algae. Too high a level
of BOD in a lake or stream will deplete it of oxygen and suffocate
the aquatic life residing in it.
Toxic pollutants usually have more subtle effects on the health of
the ecosystem and those who come into contact with it. These effects
can range from harming the reproductive cycle of the wildlife in a
stream to causing cancer in people who drink that water. In
addition, these substances often produce their toxic effect at very
low concentrations that may be beyond the capabilities of laboratory
instruments to detect.
Examples of toxic pollutants are some metals such as mercury and
chromium; pesticides such as DDT and 2,4,5-TP; and organic chemicals
such as PCBs and dioxins. Toxic substances can affect all organisms
that dwell in water. When toxic substances settle, they can also
harm the organisms that dwell in sediments. Further, there is
variation in the amount of time it takes for toxic substances to
produce harmful effects. Some toxic substances have an immediate
action; others take a longer period of time. A long-term effect of
some toxic substances is bioaccumulation, or the transfer of a
substance from the environment into living tissues through feeding
and bioconcentration (the passing of a substance from water into an
organism). In fish, for example, bioconcentration occurs by
absorption through the skin or by uptake through the gills. Living
tissues reach higher concentrations than the surrounding water when a
substance bioaccumulates. Thus, contaminated fish and shellfish
present a health hazard to people who eat them. Determining the
magnitude of risk associated with individual pollutants is a highly
complicated task involving complex technical studies.
--------------------
\7 Within this report, we define a water pollutant as any constituent
or property of water that undermines the ability of the body of water
to support the uses to which it is put. Thus, even fresh water can
be a pollutant in a salt water ecosystem, where the organisms that
reside there are adapted to highly saline water.
CONTROLLING POLLUTANTS: THE
CLEAN WATER ACT
-------------------------------------------------------- Chapter 1:1.4
The Federal Water Pollution Control Act of 1972, known as the Clean
Water Act, established the general framework for controlling these
pollutants. Surface water can be characterized according to a number
of uses. It is a source of drinking water for municipalities. It
has recreational uses, such as boating, swimming, and fishing. It is
the repository for wastes of numerous dischargers of pollutants such
as factories and municipal sewage treatment plants. Much of the act
is designed to provide a way to limit discharges in order to minimize
the degree to which they interfere with the other uses of water
bodies. In general, limits are placed on a pollutant if it is
thought to threaten human health or aquatic life.
The general framework of the 1972 Clean Water Act consists of a set
of interrelated program activities based on provisions of the act and
subsequent amendments to it. The process for controlling discharges
of pollutants is administered through the National Pollutant
Discharge Elimination System (NPDES). NPDES was created through the
act as a way of controlling the pollutants discharged by the
thousands of facilities releasing their wastes into surface waters.
The cornerstone of the NPDES framework is a process for issuing
permits to point source facilities: That is, discharging any
substance into any body of water without a permit is forbidden. When
a permit is issued, the pollutants requiring control are identified,
limits for their discharge are established, and then through
monitoring and enforcement EPA and the states are able to ensure
compliance with the permit. So, placing a limit on a pollutant in a
permit constitutes the first step in the Clean Water Act framework
for controlling discharges from a particular point source.
Permits are to be issued for every facility for up to 5 years.
Prominent among the conditions contained within most permits are
requirements that the total amount of a discharge be monitored by a
discharging facility and that the amount of pollutants released be
limited to values established through the permit process. For
example, chloroform might be limited through a permit by requiring
that it be monitored daily and that no more than 10 pounds be
discharged each day. Other pollutants would have similar types of
discharge requirements defined on the permit. Although issuing the
NPDES permit is the centerpiece of the current approach for limiting
discharges into surface waters, a related process uses information
provided by several other activities (described in chapter 2).
Implementation of NPDES rests with 39 delegated states and EPA's 10
regional offices. Each regional office retains implementation
authority for nondelegated states in that region. Regions and
delegated states issue permits with effluent limits and monitoring
requirements, track compliance with the permit conditions, and carry
out enforcement actions.
Controlling toxic substances has been one of the major challenges for
the Congress and EPA since the passage of the Clean Water Act. The
number of toxic pollutants is large, the related risk is often
difficult to demonstrate, and measurement is expensive or
inconclusive at the low concentrations that pose a threat. In 1977
and 1987, the Congress passed amendments to the act focusing renewed
attention on controlling toxics in surface water.
In section 307 of the Clean Water Act as amended in 1977, the
Congress directed EPA to focus its regulatory activities on a list
(published elsewhere) containing 31 chemical groups (such as
halomethanes and polynuclear aromatic hydrocarbons) and 38 individual
substances (such as isophorone and thallium). From this requirement,
the agency developed a list of 129 individual pollutants for priority
attention, but this list has been reduced to 126. These pollutants
are termed the "priority pollutants," and EPA has directed most of
its attention with respect to toxics within the water protection
framework to these 126. According to EPA,
"The priority pollutant list identifies toxic pollutants of
concern on a national basis. It has served as a basis for
numerous EPA actions, including the selection of pollutants for
development of water quality criteria under section 304(a) of
the Clean Water Act; the development of technology-based
effluent guidelines under section 301 of the Clean Water Act;
the listing of impaired waters under section 304(l) of the Clean
Water Act."\8
EPA has recently been augmenting its approach for controlling toxic
pollutants under the Clean Water Act. An important initiative in
this area is an approach called "whole effluent toxicity" (WET).\9
The agency has recognized that discharges from facilities often
consist of a "chemical stew" of many different pollutants, each with
its own "toxic signature," with a combined effect on a receiving
water body that is impossible to deduce from the individual
components. EPA has found that the toxicity of the "whole effluent"
is what poses a human health risk and affects the environment and
has, thus, decided that the "stew" itself, and not just the
ingredients, is a proper subject for regulation and control through
the NPDES process.
In the view of the agency, a WET limit (a limit based on the results
of testing the toxicity of the wastewater through a bioassay)
"provides a way to evaluate an effluent in the absence of detailed
information about the chemicals it contains."\10 However, because WET
test results do not specify which substances in the effluent are
toxic, they are not prescriptive: That is, by themselves, they do
not indicate how to treat the effluent's toxicity to protect human
health and aquatic life. Therefore, limits on individual pollutants
remain a critical component of the water quality approach in spite of
advances in WET testing.\11 Other initiatives EPA mentioned in this
area are toxicity identification evaluations and toxicity reduction
evaluations, which are intended to assist those holding permits to
achieve compliance with WET limits.\12
--------------------
\8 Environmental Protection Agency, "Water Quality Guidance for the
Great Lakes System and Correction; Proposed Rules," 58 Fed. Reg.
20801, at 20843 (April 16, 1993).
\9 Environmental Protection Agency, Technical Support Document for
Water Quality Based Toxics (Washington, D.C.: April 1991).
\10 Environmental Protection Agency, Introduction to Water Quality
Based Toxics Control for the NPDES Program (Washington, D.C.: March
1992), p. 6.
\11 Environmental Protection Agency, Technical Support Document, p.
1-4.
\12 Environmental Protection Agency, Technical Support Document, p.
114.
OBJECTIVES, SCOPE, AND
METHODOLOGY
---------------------------------------------------------- Chapter 1:2
OBJECTIVES
-------------------------------------------------------- Chapter 1:2.1
In the winter of 1992, the Chairman of the Subcommittee on Superfund,
Ocean, and Water Protection of the Senate Committee on Environment
and Public Works asked us to evaluate EPA's effort to ensure the
quality of the information used to control the discharge of toxics to
surface water and the extent to which toxic substances that are
discharged are listed on water quality permits. In responding to the
request, we addressed four questions:
1. What are the EPA water quality activities that control discharges
of toxic substances from point sources into surface waters and what
are their critical information requirements?
2. To what extent do these activities meet critical information
quality assurance requirements?
3. To what extent are limits for actual toxic substance discharges
included on NPDES permits?
4. What are the effects on water quality of unlisted discharges of
toxic pollutants?
SCOPE
-------------------------------------------------------- Chapter 1:2.2
For questions 1 and 2, we examined activities across programs
administered by EPA's water quality office that are directly
responsible for controlling discharges of toxics from point sources.
A number of these activities are overseen by EPA and implemented at
the state level.
The scope for question 3 was comprised of facilities selected to
include three industrial categories from which EPA's industrial
technology division has recently gathered discharge monitoring data.
These categories--pulp and paper, pharmaceutical manufacturing, and
pesticide manufacturing--are among the most significant contributors
of toxic substance discharges in the nation in terms of the amount of
toxic pollutants discharged. We addressed question 4 using the same
set of facilities.
METHODOLOGY
-------------------------------------------------------- Chapter 1:2.3
1. PRINCIPAL ACTIVITIES
AND THEIR INFORMATION
REQUIREMENTS
------------------------------------------------------ Chapter 1:2.3.1
To identify and characterize the activities that EPA has established
to control toxic discharges to surface water and their critical
information needs, we used program documents and literature reviews
to develop a list of activities conducted across a number of water
quality programs. We then submitted this list to EPA officials,
asking them to identify those that directly affect the control of
toxics.
Our next step was to administer a structured interview to the
managers of the "core" activities to determine (1) the types of
information (for example, hydrology information about receiving
bodies of water and information about pollutants discharged) that are
required for conducting the activities and (2) the principal sources
of the information. We validated our findings by asking an expert
panel to review the core programs and information requirements
identified (see appendix I for a list of the experts).
2. THE EXTENT TO WHICH
INFORMATION MEETS QUALITY
ASSURANCE CRITERIA
------------------------------------------------------ Chapter 1:2.3.2
We could not evaluate the quality of the EPA data directly, because
of the diversity of data used and the amount of variation in the same
type of information from case to case. Instead, we used commonly
accepted standards of quality assurance to assess the degree to which
the EPA process adhered to such standards in collecting, analyzing,
and reporting the required information (identified in question 1).
This evaluation approach was feasible because EPA's official policy
for conducting studies requires that quality assurance criteria be
established and followed by every agency data collection effort.
According to agency documents, quality assurance is indispensable for
producing valid data.\13 For these reasons, we were able to use an
evaluation of the EPA quality assurance effort as a surrogate for
directly examining data quality.
We developed quality assurance criteria from a review of the
scientific literature and EPA's quality assurance/quality control
(QA/QC) guidelines. Members of EPA's quality assurance management
staff then reviewed the list of criteria for relevance and
completeness. We then determined which quality assurance criteria
were relevant and critical to each information requirement (from
evaluation question 1) for each of the core water quality protection
program activities, resulting in separate sets of criteria from our
overall set for each requirement. We asked our panel of experts to
help us identify the quality assurance criteria relevant and critical
to each information requirement. We then interviewed the managers of
the activities to determine the extent to which the criteria assigned
to each information type were met. We interviewed other individuals
in EPA and the U.S. Geological Survey (USGS) to obtain responses to
questions that program managers were unable to answer.\14 We obtained
detailed explanations for each response during the interviews and
independently reviewed supporting agency studies and data bases for
particular responses.
--------------------
\13 Environmental Protection Agency, Minimal Requirements for a Water
Quality Assurance Program (Washington, D.C.: 1975), p. 1-3.
\14 We obtained responses to a number of criteria from USGS staff.
We also surveyed staff within EPA's 10 regional offices for responses
to several criteria.
3. THE EXTENT TO WHICH
THE NPDES PERMIT PROCESS
CONTROLS TOXIC SUBSTANCES
------------------------------------------------------ Chapter 1:2.3.3
The third evaluation question addresses the extent to which toxic
substances receive even the first level of control provided by the
NPDES permit process. We answered this question by comparing reports
of actual discharges against Permit Compliance System (PCS) permit
files for a sample of facilities and identifying the "unlisted"
pollutants--that is, toxic pollutants that were discharged but not
listed on the permit and were thus not directly controlled.\15 We
obtained the discharge data from (1) studies of discharges conducted
by the industrial technology division of EPA, (2) reports of
discharges by industrial facilities under section 313 of the
Emergency Planning and Community Right-to-Know Act (otherwise known
as the Toxic Release Inventory), and (3) NPDES permit applications
listing pollutants identified in facilities' effluent.
--------------------
\15 The Permit Compliance System is EPA's data storage and retrieval
system for NPDES permit files.
4. THE IMPACT OF
UNLISTED DISCHARGES ON
WATER QUALITY
------------------------------------------------------ Chapter 1:2.3.4
To address our last evaluation question, we analyzed the human health
and aquatic life risks associated with the uncontrolled toxic
pollutant discharges identified in question 3. In order to determine
these risks, we used a method similar to that employed by EPA regions
and states when they review permit applications.\16 We developed
in-stream concentrations of unlisted pollutants using a water quality
model and used reported stream flow data. We obtained these from
EPA. The model is a "static" water quality model in which the
quantity discharged by the facility is diluted into the stream at
either low flow or mean flow conditions.\17 We then compared the
resultant concentrations against applicable water quality criteria to
determine which of the discharges posed a risk to either human health
or aquatic life.
--------------------
\16 We reviewed our methodology with staff from EPA's office of water
to ensure its similarity to that employed by permit authorities.
\17 Whether we estimated concentration at low flow or mean flow
conditions depended on whether we were comparing the value against an
"aquatic life" or "human health" criterion. The general methodology
for conducting this evaluation is presented in EPA's Technical
Support Document and in Versar, Inc., ReachScan User's Manual
(Springfield, Va.: September 1991).
STRENGTHS AND LIMITATIONS
---------------------------------------------------------- Chapter 1:3
One strength of the work performed for this evaluation is the
comprehensiveness of the effort to determine the principal activities
EPA has designed to control toxic discharges. Our review of both
literature and documentation was designed to ensure that we would
identify every activity implemented in the overall EPA water quality
program. Through indepth review of all activities identified,
interviews with EPA staff to determine which activities were directly
related to toxics control, and with the use of an expert panel to
validate our results, we were certain to identify the activities that
are absolutely essential to the toxics control program.
To address EPA's steps to ensure data quality, our effort again was
comprehensive. We used both EPA's quality assurance criteria and
other criteria identified through a second literature review, this
time in the specialized area of quality assurance. We asked the EPA
staff who are responsible for the agency's quality assurance efforts
to review our initial list of criteria for appropriateness. The use
of an expert panel to examine whether criteria were both relevant and
critical to ensure quality in the activities we identified served to
further validate our work in establishing the normative framework.
A limitation with regard to the measurement of data quality is that
we could not assess that quality directly. As noted earlier, we
looked at compliance with quality assurance criteria as a proxy for
direct measurement, because the complexity of direct measurement far
transcended the scope of this study. Although compliance with
criteria for quality is an indirect measure, such compliance is a
necessary (if not sufficient) condition for quality, and it is EPA's
official policy concerning all agency data collection.
Finally, our work in answering questions 3 and 4--determining the
extent to which the permit process controls toxic discharges and
determining the risk of uncontrolled discharges--has the strength of
being a national level assessment. That is, the sample of facilities
whose permits we reviewed was geographically dispersed across the
country. Further advantages were the comprehensiveness of our three
data bases (which defined the types of toxic pollutant discharges and
their amounts) and the importance of the industrial sectors we chose
to examine (pulp and paper, pharmaceutical, and pesticide
manufacturing). These three sectors are among the most significant
across the nation for discharging toxic pollutants, in terms of both
the amount and types of pollutants discharged. However, although our
sample was fairly large (236 facilities), our study was not designed
to allow its conclusions to be generalized beyond that sample to the
total population of the dischargers.
EPA ACTIVITIES THAT LIMIT
DISCHARGES OF TOXICS INTO SURFACE
WATERS AND THEIR INFORMATION
REQUIREMENTS
============================================================ Chapter 2
In this chapter, we address our first evaluation question: What are
the EPA water quality activities that control discharges of toxic
substances from point sources into surface waters and what are their
critical information requirements? To answer this question, we
refined the larger number of activities EPA conducts across a number
of water quality programs to the 7 "core" activities that are
fundamental to EPA's basic approach to control the discharge of toxic
water pollutants from point sources. We then identified 13 types of
information required by these activities. In this chapter, we
describe each activity, the information it requires, and the roles
and functions of the information collected and list the data sources
for the various information types.
ACTIVITIES TO LIMIT DISCHARGES
OF TOXICS
---------------------------------------------------------- Chapter 2:1
EPA is responsible for many activities that are designed to limit
discharges of pollutants into surface waters. Through a literature
review, we identified 46 activities that EPA conducted within the
agency's office of water. We first eliminated the 11 that deal with
drinking water or groundwater issues and, therefore, have no direct
relationship with controlling toxic discharges from point sources.
We then administered a questionnaire and conducted a series of
interviews with EPA officials to identify the activities from the
remaining 35 that have a direct effect on controlling toxics. In
this manner, we arrived at a set of 7 activities that form the
nucleus of EPA's approach for controlling toxic discharges from point
sources. Our panel of experts then validated this list of 7
activities.
The 46 activities are presented in table 2.1, which distinguishes the
7 that we retained because of their direct influence on controlling
discharges of toxic substances from the 39 others. The principal
reasons for eliminating each of the 39 are also presented in the
table. The 7 activities that are the focus of this chapter are then
described in table 2.2.
Table 2.1
Activities Conducted Within EPA's Office
of Water
Principal reason for
No. Activity Status elimination
-------- ----------------------------- -------- -----------------------------
1 Assess progress in Eliminat Administrative activity
implementing the Coastal Zone ed
Management Act to control
nonpoint source pollution in
coastal waters
2 Assess toxicity control needs Eliminat Duplicates water-quality-
for water-quality-based ed based permit activity number
permits 28
3 Conduct water quality Eliminat Regulatory support activity
standards triennial reviews ed
4 Demonstrate accomplishments Eliminat Not related to controlling
in maintaining active state ed toxic discharges
and federal enforcement
programs for drinking water
5 Develop ecological criteria Eliminat Developmental activity
guidance ed
6 Develop stormwater permit Eliminat General regulatory activity
strategies ed with small need for data
7 Draft a strategy for Eliminat Developmental activity
addressing contaminated ed
sediment
8 Draft plans for stormwater Eliminat Administrative activity
permits ed
9 Enforce state testing for Eliminat Not related to controlling
lead in drinking water ed toxic discharges
10 Establish drinking water Eliminat Not related to controlling
laboratory certification ed toxic discharges
regulations
11 Establish national primary Eliminat Not related to controlling
drinking water regulations ed toxic discharges
12 Establish total maximum daily Retained
loads
13 Establish water quality Retained
criteria
14 Focus on reauthorization of Eliminat Nonspecific regulatory
Clean Water Act ed activity
15 Identify and track water Eliminat Administrative activity
quality improvements of ed
targeted water bodies
16 Identify compliance problems Retained
17 Identify priority issues in Eliminat General regulatory activity
stormwater application ed with small need for data
process
18 Implement nonpoint source Eliminat Not related to controlling
watershed control programs ed toxic discharges from point
sources
19 Improve effectiveness of Eliminat Regulatory developmental
inspection activities ed activity
20 Improve quality and Eliminat Regulatory developmental
timeliness of enforcement ed activity
responses
21 Issue health advisories for Eliminat Not related to controlling
drinking water ed toxic discharges
22 Issue administrative Eliminat Not related to controlling
compliance orders for ed toxic discharges
drinking water
23 Issue drinking water Eliminat Not related to controlling
regulations ed toxic discharges
24 Develop effluent guidelines Retained
25 Issue fines and negotiate Eliminat Not directly related to
consent decrees for permit ed toxics control
violations
26 Issue lists of impaired Retained
waters and related point
sources
27 Issue regulations to control Eliminat Not related to point source
pollutants in urban runoff ed control
28 Issue water quality permits Retained
29 Make revisions to permits Eliminat Duplicates number 28
codifying changes made by ed
1987 Clean Water Act
amendments (antibacksliding
provisions)
30 Manage state revolving funds Eliminat Administrative activity
ed
31 NPDES regulatory revisions Eliminat Nonspecific regulatory
contained in 1987 Water ed activity
Quality Act
32 Prepare sewage sludge use and Eliminat Regulatory development
disposal regulations ed activity
33 Propose revisions to Eliminat Not related to controlling
monitoring requirements for ed toxic discharges
drinking water
34 Reduce noncompliance with Eliminat Not related to controlling
existing drinking water ed toxic discharges
standards
35 Reduce exposure of population Eliminat Not directly related to
to contaminants in drinking ed controlling toxic discharges
water
36 Regulate sources of toxic Retained
discharges to publicly owned
treatment works from indirect
dischargers
37 Regulate substances not on Eliminat Not related to controlling
drinking water priority list ed toxic discharges
38 Revise stormwater Eliminat Program in development
implementation rules ed
39 Revision of denial or Eliminat Not directly related to
restriction of disposal sites ed controlling toxic discharges
40 Set federal NPDES fees in Eliminat Administrative activity
states where EPA administers ed
NPDES program
41 Strengthen the scientific Eliminat Not directly related to
basis of water quality ed controlling toxic discharges
standards to protect critical
aquatic resources
42 Track fish consumption Eliminat Administrative activity
advisories ed
43 Track pretreatment programs Eliminat Administrative activity
ed
44 Track sludge facilities Eliminat Administrative activity
ed
45 Track stormwater permit Eliminat Not directly related to
activity ed controlling toxic discharges
46 Water quality standards Eliminat Not of national scope
revisions for Indian tribes ed
--------------------------------------------------------------------------------
Table 2.2
Activities Designed to Control
Discharges of Toxic Substances
No. Activity\a Objective
-------- ------------------ ------------------------------
1 Effluent Develop national standards,
guidelines setting out baseline
development (no. requirements for pollution
24) removal within individual
industries
2 Water quality Develop statements of the
criteria maximum permissible
development (no. concentrations of individual
13) toxic pollutants in water that
will not harm human health and
aquatic life
3 Impaired water and Identify water threatened by
point source point source discharges of
identification\b toxic substances and the
(no. 26) facilities causing the
impairments
4 Total maximum Estimate the amount of a
daily load pollutant that can be safely
development (no. discharged from all sources
12) (including natural sources)
into a body of water
5 Water-quality- Issue the legal documents
based permit containing the conditions and
issuance (no. 28) limits controlling a point
source discharger's release of
pollutants into a body of
water based on the expected
effect of the release on water
quality
6 "Local limits" Restrict the amount of
development (for pollutants that "indirect
reducing dischargers" can release into
discharges of sewer systems
pollutants into
sewage treatment
plants) (no. 36)
7 Compliance problem Determine whether a discharger
identification is meeting its permit
(no. 16) conditions; used as a basis
for possible future
enforcement action
------------------------------------------------------------
\a Numbers in parentheses identify activities listed in table 2.1.\
\b This was a one-time exercise mandated by the Water Quality Act of
1987. Section 304(l) called for three lists to be prepared by each
state. Only one of them featured streams with impairment primarily
from toxic pollution from point sources (the others being broader
delineations of impaired water bodies). We used the preparation of
this list, and not the two others, as the activity.
Across these 7 activities, the essential data for implementing EPA's
approach are gathered and processed. In many cases, information that
has been collected and decisions that have been reached within one
activity are then used within another. For example, water quality
criteria (activity 2) are used to generate water-quality-based
permits (activity 5). In terms of their information function, the
core program activities can be grouped into four categories by the
role each plays in the overall process: (1) uniform guideline
development, (2) geographic targeting, (3) permit development, and
(4) compliance assessment.
The overall process for controlling toxics from point sources begins
with uniform guideline development: the development of general
practices for making decisions with respect to the thousands of
facilities subject to NPDES. In this first category, EPA develops
national guidelines or criteria--specifically effluent guidelines and
water quality criteria (activities 1 and 2 in table 2.2). They
provide the rationale for determining what substances should be
limited on permits and what those limits should be. The second step
in the process is to identify specific facilities that require
expedited review of their permit. The one time that geographic
targeting was employed by the Congress within the Clean Water Act
occurred in section 304(l) of the act as amended in 1987 (activity
3). The third category, permit development, uses the national
criteria and standards (from the first category) along with
site-specific information to determine the conditions and limitations
needed on individual permits--specifically, total maximum daily load
(TMDL) development, NPDES permit issuance, and "local limits"
development (activities 4, 5, and 6). The fourth category,
compliance assessment (that is, activity 7, compliance problem
identification), uses information about patterns of discharges to
determine whether facilities have complied with the conditions and
limits of their permits. This activity compares actual monitoring
and test data against the limits and conditions imposed in permits.
If problems are found, the results of this comparison can lead to
either an enforcement action against a facility for violating a
permit condition or to a revision of its permit.
INFORMATION REQUIRED WITHIN THE
SEVEN CORE ACTIVITIES
---------------------------------------------------------- Chapter 2:2
We identified critical information needed to support the core
activities through interviews with EPA program managers and others.
"Critical" information was defined as having primary importance to
conducting the activity with respect to meeting EPA's water quality
goals for controlling point source discharges of toxics into surface
water. Weaknesses in critical information clearly undermine the
effectiveness of an activity. Each of the 13 information types we
identified as required by the 7 activities is briefly described in
table 2.3.
Table 2.3
The Thirteen Types of Information
Required Across the Seven Activities
No. Information type Description
-------- ------------------ ------------------------------
1 Hydrology Characteristics of the
properties of lakes and rivers
such as stream flow
2 In-stream water The chemical properties of the
quality receiving body of water
(typically the concentration
of the chemical constituents)
3 Concentration in Concentrations of chemical
sediments constituents found within the
sediments of receiving water
bodies
4 Concentration in Concentrations of toxics
biota within the aquatic creatures
inhabiting the water body
5 Aquatic life The extent to which aquatic
exposure life is exposed to pollutants
6 Discharges from Substances and amounts being
individual sources discharged from specific
plants and facilities
7 Removal Characterizations of the
technologies technologies that are
effective within the industry
for removing pollutants from
wastewater
8 Water use Characterizations of the
reduction techniques that can be used to
technologies reduce discharges of
pollutants through water reuse
in the industrial production
process
9 Aquatic life The nature and extent of
effects effects upon aquatic organisms
that come into contact with
substances in the
environment\a
10 Human health The nature and extent of human
effects effects associated with
exposure to the substance in
the environment
11 Fate and transport The mechanisms and rate at
which a substance undergoes
change and moves through the
environment (or within a
publicly owned treatment
works)
12 Human health The extent and way in which
exposure humans are exposed to water
pollutants through drinking
water and eating fish
13 Economic data The effect on an industry's
profitability of changes in
categorical technologically
based effluent standards
------------------------------------------------------------
\a In connection with the section 304(l) exercise, this took on the
meaning of assigning "biocriteria," or using indicator species, to
measure environmental impairments.
THE SEVEN CORE ACTIVITIES AND
THEIR INFORMATION NEEDS
---------------------------------------------------------- Chapter 2:3
We found, as expected, that all 7 activities depend heavily on
information. Each requires a different set of specific data for its
implementation. As already noted, 13 different types of information
in all are required across the 7 activities. The types of
information listed as 1-13 in table 2.3 are labeled A-M in figure 1
and elsewhere in the report. The types of information used to
support the 7 activities vary significantly.
THE ROLES AND FUNCTIONS OF THE
INFORMATION
---------------------------------------------------------- Chapter 2:4
In the series of summary tables 2.4 through 2.10, we present the
reasons why the 13 types of information we identified in the
preceding section (types A-M) are required across the 7 activities.
Each table characterizes the determinations, within each activity,
that the information is required to support (its purposes) and states
the source of the information and how it is developed. In appendix
II, we present a more detailed description of each activity,
highlighting its statutory and regulatory basis in addition to the
role and principal information sources of each information type.
Table 2.4
The Purposes and Sources of Critical
Information Within Activity 1: Effluent
Guidelines Development
Typical principal
Information type Purpose source
-------------------- ------------------ ------------------
Discharges from To identify the Questionnaire to a
individual sources pollutants that sample of
(F) are typical of facilities within
facilities within industry; site
an industry visits
Removal technologies To identify the Questionnaire to a
(G) most efficient sample of
technologies facilities within
available within industry; site
an industry to visits
remove pollutants
Water use reduction To identify the Questionnaire to a
technologies (H) most efficient sample of
technologies facilities within
available within industry; site
an industry to visits
reduce discharges
through efficient
water use
Economic data (M) To identify and Questionnaire to a
characterize the sample of
economic effect on facilities within
an industry of industry
various
alternatives
------------------------------------------------------------
Table 2.5
The Purposes and Sources of Critical
Information Within Activity 2: Water
Quality Criteria Development
Typical principal
Information type Purpose source
-------------------- ------------------ ------------------
Concentration in To determine how Peer reviewed
biota (D) pollutants scientific
bioconcentrate in literature; EPA's
the environment field studies
Aquatic life To provide a Criteria from EPA
exposure (E) measure allowing (on magnitude,
the results of the duration, and
effects studied to frequency of
be related to exposure)
"real world"
exposures
Aquatic life effects To identify the Aquatic toxicology
(I) nature and studies,
magnitude of the internally
effects of developed or from
exposure of the the scientific
pollutant to literature; EPA
aquatic species testing
Human health effects To identify the Peer reviewed
(J) nature and scientific
magnitude of the literature; EPA's
effects of field studies
exposure of the
pollutant to
people
Human health To provide a Peer reviewed
exposure (L) measure allowing scientific
the results of the literature; EPA's
effects studied to field studies
be extrapolated to
the exposures of
people
------------------------------------------------------------
Table 2.6
The Purposes and Sources of Critical
Information Within Activity 3: Impaired
Water and Point Source Identification
Typical principal
Information type Purpose source
-------------------- ------------------ ------------------
In-stream water To identify areas State monitoring
quality (B) that are programs;
contaminated by STORET\a; state-
toxics specific data
bases
Concentration in To identify areas State monitoring
sediments (C) that have programs; STORET
sediments state-specific
contaminated by data bases
toxics
Concentration in To determine areas State monitoring
biota (D) that have aquatic programs and
life contaminated assessments
by toxics
Discharges from To identify areas Permit
individual sources that may be applications and
(F) contaminated by discharge
toxics and the monitoring reports
contributing
sources
Aquatic life effects To determine State monitoring
(I) whether aquatic programs and
life is being section 305(b)
harmed by reports
discharges
Fate and transport To provide precise State section
(K) estimates of in- 305(b)
stream assessments; state
concentrations surveys; state and
EPA compliance
monitoring for
point sources
------------------------------------------------------------
\a STOrage and RETrieval (STORET).
Table 2.7
The Purposes and Sources of Critical
Information Within Activity 4: TMDL
Development
Typical principal
Information type Purpose source
-------------------- ------------------ ------------------
Hydrology (A) To estimate the USGS gaging
"critical flow" of stations and
the receiving regression
water body equations from
gages on
neighboring
streams
In-stream water To estimate the State monitoring
quality (B) background programs; STORET
concentration of
the substance
being discharged
Concentration in To determine State monitoring
sediments (C) whether the programs; STORET
sediment is
contaminated
Concentration in To determine State monitoring
biota (D) whether there is a programs; STORET
contamination
problem
Discharges from To estimate the Permit
individual sources amount discharged applications and
(F) by the facility discharge
monitoring reports
Fate and transport To provide precise Textbooks, journal
(K) estimates of in- articles, EPA and
stream U.S. Fish and
concentrations Wildlife Service
publications; EPA-
supported water
quality models
------------------------------------------------------------
Table 2.8
The Purposes and Sources of Critical
Information Within Activity 5: Water-
Quality-Based Permit Issuance
Typical principal
Information type Purpose source
-------------------- ------------------ ------------------
Hydrology (A) To determine the USGS gaging
dilution limits stations and
for aquatic life regression
and human health equations from
effects gages on
neighboring
streams
In-stream water To determine the STORET and state
quality (B) background water quality data
concentrations for bases
the substances
being considered
Discharges from To determine the Permit
individual sources amount discharged applications and
(F) from the facility discharge
monitoring reports
Fate and transport To provide precise Textbooks, journal
(K) estimates of in- articles, EPA and
stream U.S. Fish and
concentrations Wildlife Service
publications; EPA-
supported water
quality models
------------------------------------------------------------
Table 2.9
The Purposes and Sources of Critical
Information Within Activity 6: "Local
Limits" Development
Typical principal
Information type Purpose source
-------------------- ------------------ ------------------
Discharges from To determine the Monitoring of the
individual sources substances and source's effluent,
(F) amounts being by the indirect
discharged discharger and the
publicly owned
treatment works
Removal technologies To determine the Self-monitoring of
(G) effectiveness of the publicly owned
the publicly owned treatment works'
treatment works at effluent
removing the
substance
Fate and transport To determine the Self-monitoring of
(K) amount of each the publicly owned
substance treatment works'
remaining in wastestream
wastewater,
volatilizing into
the atmosphere,
and precipitating
in sludge as a
result of
treatment at the
publicly owned
treatment works
------------------------------------------------------------
Table 2.10
The Purposes and Sources of Critical
Information Within Activity 7:
Compliance Problem Identification
Typical principal
Information type Purpose source
-------------------- ------------------ ------------------
In-stream water To compare against Self-monitoring by
quality (B) permit limits in facility; data for
order to evaluate major facilities
compliance (in- entered into PCS
stream monitoring
is rarely
included)
Aquatic life To compare against Self-monitoring by
exposure (E) permit limits in facility; data for
order to evaluate major facilities
compliance (WET entered into PCS
limits)
Discharges from To compare against Self-monitoring by
individual sources permit limits in facility; data for
(F) order to evaluate major facilities
compliance entered into PCS
------------------------------------------------------------
SUMMARY
---------------------------------------------------------- Chapter 2:5
Seven activities across various EPA water quality programs form the
core of the agency's effort to control toxic discharges from point
sources into surface waters: effluent guidelines development, water
quality criteria development, impaired water and point source
identification, total maximum daily load development,
water-quality-based permit issuance, "local limits" development, and
compliance problem identification.
A wide variety of information is employed within the 7 core
activities. All 7 use several different types of information for
making decisions. In all, 13 different types of information are used
at 31 points across the 7 activities (tables 2.4 through 2.10).
Information plays a critical role across these activities. From
setting national standards to ensuring compliance with individual
NPDES permits, information forms the basis for these efforts. Each
one of the 13 types of information is critical to the activities.
For example, without reliable information about efficient and
feasible removal technologies, the effluent guidelines activity would
not produce technological requirements that meet congressional
objectives in the Clean Water Act. Without reliable information
about the actual discharges of facilities with permits, the NPDES
compliance program would not be able to identify the facilities that
are not conforming to their permit requirements.
Having established the core activities and their related information
requirements, we now address our second evaluation question. In
chapter 3, we evaluate the extent to which EPA ensures that critical
quality assurance steps are taken in developing the required
information.
THE EXTENT TO WHICH REQUIRED
INFORMATION MEETS QUALITY
ASSURANCE REQUIREMENTS
============================================================ Chapter 3
In chapter 2, we identified the EPA activities that control toxic
discharges to water and the types of information that support the
activities. In this chapter, we address our second evaluation
question: To what extent do these activities meet critical
information quality assurance requirements? Here, we compare EPA's
official policy for quality assurance in data collection with
reported practice.
INTRODUCTION
---------------------------------------------------------- Chapter 3:1
As we discussed in chapter 2, information is a critical component of
the nation's approach for controlling discharges of toxics, and the
quality of that information is central to the effectiveness of the 7
core activities. In general (that is, within each activity),
information quality is promoted through adopting and practicing
quality assurance standards. Quality assurance practices and
standards control how data are collected, managed, and analyzed so as
to ensure their reliability, validity, and completeness. These
standards also regulate whether the information base is sufficient
for drawing inferences and making decisions. We evaluated
information quality according to two types of criteria. The first
type concerns the methodological steps taken to ensure information
quality.\1 The second type concerns the completeness of the data
gathered vis-a-vis the envisaged purpose--that is, whether all the
information required to support the activity was in fact acquired.
EPA's policy is to protect information quality by incorporating
quality assurance practices into individual studies. EPA order
5360.1 requires a quality assurance program for all environmentally
related measurements performed by or for the agency. The primary
goal of the program is to ensure that all such measurements yield
data of known quality that can be verified as being reliable and
appropriate to the program's objectives. The quality assurance
management staff serves as the central management authority for this
program within EPA.
The staff is charged with developing quality assurance policy and
training program managers in quality assurance practices. Program
managers are responsible for specifying the quality of the data they
require and for providing sufficient resources to ensure that an
adequate level of quality assurance is performed. All routine or
planned projects involving environmental measurement must begin with
a "quality assurance project plan" that specifies acceptable quality
goals.
--------------------
\1 Concerning methodological adequacy, EPA and USGS have recently
determined that the analytical methods used to collect and analyze
the in-stream data for metals may be flawed. As a result, concerns
have been raised about the soundness of the information on which the
states have been basing their water quality programs. At a workshop
that EPA convened to study the issue, an agency official raised the
possibility that the numbers are problematic. A professional
hydrologist suggested that much of the data collected in the past to
determine whether water bodies are exceeding standards is "almost
certainly useless" (Water Policy Report, February 3, 1993, p. 3).
EPA and USGS, the two principal agencies involved in developing
protocols for sample collection, are currently studying how to deal
with these questions.
EVALUATING QUALITY ASSURANCE
PRACTICES
---------------------------------------------------------- Chapter 3:2
Each of the 7 activities we identified in our evaluation has resulted
in a large number of program decisions and products. For example,
dozens of effluent guidelines have been issued over the course of the
clean water program and thousands of NPDES permits are issued and
reissued each year. Because of the large number of decisions and
products, which went far beyond the scope of our evaluation, we
focused our evaluation at the level of the 7 activities.
DEVELOPING CRITERIA
STATEMENTS
-------------------------------------------------------- Chapter 3:2.1
Through a review of the scientific literature on quality assurance,
EPA's quality assurance guidelines, and interviews with experts in
the field, we identified a set of 40 criteria that define a basis for
ensuring both information quality and completeness for the toxic
pollutant control activities we are addressing. Thirty-one of the 40
reflect consensus criteria developed by EPA itself or industrial
groups such as the American Society of Testing and Materials (ASTM)
relating to methodological adequacy. That is, they deal with how
information collection activities should be planned and implemented
and how the information should be reviewed. Most of them apply to
whether safeguards were taken to ensure the validity and reliability
of the information collected in the field and analyzed in the
laboratory. The remaining 9 criteria relate to completeness--that
is, whether the information base of all studies is sufficient (aside
from data quality concerns of individual studies) to reach
conclusions about the subject of interest, such as the in-stream
concentration of dioxin (for setting a limit on an individual
permit).
We converted the 40 quality assurance criteria into a set of
questions, using a response scale of frequency. The questions,
presented in table 3.1, are divided into those dealing with
methodological adequacy and those that ensure a complete information
base.
Table 3.1
Quality Assurance Criteria for
Evaluating the Seven Core Toxic
Pollutant Control Activities
Criterio
Type n Question
------------ -------- ------------------------------------
Methodologic 1 How often are data quality
al adequacy objectives instituted at the
initiation of a project?
2 How often is the design for data
collection and analysis reviewed by
an individual not directly
associated with the program who is
familiar with the methodological
issues involved in data collection
and analysis?
3 How often is a quality assurance
project plan employed within the
data collection process?
4 How often are standard operating
procedures for field and laboratory
sampling and reporting followed?
5 How often are QA/QC procedures for
data storage and maintenance
employed?
6 How often are questionnaires
pretested?
7 How often are chemical analyses and
bioassays conducted according to
EPA-approved methods or other
methods specified in the project
plan?
8 How often are chemical analyses and
bioassays performed by laboratories
that follow good laboratory
practices?
9 How often is an independent analysis
conducted to assess the quality of
the data that went into water
quality criteria?
10 How often do assessments of
environmental quality that are based
on best professional judgment
undergo quality control checks and
independent verification?
11 How often were subject area experts
consulted and drafts distributed for
public review before the data were
accepted?
12 How often are multiple, independent
sets of data used to identify the
toxic substances that were included
within the data collection effort?
13 How often are multiple, independent
sets of data used to identify the
facilities to gather data from
within the data collection effort?
14 Where sampling data were
unavailable, how often were
assessments, based on observations
of water, biota, and sediment by an
experienced professional?
15 How often do individual studies
present missing data exceeding 10
percent of planned sample size?
16 How often is a probabilistic
sampling scheme (a random sample)
employed and documented?
17 How often are all requirements of
the model, such as times and
locations for in-stream or discharge
monitoring, adhered to?
18 How often are the data checked for
age and only those that are
appropriately current retained and
used in the analysis?
19 How often are data from different
time periods gathered in such a way
as to ensure that they are
comparable?
20 How often are independent studies
conducted to assess the degree to
which self-reported samples
represent discharges at different
production and discharge levels?
21 How often is the critical flow based
on data from gaging stations
(whether on the reach itself or
estimated from gaging stations on
neighboring reaches)?
22 When substituting one variable for
another within a model, how often
were steps taken to ensure that the
surrogate variable is correlated
with the original with respect to
the function it performs in the
model?
23 For critical model parameters, how
often were laboratory findings
replicated with field studies?
24 How often are estimates of in-
stream concentrations of discharges
based on water quality models of
proven validity with respect to
local stream conditions?
25 How often do major externally
published reports of findings
discuss significant components of
bias and measurement error?
26 How often was the laboratory "in
control" during the period in which
it performed the tests?
27 How often are field-derived data
obtained under conditions that
adequately represent the range and
variability of biotic and abiotic
conditions?
28 How often is the design for data
collection and analysis reviewed by
an individual within the
organization who is familiar with
the methodological issues involved
in data collection and analysis?
29 How often is the survey design
reviewed by an individual familiar
with methodological issues involved
in survey methods?
30 How often are manufacturers'
specifications adhered to in
calibrating gages?
31 How often was the laboratory
instructed about detection and
quantitation levels and how to
report data below these levels?
Completeness 32 How often is a judgment made that
sampling and analysis included all
pollutants that were likely to be
present at measurable levels?
33 How often are data gathered on all
relevant effects when developing
criteria?
34 How often are there enough data of
proven quality concerning the most
likely effects to support the
contention that criteria are based
on the most sensitive effect?
35 How often are enough data of proven
quality collected on a variety of
representative species or cell lines
to support the contention that
criteria are based on impacts to the
most sensitive species that will
encounter the toxic substance in the
environment?
36 How often is an assessment made
downstream of each "major"
discharger?
37 How often are sampling data
available for waters on which
"major" dischargers of toxics are
situated?
38 How often are minimum data set size
requirements stipulated?
39 How often are sampling frequency and
size based on an assessment of the
variability of discharge loads?
40 How often is the data set of
adequate size to reflect variations
and differences for this regulatory
purpose?
------------------------------------------------------------
DETERMINING THE RELEVANCE OF
THE QUALITY ASSURANCE
CRITERIA
-------------------------------------------------------- Chapter 3:2.2
The list of criteria for ensuring the methodological adequacy and
completeness of information is broad. Whether a criterion is
relevant to an information requirement depends on the activity that
the information requirement supports. Consequently, our first task
was to determine which criteria were relevant to and critical for
each information requirement as it supports each specific activity.
We did this with the assistance of a panel of five experts familiar
with the information requirements of these activities. That is, for
each of the 13 types of information used by the activities (figure
2.1), we identified the criteria that are both relevant and critical
for ensuring information completeness and quality. We defined a
critical criterion as one that, if not routinely met, would
jeopardize the quality of the information and threaten EPA's ability
to draw conclusions from it. The critical criteria are presented in
table 3.2. Thirty-eight of the 40 criteria were applied to at least
one information type.
Table 3.2
Quality Assurance Criteria Applicable to
Information Types
Activity Information type Criteria
----------------------------- ------------------ -----------------------------
Effluent guidelines Discharges from 1-5, 7, 8, 12, 13, 19, 20,
development individual sources 28, 39, 40
Removal 1-5, 7, 8, 10, 13, 19, 25,
technologies 28, 39
Water use 1-5, 7, 8, 10, 13, 19, 25,
reduction 28, 39
technologies
Economic data 1, 3, 5, 6, 18, 28, 29
Water quality criteria Concentration in 1-5, 9-11, 26-28, 34
development biota
Aquatic life 1-5, 9, 11, 26, 28
exposure
Aquatic life 1-5, 7-9, 11, 26, 28, 33-35,
effects 38
Human health 1-3, 5, 9-11, 18, 28, 33, 34,
effects 38
Human health 1-3, 5, 9, 11, 28
exposure
Impaired water and point In-stream water 1, 3-5, 7, 8, 18, 26, 28, 32,
source identification quality 40
Concentration in 1, 3-5, 7, 8, 18, 25, 26, 28,
sediments 40
Concentration in 1, 3-5, 7, 8, 18, 26, 28, 40
biota
Discharges from 1, 3-5, 7, 8, 18, 26, 28, 32,
individual sources 40
Aquatic life 1, 3-5, 7, 8, 28, 33, 40
effects
Fate and transport 1-3, 5, 18, 21, 22, 24, 28,
37, 39, 40
Total maximum daily load Hydrology 21, 22, 24, 28, 38
development
In-stream water 1-5, 7, 8, 18, 28, 37, 39
quality
Concentration in 1-5, 7, 8, 18, 28, 32, 37, 39
sediments
Concentration in 1-5, 7, 8, 28, 37
biota
Discharges from 1-5, 7, 8, 18, 28, 39, 40
individual sources
Fate and transport 1-5, 7, 8, 18, 21, 22, 24,
26, 28, 32, 37, 39, 40
Water-quality-based permit Hydrology 1-5, 19, 24, 25, 27, 28, 30,
issuance 36, 40
In-stream water 1-5, 7-12, 17-19, 25-28, 31,
quality 32, 36, 37, 40
Discharges from 1-5, 7-12, 15, 16, 18, 19,
individual sources 25, 26, 28, 31, 32, 39, 40
Fate and transport 1-5, 7-9, 17-19, 24-28, 31,
32, 36, 39, 40
"Local limits" development Discharges from 1-5, 7, 8, 12, 15, 16, 18,
individual sources 25, 26, 28, 31, 32, 39, 40
Removal 1-5, 7, 8, 10, 19, 25, 26,
technologies 28, 39
Fate and transport 1-5, 7, 8, 10, 18, 19, 25,
26, 28, 39
Compliance problem In-stream water 1, 3-5, 7, 8, 18, 26, 32, 40
identification quality
Aquatic life 1, 3-5, 7, 8, 18, 19, 26, 40
exposure
Discharges from 1, 3-5, 7, 8, 18, 26, 32, 38,
individual sources 40
--------------------------------------------------------------------------------
QUALITY ASSURANCE PRACTICES
---------------------------------------------------------- Chapter 3:3
To determine the extent to which the completeness and quality
assurance criteria were employed within the 7 activities, we
interviewed the relevant EPA managers, using our 40 criteria as a
questionnaire with a five-point response scale rating the extent to
which the activity is conducted "always," "usually," "sometimes,"
"seldom," and "never." A criterion should be met at least "usually"
if quality assurance criteria are to be met. Therefore, we
classified the responses by grouping them into two categories: met
("always" and "usually") and not met ("sometimes," "seldom," and
"never"). We also asked for any supporting data and requested that
each response be explained as we conducted our interviews.
We examined the extent to which the quality assurance criteria were
met from two different perspectives: first, from the overall
perspective of the 7 activities, and, second, from the perspective of
the information types (that is, to what extent were they supported by
necessary quality assurance efforts?).
A preliminary finding was that the officials we interviewed were
aware of EPA's position concerning quality assurance policies and
practices. From our review, it is clear that these managers
recognize and have accepted the principle of quality assurance within
their activities.
Table 3.3 presents the results of our interviews with EPA officials
concerning the degree of implementation of the quality assurance
criteria. Table 3.4 summarizes our results for the activities and
information types. As presented in table 3.4, 2 activities (those
that establish effluent guidelines and water quality criteria) were
exceptional in meeting quality assurance criteria that are critical
to the effort. They met 93 percent of their quality assurance
requirements. However, the 5 other activities fall short in taking
the necessary steps to ensure information quality. That is, impaired
water and point source identification, TMDL development,
water-quality-based permit issuance, "local limits" development, and
compliance problem identification are less than optimal in ensuring
that the required quality assurance steps are taken in developing,
collecting, or maintaining their needed information. Collectively,
about 58 percent of their quality assurance criteria were met.
Table 3.3
Responses to Individual Quality
Assurance Criteria
No.\
b F G H M D E I J L B C D F I K A B C D F K A B F K F G K B E F
---- ---- ---- ---- ---- ---- ---- ---- ---- ---- ---- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
1 5 5 5 5 5 5 5 5 5 4 4 4 2 3 2 4 4 4 4 4 5 4 5 4 4 4 4 3 4 4
2 4 4 4 5 5 5 2 2 2 3 2 2 2 3 5 3 4 4 4 3 3
3 5 5 5 4 5 5 5 4 4 4 4 4 2 3 2 4 4 4 4 4 5 4 3 4 4 4 4 3 4 5
4 5 5 5 5 5 5 5 4 4 3 4 4 4 4 4 4 5 4 4 3 4 4, 4, 4 4 3
5 5
5 5 5 5 5 5 5 5 4 4 4 4 4 3 4 4 4 4 4 4 4 5 4 4 3 2 2 2 1 3 3
6 5
7 5 5 5 5 4 3 4 5 4 4 4 3 4 4 4 4 4 4 4 4 2 4 4
8 5 5 5 5 4 4 4 3 4 4 4 4 4 4 4 4 4 4 4 4 3 4 4
9 5 5 5 5 5 3 3 3
10 5 5 5 5 3 2 3 3
11 5 5 5 5 5 4 4
12 5 3 2 3
13 4 4 4
14
15 2 2
16 2 1,
2
17 4 4
18 5 5 3 4 4 4 4 4 4 4 4 4 4 4 5 3 2 5 4
19 5 5 5 4 3 4 4 4 4 3
20 5
21 3 3 3
22 3 4 4
23
24 4 3 3 2 4
25 2 5 3 2 3 4 4 2 2 2
26 4 4 4 4 4 4 3 4 4 3 4 4 4 4 4 4 4
27 5 4 4 3
28 5 5 5 5 5 5 5 4,5 4,5 3 2 3 2 3 3 3 3 2 2 3 3 4 4 4 4 3 4 4
29 5
30 5
31 3 3 3 4
32 2 2 3 3 4 2 3 4 3 4,
5
33 5 5 3
34 3 3 3
35 3
36 3 4 5
37 4 4 3 2 4 3
38 5 5 2 5
39 5 5 5 3 3 2 3 3 3 3 3 4 4
40 5 2 2 3 2 2 2 3 2 4 2 3 4 3 2 3 3
----------------------------------------------------------------------------------------------------------------------------------------------------
\a Letters in column headings under each activity refer to
information types displayed in figure 1. Numbers in cells are 1,
never; 2, seldom; 3, sometimes; 4, usually; 5, always.
\b Criteria follow numbers in table 3.1.
Table 3.4
The Number of Criteria Met Within Each
Information Type
Assign Not
Activity Information type ed Met met
---------------- ------------------ ------ ------ ------
Effluent Discharges from 14 14 0
guidelines individual
development sources
Removal 13 12 1
technologies
Water use 13 13 0
reduction
technologies
Economic data 7 7 0
Water quality Concentration in 12 11 1
criteria biota
development
Aquatic life 9 9 0
exposure
Aquatic life 15 13 2
effects
Human health 12 10 2
effects
Human health 7 6 1
exposure
Impaired water In-stream water 11 7 4
and point quality
source
identification
Concentration in 11 7 4
sediments
Concentration in 10 8 2
biota
Discharges from 11 2 9
individual
sources
Aquatic life 9 4 5
effects
Fate and transport 12 4 8
TMDL development Hydrology 5 1 4
In-stream water 11 8 3
quality
Concentration in 12 7 5
sediments
Concentration in 9 5 4
biota
Discharges from 11 7 4
individual
sources
Fate and transport 17 9 8
Water-quality- Hydrology 13 10 3
based permit
issuance
In-stream water 23 14 9
quality
Discharges from 22 12 10
individual
sources
Fate and transport 21 14 7
"Local limits" Discharges from 18 11 7
development individual
sources
Removal 13 9 4
technologies
Fate and transport 14 9 5
Compliance In-stream water 10 2 8
problem quality
identification
Aquatic life 10 7 3
exposure
Discharges from 11 8 3
individual
sources
============================================================
Total 386 260 126
Percent 66% 34%
------------------------------------------------------------
\a Criteria numbers are in table 3.1.
MEANINGFULNESS OF THE RESULTS
---------------------------------------------------------- Chapter 3:4
The work presented above shows that practice within 5 of the 7 core
activities does not measure up to EPA's enunciated information
quality assurance policies and principles. This raises questions
about the quality of information generated and used within these 5
activities.
We initiated a separate qualitative review of agency documents and
information bases and were able to confirm at least some of these
findings. For example, many of our EPA interviewees had noted that
data on in-stream water quality for toxic pollutants are generally
lacking (criterion 37). We were able to confirm the accuracy of
these responses through a review of the data in STORET, EPA's
principal repository of water quality data. For a group of streams
around the nation on which large manufacturing facilities are
situated, we found very few with any data for toxic pollutants.\2 In
fact, we found that only 9 percent of these streams had had any
in-stream data for organic toxic pollutants or pesticides entered in
STORET since 1988, despite the presence of major industrial
facilities discharging a variety of toxic pollutants into all these
waterways.
A lack of in-stream data can result in pollutants failing to be
identified or limited on NPDES permits. For example, when there is
no information about in-stream concentrations of pollutants being
considered for a water-quality-based limit (activity 5), the
in-stream concentration parameter is set to zero within the modeling
routine used to derive permit limits. For streams in which the
pollutant is present but is not reported, the absence of data
signifies that a greater amount of pollutant discharge from
facilities will be allowable than if valid in-stream data had been
available.\3 It may also lead to facilities not receiving limits
(restrictions) covering toxic pollutants although there is a water
quality concern in the receiving body of water. EPA acknowledges
some of these problems and has embarked on a program to modernize
STORET to address them.
--------------------
\2 These streams (river reaches) were selected because dischargers
within three significant industrial categories (pesticide
manufacturing, pharmaceutical manufacturing, and pulp and paper
manufacturing) are situated on them.
\3 Environmental Protection Agency, Technical Support Document for
Water Quality Based Toxics (Washington, D.C.: March 1991), p. 130.
SUMMARY
---------------------------------------------------------- Chapter 3:5
Our findings raise concerns about the quality of the information
supporting EPA's core activities controlling toxic discharges into
surface waters. Although 2 of the activities had strong quality
assurance practices associated with them, the 5 others were much
weaker, meeting about 58 percent of their critical quality assurance
criteria.
THE EXTENT TO WHICH LIMITS FOR
TOXIC SUBSTANCES ARE INCLUDED ON
NPDES PERMITS
============================================================ Chapter 4
In this chapter, we address our third evaluation question: To what
extent are limits for toxic substance discharges included on NPDES
permits? We answered this question by comparing information on
actual toxic pollutant discharges from three data sources with a list
of toxic pollutants controlled through NPDES permits for a sample of
facilities.
IDENTIFYING TOXICS TO BE
LIMITED ON NPDES PERMITS
---------------------------------------------------------- Chapter 4:1
We based the development of our sample of facilities to examine on
three industrial categories: pesticide manufacturing, pharmaceutical
manufacturing, and pulp and paper manufacturing.\1 We reviewed the
information on actual toxic effluent discharges for 236 facilities
within these categories and compared them to the facilities' permits.
Identifying which pollutants to control through the NPDES permit
process is the critical step in issuing permits. As stated earlier,
without an explicit permit limit on a pollutant, a facility is free
to discharge any amount. Hence, identifying the range of pollutants
that should be included on NPDES permits is essential for the toxic
control program to succeed.
It has been generally conceded that a discrepancy exists between the
number of pollutants that are regulated through permits and the
number that should be regulated.\2 However, this gap has not been
systematically examined. We do so here and report our findings in
this chapter for a sample of dischargers. It should be noted that
the existence of such a gap would not in itself suggest problems with
the permit process. That would be the case only if the discharge of
uncontrolled toxicants resulted in unacceptable risks to human health
and aquatic life. We address this issue in chapter 5.
--------------------
\1 We selected these three because EPA's effluent guidelines program
had recently conducted monitoring studies for facilities within these
categories.
\2 U.S. General Accounting Office, Water Pollution: Stronger
Efforts Needed by EPA to Control Toxic Water Pollution,
GAO/RCED-91-154 (Washington, D.C.: July 1991), p. 4.
INFORMATION SOURCES ABOUT
DISCHARGES
---------------------------------------------------------- Chapter 4:2
Our sources of information about toxic pollutant discharges for our
sample of facilities were (1) each facility's permit application (EPA
Form 2c); (2) EPA's Toxic Release Inventory (TRI), an annual
inventory of discharges from the largest manufacturing facilities in
the nation; and (3) discharge data collected under effluent
guidelines.\3 We pooled these three sources of information to
identify the toxic pollutants being discharged at the facilities in
our sample.\4 Each of these sources employs a different data
collection method. Consequently, using all three sources provides us
a more comprehensive characterization of the toxic pollutant
discharges from our sample of facilities. TRI in particular has a
different data collection method than the permit applications. Much
of the data it contains were derived from mass balance calculations
or published emissions factors. These data collection methods
complement the monitoring information contained within the other data
sources. Table 4.1 presents the major attributes of the three data
sources.
Table 4.1
Major Attributes of Three Sources of
Discharge Information
Attribute Permit applications Effluent guidelines TRI
----------- ------------------- ---------------------- ----------------------
Primary To identify To identify pollutants To identify the
purpose pollutants being and concentrations amounts of "hazardous
discharged for discharged by the most chemicals" that are
purposes of permit efficient facilities annually and routinely
issuance within an industry discharged into the
environment
Data source Approximately 7,000 Small sample within Approximately 2,300
major and 57,000 selected industries facilities discharging
minor facilities into surface waters
Frequency Every 5 years No schedule Annually
of
reporting
Time period Daily Daily and annually Annually
covered by
estimate
Number of Varies Typically, a series No requirements
samples for over 3 days
each
estimate
Sampling Not applicable Individually selected All facilities meeting
frame industries size requirements
discharging toxics
Pollutants Priority pollutants More than 450 Approximately 300
covered and pollutants pollutants and
80 others "if chemical groups
expected to
be present"
Principal Monitoring for Monitoring Mass balance,
basis of priority engineering
estimate pollutants; calculations, or
identification of monitoring
80 others "if
expected to be
present"\a
Who Facility EPA Facility
collects
sample
--------------------------------------------------------------------------------
\a Details of monitoring requirements are contained in section 122.21
of title 40 of the Code of Federal Regulations.
--------------------
\3 We used the TRI responses from 1989, the most recent available
when we conducted our analyses.
\4 We included 234 facilities from TRI. Because of the difficulty of
obtaining and reviewing permit applications, we drew a random sample
of 72 facilities for data from that source. The third source of
data, effluent guidelines studies, contributed data for approximately
32 facilities, which was the number of direct dischargers available
within that data set.
TOXICS DISCHARGED WITHOUT
PERMIT LIMITS
---------------------------------------------------------- Chapter 4:3
We found a large number of different pollutants being discharged by
one or more of the 236 facilities. One hundred and eighty-five
different toxic pollutants were reported to have been discharged from
the 236 facilities. Whereas some pollutants were discharged from a
single facility, others were reported from numerous facilities. For
example, chloroform, a common byproduct of industrial processes that
EPA views as a "probable" carcinogen, was reported to be discharged
by 91 of the 236 facilities. Table 4.2 presents the toxic pollutant
discharges we identified, indicates whether they are priority
pollutants or nonpriority pollutants, and shows the number of
facilities they were discharged from and the number that were
uncontrolled through the permit process across the 236 facilities in
our sample. Because of the proprietary nature of some of the
discharge data, we cannot present a facility-by-facility breakdown of
our results.
Table 4.2
The Extent of Discharge and Control of
Individual Toxic Pollutants, Priority
and Nonpriority
Faciliti
Faciliti es where
Priority es listed
pollutan releasin on
Pollutant t g permit
------------------------------ -------- -------- --------
1,1-dichloroethane No 1 1
1,1-dichloroethylene Yes 3 0
(vinylidene chloride)
1,1,1-trichloroethane Yes 7 3
1,1,2-trichloroethane Yes 1 1
1,1,2,2-tetrachloroethane Yes 1 1
1,2-dichloroethane Yes 12 6
1,2-butylene oxide No 2 0
1,2-dichlorobenzene Yes 1 1
1,2-dichloroethylene No 1 0
1,2-dichloropropane Yes 1 1
1,2,4-trimethylbenzene No 2 0
1,2,4-trichlorobenzene No 1 0
1,3-butadiene No 1 0
1,3-dichlorobenzene Yes 1 1
1,3-dichloropropylene Yes 1 0
1,4-dichlorobenzene Yes 1 1
2-hexanone No 1 0
2-phenylphenol No 1 0
2,3,7,8-TCDD (dioxin) Yes 5 1
2,4-D No 3 2
2,4-diaminotoluene No 1 0
2,4-dichlorophenol Yes 2 1
2,4-dimethylphenol Yes 1 0
2,4-dinitrophenol Yes 1 0
2,4,5-TP No 1 0
2,4,6-trichlorophenol Yes 1 0
3,3-dichlorobenzidine Yes 1 1
3,4,5-trichloroguaiacol No 1 0
3,4,5-trichlorocatechol No 2 0
4,4'-isopropylidene diphenol No 1 0
4,5-dichlorocatechol No 2 0
4,5-dichloroguaiacol No 2 0
4,5,6-trichloroguaiacol No 1 0
5,6-dichlorovanillin No 1 0
6-chlorovanillin No 1 0
Acetaldehyde No 5 0
Acetone No 108 1
Acetonitrile No 4 0
Acrolein Yes 1 0
Acrylamide No 2 0
Acrylic acid No 3 0
Acrylonitrile Yes 3 2
Alachlor No 1 1
Aluminum No 7 1
Ametryn No 1 0
Ammonia No 147 56
Ammonium nitrate No 7 0
Ammonium sulfate No 9 1\a
Aniline No 4 0
Antimony Yes 3 1
Arsenic Yes 9 1
Asbestos Yes 1 0
Atrazine Yes 1 0
Barium No 6 0
Benomyl No 1 0
Benzene Yes 5 2
Biphenyl No 6 0
Bis(2-chloroethyl) ether Yes 1 1
Bis(2 ethylhexyl) phthalate Yes 9 1
Boron No 3 0
Bromacil No 1 1
Bromoform Yes 2 1
Bromomethane Yes 1 1
Butachlor No 1 1
Butylamine No 1 0
Cadmium Yes 1 0
Calcium No 3 0
Captan No 2 0
Carbaryl No 3 1
Carbon disulfide No 2 1
Carbon tetrachloride Yes 6 4
Catechol No 72 0
Chlorinated phenols No 3 1\b
Chlorine No 44 19
Chlorine dioxide No 2 2\c
Chloroacetic acid No 2 0
Chlorobenzene Yes\d 4 3
Chloroethane No 2 0
Chlorofluoromethane No 1 0
Chloroform Yes 91 28
Chloromethane Yes 5 0
Chlorothalonil No 1 0
Chromium Yes 22 4
Cobalt No 1 0
Copper Yes 17 4
Cresol No 1 0
Cyanide Yes 12 8
Cyclohexane Yes 2 0
Dacthal No 1 0
DDT Yes 1 1
DEET No 1 1
Di-n-butyl phthalate Yes 1 0
Dicamba No 2 1
dichlorobromomethane No 1 0
Dichloromethane Yes 24 11
Dicofol No 1 0
Diethanolamine No 3 0
Dimethyl sulfone No 2 0
dimethylamine No 1 1
Diuron No 1 0
Epichlorohydrin No 2 0
Ethylbenzene Yes 11 2
Ethylene glycol No 27 0
Ethylene oxide No 1 0
Formaldehyde No 33 0
Gamma chlordane No 1 0
Glycol ethers No 6 0
Glyphosate No 2 2
Heptachlor epoxide Yes 1 0
Hexachlorobenzene Yes 1 0
Hexazinone No 1 1
Hydrochloric acid No 13 10
Hydrogen cyanide No 5 0
Iron No 3 1
Isopropyl alcohol No 1 0
Lead Yes 1 0
Magnesium No 16 0
Maleic anhydride No 1 0
Manganese No 18 2
Mercury Yes 6 2
Methanol No 39 0
Methomyl No 1 1
Methoxychlor No 1 0
Methyl methacrylate No 1 0
Methyl acrylate No 1 0
Methyl ethyl ketone No 9 0
(2-butanone)
Methyl isobutyl ketone No 5 0
Methyl mercaptan No 2 0
Molybdenum No 1 0
N-butyl alcohol No 11 0
Naphthalene Yes 2 1
Nickel Yes 10 1
Nitric acid No 3 3
Nitrilotriacetic acid No 1 0
Nitrobenzene Yes 2 1
Nitroglycerin No 1 0
N,n-dimethylaniline No 2 0
o-anisidine No 1 0
o-cresol No 2 0
o-toluidine No 1 0
o-xylene No 2 0
P-cresol No 1 0
P-xylene No 1 0
Pentachlorophenol Yes 1 0
Phenol Yes 24 11
Phenols No 14 6
Phosphoric acid No 3 3
Picric acid No 1 1
Prometon No 1 0
Prometryn No 1 0
Propachlor No 1 1
Propazine No 1 0
Propylene oxide No 2 0
Pyridine No 2 0
Sec-butyl alcohol No 1 0
Selenium Yes 1 0
Simazine No 1 0
Sodium No 3 0
Styrene No 5 0
Sulfuric acid No 19 19
Terbacil No 1 1
Terbuthylazine No 1 0
Terbutryn No 1 0
Terephthalic acid No 1 0
Tetrachlorocatechol No 2 0
Tetrachloroethylene Yes 1 1
Tetrahydrofuran No 2 0
Thallium Yes 1 0
Thiourea No 1 0
Tin No 1 1
Titanium No 4 0
Toluene Yes 22 9
Total organic halides No 3 0 \f
Toxaphene Yes 1 0
Tributyltin No 1 1
Trichloroethylene Yes 1 0
Trichlorosyringol No 1 0
Triethylamine No 2 0
Trifluralin No 2 1
Trimethylamine No 1 0
Vanadium No 6 0
Vinyl acetate No 1 0
Vinyl chloride Yes 2 0
Xylene (mixed isomers) No 19 2
Zinc Yes 46 18
============================================================
Total 1,217 285
------------------------------------------------------------
\a Ammonia and sulfate are controlled separately.
\b Phenol is listed on the permit.
\c Chlorine is listed on the permit for both facilities.
\d Chlorinated benzenes is a priority pollutant.
\e None of these facilities had permit limits for any organic halide.
We found that a large number of the toxic pollutants being discharged
were not listed on the NPDES permits of the facilities we studied.
Our unit of analysis for this assessment was a "pollution case." We
define pollution case as an individual pollutant released from a
single facility. When we looked at the 1,217 cases represented by
our sample, we found that 932 (77 percent) were not included on the
NPDES permit.
We also examined the discharge of uncontrolled toxic pollutants from
the perspective of individual facilities. We found that for 200 of
our 236 facilities (approximately 85 percent) the majority of the
toxic pollutants they discharged were not controlled through the
permit process.
PERMIT CONTROL BY CATEGORY OF
POLLUTANTS
---------------------------------------------------------- Chapter 4:4
One of the concerns raised about the EPA permit process is whether
there is too much emphasis on "priority" pollutants (discussed in
chapter 1). When we examined the information from the 236 facilities
to determine whether they indicated an emphasis on establishing
permit limits for priority pollutants, we found that the majority of
the discharge cases for our sample were nonpriority pollutants (67
percent): Of the 932 cases in which pollutant discharges were
uncontrolled, a high proportion (72 percent) were nonpriority
pollutants (see table 4.2). This means that the pollutant
category--nonpriority pollutants--that had the greatest number of
toxic discharges was also the least controlled.
Finding a bias toward limiting priority pollutants is to be expected
given the focus on priority pollutants in the permit application
itself (Form 2c), which asks the applicant to provide sampling
results for only priority pollutants. Other pollutants "that are
suspected to be present in the wastewater" are to be listed on a
separate page without monitoring results or even an estimate of
loading. This makes it difficult for the permit writer to assess the
pollutant for a limit. In a case such as this, the permit writer may
require the facility to monitor for the presence of the substance
over the course of the new permit and then reconsider whether to
issue a limit at the next reissuance (5 years later). However, an
EPA official told us that on the average a permit writer remains on
the job 18 months, less than one third of the 5-year permit cycle.
This is not enough time to acquire a perspective on individual
facilities. Moreover, permits have standard "reopener clauses"
allowing the agency to amend them during the course of the permit,
but we were told that these clauses are rarely used.
There is also an inconsistency within EPA's office of water in how
nonpriority pollutants are treated. The effluent guidelines activity
gathers sampling data on a broad set of pollutants (including, but
not restricted to, priority pollutants) in its monitoring
initiatives. This difference in treatment between the effluent
guidelines activity and the water-quality-based permit activity
appears to reflect an inconsistent attitude by EPA toward limiting
nonpriority pollutants on permits: a willingness to consider
technology-based limits for nonpriority pollutants (through effluent
guidelines) but little willingness to consider water-quality-based
limits for them.
Categorizing toxic pollutants as nonpriority pollutants should not
suggest that they are without human health or aquatic life risk. In
fact, the majority of the toxic pollutants reported in TRI are
nonpriority pollutants; they are listed there because they are
recognized as human health risks. What this means is that, since so
many toxic pollutants fall into the nonpriority category, EPA's
emphasis on priority pollutants is of limited value in resolving the
toxic water quality problems faced by the nation.
SUMMARY
---------------------------------------------------------- Chapter 4:5
The NPDES permit process is the major component of EPA's approach for
controlling toxic pollutant discharges to waterways. In this
chapter, we examined the extent to which permits control toxic
pollutants. We found that, for our sample of 236 facilities,
approximately 77 percent of the pollution cases we identified were
not controlled by permit limits. That is, approximately 77 percent
of the discharge of all pollutants across our sample was
uncontrolled. Through further review of these uncontrolled
discharges, we found that the vast majority were nonpriority
pollutants (72 percent) as opposed to priority pollutants. This
signifies that the pollutant category with the greatest number of
toxic discharges (812 of 1,217) was the least controlled.
In short, this finding makes clear that the permit application
process emphasizes the control of priority pollutants over others.
To determine whether and to what degree that is important, however,
one needs to examine whether not controlling these pollutants
actually results in human health or aquatic life risk. We address
this issue in the next chapter.
THE EFFECT OF THE UNCONTROLLED
DISCHARGE OF TOXIC SUBSTANCES ON
WATER QUALITY
============================================================ Chapter 5
In this chapter, we address our fourth evaluation question: What are
the effects on water quality of unlisted discharges of toxic
pollutants? Here, we estimated the in-water concentration of the
uncontrolled pollutants discharged from the 236 facilities within our
sample, using the data sources discussed in chapter 4, and then
compared the results to published water quality criteria for each
individual pollutant. In this way, we could determine whether the
cases of uncontrolled pollutant discharge violated either human
health or aquatic life criteria and thus were an unacceptable risk.
As we pointed out in chapter 4, the absence of a permit limit for a
discharged toxic pollutant is not necessarily a problem. EPA's
approach for issuing permits is based on the view that if a discharge
does not pose a "reasonable potential" to impair water quality, then
permit limits are not warranted. Whether a discharge poses a
"reasonable potential" depends on three factors: the toxicity of the
pollutant, the amount discharged, and characteristics of the water
body receiving the discharge. These three factors determine
pollutant risk and the consequent need for controlling it through
permits. In this chapter, we report that (1) of the pollution cases
we could evaluate, there were only a limited number within our sample
of facilities for which uncontrolled discharges of toxic substances
pose either a human health or aquatic life risk, and (2) EPA's
control of discharges of toxic pollutants is incomplete, because of a
widespread absence of water quality criteria needed for determining
whether a discharge is a human health or aquatic life risk. That is,
most of the toxic pollution that is discharged from all major
manufacturing facilities across the country is not covered by water
quality criteria and is therefore without regulatory oversight.
THE EFFECT OF TOXIC DISCHARGES
ON WATER QUALITY
---------------------------------------------------------- Chapter 5:1
The approach that EPA and the states use to decide whether permit
limits are necessary for a particular pollutant is to determine
whether a "reasonable potential" for a water quality standard will be
exceeded by that discharge's water concentration.\1 This
determination is made by estimating the in-stream concentration of
the discharge and comparing it against the water quality standard for
that pollutant.
--------------------
\1 Environmental Protection Agency, Technical Support Document for
Water Quality Based Toxics (Washington, D.C.: March 1991).
THE ANALYTICAL APPROACH
-------------------------------------------------------- Chapter 5:1.1
In order to evaluate the significance of the unlisted discharges from
the 236 facilities we examined, we conducted an analysis similar to
that routinely performed by permit authorities when they decide
whether or not to issue a permit limit for a particular pollutant
from a given facility. In conducting this analysis, we closely
followed the approach that is commonly used by permit authorities.
However, our approach is not identical to EPA's, as discussed below.
Any approach for assessing the effect of toxic discharges can be
evaluated only by estimating their concentration when diluted within
the receiving water body and comparing that value to a criterion of
maximum allowable concentration. Within the Clean Water Act approach
for controlling discharges of toxic substances, water quality
criteria and state standards define the maximum allowable
concentration.\2 Therefore, we could assess the risk of uncontrolled
discharges of toxic pollutants only when there was a water quality
criterion for a particular pollutant. We had proposed to answer this
question by estimating the extent to which water quality criteria are
threatened by all 932 cases of uncontrolled discharges of toxic
pollutants for the 236 facilities. However, we found that only a
relatively small portion of the uncontrolled discharges (357, or 38
percent) were pollutants that have water quality criteria.
Furthermore, for this analysis, we did not use the information
included on the permit applications because they were based on a very
small number of samples, very often one. We were not comfortable
estimating dilution concentrations on such small sample sizes.
Therefore, we did not assess 78 cases in which a pollutant with a
water quality criterion was characterized solely by a permit
application. We also excluded 102 cases in which the water quality
criterion is a function of time-dependent characteristics of the
water body. For example, the criterion for ammonia consists of a
table of values for varying temperatures and levels of pH within the
receiving water body. Calculating the criterion is a complex
exercise employing ambient data of varying availability and uncertain
quality. That left us with 177 uncontrolled cases whose risk we
could evaluate.
According to EPA publications, we found that, when establishing the
need for permit limits for a particular pollutant, EPA and the states
employ a water quality model of the following form:\3
where
C = downstream concentration
Cs = upstream concentration
Qs = upstream flow
Ce = effluent concentration
Qe = effluent flow
This model is used to estimate the diluted concentration of the
discharge when fully mixed in the receiving body of water. It
provides an estimate of ambient concentration under different
streamflow assumptions (Qs), typically at either the stream's
predicted mean flow or its predicted low flow (such as "7Q10," or the
stream flow that occurs seven times in any 10-year period). Because
of an absence of monitoring data for all cases to characterize the
upstream concentration, we dropped this variable from the model.
This is equivalent to setting upstream concentration equal to zero,
resulting in a conservative (or lower) estimate for downstream
concentration. Whether mean flow or low flow was assumed depends on
whether the equation is being used to identify threats to human
health or aquatic organisms. Low-flow is used when examining whether
there might be a violation of aquatic life criteria, and mean flow is
used when examining whether human health criteria might be violated.
We used this model and acquired estimates of low and mean flow for
rivers (on a "river reach" basis) from EPA. For water quality
criteria, we used EPA publications.
It is legitimate to use such models only for bodies of water with a
measurable flow--that is, rivers and streams rather than lakes or
estuaries. For this reason, we further restricted our assessment to
cases in which the facility discharges into a river or stream; this
eliminated 12 cases, reducing the total number of cases we could
evaluate to 165.
--------------------
\2 In a review of state standards for pollutants discharged from the
facilities we examined, we found a general agreement between them and
water quality criteria. Therefore, we employed the national water
quality criteria as the decision point within our assessment.
\3 Environmental Protection Agency, Technical Support Document.
RISK OF UNCONTROLLED CASES
-------------------------------------------------------- Chapter 5:1.2
We used the approach outlined above to assess the 165 cases and found
that for 14 of them (8 percent) the estimated in-stream concentration
exceeded the applicable water quality criterion. That is, in only 8
percent of the cases we could evaluate did we find that the
uncontrolled discharges of toxic pollutants posed a human health or
aquatic life risk. Table 5.1 presents the data for the 14 cases.
Because of the proprietary nature of these data, we are unable to
provide additional information about these releases.
Table 5.1
Fourteen Cases in Which Criteria Were
Exceeded
Aquatic Aquatic
Pollutan life life Human
Case t (acute) Result (chronic) Result health Result
---- -------- --------- --------- --------- --------- --------- ---------
1 Arsenic None None 0.14 Exceeded
2 Mercury 2.4 Exceeded 0.012 Exceeded 0.15 Exceeded
3 Chlorofo 28,900 Exceeded 1,240 Exceeded 470
rm
4 Mercury 2.4 0.012 Exceeded 0.15 Exceeded
5 Chlorine 19 11 Exceeded None
6 Chlorine 19 11 Exceeded None
7 Chlorine 19 Exceeded 11 Exceeded None
8 Clorofor 28,900 1,240 Exceeded 470
m
9 Hexachlo 6 Exceeded 3.68 Exceeded 0.00074 Exceeded
robenzen
e
10 Hydrogen 22 Exceeded 5.2 Exceeded None
cyanide
11 Chlorofo 28,900 1,240 Exceeded 470 Exceeded
rm
12 Chlorine 19 Exceeded 11 Exceeded None
13 Chlorofo 28,900 1,240 Exceeded 470 Exceeded
rm
14 Mercury 2.4 Exceeded 0.012 Exceeded 0.15 Exceeded
--------------------------------------------------------------------------------
\a Criteria are expressed in micrograms per liter.
As indicated above, we could not assess the risk posed by the
majority of our cases (62 percent) because of the absence of related
criteria. Therefore, we turned our attention to the issue of whether
there is a lack of criteria in general for toxic pollutants and
whether this constitutes a problem. EPA has not issued an aquatic
life criterion since 1980; the last human health criterion it issued
was in 1984 (for dioxin). However, toxicology research has not stood
still in the intervening years. Additional water quality needs (such
as bioaccumulation and sediments) are being identified that define
additional pollutants posing a threat to human health and aquatic
life. (This point is discussed later in this chapter.) Yet we found
that the list of pollutants with water quality criteria has remained
essentially unchanged since 1980.
There is no single delineation of toxic pollutants. The Clean Water
Act contains the nonspecific definition of any pollutant that is
harmful to organisms (at section 502). When the Congress directed
that EPA focus on toxic pollutants in 1977 amendments to the Clean
Water Act, it referred to a list of 31 chemical groups and 38
individual substances (in section 307(a)). Each of those groups of
pollutants is made up of numerous individual compounds of varying
toxicity and usage. For example, EPA selected a small number of the
numerous chlorinated benzenes in use to be on the priority pollutant
list and developed water quality criteria for very few of them.
Clearly, the coverage of pollutants within EPA's approach for
controlling discharges is incomplete. Because of the absence of a
single standard of toxicity, it is impossible to definitively
characterize the degree to which the list is incomplete.
In order to assess whether the lack of criteria posed a serious
"regulatory gap," we conducted two assessments. First, we defined a
population of toxic discharges at the national level and examined the
extent to which they were covered by criteria. Second, for the toxic
pollutants that were not covered by criteria, we examined the extent
to which they were discharged-- that is, the contribution these
pollutants make to the overall toxic pollutant discharge problem.
We used the list of pollutants listed in TRI as the basis for a
broader definition of "toxic pollutant" than that presented by the
126 priority pollutants. TRI was established through the Emergency
Planning and Community Right to Know Act of 1986 and contains a more
complete list of individual toxic pollutants than those specified by
the Congress in 1977. The inventory lists more than 300 individual
pollutants and 20 chemical categories (such as PCBs and lead
compounds). It is used within EPA and elsewhere to indicate the
incidence of toxic discharges. Again, these are toxic pollutants
that are tracked within TRI because they pose a human health risk.
TRI listed or tracked 195 toxic pollutants discharged to water. The
195 pollutants are listed in table 5.2.
Table 5.2
Water Quality Criteria for Pollutants
Discharged Into Surface Water in 1989
Pounds Freshwater Freshwater Human
Pollutant Priority discharged acute chronic health
--------------- ----------- ----------- ----------- ----------- -----------
Acetaldehyde 66,722
Acetamide 250
Acetone 1,023,408
Acetonitrile 91,876
Acrylamide 7,372
Acrylic acid 10,451
Acrylonitrile Yes 4,492 X X X
Allyl chloride 364
Aluminum 78,857 X X
Ammonia 24,546,136 X X
Ammonium 8,853,607
nitrate
Ammonium 69,031,944
sulfate
Aniline 14,844
o-Anisidine 4,949
p-Anisidine 250
Anthracene Yes 2,316 X
Antimony Yes 3,783 X X X
Arsenic Yes 1,754 X
Asbestos Yes 800
Barium 26,048
Benzamide 250
Benzene Yes 169,947 X X
Benzoyl 1,000
peroxide
Benzyl chloride 251
Beryllium Yes 372 X X X
Biphenyl 42,685
Bis(2- Yes 1,552 X
chloroethyl)
ether
Bis(2-chloro- Yes 12,000
1-methyl-
ether) ether
Bis(2- 2,453
ethylhexyl)
adipate
1,3-butadiene 143,434
Butyl acrylate 6,400
n-Butyl alcohol 943,547
sec-Butyl 6,411
alcohol
tert-Butyl 221,906
alcohol
Butyl benzyl Yes 1,028 X
phthalate
1,2-Butylene 4,139
oxide
Butyraldehyde 4,297
C.I. Basic 250
Green 4
C.I. Disperse 24
Yellow 3
Cadmium Yes 2,746 X X X
Captan 500
Carbaryl 750
Carbon 33,091
disulfide
Carbon Yes 16,396 X X
tetrachloride
Carbonyl 772
sulfide
Catechol 313,163
Chlordane Yes 4 X X X
Chlorine 2,403,657 X X
Chlorine 1,250
dioxide
Chloroacetic 1,524
acid
Chlorobenzene Yes 62,551 X
Chloroethane 71,749
Chloroform Yes 1,208,450 X X X
Chloromethane Yes 108,399 X
Chloroprene 9
Chlorothalonil 252
Chromium Yes 67,798 X X X
Cobalt 14,415
Copper Yes 101,105 X X X
p-Cresidine 250
Cresol (mixed 7,627
isomers)
m-Cresol 45
o-Cresol 311
p-Cresol 3,421
Cumene 10,088
Cumene 3,411
hydroperoxide
Cupferon 34
Cyclohexane 20,222
2,4-D 1,422
Decabromodiphen 3,450
yl oxide
4,4'- 595
Diaminodiphenyl
ether
Diaminotoluene 2,068
(mixed
isomers)
2,4- 250
Diaminotoluene
Dibenzofuran 447
1,2- 250
Dibromomethane
Dibutyl Yes 2,400
phthalate
Dichlorobenzene Yes 185 X X X
(mixed
isomers)
1,2- Yes 16,146 X
Dichlorobenzene
1,3- Yes 22 X
Dichlorobenzene
1,4- Yes 6,621 X
Dichlorobenzene
3,3- Yes 241 X
Dichlorobenzid
ine
1,2- Yes 227,614 X X X
Dichloroethane
1,2- 728
Dichloroethyle
ne
Dichloromethane Yes 229,620 X
2,4- Yes 78 X X
Dichlorophenol
1,2- Yes 14,977
Dichloropropane
1,3- Yes 340 X
Dichloropropyl
ene
Diethanolamine 591,555
Dicofol 250
Di(2- Yes 2,983 X X X
ethylhexyl)
phthalate
Diethyl Yes 9,163 X
phthalate
3,3'- 3
Dimethyoxybenz
idine
1,1-Dimethyl 250
hydrazine
2,4- Yes 218 X
Dimethylphenol
Dimethyl Yes 1,260 X
phthalate
Dimethyl 500
sulfate
4,6-Dinitro-o- 25
cresol
2,4- Yes 160,672 X
Dinitrophenol
2,4- Yes 12,657 X X X
Dinitrotoluene
2,6- 1,083
Dinitrotoluene
n-Dioctyl 1,185
phthalate
1,4-Dioxane 273,522
Epichlorohydrin 4,245
2- 96,042
Ethoxyethanol
Ethyl acrylate 1,188
Ethylbenzene Yes 16,923 X X
Ethylene 14,902
Ethylene glycol 3,773,670
Ethylene oxide 5,327
Formaldehyde 838,705
Freon 113 14,588
Heptachlor Yes 2 X X X
Hexachlorobenze Yes 338 X X X
ne
Hexachloro- Yes 622 X X X
1,3-butadiene
Hexachlorocyclo Yes 6 X X X
pentadiene
Hexachloroethan 421 X X X
e
Hydrazine 2,291
Hydrochloric 3,052,332 X
acid\a
Hydrogen 5,610
cyanide
Hydrogen 35,918
fluoride
Hydroquinone 4,884
Isobutyraldehyd 751
e
Isopropyl 11,008
alcohol
4,4'- 2,629
Isopropylidene-
diphenol
Lead Yes 33,314 X X
Maleic 2,824
anhydride
Manganese 148,561 X
Mercury Yes 1,555 X X X
Methanol 21,276,746
Methoxychlor 250 X
2- 46,428
Methyoxyethanol
Methyl acrylate 1,172
Methyl tert- 37,439
butyl ether
Methylenebis 506
4,4'- 1,305
Methylenediani
line
Methyl ethyl 67,797
ketone
Methyl iodide 1
Methyl isobutyl 449,410
ketone
Methyl 28,802
methacrylate
Molybdenum 124,535
trioxide
Naphthalene Yes 146,615 X X
Nickel Yes 86,211 X X X
Nitric acid\a 735,542 X X
Nitrilotriaceti 5,100
c acid
Nitrobenzene Yes 1,287 X X
Nitroglycerin 9,198
2-Nitrophenol 6
2-Nitropropane 2,700
N,N- 14,437
Dimethylaniline
N- Yes 9
Nitrosodipheny
lamine
Parathion 250 X X
Pentachlorophen Yes 2,559 X X X
ol
Peracetic acid 40
Phenol Yes 267,978 X X X
2-Phenylphenol 134
Phosgene 250
Phosphoric 26,961,174 X
acid\a
Phosphorus 3,033
Phthalic 2,120
anhydride
Picric acid 250
PCBs Yes 264 X X X
Propionaldehyde 411
Propylene 953
Propylene oxide 83,521
Pyridine 2,356
Quinoline 5
Quinone 12
Selenium Yes 750 X X X
Silver Yes 1,419 X X
Styrene 51,082
Sulfuric acid\a 19,800,811 X
1,1,2,2- Yes 5,429 X X
tetrachloroeth
ane
Tetrachloroethy Yes 54,940 X X
lene
Thiourea 971
Toluene Yes 179,797 X X
o-Toluidine 1,252
Trichlorfon 1
1,2,4- 4,729
Trichlorobenze
ne
1,1,1- Yes 27,549 X X
Trichloroethane
1,1,2- Yes 8,985 X X
Trichloroethane
Trichloroethyle Yes 16,065 X X X
ne
2,4,6- Yes 3,515 X X
Trichlorophenol
Trifluralin 322
1,2,4- 10,608
Trimethylbenze
ne
Vanadium 1,004
Vinyl acetate 5,449
Vinyl bromide 270
Vinyl chloride Yes 2,969 X
Vinylidene Yes 2,691 X
chloride
Xylene (mixed 185,752
isomers)
m-Xylene 2,933
o-Xylene 3,295
p-Xylene 2,225
2,6-Xylidene 1,906
Zinc Yes 134,700 X X
--------------------------------------------------------------------------------
SOURCE: Environmental Protection Agency, Toxic Release Inventory,
1989 (Washington, D.C.: 1991), and Water Quality Criteria Summary
(Washington, D.C.: 1991).
\a Criteria are for pH, not the acid itself.
Because of the pattern of permit coverage we identified in chapter 4,
we examined the extent to which criteria existed for the two
categories of toxic water pollutants (priority versus nonpriority)
monitored by TRI. We found that priority pollutants were, for the
most part, covered well by criteria. Of the 61 priority pollutants
reported, only 5 lacked any human health or aquatic life criteria.\4
(See table 5.2.)
Our analysis of the nonpriority pollutants reported in TRI had a very
different result. Here we found that 134 toxic pollutants were
reported. Of the 134, only 11 (approximately 8 percent) had any type
of criteria coverage. That is to say, 92 percent of the toxic
nonpriority pollutants reported in TRI had no criteria established
for them. (See table 5.2.)
In order to determine whether this lack of criteria coverage was
important, we analyzed the findings discussed above in terms of mass
pollutant discharges--that is, the number of pounds of toxic
pollutants reported discharged from TRI. Overall, approximately 190
million pounds of toxic pollutants were reported as discharged within
TRI. Of that amount, approximately 3.5 million pounds are priority
pollutant discharges. Consequently, approximately 98 percent of the
discharges by weight were nonpriority pollutants. Of this amount,
approximately 109 million pounds, or 56 percent, were not covered by
criteria.
We discussed the results of our assessment with EPA officials and
they agreed that the toxic control process is definitely slanted
toward controlling priority pollutants and that nonpriority
pollutants warrant coverage within the process.
In addition, there is reason to believe that the priority pollutant
list "that has served as a basis for numerous EPA actions" is too
restrictive.\5 Our review of agency documents suggests that several
pollutants beyond those EPA selected initially in 1980 and treated on
a priority status threaten the aquatic environment. Based on an
increasing concern toward the threats posed by pollutants that remain
in an ecosystem indefinitely, research has been conducted that
suggests that a number of toxic pollutants such as 4-bromophenyl
phenyl ether, octachlorostyrene, and photomirex may be of equal
concern as the older set of priority pollutants. EPA itself has
identified these substances as "bioaccumulative chemicals of
concern." And, as a result, the agency has placed them in a higher
risk category than most priority pollutants such as carbon
tetrachloride and cadmium in its recently proposed regional "Great
Lakes Water Quality Initiative." Human toxic effects and aquatic life
effects have been associated with several nonpriority pollutants for
which there are no water quality criteria. Examples are carbon
disulfide, ethylene glycol, formaldehyde, and xylene.\6 Xylene, for
example, is related to priority pollutants such as toluene and
ethylbenzene.
--------------------
\4 For the purpose of our assessment, we did not assume that every
pollutant required both a human health and an aquatic life criterion.
In some cases, a pollutant that could cause a threat to aquatic life
may not pose a human health threat. The reciprocal argument could
also be made. We determined that a pollutant was not covered by a
criterion when it lacked both human health and aquatic life criteria.
In addition, some pollutants do not have criteria established but
have, rather, a "lowest observable effect level" (LOEL). Strictly
speaking, LOELs are not criteria but are, in some cases, used in that
manner. We accepted the existence of a LOEL for a pollutant as
criteria coverage. Given these assumptions, our analysis is a
conservative, best case assessment.
\5 Environmental Protection Agency, "Water Quality Guidance for the
Great Lakes System and Correction; Proposed Rules," 58 Fed. Reg.
20801, at 20843 (April 16, 1993).
\6 Robert E. Gosselin et al., Clinical Toxicology of Commercial
Products, 5th ed. (Baltimore: Williams and Wilkins, 1984); Ernest
Hodgson, Richard B. Mailman, and Janice E. Chambers, Dictionary of
Toxicology (New York: Van Nostrand Reinhold Company, 1988).
SUMMARY
---------------------------------------------------------- Chapter 5:2
In this chapter, we addressed our fourth and last evaluation
question, concerning whether the uncontrolled toxic pollutants we
identified posed human health or aquatic life risk. We found that we
could assess only a minority of the uncontrolled toxics we identified
from our sample population of facilities in terms of risk: Most did
not have the human health or aquatic life criteria that are necessary
to determine whether a discharge is a risk. For the cases we could
evaluate, EPA and the states are generally effective at identifying
those that warrant permit limits. That is, only 8 percent of the
cases we could evaluate warranted permit limits when none existed.
However, we found that most of the pollutants discharged to the
nation's water, at least as represented by the population reported in
TRI, have no criteria established. This is a conclusion based upon
both the number of pollutants and the amount of pollution discharged.
Thus, many pollutants are not controlled through the permit process.
For the majority of toxic discharges to the nation's water, the
assessment necessary to determine whether a permit. limit is
required cannot be conducted because of the overwhelming lack of
human health and aquatic life criteria. Therefore, our conclusion is
that EPA's current approach has not yet proceeded far enough to
answer the critical threshold question of whether toxic pollutant
discharges pose a human health or aquatic life risk.
CONCLUSIONS, A MATTER FOR
CONSIDERATION, RECOMMENDATIONS,
AND AGENCY COMMENTS
============================================================ Chapter 6
CONCLUSIONS
---------------------------------------------------------- Chapter 6:1
The discharge of toxic pollutants to the nation's rivers and streams
poses both human health and aquatic life risks. This study has found
that the current EPA water toxics control program has major problems
in effectively controlling these risks. First, because the necessary
steps to ensure that program activities are supported by information
of acceptable quality are often not taken, the information produced
is questionable and the activities themselves are of uncertain
usefulness. Second, a wide range of toxic pollutants posing both
human health and aquatic life risk are not addressed by the permit
control process. Further, the EPA program has not established human
health or aquatic life criteria for many of the pollutants that are
discharged to the nation's waters. Consequently, when developing
permit limits for facilities, their risks cannot be assessed and they
are not regulated.
Given that these problems have been long-standing for the EPA water
quality program in general and that the likelihood of effectively
addressing them is uncertain, we also conclude that it is time to
question and reassess whether the basic strategy EPA uses to control
toxic discharges can be expected to produce the results envisaged by
the Clean Water Act. Therefore, we believe that the overall approach
now used for controlling toxic discharges should be reexamined. This
is especially important since controlling discharges of toxics from
point sources, although an important water quality concern, competes
with other causes of water quality impairment for scarce federal
funds. In addition, EPA and the states run their water quality
protection programs under stringent budgetary constraints. As a
result, we cannot assume that the program will be funded adequately
to effectively address the quality assurance and pollutant coverage
needs identified in this report.
Consequently, we provide a matter for congressional consideration.
In addition we make two recommendations specific to the problems we
identified. The matter for congressional consideration suggests
changing the general approach used for limiting discharges of toxic
pollutants into the nation's water, emphasizing pollution prevention.
Integrating pollution prevention principles into the current
standards approach may yield an improvement in water quality without
a significant increase in regulatory overhead.\1 A pollution
prevention approach would encourage a reduction in toxic discharges
by changing the system to make it in the interest of dischargers
themselves to limit the release of toxics. The two recommendations
address the lack of quality assurance and coverage of nonpriority
pollutants on permits.
--------------------
\1 Pollution prevention is discussed in U.S. General Accounting
Office, Pollution Prevention: EPA Should Reexamine the Objectives
and Sustainability of State Programs, GAO/PEMD-94-8 (Washington,
D.C.: January 25, 1994), and Water Pollution: Stronger Efforts
Needed by EPA to Control Toxic Water Pollution, GAO/RCED-91-154
(Washington, D.C.: July 19, 1991).
MATTER FOR CONGRESSIONAL
CONSIDERATION
---------------------------------------------------------- Chapter 6:2
The Clean Water Act emphasizes a standards approach for controlling
toxic pollutants. In its reauthorization of the act, the Congress
should consider augmenting this approach with additional authority to
allow EPA to emphasize pollution prevention as a way of managing
toxic pollutant discharges.
RECOMMENDATIONS
---------------------------------------------------------- Chapter 6:3
From our report's conclusions, we recommend that the Administrator of
EPA direct the Assistant Administrator for Water Quality to
initiate immediate efforts to address the information quality
assurance problems we identified in the 5 toxic control
activities in which these problems occur and
expand the use of the Toxic Release Inventory data base to identify
nonpriority pollutants being discharged to water that should be
considered for control through the permit process.
AGENCY COMMENTS AND OUR
RESPONSE
---------------------------------------------------------- Chapter 6:4
After its review of our draft report, EPA submitted several comments
to us, both general and specific. The latter have been incorporated
where appropriate. Here, we provide EPA's major comments and our
response to them.
EPA commented that to address the quality assurance issue, we needed
to extensively interview regional EPA and state officials. For the
most part, we did not do this. Consequently, EPA questioned the
validity of our conclusions. In conducting our evaluation, we
interviewed EPA headquarters' managers concerning the extent to which
their programs met critical quality assurance criteria. Each had the
opportunity during the interviews to indicate where they could not
provide information and whether we needed to discuss criteria
conformance with regional or other staff. When they indicated that
we did, we followed up with regional surveys and interviews. In
addition, the EPA official responsible for agency quality assurance
efforts indicated that our conclusions would have been the same had
we based our entire work on regional or state information.
EPA officials agreed with our finding that the scope of toxic
pollutant control is too narrow. However, they pointed out that the
risk to human health and the environment from not directly limiting
nonpriority pollutants is not well characterized in the report. EPA
suggested that we provide some additional information on risks.
Whether and to what degree nonpriority pollutants being discharged
from a facility pose human health and environmental risks must be
determined facility by facility. We attempted such an evaluation but
could not complete it because of the lack of standards for
nonpriority pollutants.
EPA officials further indicated that the report seems to equate the
lack of a specific effluent limit for a toxic pollutant to the lack
of control of that pollutant. They allege that considerable residual
control on nonpriority pollutants is accomplished through
specifically limiting priority pollutants. We believe that while
some residual control of nonpriority pollutants may occur, if it
does, it is a fortuitous result of the EPA program, which is focused
on the control of priority pollutants. When we pressed agency
officials concerning their evidence that such residual control did in
fact limit or eliminate human health or environmental risks
associated with nonpriority pollutants, or even references indicating
related studies, they could not provide us that information.
EPA officials commented that our report does not reflect the extent
to which the agency currently relies on TRI within the water quality
program. In 1989, EPA issued guidance that included a recommendation
to use TRI data to help identify impaired waters and point sources
for water-quality-based limits. However, this guidance focused only
on priority pollutants. EPA officials also commented that EPA is
currently revising the NPDES permit application forms for municipal
and industrial dischargers. These revised requirements may also
include reporting of TRI data. We recognize that EPA has, to some
extent, used TRI data in the water quality program; however, our
recommendation is that EPA use TRI data more extensively to identify
nonpriority pollutants that should be considered for control.
In evaluating the coverage of toxic discharges, EPA pointed out that
we focused on permits and discharges within three industries:
pesticide manufacturing, pharmaceutical manufacturing, and pulp and
paper production. EPA indicated that recently revised effluent
guidelines for these industries also regulate hundreds of nonpriority
pollutants as nonconventional pollutants. It also noted, however,
that although permits reflect national guidelines, they are renewed
only every 5 years and, as we noted, are rarely reopened. In
followup discussions with EPA, we learned that only the pesticides'
effluent guideline had been made final, in late 1993 after our
analytical work had been completed. EPA's estimate of when the pulp
and paper guideline will be final is within a year or two, while the
pharmaceutical guideline is estimated for February 1996.
Consequently, because permits have not been revised for the pesticide
industry and the other guidelines are still under development, we
believe our conclusions on the limited nature of pollutant control
coverage are valid.
EXPERT PANEL
=========================================================== Appendix I
Peter DeFur
Environmental Defense Fund
Washington, D.C.
Jeffrey Foran
Risk Sciences Institute
Washington, D.C.
Mary Jo Garreis
Maryland Department of Natural Resources
Baltimore, Md.
Steven Koorse
Hunton and Williams
Richmond, Va.
Richard Smith
U.S. Geological Survey
U.S. Department of the Interior
Reston, Va.
DESCRIPTIONS OF SEVEN ACTIVITIES
========================================================== Appendix II
We presented the principal EPA program activities designed to control
the discharge of toxic water pollutants in chapter 2. We also
identified their related information requirements in that chapter.
In this appendix, we characterize (1) what makes each activity an
important component of EPA's approach for limiting discharges of
toxics from point sources, (2) the role that the required information
plays in meeting the activity's objectives, and (3) the principal
sources of the required information.
ACTIVITY 1: EFFLUENT
GUIDELINES DEVELOPMENT
-------------------------------------------------------- Appendix II:1
Effluent guidelines are national standards establishing baseline
requirements for pollution removal for each individual industry.
They form the first line of defense within the Clean Water Act's
strategy for controlling discharges of pollutants from point sources
in that they take no factors into account for controlling effluent
discharge other than the technological capability of controlling
pollutant discharge from the facilities of an industrial sector.
THE BASIS FOR THE ACTIVITY
------------------------------------------------------ Appendix II:1.1
One of the foundations and major innovations of the Clean Water Act
strategy is to have facilities adopt a baseline level of control over
their pollutant discharges regardless of the quality of the water of
the receiving body. The result of this requirement is a system of
national technology-based effluent standards establishing baseline
treatment to be achieved by all dischargers within an industrial
sector. EPA has established effluent standards for 51 industrial
categories.
This effluent guideline in fact defines a required minimum level of
control for facilities within each industrial sector. Typically, the
standard establishes the maximum amount of particular pollutants
allowed to be discharged as a function of the production level of the
plant. These guidelines are incorporated as limits into permits for
dischargers.
The industrial technology division of EPA's office of water develops
these limits on the basis of common technological and economic
characteristics of individual industrial sectors. To select a
control technology for standard development, EPA typically delineates
a very well-defined subindustrial category within an industrial
sector--for example, the "paperboard from wastepaper" subcategory
within the "pulp, paper and paperboard point source category."\1
Thus, they select a sample of facilities within the industrial
subcategory, survey these facilities, conduct site visits, and define
the best available technology (BAT) economically achievable. (For
example, a current regulation for the BAT establishes that no more
than 0.87 pounds of pentachlorophenol and 0.30 pounds of
trichlorophenol can be discharged for each million pounds of paper
produced (40 C.F.R. 430.54).) All plants defined within this
subindustrial category would at a minimum be bound by these limits on
their NPDES permits.
Section 301(b) of the act establishes a mechanism for developing
these industry-specific effluent limits. For cases in which EPA has
not issued effluent guidelines, permit limits may be based on the
best professional judgment of the administering official.
The Clean Water Act directs that effluent limits for toxic substances
achieve best available technology economically achievable. Effluent
limits based on BAT must represent at a minimum the best control
technology performance in the subcategory that is technologically and
economically achievable.
--------------------
\1 EPA has issued guidelines for 25 subindustrial categories within
the pulp and paper point source category.
CRITICAL INFORMATION USED
WITHIN THE ACTIVITY
------------------------------------------------------ Appendix II:1.2
From interviews with EPA staff and our expert panel, we identified
four types of information required for developing effluent
guidelines:
removal (control) technology,
pollutants discharged,
water used in production, and
economic feasibility of regulatory alternatives.
THE ROLE OF EACH
INFORMATION TYPE
---------------------------------------------------- Appendix II:1.2.1
As used within the Clean Water Act, BAT determines the type of
information that EPA must gather to support this activity.
Information about the most efficient removal technologies currently
employed within an industry is critical to defining BAT. These
technologies need to be identified and evaluated so that their
applicability to the rest of the industry can be determined.
Technological efficiency is defined in terms of how well the removal
process can eliminate pollutants from the waste stream. Therefore,
the pollutants appearing in the discharges from these plants must be
identified and measured. One important method of reducing discharges
of pollutants is by using water more efficiently. In order to
identify these practices, EPA attempts to gather information from
plants that are efficient users of water, so that it can determine
whether BAT should be based on water use restrictions.
BAT must be affordable to most facilities within the industry. A
technology that would result in most plants being driven out of
business would not be consistent with the definition of BAT within
the statute. Therefore, economic information must be collected about
the industry as a whole to permit EPA to characterize the economic
effect on the industry of various BAT alternatives.
PRINCIPAL SOURCES OF
INFORMATION AND
DEVELOPMENT METHODS
---------------------------------------------------- Appendix II:1.2.2
Just as the definition of BAT within the Clean Water Act governs what
types of information should be gathered, it also influences how these
data are collected. Information about each of these data types is
derived through a three-step process. In the first step, EPA
attempts to define the size of the industrial sector that will be
covered by the standard. It does this through an examination of
extant data bases, such as Dun and Bradstreet's and U.S. census
data. This provides a sampling frame for data collection in the
second and third steps of the information gathering process. In the
second step, a questionnaire is administered to a sample of
facilities to characterize the range of the types of control or
removal technologies, their pollutant discharges and water use
employed, as well as the economic characteristics of the facilities.
In the third step of the process, EPA conducts site visits at a small
number of facilities. Through these site visits, the agency
characterizes more fully the removal technologies employed within
these specific plants and what pollutants they are discharging. It
identifies and measures all pollutants for which approved agency
methods governing wastewater analysis exist (currently over 400). It
also develops new methods for identifying pollutants as necessary.
ACTIVITY 2: WATER QUALITY
CRITERIA DEVELOPMENT
-------------------------------------------------------- Appendix II:2
Water quality criteria are standards that define the maximum
concentration of individual pollutants in water that will be
protective of human health and aquatic life.
THE BASIS FOR THE ACTIVITY
------------------------------------------------------ Appendix II:2.1
In addition to the technology-based requirements included in section
301 of the Clean Water Act, water-quality-based requirements are
established by sections 302 and 304(a)(1). The Congress recognized
that technology-based limits would not always meet water quality
needs. Indeed, there is no analytical relationship between
technology-based controls and the goals of the Clean Water Act to
prohibit "the discharge of toxic pollutants in toxic amounts"
(section 101(a)(3)). Consequently, the Congress included the
requirement that discharges must meet water quality needs.\2 That is,
if a pollutant that is discharged according to a technology-based
permit limit still results in a receiving body of water being
unacceptably polluted, then more stringent "water-quality-based"
limits are required (section 302).
Water quality criteria are numerical guidelines indicating the
maximum concentration of a substance that the agency advises is
protective of aquatic life or human health. Water quality criteria
themselves provide a basis for ambient water quality standards, which
in turn are used to establish water-quality-based permit limits (see
activity 5).
EPA develops water quality criteria for individual pollutants. There
are two types of criteria: criteria to protect aquatic life and
criteria to protect human health. There are separate aquatic life
criteria for freshwater and saltwater environments; both human health
and aquatic life criteria have separate values for "acute" (or
short-term) exposures and "chronic" (or long-term) exposures. EPA
has developed aquatic life criteria for 30 toxic pollutants and human
health criteria for 91 toxics. All the human health criteria and
most of the aquatic life criteria were developed between 1980 and
1985. The agency is currently reviewing several of them.
--------------------
\2 Water quality needs are interpreted to be support for the aquatic
life residing in, and the health of people who drink or fish from,
the body of water.
CRITICAL INFORMATION USED
WITHIN THE ACTIVITY
------------------------------------------------------ Appendix II:2.2
From interviews with EPA staff and our expert panel, we identified
five types of information necessary for the development of water
quality criteria:
aquatic life effects,
aquatic life exposure,
concentration of toxics in biota,
human health effects, and
human health exposure.
THE ROLE OF EACH
INFORMATION TYPE
---------------------------------------------------- Appendix II:2.2.1
Section 304(a)(1) of the Clean Water Act stipulates that criteria be
based on "the kind and extent of all identifiable effects on health
and welfare . . . from the presence of pollutants in any body of
water." Given that effects on aquatic life and human health can be so
varied, EPA developed an early protocol for establishing criteria.
Information about aquatic life effects and exposure is critical
within this program activity since criteria are based on the
concentration of a pollutant that is harmful to species in the
amounts at which they come into contact with the substance.
Bioconcentration information is necessary for determining in what
aquatic species and amounts specific toxic substances accumulate, so
that human exposure from consuming fish can be estimated.
Information about human health effects and exposure are critical for
calculating criteria to protect human health. As for aquatic life
criteria, these are based on estimates of toxicity and exposure.
PRINCIPAL SOURCES OF
INFORMATION
---------------------------------------------------- Appendix II:2.2.2
The aquatic life criteria are developed by EPA laboratories using
information on aquatic life effects, aquatic life exposure, and
concentration in biota. The human health criteria are developed by
EPA's Environmental Criteria and Assessment Office of the Office of
Research and Development in Cincinnati, Ohio, using human exposure
and human health effects information. The principal source of
information for both efforts are peer reviewed journal articles
describing laboratory studies. For aquatic life and human health
effects, the objective is to develop a minimum data set of studies
for a variety of different effects (and in the case of aquatic life
effects, different species). The laboratories sometimes conduct
research to fill gaps in information.
ACTIVITY 3: IMPAIRED WATER AND
POINT SOURCE IDENTIFICATION
-------------------------------------------------------- Appendix II:3
This program activity identifies waterways threatened by point source
discharges of toxic substances and the facilities causing the
impairment.
THE BASIS FOR THE ACTIVITY
------------------------------------------------------ Appendix II:3.1
Among the significant additions to the Clean Water Act in the 1987
amendments was section 304(l). In response to the concern that too
little headway was being made to control discharges of toxics from
point sources, the Congress directed each state to identify
"impaired" waters that are not expected to achieve or maintain water
quality standards because of toxic pollution from point sources, even
after application of technology-based effluent limits.\3 The lists
are to consist of the body of water and the facility causing the
impairment. Subsequently, an individual control strategy (ICS) is to
be established for the point source to remove the impairment. The
ICS is to consist of new limits or other conditions on the facility's
NPDES permit. So, section 304(l) (the short list) called for three
pieces of information: (1) an identification of bodies of water
where water quality standards for toxic pollutants are not expected
to be met because of point source discharges after technology-based
controls have been applied, (2) an identification of the contributing
facility or facilities, and (3) an identification of the new control
strategy ("individual control strategy") that would allow standards
to be met. Of these, the third is ostensibly the same as setting
permit limits, which is the fifth core program activity in our list.
Therefore, we discuss it under activity 5.
The section 304(l) exercise took place in 1988 and 1989. A total of
approximately 600 rivers and streams were named on the lists across
the country.
--------------------
\3 As we stated in chapter 2, section 304(l) called for several
lists. In this evaluation, we have identified the preparation of the
so-called "short list" because it concerns toxic pollution from point
sources rather than the broader delineations of impaired waters
contained in the other lists.
CRITICAL INFORMATION USED
WITHIN THE ACTIVITY
------------------------------------------------------ Appendix II:3.2
From interviews with EPA staff and our expert panel, we identified
six types of information necessary for issuing lists of waters and
source under section 304(l):
in-stream water quality,
aquatic life effects,
concentrations of toxics in sediments,
concentrations of toxics in biota,
discharges of toxics from individual sources, and
environmental fate and transport.
THE ROLE OF EACH
INFORMATION TYPE
---------------------------------------------------- Appendix II:3.2.1
The first four of these information components (in-stream water
quality, aquatic life effects, and concentrations in sediments and in
biota) provide different views of the health of an ecosystem and
indications of whether water quality standards are threatened. The
states use information about discharges from individual sources to
identify bodies of water that might be at risk and to identify the
sources of impairment. Information about environmental fate and
transport is used in a small number of cases to provide more reliable
estimates of in-stream concentration of toxics than is possible using
standard dilution models.
PRINCIPAL SOURCES OF
INFORMATION
---------------------------------------------------- Appendix II:3.2.2
The states are expected to use extant sources of information to
compile their lists of waters and point sources. The principal
sources of monitoring information are water quality data included in
STORET and other data bases, as well as previous assessments, notably
the biennial reports produced under section 305(b) of the Clean Water
Act. STORET (along with the related data bases: Biological
Information System and Ocean Data Evaluation System) are the major
storehouses of water quality monitoring data.\4 Information about
discharges is taken from NPDES permit applications and the PCS, which
includes monitoring data collected under NPDES. Under the Clean
Water Act, facilities must apply (and reapply every 5 years) for a
permit to discharge pollutants. On the application, the applicant is
expected to characterize the effluent discharge.\5 Required
monitoring reports from the facility are recorded on a monthly
discharge monitoring report and entered into PCS. Environmental fate
and transport information is obtained primarily from textbooks,
journal articles, EPA, U.S. Fish and Wildlife Service publications,
and EPA-supported water quality models.
--------------------
\4 EPA is undertaking a major modernization of STORET, in part
because of concerns about data quality.
\5 The permit application process is discussed in more detail in
chapter 3.
ACTIVITY 4: TOTAL MAXIMUM
DAILY LOAD DEVELOPMENT
-------------------------------------------------------- Appendix II:4
A TMDL provides an estimate of the amount of the pollutant that can
be safely discharged from all sources (including natural sources)
into a body of water. It is the sum of the individual allowable
allocations of the pollutant from point and nonpoint sources on a
stream as well as the amount transported from upstream sources.\6
--------------------
\6 For point sources, these are termed "wasteload allocations"
(WLAs), and for nonpoint sources they are termed "load allocations"
(LAs).
THE BASIS FOR THE ACTIVITY
------------------------------------------------------ Appendix II:4.1
Typically, several sources of pollution contribute to the water
quality problems of a given body of water. There may be several
industries and sewage treatment plants, as well as nonpoint sources
and the naturally occurring load (relevant for many toxic heavy
metals, although not for toxic organics) all discharging the same
pollutant into the same river. Taken individually, each of the
sources might not impair water quality enough to warrant the
imposition of permit limits for a specific pollutant. However, the
combined discharge of that pollutant could impair water quality. To
respond to the possibility that multiple sources could impair water
quality, the Congress included section 303(d) in the Clean Water Act.
This section directs the states to identify waters in which water
quality standards cannot be met through technological controls and to
estimate a "total maximum daily load" for the pollutants in question.
The objective of TMDLs is to allocate allowable discharge loads among
different sources of pollution. The TMDL estimates the pollutant
loadings from all sources and predicts the resulting pollutant
concentrations. It then establishes the permitted loads and sets the
base for establishing controls on sources. According to EPA, 70
percent of TMDLs have been developed in the course of establishing
permit limits under NPDES (see activity 5).
CRITICAL INFORMATION USED
WITHIN THE ACTIVITY
------------------------------------------------------ Appendix II:4.2
From interviews with EPA staff and our expert panel, we identified
six types of information necessary for developing TMDLs:
in-stream water quality,
discharges of toxics from individual sources,
hydrology (flow),
concentrations of toxics in sediments,
concentrations of toxics in biota, and
environmental fate and transport.
THE ROLE OF EACH
INFORMATION TYPE
---------------------------------------------------- Appendix II:4.2.1
The background concentrations of toxics in-stream and the discharges
are used to estimate the allowable maximum daily loads a receiving
water body is capable of assimilating while maintaining water
quality. Estimates of flow are employed in the mathematical models
used to establish TMDLs. Concentrations of toxics in sediments and
in biota are used to establish toxic concentrations in the background
and in water-dwelling organisms. Fate and transport information is
used in a relatively few instances in which a more precise estimate
of water quality effect is required than that provided by
straightforward dilution models and data are available to permit such
an assessment.
PRINCIPAL SOURCES OF
INFORMATION
---------------------------------------------------- Appendix II:4.2.2
Most flow data are collected at USGS gaging stations. However, there
is not a direct measure of flow for the "reach" to which most
facilities discharge. In these cases, the flow for the reach is
estimated by using data from gages on neighboring reaches.
Information about concentrations in compartments of the environment
(sediments, biota, and in-stream) are primarily acquired from
monitoring networks. Discharge data are provided by permit
applications and discharge monitoring reports. Fate and transport
information is obtained from textbooks, journal articles, EPA and
U.S. Fish and Wildlife Service publications, and EPA-supported water
quality models.
ACTIVITY 5:
WATER-QUALITY-BASED PERMIT
ISSUANCE
-------------------------------------------------------- Appendix II:5
NPDES permits are the legal documents that control point source
discharges of pollutants into the nation's waters.
THE BASIS FOR THE ACTIVITY
------------------------------------------------------ Appendix II:5.1
As we discussed above, the NPDES activity is the keystone within
EPA's program to limit discharges of toxics from point sources.
Within NPDES, all dischargers must acquire a permit to discharge. An
integral component of each permit is a set of limitations on the
amount of individual pollutants discharged.\7 The default basis for
setting permit limits is to apply the technology-based limits
established through section 301. Section 302 establishes an
alternative water- quality-based approach whenever "discharges . .
. would interfere with the attainment or maintenance of . . .
water quality." It is through this permit process that EPA and the
states control toxics.
The objective is to determine whether water quality standards are
threatened by the discharge of a particular substance and, if so,
what is the maximum amount of pollutant discharge that would maintain
them. EPA's recommended approach for writing water-quality-based
permit limits is described in the Technical Support Document for
Water Quality-Based Toxics Control. In short, it uses information
about the nature of the discharge and the receiving body of water to
determine whether there is a "reasonable potential" for water quality
standards to be exceeded by the discharge and, if so, to calculate
the amount that would protect them.
--------------------
\7 EPA and the states have begun to supplement individual limits with
requirements to test wastewater for "whole effluent toxicity" (WET).
WET limits are requirements to conduct standardized tests of the
wastewater on fish or other aquatic organisms. They give a measure
of the overall toxicity of the discharge, which is difficult to
accomplish using a chemical-by-chemical approach.
CRITICAL INFORMATION USED
WITHIN THE ACTIVITY
------------------------------------------------------ Appendix II:5.2
We identified four types of information necessary for issuing
water-quality-based permits:
discharges of toxics from individual sources,
hydrology (flow),
in-stream water quality, and
environmental fate and transport.
THE ROLE OF EACH
INFORMATION TYPE
---------------------------------------------------- Appendix II:5.2.1
A standard approach is employed to determine whether permit limits
are necessary and what those limits should be. Water quality
standards are used to justify the need for chemical-specific permit
limits and the amount allowed to be discharged through the permit.
The effluent has to be characterized (that is, the substances and
amounts) in order to identify the pollutants that warrant monitoring
requirements or limits on the permit. For this reason, an accurate
characterization of discharges is essential to the success of this
activity and the overall approach. The flow of the stream is
calculated in order to estimate the dilution concentration of the
toxic substance in the receiving water body.\8 An estimate of the
in-stream concentration of the substance prior to the discharge is
also used (where such data are available) so that the total
concentration of the toxic can be estimated. Although permit writers
usually use very simple models to estimate in-stream concentrations,
occasionally more sophisticated models employing fate and transport
information are used. These models offer a more precise estimate of
concentration.
--------------------
\8 Typically, permits employ discrete measures of flow, normally a
measure of low flow (7Q10) and a measure of average flow (the
harmonic mean flow).
PRINCIPAL SOURCES OF
INFORMATION
---------------------------------------------------- Appendix II:5.2.2
The sources for flow estimates are the same as within the TMDL
program. Information about concentrations in compartments of the
environment is primarily taken from monitoring networks. Discharge
data are taken from permit applications and discharge monitoring
reports. Fate and transport information is obtained from textbooks,
journal articles, EPA and U.S. Fish and Wildlife Service
publications, and EPA-supported water quality models.
ACTIVITY 6: "LOCAL LIMITS"
DEVELOPMENT
-------------------------------------------------------- Appendix II:6
"Local limits" POTWs (publicly owned treatment works) restrict the
amount of pollutants that indirect dischargers can release into the
sewer system to reflect site-specific concerns.
THE BASIS FOR THE ACTIVITY
------------------------------------------------------ Appendix II:6.1
POTWs are facilities that receive wastes from domestic and industrial
sources and treat them and discharge them into a body of water. The
industrial facilities that send their wastes to POTWs are termed
"indirect dischargers" to contrast them with dischargers, such as
POTWs themselves, that release their effluent directly into a
receiving body of water. A large portion of the total load of toxic
pollutants comes from indirect dischargers. According to EPA
estimates, these dischargers release a much larger volume of toxic
pollutants than do direct dischargers. The critical issue for the
program activity is how to ensure that the toxics discharged into
POTWs are effectively treated before being discharged.
The Clean Water Act instituted a separate system of controls for
indirect dischargers. The national pretreatment program was
established through section 307(b). The principal purpose of the
pretreatment program is to ensure that the wastes sent to POTWs can
be effectively treated by them without damaging them or passing
through into the receiving body of water.\9
To achieve these goals, about 1,500 POTWs have been given the
authority to write and enforce permits for their larger indirect
dischargers ("significant industrial users"), analogous to the
permits written by EPA or a state for direct dischargers under NPDES.
The first level of control is through technology-based limits. As is
the case for BAT with direct dischargers, EPA establishes categorical
limits for pretreatment industry by industry.
However, as is the case for direct dischargers, individual
circumstances vary among indirect dischargers and POTWs, preventing
categorical limits from always achieving the purposes of pretreatment
noted above. When this happens, the POTW has the authority to impose
"local limits" on individual indirect dischargers. Local limits are
derived by determining how much of each toxic pollutant is being
discharged and estimating how efficiently the POTW can remove them
from its own effluent.
--------------------
\9 Additional goals of the pretreatment program are preventing
pollutant interference with the POTW and improving opportunities to
recycle and reclaim wastewater and sludge.
REQUIRED INFORMATION USED
WITHIN THE ACTIVITY
------------------------------------------------------ Appendix II:6.2
We identified three types of information required for developing
local limits within the pretreatment program:
discharges of toxics from individual sources,
removal technologies, and
environmental fate and transport.
THE ROLE OF EACH
INFORMATION TYPE
---------------------------------------------------- Appendix II:6.2.1
This information is used to develop local limits. Information about
discharges from indirect dischargers is gathered to identify the
sources of the pollutants. Preventing interference or actual damage
to the POTW requires a more detailed understanding of the operations
of larger indirect dischargers ("significant industrial users").
Information about removal technologies is used to determine the
plant's efficiency at removing the pollutants that are found to be
discharged. Fate and transport data are gathered by POTWs to gain an
understanding of what happens to pollutants within the treatment
plant, how much precipitates, and how much volatilizes into the air.
PRINCIPAL SOURCES OF
INFORMATION
---------------------------------------------------- Appendix II:6.2.2
Information about discharges is obtained through a combination of
self-monitoring by the discharger and monitoring by the POTW (a
difference from the NPDES program in which permit authorities rarely
conduct their own monitoring at facilities). According to EPA, POTWs
conduct detailed inspections of the facilities of their significant
industrial users. Information about how the POTW is functioning and
what happens to the toxic substance once it enters the plant is
developed by the POTW itself.
ACTIVITY 7: COMPLIANCE PROBLEM
IDENTIFICATION
-------------------------------------------------------- Appendix II:7
NPDES permits require periodic self-monitoring of discharges. By
identifying compliance problems using the self-monitored data, EPA
marks the dischargers that violate their permit conditions for future
enforcement action.
THE BASIS FOR THE ACTIVITY
------------------------------------------------------ Appendix II:7.1
Once a permit has been issued, the holder of that permit is
responsible for attaining, monitoring, and maintaining compliance
with it. Failure to do so is a violation of the Clean Water Act.\10
EPA and authorized states are responsible for tracking compliance
with and enforcing permit requirements. Section 308 of the act
allows the states to impose requirements for sampling, analysis, and
recordkeeping and authorizes access for independent verification of
facilities' self-reports.
--------------------
\10 Recent GAO reports on this subject include Water Pollution
Monitoring: EPA's Permit Compliance System Could Be Used More
Effectively, GAO/IMTEC-92-58BR (Washington, D.C.: June 22, 1992),
and Environmental Enforcement: EPA Cannot Ensure the Accuracy of
Self-Reported Compliance Monitoring Data, GAO/RCED-93-21 (Washington,
D.C.: March 31, 1993).
CRITICAL INFORMATION USED
WITHIN THE ACTIVITY
------------------------------------------------------ Appendix II:7.2
We identified three types of information required for developing
water quality criteria:
discharges of toxics from individual sources,
aquatic life exposure, and
in-stream water quality.
THE ROLE OF EACH
INFORMATION TYPE
---------------------------------------------------- Appendix II:7.2.1
Decisions about compliance and noncompliance are made on the basis of
whether these self-reported data are within permit limits. Most
permit limits require monitoring of the amount of pollutant
discharged (whether pounds per day or milligrams per liter) per day.
Increasingly, permits are also including a requirement to test for
aquatic exposure or "whole effluent toxicity" (a biological test of
the overall toxicity of the effluent).\11 According to EPA,
occasionally permit limits are written that require in-stream
monitoring, which is a more direct measure of the effect of the
discharge on water quality.
--------------------
\11 The EPA respondent for this program coded these whole effluent
toxicity limits as "aquatic life exposure." WET limits are discussed
in chapter 3.
PRINCIPAL SOURCES OF
INFORMATION
---------------------------------------------------- Appendix II:7.2.2
The data used to determine compliance are gathered by each facility
and submitted to the permit authority according to the conditions
specified on the permit. The permit authority reviews the data and
enters them into EPA's PCS data base.
MAJOR CONTRIBUTORS TO THIS REPORT
========================================================= Appendix III
PROGRAM EVALUATION AND
METHODOLOGY DIVISION
------------------------------------------------------- Appendix III:1
Boris Kachura, Assistant Director
Dan Engelberg, Project Manager
Nila Garces-Osorio, Social Science Analyst