Environmental Information: Agencywide Policies and Procedures Are Needed
for EPA's Information Dissemination (Letter Report, 09/24/98,
GAO/RCED-98-245).
Pursuant to a legislative mandate, GAO determined the status of the
Environmental Protection Agency's (EPA) efforts to: (1) provide
communities with risk information on toxic chemical releases in their
areas; (2) make the Emergency Planning and Community Right-to-Know Act's
(EPCRA) chemical inventory information publicly accessible; and (3)
develop policies, procedures, and standards for disseminating
environmental information to the public.
GAO noted that: (1) EPA has three projects under way that would provide
additional data to communities on releases of toxic chemicals from
nearby manufacturing facilities; (2) although these projects are not
being designed to comprehensively define an individual community's
risks, collectively, they would substantially expand the information
available to communities; (3) in addition to the Toxic Release
Inventory's (TRI) quantities of chemical releases, this information is
to include data on individual facilities' history of compliance with
environmental laws, the relative toxicity of chemical releases, the
dispersion of the releases to surrounding areas, and the estimated
concentrations of the chemicals in the outdoor air from sources not
covered by the TRI; (4) however, each of these initiatives has a
different scope and timeframe for completion, and it could be several
years before the initiatives' full promise would be realized; (5)
although the data from the chemical inventory that is reported under
section 312 of the EPCRA are potentially useful for such purposes as a
citizen's finding out what chemicals are used at a nearby facility,
public use has been limited; (6) much of the information has not been
computerized to provide easy access and when it has, it is not available
in regional or national databases that permit comparisons among
industries or geographical areas; (7) EPA estimates that 868,500
facilities provide local emergency planning committees, fire
departments, and the states in which they are located with data on
thousands of hazardous chemicals; (8) in recent years, EPA has taken
some steps to assist local and state efforts to computerize the data,
and two EPA regions have initiated efforts to consolidate computerized
state databases; (9) while EPA believes that such efforts might prove to
be worthwhile, it has not provided funding for nor assessed the
potential benefits and costs of developing a national computerized
database for this information; (10) EPA has not developed policies,
procedures, and standards to govern key aspects of its projects to
disseminate information, such as the Sector Facility Indexing Project;
(11) EPA also has not developed standards to assess the data's accuracy
and mechanisms to determine and correct errors; and (12) while EPA has
several initiatives under way to improve its data management practices,
it has no specific plans to provide its program offices with guidance
for designing, developing, and implementing their information
dissemination projects.
--------------------------- Indexing Terms -----------------------------
REPORTNUM: RCED-98-245
TITLE: Environmental Information: Agencywide Policies and
Procedures Are Needed for EPA's Information Dissemination
DATE: 09/24/98
SUBJECT: Environmental law
Environmental impact statements
Federal/state relations
Toxic substances
Industrial wastes
Government information dissemination
Freedom of information
Health hazards
Pollution monitoring
Data bases
IDENTIFIER: EPA Toxic Release Inventory
EPA Relative Risk-Based Environmental Indicators Project
EPA Cumulative Exposure Project
EPA Sector Facility Indexing Project
******************************************************************
** 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. **
** **
** Please see the PDF (Portable Document Format) file, when **
** available, for a complete electronic file of the printed **
** document's contents. **
** **
** A printed copy of this report may be obtained from the GAO **
** Document Distribution Center. For further details, please **
** send an e-mail message to: **
** **
** **
** **
** with the message 'info' in the body. **
******************************************************************
Cover
================================================================ COVER
Report to Congressional Committees
September 1998
ENVIRONMENTAL INFORMATION -
AGENCYWIDE POLICIES AND PROCEDURES
ARE NEEDED FOR EPA'S INFORMATION
DISSEMINATION
GAO/RCED-98-245
EPA's Information Dissemination Activities
(160416)
Abbreviations
=============================================================== ABBREV
EPA - Environmental Protection Agency
EPCRA - Emergency Planning and Community Right-to-know Act
NAPA - National Academy of Public Administration
TRI - Toxic Release Inventory
Letter
=============================================================== LETTER
B-280580
September 24, 1998
The Honorable Christopher S. Bond
Chairman
The Honorable Barbara A. Mikulski
Ranking Minority Member
Subcommittee on VA, HUD,
and Independent Agencies
Committee on Appropriations
United States Senate
The Honorable Jerry Lewis
Chairman
The Honorable Louis Stokes
Ranking Minority Member
Subcommittee on VA, HUD,
and Independent Agencies
Committee on Appropriations
House of Representatives
U.S. industry uses hundreds of millions of pounds of toxic chemicals
each year in producing the nation's goods and services. Release of
these chemicals during their transport, storage, use, or disposal as
waste can potentially harm human health and the environment. In
1984, a catastrophic chemical accident at a plant in Bhopal, India,
killed thousands of residents, injured many others, and displaced
many more from their homes and businesses. In the wake of this
event, the Congress passed the Emergency Planning and Community
Right-to-Know Act of 1986 (EPCRA). Among other things, the act
provides access by individuals and communities to information
regarding hazardous materials in their localities. Section 313 of
the act generally requires facilities that manufacture, process, or
otherwise use toxic chemicals to report the amounts of various toxic
chemicals that they release to the environment and requires the
Environmental Protection Agency (EPA) to make this information
available to the public, which EPA has done in a national database
known as the Toxic Release Inventory (TRI). Under section 312 of the
act, employers must provide an inventory of hazardous chemicals
present at their facilities. This information must also be made
available to the public through state and local agencies.
Over the past several years, EPA has expanded the number of reporting
facilities and the number of chemicals for which data on their
releases are reported to the TRI. The agency is also considering
expanding the type of information reported for the TRI to include
detailed data on the use of these chemicals at these facilities.
Reflecting concern that communities need data on the risks posed by
local toxic releases and that EPCRA may already provide communities
with sufficient data on the use of chemicals at these facilities, the
conference report accompanying the VA, HUD, and Independent Agencies
fiscal year 1997 appropriations bill\1 and subsequent discussions
with your staff directed us to determine the status of EPA's efforts
to (1) provide communities with risk information on toxic chemical
releases in their areas; (2) make EPCRA's chemical inventory
information publicly accessible; and (3) develop policies,
procedures, and standards for disseminating environmental information
to the public.
--------------------
\1 House Report 104-812.
RESULTS IN BRIEF
------------------------------------------------------------ Letter :1
EPA has three projects under way that would provide additional data
to communities on releases of toxic chemicals from nearby
manufacturing facilities. Although these projects are not being
designed to comprehensively define an individual community's risks,
collectively, they would substantially expand the information
available to communities. In addition to the Toxic Release
Inventory's quantities of chemical releases, this information is to
include data on individual facilities' history of compliance with
environmental laws, the relative toxicity of chemical releases, the
dispersion of the releases to surrounding areas, and the estimated
concentrations of the chemicals in the outdoor air from sources not
covered by the Toxic Release Inventory, such as other facilities and
motor vehicles. However, each of these initiatives has a different
scope and time frame for completion, and it could be several years
before the initiatives' full promise would be realized. In addition,
the availability to the public of certain data in the first of these
projects--the Sector Facility Indexing Project--was delayed because
of concerns from state and industry officials about the data's
accuracy and how the information may be interpreted by the public.
Although the data from the chemical inventory that is reported under
section 312 of the Emergency Planning and Community Right-to-Know Act
are potentially useful for such purposes as a citizen's finding out
what chemicals are used at a nearby facility, public use has been
limited. Much of the information has not been computerized to
provide easy access and when it has, it is not available in regional
or national databases that permit comparisons among industries or
geographical areas. EPA estimates that 868,500 facilities provide
local emergency planning committees, fire departments, and the states
in which they are located with data on thousands of hazardous
chemicals. Industry representatives have stated that the substantial
costs of providing the information would be better justified if the
public made more use of it. In recent years, EPA has taken some
steps to assist local and state efforts to computerize the data, and
two EPA regions have initiated efforts to consolidate computerized
state databases. While EPA believes that such efforts might prove to
be worthwhile, it has not provided funding for nor assessed the
potential benefits and costs of developing a national computerized
database for this information.
EPA has not developed policies, procedures, and standards to govern
key aspects of its projects to disseminate information, such as the
Sector Facility Indexing Project. For example, EPA does not have in
place agencywide policies and procedures specifying standards and
detailed guidance for analyzing whether the information to be
disseminated meets users' needs and for obtaining stakeholders'
involvement in designing the projects. EPA also has not developed
standards to assess the data's accuracy and mechanisms to determine
and correct errors. While EPA has several initiatives under way to
improve its data management practices, it has no specific plans to
provide its program offices with guidance for designing, developing,
and implementing their information dissemination projects.
BACKGROUND
------------------------------------------------------------ Letter :2
Section 313 of EPCRA generally requires facilities at which toxic
chemicals are manufactured, processed, or otherwise used to report
annually to EPA and the states on, among other things, releases of
these substances. The requirement applies to facilities with 10 or
more full-time employees in specified industries that exceed the
chemical reporting thresholds specified in the act. EPA makes the
data in the TRI available and accessible to the public in various
formats, including a computerized database on the Internet. The
Pollution Prevention Act of 1990 expanded the information collected
in the TRI to include data on the industries' efforts to reduce
pollution at its source and on recycling. EPA has further expanded
the TRI by, in November 1994, requiring reports on additional
chemicals and, in May 1997, requiring reports by additional
industrial groups. In October 1996, EPA also announced that it was
considering an expansion that would require industries to report the
amounts of toxic chemicals entering a facility, transferred into
products and waste, and leaving the facility. This concept has been
referred to as "materials accounting" or "chemical use data." EPA
expects to propose a rule on this requirement in 1998.
The TRI currently contains data on the amounts of over 600 chemicals
that have been emitted to the environment (the air, water, or land)
and/or transferred off-site as waste. Although this information on a
nearby facility's releases provide some indication of potential risk
to human health and the environment, local communities also need
information on the chemicals' toxicity--the degree of danger to
animal or plant life--and the extent of their exposure to the
releases to more fully understand the risks. Even the most toxic
chemicals do not cause harm to an individual unless sufficient
exposure occurs. More specifically, comprehensive risk information
includes data on (1) what chemicals have been released, how, where,
and in what amounts; (2) what toxicities are associated with exposure
to each chemical and how toxic that chemical is; and (3) who has been
exposed to the chemical and how often, to how much, and for how long.
The TRI's information on the amount of releases represents estimated
aggregate amounts for a full reporting year. Additional details on
the duration or timing of these releases would be needed to more
fully understand the risks. For example, a chemical release may be
more severe if the releases are concentrated over a short period,
rather than occurring in smaller amounts over an entire reporting
year.
Almost from the start of the TRI program in 1987, EPA has been asked
by communities and other users to expand the inventory to provide
them with more information on the risks posed by the chemical
releases. More recently, in its April 1995 report on EPA, the
National Academy of Public Administration (NAPA) recommended that the
agency add risk factors to the TRI.\2 According to NAPA, the TRI is a
useful, but incomplete, tool to inform the public and company
officials about toxic releases. NAPA said that, by linking risk
information to the inventory's chemical reports, businesses would
have an incentive to reduce the most hazardous emissions first.
Under section 312 of EPCRA, employers must annually submit a
hazardous chemical inventory form to designated state and local
emergency planning organizations, as well as local fire
departments.\3 The form generally contains information regarding the
amount of hazardous chemicals present at a facility, by category, as
well as their general location. The designated recipients of the
inventory form may request more detailed information, such as
specific chemical identities and exact locations, from individual
facilities. The public may generally obtain information submitted
under section 312, although the exact locations of chemicals must be
withheld under certain circumstances. While section 313 of EPCRA
requires TRI reporting on the emissions of over 600 chemicals by
approximately 22,000 facilities, section 312 of EPCRA requires
reporting on thousands of hazardous chemicals that are present at an
estimated 868,500 manufacturing and nonmanufacturing facilities.
In September 1997, EPA announced, as one of its strategic goals, the
expansion of the public's right to know about the environment. In
addition, EPA announced, as a principle to guide senior management's
decision-making and priority setting, that agency actions should
maximize public participation and community right-to-know efforts.
In making this announcement EPA stated its intent to empower state,
local, and tribal governments and the American public by providing
citizens with information to help them make informed decisions
regarding environmental issues affecting their communities. To do
this, EPA said that it would expand the content of its databases,
improve the data's quality and usability, and make the data widely
available through the Internet and other sources.
--------------------
\2 Setting Priorities, Getting Results: A New Direction for EPA,
NAPA (Apr. 1995).
\3 Under section 301 of EPCRA, the governor of each state appoints a
state emergency response commission, which then designates local
emergency planning districts and appoints a local emergency planning
committee for each district.
EPA'S EFFORTS TO PROVIDE
COMMUNITIES WITH INFORMATION ON
THE RISKS POSED BY TOXIC
CHEMICAL RELEASES
------------------------------------------------------------ Letter :3
Although EPA does not have any plans to expand the TRI to include
information on the human health and environmental risks posed by
toxic chemical releases, the agency has three projects under way that
will provide communities with substantially more data on nearby
facilities, the relative toxicity of their chemical releases, and the
potential exposure to the releases. Two of these projects--the
Relative Risk-Based Environmental Indicators Project and the
Cumulative Exposure Project--are efforts by EPA to use toxic release
data and other information to improve the agency's consideration of
potential health and environmental risks in setting priorities and
developing policies, primarily to identify specific chemicals,
sources of chemical emissions, or geographic areas for priority
action to reduce risks. The other project--the Sector Facility
Indexing Project--provides information on the facilities and their
environmental performance, including the amounts of their toxic
releases and their history of compliance with environmental laws.
EPA recognizes that these projects separately and collectively are
not definitive assessments of risks to individual communities.
Nonetheless, the projects could expand the information available to
the public on the facilities and toxic releases in their areas.
EPA's plans to make the earliest of these projects--the Sector
Facility Indexing Project--publicly available generated considerable
concern on the part of industry and some states over the accuracy and
appropriateness of the data for public use. EPA made substantial
efforts to identify and correct the data's inaccuracies and decided
to delay including the information on the toxicity of chemical
releases, which was controversial with industry and the states,
before the project's data were made available on the Internet in May
1998. Unless addressed first, similar issues about stakeholders'
involvement, the data's accuracy, and how the information may be
interpreted by the public are likely as EPA makes the results of
other projects available to the public. For example, the toxicity
information that EPA was considering for the Indexing Project is a
component of the Relative Risk-Based Environmental Indicators
Project.
EPA DOES NOT PLAN TO ADD
COMPREHENSIVE RISK
INFORMATION TO THE TRI
---------------------------------------------------------- Letter :3.1
According to TRI program officials, EPA has no plans to expand the
data in the TRI to incorporate toxicity and exposure information.
The officials said that, when the Congress was considering the
passage of EPCRA, the issue of whether the TRI should contain risk
information was debated and it was decided that information on the
amounts of toxic releases, which are referred to as hazard
information, would be sufficient. The officials further stated that
the TRI has worked well in encouraging industry to reduce toxic
releases and that it is not practicable to develop and add the
information that would be needed to provide the public with accurate
assessments of the risks from TRI releases in specific communities.
In its annual summary reports of TRI releases, EPA has added
discussions on the potential health and environmental effects of the
covered chemicals and on the factors involved in assessing risks.
General information on the use and the potential effects of chemicals
is also available from various other sources within and outside EPA.
For various toxic chemicals, the agency is preparing fact sheets that
describe how they are generally used, how exposure to them might
occur, what happens to them in the environment, and how they affect
human health and the environment.
The information needed to assess risks is often not available. Data
on the amounts, the durations, and the methods of individuals'
exposure to chemicals is generally limited, and little is known about
the toxic effects of many of the chemicals used in commerce.
According to EPA, for 43 percent of the chemicals produced in high
volumes, no data from tests on their basic toxicity currently exist.
In April 1998, Vice President Gore directed EPA to proceed with the
Chemical Right-to-Know Initiative, which is to accelerate the
collection and dissemination of information about widely used
chemicals to which people, especially children, may be exposed.
Major aspects of the initiative involve getting industries to provide
more complete test data on chemicals that are produced in high
volumes and to perform additional testing for chemicals that children
are most likely to encounter as well as having EPA review persistent
chemicals that accumulate in body tissues to determine whether these
chemicals should be subject to TRI reporting or to lower thresholds
for reporting.
THE SECTOR FACILITY INDEXING
PROJECT
---------------------------------------------------------- Letter :3.2
The Sector Facility Indexing Project, which was initiated by EPA's
Office of Enforcement and Compliance Assurance in 1995, provides
extensive information through the Internet on over 600 facilities in
five major industries: (1) auto assembly; (2) iron and steel
production; (3) petroleum refining; (4) pulp manufacturing; and (5)
primary smelting and refining of aluminum, copper, lead, and zinc.
The project consolidates information that has been available to the
public through different data systems, publications, and several
places on the Internet. For each facility, the project provides
information on its location, production or production capacity,
surrounding populations, and permits held under major environmental
programs; the number of inspections received; its record of
compliance with federal regulations; and any chemical releases,
spills, or transfers off-site. According to EPA, facilities can use
this project to compare their data against those of similar
facilities or simply to monitor their own regulatory performance.
Government agencies can use the information as a planning tool, for
example, to identify facilities for assistance in complying with
regulations. Environmental and community groups will have easier
access to information that they can use to learn about the
environmental performance of facilities near them.
EPA plans to evaluate the project and may then expand it to include
other industry sectors and more types of data. For example, the
agency is examining ways to add data about chemicals that are
reported under other statutes, such as the Clean Air Act, and data on
the toxicity and relative risks of toxic releases--this information
is being considered under the Relative Risk-Based Environmental
Indicators Project. EPA had earlier proposed to include the toxicity
information, but industry and states expressed concerns about the
accuracy of the information and whether it would mislead the public
about the risks posed by chemical releases. According to a project
official, the agency received comments from many stakeholders that
the toxicity data did not go far enough in examining the potential
risks and that the risk components should be factored into the
project along with the toxicity information. The project official
said that toxicity information allows users to examine where
potential hazards might be without respect to whether the population
might be affected, whereas relative risk-based analysis examines
potential interactions between chemical releases, toxicity, weather
patterns, chemical dispersion properties, and surrounding
populations. The official stated that incorporating relative
risk-based information into the project is a long-term goal.
THE RELATIVE RISK-BASED
ENVIRONMENTAL INDICATORS
PROJECT
---------------------------------------------------------- Letter :3.3
EPA's Office of Pollution Prevention and Toxics initiated the
Relative Risk-Based Environmental Indicators Project to use TRI's
release information as a measure of the impact of EPA's efforts to
improve the environment. The project involves developing a computer
model that assigns numeric values to individual risk elements, such
as the amount of chemical releases, the chemical's toxicity, and
estimates of exposure and exposed populations, so one chemical
release can be compared to another. For example, a value is assigned
for the toxicity of the chemical release from a weighting index that
reflects the toxicity of the chemical relative to others. Similar
values are calculated for the other elements and these can then be
added together to arrive at a "risk-based" value for a particular
release at a facility. In turn, the values for all of a facility's
releases can be summed to be analyzed for trends over time or
compared with those of other facilities.\4
According to EPA, the project will enable potential users to analyze
the relative risks of releases by medium (i.e., air, water, or land),
chemical, geographic area, industry sector, specific facility, or a
combination of these and other variables. As a result, users will be
able to examine trends or to rank and prioritize the releases for
strategic planning, risk-related targeting for enforcement and
compliance, and community-based environmental protection purposes.
The model is being tested by EPA's regions, states, and tribal groups
and may be made available to a wide audience of users later this
year. The model, however, will only provide an indicator value for
chronic human health effects through exposure via the air. It will
not address acute human health or ecological effects or other
pathways of exposure, such as water.\5 The model is expected to
eventually (1) provide for potential chronic and acute human health
and ecological effects and other exposure pathways and (2)
incorporate additional census data that will allow users to analyze
the effects of chemical releases on exposed populations by such
demographics as race, age, and income. According to EPA, the
indicators model must often rely heavily on certain assumptions about
individual sites because the information is generally not available.
Because the model does not produce a formal risk assessment, it is to
be used primarily as a risk-screening tool rather than an attempt to
quantify the potential health risks to individuals at the community
level.
--------------------
\4 The project's purpose is not to formally assess the risks from
chemical releases in order to estimate the number of illnesses
potentially resulting from a specific toxic release, for example.
\5 Acute toxicity refers to any poisonous effect produced by a single
short-term exposure that results in severe biological harm or death.
Chronic toxicity is the capacity of a substance to cause long-term
poisonous human health effects.
THE CUMULATIVE EXPOSURE
PROJECT
---------------------------------------------------------- Letter :3.4
EPA's Office of Policy, Planning, and Evaluation is developing the
Cumulative Exposure Project as a priority-setting and policy
development tool. The project is intended to use existing data and
methods to estimate a national distribution of cumulative exposures
to environmental pollutants, including those reported under the TRI.
Because people tend to be exposed through multiple ways to numerous
pollutants originating from a variety of sources, EPA proposes to
estimate cumulative exposures by combining measured and modeled
concentrations of pollutants in the air, food, and drinking water
with human activity and consumption patterns. (The TRI's data are
used only for the project's air pollutants component.) The ultimate
goal is to develop analyses of multiple exposures and multiple
pollutants, thereby providing EPA with the ability to identify the
potentially most significant environmental exposures (among those
considered in the project) and the most affected communities or
demographic groups. According to EPA, this information will enhance
the consideration of cumulative exposures to pollutants in developing
environmental policy.
The project's air pollutants component uses the TRI, other data from
EPA on emissions, and a model from EPA on the dispersion of
pollutants in the atmosphere to estimate the outdoor concentrations
of 148 hazardous air pollutants (also referred to as air toxics) that
are regulated under the Clean Air Act. In addition to enabling users
to compare concentrations across regions, states, and census tracts,
estimates can be developed for subpopulations and for the relative
contributions to outdoor concentrations from broad sectors of the
economy, such as transportation, manufacturing, and waste
management.\6
The significance of the concentrations for a specific census tract
can be determined by comparing them to a set of benchmark
concentrations derived by EPA from available data for carcinogens
(cancer-causing agents) and noncarcinogens.\7
Officials from the Cumulative Exposure Project expect to complete the
air toxics component and make the results publicly available through
the Internet by the end of calendar year 1998. This component of the
project will provide estimates of the concentrations in the outdoor
air of the 148 hazardous air pollutants for 1990 only. EPA's Office
of Air and Radiation plans to use the project's model to measure the
success of its efforts to control air pollution in reducing exposure
to hazardous air pollutants. Data for succeeding years will be
needed to determine trends and progress, and the office is planning
updating the data. According to project officials, the database is
large and adding additional years will be labor-intensive. In
addition, EPA's Science Advisory Board, which was established in the
Office of the Administrator to provide advice on scientific matters,
urged the agency to expand its efforts to measure concentrations of
air toxics as part of its work to assess cumulative outdoor levels of
air toxics. The Board concluded that the overall conceptual
framework for the project was sound but noted that the project
suffers or will be handicapped, at least in the near term, from
limitations in available data based on actual measurements of
concentrations of air toxics.
--------------------
\6 For census-taking purposes, the United States is divided into over
60,000 tracts.
\7 These previously defined benchmark concentrations for cancer and
noncancer health effects are based on standard toxicological
references and represent levels of air toxics above which health
risks may occur.
OPPORTUNITIES EXIST TO INCREASE
ACCESSIBILITY OF CHEMICAL
INVENTORY DATA
------------------------------------------------------------ Letter :4
Although the data reported under section 312 of EPCRA can be valuable
to local emergency planning committees in their efforts to develop
emergency plans and to reduce emergency personnel's exposure to
harmful chemicals, the use of the data by the wider public has been
limited. The data are compiled on a community and state basis and
often are not computerized for easy access. Moreover, the lack of an
integrated database makes it extremely difficult, if not impossible,
to compare facilities within an industry or to perform regional or
national studies or comparisons. During the past few years, two EPA
regions have initiated projects to integrate state databases. One of
these projects also was intended to assess the potential for a
national chemical inventory database. This project was terminated
due to other priorities and then resumed with its purpose solely to
create a regional database. Although EPA considers a national
computerized database potentially worthwhile, the agency has no
specific plans to assess the feasibility for such a database.
CHEMICAL INVENTORY DATA CAN
BE VALUABLE, BUT THEIR USE
IS LIMITED
---------------------------------------------------------- Letter :4.1
Local emergency planners can use the chemical inventory data to
develop plans needed to respond to emergencies, such as spills of
hazardous materials in factories. Fire departments and other
emergency responders have access to the data to help develop response
plans before they arrive at the scene of a chemical accident or at a
fire at a facility using hazardous chemicals. People who are
considering buying or renting housing nearby also can use the data to
learn about the chemicals that are present in that community.
Individual citizens, as well as various local groups, can also use
chemical inventory data to improve their ability to protect human
health and the environment by engaging in dialogues with industry
representatives about reducing chemical risks, preventing accidents,
and limiting chemical exposure. According to local emergency
planning officials, environmental consultants and attorneys also have
requested this information to perform environmental site assessments
in compliance with federal and state laws, and environmental groups
have requested it to perform studies of chemical risks. The news
media may also be interested in this information to inform the public
about chemical releases that may have occurred during accidents.
Although chemical inventory data can be useful to local citizens, the
information has not been used extensively. A 1994 nationwide study
performed for EPA under a cooperative agreement found that most local
emergency planning committees received few inquiries from local
residents for the data. During the period from June 1993 through
June 1994, about 80 percent of the local emergency planning
committees that were considered to be functioning received six or
fewer inquiries, and more than 40 percent received no inquiries.
Eight of the 10 officials of state emergency planning commissions and
19 of the 20 officials of local emergency planning committees that we
talked to said that demand for the data was low in their geographic
areas.\8
Industry representatives have expressed concern about this low use,
considering their costs to report the data. EPA has estimated that
providing the data will cost industry $247 million during the period
from February 1997 through January 2000, and industry representatives
have maintained that the costs would be better justified if EPA took
actions to improve access to and use of the data. For example, a
major oil refiner, concerned about the cost and limited use of the
data, suggested that EPA take responsibility for ensuring that the
data are computerized to improve access to them. Furthermore, a
major industrial trade association said that EPA should make better
use of the data before it seeks additional information from industry,
such as through its plans to expand reporting requirements for the
TRI.
--------------------
\8 For each of the 10 states having the highest emissions of
chemicals according to EPA's 1995 report on the TRI, we interviewed
an official of the state emergency response commission and officials
of two local emergency planning committees. The planning committees
selected were among those most notable for their efforts in
conducting public outreach on the availability of chemical inventory
data, according to the state emergency planning commissions.
COMPUTERIZATION COULD
ENHANCE THE DATA'S
USEFULNESS
---------------------------------------------------------- Letter :4.2
While local emergency planning committees and state emergency
response commissions are not required to computerize their data,
computerization could make the chemical inventory data more useful to
the public. For example, according to state and local officials,
potential users can more easily aggregate and manipulate the data.
EPA has supported efforts to computerize the chemical inventory data.
For example, in 1996, EPA's Chemical Emergency Preparedness and
Prevention Office provided $822,000 in computer software to local
emergency planning committees and state emergency response
commissions to increase their capabilities to computerize their data.
A 1997 nationwide study performed for EPA showed that 39 percent of
the local emergency planning committees had computerized their data
and an additional 42 percent planned to do so. Because of the lack
of computerized data, copies of the individual completed reporting
forms have to be located and reviewed.
The officials from the local emergency planning committees and from
the state emergency response commissions that we interviewed
generally believed that access to and use of the chemical inventory
data would potentially improve if the data were made available
through the Internet. Once available, the data could be used to
present different environmental scenarios. For example, in early
1998, an environmental organization used the data to make available
on the Internet an accident scenario that showed the number of people
vulnerable to potential accidental releases at 10 facilities that
were operated by a major chemical manufacturer. The same analysis
also provided environmental data showing the percentage of minorities
at risk near each of those facilities.
Although none of the local or state officials in our sample had
placed chemical inventory data on the Internet, they generally said
that they would consider doing so to potentially increase public use
of the data. Some states not included in our sample are using or
planning to use the Internet to provide such data. Recently, Idaho
has used $75,000 in grant funds from EPA's Chemical Emergency
Preparedness and Prevention Office to computerize the state's
chemical inventory data and put that information on the Internet.
Oregon plans to make its chemical inventory data available on the
Internet within the next year.
A June 1998 EPA-proposed rule on EPCRA discusses, among other things,
the potential for streamlining facilities' reporting of the data by
reducing the requirement for reporting to state emergency response
commissions, local emergency planning committees, and fire
departments to one central database that would be accessible to all
three entities. EPA also suggests that a statewide database on the
Internet would provide greatly expanded public access.
EPA'S EFFORTS TO INTEGRATE
DATABASES COULD IMPROVE THE
USEFULNESS OF CHEMICAL
INVENTORY DATA
---------------------------------------------------------- Letter :4.3
Once chemical inventory data have been computerized, their usefulness
can be enhanced by integrating the data from various local and state
databases to obtain a fuller understanding of the chemicals being
stored and used throughout the country. We noted that officials in
EPA regions I and IV recently initiated projects designed to
integrate chemical inventory data from state databases and to make
the data available on the Internet.
In 1995, Region IV (Atlanta) initiated work to integrate data from
eight states in the region and to make the integrated database
available to users of the data in each of those states. According to
a regional official, it is expected that the automated database will
be available by the end of fiscal year 1999. While the Region IV
project is designed to provide a regional database, another project
started by EPA Region I (Boston) envisioned a national database of
chemical inventory data. In January 1995, Region I initiated a
feasibility study for the database and, in October 1996, awarded a
contract for developing software needed to integrate the databases of
the six New England states located within the region. Region I
intended that the integrated database for the six states would be a
prototype leading to a computer system for EPA, the states,
environmental groups, and the public to use in accessing national
chemical inventory data. With an integrated system, data formats
would be the same throughout the country and this would enable users
to make comparisons among individual facilities throughout the
nation; would provide for local, state, and regional comparisons of
chemical inventories; and would enable users to discern national
trends for the quantities of individual chemicals and groups of
chemicals.
Although EPA's Chemical Emergency Preparedness and Prevention Office
endorsed the Region I project, it did not provide any funding, and in
June 1998, the region terminated the software development contract
because of higher regional priorities. Subsequently, in July 1998,
Region I's manager for that project told us that the region had
resumed the project by using regional staff resources. According to
the project manager, five of the six states in the region had
information in the database and data from the sixth state will be
added. At this time, it is uncertain whether EPA will use the
regional database as a prototype for a national database.
An official from the Chemical Emergency Preparedness and Prevention
Office told us that it is understandable that Region I had earlier
terminated the project to focus on other priorities because no
legislative mandate exists for EPA to develop either a regional or
national database for this information. Nonetheless, the official
said it would be regrettable if EPA did not have the opportunity to
examine the effectiveness of such databases in making chemical
inventory data more accessible and useful to current and potential
users. According to the official, his office is concerned that,
although EPCRA does not provide for EPA to receive the chemical
inventory data and does not make the agency responsible for ensuring
that they are accurate, states and industry may hold EPA accountable
for the data's quality if the agency aggregates the data from state
databases and makes the information available to the public in either
regional or national databases. The official said that the data
would have to be aggregated from states using varying formats and
data elements for the information.
AGENCYWIDE POLICIES,
PROCEDURES, AND STANDARDS FOR
PUBLIC AVAILABILITY OF PROJECT
RESULTS HAVE NOT BEEN
ESTABLISHED
------------------------------------------------------------ Letter :5
EPA's recent efforts to publicly communicate environmental
information have brought objections from some state and industry
stakeholders who provide and/or use the data. Such stakeholders have
stated that, while they generally favor EPA's publicly disseminating
data under the agency's "right-to-know" authorities, they are
concerned about the manner in which EPA's data dissemination projects
have been managed. For example, in questioning the accuracy of EPA's
data, state representatives have stated that inaccurate or misleading
information provided to the public would result in the unproductive
use of federal and state resources in clarifying the data.
Industrial stakeholders stated that EPA has not adequately involved
them in its information dissemination initiatives and does not have
appropriate agencywide policies, procedures, and standards governing
decisions about disseminating data, including mechanisms to identify
and correct errors, such as outdated information on facilities in
violation of the environmental terms of their operating permits.\9
State and industrial stakeholders have discouraged EPA from publicly
reporting environmental data without assessing their accuracy and
effectively communicating to potential users the limitations that
should be placed on using the data (e.g., communicating the
limitations that apply to using the data as an indicator of risk to
human health).
According to industrial stakeholders, EPA has not collaborated with
them to identify and resolve concerns prior to disseminating
environmental information. For example, industry stakeholders told
us that the Sector Facility Indexing Project had been in existence
for more than a year before they became aware of it and requested
that EPA hold a public meeting to invite their input on the project.
Comments from the public meeting, held in May 1997, raised concerns
about inaccuracies in the data and a lack of clarifications
accompanying the data to help users understand their potential uses
and limitations. The stakeholders also maintained that, when they
brought errors to EPA's attention, they found that the agency had not
established procedures to identify such errors or to correct them
after they are found and reported by others.
In response to stakeholders' concerns, EPA held meetings with state
and industry representatives to discuss the accuracy of specific data
and procedural problems and then incorporated changes based on these
meetings, prior to releasing the data from the Sector Facility
Indexing Project on the Internet in May 1998. The changes included
correcting errors in the data, explaining limitations on how the data
can be used, and establishing a feedback mechanism to report errors
in the data. However, industry representatives told us that EPA
still lacks agencywide policies, procedures, and standards necessary
to govern future data dissemination activities, including a clear set
of ground rules for stakeholders' participation in data
administration. They believe that individual EPA offices currently
are given too much authority in determining the value of their
projects to potential users and in deciding procedural issues, such
as the requirements for the data's accuracy and the extent of
stakeholders' involvement in the projects' design and development.
To discuss the concerns about the data's accuracy and procedures that
had been raised by state, industry, and other stakeholders, we met
with EPA officials responsible for each of the three projects.
Although EPA had not established agencywide policies, procedures, and
standards for guiding the design of the projects and the release of
the information to the public, the officials maintained that, in
carrying out the projects, they have made efforts to consider the
users' needs, to obtain outside review, and to respond to concerns
over accuracy and other issues as they arose. For example, an
official of the Sector Facility Indexing Project said that EPA worked
for 3 years to identify the facilities to be included in the project
and to collect and verify the data. According to the official, each
of the facilities received a copy of its compliance and enforcement
data for review to help ensure that any problems were identified
before the information was distributed, and before the facilities'
review, EPA had asked the states to review the data and make
appropriate changes. According to the official, facilities commented
on the accuracy of 4 percent of the 38,000 major data elements they
received for review and about half of their comments were accepted
for changes. The official told us that this high rate of accuracy
indicates that industry was objecting to the project because of how
the information may be interpreted rather than its inaccuracy.
EPA has taken several agencywide actions to address concerns that
stakeholders have raised about EPA's information dissemination
processes. For example, in April 1998, EPA's Deputy Administrator
announced that the agency's Chief Information Officer will lead an
effort to develop a strategic plan to implement an agencywide
approach to improve the quality of EPA's data. He said that the plan
should address the specific roles and responsibilities of program
offices and stakeholders and that one of the principal components of
the plan should be a strategy to help ensure that "our error
correction process is well-defined, efficient, and transparent to our
partners, the public, and the regulated community." Initially, the
plan was targeted for completion by September 30, 1998, but the Chief
Information Officer told us that EPA now plans to have a draft
completed by that date and to submit it for review by EPA's Common
Sense Initiative Council during its October 1998 meeting.\10 The
Chief Information Officer and other EPA officials responsible for the
plan's development told us that the plan is a high priority within
the agency and that it is being closely coordinated with the Office
of Reinvention and the agency's newly established Center for
Environmental Information and Statistics, which also plays a key role
in addressing stakeholders' concerns.
In a February 1997 announcement of plans to establish the agency's
Center for Environmental Information and Statistics, the EPA
Administrator noted that the Center was being created to improve the
agency's collection and management of information and to provide for
better public access to "quality-assured" environmental statistics
and information. The Center's principal responsibilities include
enhancing access to EPA's databases, integrating information across
agency programs, boosting stakeholders' participation in EPA's
information policy, and helping communities better understand
environmental information.
An important function of the Center is to review the degree to which
EPA's existing databases can meet the varying demands of a wide range
of information users, including community groups, nongovernmental
organizations, and state and federal agencies. The Center currently
is leading EPA's efforts to assess the overall quality and
applicability of 31 of EPA's major national databases. These reviews
include assessments of the data's accuracy and limitations. The
assessments for accuracy will include quality checks performed by
EPA's program offices as well as statistical reviews performed by the
Center. Limitation assessments being performed by the Center will
focus on identifying databases' constraints with respect to their
primary purposes as well as their suitability for alternate uses.
EPA's Chief Information Officer told us that his office has provided
program offices with general guidance on issues relating to
information resources management. For example, chapter 21 of EPA's
Information Resources Management Policy Manual establishes the
agency's policy on the public's access to EPA's information. This
policy statement establishes the general principles to govern the
public's access to and dissemination of information gathered and
maintained by EPA and defines the information resources management
responsibilities of the agency's various offices. One of the
principles set forth is that new and enhanced data systems, data
collections, and databases are to be designed with consideration of
the need to permit and promote the public's access. However,
policies and procedures for program offices to follow in designing,
developing, and implementing information dissemination projects have
not been issued.
In the absence of such policies and procedures, as noted by a January
1998 EPA advisory council study,\11 information management has been
administered largely through EPA's program offices by using a
decentralized organization and management structure and has typically
been shaped by the program offices' policies and procedures to meet
the needs of their various internal and external users. The study
concluded that inconsistencies have arisen among the programs
regarding procedures to determine the data's accuracy, communicate
the limitations of their use, and involve stakeholders in information
management decisions.
The Deputy Director of the Center for Environmental Information and
Statistics and the Chief Information Officer told us that the
Center's role and EPA's strategy for improving data's quality are
evolving and, in the future, could involve developing guidance for
program offices to follow in their information dissemination
activities. The Deputy Director told us that such guidance could
include policies, procedures, and standards for (1) setting
priorities and performing cost-benefit analyses to determine which
information projects should receive agency resources, (2) developing
standards for the accuracy of data and mechanisms to determine and
correct errors, (3) obtaining stakeholders' involvement and analyzing
users' needs and (4) establishing other protocols that program
offices should follow in designing information dissemination
projects. He told us, however, that the Center has no specific plans
to develop such policies, procedures, and standards.
--------------------
\9 During our review, we interviewed representatives of four industry
sectors included in the Sector Facility Indexing Project and
representatives of the Coalition for Effective Environmental
Information. The coalition represents a variety of industry
segments, including petroleum refining, plastics, electronics, forest
products, chemicals, and consumer products.
\10 The Common Sense Initiative Council was established by EPA in
October 1994 as the agency's national advisory committee for
formulating recommendations and advice on the nation's pollution
control and prevention programs that relate to industrial sectors and
was directed by EPA to operate by consensus decision-making.
\11 Managing Information as a Strategic Resource: Final Report and
Recommendations of the Information Impacts Committee, The National
Advisory Council for Environmental Policy and Technology (NACEPT),
EPA (EPA 100-R-98-002, Jan. 1998).
CONCLUSIONS
------------------------------------------------------------ Letter :6
EPA is making progress in its efforts to provide communities with
more information on releases of toxic chemicals in their areas. The
data on hazardous chemicals that facilities must provide under
section 312 of EPCRA could be another substantial source of
information for communities, if access to the data could be improved
through greater computerization. Furthermore, the value of the data
could also be increased if they were contained in an integrated
regional or national database that allows for comparing nearby
facilities with others within an industry or in other geographic
locations. Although EPA's Chemical Emergency Preparedness and
Prevention Office believes that such databases could have substantial
benefits, the agency has not assessed the potential costs and
benefits of developing them.
While state and industrial stakeholders have expressed support for
EPA's efforts to make more environmental information publicly
available, some stakeholders are concerned about how it is being
done. Principally, they are concerned about the data's accuracy, the
appropriateness of some of the information for the public's use, and
how they have been involved in the design and implementation of the
projects. Industry representatives have also expressed concern that
the data required by EPCRA section 312 are costly to report but used
little by the public. EPA has issued a policy statement on public
access to the agency's information that provides general principles
for its offices to follow and has recently initiated steps to develop
a strategic plan to improve its information management. However, it
currently has no plans to develop implementing policies, procedures,
and standards to help ensure that its offices' information
dissemination activities are carried out in accordance with the
policy statement.
RECOMMENDATIONS
------------------------------------------------------------ Letter :7
To help ensure that EPA provides the public with data that are
accurate, complete, and relevant to its needs, we recommend that the
EPA Administrator supplement the agency's existing policies on
information resources management by developing agencywide policies
and procedures that specify guidance and standards for program
offices involved in designing, developing, and implementing
information dissemination projects. Such guidance and standards
should address obtaining stakeholders' involvement in the projects'
design and development, testing for and correcting errors in the
data, and communicating contextual information on the data's uses and
limitations.
Given the potential usefulness of EPCRA section 312 data to the
public, we recommend that the EPA Administrator evaluate options to
make the data more accessible and implement the most cost-effective
option that provides availability on a regional and national basis.
In implementing the project, EPA should use the policies and
standards for dissemination projects that we have recommended.
AGENCY COMMENTS
------------------------------------------------------------ Letter :8
We provided copies of a draft of this report to EPA for its review
and comment. Representatives of EPA's offices responsible for the
activities discussed in the report, including the Chief Information
Officer and the Acting Assistant Administrator for Solid Waste and
Emergency Response, said that the report, in general, accurately
describes the agency's efforts to make information on risks from
releases of toxic chemicals available to local communities, to
provide the public with chemical inventory information, and to
develop policies for publicly disseminating environmental
information. The officials concurred with our recommendation that
the EPA Administrator supplement EPA's existing information resources
management policy by developing agencywide policies and procedures
that specify guidance and standards for program offices involved in
designing, developing, and implementing information dissemination
projects. However, the Acting Assistant Administrator for Solid
Waste and Emergency Response disagreed with our recommendation that
the EPA Administrator evaluate options for making the data collected
under EPCRA section 312 more accessible and implement the most
cost-effective option to make that information available on a
regional and national basis.
The Acting Assistant Administrator said that there is no legislative
mandate for requiring industry to submit EPCRA 312 data to EPA or for
using that data to develop a national database. He also said that it
would be difficult to aggregate the data from all the states, many of
which have different reporting formats and many of which do not
currently computerize the data. In addition, the Acting Assistant
Administrator said that the Chemical Emergency Preparedness and
Prevention Office does not currently have the resources to develop
and maintain such a database. He said that the office's current
focus is on implementing section 112(r) of the Clean Air Act, which
includes developing a national database of risk management plans that
contain a wealth of information on the chemical risks at various
facilities. Furthermore, the Acting Assistant Administrator said
that any effort to develop a regional and national EPCRA section 312
database should include the involvement of the public and that it is
possible, if not probable, that the public would like to know some
information not included in the section 312 database and would not be
interested in certain data that are included.
Nonetheless, we have retained the recommendation in the report and
note that, by calling for EPA to "evaluate options" and "implement
the most cost-effective option," the recommendation gives EPA
considerable flexibility to overcome the obstacles that the agency
described. We recognize that EPA does not have a specific
legislative mandate to create a regional or national database of
EPCRA section 312 data. However, one of EPA's 10 strategic goals is
"expansion of Americans' right to know about their environment," and
the agency has been and is currently involved in various activities
across its programs to make more environmental information available
to the public. Not all of these activities are specifically directed
by legislative mandates. We further recognize that EPCRA does not
require industry to submit section 312 reports to EPA. However, as
agency officials noted, EPA can request the data from the states.
We are aware that the Chemical Emergency Preparedness and Prevention
Office has limited resources and is currently working to implement
section 112(r) of the Clean Air Act. Given the potential usefulness
to the public, we believe that making EPCRA section 312 data more
accessible should be considered in the context of agencywide
right-to-know priorities and resources rather than those of the
Chemical Emergency Preparedness and Prevention Office. Although
certain options, such as EPA's creating and maintaining a unique
national database of EPCRA section 312 information, could be costly,
the efforts of regions I and IV in integrating the various databases
of the states in the regions indicate that low-cost options could be
available. Although not all states have computerized databases, EPA,
in recent years, has provided grant funds to assist the states in
computerizing their data, and about two-thirds of the states have
done so.
We agree that stakeholders should be involved in any effort to make
EPCRA section 312 data more accessible. In addition to the public,
these stakeholders would include emergency planning and response
personnel, who need to use the data, and industry, which must report
the data. Making more or less of this information available to meet
the public's needs would be options to be considered in carrying out
our recommendation. In calling for EPA to adopt the most
cost-effective option, we recognize that the public's desire for
additional information would need to be weighed against the costs of
reporting, compiling, and maintaining it.
EPA also provided some technical comments on our draft report. We
have revised our report, as appropriate, in response to these
comments.
---------------------------------------------------------- Letter :8.1
We performed our review from November 1997 through August 1998 in
accordance with generally accepted government auditing standards.
(See app. I for a detailed description of our scope and
methodology.) We are providing copies of this report to other
appropriate congressional committees; the Director, Office of
Management and Budget; and the Administrator, EPA. We will also make
copies available to others on request.
If you or your staff have any questions, please call me at (202)
512-6111. Major contributors to this report are listed in appendix
II.
Peter F. Guerrero
Director, Environmental
Protection Issues
SCOPE AND METHODOLOGY
=========================================================== Appendix I
To identify the Environmental Protection Agency's (EPA) initiatives
to provide additional information on the risks posed by toxic
chemical releases to local communities, we held discussions with
EPA's Chief Information Officer and officials of EPA's Office of
Pollution Prevention and Toxics, which has responsibility for the
Toxic Release Inventory (TRI) program; Office of Policy, Planning,
and Evaluation; Center for Environmental Information and Statistics;
and Reinvention Office. To determine the status of these
initiatives, we interviewed officials responsible for implementing
the projects and reviewed documents obtained from them. We also
reviewed reports and written comments on the projects by EPA's
Science Advisory Board, state officials, and industry
representatives.
We discussed the projects and EPA's overall efforts to provide
environmental information to the public with the Coalition for
Effective Environmental Information, which was formed by various
industry firms and groups to monitor and provide input on EPA's
public information efforts and representatives of various firms and
associations, such as those in the mining, petroleum, and chemical
industries. In addition, we attended a September 1997 conference of
TRI users and held discussions with representatives of the Unison
Institute and OMB Watch, who support and are involved in making
environmental information available to the public.\1 To determine
EPA's policies and standards for providing information to the public,
we interviewed the Chief Information Officer and the Deputy Director
of the Center for Environmental Information and Statistics.
We discussed how data collected under EPCRA section 312 are made
available to the public, the public's requests for access to the
data, and ongoing and planned efforts to improve access with selected
State Emergency Response Commissions and Local Emergency Planning
Committees. Because state-by-state data on the number of facilities
reporting under section 312 are not available, we selected the State
Emergency Response Commissions of the 10 states that had the largest
amounts of TRI chemical releases. For each of these states, we
selected two Local Emergency Planning Committees that representatives
of their respective commissions indicated were the most active in
improving public access to the data. We also contacted state
emergency response commissions in two other states that were
undertaking initiatives to improve public access.
We discussed EPA's efforts to improve access to section 312 data with
officials of the agency's Office of Chemical Emergency Preparedness
and Prevention, as well as EPA Regions I and IV, which at the time of
our review had their own projects under way. We also discussed the
public's access to and the potential use of the data with various
people knowledgeable about that data at EPA as well as in academia,
industry, and public interest groups.
--------------------
\1 Building for the Future: Toxic Release Inventory and
Right-to-Know Conference, organized by the Unison Institute in
cooperation with EPA, Sept. 8-10, 1997, Washington, D.C.
MAJOR CONTRIBUTORS TO THIS REPORT
========================================================== Appendix II
RESOURCES, COMMUNITY, AND ECONOMIC
DEVELOPMENT DIVISION WASHINGTON,
D.C.
David G. Wood, Associate Director
Edward A. Kratzer, Assistant Director
Raymond H. Smith, Jr., Evaluator-in-Charge
J. Kenneth McDowell, Senior Evaluator
Donald E. Pless, Senior Evaluator
BOSTON FIELD OFFICE
Bruce Skud, Senior Evaluator
*** End of document. ***