Accidental Release Prevention Requirements; Risk Management
Programs Under the Clean Air Act Section 112(r)(7); Distribution of
Off-Site Consequence Analysis Information
Related Material
[Federal Register: August 4, 2000 (Volume 65, Number 151)]
[Rules and Regulations]
[Page 48107-48133]
From the Federal Register Online via GPO Access [wais.access.gpo.gov]
[DOCID:fr04au00-18]
[[Page 48107]]
-----------------------------------------------------------------------
Part III
Department of Justice
Environmental Protection Agency
-----------------------------------------------------------------------
40 CFR Chapter IV
Accidental Release Prevention Requirements; Risk Management Programs
Under the Clean Air Act Section 112(r)(7); Distribution of Off-Site
Consequence Analysis Information; Final Rule
[[Page 48108]]
-----------------------------------------------------------------------
DEPARTMENT OF JUSTICE
40 CFR Chapter IV
[AG Order No. 2318-2000]
RIN 1105-AA70
ENVIRONMENTAL PROTECTION AGENCY
40 CFR Chapter IV
RIN 2050-AE80
Accidental Release Prevention Requirements; Risk Management
Programs Under the Clean Air Act Section 112(r)(7); Distribution of
Off-Site Consequence Analysis Information
AGENCIES: Department of Justice and Environmental Protection Agency.
ACTION: Final rule.
-----------------------------------------------------------------------
SUMMARY: The Environmental Protection Agency (EPA) and the Department
of Justice (DOJ) are promulgating a rule that provides for access to
information concerning the potential off-site consequences of
hypothetical accidental chemical releases from industrial facilities.
Under section 112(r) of the Clean Air Act (CAA), facilities handling
large quantities of extremely hazardous chemicals are required to
include that information in a risk management plan (RMP) submitted to
EPA. As required by the Chemical Safety Information, Site Security and
Fuels Regulatory Relief Act (CSISSFRRA), this rule provides members of
the public and government officials with access to that information in
ways designed to minimize the likelihood of accidental releases, the
risk to national security associated with posting the information on
the Internet, and the likelihood of harm to public health and welfare.
DATES: This rule is effective on August 4, 2000.
ADDRESSES: Supporting information used to develop the proposed rule and
the final rule is contained in Docket No. A-2000-20. The docket is
available for public inspection and copying between 8 a.m. and 5:30
p.m., Monday through Friday (except government holidays), at Waterside
Mall, Room M1500, 401 M Street, S.W., Washington, DC 20460. A
reasonable fee may be charged for copying. The assessments upon which
this rule is based are also available on the Internet at http://
www.usdoj.gov and http://www.epa.gov/ceppo.
FOR FURTHER INFORMATION CONTACT: Brenda Sue Thornton, Trial Attorney,
Criminal Division, Terrorism and Violent Crime Section, Department of
Justice, 601 D Street, N.W., Room 6500, Washington, DC 20530, (202)
616-5210; John Ferris, Chemical Engineer, (202) 260-4043, or Vanessa
Rodriguez, Chemical Engineer, (202) 260-7913, Chemical Emergency
Preparedness and Prevention Office, Environmental Protection Agency
(5104), 1200 Pennsylvania Avenue, N.W., Washington, DC 20460; or the
Emergency Planning and Community Right-to-Know Hotline at (800) 424-
9346 (in the Washington, DC, metropolitan area, (703) 412-9810). You
may wish to visit the Chemical Emergency Preparedness and Prevention
Office (CEPPO) Internet site at http://www.epa.gov/ceppo.
SUPPLEMENTARY INFORMATION: This rule was published in the Federal
Register as a proposed rule on April 27, 2000 (65 FR 24834). This
Federal Register action announces EPA and DOJ's final decisions on the
rule.
Table of Contents
I. Introduction
A. Statutory Authority and Background
B. The Proposed Rule
II. Discussion of Comments on the Proposed Rule
A. Risk and Benefits Assessments
1. Benefits Assessment
2. Risk Assessment
3. The Assessments and the Proposed Rule
B. General Comments on the Rule
C. Rule's Impact on Risk Reduction
D. Reading Rooms
1. General Comments on Reading Rooms
2. The Number of Paper Copies
3. Operation of Reading Rooms
4. The Number of Reading Rooms
5. The Location of Reading Rooms
6. Security Measures at Reading Rooms
7. Alternatives to Reading Rooms
E. Enhanced Local Access to OCA Information
F. Risk Indicator System (Vulnerable Zone Indicator System)
G. Internet-Accessible OCA Information
H. Access to OCA Information by Government Officials
I. Other Provisions
III. Discussion of Final Rule
A. Access to Paper Copies of OCA Information
B. Enhanced Access to Local OCA Information
C. Vulnerable Zone Indicator System
D. Internet Access to Selected OCA Information
E. Additional Information on Chemical Accident Risk
F. Access to OCA Information by Government Officials and Other
Provisions
G. Effective Date and Implementation Schedule
IV. Administrative Requirements
A. Docket
B. Executive Order 12866
C. Executive Order 12988
D. Executive Order 13045
E. Executive Order 13084
F. Executive Order 13132
G. Regulatory Flexibility Act
H. Paperwork Reduction Act
I. Unfunded Mandates Reform Act of 1995
J. Small Business Regulatory Enforcement Fairness Act of 1996
K. Congressional Review Act
V. Judicial Review
I. Introduction
A. Statutory Authority and Background
As more fully described in the notice of proposed rulemaking (NPRM)
(65 FR 24853 (April 27, 2000)), the federal government's efforts to
prevent and mitigate chemical accidents are reflected in several pieces
of legislation, including section 112(r) of the CAA, 42 U.S.C. 7412(r).
In that section, Congress imposed a general duty on industrial
facilities handling any extremely hazardous chemicals to do so safely
(CAA section 112(r)(1)), and required EPA to establish a regulatory
program for facilities that pose the greatest risk (CAA section
112(r)(7)). Congress directed that the regulatory program require
covered facilities to develop and implement a risk management program
for preventing accidental chemical releases and minimizing the
consequences of releases that do occur. Congress further mandated that
facilities perform an off-site consequences analysis (OCA) for one or
more hypothetical accidental worst case and/or alternative release
scenarios and report the results of the analysis in a risk management
plan (RMP) to be submitted to federal, state, and local government
agencies and made available to the public.
EPA issued the rules establishing the regulatory program required
by CAA section 112(r) on January 31, 1994 (59 FR 4478) and June 20,
1996 (61 FR 31668, the ``RMP rule''). In those rules, EPA continued the
philosophy that EPA embraced in implementing the Emergency Planning and
Community Right-to-Know Act of 1986 (EPCRA). Specifically, EPA
recognized that regulatory requirements by themselves will not
guarantee safety, and that providing the public with information about
hazards in a community can and should lead government officials and the
public to work with industry to prevent accidents. EPA thus relied on
the public availability of RMPs to stimulate further chemical risk
reductions efforts, which occur primarily at the local level where the
risk is found.
Over 15,000 facilities are subject to the RMP rule. In an effort to
reduce the burden of collecting and disseminating RMPs, EPA designed an
electronic RMP
[[Page 48109]]
form that could be placed on the Internet for purposes of public
access. However, the Federal Bureau of Investigation and other
representatives of the law enforcement and intelligence communities
raised concerns that releasing the OCA portions of RMPs via the
Internet would enable individuals anywhere in the world anonymously to
search electronically for industrial facilities in the U.S. to target
for purposes of causing an intentional industrial chemical release. In
response to those concerns, EPA posted RMPs on the Internet
(www.epa.gov/ceppo/) without the sections of the RMP that contain OCA
results (sections 2 through 5). However, those OCA sections, and any
EPA electronic database created from those sections, were still subject
to public release in electronic format pursuant to the Freedom of
Information Act (FOIA), 5 U.S.C. 552. On August 5, 1999, CSISSFRRA was
enacted (Pub. L. No. 106-40) to provide at least a one-year exemption
from FOIA for ``OCA information,'' including the OCA portions of RMPs
and any EPA database created from those portions. CSISSFRRA amended
section 112(r)(7) of the CAA by adding a new subparagraph (h).
CSISSFRRA requires the President, by the end of the one-year period
of the FOIA exemption, to decide how to disseminate OCA information.
Specifically, CSISSFRRA requires the President to assess ``the
increased risk of terrorist and other criminal activity associated with
the posting of [OCA] information on the Internet'' and ``the incentives
created by public disclosure of [OCA] information for reduction in the
risk of accidental releases'' (CAA section 112(r)(7)(H)(ii)(I)). Based
on those assessments, the President is required by August 5, 2000, to
promulgate a regulation governing access to OCA information in a manner
that minimizes the likelihood of chemical releases, however caused.
Until that time, CSISSFRRA limits public access to OCA information but
provides government officials access for purposes of preventing,
planning for, or responding to chemical releases. The President
delegated to the Attorney General and the Administrator of EPA the
authority to conduct the required assessments and rulemaking (see the
delegation memorandum at 65 FR 8631 (February 22, 2000)). The proposed
and final rules are subject to approval by the Director of the Office
of Management and Budget (OMB).
The risk and benefits assessments were completed and used as the
basis for the proposed rule. The conclusions of those assessments are
fully described in the NPRM. Briefly, the risk assessment found that an
increased risk of terrorist or other criminal activity would accompany
the release of certain items of OCA information via the Internet. That
information could be used by terrorists or other criminals for purposes
of targeting or maximizing the results of industrial chemical releases.
The benefits assessment concluded that public disclosure of OCA
information would likely lead to a significant reduction in the number
and severity of accidental chemical releases. It also found that ease
of public access to information is important to the public's use of
that information. The risk and benefits assessments are available in
the docket for this rulemaking and on the EPA and DOJ websites
(www.epa.gov/ceppo/ and www.usdoj.gov).
B. The Proposed Rule
Based on the risk and benefits assessments, EPA and DOJ proposed
providing the public with several means of obtaining access to OCA
information and information about the risk expressed by OCA
information. The complete proposal is contained in the NPRM. A brief
summary follows.
In order to minimize the risk of Internet dissemination of OCA
information while still providing public access to that information, we
proposed to provide the public with access to paper copies of OCA
information for covered facilities at 50 or more federal reading rooms
geographically distributed across the United States. At the reading
rooms, members of the public would have access to OCA information for a
limited number of facilities, located anywhere in the country, and
would be able to read the information and take notes from it, but not
remove or mechanically reproduce it. Reading rooms would be authorized
to provide any member of the public with access to OCA information for
up to 10 stationary sources per calendar month. Based upon an analysis
of the geographic distribution of RMP-covered facilities, we concluded
that the 10 per individual per calendar month limit would still permit
most members of the public to have access to OCA information for
facilities in whose ``vulnerable zone'' they live or work, as well as
to OCA information for a few other facilities located elsewhere.
In addition, we proposed making the less sensitive items of OCA
information available to the public on the Internet by posting them on
EPA's website. Those items of OCA information included information
about passive and active safety systems used by facilities; we
explained that that information would facilitate risk reduction
dialogues among members of the public, state and local officials, and
facilities. Only the items of OCA information for which the risk
assessment found there was a significant risk of use for terrorist or
other criminal purposes would be excluded from Internet posting.
We also proposed creating a ``risk indicator'' system as a tool for
providing the public with a means of understanding, via Internet
inquiry, some aspects of the risk expressed by OCA information. Members
of the public would be able to enter a specific address (such as that
of a home, school, or place of employment) into the risk indicator
system and learn if that address might be within the ``vulnerable
zone'' of at least one facility that submitted an RMP to EPA. Members
of the public who do not have access to the Internet would be able to
obtain the same information by calling the EPA hotline or by mailing a
request to the Administrator of EPA. The risk indicator system also
would inform individuals of several means by which they could obtain
the names of the facilities and additional information.
Further, we proposed authorizing and encouraging members of local
emergency planning committees (LEPCs), state emergency response
committees (SERCs), or local fire departments to allow members of the
public to read, but not remove or mechanically copy, paper copies of
OCA information for all of the covered facilities in the LEPC's
jurisdiction and for any facilities whose vulnerable zone extended into
the LEPC's jurisdiction. To further supplement public access, under the
proposed rule, EPA would make available to the public additional
information on chemical accident risk through an Internet website. The
proposal also addressed how EPA would provide access to OCA information
to federal, state, and local government officials for their ``official
use'' by codifying the provisions of CSISSFRRA that appear in CAA
section 112(r)(7)(H)(ii)(II)(cc)-(ee). Finally, the proposal called for
establishing further provisions as needed to implement CSISSFRRA, such
as prohibiting the unauthorized release of OCA information and
authorizing the Administrator to provide OCA information to qualified
researchers under CAA section 112(r)(7)(H)(vii).
II. Discussion of Comments on the Proposed Rule
The proposed rule was published for comment in the Federal Register
on April 27, 2000. The comment period ended on June 8, with 68 comments
[[Page 48110]]
submitted. Commentors represented industry, trade associations, public
interest groups, journalists, environmental groups, law enforcement,
emergency response groups, state/local entities, and the general
public. In addition, on May 9, 2000, EPA and DOJ held a public hearing
on the proposed rule at which nine presenters representing public
interest groups, environmental research groups, state and local
emergency planning groups, and the general public provided comments
about the proposed regulation. We are responding to most comments on
the proposed rule in this preamble. We respond to additional comments
in a supplemental document included in the public docket for this
rulemaking.
A. Risk and Benefits Assessments
As noted above, the assessments were available on the EPA and DOJ
websites. We received comments on both the benefits assessment and the
risk assessment expressing a wide range of opinion. We note at the
outset that CSISSFRRA did not call for the assessments to be developed
through a public rulemaking process (see CAA section 112(r)(7)(H)(ii)).
Instead, CSISSFRRA required the President to conduct the assessments
and then, ``based on the assessments,'' to promulgate regulations
governing the distribution of OCA information. In requiring
regulations, CSISSFRRA ensured the public an opportunity to participate
in the government's consideration of the extent to which and the manner
in which OCA information should be made available based on the
assessments. Preparation of the risk assessment, however, necessarily
called for the exercise of expert judgment in sensitive areas of law
enforcement and national security, areas in which the President is
typically accorded broad discretion. We thus believe that Congress did
not intend the assessments to be subject to public evaluation except to
the extent they do, or do not, adequately support the rule being
promulgated. We nonetheless appreciate the careful consideration that
the assessments received from the public and respond below to the
significant concerns that were raised.
In regard to the approach taken by both assessments, several
commentors asserted that the assessments were fundamentally flawed
because they failed to quantify the risk and benefit of disseminating
OCA information. EPA and DOJ disagree with that comment. Given the
short time frame the agencies had to develop the assessments and lack
of a clear basis for estimating the probability of a chemical accident
or criminal incident involving an industrial facility, it would have
been difficult, if not impossible, to obtain or develop sufficient data
to support such an analysis. To begin with, since OCA information is
not yet publicly available, its effect on the risk and benefits to be
assessed cannot be measured directly. In addition, because the RMP
program took effect only last year and trends in terrorism are
changing, there is little other data regarding the precise issues that
the assessments were required to address.
Furthermore, EPA and DOJ believe that statistical evaluation of the
benefits and costs relating to the release of OCA information on the
Internet was not necessary to determine how OCA information should be
disseminated, which is the purpose of this rulemaking exercise. In the
benefits assessment, an analysis of the effect of public release of
Toxic Release Inventory (TRI) data indicated that information
dissemination leads to further risk reduction efforts. In the risk
assessment, an analysis was conducted of trends related to weapons of
mass destruction and recent terrorist events. Each assessment used
those analyses as the basis for assessing the benefits and risks
related to dissemination of OCA information. The findings that resulted
from those analyses informed the rule. We believe that that methodology
was appropriate for purposes of determining how best to disseminate OCA
information.
1. Benefits Assessment
As noted above, the reaction to the benefits assessment was mixed.
Many commentors agreed with the conclusions regarding the benefits of
public disclosure of OCA information. Other commentors took issue with
some of the assessment's findings.
Several commentors contended that there was no basis for drawing an
analogy between the TRI program experience and what might be expected
for OCA information because TRI data records are based on anticipated
lawful releases, derived from estimates or actual measurements, while
OCA information is based on hypothetical, unanticipated releases. We
disagree that an analogy between TRI data and OCA information is
inappropriate. As noted in the benefits assessment, although TRI data
represent actual releases while OCA information represents hypothetical
releases, our reason for examining the TRI program experience was the
fact that TRI data are made publicly available in an easily used and
understood format. The assessment noted a correlation between the ready
accessibility of TRI data and the extensive use made of it by community
and environmental groups, the news media, state and local governments,
and industry, and concluded that a similar correlation might reasonably
be expected from the dissemination of OCA information.
Some commentors disagreed with the benefits assessment's
contentions that the publication of TRI data contributed to reductions
in TRI emissions. They attributed TRI emissions reductions mainly to
economic incentives, technical considerations, and CAA regulatory
programs. The commentors were also critical of the methodology and the
conceptual and statistical support for the assessment's analysis of the
effect of negative media attention on TRI emissions reductions. They
criticized, for example, the assessment's focus on the ``worst
polluting'' facilities selected by EPA. They took issue with
comparisons between large facilities with correspondingly large
releases to small facilities with small releases, and comparisons
between supposedly similar facilities that may have differed in terms
of process or industry classification. They questioned, moreover,
whether the analysis captured the full range of TRI data available
beginning in 1987. Based on those criticisms, the commentors viewed the
TRI analogy as an invalid basis for the benefits assessment's
conclusion that wide public access to OCA information would help reduce
the risk of chemical accidents.
As the assessment noted several times, a number of different
factors contributed to TRI emissions reductions. Nonetheless, according
to the literature reviewed for the assessment, the interest in TRI
data--either in the form of published reports, negative press accounts,
or the publication of TRI data by a company--was one of the factors
affecting TRI emissions reductions. As explained in Appendix D of the
assessment, the media relied on total emissions data to label certain
facilities as the ``worst polluters.'' EPA compared the total TRI
emissions reduction rates of those ``worst polluting'' facilities with
the overall TRI emissions reduction rates for all other facilities
(both large and small) since TRI data were first published in 1989
(which includes data collected in 1987 and 1988). The ``worst
polluting'' facilities featured in news accounts appeared to have
reduced their emissions significantly more than did the other
facilities. EPA also compared ``worst polluting'' facilities to others
listed under the same TRI industry
[[Page 48111]]
classification because facilities in the same industry classification
were likely to have similar processes. EPA recognizes that within a
single industry classification there could be differences in chemical
processes that might account for some of the differences in TRI
emissions. Those variations, however, would not affect the results of
the assessment's comparative analysis of TRI emissions reduction rates
for facilities that were subject to significant negative publicity and
those that were not. As indicated in Appendix D, moreover, even the
reduction rates of facilities with relatively low levels of emissions
that were the subject of negative press accounts were significantly
greater than those of other facilities not subject to negative
publicity. In light of that evidence, we continue to believe that if
OCA information, like TRI data, were made publicly available in an
easily understood format, there would be increased public understanding
and dialogue about accidental release risk and risk reduction. We
further believe that the resulting public pressure could lead to the
adoption of additional risk reduction measures.
Other commentors contended that the benefits assessment should not
use the term ``risk reduction'' when referring to the TRI program since
TRI data does not communicate ``risk,'' which is often understood to be
the consequence of an event multiplied by the probability that the
event will occur. They also questioned whether the OCA information has
value for risk reduction. As the benefits assessment explained in
detail in Chapter 6, however, OCA information by itself does not
communicate risk; rather, OCA information in context and in comparison
with other information can provide insights about risk. As stated in
Chapter 6, ``[F]rom this comparison and understanding of potential
risk, unacceptable risks can be reduced. * * *''
Several commentors also claimed that the assessment's figures for
the costs of chemical accidents were outdated and likely overstated
because they did not take into account the significant risk reduction
benefits of the RMP rule. They suggested, for example, that many
companies reduced their inventories of hazardous chemicals in order to
avoid being subject to the RMP rule. We believe that the costs of
chemical accidents reported in the benefits assessment are based on the
most current accurate data available. Some of the data come from the
RMP five-year accident histories--data provided by the RMP facilities
themselves.
We recognize that before RMPs were required, many responsible
chemical facility owners and operators were aware of the need for
chemical accident prevention as the result of efforts by a variety of
organizations, including the Center for Chemical Process Safety, the
American Chemistry Council (formerly the Chemical Manufacturers
Association) via the Responsible Care\TM\ program, the Occupational
Safety and Health Administration, and others. The objective of the
benefits assessment was not to quantify the cumulative impact of
voluntary process safety initiatives or of the 1996 RMP rule. Instead,
as required by CSISSFRRA, the focus of the assessment was to evaluate
the nature and extent of risk reduction benefits that would likely
occur if OCA information were widely available and easily accessible to
the public. We remain convinced that the assessment correctly concluded
that readily available, easily accessible and interpreted OCA
information, in combination with RMP information, would stimulate
public dialogue about chemical risks and would result in at least some
of the 15,000 covered facilities implementing additional risk reduction
measures.
Lastly, several commentors asserted that the benefits assessment
overstated the importance of OCA information and underestimated the
value of the data already released in executive summaries or available
through local sources of risk information. The benefits assessment
acknowledged that many facilities have provided OCA data in their
executive summaries, and that individuals with sufficient effort and
know-how could generate their own offsite consequence data from
publicly available information. In fact, some organizations have
already published their own databases of ``worst-case'' scenarios based
upon data less accurate than OCA information. However, the assessment
also noted that the amount of OCA data included in executive summaries
varies widely, and that OCA data in executive summaries cannot be
easily sorted or compared. In addition, OCA results prepared by those
outside the company are often erroneous because they are based on
incomplete or inaccurate information. The OCA information in an RMP is
generated by the company submitting it, and takes into account site-
specific information; consequently, the OCA information portions of
RMPs contain the most reliable data for comparison purposes and for
understanding risks. The assessment made clear, however, that OCA
information for a single facility is of limited value, and is far more
useful when evaluated in the context of the facility's entire RMP, and
compared to OCA information reported by similar facilities or by
facilities handling similar chemicals.
2. Risk Assessment
Like the benefits assessment, the risk assessment prompted a range
of comments. Some commentors generally agreed with its conclusions.
Others, citing DOJ's expertise, deferred to the assessment's findings,
but urged DOJ to consider additional risks. In contrast, some
commentors claimed that the assessment's conclusions were overstated in
light of the availability of data comparable to OCA information or that
it failed to consider factors that would reduce the risk assessed.
Some commentors expressed concern that the risk assessment
understated the security concerns posed by the dissemination of OCA
information. Some of those commentors asserted that the assessment
should have considered the potential danger that OCA information could
be disseminated by persons taking handwritten notes that could be
posted on the Internet. In fact, the risk assessment addressed that
potential risk. While the assessment recognized that dissemination of
handwritten notes was cause for concern, it concluded that the risk
posed by that was less than that posed by release of government
documents containing OCA information. Handwritten notes would not carry
the same presumption of accuracy and reliability generally associated
with government documents. Handwritten notes also would require
significant time and effort to transcribe, making them less likely to
be used for purposes of creating a large electronic OCA database that
could be posted on the Internet.
Other commentors stated that the risk assessment did not discuss
other potential risks associated with the release of OCA information,
such as exploitation of the data for purposes of conducting industrial
espionage or locating precursor chemicals for purposes of creating
illicit drugs. We note, however, that CSISSFRRA requires the risk
assessment to weigh whether posting OCA information on the Internet
would increase the risk of criminally-caused chemical releases. While
the release of OCA information may pose other risks as well, we did
not, and, given time constraints, could not, assess those risks.
By contrast, a number of commentors asserted that the assessment
overstated or mischaracterized the risk posed by the dissemination of
OCA information.
[[Page 48112]]
Those commentors made several points. First, one stated that the
assessment did not discuss the ``increased risk'' posed by
dissemination of OCA information on the Internet (quoting CAA section
112(r)(7)(H)(ii)(I)(aa)). Rather, the commentor offered, it merely
concluded that OCA information would be helpful to a terrorist or
criminal. In fact, however, the risk assessment did address the issue
of ``increased risk,'' as required by CSISSFRRA. It concluded that OCA
information would provide someone seeking to target or maximize an
industrial chemical release with helpful information that is not
currently available, and, therefore, that posting OCA information on
the Internet would increase the risk of a terrorist using the
information for that purpose.
Other commentors argued that information identical or similar to
OCA information is already publicly available, and, therefore, the risk
assessment overstated the risk posed by posting OCA information on the
Internet. The risk assessment acknowledged that some items of OCA
information and information comparable to OCA information are currently
available to the public. However, the risk assessment also found that
the items of OCA information most likely to be used by a terrorist to
plan or execute an attack (e.g., the distance to endpoint, the
population within the distance to endpoint, and public and
environmental receptors affected) have not been assembled into a
publicly available resource that would be as comprehensive and
accessible as OCA information would be if posted on the Internet,
particularly in its database form. While several commentors noted that
RMP executive summaries are currently available on the Internet, both
the risk and benefits assessments found that the quantity and quality
of OCA data contained in the posted executive summaries vary
considerably. Some executive summaries include all of the OCA data
elements while others include little or none. Consequently, OCA data
that have been released through the executive summaries do not
constitute a comprehensive collection of OCA information. Moreover, OCA
data included in the executive summaries cannot be electronically
searched in a manner that would allow the sort of comparisons among RMP
facilities that would facilitate targeting. The risk assessment thus
reasonably concluded that full publication of OCA information on the
Internet would pose a significantly greater risk than that currently
posed by the public availability of executive summaries and other
information, even though executive summaries have been posted on the
Internet.
Similarly, some commentors questioned the risk posed by OCA
information, since data similar to OCA information could be calculated
using publicly available sources of information. The risk assessment
found that calculating information like OCA information using available
sources of data would be possible but would require significant effort
and know-how. To date, no comprehensive collection of data on the off-
site consequences of chemical releases is available on the Internet. To
the extent that EPCRA information is available on the Internet, the
risk assessment found that such information does not pose the same
degree of risk as would OCA information because EPCRA information does
not furnish the type of targeting data (such as the distance a chemical
release would travel and the population that lives within that area)
that could be used to plan terrorist events. Furthermore, the
assessment found that some publicly available information similar to
the key items of OCA information is only available through SERCs and
LEPCs, and is not Internet-accessible. The risk assessment found that
the ability to access information anonymously posed significant
security concerns and that, for information attainable only through
personal contact, for example, by contacting a SERC or LEPC, there is
less of a risk that the information would be misused by criminals, who
typically avoid such contact in executing their plans. Thus, to the
extent that information similar to OCA information is currently
available, it can be obtained only through means that do not pose a
risk comparable to that which would be created by Internet access to
OCA information.
Another commentor maintained that OCA information has already
entered the public domain because every covered facility in its state
had held the public meeting required by CSISSFRRA section 4. That
section specifies that every covered facility must provide the public
with a summary of the OCA portions of its RMPs at a meeting or in a
public notice no later than February 1, 2000. We do not believe,
however, that those meetings (and notices) provided OCA information in
a way that presents a significant risk. Facilities were required to
share only a summary of their OCA information, and facilities were free
to do so in various ways, making it unlikely that the information they
shared with the public was sufficiently detailed or uniform to make it
easy to assemble and distribute over the Internet. Also, the meetings
were a one-time requirement and thus are not an ongoing source of OCA
information.
Several commentors questioned the risk assessment's conclusions
regarding the helpfulness of OCA information to terrorists and
criminals; they asserted that it does not provide a ``roadmap'' for
terrorists and that it fails to provide all of the information that a
terrorist would need to conduct an attack. The risk assessment,
however, did not claim that OCA information provides a comprehensive
``how-to'' manual for attacks on chemical facilities. Nor did it claim
that OCA information provides all of the information that would be
sought by someone seeking to cause an intentional chemical release.
Rather, the risk assessment found that OCA information supplies some
pieces of information that would be useful to someone seeking to target
or maximize an industrial chemical release. The risk assessment noted
that information such as the population that could be affected, the
distance that a plume of chemical could radiate, and the types of
buildings and landmarks in the local area are precisely the type of
information that would be of interest to a terrorist seeking to
maximize the effect of an industrial chemical attack. Thus, even if OCA
information does not provide a ``roadmap'' for terrorists or all of the
necessary information for an attack, it still provides crucial pieces
of information that would increase the risk of terrorist or other
criminal activity.
A few commentors argued that several of the examples cited in the
risk assessment were irrelevant to whether terrorists or criminals in
the United States might seek to cause an industrial chemical release.
In particular, those commentors considered irrelevant the examples of
chemical releases that occurred in Bosnia and the incidents involving
criminals in the United States who had personal knowledge of the
industrial facilities they targeted. We disagree. Those incidents were
included in the risk assessment because they establish specific,
important points relevant to the risk assessment. The examples in
Bosnia demonstrate that it is in fact possible to cause large-scale
chemical releases using explosives or other means; and the two criminal
incidents that occurred in the United States demonstrate that criminals
in this country have indeed considered using--although they have not
successfully caused--chemical releases to inflict mass casualties.
Lastly, two commentors asserted that the risk assessment should
have taken into account the risk reduction that would be achieved by
informing the
[[Page 48113]]
community of OCA information. We agree that the dissemination of OCA
information can assist the community in preventing, preparing for, and
responding to chemical releases, regardless of how they are caused, and
thereby may mitigate the damage that such releases could cause.
However, that point does not contradict the risk assessment's finding
that the release of OCA information on the Internet would increase the
risk of an intentional chemical release or other related criminal
conduct. Moreover, while the benefits assessment concluded that public
release of OCA information would likely result in a significant
reduction in chemical risk, it did not find that the reduction in risk
would offset the increase in risk that would accompany Internet
dissemination of OCA information. As explained above, we do not have
sufficient data to estimate the number of lives that could be lost or
saved by various approaches to the dissemination of OCA information.
But we are concerned that terrorists or criminals would use anonymous
Internet access to OCA information to maximize the effects of a
release, and that those effects are likely to be large compared to the
effects of unintentional releases. Moreover, it will take time for the
public release of OCA information to create the incentives that will in
turn lead to risk reduction. The increased risk created by Internet
dissemination of OCA information, by contrast, would be immediate. For
those reasons, we do not believe that unfettered release of OCA
information would achieve the statutory objective of minimizing the
risks of chemical releases, however caused.
3. The Assessments and the Proposed Rule
We received a number of comments related to the assessments and
their role in informing the proposed rule. Some commentors believed
that the proposed rule appropriately balanced the findings of the
assessments. Those commentors noted the tension between the concerns
raised in the assessments, but offered that the proposed rule
represented a reasonable accommodation of those concerns. Others
asserted that the conclusions of the risk assessment were given too
much weight in view of the evidence presented, or that the conclusions
of the benefits assessment were given too little weight.
One commentor noted that while the benefits assessment chronicled
actual, significant damages from accidental releases in terms of
casualties, evacuations, and property damage, the risk assessment did
not cite a successful terrorist attack on an industrial facility in the
United States. The commentor was thus concerned that the proposed rule
ignored the ``very real risks'' of chemical accidents in favor of what
the commentor characterized as ``greatly exaggerated fears of the
unknown.''
While there have thankfully been no successful terrorist or
criminal chemical releases in the United States (although there have
been several abroad), the risk assessment discussed two recent plots to
cause chemical releases that were thwarted by law enforcement. As the
risk assessment also pointed out, it is important to recognize that the
consequences of an intentional release could be devastating. A chemical
release intended and designed to cause maximum damage to property and
life--as terrorist events increasingly are--would have dire
consequences. The fact that an intentional release has not yet occurred
in the United States does not mean that the risk of such an incident
should be discounted or ignored. Nor does it mean that steps to prevent
such an incident should not be taken. As the risk assessment concluded,
trends suggest that the odds of such an event are increasing. The rule
recognizes that fact and balances that concern with the benefits to be
gained from providing the public with access to OCA information.
Similarly, one commentor asserted that the proposed rule sought to
eliminate the risk associated with posting OCA information on the
Internet rather than balancing that risk with the incentives for risk
reduction that would be created by making the information available to
the public. We disagree. CSISSFRRA requires the government to
promulgate a regulation that ``minimizes'' the likelihood of accidental
and intentional releases based upon the findings of the risk and
benefits assessments. The proposed rule was designed to do so. It would
not have eliminated all the risks cited by the risk assessment. To
further reduce the risk, the proposal could have called for any member
of the public to have access to OCA information for no more than one
facility per month or even per year, or could have made reading rooms
less numerous. Instead, the proposal called for any member of the
public to obtain OCA information for up to 10 facilities per month at
the 50 or more reading rooms across the country. It also called for an
Internet-based risk indicator system to stimulate the public's interest
in OCA information and the potential for risk reduction. The proposed
rule thus was an attempt to minimize the risk of chemical releases,
however caused, by providing the public with access to OCA information
while establishing safeguards intended to discourage criminal use of
the information.
Several commentors asserted that the proposed rule was ``arbitrary
and capricious'' because it failed to make a rational connection
between the facts found in the benefits assessment and the decisions
made in regard to the rule. In particular, the commentors pointed to
the benefits assessment's findings that the public will use information
to reduce risks to the extent the information is easy to access,
understandable, and in a format that facilitates comparison and
analysis. They claimed that the proposed rule would make OCA
information difficult to obtain. They argued that the proposed
restrictions would thus undermine the potential benefits of releasing
OCA information, and that EPA and DOJ essentially disregarded the
benefits assessment's findings.
We agree that there must be a rational connection between the
regulatory limitations established in this rulemaking and the findings
in the benefits and risk assessments. However, the final rule should
not, and cannot, respond to each of the assessments' findings standing
alone. CSISSFRRA requires the final regulations to govern the
distribution of OCA information in a manner that ``minimizes the
likelihood of accidental releases and the [increased risk of terrorist
and other criminal activity associated with the posting of OCA
information on the Internet] and the likelihood of harm to public
health and welfare,'' in light of the assessments. To meet that
requirement, the findings of both assessments must be considered to
determine how best to distribute OCA information in a way that reduces
the risk to public health and welfare of chemical releases, however
caused. We believe the final rule is informed by the findings of both
the benefits and the risk assessments.
Another commentor asserted that EPA and DOJ's justification for
withholding OCA information from the Internet is ``unique and
arbitrary.'' The commentor argued that, if posting OCA information on
the Internet is unacceptably dangerous due to the assistance it could
give a terrorist in identifying a potential target and planning an
attack, then many other types of information on the Internet could be
seen as equally dangerous, such as baseball schedules and stadium
seating capacities. The commentor explained that a terrorist could use
that information to determine potential casualty figures for a planned
attack during a game.
[[Page 48114]]
This criticism misinterprets the basis of our concern about OCA
information. The risk assessment found evidence that terrorists are
increasingly interested in using weapons of mass destruction (WMD) and
that chemical releases can be triggered from an industrial facility,
thereby converting that facility into a WMD. Based in part on that
evidence, the assessment concluded that posting OCA information on the
Internet would increase the risk of terrorists or criminals targeting
chemical facilities for attack. OCA information provides data that is
qualitatively superior to the sort of information cited by the
commentor. In particular, OCA information includes the number of
people, the size of the area, and the types of buildings and landmarks
that could be affected by a chemical release. As the assessment
emphasizes, that is precisely the type of information that terrorists
seek for purposes of planning an attack. Stadium seating capacities and
schedules, by contrast, provide information only about the number of
people that could be affected. For those reasons, EPA and DOJ conclude
that the release of some items of OCA information presents a terrorism
risk that warrants their exclusion from the Internet. Moreover, the
fact that chemical facilities, as opposed to baseball stadiums and many
other places where the public congregates, are themselves potential
WMD, makes clear that there is heightened risk in making OCA
information easily available to terrorists or other criminals.
We also received a comment that the proposed rule makes OCA
information more difficult to access than information currently
reported under EPCRA section 312, even though the benefits assessment
found that EPCRA section 312 information was not widely used because it
was difficult to obtain. However, the commentor did not correctly
characterize the benefits assessment's findings. The benefits
assessment found that several reasons account for the infrequent use of
EPCRA section 312 information. First, the public is not aware of the
availability of the EPCRA information because limited resources have
allowed only about half of the SERCs and LEPCs to publicize its
availability. Second, the effort required by members of the public to
locate their SERC or LEPC and request that information has been a
disincentive. Lastly, EPCRA data is not in a format that is easily
understood by the public.
As will be described in more detail later, under the final rule,
the public will more likely be aware of OCA information and have the
means to access and understand it. First, the EPA's website on the
Internet--a widely accessible medium--will inform the public of the
existence and availability of OCA information. Second, the website will
provide contact information and instructions for obtaining access to
OCA information, so members of the public will not have to locate that
information for themselves. Third, OCA information will be accessible
from more sources than is EPCRA section 312 information; while EPCRA
section 312 information is available only through SERCs and LEPCs, OCA
information will be available through federal reading rooms, as well as
through SERCs, LEPCs, and other related state and local agencies that
opt to provide access to local OCA information, as described in more
detail later. Fourth, some OCA information will be readily accessible
on the Internet. Finally, the public is more apt to use OCA information
because it is easier to comprehend than is the EPCRA section 312 data.
OCA information does not require calculations or analysis to determine
the potential consequences of potential releases; it communicates that
information directly and is designed to allow easy comparisons among
RMP facilities.
B. General Comments on the Rule
We received comments raising a variety of general or overarching
concerns with the proposed rule. One commentor asserted that the
proposed rule does not further right-to-know efforts. Other commentors
argued that terrorists will be able to get OCA information while the
proposed rule's restrictions on OCA information will only harm the
public. As stated above, DOJ and EPA agree that the public's right-to-
know is an important element in the reduction of accidental releases
and that risk reduction benefits will flow from the public's access to
OCA information. Accordingly, the proposed rule provided the public
with multiple avenues for obtaining access to OCA information,
including federal reading rooms, LEPCs, SERCs, and fire departments
that opt to provide read-only access. It also provided the public with
hazard information through a risk indicator system and clarified that
state and local government officials, as well as federal officials, can
communicate the substance of OCA information to the public as long as
they do not release the restricted forms of that information. While the
proposed rule would not have permitted unfettered release of OCA
information, it would have provided for dissemination of OCA
information in ways that are consistent with right-to-know efforts and
would have allowed the public and industry to better prevent and
prepare for chemical releases, whether or not intentionally caused. As
explained further in this preamble, the final rule adopts and improves
on those public access provisions.
One commentor argued that the proposed system for providing the
public with access to OCA information would undermine the utility of
the CAA's citizen suit enforcement provision by denying members of the
public the information they need to prosecute such suits. But as noted
above, the system would not deny public access to OCA information, only
control it. Federal, state, and local reading rooms, and the Internet
would all be potential outlets for the information. As described later,
we have also sought to improve the proposed system's ability to assure
reasonable access to OCA information by all members of the public.
One commentor expressed concern that the proposed regulation would
``disenfranchise'' U.S. citizens located outside the country by
withholding access to OCA information from them. The basis for that
concern was our proposal to define ``member of the public or person''
as an individual located in the United States. We did not intend to
withhold access to OCA information from any U.S. citizen. Rather, we
intended only to limit our reading room obligation to establishing
rooms in the United States, where the vast majority of persons affected
by RMP facilities are located. Given the resource implications of
establishing federal reading rooms, we considered it appropriate to
commit to locating at least 50 rooms in the United States and retain
discretion to locate more elsewhere. We continue to believe that that
is the appropriate course to take.
As described later, we are developing an approach to operating
reading rooms that will give us flexibility in where we locate them; to
the extent we learn that there is demand for reading room access by
U.S. citizens abroad, we will consider providing reading room access in
appropriate locations. Nonetheless, we realize that the definition of
``member of the public or person'' need not be limited in the way
proposed to accomplish our objective. The reading room provision of the
rule itself specifies that the required reading rooms be located across
the United States. Moreover, we realize that the proposed definition
would have been problematic for some other rule provisions that used
the terms ``public'' or ``person.'' We have thus deleted the phrase
``located in the United States'' from the definition. At the same time,
[[Page 48115]]
we have revised the rule provision calling for a system that indicates
whether an address is within a facility's vulnerable zone so that our
obligation extends only to persons located within any state (defined to
include the 50 states, the District of Columbia, and U.S. territories).
The vast majority of persons affected by vulnerable zones are within a
state, and we consider it reasonable and prudent to limit our
obligation in order to limit the potential impact of that obligation on
our resources. We expect, however, to answer inquiries from persons
located outside the U.S. unless those inquiries become voluminous.
Several commentors voiced concern that, without ready access to OCA
information, the public would be unable to hold EPA accountable for the
effectiveness of the RMP program. We disagree. We do not believe that
changes over time in any single set of data (e.g., distance to
endpoint) are sufficient to measure the effects of the RMP program on a
facility's practices. Differences in OCA data may reflect differences
in assumptions and models used in conducting the analysis. Other RMP
information, including accident histories and information about
prevention and response programs, offers a more comprehensive basis for
measuring a facility's progress or comparing facilities' safety
practices. Moreover, RMP information except for OCA information is
already available on the Internet. Consequently, there is already a
wealth of information that an individual can use to determine the
compliance status of an individual facility, even without the
additional OCA information offered by the proposed rule. We thus
believe that the ready access to that information sufficiently enables
interested individuals to evaluate the effectiveness of the RMP
program.
One commentor claimed that the proposed rule distorted the notion
of a ``public record'' because the proposed rule would not allow
publicly released OCA information to be copied or carried away from
reading rooms. The commentor noted that the proposal treated OCA
information as ``public'' in the setting of the reading rooms but
prohibited it from release to the public in the context of the
Internet. We find that the proposal's treatment of OCA information is
consistent with CSISSFRRA's statutory framework. Congress anticipated
that OCA information in different forms could be disseminated
differently; under CSISSFRRA the government is required to provide the
public with access to paper copies of OCA information in limited
quantities, and in addition the government is required to assess
whether and how to provide OCA information on the Internet (CAA section
112(r)(7)(H)(ii)(II)). Thus, the proposed rule's approach to
dissemination of OCA information was well within the scheme
contemplated by the CSISSFRRA.
We received a comment that the proposed rule is illogical because
it tracks members of the public who review OCA information at federal
reading rooms but allows companies to release OCA information to the
public without restriction. Both of those aspects of the proposed rule,
however, flow from the statute itself. First, CAA section
112(r)(7)(H)(ii)(II)(aa) specifies that the final rule must provide
access to paper copies of OCA information for a ``limited number'' of
facilities. The only way the government can implement the ``limited
number'' provision is to limit the number of facilities for which an
individual can receive access to OCA information. Second, CAA section
112(r)(7)(H)(v)(III) contemplates that facilities will release their
OCA information to the public if they so choose. It provides that the
statute's restrictions on dissemination of OCA information do not apply
to information released without restriction by facilities, and it
requires facilities that provide OCA information to the public under
those terms to notify the Administrator, who is directed to maintain a
public list of such facilities. Congress thus clearly intended to allow
facilities to release their OCA information as they consider
appropriate. Congress' approach to facilities' release of their own
information does not conflict with the concern expressed in the risk
assessment that large quantities of OCA information would be
disseminated in a searchable format on the Internet. Individual
facilities separately releasing their OCA information does not
significantly raise that concern.
Relatedly, one commentor asserted that the federal government
should provide access to OCA information that facilities release
without restriction. As noted above, CSISSFRRA requires EPA to make
publicly available a list of the facilities that have notified EPA that
they have released their OCA information without restriction.
Approximately 1,000 facilities have notified EPA, and EPA has made a
list of those facilities available on its website. That list will
enable members of the public to obtain OCA information from those
facilities. At the same time, CSISSFRRA does not require that EPA and
DOJ make publicly available the OCA information released by listed
facilities. Neither EPA nor DOJ will provide the OCA information merely
because it has been released by the listed facilities, for the security
reasons cited above.
Several commentors asserted that even greater restrictions should
have been proposed because the rule would not stop OCA information from
being hand-copied and posted on the Internet. We do not believe
Congress intended for us to prevent members of the public from hand-
copying the OCA data that they view. CAA section
112(r)(7)(H)(ii)(II)(aa) guarantees the public ``access'' to paper
copies of OCA information for at least a limited number of facilities,
and the utility of ``access'' would be greatly diminished if the public
had to rely on memory alone to recall that information. Also, CAA
section 112(r)(7)(H)(viii) expressly precludes mechanical and
electronic copying of the electronic OCA information made available
under that provision. It is silent with regard to copying by hand. The
fact that Congress expressly precluded mechanical and electronic copies
suggests that it was aware of the problem of copying and made an
affirmative decision to prohibit only certain forms of copying. We thus
believe that Congress' silence with regard to copying by hand is
properly interpreted to mean that hand copies are to be permissible.
Another commentor claimed that the best manner of determining
whether the proposed rule provided adequate public access to OCA
information compared to other alternatives was to give the full RMP
database to qualified researchers so that they could use it to conduct
a peer review analysis of the proposal. CSISSFRRA mandates that the
means of disseminating paper copies of OCA information be based upon
assessments conducted by the government; it does not appear to
contemplate the sort of peer review process that the commentor
proposed. Further, it is unlikely that the short time frame provided by
the statute would have allowed for such a process. Moreover, we do not
believe that the commentor's method of assessing the various
alternatives for providing the public with access to OCA information
would be preferable to the method of analysis that we conducted through
our assessments. We agree, however, that there are public benefits to
providing qualified researchers with access to OCA information.
CSISSFRRA does not require that this rulemaking establish a means of
doing so, but we are working on devising and implementing a system for
giving qualified researchers access to
[[Page 48116]]
OCA information, as required by CAA section 112(r)(7)(H)(vii).
One commentor asserted that it was unnecessary for the public to
receive information about facilities outside their communities, and
that a facility's OCA information should only be available to members
of the community in which it is located. Such an approach, however,
would be inconsistent with CSISSFRRA and the findings of the benefits
assessment. CAA section 112(r)(7)(H)(ii)(II)(aa) expressly guarantees
access to paper copies of OCA information for a limited number of
facilities ``located anywhere in the United States, without any
geographical restriction.'' The benefits assessment also notes that a
person interested in assessing a local facility's safety practices may
find it useful to compare that facility's OCA information with that of
similar facilities located elsewhere.
Some commentors suggested that the creation of 50 federal reading
rooms, or approximately one per state, has environmental justice
implications. The Environmental Justice Executive Order (Exec. Order
No. 12898, 59 FR 7629 (1994)) requires that each federal agency conduct
all activities affecting the environment or human health in a manner
that does not discriminate by race, color, or national origin, and
address, as appropriate, any disproportionately high and adverse human
health or environmental effects on minority and low-income populations.
Executive Order 12898 also encourages agencies to work to ensure that
public documents relating to human health and the environment are
readily accessible to the public. We believe that our approach,
including various means of access in addition to federal reading rooms,
will not have a disparate impact upon minority groups or low-income
groups. As discussed below, we are committed to providing reasonable
access to everyone seeking to view OCA information and have made
changes to the rule reflecting that intention. We expect that the vast
majority of federal reading rooms will be placed in urban areas with
relatively large minority and low-income communities. Those locations
will provide practical access to OCA information for those communities,
some of which have historically suffered from a disproportionate
environmental hazard burden. The rule provides for additional access to
OCA information by allowing state and local government agencies to
provide access under the ``enhanced local access'' section of the rule.
Also, the vulnerability zone indicator system, which is accessible via
email, telephone, and U.S. mail, will provide an individual with
additional data on some aspects of the risk expressed by OCA
information.
Some commentors also expressed concern that little had been done to
involve minority and poor communities in the development or public
review of the proposed rule, contrary to the Environmental Justice
Executive Order. EPA and DOJ disagree. Especially in light of the
relatively short period of time we had to conduct the risk and benefits
assessments, as well as to propose and finalize this rule, we believe
that we provided a reasonable opportunity for review of the proposed
rule by minority and poor communities in compliance with that Executive
Order. The proposed rule outlining the federal government's policy was
published in the Federal Register and available on the EPA website. In
addition, we provided additional notice of the proposal by holding a
public hearing and providing individual notification to thousands of
individuals across the country, including state and local government
agencies.
Another commentor faulted the proposed rule for not acknowledging
Indian country, tribal governments, or tribal equivalents of SERCs and
LEPCs. CSISSFRRA itself does not address Indian country or tribes. It
amends the CAA, which defines ``state'' in a way that does not include
Indian country. However, CAA section 301(d) authorizes EPA to
promulgate regulations specifying those CAA provisions for which it is
appropriate to treat Indian tribes as states. EPA has promulgated that
regulation (63 FR 7271 (Feb. 12, 1998)), which provides that tribes can
take delegation of programs under CAA section 112, including the RMP
program, if EPA finds they meet specified criteria. Thus, a tribe found
to meet those criteria may be treated as a state and receive and
disseminate OCA information to the same extent and in the same manner
as any state under the rule being promulgated.
C. Rule's Impact on Risk Reduction
A number of commentors agreed that the proposed rule generally
provided for public access to OCA information in a way that would
minimize the likelihood and consequences of chemical releases, however
caused. Some of those commentors noted that other information available
in RMPs, under EPCRA or other programs, would, on their own or in
tandem with OCA information, allow the public to learn about and
understand the hazards and risks posed by chemical plants in their
communities. In contrast, some commentors expressed concern that the
proposed rule would not minimize overall risk, and even more
significantly, might increase overall risk by making it too difficult
for the public to access OCA information that could be used to reduce
the likelihood of accidents.
Some commentors argued that the proposed rule would take away a
risk reduction tool without decreasing existing dangers. We disagree
with that statement. The agencies did not propose to ``take away a risk
reduction tool,'' since there still would be public access to OCA
information. In order to reduce the risk associated with Internet
posting of OCA information, the proposed rule delineated procedures for
obtaining access to the information and limitations on the amount of
information that could be obtained by any member of the public. It
provided for access to up to 10 facilities' OCA information per
individual per month, access that would allow members of the public in
the vast majority of counties to obtain information for local
facilities and a few additional facilities for a basis for comparison.
In addition, the proposal in no way attempted to restrict the use of
that risk reduction tool once obtained.
A few commentors argued that the proposed rule encouraged secrecy,
which would breed incompetence and complacency. While we agree that
secrecy can have such an effect, in this case the public will have
access to OCA information, so facilities' information will be far from
secret. In addition, other RMP information currently available on the
Internet, including information concerning facilities' accident
prevention programs, provide important information for assessing and
comparing facilities' practices. Likewise, other publicly available
environmental reports--such as those concerning accidents reported
under EPCRA and the Comprehensive Environmental Response, Compensation,
and Liability Act--are useful in evaluating a facility's safety
practices. OCA information provides a particularly simple way of
roughly assessing and comparing the hazards facilities pose, but it is
not the only information capable of communicating such hazards, as a
number of commentors pointed out.
Several commentors argued that, by making OCA information difficult
to access, the proposed rule would force the public to rely on
government officials for risk information without being able to check
the accuracy of that information. Other commentors claimed that the
public might resort to other forms of less reliable, more exaggerated
information that would make local risk
[[Page 48117]]
reduction efforts more difficult. Relatedly, another commentor argued
that, to the extent other, more exaggerated information is generated as
a substitute for OCA information, terrorists and other criminals may be
led to believe that consequences of a release would be greater, thereby
increasing the risk of a release. The fundamental premise of those
comments is that the rule would render OCA information inaccessible. We
disagree. As mentioned above, we are committed to making OCA
information reasonably available to the public and have made changes in
the final rule to ensure such access. Consequently, local and state
governments need not ask the public to trust their representations but
may provide access to OCA information and other information that the
public may use to verify government assertions about the risk of
chemical releases.
Several commentors asserted that the proposed rule, in validating
the idea that public dissemination of OCA information poses a risk,
would have a ``chilling effect'' on local officials' communication of
OCA data, thus curtailing accident prevention efforts that result from
public awareness and pressure. We did not intend to create such a
chilling effect. Indeed, we believe dialogue among government, the
public, and industry is essential to further risk reduction efforts. As
we explained in the proposal, we have attempted to address the concern
about CSISSFRRA's perceived chilling effect by explaining in the rule
the ways in which state and local government agencies may legitimately
disseminate OCA information, or descriptions thereof, to the public. In
fact, the rule encourages appropriate local and state agencies to
provide public access to such information, which should counter any
inference to the contrary. Further, it is worth reiterating that
government officials may be held criminally liable only for ``willful''
violations of the restrictions on OCA information dissemination. In
other words, the government would be required to demonstrate that the
official knew his or her actions to be unlawful. EPA and DOJ moreover,
will continue to provide guidance to state and local covered persons to
explain the extent to which they may lawfully disseminate OCA
information, or communicate the substance of that information, under
the final rule.
Similarly, one commentor expressed concern that the proposed rule
might discourage members of industry from participating in SERCs and
public meetings at which OCA information is discussed. In particular,
the commentor asserted that proposed section 1400.6(b) could be
interpreted to render it unlawful for industry members serving on SERCs
to provide OCA information for their facilities to the public, if those
facilities have not formally decided to release that information. In
many instances, whether CSISSFRRA is applicable will depend upon the
context in which OCA information is being disseminated. For example, in
the instance cited by the commentor, 1400.6(b)'s restrictions on
dissemination apply only if the member of industry is distributing OCA
information to the public in his or her capacity as a representative of
the SERC. In addition, CSISSFRRA does not restrict his or her ability
to participate in public discussions about OCA information; in fact,
CSISSFRRA section 4 anticipates that members of industry will engage in
such discussions with the public.
Several commentors argued that, if the rule makes public access to
OCA information difficult, it should compensate for any resulting
decrease in risk reduction incentives by requiring facilities to secure
their sites and/or take prescribed risk reduction steps, such as
reducing their inventory of dangerous chemicals or substituting safer
chemicals to the extent feasible. Other commentors disagreed, asserting
that requiring facilities to make themselves secure from terrorist
attacks or to take other risk reduction measures would be an
inappropriate remedy for the risk posed by broad release of OCA
information. To begin with, we note that CSISSFRRA requires the final
rule to ``govern[] the distribution of [OCA] information.'' It does not
call on the government to decide whether to impose further substantive
requirements on facilities to reduce the risk of chemical releases,
however caused. In the short time available to conduct the assessments
and rulemaking on the distribution of OCA information, it was not
possible for us to address the broader policy, programmatic, and legal
issues posed by the commentors' suggestion for additional regulatory
requirements. CSISSFRRA does, however, include a requirement that DOJ,
in consultation with relevant federal, state, and local agencies, as
well as members of industry and the public, conduct studies to examine
the issue of site security at RMP facilities and the extent to which
the RMP rule effectively addresses that issue. DOJ is working to comply
with that requirement. In the meantime, EPA has issued a site security
alert informing industry of various risks posed by criminal activity
related to chemical facilities.
D. Reading Rooms
1. General Comments on Reading Rooms
As indicated above, the proposed rule called for providing the
public with access to paper copies of OCA information through the
creation of at least 50 federal reading rooms geographically
distributed across the United States. Several commentors expressed
concern that the costs of creating federal reading rooms could outweigh
the benefits. Further, several other commentors suggested that it would
be more appropriate for LEPCs, SERCs and/or other local groups to be
the principal providers of OCA information; some commentors also urged
EPA to help fund such efforts. Some commentors recommended that the
reading room approach be abandoned or scaled down out of concern that
reading rooms would not adequately safeguard the OCA information and
could result in the widespread dissemination of OCA material. Other
commentors questioned whether federal reading rooms would provide
reasonable access, particularly for people who live some distance from
reading rooms. Finally, other commentors supported the federal reading
room approach but made suggestions about how to make reading rooms more
effective and secure.
For the reasons discussed below, we continue to believe that
providing the public with access to paper copies of OCA information is
best done through reading rooms. We are developing an implementation
approach for federal reading rooms that will allow read-only access to
OCA information in a reasonably secure manner that is convenient for
the public and efficient for the government. We do not believe that
existing federal statutes authorize us to rely solely on LEPCs, SERCs,
or other state or local entities to provide reading room access;
requiring such agencies to do so, moreover, might raise constitutional
concerns regarding the appropriate relationship of federal and state
power. CSISSFRRA makes the federal government responsible for
distributing OCA information. Nevertheless, LEPCs, SERCs, and other
emergency prevention, planning, and response agencies can play an
important part in facilitating public access to OCA information, and
the final rule being promulgated encourages them to do so. We also
intend to provide assistance to interested state and local agencies.
As for whether reading rooms can provide reasonable access, we are
committed to establishing a network of federal reading rooms and other
potential state and local outlets (further
[[Page 48118]]
described below) that would ensure that every member of the public has
a reasonable opportunity to obtain access to OCA information. We
believe that federal reading rooms can and will be an appropriate and
cost-effective mechanism for providing the required public access to
OCA information.
2. The Number of Paper Copies
We received a comment interpreting the limit on the number of RMP
facilities for which an individual may view paper copies of OCA
information as 10 per person per visit. Today's notice clarifies that
the limit is 10 per person per month, regardless of the number of
reading room visits a person makes. Any person may visit a reading room
multiple times during a single calendar month to view the OCA
information for the same 10 facilities. A person may not visit multiple
reading rooms to view OCA information for more than 10 different
facilities in a single month. We have changed the text of the
regulation to clarify that point.
We received many comments on what the appropriate ``limited
number'' should be. Some commentors expressed concern that the proposed
limit of 10 per month was too generous considering the potential
criminal use of that information and suggested a lower number, such as
10 per year. Several commentors indicated that the proposed limit of 10
was arbitrary, unreasonable, and/or would hamper the goal of providing
the public with access to paper copies of OCA information because the
proposed limit of 10 per month would be insufficient for citizens
living or working in areas with high concentrations of RMP reporting
facilities, or would hinder individuals wishing to conduct nationwide
comparative research. Finally, some commentors stated that the limit of
10 per month was appropriate.
Several commentors also raised issues concerning the application of
the 10 per month limit. One commentor suggested that the limit apply
not to individuals but to organizations, so that an organization could
not use its employees or members to compile collectively OCA
information for more than the prescribed ``limited number'' of
facilities. Another commentor argued that members of the public have a
legitimate interest only in OCA information for facilities in their
community, and that the limited number should thus be applied in a way
that provides access to information only for such facilities. Two other
commentors recommended that OCA information be provided only to state
and local officials with emergency planning, prevention, or response
responsibilities.
We note at the outset that CSISSFRRA requires that these
regulations provide access for ``any member of the public'' to paper
copies of OCA information for a limited number of facilities ``located
anywhere in the United States, without any geographical restriction''
(CAA section 112(r)(r)(H)(ii)(II)(aa)). We thus do not have the
discretion to deny the public access to paper copies of OCA
information, to establish a limit that applies to organizations instead
of individuals, or to restrict the geographical scope of the facilities
for which a member of the public may request OCA information. The
benefits assessment also makes clear that public access to OCA
information would stimulate further risk reduction and that the
public's ability to compare the hazards and safety practices of similar
facilities located in different places is important to stimulating that
risk reduction.
With respect to the appropriate numbers limit, we explained in the
proposal that we chose a limit of 10 facilities per individual per
month based on consideration of many of the issues expressed in the
comments received. As required by the law, we weighed the risks that
would result from unlimited reading room access to paper copies against
the benefits that would accrue from public awareness of potential
release hazards, as communicated through OCA information. A limit was
proposed that would hinder the ability of an individual or group to
gather large quantities of OCA information to post on the Internet,
while allowing individuals in most parts of the country or in most
counties to gain access to OCA data for all the facilities in their
community and a few more for purposes of comparison.
In determining that limit, we conducted an analysis of the
geographic distribution of RMP facilities across the nation. The
analysis showed that 82% of all counties that have RMP facilities have
no more than 10 such facilities. Because residents of most counties
would be able to review OCA information for all the facilities in their
county in a single visit to a federal reading room, EPA and DOJ
believed that a limit of 10 per month would provide reasonable access
for persons living or working in areas with RMP facilities. Moreover,
under the 10 per month limit, in the great majority of those counties,
residents would also be able to review OCA data for RMP facilities
located outside their county.
At the same time, we recognized that the proposed limit of 10 per
individual per month would not permit all members of the public to
obtain OCA information for every facility in their own communities. The
proposed rule, therefore, included provisions to authorize and
encourage LEPCs, SERCs, and fire departments to supplement the access
provided by federal reading rooms by providing read-only access to OCA
information for facilities located in the LEPC's jurisdiction and
facilities with vulnerable zones that extend into that jurisdiction.
However, as discussed more fully below, we received comments that many
LEPCs and SERCs would be unwilling and/or unable to provide such
access.
In passing CSISSFRRA, Congress emphasized that members of the
public should have access to OCA information, particularly for
facilities in their local communities (see 145 Cong. Rec. S7545, daily
ed. June 23, 1999 (statement of Sen. Chafee)). We agree that every
member of the public should be able to access OCA information for
facilities in the communities where he or she lives or works without
making multiple trips to a federal reading room. We have thus decided
to require federal reading rooms to provide any person with access to
OCA information that the LEPC in whose jurisdiction the person lives or
works is authorized to provide (i.e., access to OCA information for
facilities located in the jurisdiction of the LEPC and facilities with
a vulnerable zone that extends into that jurisdiction). That access
will be in addition to access to OCA information for up to 10
facilities located anywhere in the country, without geographical
restriction. With reading room access to OCA information for local
facilities assured, access to OCA information for 10 facilities located
anywhere will allow members of the public to compare facilities in
their community with similar facilities located elsewhere and to learn
about facilities in communities where they might move or where
relatives or friends live or work.
In providing federal reading room access to OCA information for a
person's local facilities, we do not want to discourage LEPCs, SERCs,
and others from providing local access to the same. Obviously, it will
be more convenient for a member of the public to access information
locally than at a federal reading room that may be located many miles
away. Also, we want to encourage dialogue between members of the public
and their local officials responsible for chemical emergency planning
and response. By making local OCA information available locally, LEPCs,
[[Page 48119]]
SERCs, and other state and local agencies can encourage the public to
become involved in chemical risk reduction efforts. As more fully
discussed in the next section of this notice, we are committed to
helping LEPCs, SERCs, and others provide that local access.
3. Operation of Reading Rooms
Some commentors suggested that federal reading rooms be open at
nights and on weekends. We understand that some members of the public
may find it difficult to reach reading rooms during the normal work
week. However, due to cost, personnel, and security concerns, reading
rooms will be located in federal buildings, which are typically open
only during normal business hours. We will explore the extent to which
reading rooms can also be open at other hours to accommodate members of
the public.
As urged by several commentors, we have endeavored to develop a
cost-effective and secure means of operating federal reading rooms. At
some reading rooms, access will be available on a walk-in basis because
the OCA data will be maintained at the reading room. At other reading
rooms, however, the OCA data will not be maintained on-site, and
therefore a person wishing to view OCA data at those reading rooms will
need to contact a central office at a toll-free number at least three
days prior to the date on which the person would like to view the OCA
information at the reading room. During the toll-free call, the
requestor will be asked to provide his or her name, telephone number,
and the names of the facilities for which he or she is requesting OCA
information. That information will enable the central office to
schedule an appointment for the requestor at a reading room, relay the
requested copies of OCA information to that reading room, and, if
necessary, contact the requestor. That information will not be retained
beyond the requestor's appointment date.
As discussed below, at the reading room, the requestor will need to
display photo identification issued by a federal, state, or local
government agency, sign a sign-in sheet, and certify that the requestor
has not received access to OCA information for more than 10 facilities
during that calendar month. The requestor will then receive access to
the requested OCA information. Requestors will be limited to access to
paper copies of OCA information for a total of 10 facilities during a
calendar month, regardless of how many reading rooms they visit during
a single month.
As discussed above, any person will also receive access at a
federal reading room to OCA information that the LEPC in whose
jurisdiction the person lives or works is authorized to provide (i.e.,
access to OCA information for facilities located in the jurisdiction of
the LEPC and facilities with a vulnerable zone that extends into that
jurisdiction). Persons seeking such access will also be asked to sign
in and to provide proof demonstrating that he or she lives or works in
the LEPC jurisdiction for which the OCA information has been requested.
They will not, however, be required to sign a certification.
4. The Number of Reading Rooms
We received a range of comments on the appropriate number of
reading rooms. Several commentors suggested that fewer reading rooms
would be adequate and appropriate while many commentors expressed
concern that 50 reading rooms would not provide reasonable public
access due to issues such as time and travel costs, especially in large
states and for low income groups.
We are committed to providing reasonable access to OCA information.
We intend to establish reading rooms in virtually every state, the
District of Columbia, and outlying territories having RMP facilities.
In addition, we will work to set up additional reading rooms in states
that have a significant number of RMP facilities, such as California
and Texas. While we anticipate establishing more than 50 reading rooms,
we have not increased the number of rooms required by the rule because
the need for additional rooms may be affected by the extent to which
state or local government agencies provide access under the enhanced
access provisions of the rule. Moreover, as we implement the reading
room provision and learn more about the demand for reading rooms in
different parts of the country, it may become appropriate to relocate
reading rooms.
5. The Location of Reading Rooms
As for the specific locations of the federal reading rooms, a
number of commentors suggested a number of factors to consider in
determining locations. We agree with those suggestions and have decided
to use the following criteria in making our decisions: equitable
distribution across the United States and its territories; the density
of the population surrounding the location; the availability of public
transportation to the location; the ability to provide security at the
location; and the availability of federal offices that could readily
implement the reading room requirement at reasonable cost. Federal
offices, it should also be noted, are handicapped accessible. The
location of federal reading rooms will be posted on EPA's and DOJ's
websites when they are determined.
6. Security Measures at Reading Rooms
We proposed that a reading room representative be required to view
a government document identifying that individual before granting that
individual access to OCA information. Some commentors stated that that
requirement would have a chilling effect on the public's use of reading
rooms because some people may be reluctant to show identification to
the government. Other commentors urged that we require photo
identification to ensure that the person presenting the identification
is in fact the person to whom the identification was issued.
We recognize that some individuals may be reluctant to show
identification to a government official. However, the personal
identification requirement is a reasonable means of accomplishing the
statutory requirement that individuals have access to ``a limited
number'' of paper copies of OCA information. Further, as noted in the
risk assessment, EPA and DOJ believe that the identification
requirement will also decrease the likelihood that OCA information
would be obtained by individuals seeking it for criminal purposes
because such individuals prefer to conceal their activities. With
respect to the type of identification, EPA and DOJ agree that photo
identification issued by a local, state, or federal government agency
(e.g., a driver's license or passport) should be required. That
requirement will significantly reduce the risk that someone will
attempt to use identification not his or her own.
One commentor suggested that there should be some type of
identification validation system to ensure the accuracy of an
individual's identification document. EPA and DOJ have concluded that
it would be too costly to create an independent identification
validation mechanism. The responsibility for checking individuals'
identification documents will be left to those operating the federal
reading rooms. EPA and DOJ do not consider that to be a significant
problem, since the majority of locations at which the reading rooms
will be located are federal agencies that have security staff that
already visually check the identification of all persons seeking entry
to the federal facility or other areas of limited access. Individuals
using the federal reading rooms will have their identification checked
in the same manner as would any member of
[[Page 48120]]
the public seeking entry into federal buildings.
As discussed above, the final rule will require federal reading
rooms to provide any member of the public with access to the OCA
information that the LEPC where the person lives or works would be able
to provide to them. To implement that provision, it will be necessary
for reading rooms to check identification and documentation to ensure
that a requestor receives access only to the local OCA information to
which he or she is entitled (i.e., OCA information for stationary
sources located within the jurisdiction of the LEPC in which the
individual lives or works and for any other stationary sources that
have a vulnerable zone that extends into that LEPC's jurisdiction). We
will create guidelines for federal reading room personnel regarding
such procedures.
In the NPRM we described procedures by which reading rooms would
determine whether a requestor had exceeded the 10 per month allotment.
We anticipated that reading rooms would keep daily sign-in sheets to
record the name of any person who received access to OCA information
and the name and number of facilities to which that person had received
access. Whenever someone requested access to OCA information, reading
room personnel would review the sign-in sheets for that day and the
previous days during the month to determine how many, if any,
facilities' OCA information that person had already received that
month. We noted that sign-in sheets would be protected under the
Privacy Act (5 U.S.C. 552a) and would be retained for three years.
We received several comments on the record keeping aspect of the
proposed rule. Several commentors expressed concern that the use of
sign-in sheets would raise privacy concerns, and one commentor
expressed a related concern that the proposed rule was silent as to how
the federal government would use the information. Other commentors
agreed with the identification requirement and the concept of keeping
some type of record, but recommended that the final rule require record
keeping and a corresponding check on people using a reading room in
order to ensure that they have not had access to OCA information for
more than 10 facilities per month. Two commentors suggested that EPA
and DOJ establish a national database as a means of enforcing the 10
paper copy per month limit.
We recognize that privacy concerns are raised whenever the
government collects information about individuals. We also are mindful
of the need, identified by the risk assessment and required by
CSISSFRRA, to limit the number of facilities for which individuals can
access OCA information in paper form. We thus have endeavored to design
a system that will effectively implement the limitation but minimize
the need for keeping records on individuals' access to OCA information.
Specifically, we will use the sign-in sheet system discussed in the
NPRM (65 FR 24853 (April 27, 2000)), and keep the sign-in sheets in a
manner that will minimize privacy concerns and that will not entail the
creation of a system of records under the Privacy Act. The Privacy Act
applies to records retrieved by name within systems of records. Federal
reading rooms will not create an elaborate tracking system; they will
not index or otherwise manipulate the sign-in sheets according to
individuals' names. Instead, a reading room representative will
visually inspect the sign-in sheet(s), which will be organized
chronologically, for the month in which an individual seeks access to
paper copies to see if that individual's name appears on the sign-in
sheet(s) for dates earlier in the month and, if so, if that individual
has already received OCA information for the allotted 10 facilities
without geographical restriction.
We believe that the sign-in system will help deter individuals from
seeking improperly to obtain OCA information exceeding the 10
facilities per month national limit. To further deter individuals from
attempting to exceed their allotment by visiting more than one federal
reading room in a month, reading room personnel will be instructed to
provide access to OCA information only to individuals who have signed a
certification that they have not exceeded their allotment. The
certification will inform members of the public that they may be
subject to criminal penalties under federal law for falsely certifying
that they have not received OCA information for more than 10 facilities
that month.
It should also be noted that the information recorded on sign-in
sheets may be used by law enforcement in the event of a duly authorized
investigation of a violation of civil or criminal law. For that reason,
the reading rooms will retain the sign-in sheets for three years. In
the event that the sign-in sheets are compiled into a system retrieved
by name for purposes of such an investigation, they will be subject to
the Privacy Act and will be handled accordingly. Federal law
enforcement agencies have already established Privacy Act systems
applicable to their indexed investigative records, and if the
information from sign-in sheets were so compiled, it would receive
those protections.
The reading room records will not be used beyond the purposes
outlined above (i.e., to ensure compliance with the 10 facility per
month limit and to carry out authorized law enforcement
investigations).
In deciding to adopt the sign-in certification approach, EPA and
DOJ have decided not to institute a national database for enforcing the
10 facility per month limit, as some commentors recommended. We
anticipate sign-in sheets with certifications should provide adequate
assurance that the monthly limit on OCA information is not exceeded.
However, after gaining experience with the federal reading rooms, we
will evaluate whether the sign-in sheet system is in fact effective.
For that purpose, we will review a sample of sign-in sheets for several
reading rooms to determine if the existing system is adequately
enforcing the limit. Based on that review, DOJ, EPA, and OMB will
consider whether a national database or other tracking system should be
instituted to enforce the limit.
One commentor asserted that the establishment of such records would
violate the Paperwork Reduction Act of 1995 because it would not
provide the government with information that has practical utility.
That assertion is not correct. As discussed above, the information
collected would have practical utility, namely to ensure that the
statutory and regulatory limit on access to OCA information in paper
form is properly applied.
7. Alternatives to Reading Rooms
We expressly asked for comments on whether, as an alternative to
reading room access to information, paper copies should be released to
the public upon request. Some commentors stated that there should be an
alternative system of direct delivery of OCA paper copies to interested
parties. They asserted that the proposed federal reading room system
would be insufficient to provide OCA paper copy access to all
interested citizens. In addition, they indicated that, because only a
limited number of federal reading rooms would be established, some
citizens would find it inconvenient to travel the distances necessary
to access the information.
Other commentors opposed off-site distribution of paper copies or
allowing individuals to take away paper copies from reading rooms. Some
noted that such a system would pose a significant security risk because
it would increase the risk of OCA information being
[[Page 48121]]
disseminated widely, thus violating the intent of CSISSFRRA. Some
emphasized that paper copies, once outside the control of the
government, could easily be scanned into an Internet database and that
such a system would provide potential terrorists with the type of
Internet access to OCA information that the proposed rule was designed
to prevent. Further, in noting that potential terrorists may forgo
attempts to gain access to OCA information if they must do so in person
and submit to an identification check, some commentors stated that the
mail delivery alternative would lessen the deterrence benefit of on-
site access.
We have considered the alternatives of mail delivery of OCA
information to interested citizens and the distribution of take-away
copies, and have determined that both would present an unacceptable
security risk. With respect to mail delivery upon request, any
safeguards, such as a requirement of proof of residence at the delivery
location, could easily be circumvented by an individual or group
establishing a ``phantom residence.'' We also agree with the commentors
who noted that requiring persons to go to a federal reading room and
provide identification would provide some deterrence to those potential
terrorists who might wish to keep their interest in the information
hidden. We further agree that, once paper copies have left the federal
reading rooms, they can easily be scanned onto the Internet where they
could be viewed anonymously by those with criminal intent. Anonymous
access to significant amounts of OCA information is precisely what this
rule is designed to prevent. As a result, the final rule will use
reading rooms to provide access to paper copies because reading rooms
allow for that access to occur within a controlled setting.
E. Enhanced Local Access to OCA Information
Commentors generally supported the proposed rule's provisions for
enhanced local access as a promising means of facilitating the public's
access to OCA information and public-private dialogue about chemical
safety in their communities. Many of those commentors, however, also
pointed out a number of obstacles to making enhanced local access a
reality and suggested ways of overcoming those obstacles.
A key element of the proposal for enhanced local access was
clarification that state and local government officials (as well as
federal officials) may communicate to the public the substance of OCA
information (i.e., the OCA data elements reported in RMPs), even though
they may not disseminate the official forms in which the data is
reported and compiled (i.e., the OCA portions of RMPs and EPA's OCA
database). While developing the proposed rule, we learned that many
state and local officials were concerned that CSISSFRRA may preclude
them from communicating OCA data in any form. As we explained in the
proposal, the ``scope'' section of ``CSISSFRRA'' (CAA section
112(r)(7)(H)(xii)(II)) expressly provides that the statute ``does not
restrict the dissemination of [OCA] information by any covered person
[defined by CSISSFRRA as government officials and qualified
researchers] in any manner or form except in the form of a risk
management plan or an electronic data base created by the Administrator
from [OCA] information.'' In other words, while covered persons may not
disseminate the OCA portions of RMPs or any EPA database created from
those portions, they may discuss or otherwise communicate the data
reported in those portions. We thus proposed capturing that important
point in the proposed regulations.
We received comments supporting and questioning the proposed
clarification. Several commentors from LEPCs and SERCs indicated the
clarification was helpful but sought further guidance on how OCA data
could be lawfully disseminated. Other commentors were concerned that
the clarification was not consistent with the law, and that
communication of OCA data was risky because it is dissemination of the
information's content, not its format, that they believed poses the
risk. Another commentor expressed concern that the clarification could
be interpreted to allow dissemination of the restricted portions of
RMPs with only minor changes in format, which would undermine the
protections of CSISSFRRA.
After revisiting CSISSFRRA and its legislative history, we have
concluded that the proposed rule's clarification not only is consistent
with the law but virtually mirrors it. As noted above, CSISSFRRA itself
provides that it does not restrict the dissemination of OCA information
in any manner or form except in two specified forms--the OCA portions
of an RMP and any EPA database created from those portions. RMPs,
including the sections containing OCA data, are designed to make
information contained therein easy to compile into an electronic
database, which would be capable of Internet posting. The legislative
history confirms that Congress intended to make clear that government
officials could communicate the substance of OCA information if not the
restricted forms of that information--in order to allow the type of
public dialogue that is important to chemical emergency prevention,
planning, and response. As one House member explained, CSISSFRRA was
passed to address the risk posed by Internet posting of a large OCA
database, not to prevent public officials from sharing OCA data for
individual plants with their communities. (See 145 Cong. Rec. H6083,
daily ed. July 21, 1999 (statement of Rep. Dingell)).
We share the concern voiced by one commentor that the protections
provided by CSISSFRRA would be undone if minor changes in the format of
OCA information were sufficient to allow a government official to
disseminate lawfully the OCA portions of RMPs or EPA's OCA database. We
believe CSISSFRRA's scope provision must be interpreted in a common
sense manner that achieves Congress' intent both to protect OCA
information from Internet dissemination and to allow government
officials to discuss risk. As noted above, Congress' concern with the
OCA portions of RMPs arose from the fact that they are easy to compile
into an electronic database. Minor changes in format most likely would
not change that problematic characteristic. We have thus removed the
word ``replicate'' from the relevant provision of the final rule in
order to avoid the implication that minor changes in the format of OCA
information would be sufficient to permit their release. That change is
consistent with the point made by the House member cited above who
stated that OCA information may be used ``in any other format that
avoids compilation of a national database.'' Under that view, for
example, discussion of OCA data at a public meeting would be
appropriate because it would not be a form of communication amenable to
the creation of such a database.
Several LEPCs asked us to further clarify how they may communicate
the substance of OCA information (referred to as ``OCA data elements''
in the rule). We appreciate their concerns and plan to provide
additional guidance in the future. Because it is impossible to foresee
all the ways in which government officials may wish to communicate OCA
data elements, we believe it would be most efficient and productive to
work with representatives of LEPCs, SERCs, and other relevant
government agencies in reviewing possible means of communication and
responding to inquiries about the same.
[[Page 48122]]
Many commentors expressed doubt that the enhanced local access
provisions would work as proposed. They noted that, because many LEPCs
are inactive or have limited funding, few LEPCs would be willing or
able to afford to provide secure OCA read-only access. Relatedly, a
national organization of fire department officials expressed strong
opposition to the proposed specification of fire departments as
institutions that could volunteer to provide the public with local OCA
information. One commentor suggested we authorize not only LEPCs and
fire departments but also other local government agencies involved in
chemical emergency planning, prevention, or response, such as police
and planning departments. Local governments would then have several
options for providing the public with read-only access.
We recognize that a large number of LEPCs are currently inactive,
but EPCRA survey data indicate that most heavily populated industrial
areas have active LEPCs. Those LEPCs are providing EPCRA information
(chemical inventory data and contingency plans, some of which include
possible consequences of hypothetical accidents) to the public.
Although the final rule does not require LEPCs to disseminate OCA
information, we expect that those with active EPCRA public information
programs could easily provide enhanced local access to OCA information.
Since some areas of the country do not have active LEPCs, we have
decided to expand the types of entities that are authorized to provide
read-only access, as suggested by a commentor. The final rule provides
that LEPCs and any other ``related local government agency'' may
provide the public with read-only access to OCA information for
facilities in the LEPC's jurisdiction and any other facilities with a
vulnerable zone that extends into that LEPC's jurisdiction. Related
local government agencies include fire, police, and planning
departments and any other local government agency involved in chemical
emergency planning, prevention, or response.
One commentor asked whether state agencies that take delegation of
the CAA 112(r) program would be authorized to provide read-only access.
The final rule expands the types of state entities that may provide
read-only access to OCA information. Along with SERCs, any ``related
state government agency'' (e.g., emergency management, environmental
protection, and natural resources departments involved in chemical
emergency planning, prevention, or response) would be authorized to
provide a person with access to OCA information that the LEPC in whose
jurisdiction that person lives or works could provide. Thus, a state
agency that takes responsibility for implementing the RMP program under
CAA section 112(r) may provide that access. It is also worth noting
that the final rule does not prescribe the locations where read-only
access to OCA information may be provided by LEPCs, SERCs, and state
and local government agencies. They may provide access at any facility
they choose, including municipal buildings and courthouses.
As described earlier, to further address concerns that enhanced
local access may not become a reality in every part of the country, we
have also decided to require federal reading rooms to provide any
member of the public with access to OCA information that the LEPC in
whose jurisdiction the person lives or work would be authorized to
provide. By expanding the number of state and local entities that may
provide enhanced access and the scope of access to OCA information that
federal reading rooms are required to provide, we believe the final
rule will provide reasonable access to OCA information for all members
of the public.
Several commentors recommended that the federal government provide
LEPCs and SERCs with the resources necessary to provide local access,
including a binder containing all of the OCA information that a
particular LEPC would be authorized to show the public. A commentor
also requested model procedures for operating a local OCA reading room.
Further, a few commentors suggested that, in those communities with RMP
reporting facilities that do not have LEPCs, EPA work with the local
governments to establish them. We agree that federal assistance and
guidance are warranted. As explained below, we intend to supply the
binders suggested by one of the commentors to LEPCs, SERCs, and related
local and state agencies that decide to provide enhanced local access.
Providing that and other support to local access efforts will become an
important component of the EPA's chemical accident prevention program.
Additional commentors stated that most LEPCs and SERCs would be
unable to determine whether a facility outside their jurisdiction has a
vulnerable zone that would affect their area. One commentor suggested
that the final rule should simply authorize LEPCs to distribute the OCA
for any facility within 25 miles of their local boundaries. We are not
changing our approach in today's final rule. As noted above, we intend
to provide any LEPC or related local agency willing to provide local
access with a binder that contains the OCA information it is authorized
to show the public. We will also work with SERCs and related state
agencies to provide them with a similar resource (depending on the
number of facilities in a state, binders may be too cumbersome, so
there may be a need to explore other means of providing the
information). Moreover, contiguous LEPCs can, and often do, work
together to determine which RMP facilities have vulnerable zones that
affect their areas. LEPCs, SERCs and other emergency planning
organizations have historically engaged in joint planning activities to
better prepare for emergencies. We are thus confident that the rule's
provision allowing local or state agencies to share OCA information
with adjoining jurisdictions can be implemented in a manner that would
assist LEPCs and SERCs to determine which facilities outside their
jurisdictions have vulnerable zones that extend into their
jurisdiction.
Two SERCs and one LEPC commented that the proposal to authorize
SERCs to provide individuals with OCA information on the basis of that
individual's residence or workplace was too burdensome. They questioned
whether SERCs would be able to verify the requestor's place of
residence or workplace. We understand that SERCs and related state
agencies will have to request and review proof of residence and/or
workplace. Federal reading rooms will have to do the same for any
person requesting OCA information on those bases. We believe that that
requirement is necessary, however. SERCs have much broader
jurisdictions than do the vast majority of LEPCs. Thus, the number of
facilities within their jurisdictions is typically much greater. If
SERCs were allowed to share OCA information for all the facilities in
their jurisdictions with any member of the public, the risk of persons
using SERCs to amass OCA information would be significant. To avoid
that risk, we must limit the amount of OCA information a SERC or
related state agency can share. We appreciate the extra work that that
may involve, but believe it would be manageable. A driver's license or
other identification can establish someone's home address while a pay
stub can establish a work address. As we address the same issue in
federal reading rooms, we will share our ideas and experiences with the
states.
One commentor also questioned our authority to limit the release of
OCA information to individuals on the basis of their residence or
workplace. The
[[Page 48123]]
commentor claimed that there is no statutory authority for such a
limitation. In fact, the local enhanced access provision is being
implemented under CAA section 112(r)(7)(H)(ii)(II)(bb), which
authorizes the regulation to allow public access to OCA information
``as appropriate.'' In light of the previously discussed concerns that
would arise were SERCs allowed to provide OCA information for the
entire state, we believe that it is appropriate to adopt the residence
and workplace limitation for local agency dissemination of OCA
information.
Several commentors from LEPCs and SERCs expressed great reservation
about the potential criminal liability associated with the improper
disclosure of OCA information. Some stated that, because of those
concerns, they have not requested the OCA information that they are
entitled to obtain and are authorized to show the public. The final
rule is intended to address those concerns. It makes clear that state
and local (as well as federal) officials may communicate OCA data
elements to the public in a form other than the OCA sections of RMPs
and EPA's OCA database. It also authorizes LEPCs, SERCs, and related
local and state agencies to show the OCA sections of RMPs to members of
the public in accordance with specified geographical limitations. In a
subsequent section of this preamble, we discuss what OCA information
state and local officials may share with one another. Moreover, as we
noted earlier, government officials may be held criminally liable for
unlawfully disseminating OCA information only if they ``willfully''
violate CSISSFRRA (i.e., by distributing OCA information with the
knowledge that they are doing so unlawfully).
One commentor asserted that only local persons who live or work
within the vulnerable zone of a facility should have access to local
reading rooms. Several commentors also recommended that local reading
room staff be required to implement the same security procedures that
federal reading rooms will follow--asking users for photo
identification and recording information about their access to OCA
information. That, the commentors argued, would close a loophole in the
proposal that would allow persons to obtain OCA information without
being tracked. We understand that asking local providers of OCA
information to follow security procedures would further reduce the
risks identified by the risk assessment. However, we did not propose
those security procedures at the local level because of the burden that
that would create and the effect that that burden might have on the
ability and willingness of local entities to provide OCA information
access. We also took into account the fact that the vast majority of
LEPCs have a relatively small number of RMP facilities located in, or
affecting, their jurisdiction. We thus concluded that any risk posed by
local read-only access without additional security procedures was
small. The comments we received from LEPCs, SERCs, and others confirm
our concerns about requiring local agencies to follow the type of
security procedures that federal reading rooms will follow. Indeed, the
comments indicate that local agencies will find it a challenge to
provide local access, even with the help we intend to provide. We have
thus decided not to impose any further requirements on local agencies
willing to provide read-only access to local OCA information.
We also do not agree that local access should be restricted to
local residents. First, implementing such a restriction would require
local agencies to institute much, if not all, of the security
procedures that we have decided would be too burdensome. Second,
members of the public who do not live or work in a community may
nevertheless have a legitimate interest in obtaining OCA information
for that community. For example, a requestor may have relatives who
live in the community, or may be considering purchasing a home or
working in the area.
Lastly, several commentors recommended that LEPCs be authorized to
provide take-away paper copies of local OCA information. Several others
recommended against permitting LEPCs to do so. We have concluded that
if users were permitted to obtain paper copies of OCA information LEPC-
by-LEPC, it would not be long before a large collection would be
accumulated and possibly posted on the Internet. For that reason, the
final rule retains the proposed prohibition on LEPC and SERC
dissemination of take-away paper copies of local OCA information.
F. Risk Indicator System (Vulnerable Zone Indicator System)
Many of the comments on the proposed risk indicator system were
positive, stating that the system would provide useful information that
would encourage the general public to become more active in addressing
chemical safety concerns in their communities. At the same time, those
and other commentors raised various concerns with the system and made
suggestions for improving it. A few commentors considered the system so
troublesome that they urged us to abandon it altogether.
Several commentors thought the proposed indicator system might
frighten recipients of the information and had the potential for
depressing property values. They noted that the system would
communicate information based on worst-case release scenarios that are
highly unlikely and that the information provided would necessarily be
imprecise given the nature of RMP data. Based on those concerns, some
commentors urged us not to implement the system, or to convert the
system so that it would identify the RMP facilities near a particular
address, but would not indicate whether facilities' vulnerable zones
extend to that address. Other commentors recommended that we avoid
potential misunderstandings by including in the system caveats
explaining the nature and limitations of the vulnerable zone
derivations.
We continue to believe that an indicator system can help spark the
public's awareness of chemical risks in its community and interest in
working with government and industry to reduce them. Members of the
public can already use RMP*Info to locate nearby facilities by asking
the system to search for facilities by zip code or county. We proposed
an indicator system to allow members of the public to determine if
their homes, schools, or other places of interest might be affected by
a worst-case or alternative scenario release from a facility. The
benefits assessment found that the public is more apt to use such
interpreted data, and we thus developed the indicator system as a way
of providing the public with information that communicates risk without
disseminating OCA information itself. At the same time, we agree that
it is important that users of the indicator system understand the
nature and limitations of the information thereby provided. We will
therefore design the system to include sufficient explanatory
information so that users will not become unduly alarmed if the system
reports that their address might be in a vulnerable zone. The system
will display a notice explaining that it is designed to perform the
limited function of helping users quickly determine whether the off-
site consequences of any facility's worst-case or alternative release
scenarios might affect a particular address. It will also explain the
limitations of the data used to calculate the vulnerable zones.
Relatedly, several commentors thought the proposed name, ``Risk
Indicator System,'' was inaccurate because it would not provide an
indication of ``risk,'' understood to be
[[Page 48124]]
the probability of an event multiplied by the consequences of that
event. Those commentors suggested changing the name of the indicator
system to ``Hazard Indicator System'' or ``Vulnerable Zone Indicator.''
We agree with those comments, and have decided in the final rule to
change the name of the system to ``Vulnerable Zone Indicator System''
(VZIS). That name more accurately reflects the limited purpose and
capabilities of the system.
Several commentors expressed concern that the proposed indicator
system could be used to determine distance to endpoints and thus would
provide useful targeting information. We do not agree. The indicator
system will consist of very limited query and response software located
in RMP*Info. The information provided by the system will be whether an
address might be within a vulnerable zone. There will be no indication
whether the address is at or near the outer boundary of a vulnerable
zone. Nor will the system provide the name or location of the facility
that is the origin of the vulnerable zone. Thus, no one would be able
to determine from the indicator system the distance to endpoints
reported as part of OCA information.
A number of commentors asserted that the proposed indicator system
should be deployed only if it identifies the facility that is the
origin of the vulnerable zone and/or the chemical involved in the
hypothetical release defining the zone. They were concerned that,
without that information, the system would alarm users without
providing them with the information necessary to address their
concerns. A number of other commentors recommended strongly against
identifying facilities, arguing that to do so would compromise the
security achieved by the rule's restrictions on access to OCA
information. Some commentors suggested that the indicator system
instruct users on how to obtain facility identities; one recommended
including instructions on how to contact the facility or facilities
directly.
We recognize that system users who learn that their address might
be within a vulnerable zone would likely want to learn more about the
hazards they may face. Indeed, we hope that that would be their
reaction. However, we remain concerned that the indicator system would
pose security concerns if the public could immediately obtain, on an
anonymous basis, the name of the facility and chemical involved.
Instead, we intend that the system furnish instructions on how to
obtain the names of facilities in whose vulnerable zones they live or
work.
Several commentors stated that the indicator system should not
direct recipients of the indicator system data to LEPCs or SERCs for
further information unless those agencies have agreed to provide access
to such information. We agree in part with those commentors. We believe
that chemical safety is most effectively addressed at the local level.
SERCs, LEPCs, and other state and local entities are generally in
closer contact than is EPA with local facilities and communities that
would be affected by releases. For more than a decade, EPA has
endeavored to work cooperatively with local agencies so that they can
realize their potential to help prevent and respond to accidental
releases. We therefore believe that SERCs, LEPCs, and other local
entities can and should be encouraged to assume an important role in
communicating OCA information to members of the public. While we do not
intend for the indicator system to direct users specifically to SERCs
and LEPCs, the indicator system will inform users of the several ways,
including through their SERCs and LEPCs, through which they can obtain
additional information about the facilities whose vulnerable zones
might affect an address of interest. We have thus revised the last
sentence in proposed Sec. 1400.4(a) accordingly.
While we cannot at this time name all potential sources of
information, at least facility names, locations, and vulnerable zones
will be available at all federal reading rooms and all SERCs, LEPCs, or
other state and local agencies that opt to provide local access to OCA
information. The indicator system will note specifically state and
local entities that do not seek and/or provide that information. The
system will also advise users that, once they know the name of a
facility, they can turn to RMP*Info to learn more about the facility's
chemical accident history and the steps the facility is taking to
prevent such accidents. Individuals may also contact a facility
directly to request access to OCA information. The system will also
inform users that they can obtain not only OCA information but further
information on risk through contacting a SERC, LEPC, or other state or
local ``covered person.'' Federal, state, and local government
officials are authorized and encouraged in the proposed rule to provide
reading-room access to OCA information, and are permitted to convey and
discuss the substance of OCA information, as long as they do so in a
manner that does not disseminate the OCA sections of the RMPs or EPA's
OCA database.
Several commentors also expressed concern about whether the
indicator system would be easy for local covered persons to operate.
EPA intends to provide an enhanced version of the RMP*Review software
to those federal, state, and local covered persons providing local
access so that they can easily identify the facilities whose vulnerable
zones extend to a particular address, and provide that facility
identification information to individuals who request it.
Some commentors worried that the indicator system would ``rate''
facilities for potential risk. Nothing in the proposed rule required
the indicator system to include rating information, and no such
requirement has been added to the final rule. The risk a facility poses
is a function of many factors, at least some of which are site-
specific. No computer system could adequately account for all relevant
factors. As discussed below, we intend to maintain a website of
chemical safety-related information that will assist the public in
assessing hazards posed by facilities and measures that can reduce
those hazards. In addition, RMP*Info already allows the public to learn
about facilities' prevention and response programs.
G. Internet-Accessible OCA Information
As explained in detail in the NPRM, the risk assessment segregated
the OCA information that would be helpful to terrorists or other
criminals into three categories. The first category of OCA information
provided a general account of the consequences of a chemical release in
terms of the damage that might be inflicted on the community. It was
composed of the distance to endpoint, the residential population within
the distance to endpoint, the public receptors, the environmental
receptors, and the map or graphic of the worst-case or alternative
release scenario. The second category of information consisted of OCA
information that provided a rough sketch of what is involved in
triggering a release from an RMP facility. Included in this category
were the name of the chemical involved in the worst-case or alternative
release scenario; the projected quantity of chemical released; the
release rate; the duration of the release; and the scenario that
results in the release. The third category of information consisted of
OCA information on passive and active mitigation measures.
The risk assessment concluded that Internet access to categories
one and two of OCA information posed the greatest risk of being used in
relation to an attempted industrial chemical release. However, there
were certain
[[Page 48125]]
items of OCA information within category two that posed less risk
because they were fixed values that were widely known. Thus, the
proposed rule would have posted on the Internet the OCA information in
category three and parts of the information in category two, but
withheld the remaining information in category two and all of category
one.
We solicited public comment on whether any additional items of OCA
information should be placed on the Internet or whether any items of
OCA information that we have proposed posting should not be. The
comments we received were divided. Some commentors asserted that the
risk assessment's findings in regard to the dangers of posting category
2 information should be heeded and that no category 2 information
should be placed on the Internet. Others argued that category 3
information should not be posted because the risk assessment found that
it would be helpful to terrorists (although the assessment found that
it would be much less so than would category 1 or 2 information). Still
others argued that no OCA information, especially the passive and
active mitigation system information in category 3, should be placed on
the Internet.
Other commentors maintained that the OCA information that we
proposed posting would be meaningless unless viewed in the context of
the rest of the OCA information. Several commentors similarly argued
that all OCA information should be placed on the Internet without
restriction. Still another commentor believed that at least the
chemical name should be included in the information posted.
We have considered those comments and still believe that the public
will benefit from posting the items of OCA information that we
proposed. Such information can be used for purposes of comparing
various risk reduction characteristics of RMP facilities. Further,
posting it would not create an unacceptable security risk. While some
commentors have expressed concern about the release of information
about active and passive mitigation measures, similar RMP information
has already been released on the Internet and the release of that
information was found by the risk assessment to pose the least degree
of risk. Furthermore, such information is precisely the type of
information that could be used by the public to further its dialogue
with industry.
In regard to the comments that all OCA information be placed on the
Internet, the risk assessment found that wholesale release of OCA
information in that manner would unacceptably heighten the risk of
intentional releases. Similarly, we disagree with the comment
concerning the names of chemicals. While we recognize that there would
be public benefit resulting from the posting of that information, we
find that the risk that it could be used in concert with other OCA
information for illicit purposes is too great to permit it to be
posted. As one commentor noted, while an individual item of OCA
information may not appear to pose a significant risk standing alone,
its release could raise ``mosaic'' concerns: some items of OCA
information may not raise significant security concerns considered
individually but pose greater concerns when assembled with other items
of OCA information. For example, some items of OCA information in
category 2 can be used to calculate items of information that are in
category 1. We believe that while the items of OCA information that we
proposed posting will not pose mosaic problems, others would. Thus,
only the items of OCA information that were proposed to be posted will
be placed on the Internet.
H. Access to OCA Information by Government Officials
The proposed rule called for codifying CSISSFRRA's provisions
regarding access to OCA information by state and local governmental
officials for ``official use.'' We received comments raising questions
and concerns about various aspects of the proposed codification.
One commentor criticized the proposed definition of ``official
use,'' claiming that it would exclude the use of OCA information for
purposes of enforcing the RMP rule or other legal requirements. We
disagree. The proposed definition of ``official use'' is substantively
identical to the statutory definition of that term. Consequently, to
the extent that definition limits the use of OCA information, we have
no discretion to change that result. However, we believe that the
statutory and regulatory definition of ``official use'' does permit the
use of OCA information in enforcement actions against facilities.
``Official use'' is defined as ``an action of a federal, state, or
local government agency or an entity [such as LEPC, SERC or volunteer
fire or police department] intended to carry out a function relevant to
preventing, planning for, or responding to accidental releases.''
(Final rule, Sec. 1400.2(h)). Determining compliance with, and
enforcing the terms of, the RMP rule is surely carrying out a function
relevant to preventing, planning for, or responding to accidental
releases. The same can be said about determining compliance with, and
enforcing, EPCRA and other legal requirements related to chemical
accident prevention, planning, and response.
Several commentors raised concerns about the proposed restrictions
on state and local officials' dissemination of OCA information to their
counterparts in other states. One commentor considered the restrictions
arbitrary and claimed they would interfere with useful communications
among states. Another commentor urged us to avoid hindering OCA
information sharing between fire and emergency service personnel from
jurisdictions involved in joint planning. By contrast, another
commentor recommended that the rule not allow a state or local official
access to OCA information for facilities not located in the official's
state.
Based on our review of the statute and its legislative history, we
believe that the proposed provisions for state and local official
access are legally required. CSISSFRRA itself expressly provides that
the final rule must allow for state and local officials to gain access
to OCA information for facilities not only in their own state but in
other states as well. EPA will provide state or local government
officials with OCA information for their state upon request. In
addition, to avoid unnecessarily broad dissemination of OCA information
to state and local officials, CSISSFRRA requires that those officials
specifically request information for facilities in other states, rather
than provide that the federal government unilaterally distribute it to
them. CSISSFRRA leaves no doubt, however, that the final rule must
allow a state or local official, upon request, to access OCA
information for official use for his or her state or any other states.
Morever, as the benefits assessment points out, persons interested in
evaluating the safety practices of local facilities may find it helpful
to compare OCA information for those facilities with that of similar
facilities located elsewhere. This statement is as true for government
officials as it is for members of the general public.
Similarly, CSISSFRRA itself limits the extent to which a state or
local official can share OCA information with officials of other states
or of localities in other states. It specifies that the regulations
allow such officials to share OCA information for their states with
officials of contiguous states. We do not anticipate that this
limitation will hinder useful communication among officials of
different states and localities. Since under CSISSFRRA and the final
[[Page 48126]]
rule any state or local official may request OCA information for
facilities in any state, it will not be necessary for state and local
officials to disseminate their own information. A state or local
official interested in obtaining information for a noncontiguous state
may simply request it from EPA, and an official interested in sharing
that information with another state's officials may suggest to those
officials that they request it themselves.
The commentors' general point that the rule not hinder
communications among government officials nevertheless is well taken.
We have reviewed the relevant regulatory provisions and made several
changes to improve their clarity and practicality. While the proposed
rule authorized EPA to provide a state or local official with OCA
information for ``his or her'' official use, the final rule deletes the
quoted language so that every official in a state or locality with an
official use for the information need not request it separately.
Relatedly, we have revised the regulatory language to make clear that
officials within a state or locality may share OCA information with one
another for official purposes. Consequently, an official from a county
planning department, for example, may request OCA information for
official use and distribute it to his or her colleagues who also need
to review the information ``to carry out a function relevant to
preventing, planning for, or responding to accidental releases.''
As indicated above, CSISSFRRA provides government officials with
access to OCA information for ``official use.'' One commentor suggested
that EPA ensure that the government official requesting OCA information
has an ``official `need to know.' '' We believe that approach is
unnecessary and impracticable. CSISSFRRA contains a definition of
``official use'' that describes the purposes for which such officials
may lawfully use OCA information. The final rule adopts the statutory
definition verbatim. Before providing OCA information to a government
official as required under CAA section 112(r)(7)(H)(iv) (regarding
availability of OCA information during the first year following
enactment of CSISSFRRA), we ask the official to state in writing that
access is for ``official use'' as defined by the statute. If the
official uses OCA information for other than official purposes, he or
she might be exposed to administrative, and possibly criminal,
sanctions. As an added precaution, and as required by CSISSFRRA, we
will continue to provide officials receiving OCA information with a
security notice that includes examples of what constitutes ``official
use.''
Finally, several states and LEPCs commented on the logistics of
obtaining and safeguarding OCA information. One commentor urged us not
to charge local officials for paper copies of OCA information,
particularly in light of the proposal that LEPCs and other state and
local entities be allowed to make paper copies of OCA information
available to the public in read-only form. Another commentor urged us
to provide OCA information in an ``organized'' way, that is, according
to LEPC jurisdiction. As stated above we intend to provide paper copies
of OCA information, free of charge, for facilities on the basis of LEPC
jurisdiction to LEPCs, SERCs, and others interested in providing read-
only access. For local and state officials with limited electronic
resources, we also intend to provide paper copies of OCA information
for facilities within their state.
I. Other Provisions
The proposed rule also included provisions prohibiting government
officials, as well as researchers who receive OCA information under CAA
section 112(r)(7)(h)(vii), from disseminating OCA information and ``OCA
rankings'' to the public except as authorized by the rule or a
specified provision of CSISSFRRA. The proposed rule defined ``OCA
rankings'' as ``any statewide or national ranking of identified
stationary sources derived from OCA information.'' One commentor
criticized that definition, claiming that it is vague and raises due
process issues. The commentor also was concerned that the definition
would prevent state or local officials from ranking facilities based on
parameters similar or even identical to the data reported in the OCA
sections of RMPs.
The proposed definition was drawn virtually verbatim from
CSISSFRRA, which prohibits government officials and qualified
researchers from disseminating to the public OCA information ``or any
statewide or national ranking of identified stationary sources from
such information'' (CAA section 112(r)(7)(H)(v)(I)). We believe the
statutory language, and thus the regulatory definition, are not
unconstitutionally vague, as individuals clearly can identify in
advance what constitutes a ranking of stationary sources, on a
statewide or nationwide basis, and whether the OCA information provided
to them was used to create the ranking. We do not believe the
definition prevents state or local officials from using information
other than OCA information to rank facilities. ``OCA information'' is
defined by CSISSFRRA and the rule as the OCA portions of RMPs and any
EPA database created from those portions; ``RMP'' is defined as the
risk management plan submitted to EPA pursuant to the RMP rule. If
state or local officials, without resort to OCA information, have
developed or gained access to data similar or even identical to the OCA
data reported in RMPs, they are not precluded from using that data to
rank facilities.
III. Discussion of Final Rule
After considering the comments received, we have sought to craft a
final rule that meets CSISSFRRA's requirements and reflects
consideration of both assessments' findings. CSISSFRRA's requirements
include providing any member of the public with access to paper copies
of OCA information for a ``limited number'' of facilities (CAA section
1129r)(7)(H)(ii)(II)(aa)) and other access ``as appropriate'' (CAA
section 112(r)(7)(H)(ii)(II)(bb)). The risk assessment concluded that
posting certain portions of OCA information on the Internet would
increase the risk that terrorists or other criminals will attempt to
cause an industrial chemical release in the United States. Easy access
to OCA information would assist someone seeking to identify the most
lethal potential targets from among the 15,000 facilities that have
submitted OCA information. The benefits assessment, however, concluded
that public disclosure of OCA information would likely lead to a
significant reduction in the number and severity of accidental chemical
releases. Widespread access to OCA information would serve the
functions Congress originally intended in enacting the CAA and
requiring the collection of OCA information to inform members of the
public of potential environmental hazards and to allow them to
participate in decisions that affect their lives and communities.
While chemical accidents take a significant toll on life, property,
and the environment each year, we believe that the property damage,
personal injuries, and loss of life resulting from a single, successful
terrorist attack on a chemical facility could be considerable and would
likely cause more damage than would many accidental chemical releases.
We therefore have attempted to balance those concerns by making as much
OCA information as appropriate available online, but not posting the
information that the risk assessment found would, if disseminated
without restriction, pose a significant risk for terrorist or criminal
purposes. Although
[[Page 48127]]
the Internet provides a tremendous benefit by offering people easy
access to a wealth of information, we also recognize that it provides a
new means for criminals and terrorists to carry out traditional
criminal activities. The final rule provides several means for
individuals to obtain OCA information not only for facilities within
their community but also for a sufficient number of facilities located
elsewhere, thereby enabling individuals to compare facilities' safety
and prevention measures and records. Those means are described below.
Both the proposed and final rules have been approved by the
Director of OMB.
A. Access to Paper Copies of OCA Information
The final rule creates federal reading rooms to fulfill CSISSFRRA's
requirement to provide individuals with access to paper copies of OCA
information of a limited number of facilities. A minimum of 50 federal
reading rooms will be geographically located across the United States,
with approximately one federal reading room per state. The number and
location of those reading rooms may be adjusted based upon public
demand and the agencies' experience in administering them.
Under the rule, any person shall be provided with access to a paper
copy of the OCA information for up to 10 stationary sources per
calendar month located anywhere in the country, without geographical
restriction. In addition, the final rule directs federal reading rooms
to provide access to paper copies of OCA information for facilities
located within the LEPC jurisdictions where the individual lives or
works and for any additional facilities that have vulnerable zones that
extend into those LEPC jurisdictions. Individuals will be allowed to
read and take handwritten notes from, but not remove or mechanically
reproduce, the paper copy of OCA information.
Reading room personnel will be required to ascertain a requestor's
identity by viewing a photographic identification for an individual
issued by a government agency and obtain a signature on a sign-in sheet
and a certification before providing that person with access to OCA
information for up to 10 facilities per month without geographical
restriction. Similarly, reading room personnel will be required to view
documentation of where an individual lives or works and obtain a
signature on a sign-in sheet before providing any person with access to
the OCA information that the LEPC in whose jurisdiction lives or works
would be authorized to provide. Reading rooms will also be required to
keep records to ensure that no individual receives OCA information
beyond the limits established by the rule.
B. Enhanced Access to Local OCA Information
Several provisions of the final rule are designed to enhance the
public's access to OCA information for local stationary sources. In
response to comments regarding the appropriate governmental agencies to
provide enhanced access, EPA and DOJ have modified the final rule to
permit related local government agencies and related state government
agencies, as defined in the regulation, to provide access. The rule
authorizes and encourages LEPCs and related local government agencies
to provide read-only access to OCA information for sources located
within an LEPC's jurisdiction and for any other stationary sources that
have a vulnerability zone extending into that jurisdiction. Likewise,
SERCs and related state government agencies are authorized and
encouraged to provide read-only access to the same OCA information that
the LEPC in whose jurisdiction the person lives or works would be
authorized to provide. Federal reading rooms are similarly authorized
to provide read-only access to OCA information. Such information will
not be subject to the 10 facility per month limit.
The final rule also codifies the statutory provisions of CSISSFRRA
that allow any member of an LEPC or SERC or any other state or local
government official to convey to the public any OCA data elements
orally or in writing, provided that the data elements are not conveyed
in the format of sections 2 through 5 of an RMP or any electronic
database that EPA has developed that includes OCA data elements.
C. Vulnerable Zone Indicator System
The final rule establishes a ``vulnerable zone indicator system''
(VZIS) which provides persons located in any state with a means of
obtaining, via electronic mail or other inquiry, information regarding
the risk expressed by OCA information without providing Internet access
to the OCA information itself. Members of the public will be able to
learn whether a specific address (such as that of a home, school, or
place of employment) falls within a reported ``vulnerable zone'' (i.e.,
within any RMP facility's worst-case or alternative release scenario's
``distance to endpoint''). Electronic mail inquiries will usually
receive a response within two working days. Members of the public who
do not have access to the Internet will be able to obtain the same
information by calling an EPA toll-free number or by sending regular
mail to the Administrator of EPA. VZIS will consist primarily of query
and response software located in RMP*Info.
VZIS will also provide individuals with information on how to
identify the specific facilities affecting the address submitted to
VZIS. It will also provide contact information and sources of
additional information explaining chemical accident risk. Any federal
reading room or local reading room providing enhanced access under this
rule, for example, may be a source for identifying the facility or
facilities whose vulnerable zones extend to the address entered into
the indicator system, as well as the location of the facilities. System
users will be provided with the addresses and telephone numbers of the
federal reading rooms. The system will also supply users with up-to-
date contact information for the SERCs and LEPCs, and note that only
some LEPCs provide local OCA information access services. The indicator
system will advise users that, once they know the name of the facility,
they can use RMP*Info to learn more about the facility's chemical
accident history and its accident prevention measures, and they may
contact the facility directly to gain access to OCA information.
D. Internet Access to Selected OCA Information
The final rule makes some items of OCA information available to the
public through the Internet by posting it on EPA's website. Those
provisions of the final rule are identical to those in the proposed
rule. The items of information that will be posted on the Internet are
those that the risk assessment found would pose the least serious
security risk if posted on the Internet. The following items of OCA
information will be posted on the Internet, along with other RMP data
elements available in EPA's RMP*Info:
The concentration of the chemical (RMP Sections 2.1.b;
3.1.b);
The physical state of the chemical (RMP Sections 2.2;
3.2);
The duration of the chemical release for the worst-case
scenario (RMP Section 2.7);
The statistical model used (RMP Sections 2.3; 3.3; 4.2;
5.2);
The endpoint used for flammables for the worst-case
scenario (RMP Section 4.5);
The wind speed during the chemical release (RMP Sections
2.8; 3.8);
[[Page 48128]]
The atmospheric stability (RMP Sections 2.9; 3.9);
The topography of the surrounding area (RMP Sections 2.10;
3.10);
The passive mitigation systems considered (RMP Sections
2.15; 3.15; 4.10; 5.10); and
The active mitigation systems considered (RMP Sections
3.16; 5.11).
The final rule precludes the following items of OCA information
from being posted on the EPA website based upon the risk assessment's
findings that their release on the Internet would pose significant
security concerns:
The name of the chemical involved (RMP Sections 2.1.a;
3.1.a; 4.1; 5.1);
The scenario involved (RMP Sections 2.4; 3.4; 4.3; 5.3);
The quantity of chemical released (RMP Sections 2.5; 3.5;
4.4; 5.4);
The release rate of the chemical involved for the worst-
case scenario (RMP Section 2.6);
The release rate of the chemical involved in the
alternative release scenario (RMP Section 3.6);
The duration of the chemical release in the alternative
release scenario (RMP Section 3.7);
The distance to endpoint (RMP Sections 2.11; 3.11; 4.6;
5.6);
The endpoint used for flammables for the alternative
release scenario (RMP Section 5.5);
The residential population within the distance to endpoint
(RMP Sections 2.12; 3.12; 4.7; 5.7);
The public receptors within the distance to endpoint (RMP
Sections 2.13; 3.13; 4.8; 5.8);
The environmental receptors within the distance to
endpoint (RMP Sections 2.14; 3.14; 4.9; 5.9); and
Any map or other graphic used to illustrate a scenario
(RMP Sections 2.16; 3.17; 4.11; 5.12).
E. Additional Information on Chemical Accident Risk
As a supplement to the provisions of this rule, EPA will make
available to the public additional information on chemical accident
risk through an Internet website. Some of that information is currently
available through EPA's website. RMPs (except for the OCA information,
sections 2 through 5) are currently accessible to the public through
RMP*Info. Through Envirofacts, the public can easily access other
information about facilities that have submitted RMPs. EPA's website
also has links to a web-based chemical guide (http://
chemicalguide.com). Another helpful link found on the EPA website that
provides valuable information to the public is the NSC website (http://
www.nsc.org/xroads.cfm), which is aimed at the news media and provides
suggestions for information to request of facility management and local
officials, for approaches to sifting through the information, and for
presenting the information in a way that helps communities interpret
local RMPs.
EPA is also developing new sources of information through which the
public can learn about chemical accident risk. Research on accident
histories based on the data provided in RMPs and other sources, both
national and international in nature, will be posted on the EPA
website. Moreover, EPA will expand the number of links to environmental
organizations, industry trade groups, and academic institutions to
provide the public with a comprehensive means of finding chemical risk
and safety information. EPA will also provide guidance that it, along
with other organizations, has developed to assist community members and
interested groups to work with facility management and local officials
to better understand and manage the risks posed by the storage of large
quantities of toxic or flammable chemicals. EPA is developing examples
of facilities and industries that can serve as models for ``best
practices'' in chemical accident risk prevention and successful
practices in RMP implementation. EPA and other organizations are
developing background information about the nature of chemical accident
risk, and that information will be posted on EPA's website when it
becomes available. In addition, through a cooperative agreement, EPA
and Clean Air Action (a non-profit organization) will develop a primer
for lay persons on basic risk management terms and principles that help
to provide a basis for understanding chemical accident risks. EPA will
be making available an updated list of LEPC, SERC, and other emergency
response contacts.
That information is intended to give the public a better
understanding of the general nature of the risks associated with
potential accidental releases posed by hazardous chemicals. In
combination with OCA data about specific facilities, that information,
we expect, will better enable the public to engage in productive
dialogues at the local, state, and federal levels to prevent chemical
accidents and to minimize the consequences of accidents that occur. EPA
will provide that information through its Internet website, http://
www.epa.gov/ceppo. Much of that information is already available there.
EPA will continue to supplement that information as necessary or
appropriate to provide the public with a full understanding of chemical
accident risk and prevention.
F. Access to OCA Information by Government Officials and Other
Provisions
The final rule adopts the proposed provisions for access to OCA
information by federal, state, and local government officials, as well
as qualified researchers. In accordance with CAA section
112(r)(7)(H)(ii)(II)(cc)-(ee), the final rule provides state or local
government officials with access, for official use, to OCA information
for facilities located in their states, and, at the officials' request,
for facilities located in other states. Also in keeping with that
section, the final rule allows state or local government officials to
share for official use OCA information for facilities within their
state with one another and with state or local government officials in
contiguous states. Similarly, the final rule allows federal government
officials to share OCA information with each other for official use.
The final rule also establishes the other necessary provisions of
the distribution system for OCA information. Specifically, it prohibits
the dissemination of OCA information by government officials and
qualified researchers (researchers who receive OCA information under
CAA section 112(r)(7)(H)(vii)) to the public and to state and local
officials except as authorized by the rule and a related CAA provision.
It also authorizes the Administrator to disseminate OCA information as
required by two other CAA provisions concerning qualified researchers
and a read-only information technology system (CAA section
112(r)(7)(H)(viii)).
G. Effective Date and Implementation Schedule
The final rule is effective immediately so that we may continue to
make OCA information available to government officials (``covered
persons'') without interruption. CSISSFRRA and its legislative history
make clear that Congress intended government officials to have ongoing
access to OCA information to help them perform their jobs, as related
to chemical emergency planning, prevention, and response. CAA section
112(r)(7)(H)(iv) requires EPA to make OCA information available to
government officials during the ``transition period,'' the year
following the enactment of CSISSFRRA when the assessments and the
rulemaking must be conducted. (see 145 Cong. Rec. S7545, daily ed. June
23, 1999 (statement of Sen. Chafee)). However, that authority
[[Page 48129]]
ends on the earlier of the date of promulgation of the regulations or
the one-year anniversary of the enactment of CSISSFRRA, August 5, 2000.
In order to avoid a gap in government officials' access to OCA
information, we believe that there is good cause to make the final rule
effective immediately, pursuant to 5 U.S.C. 553(d)(3).
We will need time to implement and coordinate the operation of the
federal reading room system. We believe we can complete that process
within three months and begin opening reading rooms soon thereafter. We
anticipate that federal reading room access will be available by
December 31, 2000. To provide public access to OCA information as soon
as possible, we will not wait for every reading room to be operational
before opening any reading room. We will begin operating reading rooms
as they become available, and will post on EPA's website the locations
of reading rooms as they open.
The vulnerable zone indicator system will begin operation no later
than October 5, 2000. That will permit us to develop, test, and deploy
the software systems necessary for the implementation of VZIS. Further,
the OCA information to be disseminated on the EPA website will be
posted by December 31, 2000.
IV. Administrative Requirements
A. Docket
The docket is an organized and complete file of all the information
that we considered in the development of this rule. The docket is a
dynamic file because it allows members of the public and industry
readily to identify and locate documents so that they can effectively
participate in the rulemaking process. Along with the proposed and
promulgated rules and their preambles, the contents of the docket serve
as the record for purposes of judicial review. See CAA section
307(d)(7)(A), 42 U.S.C. 7607(d)(7)(A).
The official record for this rulemaking has been established under
Docket No. A-2000-20 (including comments and data submitted
electronically). A public version of this record, including printed,
paper versions of electronic comments, which does not include any
information claimed as Confidential Business Information, is available
for inspection from 8:00 a.m. to 5:30 p.m., Monday through Friday,
excluding legal holidays. The official rulemaking record is located at
the address specified in the ADDRESSES section at the beginning of this
document.
B. Executive Order 12866
OMB has determined that this rule is a ``significant regulatory
action'' under Executive Order 12866, section 3(f), ``Regulatory
Planning and Review'' (58 FR 51735, October 4, 1993). OMB also has
determined that this rule would not be economically significant because
it would have an annual effect on the economy of less than $100 million
and would not affect in a material way the economy, a sector of the
economy, productivity, competition, jobs, the environment, public
health or safety, or state, local, or tribal governments or
communities. Under the terms of Executive Order 12866, OMB has reviewed
the rule.
C. Executive Order 12988
This rule meets the applicable standards set forth in sections 3(a)
and 3(b)(2) of Executive Order 12988, ``Civil Justice Reform'' (61 FR
4729, February 5, 1996).
D. Executive Order 13045
Executive Order 13045, ``Protection of Children from Environmental
Health Risks and Safety Risks'' (62 FR 19885, April 23, 1997), does not
apply to this rule because it is not economically significant under
Executive Order 12866.
E. Executive Order 13084
Under Executive Order 13084, ``Consultation and Coordination with
Indian Tribal Governments,'' section 3, Consultation (63 FR 27655, May
19, 1998), federal agencies may not promulgate a regulation that is not
required by statute, that significantly or uniquely affects the
communities of Indian tribal governments, and that imposes substantial
direct compliance costs on those communities, unless the federal
government provides the funds necessary to pay the direct compliance
costs incurred by the tribal governments, or the regulating agencies
consult with those governments before formal promulgation of the rule.
This rule does not significantly or uniquely affect the communities of
Indian tribal governments or impose substantial direct compliance costs
on those communities. Accordingly, the requirements of section 3(b) of
Executive Order 13084 do not apply to this rule. Nonetheless, we
consulted two tribal organizations that represent tribal environmental
officials (Tribal Association on Solid Waste & Emergency Response, and
National Tribal Environmental Council) and neither expressed any
concerns with the provisions of this rule.
F. Executive Order 13132
Executive Order 13132, ``Federalism'' (64 FR 43255, August 10,
1999), requires federal agencies to develop an accountable process to
ensure ``meaningful and timely input by state and local officials in
the development of regulatory policies that have federalism
implications.'' ``Policies that have federalism implications'' is
defined in the Executive Order to include regulations that have
``substantial direct effects on the states, on the relationship between
the national government and the states, or on the distribution of power
and responsibilities among the various levels of government.''
Under section 6 of Executive Order 13132, a federal agency may not
issue a regulation that has federalism implications, that imposes
substantial direct compliance costs, and that is not required by
statute, unless the federal government provides the funds necessary to
pay the direct compliance costs incurred by state and local
governments, or the agency issuing the regulation consults with state
and local officials early in the process of developing the proposed
regulation. A federal agency also may not issue a regulation that has
federalism implications and that preempts state law unless the agency
consults with state and local officials early in the process of
developing the proposed regulation.
This final rule does not have federalism implications. It will not
have substantial direct effect on the states, on the relationship
between the national government and the states, or on the distribution
of power and responsibilities among the various levels of government,
as specified in Executive Order 13132. The statute itself--CSISSFRRA--
currently restricts the dissemination of OCA information by state and
local officials and supersedes inconsistent provisions of state or
local law. This rule only slightly narrows those statutory
restrictions, allowing certain state and local entities to provide the
public with read-only access to OCA information for local facilities.
Nevertheless, we have consulted with seven organizations that represent
state and local elected officials in developing this rule (i.e.,
National Governors Association, National Conference of State
Legislatures, U.S. Conference of Mayors, National League of Cities,
Council on State Governments, International City/County Management
Association, National Association of Counties, and National Association
of Towns and Townships). We have also consulted with state and local
[[Page 48130]]
representatives of the Accident Prevention Subcommittee of the CAA
Advisory Committee (under the Federal Advisory Committee Act (FACA))
about the implementation of the OCA provisions of CSISSFRRA. In
response to concerns some have raised about the potential chilling
effect of CSISSFRRA's restrictions on state and local officials'
willingness to obtain OCA information and to communicate the substance
of that information to the public, this rule includes a provision
clarifying that state and local officials can share OCA data with the
public as long as they do so in a way that does not disseminate or
permit mechanical replication of the OCA sections of RMPs or provide
access to EPA's OCA database. As noted above, this rule also authorizes
some state and local officials to share OCA information itself in
certain ways.
Section 4 of the Executive Order contains additional requirements
for rules that preempt state or local law, even if those rules do not
have federalism implications (i.e., the rules will not have substantial
direct effects on the states, on the relationship between the national
government and the states, or on the distribution of power and
responsibilities among the various levels of government). Those
requirements include providing all affected state and local officials
notice and an opportunity for appropriate participation in the
development of the regulation. If the preemption is not based on
express or implied statutory authority, EPA also must consult, to the
extent practicable, with appropriate state and local officials
regarding the conflict between state law and federally protected
interests within the agency's area of regulatory responsibility.
Consequently, we consulted to the extent practicable with the seven
organizations mentioned above. Other than requesting further
clarification on the proposed rule, none of those organizations raised
federalism concerns with the rule's approach.
G. Regulatory Flexibility Act
Under the Regulatory Flexibility Act of 1980 (5 U.S.C. 601 et
seq.), as amended by the Small Business Regulatory Enforcement Fairness
Act of 1996, agencies are required to give special consideration to the
effect of federal regulations on small entities and to consider
regulatory options that might mitigate any such effect. However, an
agency need not prepare a regulatory flexibility analysis if the rule
would not have a significant economic impact on a substantial number of
small entities. Small entities include small businesses, small not-for-
profit enterprises, and small government jurisdictions.
In accordance with 5 U.S.C. 605(b), we certify that this rule does
not have a significant economic impact on a substantial number of small
entities. Although the rule authorizes small government jurisdictions
to provide read-only access to OCA information, it does not require
those jurisdictions to provide that access. This rule contains a
prohibition on local government officials (and other government
officials) disclosing OCA information to the public except in
authorized ways, but that prohibition already existed under CAA section
112(r)(7)(H)(v). Moreover, we do not expect that any burden resulting
indirectly from the provisions of this rule will have a significant
economic impact on the operations of local governments.
H. Paperwork Reduction Act
The information collection requirements in this rule have been
submitted for approval to OMB under the Paperwork Reduction Act, 44
U.S.C. 3501 et seq. An Information Collection Request (ICR) document
has been prepared by EPA (ICR No. 1981.01) and a copy may be obtained
from Sandy Farmer by mail at Collection Strategies Division, U.S.
Environmental Protection Agency (2822), 1200 Pennsylvania Ave., N.W.,
Washington, DC 20460; by e-mail at farmer.sandy@epamail.epa.gov; or by
calling (202) 260-2740. A copy may also be downloaded off the Internet
at http://www.epa.gov/icr. The ICR for the proposed rule was listed as
ICR No. 1656.08. To avoid confusion with the ICR for the full RMP
Program (i.e., RMP Program Requirements and Petitions to Modify the
List of Regulated Substances under section 112(r) of the CAA), the ICR
has been changed for the final rule. The information requirements are
not enforceable until OMB approves them.
This rule will impose minimal information collection requirements
but will require record keeping. The respondent universe for this rule
is state and local officials and members of the public.
None of the respondent activities for state and local agencies are
mandatory and all depend on the state or local agency deciding to
obtain OCA information and/or communicating the substance of the
information or the information itself to the public. The respondent
activities for those agencies include reading and understanding the
Security Notice to federal, state, and local officials and researchers;
requesting OCA information and certifying that they are covered
persons; providing secure storage for the CD Rom or paper copies when
not in use; learning how to use the database and software, if needed,
to produce a copy of OCA information; providing a location for the
public to review OCA information for local facilities; ensuring that
members of the public do not remove or mechanically copy OCA
information they review; and making OCA data available in formats other
than the RMP format.
The number of respondents undertaking one or more of these
activities is estimated to be at least one agency in each state,
territory, and the District of Columbia. These agencies are assumed to
be the SERCs and may be environmental protection agencies, emergency
management agencies, or both. Based on a recent survey, EPA estimates
that there are 1,500 active LEPCs (in compliance with EPCRA). These
agencies may request OCA information from EPA for their own use for
emergency planning. Out of these, we estimate that only 1,000 LEPCs
will be providing local access by the third year covered by this ICR.
EPA estimates the total burden hours for state and local agencies to be
86,000 hours annually (258,000 hours for three years) at a cost of
$2,400,000 annually ($7,200,000 for three years).
For members of the public, the respondent activity includes calling
for an appointment, displaying photographic identification, and signing
a sign-in sheet and a certification form at a federal reading room. If
an individual would like to obtain information on local facilities, he
or she would need to provide documentation demonstrating his or her
place of residence or employment. In addition, members of the public
are assumed to use the VZIS system and to make follow-up calls to
obtain additional information. It is assumed that approximately 20,000
people will use the VZIS system each year and that 5,000 of those will
seek additional information. Those individuals without access to the
Internet will be able to call an EPA toll-free number or send the
request by mail. The total burden hours for the public are estimated to
be 14,000 hours annually (42,000 hours for three years) at a cost of
$293,000 annually ($879,000 for three years).
Burden means the total time, effort, or financial resources
expended by persons to generate, maintain, retain, disclose, or provide
information to or for a federal agency. That includes the time needed
to review instructions to develop, acquire, install, and utilize
technology and systems for the purposes of collecting, validating,
verifying, processing, maintaining, disclosing, and
[[Page 48131]]
providing information; to adjust existing ways to comply with any
previously applicable instructions and requirements; to train
personnel; to search data sources; to complete and review the
collection of information; and to transmit or otherwise disclose the
information.
An agency may not conduct or sponsor, and a person is not required
to respond to, a collection of information unless the collection
displays a currently valid OMB control number. The OMB control numbers
for EPA's regulations are listed in 40 CFR Part 9 and 48 CFR Chapter
15. The OMB control numbers for the information collection requirements
in this rule will be listed in an amendment to 40 CFR part 9 in a
subsequent Federal Register document after OMB approves the ICR.
I. Unfunded Mandates Reform Act of 1995
This rule will not result in the expenditure by state, local, and
tribal governments, in the aggregate, or by the private sector, of $100
million or more in any one year, and it contains no requirements that
might significantly or uniquely affect small governments. This rule
requires small governments that wish to obtain OCA information to
request it, and once they obtain it, they will be prohibited from
disseminating it except in accordance with the rule. We do not expect
that those provisions will impose a significant burden. Moreover,
certain members of small governments would be authorized, but not
required, to provide public access to OCA information in a manner that
is less burdensome than would be required of federal covered persons.
Therefore, no actions are deemed necessary under the Unfunded Mandates
Reform Act of 1995.
J. Small Business Regulatory Enforcement Fairness Act of 1996
This rule is not a major rule as defined by section 251 of the
Small Business Regulatory Enforcement Fairness Act of 1996, 5 U.S.C.
804. This rule will not result in an annual effect on the economy of
$100 million or more; a major increase in costs or prices; or
significant adverse effects on competition, employment, investment,
productivity, innovation, or on the ability of United States-based
companies to compete with foreign-based companies in domestic and
export markets.
K. Congressional Review Act
The Congressional Review Act, 5 U.S.C. 801 et seq., as added by the
Small Business Regulatory Enforcement Fairness Act of 1996, generally
provides that before a rule may take effect, the agency promulgating
the rule must submit a rule report, which includes a copy of the rule,
to each House of the Congress and to the Comptroller General of the
United States. EPA will submit a report containing this rule and other
required information to the U.S. Senate, the U.S. House of
Representatives, and the Comptroller General of the United States prior
to publication of the rule in the Federal Register. A major rule cannot
take effect until 60 days after it is published in the Federal
Register. This rule is not a ``major rule'' as defined by 5 U.S.C.
Sec. 804(2). This rule will be effective August 4, 2000.
V. Judicial Review
Under section 307(b)(1) of the CAA, judicial review of this final
rule is available only by filing a petition for review in the U.S.
Court of Appeals for the District of Columbia Circuit within 60 days of
publication of this rule. Under section 307(b)(2) of the CAA, the
requirements established by the final rule may not be challenged later
in civil or criminal proceedings brought to enforce these requirements.
This rule has been promulgated pursuant to CAA section 307(d).
List of Subjects in 40 CFR Part 1400
Environmental protection, Chemicals, Chemical accident prevention.
Dated: July 31, 2000.
Janet Reno,
Attorney General.
Dated: July 31, 2000.
Carol M. Browner,
Administrator.
For the reasons set forth in the preamble, EPA and DOJ establish
chapter IV of title 40 of the Code of Federal Regulations, consisting
of subchapter A, part 1400, as follows:
CHAPTER IV--ENVIRONMENTAL PROTECTION AGENCY AND DEPARTMENT OF JUSTICE
SUBCHAPTER A--ACCIDENTAL RELEASE PREVENTION REQUIREMENTS; RISK
MANAGEMENT PROGRAMS UNDER THE CLEAN AIR ACT SECTION 112(r)(7);
DISTRIBUTION OF OFF-SITE CONSEQUENCE ANALYSIS INFORMATION
PART 1400--DISTRIBUTION OF OFF-SITE CONSEQUENCE ANALYSIS
INFORMATION
Subpart A--General
Sec.
1400.1 Purpose.
1400.2 Definitions.
Subpart B--Public Access
1400.3 Public access to paper copies of off-site consequence
analysis information.
1400.4 Vulnerable zone indicator system.
1400.5 Internet access to certain off-site consequence analysis
data elements.
1400.6 Enhanced local access.
Subpart C--Access to Off-Site Consequence Analysis Information by
Government Officials
1400.7 In general.
1400.8 Access to off-site consequence analysis information by
federal government officials.
1400.9 Access to off-site consequence analysis information by
state and local government officials.
Subpart D--Other Provisions
1400.10 Limitation on public dissemination.
1400.11 Limitation on dissemination to state and local government
officials.
1400.12 Qualified researchers.
1400.13 Read-only database.
Authority: 42 U.S.C. 7412(r)(7)(H)(ii).
Subpart A--General
Sec. 1400.1 Purpose.
Stationary sources subject to the Chemical Accident Prevention
Provisions of 40 CFR part 68 are required to analyze the potential harm
to public health and welfare of hypothetical chemical accidents and
submit the results of their analyses to the U.S. Environmental
Protection Agency as part of risk management plans. This part governs
access by the public and by government officials to the portions of
risk management plans containing the results of those analyses and
certain related materials. This part also restricts dissemination of
that information by government officials.
Sec. 1400.2 Definitions.
For the purposes of this part:
(a) Accidental release means an unanticipated emission of a
regulated substance or other extremely hazardous substance into the
ambient air from a stationary source.
(b) Administrator means the Administrator of the U.S. Environmental
Protection Agency or his or her designated representative.
(c) Attorney General means the Attorney General of the United
States or his or her designated representative.
(d) Federal government official means--
(1) An officer or employee of the United States; and
(2) An officer or employee of an agent or contractor of the federal
government.
(e) State or local government official means--
(1) An officer or employee of a state or local government;
[[Page 48132]]
(2) An officer or employee of an agent or contractor of a state or
local government;
(3) An individual affiliated with an entity that has been given, by
a state or local government, responsibility for preventing, planning
for, or responding to accidental releases, such as a member of a Local
Emergency Planning Committee (LEPC) or a State Emergency Response
Commission (SERC), or a paid or volunteer member of a fire or police
department; or
(4) An officer or employee or an agent or contractor of an entity
described in paragraph (e)(3) of this section.
(f) LEPC means a Local Emergency Planning Committee created under
the Emergency Planning and Community Right-to-Know Act, 42 U.S.C. 11001
et seq.
(g) Member of the public or person means an individual.
(h) Official use means an action of a federal, state, or local
government agency or an entity described in paragraph (e)(3) of this
section intended to carry out a function relevant to preventing,
planning for, or responding to accidental releases.
(i) Off-site consequence analysis (OCA) information means sections
2 through 5 of a risk management plan (consisting of an evaluation of
one or more worst-case release scenarios or alternative release
scenarios) for an identified facility and any electronic database
created by the Administrator from those sections.
(j) Off-site consequence analysis (OCA) data elements means the
results of the off-site consequence analysis conducted by a stationary
source pursuant to 40 CFR part 68, subpart B, when presented in a
format different than sections 2 through 5 of a risk management plan or
any Administrator-created electronic database.
(k) Off-site consequence analysis (OCA) rankings means any
statewide or national rankings of identified stationary sources derived
from OCA information.
(l) Qualified researcher means a researcher who receives OCA
information pursuant to 42 U.S.C. 7412(r)(7)(H)(vii).
(m) Related local government agencies means local government
agencies, such as police, fire, emergency management, and planning
departments, that are involved in chemical emergency planning,
prevention, or response.
(n) Related state government agencies means state government
agencies, such as emergency management, environmental protection,
health, and natural resources departments, that are involved in
chemical emergency planning, prevention, or response.
(o) Risk management plan (RMP) means a risk management plan
submitted to the Administrator by an owner or operator of a stationary
source pursuant to 40 CFR part 68, subpart G.
(p) SERC means a State Emergency Response Commission created under
the Emergency Planning and Community Right-to-Know Act, 42 U.S.C. 11001
et seq.
(q) State has the same meaning as provided in 42 U.S.C. 7602(d) (a
state, the District of Columbia, the Commonwealth of Puerto Rico, the
Virgin Islands, Guam, American Samoa, and the Commonwealth of the
Northern Mariana Islands).
(r) Stationary source has the same meaning as provided in 40 CFR
part 68 subpart A, Sec. 68.3.
(s) Vulnerable zone means the geographical area that could be
affected by a worst-case or alternative scenario release from a
stationary source, as indicated by the off-site consequence analysis
reported by the stationary source in its risk management plan pursuant
to the applicable requirements of 40 CFR Part 68. It is defined as a
circle, the center of which is the stationary source and the radius of
which is the ``distance-to-endpoint,'' or the distance a toxic or
flammable cloud, overpressure, or radiant heat would travel after being
released and before dissipating to the point that it no longer
threatens serious short-term harm to people or the environment.
Subpart B--Public Access
Sec. 1400.3 Public access to paper copies of off-site consequence
analysis information.
(a) General. The Administrator and the Attorney General shall
ensure that any member of the public has access to a paper copy of OCA
information in the manner prescribed by this section.
(b) Reading-room access. Paper copies of OCA information shall be
available in at least 50 reading rooms geographically distributed
across the United States and its territories. The reading rooms shall
allow any person to read, but not remove or mechanically reproduce, a
paper copy of OCA information, in accordance with paragraphs (c)
through (g) of this section and procedures established by the
Administrator and Attorney General.
(c) Limited number. Any person shall be provided with access to a
paper copy of the OCA information for up to 10 stationary sources
located anywhere in the country, without geographical restriction, in a
calendar month.
(d) Additional access. Any person also shall be provided with
access to a paper copy of the OCA information for stationary sources
located in the jurisdiction of the LEPC where the person lives or works
and for any other stationary source that has a vulnerable zone that
extends into that LEPC's jurisdiction.
(e) Personal identification for access to OCA information without
geographical restriction. Reading rooms established under this section
shall provide a person with access to a paper copy of OCA information
under paragraph (c) of this section only after a reading room
representative has
(1) Ascertained the person's identity by viewing photo
identification issued by a federal, state, or local government agency
to the person; and
(2) Obtained the person's signature on a sign-in sheet and a
certification that the person has not received access to OCA
information for more than 10 stationary sources for that calendar
month.
(f) Personal identification for access to local OCA information.
Reading rooms established under this section shall provide a person
with access to a paper copy of OCA information under paragraph (d) of
this section only after a reading room representative has
(1) Ascertained where the person lives or works by viewing
appropriate documentation; and
(2) Obtained the person's signature on a sign-in sheet.
(g) Record keeping. Reading room personnel shall keep records of
reading room use and certifications in accordance with procedures
established by the Administrator and the Attorney General. These
records shall be retained for no more than three years. Federal reading
rooms will not index or otherwise manipulate the sign-in sheets
according to individuals' names, except in accordance with the Privacy
Act.
Sec. 1400.4 Vulnerable zone indicator system.
(a) In general. The Administrator shall provide access to a
computer-based indicator that shall inform any person located in any
state whether an address specified by that person might be within the
vulnerable zone of one or more stationary sources, according to the
data reported in RMPs. The indicator also shall provide information
about how to obtain further information.
(b) Methods of access. The indicator shall be available on the
Internet or by request made by telephone or by mail to the
Administrator to operate the indicator for an address specified by the
requestor. SERCs, LEPCs, and other related state or local government
agencies are authorized and encouraged to operate the indicator as
well.
[[Page 48133]]
Sec. 1400.5 Internet access to certain off-site consequence analysis
data elements.
The Administrator shall include only the following OCA data
elements in the risk management plan database available on the
Internet:
(a) The concentration of the chemical (RMP Sections 2.1.b; 3.1.b);
(b) The physical state of the chemical (RMP Sections 2.2; 3.2);
(c) The statistical model used (RMP Sections 2.3; 3.3; 4.2; 5.2);
(d) The endpoint used for flammables in the worst-case scenario
(RMP Section 4.5);
(e) The duration of the chemical release for the worst-case
scenario (RMP Section 2.7);
(f) The wind speed during the chemical release (RMP Sections 2.8;
3.8);
(g) The atmospheric stability (RMP Sections 2.9; 3.9);
(h) The topography of the surrounding area (RMP Sections 2.10;
3.10);
(i) The passive mitigation systems considered (RMP Sections 2.15;
3.15; 4.10; 5.10); and
(j) The active mitigation systems considered (RMP Sections 3.16;
5.11).
Sec. 1400.6 Enhanced local access.
(a) OCA data elements. Consistent with 42 U.S.C.
7412(r)(7)(H)(xii)(II), members of LEPCs and SERCs, and any other state
or local government official, may convey to the public OCA data
elements orally or in writing, as long as the data elements are not
conveyed in the format of sections 2 through 5 of an RMP or any
electronic database developed by the Administrator from those sections.
Disseminating OCA data elements to the public in a manner consistent
with this provision does not violate 42 U.S.C. 7412(r)(7)(H)(v) and is
not punishable under federal law.
(b) OCA information. (1) LEPCs and related local government
agencies are authorized and encouraged to allow any member of the
public to read, but not remove or mechanically copy, a paper copy of
the OCA sections of RMPs (i.e., sections 2 through 5) for stationary
sources located within the jurisdiction of the LEPC and for any other
stationary source that has a vulnerable zone that extends into that
jurisdiction.
(2) LEPCs and related local government agencies that provide read-
only access to the OCA sections of RMPs under this paragraph (b) are
not required to limit the number of stationary sources for which a
person can gain access, ascertain a person's identity or place of
residence or work, or keep records of public access provided.
(3) SERCs and related state government agencies are authorized and
encouraged to allow any person to read, but not remove or mechanically
copy, a paper copy of the OCA sections of RMPs for the same stationary
sources that the LEPC in whose jurisdiction the person lives or works
would be authorized to make available to that person under paragraph
(b)(1) of this section.
(4) Any LEPC, SERC, or related local or state government agency
that allows a person to read the OCA sections of RMPs in a manner
consistent with this paragraph (b) shall not be in violation of 42
U.S.C. 7412(r)(7)(H)(v) or any other provision of federal law.
Subpart C--Access to off-site consequence analysis information by
government officials.
Sec. 1400. 7 In general.
The Administrator shall provide OCA information to government
officials as provided in this subpart. Any OCA information provided to
government officials shall be accompanied by a copy of the notice
prescribed by 42 U.S.C. 7412(r)(7)(H)(vi).
Sec. 1400.8 Access to off-site consequence analysis information by
federal government officials.
The Administrator shall provide any federal government official
with the OCA information requested by the official for official use.
The Administrator shall provide the OCA information to the official in
electronic form, unless the official specifically requests the
information in paper form. The Administrator may charge a fee to cover
the cost of copying OCA information in paper form.
Sec. 1400.9 Access to off-site consequence analysis information by
state and local government officials.
(a) The Administrator shall make available to any state or local
government official for official use the OCA information for stationary
sources located in the official's state.
(b) The Administrator also shall make available to any state or
local government official for official use the OCA information for
stationary sources not located in the official's state, at the request
of the official.
(c) The Administrator shall provide OCA information to a state or
local government official in electronic form, unless the official
specifically requests the information in paper form. The Administrator
may charge a fee to cover the cost of copying OCA information in paper
form.
(d) Any state or local government official is authorized to
provide, for official use, OCA information relating to stationary
sources located in the official's state to other state or local
government officials in that state and to state or local government
officials in a contiguous state.
Subpart D--Other Provisions
Sec. 1400.10 Limitation on public dissemination.
Except as authorized by this part and by 42 U.S.C.
7412(r)(7)(H)(v)(III), federal, state, and local government officials,
and qualified researchers are prohibited from disseminating OCA
information and OCA rankings to the public. Violation of this provision
subjects the violator to criminal liability as provided in 42 U.S.C.
7412(r)(7)(H)(v) and civil liability as provided in 42 U.S.C. 7413.
Sec. 1400.11 Limitation on dissemination to state and local government
officials.
Except as authorized by this part and by 42 U.S.C.
7412(r)(7)(H)(v)(III), federal, state, and local government officials,
and qualified researchers are prohibited from disseminating OCA
information to state and local government officials. Violation of this
provision subjects the violator to civil liability as provided in 42
U.S.C. 7413.
Sec. 1400.12 Qualified researchers.
The Administrator is authorized to provide OCA information,
including facility identification, to qualified researchers pursuant to
a system developed and implemented under 42 U.S.C. 7412(r)(7)(H)(vii),
in consultation with the Attorney General.
Sec. 1400.13 Read-only database.
The Administrator is authorized to establish, pursuant to 42 U.S.C.
7412(r)(7)(H)(viii), an information technology system that makes
available to the public off-site consequence analysis information by
means of a central database under the control of the federal government
that contains information that users may read, but that provides no
means by which an electronic or mechanical copy of the information may
be made.
[FR Doc. 00-19785 Filed 8-3-00; 8:45 am]
BILLING CODE 4410-14-U
![[logo] US EPA](https://webarchive.library.unt.edu/eot2008/20090514070204im_/http://www.epa.gov/epafiles/images/logo_epaseal.gif){kind=logo}