[Federal Register Volume 60, Number 44 (Tuesday, March 7, 1995)]
[Proposed Rules]
[Pages 12492-12519]
From the Federal Register Online via the Government Publishing Office [www.gpo.gov]
[FR Doc No: 95-5021]



=======================================================================
-----------------------------------------------------------------------

ENVIRONMENTAL PROTECTION AGENCY

40 CFR Parts 51 and 58

[AD-FRL-5157-7]


Proposed Requirements for Implementation Plans and Ambient Air 
Quality Surveillance for Sulfur Oxides (Sulfur Dioxide) National 
Ambient Air Quality Standards

AGENCY: Environmental Protection Agency (EPA).

ACTION: Proposed rule.

-----------------------------------------------------------------------

SUMMARY: Today's action proposes implementation strategies for reducing 
short-term high concentration sulfur dioxide (SO2) emissions in 
the ambient air. The EPA is concerned that a segment of the asthmatic 
population may be at increased health risk when exposed to 5-minute 
peak concentrations of SO2 in the ambient air while exercising. 
``Exercising'' in this case can include walking up stairs or hills, as 
well as more strenuous activities.
    In a related document published on November 15, 1994 in the Federal 
Register (part 50/53 document), EPA proposed not to revise the current 
24-hour and annual primary national ambient air quality standards 
(NAAQS) for sulfur oxides (measured as SO2) while soliciting 
comment on the possible need to adopt additional regulatory measures to 
address short-term peak SO2 exposures. The three alternatives 
under consideration include: Augmenting the implementation of the 
existing standards by focusing on those sources or source types likely 
to produce high 5-minute peak SO2 concentrations; establishing a 
new regulatory program under the authority of section 303 of the Clean 
Air Act (Act) to supplement protection provided by the existing 
SO2 NAAQS; and revising the existing SO2 NAAQS by adding a 
new 5-minute NAAQS of 0.60 ppm SO2, 1 expected exceedance. All 
three regulatory alternatives would be implemented through a risk-based 
targeted strategy designed to protect the population at risk while 
minimizing the burden on the States for implementation.
    This document presents EPA's proposed targeted implementation 
strategy and the associated regulatory requirements for implementing 
each of the regulatory measures under consideration. Also in this 
document, EPA solicits comments on appropriate changes to the new 
source review (NSR) programs as they relate to the 5-minute NAAQS 
regulatory alternative, and EPA [[Page 12493]] proposes to incorporate 
appropriate changes to the ambient air quality surveillance 
requirements.

DATES: Written comments on this proposal must be received by June 6, 
1995. The EPA will hold a public hearing on this document in 
approximately 30 days and will announce the time and place in a 
subsequent Federal Register document.

ADDRESSES: Submit comments on the proposed revisions to the 
requirements for the preparation, adoption, and submittal of 
implementation plans (two copies are preferred) to: Office of Air and 
Radiation Docket and Information Center (Air Docket 6102), Room M 1500, 
U.S. Environmental Protection Agency, Attention: Docket No. A-94-55 
(for part 51 comments) or A-94-56 (for part 58 comments), 401 M Street, 
S.W., Washington, DC 20460. The docket may be inspected between 8:00 
a.m. and 5:30 p.m. on weekdays, and a reasonable fee may be charged for 
copying. The Air Docket may be called at 202-260-7548.

FOR FURTHER INFORMATION CONTACT: Laura D. McKelvey, Information 
Transfer and Program Integration Division (MD-12), U.S. Environmental 
Protection Agency, Research Triangle Park, NC 27711, telephone (919) 
541-5497, for the part 51 SIP. For parts 51 and 52 new source review 
programs, contact Dan deRoeck, Information Transfer and Program 
Integration Division (MD-12), U.S. Environmental Protection Agency, 
Research Triangle Park, NC 27711, telephone (919) 541-5593. For part 58 
ambient air quality surveillance, contact David Lutz, Emissions 
Monitoring and Analysis Division (MD-14), U.S. Environmental Protection 
Agency, Research Triangle Park, NC 27711, telephone (919) 541-5476.

Table of Contents

I. Background
II. Targeted Implementation Strategy
    A. Background
    1. Modeling
    2. Ambient Monitoring
    B. Implementing the Targeting Strategy
    1. Ranking of Source Categories
    2. Other Considerations
    3. States' Targeted SO2 Monitoring Program
    4. Addressing the Problem
    C. Relocating Monitors
    1. Resource Concerns
    2. Siting Concerns
    3. Trends Data Concerns
    4. Barriers
    5. Conclusion
    D. Compliance and Enforcement Issues
    1. Averaging Times for Emission Limits
    2. Malfunction Policy
    3. Conclusion
III. Requirements Associated with Retention of Existing NAAQS
IV. Requirements Associated with Retention of Existing NAAQS and 
Implementation of a Section 303 Program
V. Requirements Associated with a New 5-Minute SO2 NAAQS
    A. Targeted Implementation Strategy
    B. Designations--Section 107
    1. Statutory Requirements
    2. Timeframe for Submittal of Designations by State
    3. Determining Initial Designation of an Area
    4. Determining the Boundaries of Designated Areas
    5. Promulgation of Designations by EPA
    6. Failing to Submit Designations
    C. State Implementation Plans (SIP's)
    1. General SIP Requirements--Section 110(a)
    2. General SIP Requirements--Section 110(a)(2)
    a. Statutory and Existing Regulatory Requirements
    b. Statewide SIP's for the Revised SO2 NAAQS
    c. New Source Review Issues
    d. Schedule for Submittal of Section 110(a)(1) SIP's
    D. Nonattainment Area Requirements
    1. Attainment and SIP Submittal Dates
    2. Classifications--Section 172(a)(1)
    3. Nonattainment Plan Provisions--Section 172(c)
    a. Statutory and Existing Regulatory Requirements
    b. Reasonably Available Control Measures (Including Reasonably 
Available Control Technology)
    c. Emission Inventory
    d. Control Strategy Demonstration
    e. Reasonable Further Progress
    f. Permits for New and Modified Major Stationary Sources
    g. Contingency Measures
    E. SIP Processing Requirements
    1. SIP Completeness
    2. Approval/Disapproval of Plan
    3. Sanctions and Other Consequences of SIP Deficiencies
VI. Significant Harm Levels and Episode Criteria
VII. Proposed Revisions to Part 58 Monitoring Regulations
    A. Section 58.1 Definitions
    B. Appendix C--Ambient Air Quality Monitoring Methodology
    C. Appendix D--Network Design for State and Local Air Monitoring 
Stations (SLAMS) and National Air Monitoring Stations (NAMS)
    D. Appendix F--Annual SLAMS Report
    E. Appendix G--Air Quality Index Reporting and Daily Reporting
VIII. Transition Issues
IX. Other Clean Air Act Amendment Authorities Affecting SO2 
Sources
X. Public Participation
    A. Comments and the Public Docket
    B. Public Hearing
XI. Administrative Requirements
    A. Regulatory Impact Analysis
    B. Impact on Reporting Requirements
    C. Impact on Small Entities
    D. Reduction of Governmental Burden
    E. Environmental Justice
Appendix A--References

SUPPLEMENTARY INFORMATION:

I. Background

    As required under sections 108 and 109 of the Act, EPA has 
completed a thorough review of the air quality criteria and the current 
SO2 NAAQS. Based on the health effects information assessed in the 
air quality criteria, EPA provisionally concludes that the current 24-
hour and annual primary standards provide adequate protection against 
the effects associated with those averaging periods. As discussed in 
detail in the part 50/53 document (59 FR 58958), the key issue that 
emerged from the review is whether additional regulatory measures are 
needed to provide additional protection for asthmatic individuals that 
may be exposed to high 5-minute peak SO2 concentrations.
    As discussed in the part 50/53 document, the available air quality 
and exposure data indicate that the likelihood that the asthmatic 
population as a whole would be exposed to 5-minute peak SO2 
concentrations of concern, while outdoors and at exercise, is very low 
when viewed from a national perspective. The data indicate, however, 
that high peak SO2 concentrations can occur around certain sources 
or source types with some frequency, suggesting asthmatic individuals 
that reside in the vicinity of such sources or source types will be at 
greater health risk than indicated for the asthmatic population as a 
whole. These assessments lead EPA to conclude that if any additional 
regulatory measures are adopted to provide additional protection, they 
should be implemented through a risk-based targeted strategy that 
focuses on those individual sources most likely to produce high 5-
minute peak SO2 concentrations.
    Based on these consideration, EPA is soliciting comment on the part 
50/53 document on three regulatory alternatives: (1) Augmenting 
implementation of the existing standards by focusing on those sources 
or source types likely to produce high 5-minute peak SO2 
concentrations; (2) establishing a new regulatory program under section 
303 of the Act to supplement the protection provided by the existing 
NAAQS; and (3) revising the existing NAAQS by adding a new 5-minute 
standard of 0.60 ppm, 1 expected exceedance. Because the risk-based 
targeted strategy is an integral part of each of the three alternatives 
being proposed for comment, this notice will first present EPA's 
approach for targeting sources with a high potential [[Page 12494]] for 
causing or contributing to high 5-minute peak SO2 concentrations. 
As discussed below and in the part 58 notice, a key element of this 
strategy will be to relocate existing SO2 monitors to areas in 
proximity of point sources of concern. The relocation of monitors is 
necessary because the existing SO2 monitoring network is designed 
to characterize urban ambient air quality associated with 3-hour, 24-
hour, and annual SO2 concentrations. These monitors are not 
located to measure peak SO2 concentrations from point sources. As 
a result, EPA's existing guidance on siting criteria, the spanning of 
SO2 instruments, and instrument response time likely leads to 
underestimates of high 5-minute peak SO2 concentrations. To 
address these concerns, EPA is proposing revisions to the ambient air 
quality surveillance requirements (40 CFR part 58) and proposed certain 
technical changes to the requirements for Ambient Air Monitoring 
Reference and Equivalent Methods (40 CFR part 53) in the part 50/53 
document.
    In addition to outlining the targeted implementation strategy, this 
notice presents EPA's proposed program for implementing the section 303 
program and the 5-minute SO2 NAAQS alternative. Regardless of the 
alternative selected (i.e., retain the existing standards but augment 
their implementation, establish a new 303 program, or add a new 5-
minute NAAQS), the targeted implementation strategy would be used to 
identify areas that may be subject to high 5-minute SO2 
concentrations. The measures that sources must take if they cause or 
contribute to such high peaks and the actions that the States must take 
will vary depending on the proposed alternative, if any, selected.
    The following discussion gives statutory background information on 
the regulatory approach used in addressing air pollution. Under 
sections 108 and 109 of the Act, EPA is responsible for issuing air 
quality criteria and for proposing and promulgating NAAQS. Under 
section 110(a)(1) and part D of title I, the States then have primary 
responsibility for implementing the NAAQS. In broad outline, each State 
must develop and submit to EPA a plan that provides for attainment of 
each NAAQS within certain time limits. The EPA must review the SIP 
submittal and approve or disapprove its provisions. If States fail to 
submit required SIP's or submit inadequate SIP's, and the deficiencies 
are not cured within specified time periods, the States become subject 
to certain sanctions under section 179, and EPA ultimately becomes 
subject to an obligation to promulgate a Federal implementation plan 
(FIP). For a more complete discussion of the provisions of title I of 
the Act, see the General Preamble for the Implementation of Title I of 
the Clean Air Act Amendments of 1990 published in the Federal Register 
on April 16, 1992 (57 FR 13498).
    The 1990 Amendments preserved the existing framework of the SIP 
process, i.e., States are still responsible for preparing and 
submitting SIP's, and EPA is still responsible for reviewing and 
approving or disapproving SIP's. In addition, the 1990 Amendments, 
among other things, provide EPA with the unilateral authority to 
designate areas as either attainment, nonattainment or unclassifiable 
with respect to any NAAQS (see generally, section 107(d)(1)). States 
with areas designated nonattainment for a NAAQS are required to submit 
SIP's which provide for attainment of that NAAQS. States can face 
sanctions and other repercussions if they fail to meet the various SIP 
requirements of title I.
    In general, for each of the proposed regulatory alternatives, the 
Act may or may not require specific actions on the part of EPA or the 
States. If the existing NAAQS is retained, then the Act imposes no new 
SIP requirements on EPA and the States, although EPA will use its 
discretionary authority to effectuate the Act's protective purposes by 
requiring States to implement targeted monitoring around sources 
capable of producing short-term high concentrations of SO2 to the 
extent that those sources contribute to ambient concentrations of 
SO2. If the existing NAAQS is retained along with a trigger level 
for implementing an emergency program under section 303, then the State 
would be principally responsible for developing and implementing the 
necessary prevention and/or abatement strategies. If a new 5-minute 
NAAQS is established, States would have to develop and submit SIP's 
which provide for implementation, maintenance and enforcement of the 
new NAAQS.
    Further discussion of the requirements that are to be met by the 
States is provided below with regard to each of the additional 
regulatory alternatives to be considered by EPA.

II. Targeted Implementation Strategy

    This section principally proposes EPA's strategy to identify those 
areas where the potential exists for exceedances of the current 
SO2 NAAQS as well as the potential for high 5-minute 
concentrations of SO2. This strategy has two stages. The first 
stage is to identify potential problem areas and then to conduct 
ambient monitoring at those areas. The second stage is to take 
corrective action should monitoring conducted during the first stage 
reveal concentrations in excess of the appropriate SO2 NAAQS or 
trigger level. To begin this strategy, EPA intends to refocus Agency 
monitoring resources into those areas with potential 5-minute SO2 
peaks. The development and implementation of this strategy relies on 
the ability of the States to identify the specific emission and 
operating characteristics of sources which can contribute to violations 
of the existing NAAQS as well as contribute to high 5-minute SO2 
concentrations. Successful implementation of this strategy will result 
in either the identification of additional SO2 problem areas or 
the conclusion that the ambient SO2 problem is largely solved. It 
also allows EPA to apply finite resources in an efficient way where 
public health is most likely to be jeopardized by air pollution. The 
EPA intends to pursue this targeted strategy regardless of the outcome 
of the NAAQS proposal published in the part 50/53 notice and solicits 
comments on the targeted implementation strategy.

A. Background

1. Modeling
    For implementing the current SO2 program, EPA has historically 
relied on mathematical dispersion models for predicting air pollutant 
concentrations for the following needs: (1) For redesignating areas to 
nonattainment or attainment under section 107 of the Act; (2) for 
setting emission limits for an attainment strategy as required per 14 
section 110(a)(2)(K) and part 40 of the Code of Federal Regulations, 
Sec. 51.115 (40 CFR 51.115); (3) for predicting locations of maximum 
concentrations for siting monitors; (4) for determining boundaries of 
nonattainment areas; (5) for predicting consumption of ambient air 
increments under prevention of significant deterioration (PSD); and (6) 
for determining, under nonattainment NSR, if the significance level, 
used for determining if a major source or modification is considered to 
cause or contribute to a violation of the NAAQS, is exceeded.
    The ``Guideline on Air Quality Models (Revised),'' EPA-450/2-78-
027R, hereinafter referred to as ``the Modeling Guideline,'' has 
provided a common basis for conducting such modeling. The Modeling 
Guideline was incorporated into 40 CFR part 51 on July 20, 1993 (58 FR 
38816) as appendix W. However, modeling is not currently feasible for 
predicting 5-minute ambient [[Page 12495]] air concentrations of 
SO2. This is due to present uncertainties regarding the ability of 
models to reliably predict SO2 concentrations for 5-minute periods 
and uncertainties with the accuracy of the input data needed to run the 
models. A brief summary of issues follows.
    Validation. Although models are available, they have not been 
applied in predicting 5-minute SO2 concentrations. Model 
validation studies have not been conducted to determine whether 
existing models can estimate with sufficient accuracy to be used in a 
regulatory context. Model validation studies are therefore necessary to 
determine the precision needed for input data for achieving the desired 
prediction accuracy. This would help determine, for example, whether 
on-site 5-minute meteorological data are needed or if nearby National 
Weather Service data are sufficient.
    Emissions Data. In addition to the unassessed uncertainties of 
models, the accuracy and availability of input data, such as emissions, 
meteorology, and the occurrence of a short-term release (e.g., a 
process upset or control equipment malfunction) necessary to run the 
models, limits the ability to accurately predict 5-minute SO2 
concentrations at this time. Obtaining accurate source emission data 
for 5-minute periods is of critical importance. However, it is 
difficult to obtain such data since such data often depend on trying to 
measure emissions that may occur infrequently and at unpredictable 
times, concentrations, and flow rates (estimates of both flow rates and 
pollutant concentrations are necessary to determine mass emissions 
unless a mass balance can be performed, which would be difficult on a 
5-minute basis). Moreover, emergency bypass valves, where measurements 
of emissions might be most appropriate under some circumstances, are 
infrequently used and therefore are not appropriate sites for the 
installation of monitors for continuous measurement of flow rates or 
pollutant concentrations.
    Predicting Short-term Events. Current models used for predicting 
ambient air concentrations rely on a known emission release, usually 
some steady-state emission rate, and known past meteorological data. 
Short-term models use hourly weather data from the National Weather 
Service or from on-site meteorological stations, which are preprocessed 
before being used in the model. Long-term models use joint frequency 
distribution summaries of wind speed, direction and atmospheric 
stability category. In order to model for emission releases due to 
malfunctions, a method of determining the expected frequency of these 
malfunctions would have to be employed (e.g., a Monte Carlo simulation 
which is a computer simulation using random sampling techniques to 
obtain approximate solutions to mathematical or physical problems 
especially in terms of a range of values each of which has a calculated 
probability of being the solution). To date, EPA has never attempted to 
integrate dispersion modeling with malfunction frequency data to set 
emission limits, or to perform any other regulatory modeling tasks. 
Indeed, EPA's longstanding position has been to regard malfunctions as 
violations of applicable control requirements, subject to enforcement, 
unless it can be shown that such malfunctions are truly unavoidable 
(Bennett, 1982). To allow deviations from this policy, EPA would need 
to develop a method along with policy and guidance for its use, which 
EPA does not intend to do at this time.
    Meteorological Data. On-site meteorological data are preferable, 
but National Weather Service data may be acceptable if a station is 
nearby and deemed representative of the area modeled. The 
meteorological data requirements for 5-minute SO2 modeling could 
be determined through model evaluation studies, as discussed earlier in 
this section.
    For these reasons, in contrast with longer averaging periods, 
models cannot currently be used to predict 5-minute SO2 excursions 
needed to support a 5-minute NAAQS. However, despite these limitations, 
current models may still be used as a tool in a qualitative sense in 
the decision-making process for determining boundaries of nonattainment 
areas and for siting of monitors in areas of maximum concentrations. 
Consequently, the targeted implementation strategy which is designed to 
find areas exposed to high, 5-minute concentrations of SO2 will 
rely principally on ambient air monitoring instead of modeling.
2. Ambient Monitoring
    Requirements for monitoring are established at 40 CFR Part 58--
Ambient Air Quality Surveillance. This part: (1) Contains criteria and 
requirements for ambient air quality monitoring and requirements for 
reporting ambient air quality data and information; (2) contains 
requirements pertaining to provisions for an air quality surveillance 
system in the SIP; (3) acts to establish a national ambient air quality 
monitoring network for the purpose of providing timely air quality data 
upon which to base national assessments and policy decisions; and (4) 
includes requirements for the daily reporting of an index of ambient 
air quality to ensure that the population of major urban areas are 
informed daily of local air quality conditions.
    In the early 1970's when EPA and the States first began to monitor 
for SO2 in the ambient air, SO2 emissions were greater and 
more widespread than today. Combustion of sulfur-bearing fuels occurred 
not only in industrial and utility settings but in private settings as 
well. Fuel oil and coal were burned in residences and building boilers 
for warmth. For this reason and because of the potential for exposures 
of the population, large metropolitan areas were generally selected for 
monitoring. Sulfur oxide emissions have decreased about 27 percent 
since 1970 (EPA, 1992b). Today most residences and buildings use 
electricity or natural gas for heating and nearby industrial or utility 
sources have installed control devices or have switched to lower sulfur 
fuel resulting in less sulfur emissions in the vicinity of the ambient 
air monitors. Because of these reductions in SO2 emissions in 
populated areas, only a small number of monitors are now recording 
exceedances. Even these few exceedances are due not to area sources of 
SO2 but instead to emissions from nearby industrial sources. 
Despite these changes in the profile of sources of SO2 emissions, 
the SO2 ambient air monitoring network has not been modified to 
reflect the ambient air quality for SO2 near industrial sources.
    As a result of past emphasis on urban scale air quality management, 
SO2 monitoring networks are designed to measure population 
exposure over a large area and are not generally designed to measure 
the influence of specific point sources. To an increasing extent, 
therefore, SO2 nonattainment areas have been identified by air 
quality dispersion models and defined by one or a few point sources 
with probability of causing a violation of the SO2 NAAQS when 
operating at allowable emission limits at times of unfavorable 
meteorology. Increased concerns about high short-term concentrations of 
SO2 occurring near point sources, together with the prevalence of 
low concentrations at existing networks and the inability of models to 
predict short-term concentrations, suggest a need to redirect monitor 
networks near these sources.
    As already briefly discussed, there are about 675 SO2 SLAMS 
monitors across the Nation. In this notice, EPA is proposing changes to 
40 CFR part 58 to allow for fewer SLAMS monitors per metropolitan 
statistical area. This will enable monitors and resources to be 
redirected towards placing monitors [[Page 12496]] near point sources. 
There is a higher initial cost associated with finding and setting up 
new monitoring sites than the annual operating cost of the monitor 
itself. Because of this and because of limited State monitoring 
resources, not all monitors initially freed up can be immediately 
placed around a targeted source, but will be phased in over a period of 
time.
    For the reasons stated above, EPA proposes to direct States to 
redeploy SO2 monitors around targeted sources of SO2 and 
respan the instrumentation at selected sites to measure values above 
0.5 parts per million (ppm). The monitors will be sited at microscale, 
middle, or neighborhood distance from the targeted sources in order to 
best measure high, 5-minute concentrations of SO2. Micro, middle, 
neighborhood, and urban scales are all more completely defined in 40 
CFR part 58, appendix D. The EPA and States will first monitor around 
those sources in areas with population with the greatest potential to 
exposure to 5-minute, peak SO2 levels. The EPA and States will 
consider discontinuing the operation of existing monitors and relocate 
them for the purpose of monitoring around targeted sources (see part 58 
discussion published elsewhere in this notice for monitoring 
requirements).

B. Implementing the Targeting Strategy

    As discussed earlier, the available air quality and exposure 
information indicates that a large degree of protection against 
exposure to short-term peak SO2 concentrations is provided by the 
current NAAQS. Full implementation of the Acid Rain Program will result 
in further reduction of SO2 emissions and the likelihood of peak 
SO2 concentrations. The available data indicate, however, that 
peak concentrations of SO2 can still occur around certain sources 
or source types with some frequency, suggesting asthmatic individuals 
who reside in the vicinity of such sources or source types will be at 
greater health risk than indicated for the asthmatic population as a 
whole. These assessments have led EPA to conclude that any regulatory 
measures adopted to provide additional protection should be implemented 
through a risk-based targeted strategy that focuses on those individual 
sources more likely to produce high 5-minute peaks.
    Therefore, in order to gather more information, to focus 
implementation efforts on those sources that EPA's existing data 
suggest may pose the greatest health risk, and to allocate monitoring 
resources as efficiently as possible, EPA has developed an approach to 
guide States in developing a prioritized list of sources to be targeted 
for monitoring. As further discussed below, potential sources have been 
placed in one of three groups based on the overall likelihood of the 
source category to emit high 5-minute SO2 peaks. However, before 
redeploying monitors, States must evaluate each of these facilities 
individually, basing their decision on more specific information such 
as size, configuration, compliance history and proximity to population 
centers.
    As just described, States need to review their current SO2 
monitoring networks to determine which monitor sites should continue 
operating and which should be discontinued and relocated around 
potential sources. The EPA will work with each State to develop a 
targeted SO2 monitoring plan to implement the strategy, based on 
the number of targeted sources, SO2 monitoring resources, and 
within a reasonable time horizon.
    The EPA believes that new locations for siting monitors should be 
in the vicinity of sources suspected of causing short-term SO2 
peaks. Some examples of sources which emit SO2 are petroleum 
refineries, sulfuric acid plants, fossil fuel-fired industrial boilers, 
utility boilers, pulp and paper mills, iron and steel mills, wet corn 
milling operations, nonferrous smelters, carbon black manufacturing, 
portland cement manufacturing, phosphatic fertilizer production, and 
natural gas production. This list is not exhaustive and could 
potentially include other process sources with known emissions of 
SO2. These sources have the ability to emit relatively large 
quantities of SO2 over short durations. Such large quantities of 
emissions may be due to releases from batch type operations, 
operational malfunctions or upsets requiring control equipment 
bypasses, control equipment malfunctions that can result in 
uncontrolled emissions to the atmosphere, startup/shutdown, short 
stacks subject to downwash, or fugitive emissions.
1. Ranking of Source Categories
    The information most heavily relied on in developing this ranking 
of source categories was: (1) Available 5-minute air quality data 
documenting the number of high, short-term concentrations observed in 
the vicinity of various sources by monitoring networks (Table 3-1, EPA, 
1994b); (2) estimates of exposures from various source types, which 
integrated a source's likelihood to emit short-term SO2 peaks with 
the size and activity of the surrounding population, as summarized in 
Table 3-5, Table B-1, and Table B-2 (EPA, 1994b), as well as 
accompanying documentation (Rosenbaum et al., 1992; Stoeckenius et al., 
1990; Burton et al., 1987); and (3) the Geographic Targeting Data Base 
for nonutility sources that is derived from combining a census of 
manufacturing, the EPA Facilities Index System, and the EPA Aerometric 
Information Retrieval System (AIRS) into a projected source impact data 
set. This data base, which will be available through AIRS, is a data 
set of nonutility sources sorted on the projected annual process 
emissions per source and per size category.
    In order to further refine the ranking of source categories, both 
within and between groups, EPA solicits technical information 
concerning several issues which include: (1) The likelihood of source 
categories to produce short-term SO2 peaks; (2) the 
characteristics, within a source category which cause a subset of 
facilities to be more likely to produce short-term SO2 peaks; and 
(3) the factors which are likely to drive the variability in SO2 
emissions of individual facilities within a source category.
    The ranking described here separates source categories into three 
groups: A, B, and C. In pursuit of this targeting strategy, EPA intends 
to require States to evaluate groups A, B, and C sources and produce a 
refined monitoring plan. States are free to substitute, e.g., group B 
sources for group A sources in their priority schemes, but should 
provide a reasoned justification for finding that the risks posed by 
these sources justifies such substitution. Ultimately, EPA anticipates 
that sources in all three groups will be assessed for their exposure 
potential and appropriate actions taken to address them. The EPA 
believes that there is a higher probability of finding individual 
sources that produce high, short-term ambient concentrations of 
SO2 within each source category in group A than in the other 
groups. As such, they are judged in general to pose the highest risk of 
exposing population in their vicinity to high, short-term 
concentrations of SO2, as well as potentially exposing some 
individuals to several peaks per year.
    The source categories within group A were generally found to meet 
two of the three following characteristics. Either the source category 
contained SO2 sources which: (1) Have a high emission rate, (2) 
are near monitors which measured 5-minute peaks, or (3) are estimated, 
based on exposure analysis, to expose a high number of asthmatics 
living in their vicinity at elevated ventilation rates to SO2 
concentrations greater than 0.6 ppm. In addition, these source 
categories are known to have [[Page 12497]] short-term releases due to 
events discussed later.
    Group A consists of the following source categories: Sulfite pulp 
and paper mills, primary copper smelters, primary lead smelters, 
aluminum smelters, and the top 20 percent of the petroleum refineries 
in terms of projected annual emissions of SO2 as listed in the 
Geographic Targeting Data Base.
    Source categories were selected for group B because they have high 
annual emissions or are subject to events leading to short-term 
releases of SO2. In addition, in some instances, there were air 
quality or exposure data which indicate the source category to be of 
concern for emitting short-term SO2 peaks.
    The EPA judged group B source categories to have the potential to 
produce high 5-minute peaks of SO2 but to pose less risk than 
group A because: (1) Air quality or exposure data indicated that the 
potential to emit high 5-minute peaks of SO2 was less than for 
group A; (2) the grouping was based on annual emission data, but lacked 
5-minute data to estimate risk; or (3) the overall risk posed by the 
source category was judged to be low. This was the case for industrial 
boilers because, while exposure analysis indicated that this group was 
responsible for a considerable number of exposures, the exposures were 
attributed to a very small subset of industrial boilers. The EPA 
expects that States will examine their source categories within this 
group very closely for inclusion in the targeted SO2 monitoring 
plan.
    The group B sources are as follows: Kraft sulfate pulp and paper 
mills, secondary copper smelters, secondary lead smelters, the 
remaining petroleum refineries, iron and steel mills, carbon black 
manufacturing, portland cement manufacturing, crude petroleum and 
natural gas extraction processes, phosphatic fertilizer manufacturing, 
industrial boilers, and sulfuric acid plants.
    Industrial boilers were placed in this group because they accounted 
for about 30 to 50 percent of the 5-minute SO2 exposure events 
given in the staff paper supplement (Table 3-5, EPA, 1994b). However, 
in a study by Stoeckenius et al. (Table 2-14, 1990), approximately half 
of the total industrial boiler exposures were attributed to a very 
small proportion (2 percent) of the total population of 
industrial boilers analyzed. Good engineering judgment suggests that 
the use of higher sulfur coal and short stack height would contribute 
to an increased likelihood of producing ambient SO2 peaks.
    The group C source category consists of utility boilers. Although 
utility boilers can emit large quantities of SO2, many power 
plants are not anticipated to cause 5-minute violations despite their 
high emission rates due to tall stacks and steady-state operating 
conditions. They are placed in group C because as a source category, 
utility boilers may be responsible for approximately 17 to 37 percent 
of total estimated exposures (Table 3-5, EPA, 1994b). However, the risk 
of exposures is very unevenly distributed across the sources in this 
category. Approximately 75 percent of the utility sector's post-title 
IV exposures were estimated to result from less than 10 percent of the 
power plants (Rosenbaum, 1992, Table 3, Burton et al., 1987).
    With the passage of the 1990 Amendments, Congress created under 
title IV an SO2 emission trading program as an integral part of 
the Acid Rain Program, which is designed to reduce SO2 emissions 
by 10 million tons nationwide by the year 2010. Phase I, which begins 
in 1995, reduces emissions from the 110 largest emitting power plants, 
which are identified in table A of section 404 of the Act. The Acid 
Rain Program introduces a flexibility for sources to choose the most 
cost-effective compliance strategy to achieve their emission reduction 
obligations and to maintain the national cap of 8.95 million tons of 
SO2 emissions. Compliance flexibility may involve switching to 
low-sulfur coal, scrubbing, conservation, other emission control 
technologies, or buying SO2 allowances.
    Title IV sources participating in the Acid Rain Program are under 
the obligation to match their annual SO2 emissions with their 
allowance holdings. They are also required to meet all other 
requirements of the Act and regulations that apply to them, including 
the NAAQS. Therefore, the compliance flexibility offered under the Acid 
Rain Program does not permit any source to violate regulations adopted 
to attain or maintain the SO2 NAAQS. Emissions from these sources 
will be closely tracked, because title IV sources are also required to 
install continuous emissions monitoring systems (CEMS) and report to 
EPA on a quarterly basis their emissions of SO2, nitrogen oxides, 
and carbon dioxide.
    Further improvements in air quality are expected to be realized 
from the SO2 emission reductions under Phase II of the Acid Rain 
Program to be implemented by January 1, 2000 under title IV of the Act. 
Because of the potential to have higher emissions and because of 
potential plume downwash and interaction of complex terrain, EPA is 
mainly concerned with those power plants that buy allowances rather 
than reduce emissions themselves in order to comply with title IV and 
those located in complex terrain, respectively. Complex terrain is 
defined for modeling applications as that terrain exceeding the height 
of the stack, but this definition is being applied here for monitoring 
applications as well. In a study done for EPA, that is contained in the 
docket for this rulemaking (Polkowsky, 1991), many of the predicted 
exceedances of the SO2 standards in the vicinity of power plants 
should be reduced or eliminated by allocating allowances based on a 
reduced rate under Phase II. Any remaining exceedances not addressed by 
the more restrictive Phase II emission rates will require a reanalysis 
of the SO2 NAAQS control strategy demonstration and consideration 
of more restrictive emission limits to protect the air quality 
standards.
    Because of the SO2 reductions that will occur under the Acid 
Rain Program, the accurate stack monitoring of their emissions, and the 
long-range atmospheric transport of these emissions due to taller 
stacks at most large utilities, EPA believes that higher priority in 
placing ambient monitors should be given to nonutility sources. 
However, in instances at a particular power plant where the possibility 
of high 5-minute emission peaks still exists, EPA believes that 
consideration should be given by the State to locating monitors near 
the facility.
2. Other Considerations
    In addition to the guidelines and groupings listed above, which are 
based largely on available information concerning the likelihood of a 
source type to produce concentrated peaks of SO2, States may have 
other information which may lead them to believe that a source located 
in a lower probability group should be made a higher priority for 
SO2 monitoring. Of particular importance to consider is any 
available information on potential population exposure, inferred in 
part by the population in the vicinity of the source.
    In addition, other information can be incorporated by States into 
an evaluation of the relative likelihood of sources under their 
jurisdiction to produce SO2 exposures, thus refining their 
judgments on priority of monitoring decisions. Such other information 
can include the type of process being used (i.e., one type of process 
within a source category may be less efficient and known to emit more 
SO2 than a newer one), a history of past [[Page 12498]] upsets or 
malfunctions, the type of fuel used, the type of terrain around the 
source (e.g., is the source in a river valley or on flat terrain), 
knowledge of how well the source is controlled, and a history of 
citizen complaints, and should be considered by the States when 
deciding which sources to monitor first. Such considerations would be 
noted in each State's targeted SO2 monitoring plan presented 
during the annual SLAMS review as described below.
    As part of the targeting strategy, the States will also need to 
decide how much relative weight should be given any particular source. 
For example, a State would have to determine how heavily to weigh a 
group A source in a less densely populated area versus a group C source 
burning a high sulfur fuel in a more densely populated area. In 
addition, some sources are often found collocated with other sources 
such as sulfuric acid plants with copper smelters. Industrial boilers 
may be located with any number of process sources. There may be small 
geographic areas where there is clustering of an assorted number of 
SO2 sources. In these situations there is no precise way to 
determine what source should be targeted first at this point. For this 
reason, the decision making should rest with the States who have better 
knowledge of the individual circumstances pertaining to the potential 
sources to be targeted.
3. States' Targeted SO2 Monitoring Program
    The EPA will review and take appropriate action on the States' 
targeted SO2 monitoring plans during the annual SLAMS network 
review process to ensure that States provide an adequate rationale for 
any deviations from the grouped approach. The States are then expected 
to present to EPA in a targeted SO2 monitoring plan at the annual 
SLAMS network review their listing of sources to be monitored, the 
schedule for conducting such monitoring, and the rationale for 
selecting these sources. Requirements for the targeted SO2 
monitoring plan are discussed later in this notice for part 58 but EPA 
expects the targeted SO2 monitoring plan to be a dynamic process 
that could change depending on data gathered from early rounds of 
monitoring or changes at targeted sources, such as installation of 
control equipment.
    Section 110(a)(2)(B) of the Act requires SIP's which provide for 
the establishment and operation of appropriate devices, methods, 
systems, and procedures necessary to monitor, compile and analyze data 
on ambient air quality. Should EPA determine that a State's targeted 
SO2 monitoring plan is inadequate, then EPA expects to issue a 
call for a SIP revision under section 110(k)(5) of the Act based on a 
finding that the SIP is substantially inadequate in meeting the 
requirement of section 110(a)(2)(B). The EPA solicits comments on all 
aspects of this approach to grouping of sources to investigate 
potential air quality problems.
    In the State targeted SO2 monitoring plan, EPA expects 
SO2 monitoring network reviews to be completed within 1 year of 
the effective date of promulgation of any of the three regulatory 
alternatives. Implementation of network revisions is expected to take 
longer.
4. Addressing the Problem
    Regardless of the regulatory alternative chosen by the 
Administrator, those areas which have monitored exceedances of the 
existing or revised NAAQS or of a section 303 trigger level should 
undergo a compliance inspection by the State of the targeted source. If 
the source is out of compliance, EPA expects that the responsible air 
pollution control agency will initiate appropriate enforcement action 
to bring it into compliance, e.g., by using available administrative or 
judicial enforcement authorities. If the source is in compliance, the 
State will need to pursue other appropriate solutions to the problem as 
discussed later in section III.
    The EPA encourages States to pursue, where appropriate, the 
enforcement and improved compliance options before other regulatory 
actions. In many cases, air quality problems may be due to poor 
operation and maintenance or other resolvable compliance problems. In 
these instances, enforcement action can result in timely resolution of 
violations and avoid the sometimes lengthy regulation development 
process. However, the State should pursue existing regulatory options 
where the regulations are inadequate, e.g., because the source is in 
compliance with the existing regulations and an air quality problem 
still exists.

C. Relocating Monitors

    The EPA's criteria for the network design of monitors are discussed 
in 40 CFR part 58, appendix D. Elsewhere in this notice, EPA is 
proposing changes to part 58 in order to implement the proposed 
targeting program. The EPA recognizes that it is not a trivial matter 
to relocate monitors and that there are concerns that agencies will 
need to consider in making relocation decisions.
1. Resource Concerns
    The EPA believes that the resources currently devoted to monitoring 
ambient concentrations of SO2 may be more effectively utilized 
through systematic evaluations and reconfigurations of existing 
monitoring networks. However, even if States and locals acquire no 
additional SO2 monitors and rely solely on the current number of 
monitors, there will be some costs incurred when relocating monitors. 
Costs associated with moving a monitor include the resources taken in 
locating new sites and negotiating leases along with the capital costs 
of a new shelter and associated equipment. Because of the costs for 
relocating monitors, not all monitors freed up can be immediately 
placed around a targeted source, but will be phased in over a period of 
time. The operating costs saved by not operating these monitors will be 
used toward the costs of relocating monitors.
    In more detail, the costs for moving an SO2 monitor have been 
calculated in 1994 dollars to be $60,940 per site. These costs include 
initial capital costs, operation, and amortization. The initial costs 
include network design and site selection, land lease, power drop, 
shelter, site preparation, calibration equipment, data logger, quality 
assurance plan preparation, etc. The operation costs include routine 
site visits, repairs, maintenance, data acquisition and reporting, 
quality assurance calibrations, and supervision. The amortization costs 
for replacement capital equipment were also calculated.
    The total costs for the initial 3 years are summarized as follows. 
The existing network of 679 NAMS, SLAMS, and industrial monitors costs 
about $16 million per year. The first year costs for reconfiguration 
and operation of NAMS, SLAMS, and industrial monitors in order to 
comply with changes to 40 CFR part 58, which is being proposed in this 
notice and is not a result of the targeted implementation strategy, is 
estimated to be $12.4 million per year. This will leave an available 
$3.6 million to be used toward the targeted implementation strategy the 
first year to establish and operate four monitors around 15 sources.
    The second year costs for operating the NAMS, SLAMS, industrial, 
and targeted implementation strategy monitors is estimated to be $9.6 
million dollars, making available $6.4 million for the targeted 
implementation strategy. This will allow for establishing sites around 
26 sources in addition to the 15 sources from the first year for a 
total of 41 targeted sources. [[Page 12499]] 
    The third year costs for operating monitors are estimated to be 
$11.4 million, leaving $4.6 million for the targeted implementation 
strategy. This will allow for establishing sites around 16 sources in 
addition to the 41 sources established in the first and second years 
for a total of 57 targeted sources. The EPA estimates that monitors at 
7 of the 15 sources established in the first year would be moved in the 
third year due to no monitored violations.
2. Siting Concerns
    The EPA is aware of the many considerations that arise when siting 
monitoring stations. Monitors are usually sited where electrical power 
is already available, they are reasonably secure, the immediate 
environment satisfies the siting criteria of part 58, and they are in 
proximity to the desired locations. Waiver provisions are also included 
in the regulations to deviate from siting criteria when appropriate. 
Generally, monitors are sited at or within reasonable proximity of the 
desired locations. For purposes of convenience, monitors are sometimes 
sited where other pollutants are already monitored.
    When conducting the SO2 network review, EPA-approved air 
quality models and saturation studies may be used to predict locations 
where maximum concentrations are expected within the vicinity of 
SO2 sources or clusters of sources. As discussed earlier, models 
can be used in a qualitative sense to predict relative ambient impacts 
and are useful as a tool for establishing preferred monitor locations 
for predicting 5-minute concentrations.
3. Trends Data Concerns
    A potential concern regarding the movement of monitors is the 
effect on EPA's ability to detect and evaluate trends in air quality. 
When monitors are operated in the same locations for several years, it 
is possible to account for the effects of meteorology, seasonal 
patterns in air pollutant concentrations and other variables specific 
to a monitor location. When monitors are moved, the confidence in 
detecting trends in air pollutant concentrations is compromised due to 
a new set of variables that may affect ambient concentrations at the 
new location.
    The EPA needs to maintain a certain number of monitors for 
detecting and evaluating trends in air pollutant concentrations. 
However, EPA believes that a sufficient number of monitors now used for 
trends analyses are not critical to the objectives of trends reporting 
and should be considered for relocation. Elsewhere in this notice, the 
EPA is proposing changes to 40 CFR part 58, appendix D, in which a 
minimum number of SO2 monitors in the metropolitan areas will be 
retained for trends purposes.
4. Barriers
    Certain institutional barriers may be encountered in some attempts 
to relocate monitors. These stem from the separate political entities 
responsible for implementation of air pollution control programs at the 
State and local levels throughout the U.S. Where monitor sites 
considered for relocation are within the boundaries of one political 
entity, the problems are diminished, since the resources necessary to 
maintain existing monitoring sites may be redirected to the new sites, 
providing the SO2 monitor is not sharing a site with other 
pollutant monitors. Sites in a network around targeted sources of 
SO2 emissions which are located in different States or air 
pollution control districts may present some added difficulties. In 
such cases, resources, such as grants for support of air pollution 
planning and control programs as allowed under section 105 of the Act, 
may be redirected by EPA to aid in relocating and maintaining new 
monitoring stations.
5. Conclusion
    In general, EPA believes that a portion of the monitors now 
directed to monitoring ambient air quality in population areas for 
trends purposes should be considered for relocation. While EPA may not 
normally require monitors operated by industries to be relocated and 
thus industry-operated monitors will not be candidates for relocation, 
EPA strongly encourages companies to evaluate their networks in light 
of today's notice. However, quality-assured data from such monitors 
could allow for the relocation of nearby SLAMS monitors to other 
locations if monitored air quality concentrations from industry-
operated monitors provide assurances that the SO2 NAAQS are 
maintained.

D. Compliance and Enforcement Issues

    Certain compliance and enforcement issues will arise only if either 
the section 303 alternative or the new 5-minute NAAQS alternative is 
selected. The issues are how to determine compliance to ensure 
protection of a trigger level or NAAQS that has a 5-minute averaging 
period, and what actions are appropriate by the State when the cause of 
the violation may be process upsets, startup or shutdown, batch 
operations, or other nonsteady-state sources. As is currently done with 
the NAAQS, measurement of SO2 ambient air concentrations with 
ambient air monitors under each of the three proposed regulatory 
alternatives will serve as indicators of compliance. Enforcement will 
be based on the results of compliance inspections at the source, and 
the compliance inspection will be based on requirements in the 
applicable operating permit or SIP. In most instances, EPA believes 
that in order to ensure protection of the 5-minute NAAQS or trigger 
level, compliance will need to be determined through sources meeting 
recordkeeping and reporting requirements or carrying out any other 
agreed-upon actions designed to reduce short-term emission peaks.
1. Averaging Times for Emission Limits
    Under EPA's policy for emissions averaging under the current 
SO2 NAAQS, sources are to be controlled through the imposition of 
emission limits having averaging times consistent with the averaging 
period of the air quality standard of concern. As an example, in order 
to protect the SO2 ambient air quality standard that has been 
established for a 24-hour period, mass emission limits for sources 
should normally allow averaging of emissions over no more than a 24-
hour period when determining compliance with the limits. The purpose of 
this is to restrict extreme variations in emissions of short duration 
that might otherwise be allowed to occur if emission variations are 
averaged over much longer periods (e.g., 30 days). Air quality 
concentrations in excess of the standard could be produced while 
sources are still complying with long-term average emission limits by 
reducing emissions sufficiently at other times within their emission 
averaging periods.
    A variety of emission limit averaging times had been developed by 
State and local agencies for SIP's both prior and subsequent to the 
implementation of this policy on averaging. As a result, those SIP's 
with averaging times inconsistent with the policy that were adopted 
prior to implementation of the policy are included in an effort by EPA 
to correct general SIP enforcement deficiencies. The EPA has not taken 
final action on those rules developed subsequent to the policy.
    The EPA has allowed the use of stack tests and analysis of fuel 
samples for sulfur content as surrogates for continuous compliance 
monitoring with the emission limits. In many cases, these methods will 
continue to be feasible for ensuring compliance with a 5-minute trigger 
level or NAAQS. Technically, SO2 emissions can be measured in a 
stack at intervals less [[Page 12500]] than 5 minutes using Method 6c 
(the instrumental analyzer procedure) in Appendix A of 40 CFR part 60 
or by using a CEM. However, EPA believes that in many instances 5-
minute releases of SO2 that would cause exceedances of a 5-minute 
NAAQS or trigger level will occur at unpredictable times or as fugitive 
emissions (i.e., not through a stack), making stack tests an 
impractical compliance method. Nor may sampling fuel at 5-minute 
intervals be a practicable alternative as in the case of coal in which 
sulfur content may not be homogeneous. In addition, the source of the 
emission may not be due to combustion of fossil fuel but to chemical 
process emissions.
    The EPA believes that in most instances, in order to attain a 5-
minute NAAQS or trigger level, the State will not be able to rely on 
measurable emission limits but instead on actions by the source to, for 
example, modify equipment or process or to have improved maintenance 
that will address the emission releases that are causing 5-minute 
exceedances. Because of these potential limitations to determining 
compliance of emission limits designed to protect a 5-minute NAAQS or 
trigger level, compliance will in most instances need to consist of the 
State ensuring that the source has implemented the necessary remedies. 
Verification that actions have been effective will require that ambient 
air monitoring continue for a reasonable period, e.g., another 2 years 
following the corrective action. However, in those instances where 
emissions can be feasibly measured on a 5-minute basis or it is 
determined that fuel sampling is a feasible compliance indicator, the 
State may elect to set an emission limit and use emission measurement 
or fuel sampling as the method for determining compliance.
2. Malfunction Policy
    As stated previously, EPA has on occasions used its enforcement 
discretion in determining how and whether to act on unavoidable 
violations of source emission limits during periods of startup, 
shutdown and malfunction (40 CFR 60.11(d)). This policy recognizes that 
during startup and shutdown conditions, effective pollutant control may 
sometimes not be technically feasible due to process temperatures and 
pressures that have not yet stabilized. The policy also recognizes that 
certain source malfunctions are not reasonably foreseeable and are 
unavoidable, which result in uncontrolled emissions to the atmosphere. 
Clearly, in many cases, forces of nature such as floods, tornadoes and 
lightning strikes can overwhelm a source's ability to function in a 
normal fashion and may produce conditions that preclude proper 
operation of sources or control equipment. However, some conditions may 
be reasonably anticipated and proper design of equipment can ameliorate 
their effects (e.g., grounding of equipment for lightning protection, 
observation of flood plains, etc). It is possible in some cases to 
address this through design of redundant control systems to guard 
against the release of uncontrolled emissions to the atmosphere should 
one system suffer a malfunction; however, the cost may be prohibitive 
and such systems are not uniformly required. Some SO2 control 
systems offer this protection, such as dual acid plants operated in 
parallel at petroleum refineries. Should one plant experience 
operational problems in such cases, the other is available to provide a 
continued partial level of sulfur (and ultimately SO2) removal.
3. Conclusion
    As is currently done, where there have been monitored violations of 
the 24-hour, 3-hour, or 5-minute SO2 NAAQS or trigger level, the 
State shall be required to determine the source of the SO2 
emissions and investigate the cause of the emissions at that source. 
Where the results of these investigations demonstrate that improper 
operation and maintenance practices and/or poor control equipment 
design are primarily responsible for release of uncontrolled emissions 
to the atmosphere, the State shall be expected to work with the source 
to take appropriate actions to reduce inadequately controlled source 
emissions.
    For purposes of verifying the results of any corrective actions 
taken and compliance, the EPA intends to rely on continued ambient air 
monitoring. The EPA also anticipates the need to review the 
implementation of its malfunctions policy in light of the concerns 
discussed in this document with the possible result of more stringent 
showings required to justify the conclusion that malfunctions are truly 
unavoidable. Recordkeeping based on earlier baseline assessments of the 
problem at the source should be maintained at the source to assist in 
evaluations should further exceedances be monitored.

III. Requirements Associated With Retention of Existing NAAQS

    The State is not required to revise its SIP to address 5-minute, 
high concentrations of SO2 if the existing NAAQS is retained. 
However, in concert with changes in monitoring requirements for part 58 
proposed in this document, as discussed above, EPA is proposing to 
require States to implement a targeting strategy to more aggressively 
monitor process sources that are likely producing high concentrations 
of SO2 even if for short periods of time. As described previously, 
the targeted strategy will be implemented through the annual SLAMS 
network review during which the States will report on progress made the 
previous year. The EPA believes that the results of such a targeting 
strategy will reduce the possibility and frequency of 5-minute high-
concentration SO2 exposures as an incident to more effectively 
monitoring peak SO2 concentrations and by bringing into compliance 
those sources violating the existing NAAQS. However, EPA acknowledges 
that there may be occurrences of SO2 releases which could exceed 
the 5-minute NAAQS or section 303 trigger level proposed in the part 
50/53 document and not exceed the current SO2 NAAQS. In those 
cases, the State should, nevertheless, conduct compliance inspections 
in the eventuality that the source is out of compliance with current 
SIP requirements. Beyond these measures, EPA would not have authority 
to take further actions under the title I SIP program.
    If violations of the current NAAQS cannot be resolved through 
compliance and enforcement (i.e., the source is in compliance), then 
the State will be expected to take steps to reduce emissions on its own 
initiative by revising the emission limit, by requiring process 
modifications, or other control measures. The State shall then prepare 
a SIP revision for EPA approval in order to make the emission 
reductions federally enforceable. In the event that a State does not 
take these steps, then EPA can take either of two actions: (1) If the 
area is currently designated attainment, using the authority under 
section 107(d) to redesignate the area nonattainment; and/or (2) 
issuing a SIP call under section 110(k)(5) of the Act to notify the 
Governor of the State that the SIP is inadequate to attain and maintain 
the SO2 NAAQS and to call for a SIP revision as necessary to 
correct such inadequacies.
    There are advantages and disadvantages in using either the 
nonattainment redesignation or SIP call approach. For instance, the 
nonattainment redesignation process, in addition to requiring 
expeditious attainment of the standard, imposes the requirements 
applicable under part D, title I, of the Act (e.g., reasonably 
available control measures (RACM), reasonable further progress (RFP), 
[[Page 12501]] nonattainment NSR, and contingency measures), and 
requires sanctions and FIP's if the SIP is not developed and 
implemented in a timely manner.
    While these part D requirements may well be useful in effectively 
addressing the air quality problem, plan development may proceed more 
quickly in response to a SIP call in some cases because the SIP call 
does not entail the process and time needed to undertake a 
redesignation of an area (including the notification of the Governor 
required under section 107(d)(3)). The SIP submitted in response to a 
SIP call under section 110 must also provide for expeditious attainment 
of the NAAQS. A disadvantage of relying on SIP calls for attainment 
areas is that, unless an area is otherwise subject to section 173 
permit requirements, no mandatory sanctions are applicable in the event 
the State fails to respond adequately to the SIP call. The 
discretionary air grant funding sanction under section 179 remains 
available for attainment areas, however. The requirement for EPA to 
promulgate a Federal plan if the State fails to submit an approvable 
SIP is wholly applicable for either option.
    In addition to the advantages and disadvantages just described, 
decisions about which regulatory approach to use should consider 
factors specific to the affected area. Among the factors EPA will 
consider are the following:
    (1) The magnitude of the violation.
    (2) The persistence of violations.
    (3) The exposure potential. (For example, is it near a population 
center or a school?)
    (4) The State's regulatory process. (For example, is it lengthy; 
does the legislature only meet periodically? Would the timeline of one 
option fit better within the State's regulatory frame work?)
    (5) Other sources in the area. (For example, can culpability be 
clearly determined? Would one process facilitate that determination of 
culpability over the other? Is new source growth anticipated?)
    (6) The need for a more objective level of control.
    (7) The type of information available for indicating a problem 
exists (monitoring, modeling, others).
    (8) If there is uncertainty associated with modeling and/or past 
history of failing to attain the standard, does the action taken 
provide for appropriate contingencies that can be implemented if the 
area fails to provide a SIP or to attain and maintain the standards?
    (9) Is there a need for long-range planning for the area and does 
the approach taken facilitate this planning effort?

IV. Requirements Associated With Retention of Existing NAAQS and 
Implementation of a Section 303 Program

    In attempting to address health concerns with population exposure 
to high concentrations of SO2 for short periods of time, one of 
the alternatives that EPA considered in the part 50/53 notice is to 
reaffirm the existing SO2 NAAQS and at the same time to promulgate 
a trigger level for implementation of a program under section 303 of 
the Act. The basic rationale and legal authority for that program are 
discussed in that document. What follows in more detail is the proposed 
implementation program, including the proposed regulatory text. The EPA 
believes that a targeted implementation strategy, as already discussed, 
could be used to find sources that would be subject to further 
emissions or operational control under a section 303 program. The EPA 
believes that a program to protect the public from exposure to high 
concentrations of SO2 for short periods of time may be 
successfully implemented under section 303. The type of program EPA is 
proposing to implement would require States to submit contingency plans 
to EPA that would require certain actions on behalf of the State and 
source once an established ambient SO2 concentration (``trigger 
level'') is violated. The State would be required to take certain 
actions to determine the source of the emissions and to protect against 
future violations of the trigger level.
    As described in the part 50/53 notice concerning the regulatory 
alternative of the section 303 program, EPA believes that sections 303, 
110(a)(2)(G), and 301 provide adequate legal authority to establish 
this program and to promulgate regulations to implement it. As with the 
existing section 303 program, EPA's proposed regulations require States 
to adopt contingency plans under section 110(a)(2)(G) to carry out the 
program. The EPA is proposing to require that each State submit such 
plans to EPA within 18 months of the promulgation of final regulations 
establishing a section 303 program. The EPA believes that section 
110(a)(2)(G) authorizes EPA to require these submissions and that 18 
months is an adequate period of time to develop and submit the programs 
to EPA for approval.
    Once the section 303 trigger level has been violated, EPA proposes 
that the following actions occur. First, within 30 days of a violation 
of the trigger level, the State would carry out a compliance inspection 
of the culpable source. The EPA recommends that the State not wait for 
a violation but conduct a compliance inspection after the first 
exceedance. If the source is out of compliance with its existing 
emission limits, then the State would take the necessary steps to bring 
the source into compliance within 30 days of the compliance inspection. 
If, however, the State determines that bringing the source into 
compliance with its existing emission limits would not be likely to 
prevent further exceedances of the trigger level, or the State 
determines the source to be in compliance with applicable emission 
limits, then further action would be needed. In such circumstances, the 
next step would be for the State and source to examine the cause of the 
emissions. Once that is determined, enforceable actions would need to 
be developed to address the cause of the pollution. These actions must 
eventually be made federally enforceable by adopting them as source-
specific SIP revisions. The EPA proposes to require that actions be 
taken within 60 days of the compliance inspection and provide for 
implementation of any new control measures as expeditiously as 
practicable. The EPA expects that the control measures that may need to 
be implemented to prevent recurrences of 5-minute SO2 peaks may 
include better maintenance of control equipment, better capture of 
fugitive emissions, raising the stack height, or other innovative 
control measures.
    The EPA believes that the actions required of States and sources 
would provide adequate protection against the recurrence of high, 5-
minute SO2 peaks once such emissions are identified as a problem 
for particular sources. The EPA also believes that the time periods for 
taking action that it is proposing are reasonable periods, as they 
provide sufficient time for the required actions to take place, while 
assuring that any necessary corrective actions will be taken and 
implemented as expeditiously as practicable.
    The EPA would also retain the ability to take whatever actions it 
believed appropriate directly under section 303. Thus, EPA could take 
direct action under section 303 prior to the adoption of State 
contingency plans if needed, or take action after their adoption if 
circumstances warranted such Federal action. Moreover, once the section 
303 contingency plans have been adopted and incorporated into SIP's, 
EPA may directly enforce their provisions pursuant to section 113 of 
the Act.
    However, it is EPA's position that the States are primarily 
responsible for carrying out actions under this section 
[[Page 12502]] 303 program. If a State does not exercise its 
responsibility under section 303 once a trigger level has been 
violated, EPA intends to consult with the State prior to taking action 
itself.
    The EPA is proposing to add an Appendix X to 40 CFR part 51 which 
explains the computations necessary to determine from monitoring data 
whether the 5-minute trigger level has been exceeded or violated. 
Appendix X defines several terms, among them, ``5-minute hourly 
maximum,'' ``exceedance,'' ``expected exceedance,'' and ``violation.'' 
Appendix X explains the convention used to calculate expected 
exceedances, which essentially is a procedure which makes an adjustment 
for missing monitoring data.
    In brief, the 5-minute trigger level is not violated when the 
number of expected exceedances per year is less than or equal to one. 
In general, this determination is made by recording the number of 5-
minute hourly maximum exceedances at a monitoring site for each year, 
making the adjustment for missing data (if required), averaging the 
number of exceedances over a 2-year period, and comparing the number 
calculated to the allowable number of exceedances (one). The 2-year 
period reduces the likelihood of a source being penalized for a 
violation that may be attributed to a one-time event. Aside from 
changes in terminology to make the language appropriate for a section 
303 program rather than a NAAQS, the proposed Appendix X is identical 
to the Appendix I to 40 CFR part 50 for interpreting the 5-minute NAAQS 
for SO2 that was proposed in the part 50/53 document. The EPA is 
soliciting comments on Appendix X.

V. Requirements Associated With New 5-Minute SO2 NAAQS

    The EPA proposed in the part 50/53 document a new primary 5-minute 
SO2 NAAQS which would be in addition to the 24-hour and annual 
primary SO2 NAAQS. Should this new 5-minute NAAQS be promulgated, 
EPA intends to initiate the targeted implementation strategy previously 
described to determine which areas are not meeting the new 5-minute 
NAAQS. In addition, EPA and the States will need to initially meet 
statutory requirements under sections 107 and 110. In general, these 
requirements are that the States must submit their initial suggested 
designations and statewide SIP's to EPA. Later, if areas are designated 
or redesignated to nonattainment, then EPA and the States must meet the 
requirements under section 172. The requirements under sections 107, 
110, and 172 of the Act are discussed in detail below. The rationale 
for any requirements which are discretionary, such as setting 
timeframes, or which need interpretation, are also discussed. Since the 
current annual, 24-hour, and 3-hour NAAQS are retained under this 
option, all existing requirements, such as SIP submittal and attainment 
dates, will remain in place as to the current NAAQS.

A. Targeted Implementation Strategy

    Should a new 5-minute NAAQS be promulgated, EPA intends to initiate 
the targeted implementation strategy previously described to determine 
which areas are not meeting the revised 5-minute NAAQS. And as 
described, the States should initially attempt to address any 
violations through compliance inspection and, if necessary, enforcement 
actions.
    Because of the modeling issues discussed previously (II.A.1.), the 
targeted implementation strategy relies principally on monitoring. The 
use of models is not advocated at this time for establishing section 
107 designations under a 5-minute SO2 NAAQS due to a lack of 
evaluation results concerning model performance, or defining the 
precision and bias of modeled 5-minute ambient SO2 concentrations. 
However, models may still be used under a new 5-minute SO2 NAAQS 
program for the following purposes:
    (1) Models may be useful as a tool for developing control 
strategies. When evaluating emissions from complex sources, they may 
provide information on the relative contributions to ambient SO2 
concentrations from various sources of emissions. Receptor modeling may 
be a useful tool for developing control strategies for complex sources. 
The use of tracers or ``tramp elements'' in association with these 
models would be needed for SO2 emission sources to determine 
source locations and relative contributions to ambient SO2 
concentrations.
    (2) Models can be and are recommended as a useful tool for 
evaluating the design of monitoring networks for a 5-minute SO2 
standard. They can provide useful information in a relative sense for 
determining points of maximum impact providing the characteristics of 
the emission source are not too complex or uncertain.

B. Designations--Section 107 1. Statutory Requirements

    The 1990 Amendments require EPA to promulgate designations, of 
areas for new or revised NAAQS. Section 107(d)(1)(A) of the Act 
requires States to submit designations, and section 107(d)(1)(B) 
requires EPA to promulgate designations of all areas (or portions 
thereof) with respect to new or revised NAAQS as nonattainment, 
attainment or unclassifiable. The specific requirements of section 
107(d)(1) (A) and (B) of the Act are described below. An area which is 
designated nonattainment is one that does not meet (or that contributes 
to ambient air quality in a nearby area that does not meet) the NAAQS 
for the pollutant. An area which is designated attainment is one which 
meets the NAAQS for the pollutant. An area which is designated 
unclassifiable is one that cannot be classified on the basis of 
available information as meeting or not meeting the NAAQS for the 
pollutant. Also, while section 107(d)(1) provides for States to submit 
a list of areas designated, it authorizes EPA to modify the 
designations submitted by the States. Once an area's initial 
designation is promulgated, any change in the designation status is 
accomplished pursuant to section 107(d)(3) of the Act.
2. Timeframe for Submittal of Designations by State
    As mentioned above, section 107(d)(1)(A) of the Act requires States 
to submit a list of all areas (or portions thereof) in the State 
designating them as nonattainment, attainment or unclassifiable for 
SO2. States must submit such list of areas (or portions thereof) 
in a timeframe EPA deems reasonable but not later than 1 year after the 
effective promulgation date of the new or revised NAAQS. The EPA cannot 
require the States to submit the list of areas in less than 120 days, 
however.
    The EPA intends to require that the initial SO2 designations 
be submitted not later than 1 year from the effective date of 
promulgation of the revised standard in order to allow the States as 
much time as possible to gather the necessary data to make the 
designation determinations. The EPA believes that, in most instances, 
areas will need to be initially designated unclassifiable due to lack 
of adequate ambient air monitoring data and the inability to rely on 
models for predicting 5-minute SO2 concentrations. By giving the 
maximum time allowed under the Act, States may have enough time to 
gather the data needed to make an adequate determination of an area's 
designation status. Nonetheless, EPA encourages States to submit 
designations sooner, wherever possible, in order to provide improved 
protection of public health. [[Page 12503]] 
3. Determining Initial Designation of an Area
    The EPA expects, in most instances, to initially designate areas as 
unclassifiable due to the lack of complete data or no data at all 
reported for 5-minute averaging time increments. Most of the existing 
ambient monitoring data are not reported for 5-minute averaging time 
increments, and EPA believes that those that are reported in this 
manner may not meet the data completeness criteria required by the 
proposed SO2 NAAQS (see discussion in revisions to CFR part 50, 
appendix I, published in the part 50/53 document). Revising the 
SO2 NAAQS to include an additional primary standard set at 5-
minute and 0.60 ppm necessitates that most ambient monitors be 
respanned to measure the higher concentration.
    In anticipation of a revised NAAQS, EPA has requested that the 
States respan monitors to begin measuring for higher concentrations. In 
these cases, EPA and States may have data to provide as a basis for 
initially designating an area as nonattainment.
    The EPA understands that in some instances States may want to 
request that certain areas be initially designated attainment for the 
revised SO2 NAAQS. An area will not be initially designated as 
attainment based solely on ambient monitoring data since no 
requirements have been issued to ensure complete data. Data 
completeness is a significant issue when trying to determine if an area 
is attaining the NAAQS as opposed to determining if an area is not 
attaining the NAAQS. However, areas with no SO2 sources as shown 
by their emission inventory would be likely candidates for an early 
attainment designation. Providing ambient air monitoring data does not 
indicate otherwise, EPA intends to designate an area as attainment if 
the State can show in its emissions inventory that the area does not 
contain any potential major source of SO2 as defined in the Act. 
This does not preclude the State or EPA from initially designating an 
area unclassifiable, if there is reason to believe there is an SO2 
source which may be causing a violation of the revised NAAQS in the 
area. The EPA believes this guidance gives reasonable assurance that 
the area is in attainment of the revised NAAQS. This does not prevent 
EPA or the State from redesignating an area, initially designated 
unclassifiable, to nonattainment at a later time should ambient air 
monitoring data indicate that the area is violating the NAAQS.
4. Determining the Boundaries of Designated Areas
    States should identify the boundaries of the nonattainment, 
attainment and unclassifiable areas when submitting designations for 
the revised SO2 NAAQS. In the absence of data or more specific 
boundary information, it may be more appropriate to define SO2 
nonattainment boundaries by the perimeter of the county in which the 
ambient SO2 monitor(s) recording the violation is located. 
Alternatively, it might be appropriate to define the nonattainment area 
using monitoring or other data to determine more specifically the 
geographic area that is nonattainment. In addition, if the ambient 
monitor measuring violations is located near a county boundary, then 
EPA recommends that the adjacent county also be designated as 
nonattainment for SO2. In some situations, however, a boundary 
other than the county perimeter may be appropriate. States may choose, 
alternatively, to define the SO2 nonattainment boundaries by using 
any one, or a combination, of the following techniques: (1) Qualitative 
analysis, (2) spatial interpolation of air monitoring data, (3) air 
quality simulation by dispersion modeling, or (4) saturation 
monitoring. If a State defines an SO2 nonattainment boundary using 
one of the methods above, EPA requires that it submit a defensible 
rationale for the boundary chosen with the Governor's request to 
designate the area.
    Boundaries for attainment areas can be drawn along current 
political boundaries if the State can show in its emissions inventory 
that the area does not contain any potential major source of SO2 
as defined in the Act, nor any of the sources listed in the previous 
section on determining the initial designation of an area.
    All areas of the State not designated attainment or nonattainment 
will be designated unclassifiable. The boundaries of the unclassifiable 
area will be the ``remainder of the State.''
5. Promulgation of Designations by EPA
    Section 107(d)(1)(B)(i) of the Act requires that EPA promulgate the 
designations submitted by States as expeditiously as practicable, but 
not later than 2 years from the date of promulgation of the revised 
SO2 NAAQS. This period may be extended for up to 1 year where EPA 
has insufficient information to promulgate the designations. The EPA 
may make any modifications deemed necessary to the areas (or portions 
thereof) submitted by the State (see generally section 107(d)(1)(B) of 
the Act). However, no later than 120 days before promulgating a 
modified area, EPA must notify the affected State and provide an 
opportunity for the State to demonstrate why any proposed modification 
is inappropriate.
    The EPA expects in many cases to require the full extension of 1 
year before promulgating the designations of many areas as allowed 
under section 107(d)(1)(B) of the Act. The full extension would be 
needed in these cases in order to allow States and EPA to respan or 
relocate monitors and collect complete ambient data to better ascertain 
the designation status of areas with monitors. Therefore, EPA generally 
intends to promulgate the initial area designations within 3 years from 
the effective date of promulgation of the revised SO2 NAAQS.
    Designations promulgated pursuant to section 107(d)(1) of the Act 
are exempt from the Administrative Procedures Act requirements for 
notice-and-comment rulemaking (5 U.S.C. section 553-557) (see section 
107(d)(2)(B) of the Act). Therefore, when EPA promulgates designations 
with respect to the revised SO2 NAAQS, it may or may not 
promulgate the designations through notice-and-comment rulemaking.
6. Failing to Submit Designations
    If the Governor of a State fails to submit the required SO2 
designations, in whole or in part, EPA is required to promulgate the 
designation that EPA deems appropriate for any area (or portion 
thereof) not designated by the State (see section 107(d)(1)(B)(ii) of 
the Act). The EPA will do so no later than 3 years after the date of 
promulgation of a new NAAQS.

C. State Implementation Plans (SIP's)

    Section 110(a) establishes the general requirements for SIP's. In 
addition, subparts 1 and 5 of part D of title I of the Act establish 
additional requirements concerning SIP's for areas designated 
nonattainment for SO2. These requirements concern the content of 
the SIP's, the applicable dates by which nonattainment areas must 
attain a new SO2 NAAQS, and the schedule for the submission of the 
SIP's.
1. General SIP Requirements--Section 110(a)
    All SIP's, regardless of whether they concern areas designated 
nonattainment or not, must meet the general SIP requirements of section 
110(a). Section 110(a)(1) provides that each state must submit a SIP to 
provide for the implementation, maintenance and enforcement of a 
primary NAAQS in [[Page 12504]] each air quality control region within 
the State (hereinafter referred to as ``statewide SIP's''). Section 
110(a)(2) sets forth the elements that a SIP must contain in order to 
be fully approved. These elements are discussed in the General Preamble 
for the Implementation of Title I of the Clean Air Act Amendments of 
1990 (57 FR 13556-13557).
2. General SIP Requirements--Section 110(a)(2).
    (a) Statutory and Existing Regulatory Requirements. Regulations for 
the preparation, adoption, and submission of SIP's under section 110 of 
the Act were initially published November 25, 1971 (36 FR 22369) and 
codified as 40 CFR part 51. The 40 CFR part 51 has been modified from 
time to time since then. On November 7, 1986 (51 FR 40656), EPA 
restructured and consolidated the 40 CFR part 51 regulations to make 
them easier to follow and revise in the future.
    The 1990 amended Act did not substantially change the SIP 
requirements in section 110(a)(2) of the Act. For the most part, EPA 
believes that the existing regulatory framework, i.e., 40 CFR part 51, 
defines the general section 110(a)(2) SIP requirements for SO2. 
However, as a result of a revised SO2 NAAQS, data handling 
practices, and specified SIP submittal timeframes in the Act, some 
revisions to 40 CFR part 51 are necessary. The specific revisions to 40 
CFR part 51 are discussed in another section entitled ``Regulatory 
Revisions.'' The EPA also notes that under section 193, anything in 
part 51 that is inconsistent with the 1990 Amendments is superseded 
even if EPA has not yet revised the regulations. A discussion of the 
statewide SIP requirements is provided below.
    (b) Statewide SIP's for the Revised SO2 NAAQS. For the most 
part, States have already adopted, as part of their overall SIP for 
current SO2 NAAQS, rules or regulations which satisfy the majority 
of the general SIP requirements in section 110(a)(2) of the Act and the 
existing 40 CFR part 51. At this time, the EPA does not envision that 
States will have to develop substantial new general regulations for the 
statewide SIP's for the revised SO2 NAAQS. The EPA will issue 
appropriate guidance in the future in the event that this assessment 
changes.
    There are two requirements, in particular, under section 110(a)(2) 
that must be met by the States upon promulgation of a revised SO2 
NAAQS. Section 110(a)(2)(B) requires the establishment and operation of 
appropriate ambient air monitoring systems, data from which must be 
made available to the Administrator upon request. Coupled with this is 
a requirement under section 110(a)(2)(E) that States have adequate 
resources and authority to implement the SIP.
    (c) New Source Review Issues. Section 110(a)(2)(C) of the Act 
requires States to protect the NAAQS by providing for the regulation of 
the construction and modification of stationary sources. In areas that 
are designated as attaining the NAAQS, as well as areas that are 
designated as unclassifiable under section 107 of the Act, each 
implementation plan must contain legally-enforceable requirements which 
enable the State to determine whether the construction or modification 
of stationary sources will interfere with maintenance of the NAAQS (see 
section 161 of the Act). For major stationary sources that locate in 
attainment or unclassifiable areas, the Act requires that comprehensive 
preconstruction review requirements under PSD of the air quality 
program contained in part C, title I, of the Act must be satisfied\1\ 
(e.g., sections 160-169 of the Act).

    \1\The statutory PSD requirements apply to new major stationary 
sources and modifications of existing major stationary sources. A 
``major stationary source'' is: (1) Any source from a statutory list 
of 28 source categories that emits, or has the potential to emit, 
100 tons per year (tpy) or more of a regulated pollutant; or (2) any 
other source that emits, or has the potential to emit, at least 250 
tpy of a regulated pollutant (see section 169(1) of the Act).
---------------------------------------------------------------------------

    The EPA has set forth SIP requirements at 40 CFR 51.166 containing 
the minimum requirements by which a State preconstruction review permit 
program will be considered to meet with the statutory requirements for 
PSD.\2\ In very broad terms, these requirements provide for the 
imposition of best available control technology at new and modified 
major stationary sources for each pollutant subject to regulation under 
the Act, and provide for review of the potential air quality impacts of 
such sources and modifications (e.g., section 165(a) of the Act).

    \2\The EPA has also promulgated regulations for a Federal PSD 
program at 40 CFR 52.21. The Federal program applies to States that 
do not have EPA-approved PSD programs as part of their SIP.
---------------------------------------------------------------------------

    The current PSD program requirements under 40 CFR 51.166, which 
protect the existing primary and secondary NAAQS for SO2, will 
also be protective of a new 5-minute SO2 NAAQS in that the 
regulations prevent the issuance of a PSD permit to a major source that 
would cause or contribute to a violation of any NAAQS (Sec. 51.166(k)). 
However, while no changes to the existing requirements are needed to 
ensure the new or modified PSD source must evaluate their ambient 
impacts against a new 5-minute standard for SO2, EPA has reviewed 
certain existing PSD provisions at Sec. 51.166 (and corresponding 
provisions at Sec. 52.21) to determine whether changes may be needed to 
ensure that a new 5-minute SO2 standard, as proposed in the part 
50/53 document, would be adequately protected.
    Several of the existing PSD provisions rely on Agency-prescribed 
significance levels to determine whether any pollutant that would be 
emitted by a new or modified major stationary source must undergo 
comprehensive permit review. First, EPA uses significant emissions 
rates (expressed in tons per year) to determine whether a regulated 
pollutant (other than a pollutant emitted in major amounts) to be 
emitted by a new or modified major stationary source must undergo PSD 
review3 (e.g., Sec. 51.166(b)(23)(i)).

    \3\The PSD review requirements apply to any regulated pollutant 
which a new or modified major stationary source would emit in 
significant amounts. Thus, a source may be ``major'' for only one 
pollutant, but PSD review would apply to other pollutants emitted in 
``significant'' amounts.
---------------------------------------------------------------------------

    Second, significant ambient impact concentrations are used to 
determine whether a source must undergo an impact analysis to show that 
it will not contribute to a violation of the NAAQS or PSD increments 
(Sec. 51.165(b)). Finally, significant monitoring concentrations are 
used to determine whether the reviewing authority may exempt a source 
from the ambient monitoring requirements for a particular pollutant 
(e.g., Sec. 51.166(i)(8)).
    As described below, the EPA examined each applicable significance 
level used for SO2 in order to determine whether a 5-minute 
standard for SO2 would necessitate any revisions to the existing 
levels. In each case, EPA has determined that sufficient information is 
not presently available to warrant any revision to the existing levels.
    The significant emissions rate for SO2 is currently defined as 
an emissions rate of 40 tpy or more under the PSD regulations. New or 
modified sources that would emit significant amounts of SO2 must 
undergo PSD review for that pollutant. Conversely, de minimis amounts 
of SO2 emissions are exempt from further review. The existing 
significance level for SO2 is based on the premise that an 
emissions rate that would result in ambient concentrations equaling at 
least 4 percent of the 24-hour primary standard should be considered 
significant (45 FR 52676, 52707-52708 (August 7, 1980)). In order to 
help determine whether the existing [[Page 12505]] significant 
emissions rate for SO2 would be appropriate, based on the same 
criteria, for the proposed 5-minute standard, EPA would need to predict 
the 5-minute concentration that results from a source emitting 40 tpy 
of SO2. The absence of an approved methodology for either directly 
modeling 5-minute SO2 concentrations or converting modeled 
concentrations of SO2 from a given averaging period (e.g., 3-hour, 
1-hour) to a 5-minute average precludes EPA from completing its 
analysis of the adequacy of the existing significant emissions rate. 
Should EPA adopt a 5-minute NAAQS for SO2, EPA will further study 
the need for revisions of the significant emissions rate.
    Because of the present difficulties associated with efforts to 
model 5-minute ambient concentrations of SO2, EPA has also 
determined that it would be inappropriate to establish a significant 
ambient impact level for a 5-minute SO2 NAAQS. In the event that 
adequate data and the appropriate performance evaluations become 
available to support the use of dispersion models to estimate 5-minute 
SO2 concentrations in the future, EPA will consider the 
establishment of a 5-minute SO2 significant ambient impact 
concentration.
    Under the existing regulations, the reviewing authority may exempt 
a proposed major stationary source from the PSD pre-application 
monitoring requirements (40 CFR 51.166(m)) if either the air quality 
impacts resulting from the source, or the existing ambient 
concentrations of the particular pollutant in the area of the source, 
are less than the prescribed significance level for that pollutant. For 
SO2, the significance level is 13 g/m\3\ (24-hour 
average). Since models are not available for a source to project its 
ambient impact for 5-minute averaging periods, EPA believes that 
consideration of a new significance level for SO2 based on a 5-
minute averaging time is not practical at this time. Instead, EPA 
proposes to continue using the existing 24-hour significance level in 
conjunction with the pre-application monitoring requirement at 40 CFR 
51.166(m). Thus, if a source finds that it must gather ambient data for 
SO2, based on ambient impacts and existing air quality 
concentrations exceeding the SO2 significance level, then the 
applicant will be required to gather 5-minute air quality data in 
addition to data for all other applicable averaging periods for 
SO2.
    As indicated in the preceding discussion, for several different PSD 
program elements, EPA proposes to retain existing SO2 significance 
levels instead of pursuing the possibility of revising the significance 
levels based on a new 5-minute SO2 NAAQS. The EPA requests the 
public's views about this proposed use of existing significance levels.
    The PSD program also includes specific air quality limitations, 
known as increments, which define maximum allowable increases in 
pollutant concentrations. These increments prevent unlimited increases 
in ambient pollutant concentrations beyond a determined baseline 
concentration for a particular area.4 Section 166 of the Act 
authorizes EPA to promulgate new increments within 2 years from the 
date of promulgation of new NAAQS. The existing PSD regulations include 
increments for SO2 for the 3-hour, 24-hour and annual averaging 
periods. The EPA will determine the need for a 5-minute increment for 
SO2, especially in light of the present difficulties which 
restrict the Agency's ability to use air quality dispersion models to 
determine the amount of increment that would be consumed by new and 
modified SO2 sources for a 5-minute averaging period. The EPA will 
also investigate the feasibility of developing and implementing 
alternatives to numerical air quality increments (expressed in 
g/m\3\), as authorized under section 166(d) of the Act. In any 
event, EPA will not propose new increments for SO2 until such time 
that a new 5-minute SO2 NAAQS is first promulgated.

    \4\The PSD areas (areas designated as attainment or 
unclassifiable under section 107 of the Act) are further categorized 
as Class I, II, or III areas (section 162 of the Act). Each of these 
classifications determines the ``maximum allowable increases'' or 
increment of air quality deterioration permissible (section 163 of 
the Act). Only a relatively small increment of air quality 
deterioration is permissible in Class I areas and consequently these 
areas are afforded the greatest amount of air quality protection. An 
increasingly greater amount of air quality deterioration is allowed 
in Class II and III areas.
    Air quality deterioration is measured from the date on which the 
first PSD application is submitted. This date becomes the baseline 
date after which any change in actual emissions affects the 
allowable increment. In all instances, however, the NAAQS represent 
the overarching air quality ceiling that may not be exceeded, 
notwithstanding any allowable increment.
---------------------------------------------------------------------------

    (d) Schedule for Submittal of Section 110(a)(1) SIP's. Section 
110(a)(1) states that the SIP's required by that subsection are to be 
submitted to EPA ``within 3 years (or such shorter period as the 
Administrator may prescribe) after the promulgation of a national 
primary ambient air quality standard (or any revision thereof) under 
section 109.'' Such SIP's are to provide for ``implementation, 
maintenance and enforcement'' of the new NAAQS. Section 110(a)(1), 
however, must be read in light of the timetable for designations of 
areas as nonattainment, attainment, or unclassifiable under section 
107(d)(1) described above, and the explicit timetables for SIP 
submissions for nonattainment areas under part D of title I. Section 
107(d)(1) provides that designations must occur within 3 years of the 
promulgation of a new NAAQS and the part D provisions (sections 172(b) 
and 191(a)) provide for the submission of SIP's meeting the 
requirements of section 172(c) within a specified time period following 
the designation of an area as nonattainment.
    The EPA believes that these provisions can best be harmonized in 
the context of a new 5-minute SO2 NAAQS by interpreting the 
section 110(a)(1) deadline as being satisfied by the submission of SIP 
elements whose content does not depend on the designation of an area. 
In the case of SIP's concerning a new 5-minute SO2 NAAQS, EPA 
believes that such submissions would be limited to SIP revisions 
concerning compliance with the monitoring requirements of section 
110(a)(2)(B) and the resource requirements of section 110(a)(2)(E). The 
EPA believes that, until a problem with maintaining a new 5-minute 
NAAQS is identified, it is reasonable to view the already-existing 
substantive SIP provisions as adequate and that it would be absurd to 
require areas to adopt additional control requirements or emission 
limitations prior to the identification of particular problem sources. 
The EPA notes that any areas designated nonattainment will be subject 
to further SIP submission deadlines requiring the submission of 
nonattainment area SIP's under part D of title I that satisfy the 
substantive requirements of section 172(c).
    Moreover, with respect to the monitoring and resource SIP elements, 
EPA believes that any changes to existing SIP's that would be needed 
will not be significant in terms of scope or effort. Indeed, some 
States may have to make minimal or no changes to their own rules in 
order to implement the new monitoring requirements. For this reason, 
and because the changes in monitoring requirements will assist in 
developing information about ambient air quality that will be relevant 
to designations, EPA is proposing that all States submit any needed SIP 
revisions within 1 year of final action on today's proposal.

D. Nonattainment Area Requirements

    Areas designated nonattainment must meet the SIP requirements of 
part D of title I as well as the requirements of section 110. The 
provisions of part D [[Page 12506]] pertinent to SO2 areas are 
those contained in subparts 1 and 5. These provisions have been 
described previously in the General Preamble for the Implementation of 
Title I of the Clean Air Act Amendments of 1990 (57 FR 13498), and the 
following discussion will focus on the requirements of particular 
relevance to the implementation of a new NAAQS.
1. Attainment and SIP Submittal Dates
    To determine the attainment dates and SIP submittal dates 
applicable to a new SO2 NAAQS, it is necessary to analyze the 
relationship of the relevant provisions of both subpart 1 and subpart 
5.
    The starting point for the analysis is section 172(a) in subpart 1. 
Section 172(a)(2)(A) provides that the attainment date for attaining a 
primary NAAQS is the date by which attainment can be achieved as 
expeditiously as practicable, but not later than 5 years from the date 
of designation under section 107(d). It also provides that EPA may 
extend the attainment date to the extent appropriate, for a period of 
up to 10 years after designation, considering the severity of the air 
quality problem and the feasibility and availability of pollution 
control measures. Section 172(a)(2)(D), however, provides that ``[t]his 
paragraph (paragraph (2)) shall not apply with respect to nonattainment 
areas for which attainment dates are specifically provided under other 
provisions of this part.'' This language therefore leads to the 
question of whether areas designated nonattainment with respect to a 
new SO2 NAAQS are areas for which attainment dates are provided 
elsewhere in part D of title I.
    As subpart 5 establishes attainment dates for certain SO2 
nonattainment areas, the issue is whether those provisions establish 
attainment dates for areas designated nonattainment with respect to a 
new SO2 NAAQS. Of particular relevance are sections 192(a) and 
191(a). Section 192(a) provides that SIP's required under section 
191(a) provide for attainment ``as expeditiously as practicable but no 
later than 5 years from the date of the nonattainment designation.'' 
Section 191(a) requires that ``[a]ny State containing an area 
designated or redesignated under section 107(d) as nonattainment with 
respect to the national primary ambient air quality standards for 
sulfur oxides, nitrogen dioxide, or lead subsequent to the date of the 
enactment of the Clean Air Act Amendments of 1990 shall submit to the 
Administrator, within 18 months of the designation, an applicable 
implementation plan meeting the requirements of this part.''
    One possible interpretation of the Act is that the language of 
section 191(a) applies to areas designated nonattainment with respect 
to a new SO2 NAAQS promulgated after the enactment of the 1990 
Amendments. If that interpretation is followed, section 192(a), rather 
than section 172(a)(2), would determine the attainment date for those 
areas. This is due to the language in section 172(a)(2)(D) providing 
that section 172(a)(2) does not apply to areas for which attainment 
dates are specifically provided elsewhere in part D. The language of 
section 191(a), rather than section 172(b), would also apply to the 
establishment of the SIP submittal date for nonattainment SIP's 
required to implement the new NAAQS. The consequence of this 
interpretation for the attainment deadline is that the 5-year 
attainment deadline of section 192(a) would apply, rather than the 5-
year deadline that can be extended to 10 years under certain conditions 
under section 172(a). As far as SIP submittal deadlines are concerned, 
section 191(a)'s 18-month deadline would apply rather than section 
172(b)'s 3-year deadline.
    An alternative interpretation is that the provisions of subpart 5 
were intended to apply only to attainment dates and SIP submittal 
deadlines concerning a NAAQS in existence at the time of the enactment 
of the 1990 Clean Air Act Amendments. Under this view, the general 
provisions of subpart 1 (i.e., sections 172(a)(2)(A) and 172(b)) would 
apply to the determination of attainment dates and SIP submittal 
deadlines pertaining to a new SO2 NAAQS promulgated after the 1990 
Amendments. The EPA notes, however, that it believes that an 18-month 
SIP submittal deadline would provide adequate time for the States to 
develop and submit their SIP's regarding a new NAAQS. It would also 
provide more time to implement the control strategy adopted in the SIP, 
which EPA believes is preferable. If the maximum period of 3 years were 
allowed, there would only be 2 years between the date of the submittal 
of the SIP and the 5-year attainment date, and even less time between 
EPA's final action regarding the approvability of the SIP's and the 
attainment date. Consequently, even if the provisions of section 172(b) 
were to apply to SIP submittal deadlines for a new NAAQS, EPA would 
require States to submit their SIP's within an 18-month timeframe 
pursuant to section 172(b)'s authority to establish a shorter period 
than the maximum 3-year period.
    The EPA requests comment on both of these interpretations and the 
consequences that they lead to regarding the establishment of 
attainment dates and SIP submittal deadlines for a new SO2 NAAQS.
2. Classifications--Section 172(a)(1)
    The classification provisions (section 172(a)(1)) give EPA the 
authority to classify nonattainment areas for the purposes of applying 
attainment dates (section 172(a)(2)(A)). In exercising this authority, 
EPA may consider such factors as the severity of the nonattainment 
problem or the availability and feasibility of the pollution control 
measures. Based upon the classification, EPA may set later attainment 
dates for areas with more severe air quality problems (section 
172(a)(2)(A)).
    At the present time, EPA does not intend to establish a 
classification scheme for areas which violate the new 5-minute SO2 
NAAQS. Currently the SO2 program does not have a classification 
scheme since, typically, within the SO2 program the severity of 
the SO2 ambient air quality is not a factor in attaining the NAAQS 
once the needed control measures are put in place. The EPA believes 
that in most of the areas designated nonattainment for the new 5-minute 
NAAQS, the cause of the high SO2 concentrations (usually a single 
source) will be obvious. While the method of controlling these 
emissions may not be as obvious, the control measure should result, in 
most cases, in a single step correction of any future violations. 
Consequently, EPA does not believe a classification scheme is necessary 
or appropriate.
3. Nonattainment Plan Provision--Section 172(c)
    Section 172(c) lists the requirements to be met by a nonattainment 
SIP. Some of those requirements are discussed below in the context of a 
SIP submittal for a SO2 NAAQS nonattainment area.
    a. Statutory and Existing Regulatory Requirements. As previously 
indicated, regulations for the preparation, adoption, and submission of 
SIP's were initially published November 25, 1971 and codified as 40 CFR 
part 51. The 40 CFR part 51 has been modified from time to time since 
then. However, the most current guidance on how EPA intends to 
interpret the 1990 Amendments is found in the General Preamble (57 FR 
13498, April 16, 1992).
    The 1990 Amendments added section 172(c) which prescribes the 
nonattainment SIP requirements. To the extent that the existing SIP 
regulations that have been codified in 40 CFR parts 51 and 52 do not 
conflict with section 172(c), EPA will rely on them to carry out the 
requirements of section 172(c). [[Page 12507]] As necessary EPA will 
adopt new or modify existing regulations to carry out other provisions 
of section 172(c). For further information on potential changes to 40 
CFR part 51 with respect to SO2, see the separate section entitled 
``Regulation Revisions.'' Also, as noted earlier under section 193, 
anything in part 51 that is inconsistent with the 1990 Amendments is 
superseded even if EPA has not yet revised the regulations.
    b. Reasonably Available Control Measures (Including Reasonably 
Available Control Technology). Section 172(c)(1) requires SIP's to 
``provide for the implementation of all reasonably available control 
measures (RACM) as expeditiously as practicable (including such 
reductions in emissions from existing sources as may be obtained 
through the adoption, at a minimum, of reasonably available control 
technology (RACT)) and shall provide for attainment of the national 
primary ambient air quality standards.'' Historically, EPA has defined 
RACT as ``the lowest emission limit that a particular source is capable 
of meeting by the application of control technology that is reasonably 
available considering technological and economic feasibility (Strelow, 
1976).'' In the case of a new 5-minute SO2 NAAQS, EPA believes 
that RACT should be interpreted in accordance with EPA's long-standing 
interpretation.
    The EPA notes that, as the sources of any violations of a new 
SO2 NAAQS should be readily identifiable, there should not be any 
questions about the identity of the sources to which RACT should be 
applied. Thus, in the case of a new SO2 NAAQS, compliance with 
EPA's general recommendation that available control technology be 
applied to those existing sources in the nonattainment area that are 
reasonable to control in light of the attainment needs of the area and 
the feasibility of such controls should be readily achieved (EPA 1992c, 
n. 20, 57 FR 13541).
    While a plan must require the implementation of RACM needed to 
attain within the statutory timeframes, it need not require the 
adoption of all available control measures if it demonstrates 
attainment as expeditiously as practicable without the adoption of all 
measures. The EPA believes it would be unreasonable to require that a 
plan which demonstrates attainment include all technologically and 
economically available control measures if such measures would not 
expedite attainment. Thus, it is possible that some available control 
measures may not be ``reasonably'' available, and not required by RACM, 
because their implementation would not expedite attainment (EPA 1992c, 
57 FR 13543).
    In addition to available control technology that should be fully 
considered in identifying RACT for purposes of the current SO2 
NAAQS, RACT for purposes of a new 5-minute NAAQS would also include 
consideration of maintenance and process operating procedures at 
SO2 sources that will achieve the new NAAQS within the statutory 
timeframes. The EPA believes that such available control measures 
should be fully assessed, in light of the general guidance above, in 
determining RACM (including RACT) for purposes of implementing a 5-
minute SO2 NAAQS.
    c. Emission Inventory. Section 172(c)(3) states that the SIP shall 
include a comprehensive, accurate, current inventory of actual 
emissions from all sources of SO2 in the nonattainment area and 
that EPA may require periodic revisions of the inventory as determined 
necessary to assure that the requirements of part D are met. Typically 
for most nonattainment areas, determining the nature and extent of 
specific control strategies needed requires an emissions inventory. 
Also, typically, an emission inventory should be based on measured 
emissions or documented emission factors. The more comprehensive and 
accurate the inventory, the more effective the control evaluation.
    However, in terms of a new 5-minute NAAQS, measured emissions or 
emission factors for the probable sources of 5-minute NAAQS 
exceedances, process upsets, equipment malfunctions, batch processes, 
startup/shutdown, and fugitive emissions, are almost nonexistent. It is 
anticipated that most nonattainment areas for the 5-minute SO2 
NAAQS will be defined by a single source as measured by a monitor or 
monitors close to the source. Thus, in most cases, the part D SIP for a 
nonattainment area will fulfill the inventory requirements of section 
172(c)(3) by identifying the source around which the monitors were 
located and which may have caused the monitored problem. In situations 
where it is technically feasible, emission estimates should be made 
using emission measurements or factors.
    d. Control Strategy Demonstration. The EPA has historically 
required dispersion modeling for setting emission limits. However, 
because of the limitations of models in predicting 5-minute 
concentrations, other methods may have to be used. Control strategy 
demonstrations may have to rely on monitors as evidence of adequacy of 
the implemented emission reductions as being protective of the 5-minute 
NAAQS. In certain cases, the monitors may be used for setting the 
emission limits. The EPA intends to rely on section 11.2.2 of the 
Modeling Guideline which addresses requirements for using monitoring 
networks to set emission limits.
    e. Reasonable Further Progress. As stated in the General Preamble 
(57 FR 13547), section 171(l) of the amended Act defines reasonable 
further progress as ``such annual incremental reductions in emissions 
of the relevant air pollutant as are required by this part (part D) or 
may reasonably be required by EPA for the purpose of ensuring 
attainment of the applicable national ambient air quality standard by 
the applicable date.'' This definition is most appropriate for 
pollutants which are emitted by numerous and diverse sources, where the 
relationship between any individual source and the overall air quality 
is not explicitly quantified, and where the emission reductions 
necessary to attain the NAAQS are inventorywide. The definition is 
generally less pertinent to pollutants such as SO2, particularly 
for the proposed new NAAQS, which usually have a limited number of 
sources, relationships between individual sources and air quality which 
are relatively well defined, and emissions control measures which 
result in swift and dramatic improvement in air quality. That is, for 
SO2, there is usually a single ``step'' between pre-control 
nonattainment and post-control attainment.
    Therefore, for a new 5-minute SO2 NAAQS, with its discernible 
relationship between emissions and air quality and significant and 
immediate air quality improvements, RFP will continue to be construed 
as ``adherence to an ambitious compliance schedule.''5 The 
compliance schedule for a new 5-minute NAAQS could consist of 
implementation of a maintenance program where the source of emissions 
is due to frequent malfunction of a control device. The SIP's which 
require RFP as just described for an SO2 nonattainment area will 
be considered as meeting the requirements of section 172(c)(2).

    \5\U.S. Environmental Protection Agency, Office of Air Quality 
Planning and Standards, ``Guidance Document for Correction of Part D 
SIP's for Nonattainment Areas,'' (Research Triangle Park, North 
Carolina, January 27, 1984), page 27.
---------------------------------------------------------------------------

    f. Permits for New and Modified Major Stationary Sources. Section 
172(c)(5) of the Act states that the SIP shall require permits for the 
construction and operation of new or modified major stationary sources 
(i.e., stationary [[Page 12508]] sources which emit or have the 
potential to emit at least 100 tpy of any nonattainment pollutant or 
lesser amounts in certain nonattainment areas) anywhere in a 
nonattainment area, in accordance with section 173 of the Act.6 In 
nonattainment areas, a presumption exists that emissions increases 
resulting from new and modified major stationary sources will adversely 
affect the area; thus, in lieu of a complete air quality impact 
analysis (including ambient monitoring), emissions reductions (offsets) 
from existing sources must be obtained in order to mitigate the ambient 
impacts resulting from the potential emissions from the proposed new 
source, or net emissions increase from a proposed major modification to 
an existing source (e.g., section 173(c) of the Act).

    \6\For purposes of the nonattainment NSR requirements under part 
D of title I of the Act, ``major stationary source'' is defined as 
any stationary source which emits, or has the potential to emit, 100 
tpy (or lesser amounts in certain nonattainment areas) of any 
nonattainment pollutant (see, e.g., sections 182(c-e), 189(b)(3), 
and 302(j) of the Act).
---------------------------------------------------------------------------

    Under the nonattainment NSR program (40 CFR 51.165(a)), EPA uses 
significant emissions rates (expressed in tons per year) for pollutant 
applicability purposes to determine whether a modification of an 
existing major stationary source will result in a significant net 
emissions increase (Sec. 51.165(a)(1)(x)). For the same reasons 
described in section V.C of this preamble, EPA does not now intend to 
propose to revise the significant emissions rate for SO2 
commensurate with the 5-minute SO2 NAAQS proposed in the part 50/
53 document. Public comment is requested as to whether the existing 40 
tpy significant emissions rate needs to be revised if EPA promulgates 
the proposed 5-minute SO2 standard.
    Major new or modified sources locating in the nonattainment area 
will be required to meet the lowest achievable emission rate, obtain 
emissions offsets, and satisfy other applicable requirements under 
section 173 of the Act. With implementation of a new 5-minute NAAQS, 
these requirements may be addressed by existing permit programs for 
those areas already designated nonattainment for SO2 and meeting 
the nonattainment NSR requirements under section 173 of the Act. 
However, for those States without the appropriate nonattainment NSR 
program, the State would need to develop and implement such a program 
for any newly designated nonattainment areas resulting from a new 5-
minute NAAQS for SO2.
    g. Contingency Measures. Section 172(c)(9) of the amended Act 
defines contingency measures as measures that become effective without 
further action by the State or EPA, upon determination by EPA that the 
area has failed to: (1) Make reasonable further progress, or (2) attain 
the SO2 NAAQS by the applicable statutory deadline.
    For current SO2 programs, EPA interprets ``contingency 
measures'' to mean that the State agency has a comprehensive program to 
identify sources of violations of the SO2 NAAQS and to undertake 
an aggressive followup for compliance and enforcement, including 
expedited procedures for establishing enforceable consent agreements 
pending the adoption of revised SIP's. The rationale for this 
interpretation as presented in the General Preamble (57 FR 13547) is 
the following. The EPA interprets the contingency measure provisions as 
primarily directed at general programs which can be undertaken on an 
areawide basis. First, for some criteria pollutants, the analytical 
tools for quantifying the relationship between reductions in emissions 
and resulting air quality improvements remain subject to significant 
uncertainties, in contrast with procedures for pollutants such as 
SO2 and its current NAAQS. Second, emission estimates and 
attainment analyses can be strongly influenced by overly optimistic 
assumptions about control efficiency and rates of compliance for many 
small sources. In contrast, controls for the current SO2 NAAQS are 
well understood and are far less prone to uncertainty. Since SO2 
control measures are by definition based upon what is directly and 
quantifiably necessary to attain the SO2 NAAQS, it would be 
unlikely for an area to implement the necessary emissions control yet 
fail to attain the NAAQS.
    However, for the proposed 5-minute SO2 NAAQS, EPA will need to 
interpret requirements for contingency measures different from those 
for the current NAAQS, due to the nature of sources and emissions that 
EPA considers likely to cause violations. As opposed to the current 
NAAQS, which can rely on dispersion models to predict attainment of the 
NAAQS, the State and Local agencies cannot reliably predict that 
attainment will be achieved even with proper implementation of a 
control program. It is possible that even with the control equipment 
operating properly, violations may persist. In other words, there may 
be overly optimistic assumptions about control efficiencies and 
emission rates. Therefore, contingency measures for the proposed 5-
minute NAAQS will require more than aggressive follow-up for compliance 
and enforcement as allowed for the current SO2 NAAQS. As an 
example, if the cause of the SO2 violations is due to control 
equipment failure, a SIP may require a more rigorous maintenance 
schedule. If further violations occur due to continued failures of the 
control equipment, then the contingency measures may need to invoke a 
more frequent inspection/maintenance program of the control equipment 
or even installation of backup control equipment.

E. SIP Processing Requirements

1. SIP Completeness
    Section 110(k)(1) required EPA to promulgate minimum criteria that 
any SIP submittal must meet. The EPA proposed an initial set of 
completeness criteria at 56 FR 23826 (May 24, 1991) and finalized them 
at 56 FR 42216 (August 26, 1991). Those notices describe the procedures 
for assessing whether a SIP submittal is complete and, therefore, 
adequate to trigger the Act requirement that EPA review and take action 
on the submittal. The completeness criteria provide a procedure and 
criteria that enable States to prepare adequate SIP submittals and 
enable EPA reviewers to promptly screen SIP submittals, identify those 
that are incomplete, and return them to the State for corrective action 
without having to go through rulemaking. The EPA intends to use the 
completeness criteria as amended in 40 CFR part 51, appendix V, to 
determine completeness of SIP submittals as required under section 
110(k)(1)(B).
2. Approval/Disapproval of Plan
    The Act as amended in 1990 allows for EPA to make full and partial 
approvals and disapprovals under section 110(k)(3) and conditional 
approvals under section 110(k)(4) of SIP submittals. In meeting the 
requirements under section 110(k)(3) and (4), EPA intends to follow the 
guidance for processing SIP submittals issued in the memo from Calcagni 
to the Regional Air Division Directors dated July 9, 1992.
3. Sanctions and Other Consequences of SIP Deficiencies
    The EPA intends to use sanctions consistent with the following 
stated policies and regulations as provided for by the Act in sections 
110(m) and 179 for the imposition of sanctions in the event that EPA 
finds that a State did not make a required SIP submission (in whole or 
in part), finds that a State did not submit a complete submission, 
[[Page 12509]] disapproves in whole or in part a required submission, 
or finds that any part of an approved SIP is not being implemented. 
Section 179(a) provides for the imposition of mandatory sanctions 
unless the deficiency identified by EPA (e.g., the failure to submit or 
disapproval) is corrected within 18 months. Moreover, section 110(m) 
provides EPA with the discretionary authority to impose sanctions at 
any time after a finding, disapproval or determination under section 
179(a).
    With respect to mandatory sanctions, section 179(a) provides that 
unless the State corrects the deficiency within 18 months, one of the 
two sanctions referred to in section 179(b) (i.e., highway or offset 
sanctions) shall be selected by EPA and will apply until EPA determines 
that the State has come into compliance. (In the case of a finding of 
failure to submit a required SIP revision, the sanctions would not be 
lifted until EPA determines that the State has submitted a SIP revision 
that satisfies the completeness criteria.) If 6 months after the 
imposition of the first sanction the State still has not corrected the 
deficiency, then the second sanction shall apply as well. If EPA finds 
a lack of good faith on the part of the State, then both the highway 
and offset sanctions are applied 18 months after the finding or 
disapproval.
    The EPA has discussed in detail issues concerning the imposition of 
sanctions in a number of Federal Register notices. The criteria for 
imposing discretionary sanctions on a statewide basis are discussed in 
a February 11, 1994 Federal Register notice, Criteria for Exercising 
Discretionary Sanctions Under Title I of the Clean Air Act (59 FR 
1476), and are codified at 40 CFR 52.30. The preamble to this notice 
also sets forth EPA's policy with respect to section 110(m) sanctions. 
Mandatory sanctions were discussed in a October 1, 1993 proposal (58 FR 
51270) and in the August 4, 1994 final rule (59 FR 39832) selecting the 
order of mandatory sanctions under section 179. That final rule does 
not apply to State failures to respond to SIP calls. The EPA intends to 
address sanctions for such failures in a future rulemaking.
    Apart from sanctions under sections 110(m) and 179(b), other 
consequences may also attach to a failure to comply with the Act's SIP 
submission or implementation requirements. First, section 179(a) 
authorizes EPA to withhold all or part of section 105 grants for air 
pollution control planning and control programs. Second, section 
110(c)(1)(B) provides that within 2 years of a finding that a State has 
failed to make a required submittal, a finding that a required 
submittal was not complete, or a disapproval of a submission (in whole 
or in part), EPA shall promulgate a FIP unless EPA approves a submitted 
SIP that corrects the deficiency. In support of this requirement, EPA 
intends to use its authority to withhold all or part of section 105 
grants to develop and implement FIP's where a State fails to comply 
with the Act's SIP submission or implementation requirements.

VI. Significant Harm Levels and Episode Criteria

    In a notice published in the Federal Register on April 26, 1988 (53 
FR 14926), in which the EPA proposed not to revise the SO2 NAAQS, 
the EPA at the same time proposed to revise the significant harm levels 
for SO2. Since final action was never taken on that proposal, EPA 
is reproposing to revise the 24-hour significant harm levels.
    Section 303 of the Act authorizes the Administrator to take certain 
emergency actions if pollution levels in an area constitute ``an 
imminent and substantial endangerment to public health or welfare, or 
the environment.'' The Act and EPA's regulations governing adoption and 
submittal of SIP's (section 110(a)(2)(G) and 40 CFR 51.16 and subpart H 
of part 51) require States to adopt contingency plans to prevent 
ambient pollutant concentrations from reaching specified significant 
harm levels and to take additional abatement actions if such levels are 
reached. The existing significant harm levels (40 CFR 51.16a) for 
SO2 were established in 1971 (36 FR 24002, November 21, 1971) at 
the following levels: SO2 alone--1.00 ppm (2620 g/m\3\) 
24-hour average of SO2; and SO2  x  tsp--490  x  103 
(g/m\3\) 2--24-hour average product of SO2 and tsp 
concentrations.
    On the basis of EPA's reassessment of the data upon which these 
levels were based and its assessment of more recent scientific evidence 
on sulfur oxides and particulate matter, EPA proposes to revise the 
significant harm levels for SO2.
    In actions related to the revisions of the particulate matter 
standards, EPA has already eliminated the combined tsp/SO2 
significant harm level (52 FR 24672, July 1, 1987). In doing so, EPA 
left open the possibility of reinstating an SO2/PM-10 significant 
harm level, if necessary for additional protection against SO2 
effects, at the conclusion of the SO2 review. The scientific data 
suggest that SO2 in combination with high levels of particulate 
matter have been associated with increases in daily mortality. The 
final 24-hour PM-10 significant harm level of 600 g/m3 
takes this potential interaction into account. Addition of a combined 
SO2/PM-10 significant harm level therefore appears unnecessary.
    Removal of the combined significant harm level raises the question 
as to whether the remaining SO2 significant harm level is 
sufficient. The possibility that SO2 alone or in combination with 
other pollutant or fog droplets may be in part responsible for the 
effects associated with 24-hour exposures suggests the need to continue 
a 24-hour significant harm level for SO2 alone at a substantially 
lower concentration. The EPA's assessment of studies of daily mortality 
(EPA, 1986a, Table 1 and EPA, 1986b Table 4-2) indicates greatest 
certainty of some increased daily mortality associated with high 
particle concentrations in combination with SO2 levels at or above 
750 g/m\3\ (0.29 ppm) for 24-hours. Accordingly, EPA proposes 
to revise the 24-hour SO2 significant harm level from 1.0 (2,620 
g/m3) to 0.29 ppm (750 g/m\3\).
    Appendix L to part 51 contains example air pollution episode levels 
and example contingency plans for the purpose of preventing air 
pollution from reaching the significant harm levels prescribed in 
section 51.151. The examples in appendix L serve as guides to States 
for the development of their own contingency plans. To conform with the 
proposed revisions to the significant harm level for SO2, certain 
changes to appendix L are required. The EPA proposes the following 
revisions to the example 24-hour episode levels for SO2:
    (1) That the example alert level for SO2 be changed from 800 
g/m3 to 0.19 ppm (500 g/m3), 24-hour 
average.
    (2) That the example warning level for SO2 be changed from 
1600 g/m3 to 0.23 ppm (600 g/m3), 24-hour 
average.
    (3) That the example emergency level for SO2 be changed from 
2100 g/m3 to 0.26 ppm (675 g/m3), 24-hour 
average.
    The basis for changing the episode levels for SO2 is the same 
as discussed above for the revisions to the significant harm level. 
With respect to example episode levels, the proposed alert level 
reflects the upper bound of the 24-hour range of interest for the NAAQS 
presented in the staff paper addendum (EPA, 1986b, Table 2). The staff 
paper concludes that at or above 0.19 ppm (500 g/m3) for 
24 hours, health effects are likely to occur in certain sensitive 
population groups (EPA, 1982a, page 72). Therefore, it would be 
appropriate under the episode criteria to initiate first stage control 
action when this ambient level of SO2 occurs. The proposed 24- 
[[Page 12510]] hour warning and emergency levels are set at increments 
between the proposed alert level and the proposed significant harm 
level. This approach would provide opportunity for the control actions 
associated with each episode level to take effect before the next stage 
is triggered and additional control actions become necessary. This 
proposal, if adopted, would change the 24-hour significant harm level. 
Therefore, States would be required to adopt the new numerical level, 
to evaluate the emergency episode provisions, in their current SIP's 
and any permits containing such provisions and to make any revisions 
necessary to assure their adequacy.
    All public comments on the proposed significant harm level and 
episode criteria will be considered by the Agency as it makes a 
decision on the final significant harm level.

VII. Proposed Revisions to Part 58 Monitoring Regulations

    The proposed revisions to 40 CFR part 58 are needed to allow States 
to reduce in most cases the number of NAMS SO2 monitors in the 
metropolitan areas. This, in turn, will free up monitors and resources 
that can be used toward the SO2 targeted implementation strategy. 
The following preamble details requirements which will be implemented 
regardless of the regulatory alternative that is ultimately selected 
for part 50.

A. Section 58.1  Definitions

    The number of SO2 monitors in the revised NAMS network for 
major metropolitan areas will be based on factors including population, 
historical ambient concentration measurements, and total SO2 
emissions. The SO2 emissions data are available from the AIRS for 
each county and for each consolidated metropolitan statistical area/
metropolitan statistical area (CMSA/MSA). Therefore, the requirements 
for NAMS SO2 stations have been determined on a CMSA/MSA basis, 
and the requirements for SLAMS SO2 stations have been determined 
on a county basis. Definitions are added for CMSA and MSA as provided 
by the U.S. Census Bureau.

B. Appendix C--Ambient Air Quality Monitoring Methodology

    As explained in a related notice in this issue of the Federal 
Register that proposes amendments to part 53, continuous ambient air 
monitoring analyzers designed to obtain 1-hour average SO2 
concentration measurements may not provide accurate 5-minute average 
concentration measurements. That notice proposes special supplemental 
performance specifications applicable to continuous SO2 analyzers 
that would be used for 5-minute monitoring so that the average SO2 
concentration measurements would be accurate. A companion amendment to 
appendix C of part 58 is needed to specifically require the use of 
these specially approved analyzers for 5-minute monitoring in SLAMS 
monitoring networks. Accordingly, a new section 2.4 is proposed to 
require that monitoring methods used for 5-minute average SO2 
measurements meet the special supplemental specifications proposed to 
be added to part 53.

C. Appendix D--Network Design for State and Local Air Monitoring 
Stations (SLAMS) and National Air Monitoring Stations (NAMS)

    Appendix D is being revised to change the NAMS requirements for 
SO2 monitors. The present requirements are based on measuring 
population exposure over a large area without being unduly influenced 
by point sources. Because concentrations at a significant number of 
these sites have decreased over time and many are measuring 
concentrations well below the current SO2 NAAQS, EPA believes that 
they may be put to better use if relocated. The monitors which may be 
moved could be used to complete the minimum NAMS and SLAMS requirements 
or to implement the targeted monitoring strategy for point sources of 
SO2 emissions described earlier in this notice (section II: 
Targeted Implementation Strategy). Up to three SO2 monitors would 
be required for each metropolitan area for trends purposes and general 
urban air quality analyses. The new number of NAMS monitors required 
for each metropolitan area would be based on the combination of 
population and SO2 emissions, as defined in the Air Facility 
Subsystem of AIRS and other information. The EPA solicits comments on 
reducing the requirements for the number of population-oriented NAMS 
SO2 monitors in the metropolitan areas.
    In addition to changing the criteria for the required number of 
NAMS monitors as noted above, new criteria are being included for a 
minimum number of SLAMS SO2 monitors for those counties (or parts 
of counties) not a part of any CMSA/MSA but with significant SO2 
emissions. These counties with SO2 emissions greater than 20,000 
tons/year, as defined in the Air Facility Subsystem of AIRS, would be 
required to have one to two monitors. However, EPA is proposing a 
provision which would allow for a waiver of all (or part of) these 
monitoring requirements after a 2-year monitoring period in accordance 
with EPA guidelines for network review for source-oriented SO2 
monitoring in nonurban areas. Although these guidelines have not been 
developed at this time, EPA solicits comments on the waiver provision 
criteria to be established and included in the guideline as well as the 
minimum number of years for data collection. The EPA also solicits 
comments on the requirement for SO2 SLAMS monitors in these areas.
    As discussed earlier in this notice, EPA believes there are a 
significant number of sources of SO2 emissions which can produce 
high 5-minute ambient concentrations of SO2. These 5-minute 
concentrations have the potential to exceed the level for a proposed 5-
minute SO2 NAAQS or the trigger level which may be established 
under the authority of section 303 of the Act. The sources which are 
believed to provide these high concentrations would be targeted for 
monitoring as discussed earlier in this notice. States will be required 
to prepare a targeted SO2 monitoring plan containing a listing of 
sources to be monitored, the schedule for monitoring, and the rationale 
for selecting the sources. The schedule for monitoring should be as 
expeditious as practicable. It is expected that the resources which are 
made available by the reconfiguration of the NAMS and SLAMS networks 
will be used to implement the targeting strategy around selected 
SO2 sources. The targeted SO2 monitoring plan will be 
reviewed as part of the annual network review.
    The number of SO2 monitors to be used around the targeted 
sources depends on several diverse factors, i.e. quantity of SO2 
emissions, meteorology, terrain, stack height and diameter of stack, 
temperature and velocity of stack emissions, distance from point of 
emissions to fence line and populated areas, batch operations, etc. To 
capture high peak 5-minute concentrations may require many monitors 
around the sources (Sonoma Technology Inc., 1994). However, it is not 
economically feasible to place enough monitors around the source to 
capture all potential exceedances of the NAAQS or trigger level. 
Therefore, EPA is using a more moderate approach on the number of 
monitors required.
    The EPA is proposing a minimum requirement of four SO2 
monitors to measure 5-minute, 3-hour, 24-hour, and annual average 
SO2 concentrations around the targeted sources. These monitors 
could be point SO2 monitors, open path SO2 analyzers, or a 
combination of both. If open path [[Page 12511]] analyzers with 
multiple monitoring paths are used, each monitoring path could 
potentially be substituted for one point SO2 monitor. Modeling, 
and perhaps saturation monitoring (a short term study involving the use 
of portable monitors deployed around the source), could be used to 
determine the area of expected maximum concentration based on the most 
predominant wind direction. One monitor would be placed at the fence 
line downwind of the predominant wind direction. A second monitor would 
be placed in the modeled maximum concentration area based on the 
predominant wind direction. Since wind directions around an SO2 
source may be significantly different from one season to another, this 
same procedure would be repeated for the second most frequent wind 
direction. For some cases, two or more of these locations may coincide 
and thereby reduce the number of monitors, or allow for a State or 
local agency to locate sites in alternative locations. In other cases, 
additional monitors would probably be needed for situations of complex 
terrain and/or meteorology. The EPA also encourages the use of open 
path SO2 analyzers in combination with point SO2 monitors to 
obtain better spatial coverage around the targeted sources. One open 
path SO2 analyzer using multiple monitoring paths could 
potentially replace several of the point SO2 monitors, depending 
on factors such as meteorology, terrain, and obstructions. Open path 
analyzers may be particularly useful in assessing ambient SO2 
concentrations over large populated areas, such as parks and recreation 
centers, where people are expected to jog/exercise. The EPA solicits 
comments on the location, number and type of SO2 monitors, the 
various available monitoring technologies, and the need to waive 
minimum monitoring requirements.
    The concentration gradients are expected to be sharper around these 
targeted sources of SO2 emissions. As a result, the SO2 
monitors located to measure population exposures over a wide area are 
unlikely to adequately characterize these peaks. Therefore, appendix D 
is being revised to allow the use of microscale SO2 sites for 
SLAMS monitors, and to encourage middle/neighborhood scale measurements 
as appropriate in populated areas near these targeted sources. The 
microscale measurements for SO2 would represent concentrations 
over an area ranging from several meters to up to about 100 meters. The 
EPA solicits comments on the use of micro, middle, and neighborhood 
scale monitors, both point monitors and/or open path analyzers, around 
point sources of SO2 emissions.
    The EPA is also proposing that the SO2 monitors around these 
targeted sources of SO2 emissions be classified as SLAMS monitors. 
Section 2.3 requires that monitoring be performed for a minimum of 2 
years. After that time, a decision should be made during the annual 
network review as to whether the monitoring should be continued around 
the targeted source, or the monitors redeployed around a different 
targeted source based on measured concentration levels, changes in 
plant process operations, etc. The EPA solicits comments on the SLAMS 
classification of the SO2 monitors around the targeted sources and 
a waiver provision to relocate the monitors before the full 2 years 
based on a review of the data.
    With this proposal, EPA is also requiring the collection of 5-
minute SO2 concentrations at the targeted sites. The EPA solicits 
comment on the need to require 5-minute concentrations at NAMS or other 
SLAMS sites, and if supplementary criteria should be considered for 
this additional request (e.g., require 5-minute SO2 monitor data 
if 1-hour concentration exceeds some level).

D. Appendix F--Annual SLAMS Report

    A proposed revision to section 2.1.1 of appendix F would reword 
this section to provide greater clarity and add a requirement to report 
the number of 5-minute hourly maximum observations. Section 2.1.2 would 
similarly be reworded for clarity and to require that the 24-hour 
averages reported in the annual report for SO2 be based on block 
(midnight to midnight) averaging periods and the 3-hour averages also 
to be based on block averaging periods. Reporting of the number of 
values in specified ranges of 24-hour average concentrations would be 
deleted because of new revisions to 40 CFR part 58 data reporting 
requirements.
    Reporting of 5-minute hourly maximums would also be added. The EPA 
solicits comments on the need for reporting additional summary data if 
a multiple exceedance form of the standard is adopted.

E. Appendix G--Air Quality Index Reporting and Daily Reporting

    The EPA proposes to revise the SO2 ambient concentrations 
contained in Tables 1 and 2 and in Figure 3 to correspond to the 
proposed new episode criteria and significant harm levels.

VIII. Transition Issues

    Since the existing NAAQS would be retained even if a 5-minute NAAQS 
is promulgated, all existing requirements and attainment dates will 
remain in place as to the existing NAAQS.

IX. Other Clean Air Act Amendment Authorities Affecting SO2 
Sources

    The EPA is also developing a voluntary program as part of the acid 
rain program to encourage nonutility sources to reduce their emissions 
of SO2. The voluntary entry into the acid rain program, known as 
the opt-in program, allows nonaffected sources (nonaffected under title 
IV), the opportunity to receive their own allowances, undertake 
emission reductions and trade the extra allowances they would no longer 
need for compliance with the acid rain program. Again, such 
participating sources would be under the same obligations to meet all 
other air regulatory requirements.
    These nonutility sources that could participate in the opt-in 
program are the same group of sources of concern for establishing a 5-
minute SO2 NAAQS. Assuming entry occurred prior to the imposition 
of the 5-minute standard, the source could accelerate its emissions 
reductions and offset the cost of such reductions through participation 
in the opt-in program. The EPA believes the development of options for 
a 5-minute SO2 standard and the opt-in program protects public 
health and provides an opportunity for cost reduction.

X. Public Participation

A. Comments and the Public Docket

    The EPA welcomes comments on all aspects of this proposed 
rulemaking. Commenters are especially encouraged to give suggestions 
for changing any aspects of the proposal that they find objectionable. 
All comments, with the exception of proprietary information, should be 
directed to Docket No. A-94-55 with regard to part 51 and Docket No. A-
94-56 with regard to part 58 (see ADDRESSES).
    Commenters who wish to submit proprietary information for 
consideration should clearly separate such information from other 
comments by: (1) Labeling proprietary information ``Confidential 
Business Information,'' and (2) sending proprietary information 
directly to the contact person listed (see FOR FURTHER INFORMATION 
CONTACT) and not to the public docket.
    This will help ensure that proprietary information is not 
inadvertently placed in the docket. If a commenter wants EPA to use a 
submission labeled as [[Page 12512]] confidential business information 
as part of the basis for the final rule, then a nonconfidential version 
of the document, which summarizes the key data or information, should 
be sent to the docket. Information covered by a claim of 
confidentiality will be disclosed by EPA only to the extent allowed and 
by the procedures set forth in 40 CFR part 2. If no claim of 
confidentiality accompanies the submission when it is received by EPA, 
the submission may be made available to the public without notifying 
the commenters.

B. Public Hearing

    Anyone who wants to present testimony about this proposal at the 
public hearing (see DATES) should, if possible, notify the contact 
person (see FOR FURTHER INFORMATION CONTACT) at least 7 days prior to 
the day of the hearing. The contact person should be given an estimate 
of the time required for the presentation of testimony and notification 
of any need for audio/visual equipment. A sign-up sheet will be 
available at the registration table the morning of the hearing for 
scheduling those who have not notified the contact earlier. This 
testimony will be scheduled on a first-come, first-serve basis to 
follow previously scheduled testimony.
    The EPA requests that approximately 50 copies of the statement or 
material to be presented be brought to the hearing for distribution to 
the audience. In addition, EPA would find it helpful to receive an 
advance copy of any statement or material to be presented at the 
hearing at least 1 week before the scheduled hearing date. This is to 
give EPA staff adequate time to review such material before the 
hearing. Such advance copies should be submitted to the contact person 
listed.
    The official records of the hearing will be kept open for 30 days 
following the hearing to allow submission of rebuttal and supplementary 
testimony. All such submissions should be directed to Docket No. A-94-
55 with regard to part 51 and Docket No. A-94-56 with regard to part 58 
(see ADDRESSES).
    Joseph W. Paisie is hereby designated Presiding Officer of the 
hearing. The hearing will be conducted informally, and technical rules 
of evidence will apply. A written transcript of the hearing will be 
placed in the above docket for review. Anyone desiring to purchase a 
copy of the transcript should make individual arrangements with the 
court reporter recording the proceeding.

XI. Administrative Requirements

A. Regulatory Impact Analysis

    Under Executive Order 12866, (58 FR 51735 (October 4, 1993)) the 
Agency must determine whether the regulatory action is ``significant'' 
and therefore subject to the Office of Management and Budget (OMB) 
review and the requirements of the Executive Order. The Order defines 
``significant regulatory action'' as one that is likely to result in a 
rule that may:
    (1) Have an annual effect on the economy of $100 million or more or 
adversely 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.
    (2) Create a serious inconsistency or otherwise interfere with an 
action taken or planned by another agency.
    (3) Materially alter the budgetary impact of entitlements, grants, 
user fees, or loan programs or the rights and obligations of recipients 
thereof.
    (4) Raise novel legal or policy issues arising out of legal 
mandates, the President's priorities , or the principles set forth in 
the Executive Order.
    Pursuant to the terms of Executive Order 12866, it has been 
determined that this rule is a ``significant regulatory action'' 
because of its potential to have an annual effect on the economy of 
$100 million or more as discussed in the related SO2 NAAQS 
proposal package on November 15, 1994 (59 FR 58958). As such, this 
action was submitted to OMB for review. Changes made in response to OMB 
suggestions or recommendations will be documented in the public record.
    The EPA has prepared a draft regulatory impact analysis (RIA) based 
on information developed by several EPA contractors. It includes 
estimates of costs, benefits, and net benefits associated with 
alternative SO2 NAAQS. The draft analysis, entitled Regulatory 
Impact Analysis of the National Ambient Air Quality Standards for 
SO2-Draft, is available from the address given above. The draft 
RIA estimates the cost for the short-term SO2 NAAQS regulatory 
alternative. The cost estimate for the short-term SO2 NAAQS 
alternative represent a snapshot of the estimated total industry costs 
that could be incurred at some unspecified time in the future following 
full implementation of a short-term SO2 NAAQS. The costs are based 
on the use of add-on control devices and fuel switching to lower-sulfur 
fuels. Given that EPA believes that many sources will be able to reduce 
their peaks through other, nontechnological means, this assumption may 
result in overstating costs. With this caveat in mind, nonutility 
annualized costs are estimated to be approximately $250 million for an 
ambient SO2 concentration for a 0.06 ppm, 5 annual exceedance 
concentration levels are estimated to be approximately $160 million. It 
is estimated that SO2 will be reduced by approximately 910,000 
tons, and 560,000 tons for 1 and 5 exceedance cases, respectively. 
Incremental to the title IV requirements and attainment of the existing 
SO2 NAAQS, total utility annualized costs in 2005 are estimated to 
be an additional $1.5 billion for the 0.06 ppm, 1 expected exceedance 
case, and $400 million for the 5 expected exceedance case. Estimated 
total utility SO2 emissions in 2005 are not expected to change 
given the title IV emissions trading program.
    Administrative costs are estimated to be approximately $18 million 
for the short-term NAAQS regulatory alternative. Monitoring costs are 
estimated to be minimal.
    However, EPA has not completed its cost analysis of the section 303 
regulatory alternative which EPA believes will be less than the 
SO2 NAAQS regulatory alternative. The EPA intends to complete this 
analysis and make it available to the public by the end of January 
1995. The EPA will announce the availability of this analysis in the 
Federal Register as soon as it is available. A final RIA will be issued 
at the time of promulgation of final standards. Neither the draft RIA 
nor the other contractor reports have been considered in issuing this 
proposal.
    The regulations, implementation of the revised SO2 NAAQS, the 
retained existing NAAQS, and the section 303 program, have been 
submitted to OMB for review under Executive Order 12866. Any written 
comments from OMB and any EPA responses to those comments are in the 
public docket for this rulemaking.

B. Impact on Reporting Requirements

    Air quality monitoring activities that would occur as a result of 
the SO2 NAAQS proposal could increase the costs and man-hour 
burdens to State and local agencies for conducting ambient SO2 
surveillance required by 40 CFR part 58 and currently approved under 
OMB Control Number 2060-0084. Temporarily-increased costs could result 
from the relocation of some monitors currently operated as part of the 
SLAMS networks and from the purchase and operation of additional 
monitors in a small number of agencies. [[Page 12513]] However, some or 
all of these costs could be offset by savings in existing monitoring 
networks. As a result, to the extent that additional monitoring costs 
will be incurred at all, EPA expects that these costs will be minimal.
    The information collection requirements in this proposed rule have 
been submitted for approval to OMB under the Paperwork Reduction Act, 
44 U.S.C. 3501 et seq. An Information Collection Request document has 
been prepared by EPA (ICR No.0940.11) and a copy may be obtained from 
Sandy Farmer, Information Policy Branch, EPA, 401 M St., S.W. (Mail 
Code 2136), Washington, DC 20460, or by calling (202) 260-2740.
    Send comments regarding the burden estimate or any other aspect of 
this collection of information, including suggestions for reducing this 
burden to Chief, Information Policy Branch, EPA, 401 M St., S.W. (Mail 
Code 2136), Washington, DC 20460, and to the Office of Information and 
Regulatory Affairs, Office of Management and Budget, Washington, DC 
20503, marked ``Attention: Desk Officer for EPA.'' The final rule will 
respond to any OMB or public comments on the information collection 
requirements contained in this proposal.

C. Impact on Small Entities

    Under the Regulatory Flexibility Act, 5 U.S.C., 600 et seq, the 
Agency must prepare a regulatory flexibility analysis assessing the 
impact of any proposed or final rule on small entities. Under 5 U.S.C. 
605(b), this requirement may be waived if the Agency certifies that the 
rule will not have a significant economic effect on a substantial 
number of small entities. Small entities include small businesses, 
small not-for-profit enterprises, and governmental entities with 
jurisdiction over populations of less than 50,000.
    A decision to revise the current NAAQS for SO2 or set a 
trigger level for implementation of a section 303 program would impose 
no new major requirements. It is expected that following the 
promulgation of a revised SO2 NAAQS, additional nonattainment 
areas will be designated and will thus have to submit SIP revisions 
imposing additional control requirements on affected sources.
    Furthermore, the control measures necessary to attain and maintain 
the NAAQS or implement a section 303 program are developed by the 
respective States as part of their SIP's. In selecting such measures, 
the States have considerable discretion so long as the mix of controls 
selected is adequate to attain and maintain the NAAQS or not exceed the 
section 303 trigger level. Whether a particular NAAQS would have a 
significant effect on a substantial number of small entities, 
therefore, depends on how the States would choose to implement it. For 
these reasons, any assessment performed by EPA on the costs of 
additional SIP requirements at this time would necessarily be 
speculative. On the basis of the above considerations and findings, and 
as required by section 605 of the Regulatory Flexibility Act, 5 U.S.C. 
601 et seq., the Administrator certifies that this regulation does not 
have a significant impact on a substantial number of small entities.

D. Reduction of Governmental Burden

    Executive Order 12875 (``Enhancing the Intergovernmental 
Partnership'') is designed to reduce the burden to State, local, and 
tribal governments of the cumulative effect of unfunded Federal 
mandates. The Order recognizes the need for these entities to be free 
from unnecessary Federal regulation to enhance their ability to address 
problems they face and provides for Federal agencies to grant waivers 
to these entities from discretionary Federal requirements. The Order 
applies to any regulation that is not required by statute and that 
creates a mandate upon a State, local, or tribal government. The EPA is 
required by statute to review periodically and, as necessary, revise 
the national ambient air quality standards, and to call on States to 
develop plans to attain and maintain these standards. However, this 
action also includes a request for comment on the adoption of a section 
303 program, as well as a proposal to establish a targeted monitoring 
network, neither of these actions is explicitly mandated by statute. 
Therefore, in accordance with the purposes of Executive Order 12875, 
EPA will consult with representatives of State, local, and tribal 
governments to inform them of the requirements for implementing the 
alternative regulatory measures being proposed to address short-term 
peak SO2 exposures. The EPA will summarize the concerns of the 
governmental entities and respond to their comments prior to taking 
final action.
    The EPA anticipates that there will be no additional cost burden 
imposed on States in order to implement the monitoring requirements 
proposed in this notice. In general, costs incurred for relocating 
monitors will be offset by operating costs saved from discontinuing 
SLAMS and NAMS monitors. For more detail the reader is referred to the 
section on resource concerns for relocating monitors under the targeted 
implementation strategy section discussed earlier in this notice or to 
the supporting statement for the information collection request.

E. Environmental Justice

    Executive Order 12898 requires that each Federal agency shall make 
achieving environmental justice part of its mission by identifying and 
addressing, as appropriate, disproportionately high and adverse human 
health or environmental effects of its programs, policies, and 
activities on minority and low-income populations. The requirements of 
Executive Order 12898 have been addressed in the draft RIA cited above.
    On average, approximately 25 percent of the total population and 14 
percent of total households residing in geographic areas that are 
potentially impacted by short-term SO2 peaks of 0.60 ppm or 
greater are nonwhite and below the poverty level, respectively. These 
estimates exceed the national averages of 19.7 percent and 12.7 
percent, respectively. It also follows that, on average, 25 percent of 
the asthmatics potentially exposed to short-term SO2 peaks of 0.60 
ppm or greater are nonwhite. Upon closer examination, 44 percent of 
these potentially SO2-impacted areas have a nonwhite population 
greater than the national average with 24 percent between 1 and 2 times 
greater, 10 percent between 2 and 3 times greater, 7 percent between 3 
and 4 times greater, and 3 percent between 4 and 5 times greater.

Appendix A--References

Bennett, K.M. (1982), US EPA, Assistant Administrator for Air, Noise 
and Radiation, Policy on Excess Emissions During Startup, Shutdown, 
Maintenance, and Malfunctions, Memorandum to Regional 
Administrators, Regions I-X, September 28, 1982.
Burton, C.S.; Stoeckenius, T.E.; Stocking, T.S.; Carr, E.L.; Austin, 
B.S.; Roberson, R.L. (1987), Assessment of exposures of exercising 
asthmatics to short-term SO2 levels as a result of emissions 
from U.S. fossil-fueled power plant, Systems Applications, Inc., San 
Rafael, CA., Pub. No. 87/176, September 23, 1987.
Calcagni, J. (1992), Director of Air Quality Management Division, 
Processing of State Implementation Plan (SIP) Submittals, Memorandum 
to Air Division Directors, Regions I-X, July 9, 1992.
EPA (1982a), Review of the National Ambient Air Quality Standards 
for Sulfur Oxides: Assessment of Scientific and Technical 
Information-OAQPS Staff Paper, Office of Air Quality Planning and 
Standards, Research Triangle Park, NC, EPA-450/5-82-007. 
[[Page 12514]] 
EPA (1986a), Second Addendum to Air Quality Criteria for Particulate 
Matter and Sulfur Oxides (1982): Assessment of Newly Available 
Health Effects Information, Environmental Criteria and Assessment 
Office, Research Triangle Park, NC, EPA-450/5-86-012.
EPA (1986b), Review of the National Ambient Air Quality Standards 
for Sulfur Oxides: Updated Assessment of Scientific and Technical 
Information, Addendum to the 1982 OAQPS Staff Paper, Office of Air 
Quality Planning and Standards, Research Triangle Park, NC, EPA-450/
05-86-013.
EPA (1986c), Guideline on Air Quality Models (Revised), Office of 
Air Quality Planning and Standards, Research Triangle Park, NC, EPA-
450/2-78-027R. (Codified in 40 CFR 51, appendix W.)
EPA (1992a), National Air Pollutant Emission Estimates, 1900-1991, 
Office of Air Quality Planning and Standards, Research Triangle 
Park, NC, EPA-454/R-92-013.
EPA (1992b), National Air Quality and Emissions Trends Report, 1991, 
Office of Air Quality Planning and Standards, Research Triangle 
Park, NC, EPA/450-R-92-001.
EPA (1992c), General Preamble for the Implementation of Title I of 
the Clean Air Act Amendments of 1990, April 16, 1992, 57 FR 13498.
EPA (1994b), Review of the Ambient Air Quality Standards for Sulfur 
Oxides: Updated Assessment of Scientific and Technical Information, 
Supplement to the 1986 OAQPS Staff Paper Addendum, Office of Air 
Quality Planning and Standards, Research Triangle Park, NC, EPA/452/
R-94-013.
Polkowsky, B. (1991), U.S. EPA, Standards Development Section, 
Analysis of Utility Emissions Limits for Short-term SO2 NAAQS, 
Memorandum to John Haines, Chief, Standards Development Section, 
August 2, 1991.
Rosenbaum, A.S.; Hudischewskyj, A.B.; Roberson, R.L.; Burton, C.S. 
(1992), Estimates of Future Exposures of Exercising Asthmatics to 
Short-term Elevated SO2 Concentrations Resulting from Emissions 
of U.S. Fossil-fueled Power Plants: Effects of the 1990 Amendments 
to the Clean Air Act and a 5-minute Average Ambient SO2 
Standard, Pub. No. SYSAPP-92/016, April 23, 1992.
Sonoma Technology Inc. (January 1994), Recommendations for the Use 
of Open-Path and Fixed-Point Monitors for Determining Ambient 
SO2 Concentrations, Final Report STI-94021-1402-FR, Prepared 
for U. S. Environmental Protection Agency, Research Triangle Park, 
NC.
Strelow, R. (1976), U.S. EPA, Assistant Administrator, Air and Waste 
Management, Guidance for Determining Acceptability of SIP 
Regulations in Non-attainment Areas, Memorandum to Regional 
Administrators, Regions I-X, December 9, 1976.
Stoeckenius, T.E.; Garelick, B.; Austin, B.S.; O'Connor, K.; 
Pehling, J.R. (1990), Estimates of Nationwide Asthmatic Exposures to 
Short-term Sulfur Dioxide Concentrations in the Vicinity of 
Nonutility Sources. Systems Application Inc., San Rafael, CA, 
Publication Number SYSAPP-90/129, December 6, 1990.

List of Subjects in 40 CFR Parts 51 and 58

    Environmental protection, Administrative practices and procedure, 
Air pollution control, Intergovernmental relations, SO2, Reporting 
and recordkeeping requirements, State implementation plans.

    Dated: February 15, 1995.
Carol M. Browner,
Administrator.
    For the reasons set forth in the preamble, EPA proposes to amend 
chapter I of title 40 of the Code of Federal Regulations as follows:

PART 51--REQUIREMENTS FOR PREPARATION, ADOPTION, AND SUBMITTAL OF 
IMPLEMENTATION PLANS

    1. The authority citation for part 51 continues to read as follows:

    Authority: 42 U.S.C. 7401(a)(2), 7475(e), 7502 (a) and (b), 
7503, 7601(a)(1) and 7602.

    2. In Sec. 51.151 of subpart H, the entry for ``Sulfur dioxide'' is 
revised to read as follows:


Sec. 51.151  Significant harm levels.

* * * * *
    Sulfur dioxide--0.29 parts per million (750 micrograms/cubic 
meter), 24-hour average.
* * * * *
    3. In appendix L to part 51, paragraphs 1.1 (b), (c), and (d) are 
amended by revising the entries for ``SO2'' to read as follows:

Appendix L to Part 51--Example Regulations for Prevention of Air 
Pollution Emergency Episodes

* * * * *
    1.1 * * *
    (b) * * *
    SO2--0.19 ppm (500 g/m\3\), 24-hour average.
* * * * *
    (c) * * *
    SO2--0.23 ppm (600 g/m\3\), 24-hour average.
* * * * *
    (d) * * * 
    SO2--0.26 ppm (675 g/m\3\), 24-hour average.
* * * * *

Subpart T--Conformity to State or Federal Implementation Plans of 
Transportation Plans, Programs, and Projects Developed, Funded or 
Approved Under Title 23 U.S.C. or the Federal Transit Act

    4. Section 51.465 is added to Subpart T to read as follows:


Sec. 51.465  Contingency plans.

    (a) Each plan must include a contingency plan which must, as a 
minimum, provide for taking action necessary to prevent further 
violations of the 5-minute trigger level for sulfur dioxide (SO2) 
attributable to emissions from a source once one exceedance has 
occurred. The 5-minute trigger level is 0.60 parts per million (ppm), 
not to be exceeded more than once per calendar year, as determined in 
accordance with appendix Y to this part.
    (b) Each contingency plan must provide that:
    (1) Within 30 days of determination of a violation of the trigger 
level, the State shall carry out a compliance inspection of any source 
whose emissions may have resulted in or contributed to the violation of 
the trigger level.
    (2) If the source is out of compliance with applicable SO2 
emission limits then, within 30 days of completing the compliance 
inspection in paragraph (b)(1) of this section, the State shall take 
enforcement action to bring the source into compliance.
    (3) If the source is in compliance with applicable SO2 
emission limits then, within 60 days of completing the compliance 
inspection in paragraph (b)(1) of this section, the State shall develop 
and implement an enforceable emission reduction plan with a compliance 
schedule to address the cause of the emissions producing the trigger 
level violation. The schedule shall provide for implementation of all 
actions necessary to prevent further violations of the trigger level as 
expeditiously as practicable. This emission reduction plan must be 
submitted to EPA as a revision to their State implementation plan 
within 1 year of completing the compliance inspection in paragraph 
(b)(1) of this section.
    (4) If in carrying out the compliance inspection referred to in 
paragraph (b)(1) of this section, the State determines that the source 
is out of compliance with its applicable SO2 emission limits but 
also determines that bringing the source into compliance with its 
applicable emission limits would not be likely to prevent further 
exceedances of the trigger level, then the State and source shall 
develop and [[Page 12515]] implement an emission reduction plan as 
described in paragraph (b)(3) of this section.
    5. Appendix Y is added to part 51 to read as follows:

Appendix Y to Part 51--Interpretation of the 5-Minute Trigger Level for 
Sulfur Dioxide

1.0  General

    a. This appendix explains the computations necessary for 
analyzing sulfur dioxide data to determine whether the 5-minute 
trigger level specified in Sec. 51.400(a), subpart T, has been 
exceeded and whether the 5-minute trigger level has been violated. 
Sulfur dioxide is measured in the ambient air by the reference 
method specified in appendix A of this part or an equivalent method 
designated in accordance with part 53 of this chapter.
    b. Several terms used in this appendix must be defined. A ``5-
minute hourly maximum'' for SO2 refers to the highest of the 12 
possible nonoverlapping 5-minute SO2 averages calculated or 
measured during a clock hour. The term ``exceedance'' of the 5-
minute trigger level concentration means a 5-minute hourly maximum 
value that is greater than the 5-minute trigger level after rounding 
to the nearest hundredth ppm (i.e., values ending in or greater than 
0.005 ppm are rounded up; e.g., a value of 0.605 would be rounded to 
0.61, which is the smallest value for an exceedance). The term 
``year'' refers to a calendar year. The term ``quarter'' refers to a 
calendar quarter. The 5-minute SO2 trigger level is expressed 
in terms of the number of expected exceedances per year by adjusting 
for missing data (if required) and by averaging over a 2-year 
period.

2.0  Trigger Level Determination

    a. The 5-minute trigger level is not violated when the number of 
expected exceedances per year is less than or equal to one. In 
general, this determination is to be made by recording the number of 
5-minute hourly maximum exceedances at a monitoring site for each 
year, using the calculations in section 3.2 to compensate for 
missing data (if required), averaging the number of exceedances over 
a 2-year period, and comparing the number of exceedances (rounded to 
the nearest integer) to the number of allowable exceedances.
    b. Although it is necessary to meet the minimum data 
completeness requirements to use the computational formula described 
in section 3.2, this criterion does not apply when there are obvious 
exceedance situations which contribute to a violation. For example, 
when a site fails to meet the completeness criteria, violation of 
the 5-minute trigger level can still be established on the basis of 
the observed number of exceedances in a year (e.g., three observed 
exceedances in a single year).

3.0  Calculations for the 5-Minute Trigger Level

3.1  Calculating a 5-Minute Hourly Maximum

    A 5-minute hourly maximum value for SO2 is the highest of 
the 5-minute averages from the 12 possible nonoverlapping periods 
during a clock hour. These 5-minute values shall be rounded to the 
nearest hundredth ppm (fractional values equal to or greater than 
0.005 ppm are rounded up). A 5-minute maximum shall be considered 
valid if: (1) 5-minute averages were available for at least 9 of the 
12 5-minute periods during the clock hour, or (2) the value of the 
5-minute average exceeds the level of the 5-minute trigger level.

3.2  Calculating Expected Exceedances for a Year

    a. Because of practical considerations, a 5-minute maximum 
SO2 value may not be available for each hour of the year. To 
account for the possible effect of incomplete data, an adjustment 
must be made to the data collected at a particular monitoring 
location to estimate the number of exceedances in a year. The 
adjustment is made on a quarterly basis to ensure that the entire 
year is adequately represented. In this adjustment, the assumption 
is made that the fraction of missing values that would have exceeded 
the trigger level is identical to the fraction of measured values 
above this level.
    b. For all NAMS and SLAMS sites that report 5-minute SO2 
data, the computation for incomplete data is to be made for all 
sites with 50 to 90 percent complete data in each quarter. If a site 
has more than 90 percent complete data in a quarter, no adjustment 
for missing data is required. If a site has less than 50 percent 
complete data in a quarter, no adjustment for missing data is 
required and the observed exceedances are used.
    c. The estimate of the expected number of exceedances for the 
quarter is equal to the observed number of exceedances plus an 
increment associated with the missing data.
    1. The following formula must be used for these computations:

eq=vq+[(vq/
nq) x (Nq-nq]=vq x Nq/nq  [1]

Where:

eq=the expected number of exceedances for quarter q,
vq=the observed number of exceedances for quarter q,
Nq=the number of hours in quarter q, and
nq=the number of hours in the quarter with valid 5-minute 
hourly SO2 maximums
q=the index for each quarter, q=1, 2, 3 or 4.

    2. The expected number of exceedances for the quarter must be 
rounded to the nearest hundredth (fractional values equal to or 
greater than 0.005 are rounded up).
    d.1. The expected number of exceedances for the year, e, is the 
sum of the estimates for each quarter.
[GRAPHIC][TIFF OMITTED]TP07MR95.001


    2. The expected number of exceedances for a single year must be 
rounded to one decimal place (fractional values equal to or greater 
than 0.05 are rounded up).
    e. The number of exceedances is then estimated by averaging the 
individual annual estimates over a 2-year period, rounding to the 
nearest integer, and comparing with the allowable exceedance rate of 
one per year (fractional values equal to or greater than 0.5 are 
rounded up; e.g., an expected number of exceedances of 1.5 would be 
rounded to 2, which is the lowest value for violating the trigger 
level.
    f. Example.
    1. During the most recent quarter, 1210 out of a possible 2208 5-
minute hourly maximums were recorded, with one observed exceedance of 
the 5-minute trigger level. Using formula [1], the expected number of 
exceedances for the quarter is:

eq = 1 x 2208/1210 = 1.825 or 1.83

    2. If the expected exceedances for the other 4 quarters were 0.0, 
then using formula [2], the expected number of exceedances for the year 
is:

1.83 + 0.0 + 0.0 + 0.0 = 1.83 or 1.8

    3. If the expected number of exceedances for the previous year was 
0.0, then the expected number of exceedances is estimated by:

(1.8 + 0.0)/2 = 0.9 or 1

    4. Since 1 is not greater than the allowable number of exceedances, 
this monitoring site would not violate the trigger level.

PART 58--AMBIENT AIR QUALITY SURVEILLANCE

    1. The authority citation for part 58 continues to read as follows:

    Authority: Secs 110, 301(a), and 319 of the Clean Air Act as 
amended, (42 U.S.C. 7410, 7601(a), and 7619).

    2. Section 58.1 is amended by adding and reserving paragraphs (aa) 
through (hh) and by adding paragraphs (ii) and (jj) to read as follows:


Sec. 58.1  Definitions.

* * * * *
    (ii) ``Metropolitan Statistical Area'' means the most recent area 
as designated by the U.S. Office of Management and Budget and 
population figures from the U.S. Bureau of the Census. The Department 
of Commerce defines a metropolitan area as ``one of a large population 
nucleus, together with adjacent communities which have a high degree of 
economic and social integration with that nucleus.''1

    \1\U.S. Bureau of the Census, ``Statistical Abstract of the 
United States: 1993'', (113th Edition), Washington, DC (1993).
---------------------------------------------------------------------------

    (jj) ``Consolidated Metropolitan Statistical Area'' means the most 
recent area as designated by the U.S. Office of Management and Budget 
and population figures from the Bureau of [[Page 12516]] the Census. 
The Department of Commerce provides ``that within metropolitan 
complexes of 1 million or more population, separate component areas are 
defined if specified criteria are met. Such areas are designated 
primary metropolitan statistical areas (PMSA's); and any area 
containing PMSA's is designated a consolidated metropolitan statistical 
area (CMSA).''\2\
* * * * *
    3. In appendix C to part 58, section 2.4 is added to read as 
follows:

Appendix C--Ambient Air Quality Monitoring Methodology

* * * * *
    2.4  A monitoring method for SO2 used for obtaining 5-
minute average concentrations in connection with targeted monitoring 
of an SO2 source likely to produce short-duration, high-level 
concentration peaks must be a designated reference or equivalent 
method as defined in Sec. 50.1 of this chapter and must meet the 
supplemental specifications for 5-minute monitoring given in table 
B-1 of part 53 of this chapter.
* * * * *
    4. In appendix D to part 58, section 1, the last two sentences of 
the third paragraph are removed, and replaced by four new sentences to 
read as follows:

Appendix D--Network Design for State and Local Air

Monitoring Stations (SLAMS), National Air Monitoring Stations 
(NAMS), and Photochemical Assessment Monitoring Stations (PAMS)

* * * * *
    1. * * *
    *  *  *  It should be noted that this appendix contains no 
criteria for determining the total number of stations in SLAMS 
networks. A minimum number of lead SLAMS is prescribed as well as a 
minimum required number of SO2 SLAMS for those counties not 
within the boundaries of any CMSA/MSA. Also, a minimum required 
number of SO2 SLAMS is listed for targeted sources of SO2 
emissions. The optimum size of a particular SLAMS network involves 
trade-offs among data needs and available resources which EPA 
believes can best be resolved during the annual network design 
review process.
* * * * *


Sec. 2.3  [Amended]

    5. In appendix D, the first paragraph of section 2.3 is revised, 
and a new paragraph is added between the first and second paragraphs to 
read as follows:
* * * * *
    2. * * *
    2.3 * * *
    The spatial scales for SO2 SLAMS monitoring are the micro, 
middle, neighborhood, urban, and regional scales. The most important 
spatial scales to effectively characterize the emissions of SO2 
from stationary sources are the micro, middle, and neighborhood 
scales. Because of the nature of SO2 emissions and the nature 
of distributions over metropolitan areas, the neighborhood scale is 
the most likely scale to be represented by a single measurement in 
the metropolitan area where the concentration gradients are less 
steep, but only if the undue effects from local sources (minor or 
major point sources) can be eliminated. Urban scales would represent 
areas where the concentrations are uniform over a larger 
geographical area. Regional scale measurements would be associated 
with rural areas and urban background measurements.
    Microscale--Emissions from stationary sources may, under certain 
plume conditions, result in high 5-minute and 24-hour ground level 
concentrations at the microscale level. The microscale measurements 
would represent an area impacted by the plume with dimensions 
extending up to approximately 100 meters.
* * * * *
    6. In appendix D, section 2.3, a sentence is added to the end of 
the paragraph titled ``Middle Scale'' to read as follows:
* * * * *
    2.3 * * *
    Middle Scale * * * Emissions from stationary sources that cover 
larger geographic areas may also result in high 5-minute and 24-hour 
SO2 concentrations.
* * * * *
    7. In appendix D, section 2.3, a sentence is added to the last 
paragraph to read as follows:
* * * * *
    2.3 * * *
    * * * The use of SO2 saturation monitors is encouraged to 
determine the areas of maximum concentration from sources of 
SO2 emissions as an aid to locating reference or equivalent 
SO2 monitors.
* * * * *
    8. In appendix D, Sec. 2.3, seven new paragraphs are added at the 
end of this section to read as follows:
* * * * *
    The required number of sites needed to measure SO2 
concentrations for population exposure in the metropolitan areas of 
the counties are discussed in section 3.2 of this appendix. However, 
there may be significant point source emissions in other counties 
which are not within the geographic boundaries of any CMSA/MSA. To 
determine the SO2 concentrations and exposures for these 
counties, a minimum number of SLAMS SO2 monitors will be 
required. Table 2 shows the minimum required number of SLAMS 
SO2 monitors for those counties which are not a part of any 
CMSA/MSA and also have SO2 emissions greater than 20,000 tons/
year as defined in the Air Facility Subsystem of AIRS.

        Table 2--State and Local Air Monitoring Stations Criteria       
------------------------------------------------------------------------
                                                       Minimum number of
           Area             SO2 emissions (tons/year)     SO2 stations  
------------------------------------------------------------------------
Counties (or parts of                                                   
 counties) not included in                                              
 any CMSA/MSA.............                   >100,000                   
                                       20,000-100,000                   
                                              <20,000                  2
                                                                       1
                                                                       0
------------------------------------------------------------------------

    Monitors located to meet this requirement would generally be 
either middle or neighborhood scale of representativeness to measure 
population exposure. The monitors are not necessarily required to be 
located in the county where the SO2 emissions originate, but 
should be located in the maximum concentration area. The maximum 
concentration area may be determined by modeling the SO2 
emission sources and/or in combination with SO2 saturation 
monitoring studies.

    \2\See footnote 1 in paragraph (ii) of this section.
---------------------------------------------------------------------------

    The EPA will consider a request to waive all or part of these 
requirements for these areas. If monitoring has been conducted for a 
minimum of 2 years and the measured concentrations were low, then 
EPA will consider a request to waive all or part of the monitoring 
requirement in accordance with EPA guidelines.
    In addition to the above requirement for SO2 monitors, 
SLAMS monitors are required to be deployed around targeted sources 
of SO2 emissions in order to produce 5-minute, 3-hour, 24-hour, 
and annual average concentration measurements. A listing of which 
sources are to be monitored, the schedule for monitoring, and the 
rationale for selecting the sources shall be prepared by the State 
in a targeted SO2 monitoring plan to be reviewed as part of the 
annual SLAMS network review. The implementation of this plan will be 
as expeditious as practicable.
    To adequately monitor and characterize air quality around point 
sources of SO2 emissions would require multiple point monitors 
or open path analyzers (or a combination of both). Financial and 
practical [[Page 12517]] 
constraints may prohibit the deployment of large numbers of SO2 
monitors around these targeted sources. Therefore, a modest network 
with a minimum requirement of four SO2 monitors around each 
targeted source will be used. If open path analyzers with multiple 
paths are used, each monitoring path could potentially be 
substituted for one point SO2 monitor. Modeling and/or 
saturation sampling may be used to determine the general area(s) of 
expected maximum SO2 concentrations based on the most 
predominant wind direction. One monitor will be located at the fence 
line downwind of the most predominant wind direction, and a second 
monitor will be located in the modeled maximum concentration area 
based on the most predominant wind direction. Since wind directions 
frequently change from one season to another, the second most 
predominant wind direction will be used to locate the second pair of 
monitors. The third monitor will be located at the fence line 
downwind of the second most predominant wind direction, and the 
fourth monitor will be located in the modeled maximum concentration 
area based on the second most predominant wind direction. However, 
for situations where there is complex terrain and/or meteorology, 
additional monitors may be required to adequately monitor the 
emissions.
    In some cases, it is simply not practical to place monitors at 
the indicated modeled locations of maximum concentrations. Some 
examples may include locations over open bodies of water, on rivers, 
swamps, cliffs, etc. The EPA Regional Offices and the State or local 
air pollution control agencies should determine alternative 
locations and alternative network designs on a case-by-case basis.
    The use of SO2 monitoring around targeted sources of 
SO2 emissions is intended to capture high 5-minute peak 
concentrations as well as exceedances of the 3-hour, 24-hour, and 
annual mean standards for SO2. However, there will be cases 
where this monitoring strategy will be implemented around the 
targeted sources of SO2 emissions, and the resulting measured 
SO2 concentrations will be low. Therefore, SO2 monitoring 
around a targeted source must be conducted for a minimum of 2 years 
to account for factors such as year-to-year variability in 
meteorology, change of plant processes, etc. If monitoring has been 
conducted for a minimum of 2 years, and the concentrations were low, 
then a decision could be made in the annual SLAMS network review 
between the EPA Regional Office and the State or local air pollution 
control agency to move the SO2 monitors to another targeted 
source of SO2 emissions. In general, it is more important to 
monitor around another targeted source of SO2 emissions than to 
retain monitors around a source with demonstrated low SO2 
concentrations.

    9. In appendix D, the first two sentences of the first paragraph of 
section 3 are removed and the following two sentences are added to read 
as follows:
* * * * *
    3. * * *
    The NAMS must be stations selected from the SLAMS network with 
emphasis given to urban and multisource areas. Areas to be monitored 
must be selected based on the CMSA/MSA population and pollutant 
emission concentration levels as defined in the Air Facility 
Subsystem of AIRS. * * *
* * * * *
    10. In appendix D, the first paragraph in section 3.2 and Table 3 
are revised to read as follows:
* * * * *


3.2  Sulfur Dioxide (SO2) Design Criteria for NAMS.

    It is desirable to have several NAMS in the more polluted and 
densely populated urban and multisource areas to characterize the 
national and regional SO2 air quality trends and geographical 
patterns. Table 3 shows the required number of NAMS monitors in the 
metropolitan areas to accomplish this purpose. These neighborhood 
scale monitoring stations (which would be located within the 
boundaries of the CMSA/MSA) would normally be classified as category 
(a) or (b) as discussed in section 3. The actual number and location 
of the NAMS must be determined by the EPA Regional Office and the 
State agency, subject to the approval of EPA Headquarters (OAR).

           Table 3.--National Air Monitoring Station Criteria           
------------------------------------------------------------------------
                                                                Minimum 
                                                                required
            CMSA/MSA population               SO2 emissions      number 
                                               (tons/year)        SO2   
                                                                stations
------------------------------------------------------------------------
>1,000,000................................            200,000          3
                                              100,000-200,000          2
                                                    0-100,000          1
200,000-1,000,000.........................           >200,000          3
                                              100,000-200,000          2
                                               20,000-100,000          1
                                                      >20,000          0
50,000-200,000............................           >100,000          2
                                               20,000-100,000          1
                                                      <20,000          0
------------------------------------------------------------------------

* * * * *
    11. In appendix D, section 5, Table 5 is amended by revising the 
entry for ``Micro'' to read as follows:
* * * * *
    5. * * *

                   Table 5.--Summary of Spatial Scales for SLAMS and Required Scales for NAMS                   
----------------------------------------------------------------------------------------------------------------
                                     Scale applicable for SLAMS                 Scales Required for NAMS        
        Spatial scale        -----------------------------------------------------------------------------------
                               SO2     CO     O3    NO2     Pb   PM-10   SO2     CO     03    NO2     Pb   PM-10
----------------------------------------------------------------------------------------------------------------
Micro.......................  2).
    2.1.1  Site and Monitoring Information. City name (when 
applicable), county name and street address of site location. AIRS 
site code. AIRS monitoring method code. Number of 5-minute hourly 
maximum observations. Number of hourly observations.
    2.1.2  Annual Summary Statistics. Annual arithmetic mean (ppm). 
Highest and second highest 24-hour averages (ppm) (block averages 
measured midnight to midnight) and dates of occurrence. Highest and 
second highest 5-minute hourly maximums (ppm) (block averages) and 
dates and times (hour) of occurrence when 5-minute measurements are 
required. Highest and second highest 3-hour averages (ppm) (block 
averages beginning at midnight) and dates and times (ending hour) of 
occurrence. Number of exceedances of the 24-hour primary NAAQS. 
Number of exceedances of the 5-minute primary NAAQS (if a 5-minute 
primary NAAQS is promulgated) when 5-minute measurements are 
required. Number of exceedances of the 3-hour secondary NAAQS.
* * * * *
    14. Appendix G is amended by revising tables 1 and 2 and Figure 3 
to read as follows:

Appendix G--Uniform Air Quality Index and Daily Reporting

* * * * *

                                Table 1.--Breakpoints for PSI in Metric Units\1\                                
----------------------------------------------------------------------------------------------------------------
                                                24-hr.         24-hr.                     1-hr. 03    1-hr. NO2 
          PSI value ()            PMg/  SO2g/  8-hr. COmg/  g/  g/
                                                 m\3\           m\3\           m\3\         m\3\         m\3\   
----------------------------------------------------------------------------------------------------------------
50........................................            50           \3\80             5          120     (\2\)   
100.......................................           150          \3\365            10          235     (\2\)   
200.......................................           350          \3\500            17          400        1,130
300.......................................           420          \3\600            34          800        2,260
400.......................................           500          \3\675            46        1,000        3,000
500.......................................           600          \3\750          57.5        1,200        3,750
----------------------------------------------------------------------------------------------------------------
\1\At 25 deg.C and 760 mm Hg.                                                                                   
\2\No index values reported at these concentration levels because there is no short-term NAAQS.                 
\3\All the concentration levels are used for illustrative purposes only. The actual levels will be determined at
  the time of the promulgation of the standard.                                                                 


                      Table 2.--Breakpoints for PSI                     
                           [Parts per million]                          
------------------------------------------------------------------------
                                          24-hr.   8-hr.   1-hr.   1-hr.
         PSI value ()             SO2     CO      03      NO2 
------------------------------------------------------------------------
50......................................  \2\0.0                        
                                               3     4.5     .06   (\1\)
100.....................................  \2\0.1                        
                                               2       9     .12   (\1\)
200.....................................  \2\0.1                        
                                               9      15     0.2     0.6
300.....................................  \2\0.2                        
                                               3      30     0.4     1.2
400.....................................  \2\0.2                        
                                               6      40     0.5     1.6
500.....................................  \2\0.2                        
                                               9      50     0.6     2.0
------------------------------------------------------------------------
\1\No index value reported at these concentration levels because there  
  is no short-term NAAQS.                                               
\2\All the concentration levels are used for illustrative purposes only.
  The actual levels will be determined at the time of the promulgation  
  of the standard.                                                      

* * * * *


                                                 BILLING CODE 6560-50-P
[[Page 12519]]

[GRAPHIC][TIFF OMITTED]TP07MR95.000




BILLING CODE 6560-50-C
* * * * *
[FR Doc. 95-5021 Filed 3-6-95; 8:45 am]
BILLING CODE 6560-50-P