Environmental Protection Agency,

Federal Register: July 11, 2001 (Volume 66, Number 133)Proposed RulesPage 36369-36401From the Federal Register Online via GPO Access [wais.access.gpo.gov]

DOCID:fr11jy01-36

[Page 36369]

Part II

Environmental Protection Agency

40 CFR Part 52

Approval and Promulgation of Implementation Plans; Illinois; Ozone; Proposed Rule

[Page 36370]

ENVIRONMENTAL PROTECTION AGENCY

40 CFR Part 52

IL200-1; FRL-7008-9Approval and Promulgation of Implementation Plans; Illinois; Ozone

AGENCY: Environmental Protection Agency (EPA).

ACTION: Proposed rule.

SUMMARY: EPA is proposing to approve the following as revisions to the Illinois State Implementation Plan (SIP) for the Chicago-Gary-Lake County ozone nonattainment area, i.e., for the Illinois portion of this bi-state ozone nonattainment area: an ozone attainment demonstration; a post-1999 ozone Rate-Of-Progress (ROP) plan; a contingency measures plan for both the ozone attainment demonstration and post-1999 ROP plan; a commitment to conduct a mid-course review of the ozone attainment demonstration; mobile source conformity emission budgets for Volatile Organic Compounds (VOC) and Oxides of Nitrogen ( NOX) and the State's commitment to revise these emission budgets using the MOBILE6 emissions factor model; and, a Reasonably Available Control Measure (RACM) analysis. The EPA is also proposing to revise the existing NOXemissions control waiver for the Illinois portion of the Chicago-Gary-Lake County ozone nonattainment area to the extent that the State has relied on NOXemission controls from certain Electrical Generating Units (EGUs), major non-EGU boilers and turbines, and major cement kilns in the nonattainment area to attain the ozone standard. The existing NOXemissions control waiver remains in place for Reasonably Available Control Technology (RACT), New Source Review (NSR), and certain requirements of vehicle Inspection and Maintenance (I/M) and transportation and general conformity. The EPA is proposing to deny a related citizen petition for the termination of the NSR portion of the NOXwaiver.

DATES: Written comments must be received on or before August 10, 2001.

ADDRESSES: Written comments should be sent to J. Elmer Bortzer, Chief, Regulation Development Section, Air Programs Branch (AR-18J), U.S. Environmental Protection Agency, 77 West Jackson Boulevard, Chicago, Illinois 60604.

Copies of the State's submittals addressed in this proposed rule and other relevant materials are available for public inspection during normal business hours at the following address: U.S. Environmental Protection Agency, Region 5, Air and Radiation Division, 77 West Jackson Boulevard, Chicago, Illinois 60604 (please telephone Edward Doty at (312) 886-6057 before visiting the Region 5 office).

FOR FURTHER INFORMATION CONTACT: Edward Doty, Regulation Development Section, Air Programs Branch (AR-18J), U.S. Environmental Protection Agency, Region 5, 77 West Jackson Boulevard, Chicago, Illinois 60604, Telephone Number: (312) 886-6057, E-Mail Address: doty.edward@epa.gov.

SUPPLEMENTARY INFORMATION: Throughout this document whenever ``we,'' ``us,'' or ``our'' is used, we mean EPA. Whenever ``you'' or ``me'' is used, we mean you the reader of this proposed rule or the sources subject to the requirements of the State plan as discussed in the State's submittal or in this proposed rule.

This section provides additional information by addressing the following topics and questions:

I. What Action Is EPA Proposing Today? II. Background Information

A. What is a State Implementation Plan (SIP)?

B. What is the Federal Approval Process for a SIP?

C. What Does Federal Approval of a State Regulation Mean to Me?

D. What are the Options for Action on a State SIP Submittal?

E. What Ozone Nonattainment Area is Addressed by the State Submittal Reviewed in This Proposed Rule?

F. What Prior EPA Rulemakings Relate to or Led to the State Submittal Reviewed in this Proposed Rule?

G. What is the Time Frame for EPA to Take Action on the State Submittal?

H. What are the Basic Components of the State Submittal and What are the Subjects Covered in this Proposed Rule? III. Ozone Attainment Demonstration and Emissions Control Strategy

A. Background Information and Requirements Placed on the Ozone Attainment Demonstration

1. What Clean Air Act requirements apply to the State's ozone attainment demonstration?

2. What is the history of the State's ozone attainment demonstration and how is it related to EPA's NOXSIP Call?

3. What are the modeling requirements for the ozone attainment demonstrations?

4. What additional analyses may be considered when the ozone modeling fails to show attainment of the ozone standard?

5. Besides the modeled attainment demonstration and adopted emission control strategy, what other elements must be addressed in an attainment demonstration SIP?

6. What are the relevant EPA policy and guidance documents?

B. Technical Review of the State's Submittal

1. When was the attainment demonstration addressed in public hearings, and when was the attainment demonstration submitted to the EPA?

2. What are the basic components of the submittal?

3. What modeling approach was used in the analyses to develop and validate the ozone modeling system?

4. How were the 1996 base year emissions developed?

5. What procedures and sources of projection data were used to project the emissions to the attainment year?

6. How were the 1996 and 2007 emission estimates quality assured?

7. What is the adopted emissions control strategy?

8. What were the ozone modeling results for the base period and for the future attainment period with the selected emissions control strategy?

9. What additional analyses and emissions were modeled by the State of Illinois?

10. Do the modeling results demonstrate attainment of the ozone standard?

11. Does the attainment demonstration depend on future reductions of regional emissions?

12. Has the State adopted all of the regulations/rules needed to support the ozone attainment strategy and demonstration?

C. EPA's Evaluation of the Ozone Attainment Demonstration Portion of the State's Submittal

1. Did the State adequately document the techniques and data used to derive the modeling input data and modeling results of the analyses?

2. Did the modeling procedures and input data used comply with the Clean Air Act requirements and EPA guidelines?

3. Did the State adequately demonstrate attainment of the ozone standard?

4. Has the adopted emissions control strategy been adequately documented?

5. Is the emissions control strategy acceptable? IV. Post-1999 Rate-of-Progress (ROP) Plan

A. What is a Post-1999 ROP Plan?

B. What is the ROP Contingency Measure Requirement?

C. What Illinois Counties are Covered by the Post-1999 ROP Plan?

D. Who is Affected by the Illinois Post-1999 ROP Plan?

E. What Criteria Must a Post-1999 ROP Plan Meet to be Approved?

F. What are the Special Requirements for Claiming NOX Emission Reductions in Post-1996 ROP Plans?

G. How Did Illinois Calculate the Needed ROP and Contingency Emission Reduction Requirements?

1. VOC and NOXfractions of the total emission reductions for a milestone period

2. Baseline emissions

3. Milestone emission target levels

4. Projected emission growth levels

5. Emission reductions needed to achieve ROP

6. Calculation of the required contingency measure emission reduction

[Page 36371]

H. What are the Criteria for Acceptable ROP Emission Control Strategies?

I. What are the Emission Control Measures in Illinois' Post-1999 ROP Plan?

J. Are the Emission Control Measures and Calculated Emission Reductions Acceptable to the EPA, and is the Post-1999 ROP Plan Approvable? V. Contingency Measures Plan

A. What are the Requirements for Contingency Measures Under Section 172(c)(9) of the CAA?

B. How Does the Chicago Attainment Demonstration SIP Address the Contingency Measure Requirements?

C. Does the Chicago, Illinois Attainment Demonstration Meet the Contingency Measure Requirements? VI. Emission Control Rule Adoption and Implementation Status VII. Mid-Course Review Commitment

A. Why is a Mid-Course Review Commitment Necessary?

B. Did Illinois Submit a Mid-Course Review Commitment? VIII. NOXWaiver

A. What is the History of the NOXEmissions Control Waiver in the Chicago-Gary-Lake County Ozone Nonattainment Area?

B. What are the Conclusions of the State Regarding the Impact of the Ozone Attainment Demonstration on the NOXControl Waiver?

C. What Are the Bases and Conclusions of a Petition Against the NOXWaiver?

D. What are the Conclusions That Can Be Drawn Regarding the NOXControl Waiver From Data Contained in the State's Ozone Attainment Demonstration?

E. What are the EPA Conclusions Regarding the Existing NOXWaiver Given the Petition and the Available Ozone Modeling Data? IX. Motor Vehicle Emissions Budgets for Conformity and Commitment to Re-Model Using Mobile6

A. What are the Requirements for Motor Vehicle Emissions Budgets for Conformity?

B. How Were the Illinois Attainment Demonstration and ROP Emissions Budgets Developed?

C. Did Illinois Commit to Revise the Budgets When MOBILE6 Is Released?

D. Are the Illinois Emissions Budgets Adequate for Conformity Purposes? X. Reasonably Available Control Measure (RACM) Analysis

A. What are the Requirements for RACM?

B. How Does This Submission Address the RACM Requirement?

C. Does the Chicago Attainment Demonstration Meet the RACM Requirement? XI. Responses to Public Comments XII. Administrative Requirements

I. What Action Is EPA Proposing Today?

Based on a review of all available information, Clean Air Act (CAA) requirements, and relevant EPA guidance, we are proposing to approve: (1) Illinois' 1-hour ozone attainment demonstration for the Chicago- Gary-Lake County ozone nonattainment area; (2) Illinois' post-1999 ROP plan (an ROP plan covering the time period of November 15, 1999 through November 15, 2007) for the Illinois portion of the Chicago-Gary-Lake County ozone nonattainment area; (3) Illinois' contingency measures plan for both the ozone attainment demonstration and the post-1999 ROP plan; (4) Illinois' commitment to conduct a mid-course review of the ozone attainment demonstration; (5) Illinois' mobile source conformity emission budgets for VOC and NOXin the Illinois portion of the Chicago-Gary-Lake County ozone nonattainment area; and (6) Illinois' RACM analysis/demonstration for the Illinois portion of the Chicago-Gary-Lake County ozone nonattainment area (the term ``Chicago area'' is used to refer to the Illinois portion of this ozone nonattainment area).

We are proposing to modify an existing NOXemissions control waiver (the NOXemissions control waiver has been in place since January 1996) for the Chicago area. The existing NOXemissions control waiver was based on ozone modeling data showing that NOXemission reductions in the ozone nonattainment area would not contribute to attainment of the ozone standard in this nonattainment area. Ozone modeling supporting the ozone attainment demonstration addressed in this proposed rule shows that NOXemission controls on EGUs, major non-EGU boilers and turbines, and major cement kilns in the ozone nonattainment area (and statewide) are beneficial and will contribute to attainment of the 1-hour ozone standard. The attainment demonstration further shows that the ozone standard will be attained by the applicable attainment date without the use of additional NOXemission controls \1\ (beyond other NOXemission controls already implemented and/ or modeled in the ozone attainment demonstration) in the nonattainment area. Consequently, such additional NOXemission controls are in excess of what is needed to attain the ozone standard.

\1\ The additional NOXemission controls not considered in the ozone attainment demonstration include NOXRACT, NOXNSR, and additional mobile source NOXcontrols, including vehicle inspection/ maintenance (I/M) emission cutpoints.

We are proposing to modify the existing NOXwaiver to remove from the emissions control waiver the EGUs, major non-EGU boilers and turbines, and major cement kilns for which the State included emission controls in the ozone attainment demonstration. Based on the ``excess emissions'' control provisions of section 182(f)(2) of the CAA, however, we are proposing to retain the NOXwaiver for RACT, NSR, and certain requirements of transportation and general conformity, and I/M. \2\

\2\ States with NOXwaivers are still required to prepare motor vehicle emissions budgets consistent with the ozone attainment demonstrations and to use these emissions budgets in conformity analyses.

We are proposing to deny a related citizen petition to terminate the NSR portion of the NOXemissions control waiver for the Chicago area. No data have been submitted or are available showing that the existence of the waiver for NOXNSR in the Chicago area will prevent the attainment of the 1-hour ozone standard by the November 15, 2007 deadline or will delay attainment of the ozone standard by an earlier date.

II. Background Information

A. What Is a State Implementation Plan (SIP)?

Section 110 of the CAA requires states to develop air pollution control regulations (rules) and strategies to ensure that state air quality meets the National Ambient Air Quality Standards (NAAQS) established by the EPA. Each state must submit the rules and emission control strategies to the EPA for approval and promulgation into a Federally enforceable SIP.

Each Federally approved SIP protects air quality primarily by addressing air pollution at its points of origin. The SIPs can be and generally are extensive, containing many state rules or other enforceable documents and supporting information, such as emission inventories, monitoring documentation, and modeled attainment demonstrations.

B. What Is the Federal Approval Process for a SIP?

In order for state rules and emission control strategies to be incorporated into the Federally enforceable SIPs, states must formally adopt the rules and emission control strategies consistent with state and Federal requirements. This process generally includes public notice, public hearings, public comment periods, and formal adoption by state-authorized rulemaking bodies.

Once a state rule or emissions control strategy is adopted, the state submits it to us for inclusion into the SIP. We must provide public notice and must seek additional public comment regarding our proposed action on the state submission. If adverse comments are received, they must be addressed prior to any final Federal action (they are

[Page 36372]generally addressed in a final rulemaking action).

All state rules and supporting information approved by the EPA under section 110 of the Act are incorporated into Federally approved SIPs. Records of such SIP actions are maintained in the Code of Federal Regulations (CFR) at Title 40, part 52, titled ``Approval and Promulgation of Implementation Plans.'' The actual state rules which are approved are not reproduced in their entirety in the CFR, but are ``incorporated by reference,'' which means that EPA has approved the state rules with specific effective dates, has identified the rules in the CFR, and, thereby, has identified the full texts of the rules by reference.

C. What Does Federal Approval of a State Regulation Mean to Me?

Enforcement of a state rule before and after it is incorporated into a Federally approved SIP is primarily a state responsibility. After a rule is Federally approved, however, EPA is authorized under section 113 of the CAA to conduct enforcement actions against violators. Citizens are also offered legal recourse to address violations as described in section 304 of the CAA.

D. What Are the Options for Action on a State SIP Submittal?

Depending on the circumstances unique to each of the SIP submissions, we may propose one or more of several types of approval, or disapproval in the alternative (or a combination if our rulemaking process involves separable portions of a SIP submission). In addition, these proposals may identify additional state actions that may be necessary by a state before EPA may fully approve the submissions.

The CAA provides for EPA to approve, disapprove, partially approve, or conditionally approve a state's submission. The EPA must fully approve a submission if it meets the requirements of the Act. If a submission is deficient in some way, EPA may disapprove the submission. In the alternative, if portions of the submission are approvable, EPA may partially approve and partially disapprove the submission, or may conditionally approve the submission based on a state's commitment to correct the deficiency by a date certain, not later than one year from the date of EPA's final conditional approval.

The EPA has recognized that, in some limited circumstances, it may be appropriate to issue a full approval for a submission that consists, in part, of an enforceable commitment by the state. Unlike the commitment for a submission correction under a conditional approval, such an enforceable commitment can be enforced in court by EPA or citizens. In addition, this type of commitment may extend beyond one year following EPA's final approval action. Thus, EPA may accept such an enforceable commitment where it is infeasible for the state to accomplish the necessary action(s) in the short term.

E. What Ozone Nonattainment Area Is Addressed by the State Submittal Reviewed in This Proposed Rule?

The December 26, 2000 submittal of the Illinois Environmental Protection Agency (IEPA) reviewed here primarily deals with the attainment of the 1-hour ozone standard in the Chicago area. The Illinois portion of the Chicago-Gary-Lake County ozone nonattainment area includes the counties of Cook, DuPage, Kane, Lake, McHenry, and Will, and the townships of Aux Sable and Goose Lake in Grundy County and Oswego in Kendall County. The Chicago-Gary-Lake County ozone nonattainment area also includes Lake and Porter Counties in Indiana, an Indiana submittal for which is the subject of a separate review and rulemaking.

For purposes of an ozone attainment demonstration, the Chicago- Gary-Lake County ozone nonattainment area is a sub-portion of a larger ozone modeling domain, referred to as Grid M. This ozone modeling domain is further discussed in a later portion of this proposed rule. The State's submission demonstrates that attainment of the 1-hour ozone standard will occur by November 15, 2007 throughout Grid M, including within the Chicago-Gary-Lake County ozone nonattainment area.

F. What Prior EPA Rulemakings Relate to or Led to the State Submittal Reviewed in This Proposed Rule?

On December 16, 1999 (64 FR 70496), we proposed to conditionally approve the 1-hour ozone attainment demonstration for the Chicago area submitted by the IEPA on April 30, 1998. The April 30, 1998 attainment demonstration submittal was based on a range of possible emission control measures (on a number of sets of emission control measures reflecting various emission control alternatives) and did not specify a single set of emission control measures as an adopted emissions control strategy. We based our December 16, 1999 proposed conditional approval on the State's commitment to adopt and submit, by December 31, 2000, a final ozone attainment demonstration SIP revision and a post-1999 ROP plan, including the necessary State-adopted air pollution control rules needed to support and complete the ozone attainment demonstration and post-1999 ROP plan. In the alternative, we proposed to disapprove the attainment demonstration if, by December 31, 1999, the State did not adopt an emissions control strategy as supported by its modeled ozone attainment demonstration and did not submit adequate motor vehicle emission budgets for VOC and NOXfor the Chicago area that comply with EPA's transportation conformity regulations. In addition, we conditioned our approval on the State of Illinois submitting, by December 31, 1999, an enforceable commitment to conduct a mid-course review of the ozone attainment plan in 2003.

The December 16, 1999 proposed rulemaking noted that, if the EPA issued a final conditional approval of the State's April 30, 1998 submission, \3\ the conditional approval would revert to a disapproval if the State did not adopt and submit a complete SIP submission with the following elements by December 31, 2000: (1) A final adopted ozone modeling analysis that fully assesses the impacts of regional NOXemissions reductions, models a specific local emissions reduction strategy, and reconsiders the effectiveness of the existing NOXemissions control waiver (see the discussion relating to the NOXemissions control waiver below); (2) adopted emission control measures needed to meet the post-1999 ROP requirements (an ROP plan covering the period of November 15, 1999 through the ozone attainment year); and (3) local VOC and regional NOX emission control measures sufficient to support the final ozone attainment demonstration. If the State made this complete submission by December 31, 2000, we noted that we would propose action on the new submission for the purpose of determining whether to issue a final full approval of the ozone attainment demonstration.

\3\ To date, the EPA has not issued a final rule conditionally approving the State's April 30, 1998 submittal.

As noted below, the December 26, 2000 submittal reviewed here, in part, addresses a post-1999 ROP plan for the Chicago area. The post- 1999 ROP plan provides required emission reductions in addition to Illinois' 15 percent ROP plan (ROP emission reductions occurring prior to November 15, 1996) and 9 percent post-1996 ROP plan (ROP emission reductions occurring prior to November 15, 1999) for this ozone nonattainment area. On July 14, 1997

[Page 36373](62 FR 37494), we published a final rule to approve Illinois' 15 percent ROP plan. On December 18, 2000 (65 FR 78961), we published a final rule to approve Illinois' post-1996 ROP plan. These final rules address the emission control measures selected by the State to achieve required ROP emission reductions and address the State's calculation of the 1996 VOC emission target and the 1999 VOC and NOX emission targets. The December 18, 2000 final rule also addresses the State's adopted contingency measure plan for the post-1996 ROP plan and approves the 1999 motor vehicle emissions budgets associated with the ROP plan for the Chicago area.

The December 26, 2000 submittal reviewed in this proposed rule includes, as part of the ozone attainment demonstration and the post- 1999 ROP plan, regional NOXemission reductions. These regional NOXemission reductions must be reviewed in light of the fact that a NOXemissions reduction waiver exists for the Chicago-Gary-Lake County ozone nonattainment area. On January 26, 1996 (61 FR 2428), we published a final rule approving the NOXemissions control waiver based on a showing that NOXreductions would not contribute to attainment of the 1- hour ozone NAAQS. Through the January 26, 1996 rulemaking, the EPA granted exemptions from the RACT and NSR requirements for major stationary sources of NOXand from certain vehicle I/M and general conformity requirements for NOXin the ozone nonattainment areas in the Lake Michigan Ozone Study modeling domain (the Lake Michigan Ozone Study modeling domain is a sub-portion of Grid M centered on lower Lake Michigan). On February 12, 1996 (61 FR 5291), we published a final rule granting exemption from certain transportation conformity \4\ requirements for NOXin the Chicago area. Consequently, since the NOXrequirements have been waived based on a demonstration that NOXemission controls in the ozone nonattainment area are not beneficial toward attaining the ozone standard, the State may not receive credit for NOXemission controls in the ozone nonattainment area toward ROP requirements unless the State can demonstrate the opposite, i.e., that such emission controls are beneficial for attainment of the ozone standard. The State, in its December 26, 2000 submittal, is now demonstrating that certain regional NOXemission controls (including some controls on EGUs in the Chicago ozone nonattainment area) would contribute toward attainment of the ozone standard \5\. We are proposing, based on the information submitted, to revise the NOXwaiver for the Chicago nonattainment area, as further explained below.

\4\ The NOXwaiver does not include an exemption from the need for the States to adopt mobile source NOX emission budgets for the Chicago-Gary-Lake County ozone nonattainment area to support transportation and general confomity reviews. After the State has submitted and EPA has approved a motor vehicle NOXemissions budget to be used for conformity purposes, the NOXwaiver is no longer applicable for transportation or general conformity as the State must consider the NOXemissions budget when making conformity determinations.

\5\ Statewide NOXemission controls on major non-EGU boilers and turbines and major cement kilns were also considered in the ozone attainment demonstration, but specific controls on NOXsources for these source categories were not identified for the Chicago area.

G. What Is the Time Frame for EPA To Take Action on the State Submittal?

As noted above, the EPA is providing a 30 day public comment period for this proposed rule. This comment period is typical for such proposed rules and is critical in this case given the relatively tight time constraints under which the EPA is operating. To meet the time constraints of an existing consent decree between the EPA and the Natural Resources Defense Council, the EPA must complete final rulemaking approving the December 26, 2000 submittal by October 15, 2001 or must publish a proposed Federal Implementation Plan (FIP) for the Chicago area by that date.

H. What Are the Basic Components of the State Submittal and What Are the Subjects Covered in This Proposed Rule?

The December 26, 2000 Illinois submittal reviewed in this proposed rule addresses the following required plan elements: (1) An ozone attainment demonstration for the Chicago-Gary-Lake County ozone nonattainment area and the Grid M modeling domain; (2) the post-1999 ROP plan for the Chicago area; (3) contingency measures for the post- 1999 ROP plan and for the ozone attainment demonstration; and (4) motor vehicle transportation conformity emission budgets. Besides these plan elements, this proposed rule addresses the following additional issues: (1) Illinois' commitments for a mid-course review of the ozone attainment demonstration; (2) revisions to the existing NOX control waiver for the Chicago-Gary-Lake County ozone nonattainment area and a public petition requesting a removal of the NSR portion of the NOXcontrol waiver; and (3) a RACM analysis for the Chicago area. In this notice we do not respond to the public comments submitted on our December 16, 1999 proposed rule on Illinois' April 30, 1998 ozone attainment demonstration submittal. We will address those comments when we take final action on Illinois' ozone attainment demonstration and other plan elements.

III. Ozone Attainment Demonstration and Emissions Control Strategy

A. Background Information and Requirements Placed on the Ozone Attainment Demonstration

1. What Clean Air Act Requirements Apply to the State's Ozone Attainment Demonstration?

The CAA requires the EPA to establish NAAQS for certain widespread air pollutants that cause or contribute to air pollution that is reasonably anticipated to endanger public health or welfare. Clean Air Act sections 108 and 109. In 1979, EPA promulgated the 1-hour ozone standard at a level of 0.12 parts per million (ppm) (120 parts per billion [ppb]). 44 FR 8202 (February 8, 1979). Ground-level ozone is not emitted directly by sources. Rather, emissions of NOX and VOC react in the presence of sunlight to form ground-level ozone and other secondary pollutants. NOXand VOC are referred to as precursors of ozone. Control of VOC and NOXemissions is addressed in ozone control strategies to reduce peak ozone levels.

An area exceeds the 1-hour ozone standard each day in which an ambient air quality monitor records an 1-hour average ozone concentration above 0.124 ppm. An area violates the ozone standard if, over a consecutive 3-year period, more than 3 daily exceedances are recorded or are expected to occur at any monitor in the area or in its immediate downwind environs. The highest of the fourth-high daily peak ozone concentrations over the 3-year period at any monitoring site in the area is called the ozone design value for the area. The CAA required the EPA to designate as nonattainment any area that was violating the 1-hour ozone standard, generally based on the air quality monitoring data for the 3 year period from 1987 through 1989. Clean Air Act section 107(d)(4); 56 FR 56694 (November 6, 1991). The CAA further classified these areas, based on the areas' ozone design values, as marginal, moderate, serious, severe, or extreme. Clean Air Act section 181(a). Marginal nonattainment areas were suffering the least significant air quality problems and extreme nonattainment areas had the most significant air quality problems.

[Page 36374]

The control requirements and date by which attainment of the 1-hour ozone standard needs to be achieved vary with an area's classification. Marginal areas are subject to the fewest mandated control requirements and have the earliest ozone attainment date. Moderate, serious, severe, and extreme ozone nonattainment areas are subject to more stringent planning and control requirements but are provided more time to attain the standard. Serious nonattainment areas were required to attain the 1-hour ozone standard by November 15, 1999, and severe ozone nonattainment areas are required to attain the ozone standard by November 15, 2005 or November 15, 2007 depending on the areas' ozone design values. The Chicago-Gary-Lake County ozone nonattainment area is classified as severe-17 and its attainment date is November 15, 2007.

Under sections 182(c)(2) and 182(d) of the CAA, states with serious or severe ozone nonattainment areas were required to submit, by November 15, 1994, demonstrations of how the nonattainment areas would attain the 1-hour ozone standard and how they would achieve ROP reductions in VOC emissions of 9 percent for each 3-year period until the attainment date. In some cases, NOXemission reductions can be substituted for the required VOC emission reductions to achieve ROP. 2. What Is the History of the State's Ozone Attainment Demonstration and How Is It Related to EPA's NOXSIP Call?

Notwithstanding significant efforts by the states, in 1995 EPA recognized that many states in the eastern half of the United States could not meet the November 15, 1994 time frame for submitting attainment demonstration SIP revisions because emissions of NOXand VOC in upwind states (and the ozone formed by these emissions) affected these nonattainment areas and the full impact of this effect had not yet been determined. This phenomenon is called ozone transport.

On March 2, 1995, Mary D. Nichols, EPA's then Assistant Administrator for Air and Radiation, issued a memorandum to EPA's Regional Administrators acknowledging the efforts made by states but noting the remaining difficulties in making attainment demonstration SIP submittals.\6\ Recognizing the problems created by ozone transport, the March 2, 1995 memorandum called for a collaborative process among the states of the eastern half of the country to evaluate and address transport of ozone and its precursors. This memorandum led to the formation of the Ozone Transport Assessment Group (OTAG) \7\ and provided for the states to submit the attainment demonstration SIPs based on the expected time frame for OTAG to complete its evaluation of ozone transport and to take into consideration the OTAG ozone modeling results.

\6\ Memorandum, ``Ozone Attainment Demonstrations,'' issued March 2, 1995. A copy of the memorandum may be found on EPA's web site at http://www.epa.gov/ttn/oarpg/t1pgm.html.

\7\ Letter from Mary A. Gade, Director, State of Illinois Environmental Protection Agency, to the members of the Environmental Council of States (EOCS), dated April 13, 1995.

In June 1997, OTAG completed its process. OTAG submitted to EPA the results of its technical air quality modeling efforts, which quantified the impact of the transport of ozone and its precursors. OTAG recommended consideration of a range of regional NOX emission control measures.

In recognition of the length of the OTAG process, in a December 29, 1997 memorandum, Richard Wilson, EPA's then Acting Assistant Administrator for Air and Radiation, provided until April 1998 for states to submit the following elements of their attainment demonstration SIPs for serious and severe nonattainment areas: (a) Evidence that the applicable emission control measures in subpart 2 of part D of title I of the CAA were adopted and implemented or were on an expeditious course to being adopted and implemented; (b) lists of measures needed to meet the remaining ROP emissions reduction requirements and to reach attainment; (c) for severe areas only, a commitment to adopt and submit the emission control measures necessary for attainment and the ROP plans through the attainment year by the end of 2000; \8\ (d) commitments to implement the SIP control programs in a timely manner to meet ROP emission reduction milestone targets and to achieve attainment of the ozone standard; and (e) evidence of a public hearing on each state's submittal.\9\ This submission is sometimes referred to as the Phase II submission. Motor vehicle emission budgets can be established based on a commitment to adopt the measures needed for attainment and identification of the measures needed. Thus, state submissions due in April 1998, under the Wilson policy, should have also included motor vehicle emissions budgets.

\8\ In general, a commitment for severe areas to adopt by December 2000 the control measures necessary for attainment and ROP through the attainment year applies to any additional measures necessary for attainment that were not otherwise required to be submitted earlier. This memorandum was not intended to allow states to delay submission of measures required under the Clean Air Act.

\9\ Memorandum, ``Guidance for Implementing the 1-Hour Ozone and Pre-Existing PM 10 NAAQS,'' issued December 29, 1997. A copy of this memorandum may be found on EPA's web site at http://www.epa.gov/ttn/ oarpg/t1pgm.html.

Building upon the OTAG recommendations and technical analyses, in November 1997, EPA proposed action addressing the ozone transport problem. In its proposal, the EPA found that current SIPs in 22 states and the District of Columbia (23 jurisdictions) did not meet the requirements of section 110(a)(2)(D) of the CAA because they did not adequately regulate statewide NOXemissions that significantly contribute to ozone nonattainment in downwind states. 62 FR 60318 (November 7, 1997). The EPA finalized that rule in September 1998, calling on the 23 jurisdictions to revise their SIPs to require NOXemission reductions within each jurisdiction to a level consistent with a NOXemission budget identified in the final rule. 63 FR 57356 (October 27, 1998). The final rule is commonly referred to as the NOXSIP Call. 3. What Are the Modeling Requirements for the Ozone Attainment Demonstrations?

The EPA provides that states may rely on a modeled attainment demonstration supplemented with additional evidence to demonstrate attainment of the ozone standard.\10\ In order to have complete ozone modeling attainment demonstration submissions, states should have submitted the required modeling analyses and identified any additional evidence that EPA should consider in evaluating whether areas will attain the ozone standard.

\10\ The EPA issued guidance on air quality modeling that is used to demonstrate attainment of the 1-hour ozone NAAQS. See U.S. EPA (1991), Guideline for Regulatory Application of the Urban Airshed Model, EPA-450/4-91-013 (July 1991). A copy may be found on EPA's web site at http://www.epa.gov/ttn/scram/ (file name: ``UAMREG''). See also U.S. EPA (1996), Guidance on Use of Modeled Results to Demonstrate Attainment of the Ozone NAAQS, EPA-454/B-95- 007 (June 1996). A copy may be found on EPA's web site at http:// www.epa.gov/ttn/scram/ (file name: ``03TEST'').

For the purposes of demonstrating attainment of the ozone standard, the CAA (section 182(c)(2)(A)) requires states with serious and severe ozone nonattainment areas to use photochemical dispersion modeling or an analysis method EPA determines to be as effective to assess the adequacy of emission control strategies and to demonstrate attainment of the ozone standard. The photochemical dispersion modeling system is set up using

[Page 36375]observed meteorological conditions conducive to the formation of ozone. The meteorological conditions are selected based on historical data for high ozone periods in the nonattainment area or in its associated modeling domain. Emissions for a base year and monitored ozone and ozone precursor concentrations are used to evaluate the modeling system's ability to reproduce actual monitored air quality values (ozone and other associated pollutants). Following validation of the modeling system for the base year, emissions are projected to an attainment year and modeled in the photochemical modeling system to predict air quality levels in the attainment year. Projected emission changes include source emissions growth up to the attainment year and emission controls implemented by the attainment year.

A modeling domain is chosen that encompasses the ozone nonattainment area and surrounding upwind and downwind areas. Attainment of the ozone standard is demonstrated when all predicted ozone concentrations in the attainment year in the modeling domain are at or below the ozone NAAQS or at an acceptable upper limit above the NAAQS permitted under certain conditions as explained in EPA's guidance. An optional Weight-Of-Evidence (WOE) determination may be used to address uncertainty inherent in the application of photochemical grid models. See the discussion of possible WOE determination tests and analyses below.

The EPA guidance identifies the features of a modeling analysis that are essential to obtain credible results. First, the State must develop and implement a modeling protocol. The modeling protocol describes the methods and procedures to be used in conducting the modeling analyses and provides for policy oversight and technical review by individuals responsible for developing or assessing the attainment demonstration (state and local agencies, EPA regional offices, the regulated community, and public interest groups). Second, for purposes of developing the information to put into the model, the state must select historical high ozone days (days with ozone concentrations exceeding the ozone standard) that are representative of the ozone pollution problem for the nonattainment area. Third, the state needs to identify the appropriate dimensions of the area to be modeled, i.e., the modeling domain size. The modeling domain should be larger than the designated ozone nonattainment area to reduce uncertainty in the nonattainment area boundary conditions and should include any large upwind sources just outside of the ozone nonattainment area. In general, the modeling domain is considered to be the area where control measures are most beneficial to bring the area into attainment of the ozone NAAQS. Fourth, the state needs to determine the modeling grid resolution (the modeling domain is divided into a three-dimensional grid). The horizontal and vertical resolutions in the modeling domain affect the modeled dispersion and transport of emission plumes. Artificially large grid cells (too few vertical layers and horizontal grids) may artificially dilute pollutant concentrations and may not properly consider impacts of complex terrain, meteorology, and land/water interfaces. Fifth, the state needs to generate meteorological data and emissions that describe atmospheric conditions and inputs reflective of the selected high ozone days. Finally, the state needs to verify that the modeling system is properly simulating the chemistry and atmospheric conditions through diagnostic analyses and model performance tests (generally referred to as model validation). Once these steps are satisfactorily completed, the model is ready to be used to generate air quality estimates, to evaluate emission control strategies, and to support an ozone attainment demonstration.

The modeled attainment test compares model-predicted 1-hour daily maximum ozone concentrations in all grid cells for the attainment year (2007 for the Chicago-Gary-Lake County ozone nonattainment area), with all selected emission control measures in place, to the level of the ozone NAAQS. A predicted peak ozone concentration above 0.124 ppm (124 ppb) indicates that the area may exceed the ozone standard in the attainment year under the tested base year conditions and that the tested emissions control strategy may be inadequate to attain the ozone standard. This type of test is referred to as an exceedance test. EPA's guidance recommends that states use either of two modeled attainment or exceedance tests for the ozone attainment demonstration, a deterministic test or a statistical test.

The deterministic test requires a state to compare predicted 1-hour daily maximum ozone concentrations for each modeled day \11\ to the attainment level of 0.124 ppm. If none of the predictions exceed 0.124 ppm, the test is passed.

\11\ The initial, ``ramp-up'' day for each modeled high ozone episode is excluded from this determination.

The statistical test takes into account the fact that the 1-hour ozone NAAQS allows exceedances. If, over a 3-year period, an area has an average of 1 or fewer daily exceedances per year at any monitoring site, the area is not violating the ozone standard. Thus, if the state models an extreme day, considering meteorological conditions that are very conducive to high ozone levels, the statistical test provides that a prediction of an 1-hour ozone concentration above 0.124 ppm up to a certain upper limit may be consistent with attainment of the standard.

The acceptable upper limit for modeled peak ozone concentrations in the statistical test is determined by examining the levels of ozone standard exceedances at monitoring sites which meet the 1-hour ozone NAAQS. For example, a monitoring site for which the four highest 1-hour average ozone concentrations over a 3-year period are 0.136 ppm, 0.130 ppm, 0.128 ppm, and 0.122 ppm is attaining the standard. To identify an acceptable upper limit, the statistical likelihood of observing ozone air quality exceedances of the standard of various concentrations is equated to the relative severity of the modeled day. The upper limit generally represents the maximum ozone concentration observed at a location on a single day, and would be the only ozone reading above the standard that would be expected to occur no more than an average of once a year over a 3-year period. Therefore, if the maximum ozone concentration predicted by the model is below the acceptable upper limit, in this case 0.136 ppm, then EPA might conclude that the modeled attainment test is passed. Generally, exceedances well above 0.124 ppm are very unusual at monitoring sites meeting the ozone NAAQS. Thus, these upper limits are rarely substantially higher than the attainment level of 0.124 ppm. 4. What Additional Analyses May Be Considered When the Ozone Modeling Fails To Show Attainment of the Ozone Standard?

When the ozone modeling does not conclusively demonstrate attainment of the ozone standard through either a deterministic test or a statistical test, additional analyses may be presented to help determine whether the area nevertheless will attain the standard. As with other predictive tools, there are inherent uncertainties in some of the photochemical modeling inputs, such as the meteorological and emissions data bases for individual days and in the methodology used to assess the severity of an exceedance at individual sites. EPA's guidance recognizes these

[Page 36376]limitations, and provides a means for considering other evidence to help assess whether attainment of the NAAQS is likely. The process by which this is done is the WOE determination.\12\

\12\ States may choose to submit WOE determinations even when the ozone modeling results pass either the deterministic test or the statistical test. This may be done to support the attainment demonstration, recognizing that the ozone modeling results possess a certain degree of uncertainty. Nonetheless, the submittal of WOE determinations is only needed if the ozone modeling fails to demonstrate attainment of the ozone standard through either a deterministics test or a statistical test.

Under a WOE determination, a state can rely on and EPA will consider factors such as: Other modeled attainment tests, e.g., a rollback analysis; other modeled outputs, e.g., changes in the predicted frequency and pervasiveness of ozone standard exceedances and predicted changes in an area's ozone design value; actual observed air quality trends; estimated emissions trends; analyses of air quality monitoring data; the responsiveness of the model predictions to further emission controls; and, whether there are additional emission control-- measures that are or will be approved into the SIP but that were not included in the ozone modeling analysis. This list is not an exhaustive list of factors that may be considered, and the factors considered could vary from case to case. EPA's guidance contains no limit on how close a modeled attainment test (a deterministic test or a statistical test) must be to passing to conclude that other evidence besides an attainment test is sufficiently compelling to suggest attainment. The further a modeled attainment test is from being passed, however, the more compelling the WOE determination needs to be.

EPA's 1996 modeling guidance also recognizes a need to perform a mid-course review as a means for addressing uncertainty in the modeling results, particularly if a WOE determination is needed to support an ozone attainment demonstration. Because of the uncertainty in long-term projections, EPA believes a viable attainment demonstration that relies on a WOE determination needs to contain provisions for periodic review of monitoring, emissions, and modeling data to assess the extent to which refinements to emission control measures are needed. The mid- course review is further discussed below. 5. Besides the Modeled Attainment Demonstration and Adopted Emission Control Strategy, What Other Elements Must Be Addressed in the Attainment Demonstration SIP?

In addition to the modeling analysis and WOE determination supporting the attainment demonstration, the EPA has identified the following key elements which must also be adopted by the state and approved by the EPA in order for EPA to approve the 1-hour ozone attainment demonstration SIPs.

a. Clean Air Act Measures, and Other Measures Relied on in the Modeled Attainment Demonstration. This includes adopted and submitted rules for all Clean Air Act required measures for the specific area classification. This also includes measures that may not be required given the area's ozone classification but that the state relied on in its attainment demonstration or in its ROP plan.

The state should have adopted the emission control measures required under the CAA for the area's ozone nonattainment classification. In addition, states with severe ozone nonattainment areas had until December 2000 to adopt and submit additional emission control measures needed to achieve ROP through the attainment year and to attain the ozone standard. For purposes of fully approving a state's SIP, the state needs to adopt and submit rules for all VOC and NOXcontrols within the ozone modeling domain and within the state that are relied on to support the modeled ozone attainment demonstration.

Table I presents a summary of the CAA requirements that need to be met for each severe ozone nonattainment area. These requirements are specified in section 182 of the CAA. Information on more measures that states may have adopted or relied on in their current SIP submissions is not shown in the table.

Table I.--CAA Requirements for Severe Ozone Nonattainment Areas

NSR Requirements for VOC and NOX, Including an Offset Ratio of 1.3:1 and a Major Source VOC and NOXEmissions Threshold of 25 Tons Per Year \13\. RACT for VOC and NOX\14\. Enhanced Vehicle I/M. 15 percent VOC control plan for ROP through 1996. 3 percent VOC/ NOXReduction Per Year Through the Ozone Standard Attainment Year for ROP \15\ RACM. Contingency Measures. Base Year Emissions Inventory. Emission Statement Rules Requiring Sources to Periodically Submit Summaries of Their VOC and NOXEmissions. Ozone Attainment Demonstration. Clean Fuels Fleet Program. Enhanced Ambient Monitoring (Photochemical Assessment Monitoring System [PAMS]). Stage II Gasoline Vapor Recovery At Retail Service Stations. Reformulated Gasoline. Measures to Offset Growth in Vehicle Miles Tranvelled (VMT).

b. NOXReductions Affecting Boundary Conditions. EPA completed final rulemaking on the NOXSIP Call on October 27, 1998, requiring states to address transport of NOXand ozone to other states. To address transport, the NOXSIP Call established state-specific emission budgets for NOX that 23 jurisdictions were required to meet through enforceable SIP emission control measures adopted and submitted by September 30, 1999. The NOXSIP Call is intended to reduce emissions in upwind states that significantly contribute to downwind ozone nonattainment problems. The EPA did not identify specific NOXsources that the states must regulate nor did the EPA limit the states' choices regarding where within the states to achieve the emission reductions.

\13\ The NOXNSR requirements do not currently apply in the Chicago area based on a NOXwaiver granted to Illinois on January 26, 1996 (61 FR 2428).

\14\ The NOXRACT requirements do not currently apply in the Chicago area based on a NOXwaiver granted to Illinois on January 26, 1996 (61 FR 2428).

\15\ To provide interim progress, EPA accepted 9 percent VOC/ NOXemission reduction plans to cover ROP requirements between 1996 and 1999. The States with severe nonattainment areas were required to meet the remainder (post-1999) of the ROP requirements through the submittal of a final ROP plan with adopted emission control regulations by December 2000. The Illinois post- 1999 ROP plan is reviewed later in this proposed rule.

On May 25, 1999, the U.S. Court of Appeals for the District of Columbia issued an order staying the SIP submission requirement of the NOXSIP Call. On March 3, 2000, the Court issued its decision, which largely upheld EPA's final NOXSIP Call rule, with certain exceptions that do not affect this proposed rule. On June 23, 2000, the Court lifted the stay. On August 30, 2000, the Court issued an order providing that EPA could not require SIPs to include a source control implementation date earlier than May 31, 2004.

Emission reductions that will be achieved through EPA's NOXSIP Call

[Page 36377]will reduce the levels of ozone and ozone precursors entering ozone nonattainment areas and ozone modeling domains at their boundaries and will reduce the NOXemissions generated within the ozone modeling domains. The ozone levels at the boundary of the local modeling domain are reflected in modeled attainment demonstrations and are, along with the concentrations of pollutants entering the modeling domain, referred to as boundary conditions. The boundary conditions and the ozone generated and transported within the modeling domain will be impacted by the NOXemission reductions resulting from the NOXSIP Call in many areas. Therefore, EPA believes it is appropriate to allow states to continue to assume the NOX emission reductions resulting from the NOXSIP Call in areas outside of the local ozone modeling domains. If states assume emission reductions other than those of the NOXSIP Call within their states but outside of the ozone modeling domains, the states must also adopt emission control regulations to achieve those additional emission reductions in order to have an approvable ozone attainment demonstration. States subject to the NOXSIP Call, particularly those relying on the NOXSIP Call-based emission reductions as part of their ozone attainment demonstrations, are expected to have adopted the NOXemission control regulations needed to comply with the NOXSIP Call. In these areas, approval of the ozone attainment demonstration is dependent on the approval of the NOXemission control regulations.

As provided above, any emission controls assumed by a state within a local ozone modeling domain must be adopted by the state and approved by us to achieve our final approval of the state's 1-hour ozone attainment demonstration SIP.

c. Motor Vehicle Emissions Budget. The EPA believes that attainment demonstration and ROP SIPs must necessarily estimate the motor vehicle VOC and NOXemissions that will be produced in the attainment and milestone years and must demonstrate that these emissions, when considered with emissions from all other sources, is consistent with attainment of the ozone standard and ROP. The estimate of motor vehicle emissions is used to determine the conformity of transportation plans and programs to the SIP, as described by section 176(c)(2)(A) of the Act. For transportation conformity purposes, the estimate of motor vehicle emissions is known as the motor vehicle emissions budget. EPA believes that appropriately identified motor vehicle emissions budgets are a necessary part of attainment demonstration and ROP SIPs, and that EPA must find these budgets to be adequate before we can give final approval to the attainment demonstration and ROP SIPs.

d. Mid-Course Review. An enforceable commitment to conduct a mid- course review (MCR) and evaluation of the attainment demonstration based on air quality and emissions trends at some time prior to the attainment year must be included in the attainment demonstration SIP before it can be approved by the EPA, particularly if the SIP depends on a WOE determination to demonstrate attainment of the ozone standard. The MCR shows whether the adopted emission control measures and emissions control strategy (all measures combined into a single plan) are sufficient in timing and extent to reach attainment of the ozone standard by the area's attainment deadline, or whether additional emission control measures may be necessary.

A MCR is a reassessment of the modeling analyses and more recent monitoring and emissions data to determine if a prescribed emissions control strategy is resulting in emission reductions and air quality improvements needed to attain the ozone standard as expeditiously as practicable but no later than the statutory attainment date. The EPA believes that an enforceable commitment to perform a MCR is a critical element of a WOE determination.

For severe areas, such as the Chicago-Gary-Lake County ozone nonattainment area, the state(s) must submit an enforceable commitment (Illinois has submitted such a commitment as discussed below). The commitment must provide the date by which the MCR will be completed. The EPA believes that the MCR process should be done immediately following the ozone season (April through October in Illinois) in which the states have implemented the NOXregulations resulting from the NOXSIP Call and that the states should submit the results to us by the end of that calendar year. Because the Court of Appeals ordered that EPA cannot require states to establish a NOXsource compliance date prior to May 31, 2004, EPA believes that the MCR should be performed following the 2004 ozone season and that the results should be submitted by the end of 2004.

Following submittal of MCR analysis results, we would review the results and determine whether the state(s) needs to adopt and submit additional emission control measures for purposes of attainment. We are not requesting that states commit now to adopt new emission control measures as a result of this process. It would be impractical for the states to make a commitment for such control measures that is specific enough to be considered enforceable. Moreover, the MCR could indicate that upwind states may need to adopt some or all of the additional emission controls needed to ensure that a downwind state/area attains the ozone standard. We would determine whether additional emission controls are needed in the state in which a nonattainment area is located or in upwind states, or in both. We would require the appropriate state(s) to adopt and submit new emission control measures within a period specified at that time. We anticipate that these findings would be made as SIP Calls under section 110(k)(5) of the Act and, therefore, the period for the submission of the measures would be no longer than 18 months after we make a finding. A guidance document regarding the MCR process is located on EPA's web site at http:// www.epa.gov/ttn/scram. The EPA is working on additional guidance that it expects to issue and put on its website shortly. 6. What Are the Relevant EPA Policy and Guidance Documents?

The relevant policy documents for ozone attainment demonstrations and their locations on EPA's web site are listed below:

a. U.S. EPA, Guideline for Regulatory Application of the Urban Airshed Model, EPA-450/4-91-013, (July 1991), Web site: http:// www.epa.gov/ttn/scram/ (file name: ``UAMREG'').

b. U.S. EPA, Guidance on Use of Modeled Results to Demonstrate Attainment of the Ozone NAAQS, EPA-454/B-95-007, (June 1996), Web site: http://www.epa.gov/ttn/scram/ (file name: ``O3TEST'').

c. Memorandum, ``Ozone Attainment Demonstrations,'' from Mary D. Nichols, issued March 2, 1995, Web site: http://www.epa.gov/ttn/oarpg/t1pgm.html.

d. Memorandum, ``Extension of Attainment Dates for Downwind Transport Areas,'' issued July 16, 1998, Web site: http://www.epa.gov/ ttn/oarpg/t1pgm.html.

e. Memorandum, ``Guidance for Implementing the 1-Hour Ozone and Pre-Existing PM10NAAQS,'' from Richard Wilson, issued December 29, 1997, Web site: http://www.epa.gov/ttn/oarpg/t1pgm.html.

f. ``Guidance for Improving Weight of Evidence Through Identification of

[Page 36378]Additional Emission Reductions, Not Modeled,'' U.S. EPA, Office of Air Quality Planning and Standards, November 1999, Web site: http:// www.epa.gov/ttn/scram/ (file name: ``ADDWOE1H'').

g. ``Serious and Severe Ozone Nonattainment Areas: Information on Emissions, Control Measures Adopted or Planned and Other Available Control Measures,'' Draft Report, U.S. EPA, Ozone Policy and Strategies Group, November 3, 1999.

h. Memorandum, ``Guidance on Motor Vehicle Emissions Budgets in 1- hour Attainment Demonstrations,'' from Merrylin Zaw-Mon, Office of Mobile Sources, November 3, 1999, Web site: http://www.epa.gov/oms/ transp/traqconf.htm.

i. Memorandum, ``1-Hour Ozone Attainment Demonstrations and Tier 2/ Sulfur Rulemaking,'' from Lydia Wegman and Merrylin Zaw-Mon, Office of Air Quality Planning and Standards and Office of Mobile Sources, November 8, 1999, Web site: http://www.epa.gov/oms/transp/traqconf.htm.

j. Draft Memorandum, ``1-Hour Ozone NAAQS-Mid-Course Review Guidance,'' from John Seitz, Director, Office of Air Quality Planning and Standards, Web site: http://www.epa.gov/ttn/scram/.

B. Technical Review of the State's Submittal

1. When Was the Attainment Demonstration Addressed in Public Hearings, and When Was the Attainment Demonstration Submitted to the EPA?

The State of Illinois held a public hearing on the ozone attainment demonstration on November 8, 2000. The attainment demonstration was submitted by the Illinois Environmental Protection Agency (IEPA) on December 26, 2000. 2. What Are the Basic Components of the Submittal?

Since Illinois, along with Indiana, Michigan, and Wisconsin, jointly participates in the Lake Michigan Air Directors Consortium (LADCO) and since LADCO has conducted the ozone analyses used to develop the ozone attainment demonstration, technical support documents developed by LADCO form the main bases for Illinois' ozone attainment demonstration. Three documents from LADCO provide much of the technical support for the attainment demonstration. These documents are:

a. ``Midwest Subregional Modeling: 1-Hour Attainment Demonstration for Lake Michigan Area--Summary,'' LADCO, September 18, 2000;

b. ``Technical Support Document--Midwest Subregional Modeling: 1- Hour Attainment Demonstration for Lake Michigan Area,'' LADCO, September 18, 2000; and

c. ``Technical Support Document--Midwest Subregional Modeling: Emissions Inventory,'' LADCO, September 27, 2000.

Illinois, like Indiana and Wisconsin, has included a state-specific cover letter and a state-specific synopsis of the ozone attainment demonstration. Illinois has also included additional modeling analysis results to address emissions changes not addressed in the earlier LADCO analyses. These emission changes include increased state-wide NOXand VOC emissions due to the permitting and implementation of new combustion turbine generators (peakers or peaker plants and combined cycle facilities) designed to supplement electrical power generation on high demand days (many high electricity demand days are potentially high ozone days due to high ambient temperatures) and to replace the electrical generating capacity of electrical--generating facilities taken off-line. Additional VOC and NOXemissions due to higher-than-planned vehicle miles of travel in the planning area are also considered.

A number of other related submittal components are discussed in later sections of this proposed rule. This section deals exclusively with the technical aspects of Illinois' 1-hour ozone attainment demonstration, focusing on the ozone modeling results and supporting air quality and emissions analyses. 3. What Modeling Approach Was Used in the Analyses To Develop and Validate the Ozone Modeling System?

The LADCO States, as participants in the Lake Michigan Ozone Study (designed to establish the modeling system and its base input data and to validate the modeling system) and in the Lake Michigan Ozone Control Program (designed to select and test possible emission control strategies), used the same modeling approach to develop the basis for each State's ozone attainment demonstration although each State selected a different emissions control strategy for their respective ozone attainment demonstration. The modeling approach is documented in LADCO's September 18, 2000 Technical Support Document (TSD) and is summarized in LADCO's September 18, 2000 modeling summary (see above).

The heart of the modeling system is the Urban Airshed Model-Version V (UAM-V) photochemical dispersion model developed originally for specific application in the Lake Michigan area. This is the same version of the model that was used during the OTAG analysis of ozone transport and ozone transport control measures.

For purposes of the local ozone attainment demonstration, UAM-V was applied to a local modeling domain and grid configuration that was established based on consideration of areas of high ozone concentrations (generally the ozone nonattainment areas) in the Lake Michigan States and of possible upwind source areas impacting these high concentration areas. The primary modeling domain is referred to as Grid M. This grid extends east to the most eastern portion of Michigan (and to central Ohio, eastern Kentucky, and eastern Tennessee); north to the northern end of Michigan's Lower Penninsula (and to the north of Green Bay, Wisconsin); west to include the eastern thirds of Iowa and Missouri; and south to the southern border of Tennessee. The horizontal grid is rectangular in shape (see Figure 1 of the September 18, 2000 LADCO TSD). The modeling has the following horizontal and vertical resolutions:

Horizontal Resolutions

Approximately 12 kilometers x 12 kilometers--all modeling runs.

Approximately 4 kilometers x 4 kilometers--for selected runs to give better resolution in the area along the western shore of Lake Michigan.

Vertical Resolution

7 vertical layers with the following height ranges (above terrain) in meters: 0-50; 50-100; 100-250; 250-500; 500-1500; 1500-2500; and 2500-4000.

A subregional portion of the grid, centered (east to west) on the lower portion of Lake Michigan, was also considered to allow a more detailed analysis of the high ozone areas of Grid M. The use of Grid M and the subregional portion of Grid M allowed the consideration of both urban scale analyses and ozone transport. It should be noted that the modeling results from the modeling runs with the tighter 4 kilometer resolution were generally consistent with the results for the 12 kilometer resolution.

Four high ozone episodes in the Lake Michigan area were modeled. These episodes were: June 22-28, 1991; July 14-21, 1991; June 13-25, 1995; and July 7-18, 1995. These episodes were selected because: (1) They were judged to be representative of typical high ozone episodes in the Lake Michigan area and because they respresent a variety of meteorological conditions that have been found to be conducive to high

[Page 36379]ozone concentrations in this area; (2) there is an intensive data base available for the 1991 episodes; and (3) several of these episodes (the July episodes) were modeled as part of the OTAG analyses, providing ozone transport and modeling domain boundary data.

The following input data systems and analyses were used to develop input data for the ozone model:

a. Emissions. UAM-V requires a regional inventory of gridded, hourly estimates of speciated VOC, NOX, and carbon monoxide (CO) emissions. The States provided emission inventories which were processed through the Emissions Modeling System-1995 version (EMS-95). Emissions were prepared for a 1996 base year (used to test model performance), a 2007 base year (considering growth and previously adopted emission control measures), and several 2007 emission control strategy/sensitivity scenarios. The emission inventories include 1996 state periodic inventory data for stationary point and area sources, updated state transportation data, excess NOXemissions produced by heavy-duty vehicles as a result of built-in ``defeat'' devices, updated growth and emissions control data, and EPA's latest emission reduction estimates for the mobile source Tier II/Low Sulfur program. Ambient temperature data affecting mobile source and evaporative emissions and biogenic emissions were generated using the RAMS3a meteorological model. Biogenic emissions were based on EPA's BEIS2 model, with an adjustment of the isoprene emissions in the Ozarks.\16\ Point source emissions for some sources were addressed through the use of Plume-in-Grid (PiG) \17\ techniques incorporated within UAM-V. An additional discussion of the development of the modeled emission inventories is presented below.

\16\ Analyses of initial ozone modeling results indicated that initial isoprene emission estimates for the Ozarks had unrealistic impacts on the ozone concentrations modeled for the Lake Michigan area. Background ozone monitoring data did not support the high background/transported ozone levels modeled to result from this upwind source area. A study, known as OZIE, was conducted to reanalyse the isoprene emissions for the Ozarks. Based on the preliminary results of the OZIE study, LADCO concluded that the isoprene emissions for the Ozarks should be reduced by a factor of 2 (halved).

\17\ sources to be addressed through PiG techniques were selected based on their magnitudes of NOXemissions (the top 100 ranked stacks) and locations (the next 34 topped ranked stacks in the Lake Michigan and St. Louis areas).

b. Meteorology. UAM-V requires gridded 3-dimensional hourly values of wind speed, wind direction, temperatures, air pressure, water vapor content, vertical diffusivity, and, if applicable, clouds and precipitation. Most meteorological inputs were derived through prognostic modeling with the RAMS3a model. Cloud and precipitation data were developed based on observed National Weather Service data. Preliminary analyses of the modeled meteorological data results showed adequate representation of the observed airflow features and good agreement between modeled and measured wind speeds, temperatures, and water vapor levels. LADCO, has concluded, however, that errors or uncertainties in the meteorological data may have affected the UAM-V results (albeit not significantly enough to invalidate the modeling results based on EPA recommended validation criteria). The errors have been minimized to the extent possible and suppressed through ``nudging'' using observed National Weather Service data at 12-hour intervals.

c. Boundary Conditions. Boundary conditions were developed by applying UAM-V over the OTAG modeling domain (this modeling domain covered most of the eastern half of the United States) for the selected high ozone episodes at a 36 kilometer grid resolution. The modeling was conducted to be consistent with the modeling used in the OTAG analyses.

Basecase modeling was conducted to evaluate model performance by comparing observed and modeled ozone concentrations. The model performance evaluation consisted of comparisons of the spatial patterns, temporal profiles, and magnitudes of modeled and measured 1- hour (and 8-hour) ozone concentrations.

In making the comparison of modeled and observed ozone concentrations, 1996 emissions were assumed to be reasonably similar to 1995 emissions, but significantly lower than 1991 emissions. To account for the 1991-1996 differences, a set of simple ``backcast'' emission factors were derived by comparing the county-level emissions in the 1991 Lake Michigan Ozone Control Program emissions inventory with the 1996 base year emissions inventory.

Peak daily 1-hour modeled ozone concentrations for each episode were analyzed and compared to the observed peak ozone levels in the modeling domain. For each type of comparison, the following conclusions were developed.

Spatial Patterns. This analysis showed that areas of high modeled ozone concentrations correspond acceptably with areas of high measured ozone concentrations in the Lake Michigan area. Rural (generally upwind of the Lake Michigan ozone nonattainment areas) measured and modeled ozone concentrations were found to compare favorably. Peak modeled ozone concentrations over Lake Michigan, however, appear to be underestimated on some days.

Temporal Patterns. Time series plots of 1-hour modeled and measured ozone concentrations by monitoring site were compared. The hour-to-hour and day-to-day variations of modeled and measured ozone concentrations were found to compare favorably. The modeling system seems to over-predict nighttime ozone concentrations and to under- predict peak daytime ozone concentrations, but performs within acceptable limits (see a discussion of the modeling validation below). At the monitoring sites with high measured ozone concentrations, the mid-afternoon modeled ozone concentrations are low.

Magnitude Comparisons. Ozone statistics, unpaired peak accuracy, average accuracy of peak ozone concentrations, normalized bias results, and normalized gross error results are provided in the modeling system documentation. The model performance statistics for the Lake Michigan modeling domain subregion comply with EPA's recommended acceptance ranges. The statistics of the modeling system performance, however, demonstrate the tendency of the modeling system to underestimate measured peak ozone concentrations (although the modeling system overestimated some of the peak ozone concentrations).

Other Factors. The modeling system's response to changes in ozone precursor emissions has been assessed by conducting sensitivity analyses and by comparing the differences in modeled and measured ozone concentrations and changes in emissions between 1991 and 1996. This assessment indicates that the model is responsive to changes in ozone precursor emissions and is consistent with observed air quality data and emissions data.

To assess the effects of grid resolution, analyses were conducted comparing modeling results for resolutions of 4 kilometers and 12 kilometers. Plots of predicted peak concentrations were analyzed for these two grid resolutions. In general, it appears that model performance at a resolution of 4 kilometers is comparable to that at a resolution of 12 kilometers.

The LADCO States have concluded that the modeling system performance is acceptable for air quality planning purposes (for the purposes of assessing

[Page 36380]the impacts of emission control strategies).

To test ozone attainment strategies, the LADCO States have projected emissions from the base year to 2007, the attainment year. The future emissions have been modified to reflect the various tested emission control strategies.\18\ All other inputs to the ozone modeling system have been fixed at the levels used in the validated base year modeling analyses.

\18\ For a listing of the emission control measures modeled in the various emission control strategies, see Table 6, ``Control Measures,'' in LADCO's September 27, 2000 ``Technical Support Document: Midwest Subregional Modeling: Emissions Inventory'' or Section 5, ``Strategy Modeling,'' and Table 4, ``Control Measures,'' of LADCO's September 18, 2000 ``Technical Support Document: Midwest Subregional Modeling; 1-Hour Attainment Demonstration for Lake Michigan Area,'' both of which were included in Illinois' December 26, 2000 attainment demonstration submittal.

The remainder of the questions in this section of this proposed rule address the States' efforts to demonstrate attainment using the validated ozone modeling system and focuses on evaluating the attainment strategy. For additional discussions of the efforts to validate the modeling system, you are referred to the discussions of these efforts in the December 16, 1999 proposed rule (64 FR 70496). 4. How Were the 1996 Base Year Emissions Developed?

Besides being used to develop and validate the ozone modeling system, base year emissions were also used to project the attainment year emissions and, through comparisons with the attainment year emissions and analyses of monitored and modeled ozone concentrations, to support the adequacy of the selected emissions control strategy. For the purposes of the attainment demonstration used here, 1996 was selected to be the base year of the analyses.

The September 27, 2000 LADCO emissions inventory TSD documents the development of the base year emissions as well as the projection and development of the attainment year emissions used in the attainment strategy modeling and attainment demonstration. The following summarizes the development of base year emissions as documented in LADCO's September 27, 2000 TSD.

For the 1996 base year, emission rates for point and area sources were either provided by the EPA (from the NOXSIP Call documentation) or by the States based on 1996 periodic emission inventories. Where appropriate, EPA's NOXdata were supplemented or corrected using state-specific data, as noted in LADCO's September 27, 2000 TSD.

Emission rates for on-road mobile sources were calculated through the use of EMS-95 based on a mobile source activity level, e.g., vehicle miles traveled (VMT), and the MOBILE5b emission factor model. The sources of the VMT, vehicle speed, and vehicle mix data are summarized in LADCO's September 27, 2000 TSD. Relative to previous emissions modeling, vehicle speeds were increased and vehicle mix distributions were shifted to heavier vehicles based on more recent data (the increased use of sports utility vehicles has increased the relative vehicle mixes of light duty gasoline trucks, increasing per VMT emissions rates). Mobile source emissions of NOXwere also increased for heavy-duty diesel vehicles as the result of the use of built-in ``defeat'' devices. These increased NOX emissions were estimated by applying a processor supplied by the EPA.

Day-specific biogenic emissions were calculated using EPA's BEIS2 model. As noted above, comparisons of emission estimates and measured isoprene concentrations in the Ozarks indicated that the BEIS 2 isoprene emission estimates for the Ozarks are overestimated by a factor of 2.

As noted above, a number of refinements of the emissions estimates must be made to support the ozone modeling system. These refinements include spatial, temporal, and species processing and resolution. This was accomplished through the use of EMS-95. County-level point source emissions were spatially distributed based on facility or stack coordinates. County-level area source emissions were spatially resolved based on surrogates, such as population distributions and land use data. Mobile source emissions were calculated for each modeling grid cell by EMS-95, not requiring further resolution.

Daily average point source emissions were temporally allocated based on using facility-specific reported operating schedule information. Daily average area source emissions were temporally allocated using category-specific hourly distribution profiles. Mobile source and biogenic source emissions are temporally resolved through the use of EMS-95, which includes temporal emission profiles for these source categories.

The speciation profiles in EMS-95 were obtained from the latest version of EPA's SPECIATE data base.

To quality assure the base year emissions data, a top-down evaluation of the emissions inventory was performed using ambient ozone precursor data collected from the Photochemical Assessment Monitoring Stations (PAMS) in the Lake Michigan area. The evaluation included comparisons of monitored and calculated VOC to NOXemissions ratios, the relative amounts of individual VOC species, and the measured and calculated reactivity of VOC compounds. 5. What Procedures and Sources of Projection Data Were Used To Project the Emissions to the Attainment Year?

The future year emission inventories used in the Lake Michigan Ozone Control Program and in the ozone attainment demonstration were derived from the base year emissions inventory. The base year emissions inventory was projected to 2007 by applying scalar growth factors for most source categories. Each LADCO State provided estimates of source growth and control factors by source sector. Source growth and emission control factors used in EPA's NOXSIP Call were also considered, particularly for EGUs. Table 1 of the LADCO September 27, 2000 TSD documents in detail the sources of 2007 emission estimates by source category along with the sources of 1996 emissions and emission control factors and is included by reference here. 6. How Were the 1996 and 2007 Emission Estimates Quality Assured?

To improve the reliability of the modeling source emission inventories, several quality assurance activities were performed by the State emission inventory personnel, the emission modelers (those people responsible for speciating and temporally and spatially resolving the emissions data for use in the ozone modeling system), and the photochemical modelers. These activities included:

Development and Implementation of an Emissions Quality Assurance Plan. A standardized set of data and file checks were documented in a LADCO draft emissions quality assurance (QA) plan. This plan identifies the emissions quality assurance procedures to be followed by the State emission inventory personnel. Each State was responsible for quality assurance of its own emissions inventory data before providing these data to the LADCO emission modelers. The quality assurance of the data by the States included reviewing many EMS-95 emissions reports for consistency with other State-specific emissions data.

Emission Reports. EMS-95 itself performs a number of emission checks and generates reports flagging possible emission errors and summarizing data

[Page 36381]that can be checked against alternative emission data sets/reports. Table 7 of LADCO's September 27, 2000 TSD lists the EMS-95 standardized QA reports and is included by reference here. These reports were generated in the preparation of the Grid M emissions data and were used for QA efforts.

Review by Photochemical Modelers. The photochemical modelers quality assured the emissions inventories by generating and reviewing spatial plots of emissions by source sector/type. The reviews were designed to detect spatial anomalies (misplaced or missing sources). The modelers also conducted emission total checks against EMS-95 summary reports.

Stack Parameter Checks. A contractor, Alpine Geophysics, was employed, in part, to QA the point source emissions data. Alpine Geophysics discovered errors in the stack parameters and other point source data, including potential errors in gas exit velocities, emission rates, and physical stack parameters, for many point sources in the previous versions of the modeling system emission inventories. This review was distributed to the LADCO States to get the States to correct their respective point source emissions data. Some stack data were shifted from the elevated point source data files to the ground- level data files based on adopted screening parameters. This resulted in a spatial shift in emissions from previous modeling emission inventory versions. 7. What Is the Adopted Emissions Control Strategy?

To select possible emission control strategies, the LADCO States have modeled the ozone impacts of a number of emission control strategies for VOC and NOX. After modeling and reviewing the ozone impacts of various strategies and considering CAA and EPA emission control requirements, Illinois has adopted the emission control strategy known as SR 16 (LADCO Strategy Run 16) as the emission control strategy that will be pursued to attain the 1-hour ozone standard in the Chicago-Gary-Lake County ozone nonattainment area. Table II lists the emission controls included in SR 16.

Table II.--SR 16--Emission Control Strategy

Clean Air Act Title IV Acid Rain Controls for NOX--Phase I Rate-Of-Progress Plans (15 Percent ROP Plan and 9 Percent Post- 1996 ROP Plan) National Low Emission Vehicle Standards Reformulated Gasoline--Phase II (where required) Federal Phase II Small Engine Standards Federal Marine Engine Standards Federal Heavy Duty Vehicle (50 horsepower) Standards--Phase I Federal Locomotive Standards--Including Rebuilds Federal High Compression Engine Standards Federal Tier I Light Duty Vehicle and Heavy Duty Vehicle Emission Standards Enhanced Vehicle Inspection and Maintenance (I/M) (where required) Basic Vehicle I/M (where required) Federal Clean Fuel Fleets Requirements (where required) Federal Tier II and Low Sulfur Gasoline Standards Utility 0.15 Pounds NOXPer Million Btu of Heat Input Emission Limits (20 affected States, including Illinois) 60 Percent Reduction of NOXEmissions From Large Non-Electric Generating Unit (Non-EGU) Boilers and Turbines (20 affected States, including Illinois) 30 Percent Reduction of NOXEmissions From Large Cement Kilns (20 affected States, including Illinois) Wisconsin--0.28 Pounds NOXPer Million Btu of Heat Input for Utilities (EGUs) in 8 Counties Missouri--0.25 Pounds NOXPer Million Btu of Heat Input for EGUs in the Eastern One-Third of the State Missouri--0.35 Pounds NOXPer Million Btu of Heat Input for EGUs in the Western Two-Thirds of the State

With regard to the NOXemission controls listed in Table II, several aspects of the assumed NOXemission reductions should be noted. First, the NOXemission controls for utilities (EGUs), large non-EGU boilers and turbines, and large cement kilns in Grid M were assumed for all States (other than Wisconsin and Missouri) that are subject to EPA's NOXSIP Call. In reality, the assumed NOXemission reductions only reflect the expected NOXemissions budgets for these States and not the actual NOXemission controls that may actually occur in these States. Under the NOXSIP Call, states are not restricted to specific NOXemission controls, but are required to achieve assigned NOXemission budgets. The UAM modeling system is designed to test emission reductions for specific source categories. Therefore, LADCO chose a specific emission control scenario expected to produce NOXemissions that are compliant with the NOXSIP Call.

Illinois has developed NOXemission control regulations to control emissions from EGUs, non-EGU boilers and turbines, and cement kilns at or below the emission levels assumed for Illinois in control strategy SR 16. (The NOXrules for EGUs, non-EGU boilers and turbines and cement kilns are undergoing separate review (see an EPA proposed rule addressing this State rule published on August 31, 2000, 65 FR 52967) and are expected to be approved before EPA completes final rulemaking on Illinois' ozone attainment demonstration.) Other states in Grid M have also submitted adopted or draft NOXrules to comply with the NOXSIP Call.

Second, with regard to the NOXemission reductions assumed for Wisconsin and Missouri, these States have adopted and submitted NOXrules to achieve the NOXemission controls assumed in SR 16. The EPA has approved Missouri's NOXrule (December 28, 2000, 65 FR 82285) and expects to take final action on Wisconsin's NOXrule in the future and prior to final action on Illinois' ozone attainment demonstration.

In addition to the emission controls included in the above table, the following emission changes were also reflected in the modeling results for the control strategy: (a) Use of NOXvehicle I/M cut-points in the Wisconsin ozone nonattainment areas; (b) revised traffic network vehicle miles traveled data provided by the Chicago Area Transportation Study (CATS); (c) updated MOBILE5b input data for Illinois and Wisconsin; and (d) corrected MOBILE5b input data for Ohio.

In the ozone modeling, the CAA-required emission controls were assumed for all states within Grid M and were assumed for all areas outside of Grid M in modeling used to determine the initial and boundary ozone and ozone precursor concentrations for Grid M. In the Chicago area, the CAA-required controls modeled include: Reasonably Available Control Technology (RACT) on stationary sources of VOC; enhanced vehicle I/M; Transportation Control Measures (TCM); and other emission controls included in the State's 15 percent ROP plan (for a discussion of the emission controls included in this plan see 62 FR 37494, July 14, 1997) and 9 percent post-1996 ROP plan (for a discussion of the emission controls included in this plan see 65 FR 78961, December 18, 2000).

Table III compares the VOC and NOXemission rates for major source sectors in Grid M for the 1996 base year and for the adopted emission control strategy in 2007.

[Page 36382]

Table III.--Comparison of 1996 and SR 16 (2007) Emissions in Grid M [Emissions in tons/day]

Point--Non- Area--Offroad

Biogenic Pollutant

Point--EGU EGU

mobile Area--Other Onroad--Mobile sources Total

VOC:

1996 Base Year...........................................

32 2,335

1,716

4,780

3,633 30,816 43,312

SR 16....................................................

37 1,771

1,167

4,410

2,687 30,816 40,888 NOX:

1996 Base Year........................................... 5,844 1,876

2,138

602

5,681

2,000 18,141

SR 16.................................................... 2,092 1,822

1,748

734

3,230

2,000 11,626

Source: Table 3, ``Technical Support Document--Midwest Subregional Modeling: Emissions Inventory,'' September 27, 2000.

8. What Were the Ozone Modeling Results for the Base Period and for the Future Attainment Period With the Selected Emissions Control Strategy?

Table IV presents the Grid M peak observed and modeled ozone concentrations for the high episode days selected for the modeling analysis and attainment demonstration. The following modeled peak concentrations are presented: (a) The modeled validation peak ozone concentrations for Grid M; (b) the modeled Grid M peak ozone concentrations using the 1996 base year emissions; and (c) the 2007 predicted ozone concentrations for ozone control strategy SR 16. All modeled and monitored ozone concentrations are 1-hour averages and represent peak ozone concentrations anywhere within Grid M.

Table IV.--Peak Monitored and Modeled Ozone Concentrations for Grid M [Ozone Concentrations in ppb]

Peak ozone Peak ozone modeled Peak ozone Date

Peak ozone modeled 1996 base modeled SR observed validation year

16 emissions

6-25-91.....................................................

104

123

123

110 6-26-91.....................................................

175

136

138

117 6-27-91.....................................................

118

139

127

111 6-28-91.....................................................

138

124

102

95 7-16-91.....................................................

130

129

108

103 7-17-91.....................................................

137

119

89

89 7-18-91.....................................................

170

137

108

109 7-19-91.....................................................

170

137

112

111 7-20-91.....................................................

138

168

150

128 6-21-95.....................................................

112

123

122

118 6-22-95.....................................................

119

131

131

119 6-23-95.....................................................

123

128

128

113 6-24-95.....................................................

166

136

136

126 6-25-95.....................................................

108

125

124

120 7-12-95.....................................................

146

118

118

105 7-13-95.....................................................

178

147

146

124 7-14-95.....................................................

150

140

140

127 7-15-95.....................................................

154

156

156

128

Sources: Table 1, ``Midwest Subregional Modeling: 1-Hour Attainment Demonstration for Lake Michigan Area-- Summary,'' September 18, 2000. Table 6, ``Technical Support Document--Midwest Subregional Modeling: 1-Hour Attainment Demonstration for Lake Michigan Area,'' September 18, 2000.

From the above, you can see that the ozone modeling results for the selected emissions control strategy do show four peak ozone concentrations above the 1-hour ozone standard on the following dates: July 20, 1991; June 24, 1995; July 14, 1995; and July 15, 1995. As noted in LADCO's September 18, 2000 summary of the attainment demonstration, simple modeling and assessment of the potential future peak ozone concentrations (a deterministic test) does not demonstrate attainment of the ozone standard because of these modeled ozone standard exceedances. Additional analyses were conducted to support the attainment demonstration for this and other emission control strategies.

EPA's most relevant current ozone modeling/attainment demonstration guidance (Guidance on Use of Modeled Results to Demonstrate Attainment of the Ozone NAAQS, EPA-454/B-95-007, [June 1996]) provides for a statistical test as an alternate to a deterministic test to demonstrate attainment of the ozone standard (passing a statistical test can be used to support an ozone attainment demonstration even if a deterministic test is not passed). Under a statistical test, three benchmarks must be passed.

Benchmark 1 of the statistical test requires that the number of days with modeled ozone standard exceedances in each modeling domain grid cell must be less than 3 and that any modeled ozone standard exceedances occur on a ``severe'' day (severe days are determined by ranking high ozone days over many years and considering the ranking of the days covered in the modeled ozone attainment demonstration). Ten of the days

[Page 36383]modeled by LADCO were determined to be ``severe,'' including July 20, 1991 and July 15, 1995.

Benchmark 2 of the statistical test requires that the maximum modeled ozone concentration on severe days shall not exceed 130 ppb to 160 ppb, depending on the ``severity'' of the meteorological conditions on the modeled days. For the ozone attainment demonstration addressed in this proposed rule, LADCO's analysis of the severity of the modeled days led LADCO to conclude that the peak ozone concentration limit should be 130 ppb.

Finally, benchmark 3 of the statistical test requires that the number of modeling domain grid cells with peak ozone concentrations above or equal to 125 ppb must be reduced (from the number in the modeled base period) by 80 percent on each ``severe'' day.

LADCO has determined that the SR 16 emissions control strategy (and other modeled emission control strategies not adopted by Illinois) leads to modeled peak ozone concentrations meeting all three benchmarks of the statistical test. See LADCO's September 18, 2000 ``Technical Support Document--Midwest Subregional Modeling: 1-Hour Attainment Demonstration for Lake Michigan Area.'' Therefore, attainment of the ozone standard is demonstrated through modeling for the SR 16 emissions control strategy.

In light of the inherent uncertainties in the ozone modeling and to further evaluate the ozone attainment demonstration, LADCO has also chosen to conduct two additional analyses that are components of a WOE analysis. First, using the base period observed ozone design values for various ozone monitoring sites and the modeled 2007, post-control peak ozone concentrations for the domain grid cells in the vicinities of these monitors, LADCO predicted 2007 ozone design values for these monitoring sites (this procedure is referred to as the ``relative reduction factor'' test). For the SR 16 control strategy, the relative reduction factor test leads to predicted ozone design values below the ozone standard for all ozone monitoring sites and modeling receptor locations considered, with the highest projected ozone design values being 122 ppb at an unmonitored mid-Lake Michigan location (a synthetic base period ozone design value was used for this site) and 119 ppb for a Michigan City, Indiana ozone monitoring site.

Second, LADCO conducted an ozone trends analysis, which shows a considerable amount of progress toward attaining the ozone standard. Local ozone levels have significantly declined over time, while incoming ozone concentrations (transported ozone concentrations) remain relatively high.

The WOE analyses further support the conclusions of the attainment demonstration and counter any concerns that may be raised regarding the inherent uncertainties in the ozone modeling and the tendency of the modeling system to under-predict some peak ozone concentrations (the modeling system also over-predicts some peak ozone concentrations).

Based on all of the ozone modeling data available and related emissions analyses, LADCO concludes that the best ozone control strategy would be to control local VOC emissions (within the urban nonattainment areas) and to couple this with the control of domain- wide, regional NOXemissions (the purpose of EPA's NOXSIP Call and Illinois' adoption of NOX emission control rules for EGUs, non-EGU boilers and turbines, and cement kilns). This recommended emission control strategy approach is compatible with the emission control strategy selected by Illinois. 9. What Additional Analyses and Emissions Were Modeled by the State of Illinois?

Although the December 26, 2000 submittal of the ozone attainment demonstration by the IEPA indicates that the State of Illinois has adopted SR 16 as the emissions control strategy for attaining the 1- hour ozone standard, the IEPA has also decided to test the potential impacts of several emission changes not considered by the LADCO States as a whole. The additional emissions changes include: (a) Addition of NOXemissions from recently permitted combustion turbine EGUs; and (b) incorporation of transportation conformity emissions budgets that include a greater level of Vehicle Miles Travelled (VMT) than considered in the LADCO ozone modeling.

Illinois has recently issued emission permits for 33 new combustion turbine EGUs statewide (prior to the submittal of the ozone attainment demonstration and prior to the public hearing on this attainment demonstration). Ten of these units are located within the Illinois portion of the Chicago-Gary-Lake County ozone nonattainment area. These combustion turbine units include ``combined-cycle'' installations for providing base load and intermediate to peak load electricity production, as well as ``simple-cycle'' installations for providing peak load generating capacity (peaker-plants). Some of the installations have been built to replace existing units and others have been built to reduce boiler usage at existing facilities. The IEPA has determined the peak daily VOC and NOXemissions to be added by all of these installations and has determined the existing VOC and NOXemissions that would be replaced by the new installations. Modeled emission rates are based on the turbines operating at 100 percent loads.

The attainment demonstration analyses conducted by LADCO included the 2007 Chicago link-based transportation network VMT provided by CATS. Historically and in previous ozone rate-of-progress plans, the IEPA has used higher 2007 VMT estimates for 2007 provided by the Illinois Department of Transportation. To remain consistent with these prior plans and with the base data used to derive the 1990 base year emissions (used to calculate furture year emissions and ROP plan emission reduction targets), the IEPA concluded that it should consider the extra emissions resulting from the higher VMT estimates.

To test the impacts of the extra VOC and NOXemissions resulting from the permitted turbines and the increased VMT estimates, the IEPA has re-conducted the Grid M ozone modeling for SR 16, adding the extra VOC and NOXemissions for the July 1991 modeled ozone episode days (the IEPA notes that this episode is the most constraining episode, requiring the greatest amount of ozone precursor emission reduction amongst all tested high ozone episodes). The State has re-conducted the modeling analyses for the revised Grid M emissions, and concludes that the revised modeling results pass the statistical test benchmarks. The peak modeled ozone concentrations for SR 16 and the IEPA supplemental ozone modeling are given in Table V.

Table V.--Comparison of Predicted Peak 1-Hour Ozone Concentrations \19\ [Ozone Concentrations in ppb]

IEPA LADCO SR supplemental Episode day

16

ozone results results

7-16-91.......................................

103

104 7-17-91.......................................

89

90 7-18-91.......................................

109

109 7-19-91.......................................

111

113 7-20-91.......................................

128

130

\19\ Data taken from Table 2, Chapter I, of the December 21, 2000 ``Ozone Attainment Demonstration for the Chicago Nonattainment Area'' included as part of Illinois' December 26, 2000 ozone attainment demonstration submittal.

The IEPA concludes that the added emissions do not overturn the

[Page 36384]conclusion of LADCO that the SR 16 emission control strategy will lead to attainment of the 1-hour ozone standard. The IEPA further points out that this procedure is conservative because the increased NOXemissions from the EGU turbine installations will not actually increase the total NOXemissions in the State of Illinois. Since all of these new turbines will be subject to the State's EGU NOXrule, their NOXemissions will be included in the State's NOXemissions total, which will be constrained through a statewide NOXemissions cap under EPA's NOXSIP Call. Therefore, not all of the estimated 1-2 ppb ozone increase will actually occur.

It should be noted that, although these modeling results do not affect the conclusions regarding the adopted emissions control strategy, they do potentially affect the existing NOX emissions control waiver in the Illinois portion of the Chicago-Gary- Lake County ozone nonattainment area. See the section of this proposed rule addressing the NOXemissions control waiver below. 10. Do the Modeling Results Demonstrate Attainment of the Ozone Standard?

Based on LADCO's ozone modeling results and Illinois' supplemental modeling results, EPA believes that LADCO and, in particular, the State of Illinois have demonstrated attainment of the 1-hour ozone standard for the Chicago area based on the adopted SR 16 emissions control strategy. 11. Does the Attainment Demonstration Depend on Future Reductions of Regional Emissions?

Yes. The adopted emissions control strategy includes regional NOXemission reductions for the State of Illinois as well as for surrounding states in compliance with EPA's NOXSIP Call. LADCO has concluded that regional NOXemissions reductions are crucial to attainment of the 1-hour ozone standard in the Lake Michigan area. 12. Has the State Adopted All of the Regulations/Rules Needed to Support the Ozone Attainment Strategy and Demonstration?

The State of Illinois has adopted and is implementing all emission controls required under the CAA, including the emission controls included in Illinois' 15 percent and post-1996 ROP plans. The additional emission controls needed to support the adopted emissions control strategy are the NOXrules needed to comply with EPA's NOXSIP Call. The State has adopted NOX emissions control rules for EGUs, major non-EGU boilers and turbines, and cement kilns, and EPA is in the process of reviewing these rules. The EPA expects to approve these NOXrules in final before giving final approval to the ozone attainment demonstration.

C. EPA's Evaluation of the Ozone Attainment Demonstration Portion of the State's Submittal

1. Did the State Adequately Document the Techniques and Data Used To Derive the Modeling Input Data and Modeling Results of the Analyses?

The State's submittal thoroughly documents the techniques and data used to derive the modeling input data. The submittal adequately summarizes the modeling outputs and the conclusions drawn from these modeling outputs. This includes the State's modifications to LADCO's model inputs. Therefore, EPA concludes that the ozone modeling has been successfully documented and that the State's attainment demonstration is complete from a documentation standpoint. This includes documentation of an adopted emissions control strategy, which was lacking in the State's earlier April 1998 ozone attainment demonstration submittal. 2. Did the Modeling Procedures and Input Data Used Comply With the Clean Air Act Requirements and EPA Guidelines?

Yes. The State of Illinois, through LADCO, has used the UAM to model attainment of the 1-hour ozone standard. The State has documented the modeling results and the input data considered. The modeling procedures and input data comply with the requirements of the CAA as well as with EPA policy. 3. Did the State Adequately Demonstrate Attainment of the Ozone Standard?

Illinois, in accordance with the CAA, as further clarified in EPA's December 1997 guidance, has demonstrated that attainment of the 1-hour ozone standard is achievable by November 15, 2007 (the attainment deadline for the Chicago-Gary-Lake County ozone nonattainment area) provided projected reductions in background ozone and ozone precursor concentrations occur as the result of the implementation of EPA's NOXSIP Call. The State has demonstrated that the adopted emission control strategy, including local VOC emission control measures and regional NOXemission control measures (including statewide NOXemission reductions in Illinois needed to comply with the NOXSIP Call), is adequate for attainment of the 1-hour ozone standard. 4. Has the Adopted Emissions Control Strategy Been Adequately Documented?

Yes. The emission controls included in adopted strategy have been identified and their cumulative emission impacts have been documented. 5. Is the Emissions Control Strategy Acceptable?

Yes. It is noted that the adopted emissions control strategy relies significantly on the adoption of NOXemission control regulations by Illinois to comply with the requirements of EPA's NOXSIP Call. Illinois has adopted rules to reduce NOXemissions from EGUs, major non-EGU boilers, and major cement kilns. The EPA has proposed rulemaking for the EGU NOXrule (65 FR 52967, August 31, 2000), proposing to approve the rule, and proposing to disapprove it in the alternative, if the State does not correct noted deficiencies in the rule (the State corrected the most significant deficiency in this rule through State legislation on May 31, 2001 as documented in a June 11, 2001 letter from the IEPA). The EPA is preparing proposed rulemakings for the non- EGU boiler and cement kiln NOXemissions control rules. We cannot approve the attainment demonstration until after (or at the same time) we approve all of the NOXemission control rules relied on in the State's ozone attainment demonstration. Assuming that we will approve Illinois' NOXrules prior to or by the time we promulgate final approval of the ozone attainment demonstration, we find the ozone attainment demonstration to be approvable.

IV. Post-1999 Rate-of-Progress (ROP) Plan

A. What Is a Post-1999 ROP Plan?

ROP plans are a requirement of section 182 of the CAA. Section 182(c)(2)(B) of the CAA requires states with ozone nonattainment areas classified as serious and above, including the Chicago area which is classified as severe nonattainment, to adopt and implement plans to achieve periodic reductions in ozone precursors (VOC and/or NOX) after 1996. The requirement is intended to ensure that an area makes definite and reasonable progress toward attainment of the ozone NAAQS. Since Illinois has already adopted and implemented a post- 1996 ROP plan to meet the requirements of section 182(c)(2)(B) through November 15, 1999 (EPA approved this plan on December 18, 2000, 65 FR 78961) and

[Page 36385]since the ROP plan reviewed here addresses the ROP requirements for the period after November 15, 1999, we refer to the ROP plan reviewed in this proposed rule as the post-1999 ROP plan.

The post-1999 ROP emission reductions are to occur at a rate of 9 percent of baseline emissions, \20\ net of emissions growth, averaged over each 3-year period through the attainment year (2007 for the Chicago-Gary-Lake County ozone nonattainment area). The State must achieve the first 3 year ROP milestone (i.e., 9 percent) by November 15, 2002, another 9 percent ROP milestone by November 15, 2005, and the remaining 6 percent ROP milestone by November 15, 2007.

\20\ ``Baseline emissions'' are defined in section 182(b)(1)(B) of the CAA as the total amount of actual VOC or NOX emissions from all anthropogenic sources in the area during the calendar year of the Clean Air Act Amendments of 1990, excluding emissions that would be eliminated due to: (1) Any measure relating to motor vehicle exhaust or evaporative emissions promulgated by the EPA by January 1, 1990; (2) any regulations concerning Reid Vapor Pressure promulgated by the EPA by November 15, 1990 or required to be promulgated under section 211(h) of the CAA.

The ROP plan contains: (1) Documentation showing how the State calculated the emission reductions needed to achieve the incremental ROP emission reductions for each milestone period; (2) a description of the emission control measures used to achieve the incremental emission reductions; and (3) a description of how the State has determined the emission reduction creditable to each emission control measure.

B. What Is the ROP Contingency Measure Requirement?

Section 172(c)(9) of the CAA requires states with ozone nonattainment areas classified as moderate and above to adopt contingency measures by November 15, 1993. Such measures must provide for the implementation of specific emission control measures if an ozone nonattainment area fails to achieve ROP or to attain the NAAQS within the time-frames specified under the CAA. Section 182(c)(9) of the CAA requires that, in addition to the contingency measures required under section 172(c)(9), the contingency measure portion of the SIP for serious and above ozone nonattainment areas must also provide for the implementation of specific measures if an area fails to meet any applicable milestones in the CAA. As provided in these sections of the CAA, the contingency measures must take effect without further action by the state or by EPA upon failure of the state to meet ROP emission reduction milestones or to achieve attainment of the ozone NAAQS by a required deadline.

Our policy, as provided in the April 16, 1992 ``General Preamble for the Implementation of Title I of the Clean Air Act Amendments of 1990'' (General Preamble) (57 FR 13498), states that the contingency measures, in total, must generally be able to provide for emission reductions equal to 3 percent of the 1990 baseline emissions.

While all contingency measures and rules must be fully adopted by the states, states can use the contingency measures in one of two different ways. A state can choose to implement contingency measures before a milestone deadline. Alternatively, a state may decide not to implement a contingency measure until an area has actually failed to achieve a ROP or attainment milestone. In the latter situation, the contingency measure emission reduction must be achieved within one year following identification of a milestone failure.

C. What Illinois Counties Are Covered by the Post-1999 ROP Plan?

The post-1999 ROP plan covers the emission reduction requirements for the Chicago area. As indicated above, this area includes Cook, DuPage, Kane, Lake, McHenry, and Will Counties and the townships of Aux Sable and Goose Lake in Grundy County and Oswego in Kendall County. The VOC emission reduction requirements, as discussed below are determined relative to the VOC emissions in this area. Section 182(c)(2)(C) of the CAA provides for the substitution of NOXemission controls to meet part of the VOC emission reduction requirements for ROP provided that the NOXemission reduction produces an ozone reduction equivalent to that achieved from the required VOC emission reduction. As noted below, Illinois relies on the substitution of NOXemission reductions in its post-1999 ROP plan. It should also be noted that EPA interprets the CAA to allow the substitution of VOC and NOXemission reductions occurring outside of the ozone nonattainment area for nonattainment area VOC emission reductions needed to comply with ROP requirements, and Illinois' ROP plan incorporates such emission reduction substitution.

The Illinois ROP plan documentation refers to the term ``Volatile Organic Material'' (VOM) rather than to VOC. The State's definition of VOM is equivalent to EPA's definition of VOC. The two terms are interchangeable when discussing volatile organic emissions. For consistency with the CAA and EPA policy, we are using the term VOC in this proposed rulemaking.

D. Who Is Affected by the Illinois Post-1999 ROP Plan?

The post-1999 ROP plan does not itself create any new emission control requirements. Rather, it is a demonstration that existing regulations or regulations being developed to meet other emission reduction requirements are sufficient to achieve the required ROP emission reduction requirements.

The post-1999 ROP plan refers to various emission control regulations that have contributed to achieving the required ROP emission reductions for the 1999-2002, 2002-2005, and 2005-2007 periods for the Chicago area. These regulations, both Federal and State, affect a variety of industries, businesses, and, through the vehicle I/M program and other mobile source emission reduction requirements, motor vehicle owners. Most of these regulations, however, are already Federally enforceable through SIP revisions or through federally promulgated regulations.

E. What Criteria Must a Post-1999 ROP Plan Meet To Be Approved?

Section 182(c)(2)(B) establishes certain elements a post-1999 ROP plan must contain for approval. These elements are: (1) Emissions baseline; (2) emission target levels for each of the milestone years (2002, 2005, and 2007); (3) accounting for emission growth projections; and (4) emission reduction estimates from planned emission control measures.

The EPA has issued several guidance documents for states to use in developing approvable post-1996 ROP plans, which, as noted above, includes the post-1999 ROP plan. These documents address such topics as: (1) The relationship of ROP plans to other SIP elements required by the CAA; (2) calculation of the emission baseline and milestone year emission target levels; (3) procedures for projecting emission growth; and (4) methodology for determining emission reduction estimates for various emission control measures, including Federal emission control measures.

Our January 1994 guidance document, ``Guidance on the Post-1996 Rate-Of-Progress Plan and the Attainment Demonstration,'' provides States with the appropriate methods to calculate the emission reductions needed to meet the ROP emission reduction requirements. A complete list of ROP guidance documents is provided in the Technical Support Document (TSD) for the proposed rulemaking on Illinois' 9

[Page 36386]percent post-1996 ROP plan (referred to in a March 3, 2000 proposed rule, 65 FR 11528), which can be obtained from Region 5 at the address indicated in the ADDRESS section.

F. What Are the Special Requirements for Claiming NOX Emission Reductions in Post-1996 ROP Plans?

If a post-1996 (or post-1999 in this case) ROP plan relies on NOXemission reductions, it is subject to certain additional requirements. Under section 182(c)(2)(C) of the CAA, a plan can substitute NOXreductions for VOC reductions if the resulting ozone reductions are at least equivalent to the ozone reductions that would occur under a plan that relies only on VOC emission reductions. As required by section 182(c)(2)(C), the EPA issued guidance concerning the conditions for demonstrating equivalency. Our guidance provides that the NOXsubstitution strategy must show that the sum of VOC and NOXemission reduction percentages for each analyzed period must equal the ROP emissions reduction percentage required for that period, e.g., a 9 percent reduction from the 1990 baseline emissions for a 3-year period. Moreover, the State must provide technical justification that the NOXemission reductions will reduce ozone concentrations within the nonattainment area covered by the ROP plan.

On December 29, 1997, we issued a policy memorandum entitled ``Guidance for Implementing the 1-Hour Ozone and Pre-Existing PM10 NAAQS.'' Under this policy, both VOC emission controls outside of an ozone nonattainment area and NOXemission controls may be substituted for VOC emission controls within the ozone nonattainment area to meet the ROP VOC emission reduction requirements. The geographic area for substitution of VOC emission reductions is within 100 kilometers of the ozone nonattainment area. The geographic area for substitution of NOXemission reductions is within 200 kilometers of the ozone nonattainment area with the possibility for additional expansion of the NOXsubstitution area as follows. Based on its review of public comments on this policy, EPA believes that the area for allowable NOXsubstitutions should be expanded up to an entire state for those states in the core part of the OTAG modeling domain. For the purposes of this proposed rule, the core part of the OTAG modeling domain consists of the following states: Alabama; Connecticut; District of Columbia; Delaware; Georgia; Illinois; Indiana; Kentucky; Maine; Massachusetts; Maryland; Michigan; Missouri; North Carolina; New Hampshire; New Jersey; New York; Ohio; Pennsylvania; Rhode Island; South Carolina; Tennessee; Vermont; Virginia; Wisconsin; and West Virginia, i.e., the fine grid area of the OTAG modeling domain. The OTAG modeling results provide an adequate technical justification for statewide NOXemission substitutions for ROP. All other states implementing a NOX substitution strategy for ROP are restricted to a distance of 200 kilometers from an ozone nonattainment area, unless a substitution from a greater distance is accompanied by adequate technical justification.

The December 1997 policy states that a nonattainment area which has been granted a NOXwaiver can claim NOXemission reductions occurring outside of the nonattainment area, but within the state's boundary, if such reductions will reduce ozone concentrations within the ozone nonattainment area. We granted a NOXwaiver for the Chicago-Gary-Lake County ozone nonattainment area in two final rules. On January 26, 1996 (61 FR 2428), we granted exemptions from the RACT and NSR requirements for major stationary sources of NOXand from I/M and general conformity requirements for NOXfor ozone nonattainment areas within the Lake Michigan Ozone Study (LMOS) modeling domain. On February 12, 1996 (61 FR 5291), we approved Illinois' request to exempt the Chicago area (the Illinois portion of the Chicago-Gary-Lake County ozone nonattainment area) from the applicable NOXtransportation conformity requirements. \21\ See the discussion of the NOXwaiver below. OTAG modeling has shown that several NOXwaiver areas actually benefit from NOXreductions upwind. Therefore, under the December 1997 policy, a state can credit NOXemission reductions occurring outside of a NOXwaiver area, but within the state's boundary, if the state provides a technical analysis showing that the NOXemission reductions will lower ozone concentrations within the ozone nonattainment area (i.e., the NOXwaiver area). The ozone attainment demonstration submitted by Illinois provides such