Federal Implementation Plans: Interstate Transport of Fine Particulate Matter and Ozone and Correction of SIP Approvals

AGENCY:

Environmental Protection Agency (EPA).

ACTION:

Final rule.

SUMMARY:

In this action, EPA is limiting the interstate transport of emissions of nitrogen oxides (NO_X) and sulfur dioxide (SO₂) that contribute to harmful levels of fine particle matter (PM_2.5) and ozone in downwind states. EPA is identifying emissions within 27 states in the eastern United States that significantly affect the ability of downwind states to attain and maintain compliance with the 1997 and 2006 fine particulate matter national ambient air quality standards (NAAQS) and the 1997 ozone NAAQS. Also, EPA is limiting these emissions through Federal Implementation Plans (FIPs) that regulate electric generating units (EGUs) in the 27 states. This action will substantially reduce adverse air quality impacts in downwind states from emissions transported across state lines. In conjunction with other federal and state actions, it will help assure that all but a handful of areas in the eastern part of the country achieve compliance with the current ozone and PM_2.5 NAAQS by the deadlines established in the Clean Air Act (CAA or Act). The FIPs may not fully eliminate the prohibited emissions from certain states with respect to the 1997 ozone NAAQS for two remaining downwind areas and EPA is committed to identifying any additional required upwind emission reductions and taking any necessary action in a future rulemaking. In this action, EPA is also modifying its prior approvals of certain State Implementation Plan (SIP) submissions to rescind any statements that the submissions in question satisfy the interstate transport requirements of the CAA or that EPA's approval of the SIPs affects our authority to issue interstate transport FIPs with respect to the 1997 fine particulate and 1997 ozone standards for 22 states. EPA is also issuing a supplemental proposal to request comment on its conclusion that six additional states significantly affect downwind states' ability to attain and maintain compliance with the 1997 ozone NAAQS.

DATES:

This final rule is effective on October 7, 2011.

ADDRESSES:

EPA has established a docket for this action under Docket ID No. EPA-HQ-OAR-2009-0491. All documents in the docket are listed on the http://www.regulations.gov Web site. Although listed in the index, some information is not publicly available, e.g., CBI or other information whose disclosure is restricted by statute. Certain other material, such as copyrighted material, is not placed on the Internet and will be publicly available only in hard copy form. Publicly available docket materials are available either electronically through http://www.regulations.gov or in hard copy at the EPA Docket Center, EPA West, Room B102, 1301 Constitution Avenue, NW., Washington, DC. The Public Reading Room is open from 8:30 a.m. to 4:30 p.m., Monday through Friday, excluding legal holidays. The telephone number for the Public Reading Room is (202) 566-1744, and the telephone number for the Air Docket is (202) 566-1742.

FOR FURTHER INFORMATION CONTACT:

For general questions concerning this action, please contact Ms. Meg Victor, Clean Air Markets Division, Office of Atmospheric Programs, Mail Code 6204J, Environmental Protection Agency, 1200 Pennsylvania Avenue, NW., Washington, DC 20460; telephone number: (202) 343-9193; fax number: (202) 343-2359; e-mail address: victor.meg@epa.gov. For legal questions, please contact Ms. Sonja Rodman, U.S. EPA, Office of General Counsel, Mail Code 2344A, 1200 Pennsylvania Avenue, NW., Washington, DC 20460, telephone (202) 564-4079; e-mail address: rodman.sonja@epa.gov.

SUPPLEMENTARY INFORMATION:

I. Preamble Glossary of Terms and Abbreviations

The following are abbreviations of terms used in the preamble.

AQAT Air Quality Assessment Tool

ARP Acid Rain Program

BART Best Available Retrofit Technology

BACT Best Available Control Technology

CAA or Act Clean Air Act

CAIR Clean Air Interstate Rule

CAMx Comprehensive Air Quality Model with Extensions

CBI Confidential Business Information

CCR Coal Combustion Residuals

CEM Continuous Emissions Monitoring

CENRAP Central Regional Air Planning Association

CFR Code of Federal Regulations

DEQ Department of Environmental Quality

DSI Dry Sorbent Injection

EGU Electric Generating Unit

FERC Federal Energy Regulatory Commission

FGD Flue Gas Desulfurization

FIP Federal Implementation Plan

FR Federal Register

EPA U.S. Environmental Protection Agency

GHG Greenhouse Gas

GW Gigawatts

Hg Mercury

ICR Information Collection Request

IPM Integrated Planning Model

km Kilometers

lb/mmBtu Pounds Per Million British Thermal Unit

LNB Low-NO_X Burners

MACT Maximum Achievable Control Technology

MATS Modeled Attainment Test Software

μg/m ³ Micrograms Per Cubic Meter

MSAT Mobile Source Air Toxics

MOVES Motor Vehicle Emission Simulator

NAAQS National Ambient Air Quality Standards

NBP NO_X Budget Trading Program

NEI National Emission Inventory

NESHAP National Emissions Standards for Hazardous Air Pollutants

NO_X Nitrogen Oxides

NODA Notices of Data Availability

NSPS New Source Performance Standard

NSR New Source Review

OFA Overfire Air

OSAT Ozone Source Apportionment Technique

OTAG Ozone Transport Assessment Group

ppb Parts Per Billion

PM_2.5 Fine Particulate Matter, Less Than 2.5 Micrometers

PM₁₀ Fine and Coarse Particulate Matter, Less Than 10 Micrometers

PM Particulate Matter

ppm Parts Per Million

PUC Public Utility Commission

RIA Regulatory Impact Analysis

SCR Selective Catalytic Reduction

SIP State Implementation Plan

SMOKE Sparse Matrix Operator Kernel Emissions

SNCR Selective Non-catalytic Reduction

SO₂ Sulfur Dioxide

SO_X Sulfur Oxides, Including Sulfur Dioxide (SO₂) and Sulfur Trioxide (SO₃)

TAF Terminal Area Forecast

TCEQ Texas Commission on Environmental Quality

TIP Tribal Implementation Plan

TLN3 Tangential Low NO_X

TPY Tons Per Year

TSD Technical Support Document

WRAP Western Regional Air Partnership

II. General Information

A. Does this action apply to me?

This rule affects EGUs, and regulates the following groups:

This table is not intended to be exhaustive, but rather provides a guide for readers regarding entities likely to be regulated by this action. This table lists the types of entities that EPA is aware of that could potentially be regulated. Other types of entities not listed in the table could also be regulated. To determine whether your facility would be regulated by the proposed rule, you should carefully examine the applicability criteria in proposed §§ 97.404, 97.504, and 97,604.

B. How is the preamble organized?

I. Preamble Glossary of Terms and Abbreviations

II. General Information

A. Does this action apply to me?

B. How is the preamble organized?

III. Executive Summary

IV. Legal Authority, Environmental Basis, and Correction of CAIR SIP Approvals

A. EPA's Authority for Transport Rule

B. Rulemaking History

C. Air Quality Problems and NAAQS Addressed

1. Air Quality Problems and NAAQS Addressed

2. FIP Authority for Each State and NAAQS Covered

3. Additional Information Regarding CAA Section 110(a)(2)(D)(i)(I) SIPs for States in the Transport Rule Modeling Domain

D. Correction of CAIR SIP Approvals

V. Analysis of Downwind Air Quality and Upwind State Emissions

A. Pollutants Regulated

1. Background

2. Which pollutants did EPA propose to control for purposes of PM_2.5 and Ozone Transport?

3. Comments and Responses

B. Baseline for Pollution Transport Analysis

C. Air Quality Modeling to Identify Downwind Nonattainment and Maintenance Receptors

1. Emission Inventories

2. Air Quality Basis for Identifying Receptors

3. How did EPA project future nonattainment and maintenance for annual PM_2.5, 24-hour PM_2.5, and 8-hour ozone?

D. Pollution Transport From Upwind States

1. Choice of Air Quality Thresholds

2. Approach for Identifying Contributing Upwind States

VI. Quantification of State Emission Reductions Required

A. Cost and Air Quality Structure for Defining Reductions

1. Summary

2. Background

B. Cost of Available Emission Reductions (Step 1)

1. Development of Annual NO_X and Ozone-Season NO_X Cost Curves

2. Development of SO₂ Cost Curves

3. Amount of Reductions That Could Be Achieved by 2012 and 2014

C. Estimates of Air Quality Impacts (Step 2)

1. Development of the Air Quality Assessment Tool and Air Quality Modeling Strategy

2. Utilization of AQAT to Evaluate Control Scenarios

3. Air Quality Assessment Results

D. Multi-Factor Analysis and Determination of State Emission Budgets

1. Multi-Factor Analysis (Step 3)

2. State Emission Budgets (Step 4)

E. Approach to Power Sector Emission Variability

1. Introduction to Power Sector Variability

2. Transport Rule Variability Limits

F. Variability Limits and State Emission Budgets: State Assurance Levels

G. How the State Emission Reduction Requirements Are Consistent With Judicial Opinions Interpreting the Clean Air Act

VII. FIP Program Structure to Achieve Reductions

A. Overview of Air Quality-Assured Trading Programs

B. Applicability

C. Compliance Deadlines

1. Alignment With NAAQS Attainment Deadlines

2. Compliance and Deployment of Pollution Control Technologies

D. Allocation of Emission Allowances

1. Allocations to Existing Units

2. Allocations to New Units

E. Assurance Provisions

F. Penalties

G. Allowance Management System

H. Emissions Monitoring and Reporting

I. Permitting

1. Title V Permitting

2. New Source Review

J. How the Program Structure Is Consistent With Judicial Opinions Interpreting the Clean Air Act

VIII. Economic Impacts of the Transport Rule

A. Emission Reductions

B. The Impacts on PM_2.5 and Ozone of the Final SO₂ and NO_X Strategy

C. Benefits

1. Human Health Benefit Analysis

2. Quantified and Monetized Visibility Benefits

3. Benefits of Reducing GHG Emissions

4. Total Monetized Benefits

5. How do the benefits in 2012 compare to 2014?

6. How do the benefits compare to the costs of this final rule?

7. What are the unquantified and non-monetized benefits of the Transport Rule emission reductions?

D. Costs and Employment Impacts

1. Transport Rule Costs and Employment Impacts

2. End-Use Energy Efficiency

IX. Related Programs and the Transport Rule

A. Transition From the Clean Air Interstate Rule

1. Key Differences Between the Transport Rule and CAIR

2. Transition From the Clean Air Interstate Rule to the Transport Rule

B. Interactions With NO_X SIP Call

C. Interactions With Title IV Acid Rain Program

D. Other State Implementation Plan Requirements

X. Transport Rule State Implementation Plans

XI. Structure and Key Elements of Transport Rule Air Quality-Assured Trading Program Rules

XII. Statutory and Executive Order Reviews

A. Executive Order 12866: Regulatory Planning and Review and Executive Order 13563: Improving Regulation and Regulatory Review

B. Paperwork Reduction Act

C. Regulatory Flexibility Act

D. Unfunded Mandates Reform Act

E. Executive Order 13132: Federalism

F. Executive Order 13175: Consultation and Coordination With Indian Tribal Governments

G. Executive Order 13045: Protection of Children From Environmental Health and Safety Risks

H. Executive Order 13211: Actions That Significantly Affect Energy Supply, Distribution, or Use

I. National Technology Transfer and Advancement Act

J. Executive Order 12898: Federal Actions To Address Environmental Justice in Minority Populations and Low-Income Populations

1. Consideration of Environmental Justice in the Transport Rule Development Process and Response to Comments

2. Potential Environmental and Public Health Impacts Among Populations Susceptible or Vulnerable to Air Pollution

3. Meaningful Public Participation

4. Summary

K. Congressional Review Act

L. Judicial Review

III. Executive Summary

The CAA section 110(a)(2)(D)(i)(I) requires states to prohibit emissions that contribute significantly to nonattainment in, or interfere with maintenance by, any other state with respect to any primary or secondary NAAQS. In this final rule, EPA finds that emissions of SO₂ and NO_X in 27 eastern, midwestern, and southern states contribute significantly to nonattainment or interfere with maintenance in one or more downwind states with respect to one or more of three air quality standards—the annual PM_2.5 NAAQS promulgated in 1997, the 24-hour PM_2.5 NAAQS promulgated in 2006, and the ozone NAAQS promulgated in 1997 (EPA uses the term “states” to include the District of Columbia in this preamble).

These emissions are transported downwind either as SO₂ and NO_X or, after transformation in the atmosphere, as fine particles or ozone. This final rule identifies emission reduction responsibilities of upwind states, and also promulgates enforceable FIPs to achieve the required emission reductions in each state through cost-effective and flexible requirements for power plants. Each state has the option of replacing these federal rules with state rules to achieve the required amount of emission reductions from sources selected by the state.

Section 110(a)(2)(D)(i)(I) of the CAA requires the elimination of upwind state emissions that significantly contribute to nonattainment or interfere with maintenance of a NAAQS in another state. Elimination of these upwind state emissions may not necessarily, in itself, fully resolve nonattainment or maintenance problems at downwind state receptors. Downwind states also have control responsibilities because, among other things, the Act requires each state to adopt enforceable plans to attain and maintain air quality standards. Indeed, states have put in place measures to reduce local emissions that contribute to nonattainment within their borders. Section 110(a)(2)(D)(i)(I) only requires the elimination of emissions that significantly contribute to nonattainment or interfere with maintenance of the NAAQS in other states; it does not shift to upwind states the responsibility for ensuring that all areas in other states attain the NAAQS.

The reductions obtained through the Transport Rule will help all but a few downwind areas come into attainment with and maintain the 1997 annual PM_2.5 NAAQS, the 2006 24-hour PM_2.5 NAAQS, and the 1997 ozone NAAQS. With respect to the annual PM_2.5 NAAQS, this rule finds that 18 states have SO₂ and annual NO_X emission reduction responsibilities, and this rule quantifies each state's full emission reduction responsibility under section 110(a)(2)(D)(i)(I). See Table III-1 for the list of these states. With these reductions, EPA projects that no areas will have nonattainment or maintenance concerns with respect to the annual PM_2.5 NAAQS.

With respect to the 24-hour PM_2.5 NAAQS, this rule finds that 21 states have SO₂ and annual NO_X emission reduction responsibilities, and this rule quantifies each state's full emission reduction responsibility under 110(a)(2)(D)(i)(I). See Table III-1 for the list of these states. In all, this rule requires emission reductions related to interstate transport of fine particles in 23 states. With these reductions, as discussed in section VI.D of this preamble, only one area (Liberty-Clairton) is projected to remain in nonattainment, and three other areas (Chicago, Detroit, and Lancaster) are projected to have remaining maintenance concerns for the 24-hour PM_2.5 NAAQS.

With respect to the 1997 ozone NAAQS, this rule finds that 20 states have ozone-season NO_X emission reduction responsibilities. For 10 of these states this rule quantifies the state's full emission reduction responsibility under section 110(a)(2)(D)(i)(I). For 10 additional states, EPA quantifies in this rule the ozone-season NO_X emission reductions that are necessary but may not be sufficient to eliminate all significant contribution to nonattainment and interference with maintenance in other states. See Table III-1 for the complete list of 20 states required to reduce ozone-season NO_X emissions in this rule. With the Transport Rule reductions, only one area (Houston) is projected to remain in nonattainment, and one area (Baton Rouge) to have a remaining maintenance concern with respect to the 1997 ozone NAAQS. The 10 states upwind of either of these two areas are the states for which additional reductions may be necessary to fully eliminate each state's significant contribution to nonattainment and interference with maintenance, as discussed in section VI of this preamble.

As discussed further below, EPA's analysis also demonstrates that six additional states should be required to reduce ozone-season NO_X emissions. EPA is issuing a supplemental proposal to request comment on requiring ozone-season NO_X reductions in these six states. For five of these six states, EPA's analysis identifies the state's full emission reduction responsibility under section 110(a)(2)(D)(i)(I), and for the remaining one state EPA's analysis identifies reductions that are necessary but may not be sufficient to satisfy the requirements of 110(a)(2)(D)(i)(I).

On January 19, 2010, EPA proposed revisions to the 8-hour ozone NAAQS that the Agency had issued March 12, 2008 (75 FR 2938); the Agency intends to finalize its reconsideration in the summer of 2011. EPA intends to propose a rule to address transport with respect to the reconsidered 2008 ozone NAAQS as expeditiously as possible after reconsideration is completed. EPA intends to include in that proposed rule requirements to address any remaining significant contribution to nonattainment and interference with maintenance with respect to the 1997 ozone NAAQS for the states identified in this final rule, or the associated supplemental notice of proposed rulemaking, for which EPA was unable to fully quantify the emissions that must be prohibited to satisfy the requirements of 110(a)(2)(D)(i)(I) with respect to the 1997 ozone NAAQS.

The Act requires EPA to conduct periodic reviews of each of the NAAQS. When NAAQS are set or revised, the CAA requires revision of SIPs to ensure the standards are met expeditiously and within relevant timetables in the Act. If more protective NAAQS are promulgated, in the case of pollutants for which interstate transport is important, additional emission reductions to address transported pollution may be required from the power sector, from other sectors, and from sources in additional states. EPA will act promptly to promulgate any future rules addressing transport with respect to revised NAAQS.

The Transport Rule requires substantial near-term emission reductions in every covered state to address each state's significant contribution to nonattainment and interference with maintenance downwind. This rule achieves these reductions through FIPs that regulate the power sector using air quality-assured trading programs whose assurance provisions ensure that necessary reductions will occur within every covered state. This remedy structure is substantially similar to the preferred trading remedy structure presented in the proposal. The Transport Rule's air quality-assured trading approach will assure environmental results in each state while providing market-based flexibility to covered sources through interstate trading. The final rule includes four air quality-assured trading programs: An annual NO_X trading program, an ozone-season NO_X trading program, and two separate SO₂ trading programs (“SO₂ Group 1” and “SO₂ Group 2”), as discussed further in sections VI and VII, below.

The first phase of Transport Rule compliance commences January 1, 2012, for SO₂ and annual NO_X reductions and May 1, 2012, for ozone-season NO_X reductions. The second phase of Transport Rule reductions, which commences January 1, 2014, increases the stringency of SO₂ reductions in a number of states as discussed further below.

EPA projects that with the Transport Rule, covered EGU will substantially reduce SO₂, annual NO_X and ozone-season NO_X emissions, as shown in Tables III-2 and III-3, below. This rule generally covers electric generating units that are fossil fuel-fired boilers and turbines producing electricity for sale, as detailed in section VII.B.

EPA is promulgating the Transport Rule in response to the remand of the Clean Air Interstate Rule (CAIR) by the U.S. Court of Appeals for the District of Columbia Circuit (“Court”) in 2008. CAIR, promulgated May 12, 2005 (70 FR 25162), required 29 states to adopt and submit revisions to their State Implementation Plans (SIPs) to eliminate SO₂ and NO_X emissions that contribute significantly to downwind nonattainment of the PM_2.5 and ozone NAAQS promulgated in July 1997. CAIR covered a similar but not identical set of states as the Transport Rule. CAIR FIPs were promulgated April 26, 2006 (71 FR 25328) to regulate electric generating units in the covered states and achieve the emission reduction requirements established by CAIR until states could submit and obtain approval of SIPs to achieve the reductions.

In July 2008, the Court found CAIR and the CAIR FIPs unlawful. North Carolina v. EPA, 531 F.3d 896 (D.C. Cir. 2008), modified on rehearing, North Carolina v. EPA, 550 F.3d 1176, 1178 (D.C. Cir. 2008). The Court's original decision vacated CAIR. North Carolina, 531 F.3d at 929-30. However, the Court subsequently remanded CAIR to EPA without vacatur because it found that “allowing CAIR to remain in effect until it is replaced by a rule consistent with our opinion would at least temporarily preserve the environmental values covered by CAIR.” North Carolina, 550 F.3d at 1178. The CAIR requirements have remained in place while EPA has developed the Transport Rule to replace them.

EPA's approach in the Transport Rule to measure and address each state's significant contribution to downwind nonattainment and interference with maintenance is guided by and consistent with the Court's opinion in North Carolina and addresses the flaws in CAIR identified by the Court therein. This final rule also responds to extensive public comments and stakeholder input received during the public comment periods in response to the proposal and subsequent Notices of Data Availability (NODAs).

In this action, EPA both identifies and addresses emissions within states that significantly contribute to nonattainment or interfere with maintenance in other downwind states. In developing this rule, EPA used a state-specific methodology to identify emission reductions that must be made in covered states to address the CAA section 110(a)(2)(D)(i)(I) prohibition on emissions that significantly contribute to nonattainment or interfere with maintenance in a downwind state. EPA believes this methodology addresses the Court's concern that the approach used in CAIR was insufficiently state-specific. EPA used detailed air quality analysis to determine whether a state's contribution to downwind air quality problems is at or above specific thresholds. A state is covered by the Transport Rule if its contribution meets or exceeds one of those air quality thresholds and the Agency identifies, using a multi-factor analysis that takes into account both air quality and cost considerations, emissions within the state that constitute the state's significant contribution to nonattainment and interference with maintenance with respect to the 1997 ozone or the 1997 annual or 2006 24-hour PM_2.5 NAAQS. Section 110(a)(2)(D)(i)(I) requires states to eliminate the emissions that constitute this “significant contribution” and “interference with maintenance.”

In this final rule, EPA determined the emission reductions required from all upwind states to eliminate significant contribution to nonattainment and interference with maintenance with respect to the 1997 ozone, 1997 annual PM_2.5, and 2006 24-hour PM_2.5 NAAQS, using, in part, an assessment of modeled air quality in 2012 and 2014. EPA first identified the following two sets of downwind receptors: (1) Receptors that EPA projects will have nonattainment problems; and, (2) receptors that EPA projects may have difficulty maintaining the NAAQS based on historic variation in air quality. To identify areas that may have problems attaining or maintaining these air quality standards, EPA projected a suite of future air quality design values, based on measured data during the period 2003 through 2007. EPA used the average of these future design values to assess whether an area will be in nonattainment. EPA used the maximum projected future design value to assess whether an area may have difficulty maintaining the relevant NAAQS (i.e., whether an area has a reasonable possibility of being in nonattainment under adverse emission and weather conditions). Section V.C of this preamble details the Transport Rule's approach to identify downwind nonattainment and maintenance areas.

After identifying downwind nonattainment and/or maintenance areas, EPA next used air quality modeling to determine which upwind states are projected to contribute at or above threshold levels to the air quality problems in those areas. Section V.D details the choice of air quality thresholds and the approach to determine how much each upwind state contributes. States whose contributions meet or exceed the threshold levels were analyzed further, as detailed in section VI, to determine whether they significantly contribute to nonattainment or interfere with maintenance of a relevant NAAQS, and if so, the quantity of emissions that constitute their significant contribution and interference with maintenance.

When EPA proposed this air-quality and cost-based multi-factor approach to identify emissions that constitute significant contribution to nonattainment and interference with maintenance from upwind states with respect to the 1997 ozone, annual PM_2.5, and 2006 24-hour PM_2.5 NAAQS, the Agency indicated that the approach was designed to be applicable to both current and potential future ozone and PM_2.5 NAAQS (75 FR 45214). EPA believes that the Transport Rule's approach of using air-quality thresholds to determine upwind-to-downwind-state linkages and using the air-quality and cost-based multi-factor approach to determine the quantity of emissions that each upwind state must eliminate, i.e., the state's significant contribution to nonattainment and interference with maintenance, could serve as a precedent for quantifying upwind state emission reduction responsibilities with respect to potential future NAAQS, as discussed further in section VI.A of this preamble. The Agency further believes that the final Transport Rule demonstrates the strong value of this approach for addressing the role of interstate transport of air pollution in communities' ability to comply with current and future NAAQS.

EPA thus identified specific emission reduction responsibilities for each upwind state found to significantly contribute to nonattainment or interfere with maintenance in other states. Using that information, EPA developed individual state budgets for emissions from covered units under the Transport Rule. The Transport Rule emission budgets are based on EPA's state-by-state analysis of each upwind state's significant contribution to nonattainment and interference with maintenance. Because each state's budget is directly linked to this state-specific analysis of the state's obligations pursuant to section 110(a)(2)(D)(i)(I), this approach addresses the Court's concerns about the development of CAIR budgets.

In this rule, EPA is finalizing SO₂ and annual NO_X budgets for each state covered for the 24-hour and/or annual PM_2.5 NAAQS and an ozone-season NO_X budget for each state covered for the ozone NAAQS. A state's emission budget is the quantity of emissions that will remain from covered units under the Transport Rule after elimination of significant contribution to nonattainment and interference with maintenance in an average year (i.e., before accounting for the inherent variability in power system operations).

Baseline power sector emissions from a state can be affected by changing weather patterns, demand growth, or disruptions in electricity supply from other units or from the transmission grid. As a consequence, emissions could vary from year to year even in a state where covered sources have installed all controls and taken all measures necessary to eliminate the state's significant contribution to nonattainment and interference with maintenance. As described in detail in sections VI and VII of this preamble, the Transport Rule accounts for the inherent variability in power system operations through “assurance provisions” based on state-specific variability limits which extend above the state budgets to form each state's “assurance level.” The state assurance levels take into account the inherent variability in baseline emissions from year to year. The final Transport Rule FIPs will implement assurance provisions starting in 2012 as discussed in section VII, below.

The emission reduction requirements (i.e., the “remedy”) EPA is promulgating in this rule respond to the Court's concerns that in CAIR, EPA had not shown that the emission reduction requirements would get all necessary reductions within the state as required by section 110(a)(2)(D)(i)(I). The Transport Rule FIPs include assurance provisions specifically designed to ensure that no state's emissions are allowed to exceed that specific state's budget plus the variability limit (i.e., the state's assurance level).

Each state's Transport Rule SO₂, annual NO_X, or ozone-season NO_X emission budget is composed of a number of emission allowances (“allowances”) equivalent to the tonnage of that specific state budget. Under the Transport Rule FIPs, EPA is distributing (“allocating”) allowances under each state's budget to covered units in that state. In this rule, EPA analyzed each individual state's significant contribution to nonattainment and interference with maintenance and calculated budgets that represent each state's emissions after the elimination of those prohibited emissions in an average year. The methodology used to allocate allowances to individual units in a particular state has no impact on that state's budget or on the requirement that the state's emissions not exceed that budget plus the variability limit; the allocation methodology therefore has no impact on the rule's ability to satisfy the statutory mandate of CAA section 110(a)(2)(D)(i)(I).

The Transport Rule's approach to allocate emission allowances to existing units is based on historic heat-input data, as detailed in section VII.D of this preamble. The Transport Rule SO₂, annual NO_X, and ozone-season NO_X emission allowances each authorize the emission of one ton of SO₂, annual NO_X, or ozone-season NO_X emissions, respectively, during a Transport Rule control period, and are the currency in the Transport Rule's air quality-assured trading programs. As discussed in section IX.A.2 below, EPA is creating these Transport Rule allowances as distinct compliance instruments with no relation to allowances from the CAIR trading programs. EPA agrees with the general principle that it is desirable, where possible, to provide continuity under successive regulatory trading programs, for example through the carryover of allowances from one program into a subsequent one. However, EPA is promulgating the Transport Rule as a court-ordered replacement for (not a successor to) CAIR's trading programs. In light of the specific circumstances of this case, including legal and technical issues discussed in Section IX.A.2 below, the final rule will not allow any carryover of banked SO₂ or NO_X allowances from the Title IV or CAIR trading programs. EPA will strongly consider administrative continuity of this rule's trading programs under any future actions designed to address related problems of interstate transport of air pollution. A state may submit a SIP revision under which the state (rather than EPA) would determine allocations for one or more of the Transport Rule trading programs beginning with vintage year 2013 or later allowances. Section X of this preamble discusses the final rule's provisions for SIP submissions in detail.

Table III-1 lists states covered by the Transport Rule for PM_2.5 and ozone. It also, with respect to PM_2.5, identifies whether EPA determined the state was significantly contributing to nonattainment or interfering with maintenance of the 1997 annual PM_2.5 NAAQS, the 2006 24-hour PM_2.5 NAAQS, or both. As discussed below, the Transport Rule sorts the states required to reduce SO₂ emissions due to their contribution to PM_2.5 downwind into two groups of varying reduction stringency, with “Group 1” states subject to greater SO₂ reduction stringency than “Group 2” states starting in 2014. Table III-1 also lists which SO₂ Group each of the states is in.

As explained in this preamble, EPA has improved and updated both steps of its significant contribution analysis. It updated and improved the modeling platforms and modeling inputs used to identify states with contributions to certain downwind receptors that meet or exceed specified thresholds. It also updated and improved its analysis for identifying any emissions within such states that constitute the state's significant contribution to nonattainment or interference with maintenance. Therefore, the results of the analysis conducted for the final rule differ somewhat from the results of the analysis conducted for the proposal.

With respect to the 1997 ozone NAAQS, the analysis EPA conducted for the proposal did not identify Wisconsin, Iowa and Missouri as states that significantly contribute to nonattainment or interfere with maintenance of the ozone NAAQS in another state. However, the analysis conducted for the final rule shows that emissions from these states do significantly contribute to nonattainment or interfere with maintenance of the ozone NAAQS in another state. EPA is not issuing FIPs with respect to the 1997 ozone NAAQS or finalizing ozone season NO_X budgets for these states in this rule. EPA is publishing a supplemental notice of proposed rulemaking that will provide an opportunity for public comment on our conclusion that these states significantly contribute to nonattainment or interfere with maintenance of the 1997 ozone NAAQS.

In the other direction, the analysis conducted for the proposal supported EPA's conclusion at the time that Connecticut, Delaware, and the District of Columbia significantly contributed to nonattainment or interfered with maintenance with respect to the 1997 ozone NAAQS, whereas the modeling for the final rule no longer supports that conclusion for those states.

Additionally, the modeling conducted for the final rule identified two ozone maintenance receptors that were not identified in the modeling conducted for the proposal—Allegan County (MI) and Harford County (MD). Five states that EPA identified as significantly contributing to maintenance problems at the Allegan and/or Harford County receptors in the modeling for the final rule uniquely contribute to these receptors, i.e., absent these receptors the states would not be covered by the Transport Rule ozone-season program. The five states that uniquely contribute to these receptors are Iowa, Kansas, Michigan, Oklahoma, and Wisconsin. EPA is not issuing FIPs with respect to the 1997 ozone NAAQS or finalizing ozone-season NO_X budgets for these states in this rule. EPA is publishing a supplemental notice of proposed rulemaking that will provide an opportunity for public comment on our conclusion that these states significantly contribute to nonattainment or interfere with maintenance of the 1997 ozone NAAQS.

EPA did not change its methodology between the proposed Transport Rule and the final Transport Rule for identifying upwind states that significantly contribute to nonattainment or interfere with maintenance in other states; nor did EPA change its methodology for identifying receptors of concern with respect to maintenance of the 1997 ozone NAAQS. The final rule's air quality modeling identifies the new states and new receptors described above based on updated input information (including emission inventories), much of which was provided to EPA through public comment on the proposal and subsequent NODAs. Section V of this preamble details the approach EPA used to identify contributing states and receptors of concern.

With respect to the annual PM_2.5 NAAQS, the analysis EPA conducted for the proposal supported EPA's conclusion that the states of Delaware, the District of Columbia, Florida, Louisiana, Minnesota, New Jersey, and Virginia were significantly contributing to nonattainment and interfering with maintenance of the annual PM_2.5 NAAQS while the final rule's analysis does not. Also, with respect to the 24-hour PM_2.5 NAAQS, the analysis conducted for the proposal supported EPA's conclusion that the states of Connecticut, Delaware, the District of Columbia, and Massachusetts were significantly contributing to nonattainment or interfering with maintenance in other states while the analysis conducted for the final rule did not.

In the proposal EPA also requested comment on whether Texas should be included in the Transport Rule for annual PM_2.5. EPA's analysis for the proposal showed that emissions in Texas would significantly contribute to nonattainment or interfere with maintenance of the annual PM_2.5 NAAQS if Texas were not included in the rule for PM_2.5. The proposal did not include an illustrative budget for Texas or illustrative allowance allocations. However, the budgets and allowance allocations provided for other states in the proposal were included solely to illustrate the result of applying EPA's proposed methodology for quantifying significant contribution to the data EPA proposed to use. EPA provided an ample opportunity for comment on this methodology and on the data, including data regarding emissions from Texas sources, used in the significant contribution analysis. EPA received numerous comments on and corrections to Texas-specific data. The modeling conducted for the final rule demonstrates that Texas significantly contributes to nonattainment or interferes with maintenance of the annual PM_2.5 NAAQS in another state. EPA provided a full opportunity for comment on whether Texas should be included in the rule for annual PM_2.5, as well as on the methodology and data used for the significant contribution analysis for the final rule. EPA therefore believes its determination that Texas must be included in the rule for annual PM_2.5 is a logical outgrowth of its proposal.

With respect to the 24-hour PM_2.5 NAAQS, the analysis EPA conducted for the proposal did not identify Texas as a state that significantly contributes to nonattainment or interferes with maintenance of 24-hour PM_2.5 in another state. However, the analysis conducted for the final rule shows that emissions from Texas do significantly contribute to nonattainment of the 24-hour PM_2.5 NAAQS in another state. EPA is not issuing a FIP for Texas with respect to the 24-hour PM_2.5 NAAQS in this rule. However, EPA believes that the FIP for Texas with respect to the 1997 annual PM_2.5 NAAQS also addresses the emissions in Texas that significantly contribute to nonattainment and interference with maintenance of the 2006 24-hour PM_2.5 NAAQS in another state.

The final rule, however, does not cover the states of Connecticut, Delaware, the District of Columbia, Florida, Louisiana, or Massachusetts for annual or 24-hour PM_2.5 as the analysis for the final rule does not support their inclusion.

The Transport Rule FIPs require the 23 states covered for purposes of the 24-hour and/or annual PM_2.5 NAAQS to reduce SO₂ and annual NO_X emissions by specified amounts. The FIPs require the 20 states covered for purposes of the ozone NAAQS to reduce ozone-season NO_X emissions by specified amounts. As discussed in detail in section VI, below, the 23 states covered for the 24-hour and/or annual PM_2.5 NAAQS are grouped in two tiers reflecting the stringency of SO₂ reductions required to eliminate that state's significant contribution to nonattainment and interference with maintenance downwind. The more-stringent SO₂ tier (“Group 1”) is comprised of the 16 states indicated in Table III-1, above, and the less-stringent SO₂ tier (“Group 2”) is comprised of the 7 states identified in the table. The two SO₂ trading programs are exclusive, i.e., a covered source in a Group 1 state may use only a Group 1 allowance for compliance, and likewise a source in a Group 2 state may use only a Group 2 allowance for compliance. In Group 1 states, the SO₂ reduction requirements become more stringent in the second phase, which starts in 2014.

In response to the Court's opinion in North Carolina, EPA has coordinated the Transport Rule's compliance deadlines with the NAAQS attainment deadlines that apply to the downwind nonattainment and maintenance areas. The Transport Rule requires that all significant contribution to nonattainment and interference with maintenance identified in this action with respect to the 1997 annual PM_2.5 NAAQS and the 2006 24-hour PM_2.5 NAAQS be eliminated by no later than 2014, with an initial phase of reductions starting in 2012 to ensure that reductions are made as expeditiously as practicable and, consistent with the Court's remand, to “preserve the environmental values covered by CAIR.” Sources must comply by January 1, 2012 and January 1, 2014 for the first and second phases, respectively.

With respect to the 1997 ozone NAAQS, the Transport Rule requires NO_X reductions starting in 2012 to ensure that reductions are made as expeditiously as practicable to assist downwind state attainment and maintenance of the standard. Sources must comply by May 1, 2012. The Transport Rule's compliance schedule and alignment with downwind NAAQS attainment deadlines are discussed in detail in section VII below.

Table III-2 shows projected Transport Rule emissions compared to projected base case emissions, and Table III-3 shows projected Transport Rule emissions compared to historical emissions (i.e., 2005 emissions), for the power sector in all Transport Rule states. The ozone-season NO_X results shown in Tables III-2 and III-3 are based on analysis of the group of 26 states that would be covered for the ozone-season program if EPA finalizes the supplemental proposal regarding ozone-season requirements for Iowa, Kansas, Michigan, Missouri, Oklahoma, and Wisconsin.

Notes:

The SO₂ and annual NO_X emissions in this table reflect EGUs in the 23 states covered by this rule for purposes of the 24-hour and/or annual PM_2.5 NAAQS (Alabama, Georgia, Illinois, Indiana, Iowa, Kansas, Kentucky, Maryland, Michigan, Minnesota, Missouri, Nebraska, New Jersey, New York, North Carolina, Ohio, Pennsylvania, South Carolina, Tennessee, Texas, Virginia, West Virginia, and Wisconsin). The ozone-season NO_X emissions reflect EGUs in the 20 states covered by this rule for purposes of the ozone NAAQS (Alabama, Arkansas, Florida, Georgia, Illinois, Indiana, Kentucky, Louisiana, Maryland, Mississippi, New Jersey, New York, North Carolina, Ohio, Pennsylvania, South Carolina, Tennessee, Texas, Virginia, and West Virginia) and the six states that would be covered for the ozone NAAQS if EPA finalizes its supplemental proposal (Iowa, Kansas, Michigan, Missouri, Oklahoma, and Wisconsin).

Notes:

The SO₂ and annual NO_X emissions in this table reflect EGUs in the 23 states covered by this rule for purposes of the 24-hour and/or annual PM_2.5 NAAQS (Alabama, Georgia, Illinois, Indiana, Iowa, Kansas, Kentucky, Maryland, Michigan, Minnesota, Missouri, Nebraska, New Jersey, New York, North Carolina, Ohio, Pennsylvania, South Carolina, Tennessee, Texas, Virginia, West Virginia, and Wisconsin). The ozone-season NO_X emissions reflect EGUs in the 20 states covered by this rule for purposes of the ozone NAAQS (Alabama, Arkansas, Florida, Georgia, Illinois, Indiana, Kentucky, Louisiana, Maryland, Mississippi, New Jersey, New York, North Carolina, Ohio, Pennsylvania, South Carolina, Tennessee, Texas, Virginia, and West Virginia) and the six states that would be covered for the ozone NAAQS if EPA finalizes its supplemental proposal (Iowa, Kansas, Michigan, Missouri, Oklahoma, and Wisconsin).

In addition to the emission reductions shown above, EPA projects other substantial benefits of the Transport Rule, as described in section VIII in this preamble. EPA used air quality modeling to quantify the improvements in PM_2.5 and ozone concentrations that are expected to result from the Transport Rule emission reductions in 2014. The Agency used the results of this modeling to calculate the average and peak reduction in annual PM_2.5, 24-hour PM_2.5, and 8-hour ozone concentrations for monitoring sites in the Transport Rule covered states (including the six states for which EPA issued a supplemental proposal for ozone-season NO_X requirements) in 2014.

For annual PM_2.5, the average reduction across all monitoring sites in covered states in 2014 is 1.41 microgram per meter cubed (µg/m³) and the greatest reduction at a single site is 3.60 µg/m³. For 24-hour PM_2.5, the average reduction across all monitoring sites in covered states in 2014 is 4.3 µg/m³ and the greatest reduction at a single site is 11.6 µg/m³. And finally, for 8-hour ozone, the average reduction across all monitoring sites in covered states in 2014 is 0.3 parts per billion (ppb) and the greatest is 3.9 ppb. See section VIII for further information on air quality improvements.

EPA estimated the Transport Rule's costs and benefits, including effects on sensitive and vulnerable and environmental justice communities. Table III-4, below, summarizes some of these results. Further discussion of the results is provided in preamble section VIII, below, and in the Regulatory Impact Analysis (RIA). Estimates here are subject to uncertainties discussed further in the RIA.

As a result of updated analyses and in response to public comments, the final Transport Rule differs from the proposal in a number of ways. The differences between proposal and final rule are discussed throughout this preamble. Some key changes between proposal and final rule are that EPA:

• Updated emission inventories (resulting in generally lower base case emissions). See section V.C.
• Updated modeling and analysis tools (including improved alignment between air quality estimates and air quality modeling results). See sections V and VI.
• Updated conclusions regarding which states significantly contribute to nonattainment or interfere with maintenance of the NAAQS in other states. See Table III-1 and sections V.D and VI.
• Recalculated state budgets and variability limits, i.e., state assurance levels, based on updated modeling. See section VI.
• Simplified variability limits for one-year application only. See section VI.E.
• Revised allocation methodology for existing and new units and revised new unit set-asides for new units in Transport Rule states and new units potentially locating in Indian country. See section VII.D.
• Changed start of assurance provisions to 2012 and increased assurance provision penalties. See section VII.E.
• Removed opt-in provisions. See section VII.B
• Added provisions for full and abbreviated Transport Rule SIP revisions. See section X.

EPA conducted substantial stakeholder outreach in developing the Transport Rule, starting with a series of “listening sessions” in the spring of 2009 with states, nongovernmental organizations, and industry. EPA docketed stakeholder-related materials in the Transport Rule docket (Docket ID No. EPA-HQ-OAR-2009-0491). The Agency conducted general teleconferences on the rule with tribal environmental professionals, conducted consultation with tribal governments, and hosted a webinar for communities and tribal governments. EPA continued to provide updates to regulatory partners and stakeholders through several conference calls with states as well as at conferences where EPA officials often made presentations. The Agency conducted additional stakeholder outreach during the public comment period. EPA responded to extensive public comments received during the public comment periods on the proposed rule and associated NODAs.

This Transport Rule is one of a series of regulatory actions to reduce the adverse health and environmental impacts of the power sector. EPA is developing these rules to address judicial review of previous rulemakings and to issue rules required by environmental laws. Finalizing these rules will effectuate health and environmental protection mandated by Congress while substantially reducing uncertainty over the future regulatory obligations of power plants, which will assist the power sector in planning for compliance more cost effectively. The Agency is providing full opportunity for notice and comment for each rule.

As discussed above, rules to address transport under revised NAAQS, including the reconsidered 2008 ozone NAAQS, may result in additional emission reduction requirements for the power sector. In addition, existing Clean Air Act rules establishing best available retrofit technology (BART) requirements and other requirements for addressing visibility and regional haze may also result in future state requirements for certain power plant emission reductions where needed.

On May 3, 2011 (76 FR 24976), EPA proposed national emission standards for hazardous air pollutants from coal- and oil-fired electric utility steam generating units under CAA section 112(d), also called Mercury and Air Toxics Standards (MATS), and proposed revised new source performance standards for fossil fuel-fired EGUs under section 111(b). As discussed in the EPA-led public listening sessions during February and March 2011, EPA is preparing to propose innovative, cost-effective and flexible greenhouse gas (GHG) emissions performance standards under section 111 for steam electric generating units, the largest U.S. source of greenhouse gas emissions. On April 20, 2011 (76 FR 22174), EPA proposed requirements under section 316(b) of the Clean Water Act for existing power generating facilities, manufacturing and industrial facilities that withdraw more than two million gallons per day of water from waters of the U.S. and use at least twenty-five percent of that water exclusively for cooling purposes. On June 21, 2010 (75 FR 35128), the Agency proposed to regulate coal combustion residuals (CCRs) under the Resource Conservation and Recovery Act to address the risks from the disposal of CCRs generated from the combustion of coal at electric utilities and independent power producers.

EPA will coordinate utility-related air pollution rules with each other and with other actions affecting the power sector including these rules from EPA's Office of Water and its Office of Resource Conservation and Recovery to the extent consistent with legal authority in order to provide timely information needed to support regulated sources in making informed decisions. Use of a small number of air pollution control technologies, widely deployed, can assist with compliance for multiple rules. EPA also notes that the flexibility inherent in the allowance-trading mechanism included in the Transport Rule affords utilities themselves a degree of latitude to determine how best to integrate compliance with the emission reduction requirements of this rule and those of the other rules. EPA will pursue energy efficiency improvements in the use of electricity throughout the economy, along with other federal agencies, states and other groups, which will contribute to additional environmental and public health improvements while lowering the costs of realizing those improvements.

IV. Legal Authority, Environmental Basis, and Correction of CAIR SIP Approvals

A. EPA's Authority for Transport Rule

The statutory authority for this action is provided by the CAA, as amended, 42 U.S.C. 7401 et seq. Section 110(a)(2)(D) of the CAA, often referred to as the “good neighbor” provision of the Act, and requires states to prohibit certain emissions because of their impact on air quality in downwind states. Specifically, it requires all states, within 3 years of promulgation of a new or revised NAAQS, to submit SIPs that prohibit certain emissions of air pollutants because of the impact they would have on air quality in other states. 42 U.S.C. 7410(a)(2)(D). This action addresses the requirement in section 110(a)(2)(D)(i)(I) regarding the prohibition of emissions within a state that will significantly contribute to nonattainment or interfere with maintenance of the NAAQS in any other state. EPA has previously issued two rules interpreting and clarifying the requirements of section 110(a)(2)(D)(i)(I). The NO_X SIP Call, promulgated in 1998, was largely upheld by the U.S. Court of Appeals for the DC Circuit in Michigan, 213 F.3d 663. CAIR, promulgated in 2005, was remanded by the DC Circuit in North Carolina, 531 F.3d 896, modified on reh'g, 550 F.3d. 1176. These decisions provide additional guidance regarding the requirements of section 110(a)(2)(D)(i)(I) and are discussed later in this notice.

Section 301(a)(1) of the CAA also gives the Administrator of EPA general authority to prescribe such regulations as are necessary to carry out her functions under the Act. 42 U.S.C. 7601(a)(1). Pursuant to this section, EPA has authority to clarify the applicability of CAA requirements. In this action, among other things, EPA is clarifying the applicability of section 110(a)(2)(D)(i)(I) by identifying SO₂ and NO_X emissions that must be prohibited pursuant to this section with respect to the PM_2.5 NAAQS promulgated in 1997 and 2006 and the 8-hour ozone NAAQS promulgated in 1997.

Section 110(c)(1) requires the Administrator to promulgate a FIP at any time within 2 years after the Administrator finds that a state has failed to make a required SIP submission, finds a SIP submission to be incomplete or disapproves a SIP submission unless the state corrects the deficiency, and the Administrator approves the SIP revision, before the Administrator promulgates a FIP. 42 U.S.C. 7410(c)(1).

Tribes are not required to submit state implementation plans. However, as explained in EPA's regulations outlining Tribal Clean Air Act authority, EPA is authorized to promulgate FIPs for Indian country as necessary or appropriate to protect air quality if a tribe does not submit and get EPA approval of an implementation plan. See 40 CFR 49.11(a); see also 42 U.S.C. section 7601(d)(4).

Section 110(k)(6) of the CAA gives the Administrator authority, without any further submission from a state, to revise certain prior actions, including actions to approve SIPs, upon determining that those actions were in error.

B. Rulemaking History

The Transport Rule FIPs will limit the interstate transport of emissions of NO_X and SO₂ within 27 states in the eastern, midwestern, and southern United States that affect the ability of downwind states to attain and maintain compliance with the 1997 and 2006 PM_2.5 NAAQS and the 1997 ozone NAAQS. Prior to this Transport Rule, CAIR was EPA's most recent regulatory action in a longstanding series of regulatory initiatives to address interstate transport of air pollution. The proposed Transport Rule preamble provides more information on EPA actions prior to CAIR (75 FR 45221-45225).

CAIR, promulgated May 12, 2005 (70 FR 25162), required 29 states to adopt and submit revisions to their SIPs to eliminate SO₂ and NO_X emissions that contribute significantly to downwind nonattainment of the PM_2.5 and ozone NAAQS promulgated in 1997. The states covered by CAIR were similar but not identical to the states covered by the Transport Rule. The CAIR FIPs, promulgated April 26, 2006 (71 FR 25328), regulated electric generating units in the covered states and achieved CAIR's emission reduction requirements unless or until states had approved SIPs to achieve the required reductions.

In July 2008, the DC Circuit Court found CAIR and the CAIR FIPs unlawful and vacated CAIR. North Carolina, 531 F.3d at 929-30. However, the Court subsequently remanded CAIR to EPA without vacatur in order to “at least temporarily preserve the environmental values covered by CAIR.” North Carolina, 550 F.3d at 1178. CAIR requirements have remained in place and CAIR's emission trading programs have operated while EPA developed replacement rules in response to the remand.

By promulgating the Transport Rule FIPs, EPA is responding to the Court's remand of CAIR and the CAIR FIPs and replacing those rules. The approaches EPA used in the Transport Rule to measure and address each state's significant contribution to downwind nonattainment and interference with maintenance are guided by and consistent with the Court's opinion in North Carolina and address the flaws in CAIR identified by the Court therein.

By notice of proposed rulemaking (Federal Implementation Plans To Reduce Interstate Transport of Fine Particulate Matter and Ozone, 75 FR 45210; August 2, 2010), EPA proposed the Transport Rule to identify and limit NO_X and SO₂ emissions within 32 states in the eastern, midwestern, and southern United States that affect the ability of downwind states to attain and maintain compliance with the 1997 and 2006 PM_2.5 NAAQS and the 1997 ozone NAAQS. EPA proposed to achieve the emission reductions under FIPs, which states may choose to replace by submitting SIPs for EPA approval. EPA proposed to limit emissions by regulating electric generating units in the 32 states with interstate emission trading programs and assurance provisions to ensure the required reductions occur in each covered state. EPA also requested comment on two alternative FIP remedies.

EPA supplemented the Transport Rule record with additional information relevant to the rulemaking in three NODAs for which EPA requested comments:

• Notice of Data Availability Supporting Federal Implementation Plans to Reduce Interstate Transport of Fine Particulate Matter and Ozone (75 FR 53613; September 1, 2010). This NODA provided an updated database of unit-level characteristics of EGUs included in EPA modeling, an updated version of the power sector modeling platform EPA used to support the final rule, and other input assumptions and data EPA provided for public review and comment.
• Notice of Data Availability Supporting Federal Implementation Plans To Reduce Interstate Transport of Fine Particulate Matter and Ozone: Revisions to Emission Inventories (75 FR 66055; October 27, 2010). This NODA provided additional information relevant to the rulemaking, including updated emission inventory data for 2005, 2012 and 2014 for several stationary and mobile source inventory components.
• Notice of Data Availability for Federal Implementation Plans To Reduce Interstate Transport of Fine Particulate Matter and Ozone: Request for Comment on Alternative Allocations, Calculation of Assurance Provision Allowance Surrender Requirements, New-Unit Allocations in Indian Country, and Allocations by States (76 FR 1109; January 7, 2011). This NODA provided additional information relevant to the rulemaking, including emissions allowance allocations for existing units calculated using two alternative methodologies, data supporting those calculations, information about an alternative approach to calculation of assurance provision allowance surrender requirements, allocations for new units locating in Indian country in Transport Rule states in the future, and provisions for states to submit SIPs providing for state allocation of allowances in the Transport Rule trading programs.

C. Air Quality Problems and NAAQS Addressed

1. Air Quality Problems and NAAQS Addressed

a. Fine Particles

Fine particles are associated with a number of serious health effects including premature mortality, aggravation of respiratory and cardiovascular disease (as indicated by increased hospital admissions, emergency room visits, health-related absences from school or work, and restricted activity days), lung disease, decreased lung function, asthma attacks, and certain cardiovascular problems. In addition to effects on public health, fine particles are linked to a number of public welfare effects, including (1) Reduced visibility (haze) in scenic areas, (2) effects caused by particles settling on ground or water, such as: making lakes and streams acidic, changing the nutrient balance in coastal waters and large river basins, depleting the nutrients in soil, damaging sensitive forests and farm crops, and affecting the diversity of ecosystems, and (3) staining and damaging of stone and other materials, including culturally important objects such as statues and monuments.

In 1997, EPA revised the NAAQS for PM to add new annual and 24-hour standards for fine particles, using PM_2.5 as the indicator (62 FR 38652). These revisions established an annual standard of 15 μg/m³ and a 24-hour standard of 65 μg/m³. During 2006, EPA revised the air quality standards for PM_2.5. The 2006 standards decreased the level of the 24-hour fine particle standard from 65 μg/m³ to 35 μg/m³, and retained the annual fine particle standard at 15 μg/m³.

b. Ozone

Short-term (1- to 3-hour) and prolonged (6- to 8-hour) exposures to ambient ozone have been linked to a number of adverse health effects. At sufficient concentrations, short-term exposure to ozone can irritate the respiratory system, causing coughing, throat irritation, and chest pain. Ozone can reduce lung function and make it more difficult to breathe deeply. Breathing may become more rapid and shallow than normal, thereby limiting a person's normal activity. Ozone also can aggravate asthma, leading to more asthma attacks that may require a doctor's attention and the use of additional medication. Increased hospital admissions and emergency room visits for respiratory problems have been associated with ambient ozone exposures. Longer-term ozone exposure can inflame and damage the lining of the lungs, which may lead to permanent changes in lung tissue and irreversible reductions in lung function. A lower quality of life may result if the inflammation occurs repeatedly over a long time period (such as months, years, or a lifetime). There is also epidemiological evidence indicating a correlation between short-term ozone exposure and premature mortality.

In addition to causing adverse health effects, ozone affects vegetation and ecosystems, leading to reductions in agricultural crop and commercial forest yields; reduced growth and survivability of tree seedlings; and increased plant susceptibility to disease, pests, and other environmental stresses (e.g., harsh weather). In long-lived species, these effects may become evident only after several years or even decades and have the potential for long-term adverse impacts on forest ecosystems. Ozone damage to the foliage of trees and other plants can also decrease the aesthetic value of ornamental species used in residential landscaping, as well as the natural beauty of our national parks and recreation areas. In 1997, at the same time we revised the PM_2.5 standards, EPA issued its final action to revise the NAAQS for ozone (62 FR 38856) to establish new 8-hour standards. In this action published on July 18, 1997, we promulgated identical revised primary and secondary ozone standards that specified an 8-hour ozone standard of 0.08 parts per million (ppm). Specifically, the standards require that the 3-year average of the fourth highest 24-hour maximum 8-hour average ozone concentration may not exceed 0.08 ppm. In general, the 8-hour standards are more protective of public health and the environment and more stringent than the pre-existing 1-hour ozone standards.

On March 12, 2008, EPA published a revision to the 8-hour ozone standard, lowering the level from 0.08 ppm to 0.075 ppm. On September 16, 2009, EPA announced it would reconsider these 2008 ozone standards. The purpose of the reconsideration is to ensure that the ozone standards are clearly grounded in science, protect public health with an adequate margin of safety, and are sufficient to protect the environment. EPA proposed revisions to the standards on January 19, 2010 (75 FR 2938) and anticipates issuing final standards soon.

c. Which NAAQS does this rule address?

This action addresses the requirements of CAA section 110(a)(2)(D)(i)(I) as they relate to:

(1) The 1997 annual PM_2.5 standard,

(2) The 2006 24-hour PM_2.5 standard, and

(3) The 1997 ozone standard.

The original CAIR and CAIR FIP rules, which pre-dated the 2006 PM_2.5 standards, addressed the 1997 ozone and 1997 PM_2.5 standards only.

In this action, EPA fully addresses, for the states covered by this rule, the requirements of CAA section 110(a)(2)(D)(i)(I) for the annual PM_2.5 standard of 15 μg/m³ and the 24-hour standard of 35 μg/m³. For the 1997 8-hour ozone standard of 0.08 ppm, EPA fully addresses the CAA section 110(a)(2)(D)(i)(I) requirements for some states covered by this rule, but for the remaining states EPA is conducting further analysis to determine whether further requirements are needed, as discussed in section III of this preamble.

This action does not address the CAA section 110(a)(2)(D)(i)(I) requirements for the revised ozone standards promulgated in 2008. These standards are currently under reconsideration. We are, however, actively conducting the technical analyses and other work needed to address interstate transport for the reconsidered ozone standard as soon as possible. We intend to issue as soon as possible a proposal to address the transport requirements with respect to the reconsidered standard.

This action addresses these CAA transport requirements through reductions in annual emissions of SO₂ and NO_X, and through reductions in ozone-season NO_X. The rationale for these reductions is discussed in detail later in the preamble.

d. Public Comments

EPA received comments on two issues related to the NAAQS regulated under the proposed FIPs.

A number of commenters believed that EPA's approach to ozone was inadequate, and that EPA should not have based the proposed requirements on the 1997 ozone NAAQS. These commenters cited EPA's 2008 revision to the standard which lowered the standard to 75 ppb, and noted that EPA's January 2010 proposal for reconsidered ozone NAAQS would, if finalized, further lower the primary NAAQS from 75 ppb to a value between 60 and 70 ppb. Accordingly, many of the commenters believed that EPA should have considered the 75 ppb level to be the maximum possible value moving forward, and that EPA should have used a value no greater than 75 ppb in its analysis.

EPA agrees with commenters that EPA and states should address interstate transport with respect to the tighter ozone NAAQS as quickly as possible. EPA, as commenters noted, intends to propose a second rule to address interstate transport of ozone that will be appropriately configured for the revised level of the ozone NAAQS after reconsideration of the 2008 standard is finalized. EPA is mindful of the need for SIPs to provide for continuing ozone progress to meet the 75 ppb level of the 2008 NAAQS, or possibly lower levels based on the reconsideration. EPA believes that the ozone-season NO_X requirements of this rule will provide important initial assistance to states in this regard.

Some commenters questioned whether EPA had given states the opportunity to provide SIPs addressing transport under the 2006 PM_2.5 NAAQS, and thus questioned the appropriateness of the issuance of FIPs addressing those NAAQS. Those comments, and EPA's response, are discussed in detail in section IV.C.2.

2. FIP Authority for Each State and NAAQS Covered

The CAA requires and authorizes EPA to promulgate each of the Federal Implementation Plans in this final rule. Section 110(c)(1) of the CAA requires the Administrator to promulgate a FIP at any time within 2 years after the Administrator takes one of three distinct actions: (1) She finds that a state has failed to make a required SIP submission; (2) she finds a SIP submission to be incomplete; or (3) she disapproves a SIP submission. Once the Administrator has taken one of these actions with respect to a specific state's 110(a)(2)(D)(i)(I) obligation for a specific NAAQS, she has a legal obligation to promulgate a FIP to correct the SIP deficiency within 2 years. EPA is relieved of the obligation to promulgate a FIP only if two events occur before the FIP is promulgated: (1) The state submits a SIP correcting the deficiency; and (2) the Administrator approves the SIP revision. 42 U.S.C. 7410(c)(1).

For each FIP in this rule, EPA either has found that the state has failed to make a required 110(a)(2)(D)(i)(I) SIP submission, or has disapproved a SIP submission. In addition, EPA has determined, in each case, that there has been no approval by the Administrator of a SIP submission correcting the deficiency prior to promulgation of the FIP. EPA's obligation to promulgate a FIP arose when the finding of failure to submit or disapproval was made, and in no case has it been relieved of that obligation.

Some commenters argued that EPA was relieved of its obligation to promulgate FIPs when it approved the CAIR SIPs for certain states. As an initial matter, EPA notes that this argument applies only to EPA's authority to promulgate FIPs with respect to the 1997 PM_2.5 and/or 1997 ozone NAAQS for a subset of states covered by the CAIR. It does not apply to EPA's authority to promulgate FIPs for the 2006 PM_2.5 NAAQS which was not addressed in CAIR. It also does not apply to EPA's authority to promulgate FIPs for the 1997 ozone and 1997 PM_2.5 NAAQS for states that remain subject to the CAIR FIPs, including the states that received EPA approval of abbreviated CAIR SIPs which allowed the states to allocate allowances while remaining subject to the CAIR FIPs.

Further, the CAIR SIP approvals do not eliminate EPA's obligation and authority to promulgate a FIP to address the requirements of 110(a)(2)(D)(i)(I) because the Court in North Carolina v. EPA, 531 F.3d 896 (D.C. Cir. 2008) found that compliance with CAIR does not satisfy the requirement that each state prohibit all emissions within the state that significantly contribute to nonattainment or interfere with maintenance in another state. The Court's finding that CAIR was unlawful because it did not make measureable progress towards the statutory mandate of section 110(a)(2)(D)(i)(I) meant that the CAIR SIPs were not adequate to satisfy that mandate. The CAIR SIPs thus do not correct the SIP deficiencies identified in the 2005 findings of failure to submit. The SIPs remained in force for the limited purpose allowed by the Court—that is, to achieve interim reductions until EPA promulgated a rule to replace CAIR. Given the flaws the court identified with CAIR, EPA's approval of a CAIR SIP does not relieve it of the obligation to promulgate FIPs created under section 110(c)(1) of the CAA.

Further, to avoid any confusion, EPA has decided to correct, in this notice, the full CAIR SIP approvals for states covered by this rule and the CAA 110(a)(2)(D)(i) SIP approvals for states covered by CAIR to rescind any statements suggesting that the SIP submissions satisfied or relieved states of the obligation to submit SIPs to satisfy the requirements of section 110(a)(2)(D)(i)(I) or that EPA was relieved of its obligation and authority to promulgate FIPs under 110(a)(2)(D)(I)(i).

Some commenters further argued that states should be given additional time, following promulgation of the Transport Rule, to submit a SIP to meet the requirements of section 110(a)(2)(D)(i)(I) and that CAIR should remain in place in the meantime. Some commenters specifically suggested that EPA restart the “FIP clock” to give states this additional time. EPA does not interpret the CAA as giving it authority to extend the deadline for SIP submissions or restart the FIP clock. And nothing in the Act requires EPA to give the states another opportunity, following promulgation of the Transport Rule, to promulgate a SIP before EPA promulgates a FIP. The plain language of section 110(a)(1) of the Act requires the submission of SIPs that meet the requirements of 110(a)(2)(D)(i)(I) within 3 years after the promulgation of or revision of a primary NAAQS. See 42 U.S.C. 7410(a)(1). Section 110(a)(2)(D)(i)(I) SIPs for the 1997 ozone and PM_2.5 NAAQS were due in 2000 and 110(a)(2)(D)(i)(I) SIPs for the 2006 PM_2.5 NAAQS were due in 2009. While the statute gives EPA authority to prescribe a shorter period of time for states to make these SIP submissions, it does not give EPA authority to extend the 3-year deadline established by the Act. See 42 U.S.C. 7410(a)(1). The plain language of section 110(c)(1) of the Act, in turn, provides that EPA shall promulgate a FIP at any time within 2 years after the Administrator makes a finding of failure to make a required SIP submission of disapproves, in whole or in part, a SIP submission. See 42 U.S.C. 7410(c)(1). EPA does not have authority to set aside the specific deadlines established in the statute, and neither provision allows for the deadlines to be extended or to run from promulgation by EPA of a rule to quantify the state's specific obligations pursuant to section 110(a)(2)(D)(i)(I). The Act does not require EPA to promulgate a rule or issue guidance regarding the specific requirements of section 110(a)(2)(D)(i)(I) in advance of the SIP submittal deadline, much less require EPA to promulgate such a rule a specific amount of time before the SIP submittal deadline. For these reasons, EPA has neither authority to alter the SIP submittal deadline nor authority to alter the statute provision regarding when EPA's obligation to promulgate a FIP is triggered.

Finally, EPA does not believe it would be appropriate, in light of the Court's decision in North Carolina, to establish a lengthy transition period to the rule that will replace CAIR. The Court decision remanding CAIR without vacatur stressed the court's conclusion that CAIR was deeply flawed and emphasized EPA's obligation to remedy those flaws expeditiously. North Carolina, 550 F.3d 1176. Although the Court did not set a specific deadline for corrective action, the Court took care to note that the effect of its opinion would not be delayed “indefinitely” and that petitioners could bring a mandamus petition if EPA were to fail to modify CAIR in a manner consistent with its prior opinion. Id. Given the Court's emphasis on remedying CAIR's flaws expeditiously, EPA does not believe it would be appropriate to establish a lengthy transition period to the rule which is to replace CAIR.

3. Additional Information Regarding CAA Section 110(a)(2)(D)(i)(I) SIPs for States in the Transport Rule Modeling Domain

This final rule quantifies out-of-state contributions for the 38 states that are fully contained within the 12 kilometers (km) eastern U.S. modeling domain. EPA is making no specific finding for states that are not fully contained within the eastern 12 km modeling domain. EPA did not conduct a contribution analysis or make any specific finding for New Mexico, Colorado, Wyoming, and Montana since they are only partially contained within the 12 km modeling domain. With regard to the 1997 PM_2.5 NAAQS and 2006 PM_2.5 NAAQS, EPA believes that states that are included in this 38 state modeling domain will meet their section 110(a)(2)(D)(i)(I) obligations to address the “significant contribution” and “interference with maintenance” requirements by complying with the requirements in this rule. With regard to the 1997 ozone NAAQS, EPA believes that states that are included in this 38 state modeling domain will meet their section 110(a)(2)(D)(i)(I) obligations to address the “significant contribution” and “interference with maintenance” requirements by complying with the requirements in this rule, except for the 10 states found to significantly contribute to nonattainment or interference of maintenance in either Houston or Baton Rouge (i.e., Alabama, Arkansas, Georgia, Illinois, Indiana, Kentucky, Louisiana, Mississippi, Tennessee, and Texas). States that are in the 38 state modeling domain, and that are not found to be contributing significantly to nonattainment or interfering with maintenance for any NAAQS evaluated in the modeling for the final rule, could rely on this analysis as technical support that their existing or future interstate transport SIP submittals are adequate to address the transport requirements of 110(a)(2)(D)(i)(I). For example, this rule finds that South Carolina significantly contributes to nonattainment and interferes with maintenance of the 1997 ozone NAAQS and the 1997 PM_2.5 NAAQS in downwind states. The technical support for the rule does not show that South Carolina significantly contributes to nonattainment or interferes with maintenance of the 2006 PM_2.5 NAAQS in downwind states. EPA believes that South Carolina can make a negative declaration concluding that the state does not significantly contribute to nonattainment or interfere with maintenance in other states with regard to the 2006 PM_2.5 NAAQS.

D. Correction of CAIR SIP Approvals

In this action, EPA is also correcting its prior approvals of CAIR related SIP submissions and CAA 110(a)(2)(D)(i) SIP submissions from Alabama, Arkansas, Connecticut, Florida, Georgia, Illinois, Indiana, Iowa, Kentucky, Louisiana, Maryland, Massachusetts, Minnesota, Mississippi, Missouri, New York, North Carolina, Ohio, Pennsylvania, South Carolina, Virginia and West Virginia to rescind any statements that the SIP submissions either satisfy or relieve the state of the obligation to submit a SIP to satisfy the requirements of section 110(a)(2)(D)(i)(I) with respect to the 1997 ozone and/or 1997 PM_2.5 NAAQS or any statements that EPA's approval of the SIP submissions either relieve EPA of the obligation to promulgate a FIP or remove EPA's authority to promulgate a FIP. This action is based on EPA's determination that those SIP approvals were in error to the extent they provided explicitly or implicitly that compliance with CAIR satisfies the requirements of 110(a)(2)(D)(i)(I) with respect to the 1997 ozone and 1997 PM_2.5 NAAQS. The July 2008 decision of the DC Circuit held, among other things, that the CAIR rule did not “achieve[] something measureable toward the goal of prohibiting sources `within the State' from contributing to nonattainment or interfering with maintenance in `any other State.'” North Carolina, 531 F.3d 908; see also, e.g., id. at 916 (EPA not exercising its authority to make measureable progress towards the goals of section 110(a)(2)(D)(i)(I) because the emission budgets were insufficiently related to the statutory mandate). EPA's actions to approve CAIR SIP submittals as satisfying the requirements of section 110(a)(2)(D)(i)(I), based on the flawed determination in CAIR that compliance with CAIR satisfied those statutory requirements, were thus in error as were the separate actions taken to approve section 110(a)(2)(D)(i)(I) submissions that relied wholly or in part on CAIR.

The approval for Alabama titled “Approval and Promulgation of Implementation Plans; Alabama; Clean Air Interstate Rule” which is hereby corrected was originally published in the Federal Register on October 1, 2007 (72 FR 55659).

The approval for Arkansas titled “Approval and Promulgation of Implementation Plans; Arkansas; Clean Air Interstate Rule Nitrogen Oxides Ozone Season Trading Program” which is hereby corrected was originally published in the Federal Register on September 26, 2007 (72 FR 54556).

The approval for Connecticut titled “Approval and Promulgation of Air Quality Implementation Plans; Connecticut; State Implementation Plan Revision to Implement the Clean Air Interstate Rule” which is hereby corrected was originally published in the Federal Register on January 24, 2008 (73 FR 4105) and the approval for Connecticut titled “Approval and Promulgation of Air Quality Implementation Plans; Connecticut; Interstate Transport of Pollution” which is hereby corrected was originally published in the Federal Register on May 7, 2008 (73 FR 25516).

The approval for Florida titled “Approval and Promulgation of Implementation Plans; Florida; Clean Air Interstate Rule” which is hereby corrected was originally published in the Federal Register on October 12, 2007 (72 FR 58016).

The approval for Georgia titled “Approval and Promulgation of Implementation Plans; Georgia; Clean Air Interstate Rule” which is hereby corrected was originally published in the Federal Register on October 9, 2007 (72 FR 57202).

The approval for Illinois titled “Approval of Implementation Plans of Illinois: Clean Air Interstate Rule” which is hereby corrected was originally published in the Federal Register on October 16, 2007 (72 FR 58528).

The approval for Indiana titled “Limited Approval of Implementation Plans of Indiana: Clean Air Interstate Rule” which is hereby corrected was originally published in the Federal Register on October 22, 2007 (72 FR 59480) and the approval for Indiana titled “Approval and Promulgation of Air Quality Implementation Plans; Indiana; Clean Air Interstate Rule” which is hereby corrected was originally published in the Federal Register on November 29, 2010 (75 FR 72956).

The approval for Iowa titled “Approval and Promulgation of Implementation Plans; Iowa; Clean Air Interstate Rule” which is hereby corrected was originally published in the Federal Register on August 6, 2007 (72 FR 43539) and the approval for Iowa titled “Approval and Promulgation of Implementation Plans; Iowa; Interstate Transport of Pollution” which is hereby corrected was originally published in the Federal Register on March 8, 2007 (72 FR 10380).

The approval for Kentucky titled “Approval of Implementation Plans of Kentucky: Clean Air Interstate Rule” which is hereby corrected was originally published in the Federal Register on October 4, 2007 (72 FR 56623).

The approval for Louisiana titled “Approval and Promulgation of Implementation Plans; Louisiana; Clean Air Interstate Rule Sulfur Dioxide Trading Program” which is hereby corrected was originally published in the Federal Register on July 20, 2007 (72 FR 39741) and the approval for Louisiana titled “Approval and Promulgation of Implementation Plans; Louisiana; Clean Air Interstate Rule Nitrogen Oxides Trading Program” which is hereby corrected was originally published in the Federal Register on September 28, 2007 (72 FR 55064).

The approval for Maryland titled “Approval and Promulgation of Air Quality Implementation Plans; Maryland; Clean Air Interstate Rule” which is hereby corrected was originally published in the Federal Register on October 30, 2009 (74 FR 56117).

The approval for Massachusetts titled “Approval and Promulgation of Air Quality Implementation Plans; Massachusetts; State Implementation Plan Revision to Implement the Clean Air Interstate Rule” which is hereby corrected was originally published in the Federal Register on December 3, 2007 (72 FR 67854).

The approval for Minnesota titled “Approval and Promulgation of Air Quality Implementation Plans; Minnesota; Interstate Transport of Pollution” which is hereby corrected was originally published in the Federal Register on June 2, 2008 (73 FR 31366).

The approval for Mississippi titled “Approval and Promulgation of Implementation Plans; Mississippi: Clean Air Interstate Rule” which is hereby corrected was originally published in the Federal Register on October 3, 2007 (72 FR 56268).

The approval for Missouri titled “Approval and Promulgation of Implementation Plans; Missouri; Clean Air Interstate Rule” which is hereby corrected was originally published in the Federal Register on December 14, 2007 (72 FR 71073) and the approval of Missouri titled “Approval and Promulgation of Implementation Plans; Missouri; Interstate Transport of Pollution” which is hereby corrected was originally published in the Federal Register on May 8, 2007 (75 FR 25975).

The approval for New York titled “Approval and Promulgation of Implementation Plans; New York: Clean Air Interstate Rule” which is hereby corrected was originally published in the Federal Register on January 24, 2008 (73 FR 4109).

The approval for North Carolina titled “Approval of Implementation Plans; North Carolina: Clean Air Interstate Rule” which is hereby corrected was originally published in the Federal Register on October 5, 2007 (72 FR 56914) and the approval for North Carolina titled “Approval and Promulgation of Air Quality Implementation Plans; North Carolina; Clean Air Interstate Rule” which is hereby corrected was originally published in the Federal Register on November 30, 2009 (74 FR 62496).

The approval for Ohio titled “Approval and Promulgation of Air Quality Implementation Plans; Ohio; Clean Air Interstate Rule” which is hereby corrected was originally published in the Federal Register on February 1, 2008 (73 FR 6034) and the approval for Ohio titled “Approval and Promulgation of Air Quality Implementation Plans; Ohio; Clean Air Interstate Rule” which is hereby corrected was originally published in the Federal Register on September 25, 2009 (74 FR 48857).

The approval for Pennsylvania titled “Approval and Promulgation of Air Quality Implementation Plans; Pennsylvania; Clean Air Interstate Rule; NO_X SIP Call Rule; Amendments to NO_X Control Rules” which is hereby corrected was originally published in the Federal Register on December 10, 2009 (74 FR 65446).

The approval for South Carolina titled “Approval of Implementation Plans of South Carolina: Clean Air Interstate Rule” which is hereby corrected was originally published in the Federal Register on October 9, 2007 (72 FR 57209) and the approval for South Carolina titled “Approval and Promulgation of Air Quality Implementation Plans; South Carolina; Clean Air Interstate Rule” which is hereby corrected was originally published in the Federal Register on October 16, 2009 (74 FR 53167).

The approval for Virginia titled “Approval and Promulgation of Air Quality Implementation Plans; Virginia; Clean Air Interstate Rule Budget Trading Programs” which is hereby corrected was originally published in the Federal Register on December 28, 2007 (72 FR 73602).

The approval for West Virginia titled “Approval and Promulgation of Air Quality Implementation Plans; West Virginia; Clean Air Interstate Rule” which is hereby corrected was originally published in the Federal Register on December 18, 2007 (72 FR 71576) and the approval for West Virginia titled “Approval and Promulgation of Air Quality Implementation Plans; West Virginia; Clean Air Interstate Rule” which is hereby corrected was originally published in the Federal Register on August 4, 2009 (74 FR 38536).

EPA is taking this final action without prior opportunity for notice and comment because EPA finds, for good cause, that notice and public procedure thereon are unnecessary and not in the public interest. Section 553(b)(B) of the Administrative Procedure Act provides that the notice and comment requirements in section 553 do not apply when the agency for good cause finds that notice and public procedure there on are impracticable, unnecessary, or contrary to the public interest. 5 U.S.C. 553(b)(B). Section 307(d)(1) of the CAA in turn provides that the requirements of section 307(d) do not apply in the case of a rule or circumstance referred to in section 553(b)(A) or section 553(b)(B) of the Administrative Procedure Act in Title 5. 42 U.S.C. 7607(1).

EPA finds that notice and public procedure are unnecessary because EPA has no discretion given the specific circumstances presented in this case. EPA is bound by the decisions of the courts and must act in accordance with those decisions. EPA must accept the Court's conclusion that compliance with CAIR does not satisfy the requirements of CAA section 110(a)(2)(D)(i)(I) and lacks discretion to reach a different conclusion. This correction is a ministerial matter consistent with the decisions of the courts. For these reasons, it is unnecessary to provide an opportunity for notice and comment.

V. Analysis of Downwind Air Quality and Upwind State Emissions

A. Pollutants Regulated

To address interstate transport of air pollution, EPA must choose which pollutants to regulate relevant to significant contribution to nonattainment or interference with maintenance of the NAAQS of concern downwind. This section of the preamble discusses the pollutants regulated under the final Transport Rule.

1. Background

Based on scientific and technical information, as well as EPA's air quality modeling, EPA concluded for CAIR that the most effective approach to reducing the contribution of interstate transport to PM_2.5 was to control SO₂ and NO_X emissions. For CAIR, EPA did not limit emissions of other components of PM_2.5, noting that “current information relating to sources and controls for other components identified in transported PM_2.5 (carbonaceous particles, ammonium, and crustal materials) does not, at this time, provide an adequate basis for regulating the regional transport of emissions responsible for these PM_2.5 components” (69 FR 4582).

With respect to ozone transport, EPA has previously concluded that it is proper to control ozone-season NO_X emissions. For CAIR and the NO_X SIP Call programs, EPA based this conclusion on the assessment of ozone transport conducted by the Ozone Transport Assessment Group (OTAG) in the mid-1990s. The OTAG Regional and Urban Scale Modeling and Air Quality Analysis Work Groups concluded that regional NO_X emission reductions are effective in producing ozone benefits that grow with increasing regional NO_X abatement.

The relative importance of NO_X and VOC in ozone formation and control varies with local and time-specific factors, including the relative amounts of VOC and NO_X present. In rural areas and many urban areas with high concentrations of VOC from biogenic sources, ozone formation and control is governed by NO_X. In some urban core situations, NO_X concentrations can be high enough relative to VOC to suppress ozone formation locally, but still contribute to increased ozone downwind from the city. In such situations, VOC reductions are most effective at reducing ozone within the urban environment and immediately downwind. The formation of ozone increases with temperature and sunlight, which is one reason ozone levels are higher during the summer. Increased temperature also increases emissions of volatile man-made and biogenic organics and can indirectly increase NO_X as well (e.g., increased electricity generation for air conditioning). Summertime conditions also bring increased episodes of large scale stagnation of air masses, which promote the build-up of direct emissions and pollutants formed through atmospheric reactions over large regions. Authoritative assessments of ozone control approaches have concluded that, for reducing regional scale ozone transport, a NO_X control strategy is most effective, whereas VOC reductions are generally most effective locally, in more dense urbanized areas.

Studies conducted since the 1970s established that ozone occurs on a regional scale (i.e., thousands of kilometers) over much of the eastern U.S., with elevated concentrations occurring in rural as well as metropolitan areas. While substantial progress has been made in reducing ozone in many urban areas, regional-scale ozone transport is still an important component of high ozone concentrations during the extended summer ozone season. A series of more recent progress reports discussing the effect of the NO_X SIP Call reductions can be found on EPA's Web site at: http://www.epa.gov/airmarkets/progress/progress-reports.html.

More recent assessments of ozone (including those conducted for the Regulatory Impact Analysis for the ozone standards in 2008) continue to show the importance of NO_X transport as a factor in ozone formation. For addressing interstate ozone transport in CAIR, EPA required NO_X emission reductions but did not include requirements for VOCs. EPA believes that VOCs from some upwind states do indeed have an impact in some nearby downwind states, particularly over short transport distances. EPA expects that states, typically in local nonattainment planning, would benefit from examining the extent to which VOC emissions affect ozone pollution levels within and near urban nonattainment areas, and states may identify areas where multi-state VOC strategies might assist in attainment planning for meeting the 8-hour standard. However, EPA continues to believe that the most effective regional pollution control strategy for mitigation of interstate transport of ozone remains NO_X emission reductions.

2. Which pollutants did EPA propose to control for purposes of PM_2.5 and ozone transport?

For the proposed rule, EPA concluded that its findings in CAIR regarding the nature of pollutant contributions are still appropriate. EPA proposed to require SO₂ and annual NO_X emission reductions to control PM_2.5 transport and to require ozone-season NO_X emission reductions to control ozone transport. In the proposal, EPA discussed and requested comment on the inclusion of southern states in the annual NO_X program for PM_2.5 control.

3. Comments and Responses

EPA received no adverse comments on its proposal to regulate SO₂ for addressing PM_2.5 transport, the proposal not to regulate direct PM_2.5 or organic PM_2.5 precursors, and the proposal to focus ozone-season efforts on NO_X and not to regulate VOCs.

One commenter questioned EPA's regulation of NO_X for purposes of addressing PM_2.5 transport in all states (including northern states with cooler climates and higher nitrate deposition). Several commenters, representing southern state air quality agencies and regulated sources in southern states, disagreed with EPA's proposed regulation of annual NO_X emissions for all regulated states. These commenters, while not disagreeing with the need for regulation of SO₂, observed that in EPA's modeling analysis, contributions from certain southern states' NO_X emissions to PM_2.5 in downwind states were relatively small.

Accordingly, these commenters argued that either (1) EPA should remove NO_X as a precursor analyzed for PM_2.5 contribution from those states, or (2) the required remedy for emission reductions in those states should not require reductions in annual NO_X.

For the final rule, EPA retains the approach for regulated pollutants in the proposal, which regulates annual NO_X and SO₂ for states affecting downwind state PM_2.5 nonattainment and maintenance sites, and ozone-season NO_X for states impacting downwind state ozone nonattainment and maintenance. EPA considered commenters' requests to remove some states from the annual NO_X program. However, EPA believes that it is appropriate to establish a cap on these states' annual NO_X emissions, in part to ensure the continued annual operation of existing control equipment that would prevent substantial increases in NO_X emissions. EPA believes that without these reductions, increased “nitrate replacement” could occur, a known atmospheric phenomenon whereby some of the sulfate reductions due to SO₂ emission reductions are eroded by increases in nitrate concentrations due solely to those SO₂ reductions. This is an especially pertinent concern for southern states which have significant impacts on northern receptors in colder climates where nitrate concentrations are generally higher. For example, Alabama and Tennessee are both linked to Washtenaw County, MI for 24-hour PM_2.5; North Carolina is linked to Lancaster County, PA for 24-hour PM_2.5; and Texas is linked to Madison County, IL for both annual and 24-hour PM_2.5. All of these downwind areas have appreciable nitrate deposition contributing to nonattainment and maintenance concerns for the PM_2.5 NAAQS. If the states linked to those receptors were to make SO₂ reductions only, their beneficial impact on downwind air quality would be partially eroded by nitrate replacement. EPA therefore believes that it is reasonable to seek both SO₂ and NO_X reductions from states included in the Transport Rule program that are found to significantly contribute to nonattainment or interfere with maintenance of the PM_2.5 NAAQS in downwind states.

In addition, EPA notes that there would be important disbenefits to effectively removing CAIR's existing annual NO_X requirements in those states. If EPA were to allow annual NO_X emissions to increase for those states, there would be potentially harmful effects on visibility, nitrogen deposition, and other aspects of human and environmental health.

B. Baseline for Pollution Transport Analysis

Implementing the mandate of CAA section 110(a)(2)(D)(i)(I) requires EPA to determine which states significantly contribute to nonattainment and interfere with maintenance of the NAAQS in other states, as well as to quantify the emissions in each state that must be eliminated. This process begins with an analysis of baseline emissions. Baseline emissions are the emissions that would occur in each state if EPA did not promulgate the Transport Rule. To conduct such analysis, EPA generally takes into account emission limitations that are currently, and will continue to be, in place. From that baseline, EPA analyzes whether additional reductions are necessary beyond those already mandated by existing emission limitation requirements. For example, the base case used in CAIR reflected the reductions already required by the NO_X SIP Call, which remained in effect even after the CAIR emission reduction requirements took effect.

The unique legal situation addressed by the Transport Rule necessarily affects the quantification of baseline emissions. Specifically, because the Transport Rule will replace CAIR, EPA cannot consider reductions associated with CAIR in the “base case” (i.e., analytical baseline emissions scenario). If EPA were to consider all reductions associated with CAIR in the “base case,” the baseline emissions would not adequately reflect the true 2012 baseline in each state (i.e., the emissions that would occur in each state in 2012 if the Transport Rule did not require any reductions in that state). Similarly, if EPA were to treat the capital investments that have already been made to meet the requirements of CAIR as new costs rather than treating them as “sunk” capital costs, EPA's analysis would not accurately reflect the cost of emission reductions required by the Transport Rule. As explained below, EPA's analysis both properly considered all capital investments made in response to CAIR and properly recognized that, after CAIR is terminated, the emission limitations imposed by CAIR will cease to exist.

In 2005 EPA promulgated CAIR, which required large electric generating units in 29 states to make phase I emission reductions in NO_X emissions starting in 2009, phase I emission reductions in SO₂ starting in 2010 and phase II reductions in emissions of both pollutants starting in 2015. On July 11, 2008, the DC Court of Appeals held that CAIR had “more than several fatal flaws,” North Carolina, 531 F.3d at 901, and remanded and vacated the rule, id. at 930. The Court subsequently granted EPA's petition for rehearing in part and remanded CAIR without vacatur “for EPA to conduct further proceedings consistent with” the Court's July 11, 2008 opinion. North Carolina, 550 F.3d 1176. The Court explained that it was “allowing CAIR to remain in effect until it is replaced by a rule consistent with [the July 11, 2008] opinion” because this “would at least temporarily preserve the environmental values covered by CAIR.” Id. at 1178. Moreover, the Court stated that it did not “intend to grant an indefinite stay of the effectiveness of” the July 11, 2008 order vacating CAIR. Id. In summary, the Court determined that CAIR was fatally flawed and could remain in effect only as a stopgap measure until EPA could act to replace it.

Thus, unlike most other regulatory requirements (such as the Acid Rain Program under CAA Title IV, the NO_X Budget Trading Program under the NO_X SIP Call, New Source Performance Standards, and state laws and consent orders requiring emission reductions), the emission limitations contained in CAIR are only temporary. Moreover, the duration of these limitations is directly tied to the Transport Rule. The Transport Rule replaces CAIR. Thus, CAIR itself will be terminated for the SO₂, annual NO_X, and ozone-season NO_X control periods starting in 2012 when the emission limitations established in the final Transport Rule for those control periods take effect (January 1, 2012 for the annual control periods and May 1, 2012 for the ozone-season control period). For this reason, emission reductions made to comply with CAIR cannot be treated as if they were emission reductions achieved to comply with statutory provisions, rules, consent decrees, and other enforceable requirements that establish permanent emission limitations. EPA takes reductions made to comply with permanent limitations into consideration when quantifying each state's baseline emissions for the purpose of analyzing whether its emissions significantly contribute to nonattainment or interfere with maintenance in another state. However, the unique legal status of CAIR and its replacement with the Transport Rule distinguish the emission reductions required by CAIR from those of other regulatory requirements. Since the limitations and emission reduction requirements in CAIR are temporary and will be terminated by the Transport Rule, they must be excluded from the Transport Rule's base case analysis.

Some comments on the Transport Rule proposal claim that EPA's treatment of CAIR is inconsistent with the treatment, in prior rulemakings, of the Acid Rain Program and the NO_X SIP Call. Such comments ignore the unique legal status of CAIR, and EPA therefore rejects these claims.

A simple example illustrates this point. Assume state Z's emissions before CAIR were 2,000 tons and that state Z was required by CAIR to reduce its emissions to 1,000 tons. If EPA were to determine that state Z's baseline emissions were 1,000 tons and then conclude, based on that assumption, that no additional reductions in state Z are necessary because state Z does not significantly contribute to downwind nonattainment unless its emissions exceed 1,500 tons, then state Z would not be covered by the Transport Rule. However, the Transport Rule will terminate all CAIR requirements in all CAIR states regardless of whether they are covered by the Transport Rule. Thus, after promulgation of the Transport Rule, state Z would again be allowed, and would be projected in this example, to emit 2,000 tons. In other words, state Z would be allowed to significantly contribute to nonattainment and/or interfere with maintenance in other states—a result that would be inconsistent with the statutory mandate of CAA section 110(a)(2)(D)(i)(I). On the other hand, if EPA assumes state Z's baseline emissions are 2,000 tons as projected without CAIR in place, EPA can properly determine whether, if state Z were allowed to emit that amount (i.e., the amount state Z would be projected to emit if excluded from the Transport Rule), the state would significantly contribute to nonattainment or interfere with maintenance in any other state. In other words, EPA can determine the stringency of emission limitations needed (if any) to replace those that were established by CAIR in order to ensure that state Z prohibits all emissions that significantly contribute to nonattainment or interfere with maintenance in other states.

In fact, commenters' suggestion that the Transport Rule base case should include CAIR would cause the anomalous result of excluding sources in a state from the Transport Rule because of their CAIR-required emission reductions while simultaneously eliminating those CAIR emission reduction requirements. If EPA's base case analysis were to assume erroneously that reductions from CAIR would continue indefinitely, a state currently covered by CAIR, but not covered by the Transport Rule, would have no CAIR requirements once the Transport Rule programs began and so could increase emissions beyond the CAIR limitations. Downwind areas that are in attainment (and are not experiencing interference with maintenance of such attainment) solely because of emission reductions required by CAIR could again face nonattainment or interference with maintenance problems because the current protection from upwind pollution from such an upwind state would not be replaced. In short, the analysis of whether a state should be included in a rule eliminating and replacing CAIR cannot logically assume that CAIR remains in place. For these reasons, EPA believes it is reasonable to use a base case that does not assume that the CAIR reduction requirements will continue to be achieved and so does not include CAIR-specific emission reductions.

As a result, EPA's 2012 base case shows emissions higher than current levels in some states. In the absence of the CAIR SO₂ and NO_X programs that EPA has been directed to eliminate and replace, utility emissions in CAIR states will be limited only by non-CAIR constraints including the Acid Rain Program, the NO_X SIP Call, New Source Performance Standards, any state laws and consent order requiring emission reductions, and any other permanent and enforceable binding reduction commitments. This will lead to increased emissions in some states in the 2012 base case relative to current emissions. For example, efforts to comply with the Acid Rain Program at the least cost may occur, in some cases, without the operation of existing scrubbers through use of readily available, inexpensive Title IV allowances.

It is important to note that, to the extent that emission reductions currently required by CAIR are also reflected in emission reduction requirements under the Acid Rain Program, the NO_X SIP Call, New Source Performance Standards, any state laws and consent orders requiring emission reductions, and any other enforceable binding reduction commitments, such reductions are accounted for in EPA's 2012 base case. Some commenter claimed that in excluding CAIR-specific emission reductions from the base case, EPA ignores non-CAIR legal requirements (e.g., in Title V permits) that may prevent sources from increasing emissions above CAIR levels. Such allegations are incorrect. As discussed elsewhere in this preamble, EPA accounted for any Title V permits, consent decrees, state rules, and other enforceable limitations on sources' emissions; if these non-CAIR limitations effectively restrain a state's emissions to not exceed the state's CAIR limitations, EPA's base case modeling would reflect this outcome. Commenters also assert that utilities are unlikely to dismantle or discontinue running the installed controls to the point of returning to pre-CAIR emission levels. EPA agrees that installed controls are not likely to be physically dismantled, and as discussed elsewhere in this preamble, EPA's analysis properly treats the capital investments made in emission controls attributed to CAIR as “sunk” capital costs (i.e., capital costs already obligated in the past) that are not included as costs of meeting Transport Rule requirements.

Our cost analysis for significant contribution reflects on-the-ground realities. Investments in pollution control equipment were made in response to CAIR requirements. Those expenditures are “sunk” capital costs, meaning that those investments were committed in the past, prior to the Transport Rule. Adding the capital costs of that equipment into the costs of Transport Rule emission reduction options would be incorrect; those capital investments are represented in place in the base case.

However, given ongoing costs associated with operating these controls, EPA believes sources would have an economic incentive to discontinue operating installed controls, or to operate those controls less effectively, except to the extent non-CAIR legal requirements mandate emission reductions or to the extent that sources would find it economic to operate the controls for non-CAIR market-based emission control programs. EPA properly treats the costs of operating controls installed to meet CAIR requirements as costs of meeting Transport Rule requirements. EPA's base case accounts for non-CAIR requirements and does not make the unreasonable assumption that installed controls would be operated to achieve emission reductions that are not necessary to meet non-CAIR requirements. For all of these reasons, EPA rejects commenters' claims that the base case is “unrepresentative” or lacks “a rational relationship to the real world.”

C. Air Quality Modeling To Identify Downwind Nonattainment and Maintenance Receptors

1. Emission Inventories

To inform air quality modeling for the development of the final Transport Rule, EPA developed emission inventories for a 2005 base year and for 2012 and 2014 projections. The inventories for all years include emission estimates for EGUs, non-EGU point sources, stationary nonpoint sources, onroad mobile sources, nonroad mobile sources, and biogenic (non-human) sources. EPA's air quality modeling relies on this comprehensive set of emission inventories because emissions from multiple source categories are needed to model ambient air quality and to facilitate comparison of model outputs with ambient measurements. In addition, EPA considers all relevant emissions (regardless of source category) when determining whether a state is found to be significantly contributing to or interfering with maintenance of a particular NAAQS in another state.

The emission inventories were processed through the Sparse Matrix Operator Kernel Emissions (SMOKE) Modeling System version 2.6 to produce the gridded, hourly, speciated, model-ready emissions for input to the CAMx air quality model. Additional information on the development of the emission inventories and related data sets for emissions modeling are provided in the Emission Inventory Final Transport Rule TSD.

On October 27, 2010, EPA issued a NODA on “Revisions to Emission Inventories.” The NODA's primary purpose was to notify the public about changes to emission inventories made since the proposal modeling. The affected emission sectors were non-EGU stationary point sources, nonpoint sources, and Category 3 commercial marine vessel sources. The NODA also presented a newly released model for developing onroad mobile source emissions for use in air quality modeling for the final Transport Rule.

The major comments received in response to the emission inventories and modeling included in the proposed Transport Rule and the October 27 NODA are summarized in the following subsections. EPA agreed with the comments summarized below and adopted technical corrections or updates to the emission inventories and modeling accordingly. For EPA to be able to take appropriate action, comments on the emission inventories needed to be specific enough to allow for credible alternative data sources to be located. EPA adopted corrections from comments on in-place control programs or devices where the controls were enforceable and quantifiable.

a. Foundation Emission Inventory Data Sets

EPA developed emission data representing the year 2005 to support air quality modeling of a base year from which future air quality could be forecasted. EPA used the 2005 National Emission Inventory (NEI), version 2 from October 6, 2008, as the chief basis for the U.S. inventories supporting the 2005 air quality modeling. This inventory includes 2005-specific data for point and mobile sources, while most nonpoint data were carried forward from version 3 of the 2002 NEI. The future base case scenarios modeled for 2012 and 2014 represent predicted emission reductions primarily from already promulgated federal measures.

EPA used a 2006 Canadian inventory and a 1999 Mexican inventory for the portions of Canada and Mexico within the air quality modeling domains for all modeled scenarios. Emissions from Canada and Mexico for all source sectors (including EGUs) in these countries were held constant for all base- and future-year cases. EPA made this assumption because it does not currently have sufficient data to support projections of future-year emissions from Canada and Mexico.

b. Development of Emission Inventories for EGUs

The annual NO_X and SO₂ emissions for EGUs in the 2005 NEI v2 are based primarily on data from continuous emissions monitoring systems (CEMS), with other EGU pollutants estimated using emission factors and annual heat input data reported to EPA. Although only NO_X and SO₂ are considered for control in this rule, emissions for all criteria air pollutants are necessary to model air quality. For EGUs without CEMS, EPA used data submitted to the NEI by the states. For more information on the details of how the 2005 EGU emissions were developed, see the Emissions Inventory Final Rule TSD.

Commenters stated that some point sources that were classified as non-EGUs in the proposal modeling were actually EGUs, resulting in double counting of emissions in future-year modeling. EPA reviewed its assignment of EGUs and non-EGUs and reclassified EGU sources found to be in the non-EGU inventory for the updated 2005 EGU inventory to prevent double counting of future-year emissions.

The future base case scenarios for EGUs reflect projected changes to fuel usage and economics, as described in the Emission Inventory Final Rule TSD. Future year base case EGU emissions that predict SO₂, NO_X, and PM_2.5 were obtained from version 4.10_FTransport of the Integrated Planning Model (IPM) outputs ( http://www.epa.gov/airmarkt/progsregs/epa-ipm/index.html ). The IPM is a multi-regional, dynamic, deterministic linear programming model of the U.S. electric power sector; version 4.10_FTransport reflects state rules and consent decrees through December 1, 2010, and incorporates public comments on existing controls submitted to EPA through both the Transport Rule-related notice and comment process as well as the proposed Mercury and Air Toxics Standards Information Collection Request (ICR). The operation of existing SO₂ or NO_X advanced controls (e.g., scrubber, SCR) on units that were not required to operate those controls for compliance with Title IV, New Source Review (NSR), state settlements, or state-specific rules was projected by IPM on the basis of providing least cost operation of the power generation system subject to existing regulatory requirements except CAIR (see baseline discussion in section V.B).

Additionally, IPM v.4.10_FTransport incorporates comments received during the rulemaking process. Fuel-related updates include comment-driven unit-specific limitations on 2012 coal rank selection, limiting unrestricted switching from bituminous to subbituminous coal by imposing boiler modification costs for those units shifting from bituminous to subbituminous coal without historical precedent, and a correction of waste coal prices. Pollution control-related updates include keying the performance assumptions for FGD and SCR more closely to historic performance data, and the inclusion of dry sorbent injection (DSI), a SO₂ removal technology. Other notable updates include revised assumptions on the heat rate and consequent dispatching of cogenerating units and incorporation of additional planned retirements. Further details on these updates are available in the IPM Documentation, available in the docket and at: http://www.epa.gov/airmarkets/progsregs/epa-ipm/index.html.

c. Development of Emission Inventories for Non-EGU Point Sources

Details on the development of emission inventories are available in the Emission Inventory Final Rule TSD. In both the proposal and final modeling, controls on industrial boilers installed under the NO_X SIP call were assumed to have been implemented by 2005 and captured in the 2005 NEI v2. The non-EGU point source emissions were updated from the 2005 NEI and the emissions used for the proposal modeling through the incorporation of comments on the proposal emissions values, previously unknown facility closures, and through other data improvements as identified by EPA analyses.

EPA does not factor in economic growth to develop non-EGU point source emission projections because analysis of historical emission trends and economic data did not support using economic growth to project non-EGU emissions. More details on the rationale for not applying economic growth to non-EGU industrial sources can be found in Appendix D of the Regulatory Impact Assessment (RIA) for the PM NAAQS rule ( http://www.epa.gov/ttn/ecas/regdata/RIAs/Appendix%20DInventory.pdf ). Although projections based on economic growth were not included, EPA did include reductions resulting from plant and unit closures, local and federal consent decrees, and several Maximum Achievable Control Technology (MACT) standards.

For non-EGU point sources, local control programs that may be necessary for areas to attain the annual PM_2.5 NAAQS and the ozone NAAQS are only included in the future base case projections when specific information about existing enforceable local controls was provided.

Since aircraft at airports were treated as point emissions sources in the 2005 NEI v2, we applied projection factors based on activity growth projected by the Federal Aviation Administration Terminal Area Forecast (TAF) system, published in December 2008.

A number of comments were received on the stationary non-EGU point source inventories. Below is a summary of the major comments that impacted the stationary non-EGU point source inventories for the final modeling:

Comment: Commenters stated that EPA did not properly represent some point source emissions in base-year and future-year inventories due to facility and unit closures, consent decrees, emission caps, control programs, and alternative emission estimates.

Response: EPA reviewed the sources referenced in the individual comments regarding the base-year and future-year inventories. In cases where credible alternative data were available, EPA revised the emission inventories to incorporate additional facility and unit closures, consent decrees, emission caps, control programs, enforceable local controls, and alternative emission estimates.

Comment: Commenters stated that EPA should include controls from the National Emission Standards for Hazardous Air Pollutants for Reciprocating Internal Combustion Engines (RICE NESHAP) in our modeling.

Response: EPA included reductions expected to be achieved by the RICE NESHAP across the United States in our final modeling of stationary non-EGU and nonpoint sources.

Comment: Commenters stated that EPA was not properly representing existing or planned controls for cement plants.

Response: EPA updated control and projection information for cement plants based on the latest available data and cement sector-specific modeling results.

Comment: EPA specifically requested comments on whether to incorporate emission reduction estimates from the NESHAP for Major Sources: Industrial, Commercial, and Institutional Boilers and Process Heaters (75 FR 32006). Commenters stated that emission reduction estimates should not be included until the rule became final.

Response: EPA did not incorporate emission reduction estimates from the NESHAP for Major Sources: Industrial, Commercial, and Institutional Boilers and Process Heaters (75 FR 32006) into the proposal or final modeling because the rule was not final at the time the modeling was performed. Note that reductions from this rule would not have impacted the 2012 base case due to its implementation schedule, and only the 2014 emissions would have been affected.

d. Development of Emission Inventories for Onroad Mobile Sources

The onroad emissions in the proposal modeling were primarily based on the National Mobile Inventory Model (NMIM) monthly, county, and process level emissions along with gasoline exhaust emissions from a fall 2008 draft version of the Motor Vehicle Emission Simulator (MOVES). A major comment on the proposal modeling for onroad mobile sources was the following:

Comment: Commenters stated that EPA should use a publicly released version of MOVES for its final modeling.

Response: EPA updated the final modeling to use data from the publicly released version of the MOVES 2010 model because the model became available in time for inclusion of its results in the final modeling. It was not used for the proposal modeling because it was not available at the time the modeling was performed.

In the final Transport Rule modeling, EPA used MOVES 2010 state-month level emissions for all criteria pollutants and all modes (evaporative, exhaust, brake wear and tire wear) and allocated those emissions to counties according to state-county NMIM emissions ratios. For California (the emissions for which are included to support the coarse modeling domain), the onroad mobile emissions data were derived from data provided by the state. These data were augmented with MOVES 2010 outputs for NH₃ because data for that pollutant had not been provided. Additional information on the approach to onroad mobile source emissions is available in the Emission Inventory Final Rule TSD.

In the future-year base modeling for mobile sources, all national measures available at the time of modeling were included. The future scenarios for mobile sources reflect projected changes to fuel usage, as described in the Emission Inventory Final Rule TSD. Emissions for these years reflect onroad mobile control programs including the Light-Duty Vehicle Tier 2 Rule, the Onroad Heavy-Duty Rule, the Light-Duty Vehicle Greenhouse Gas Rule, the Renewable Fuel Standards Rule, and the Mobile Source Air Toxics (MSAT) final rule.

e. Development of Commercial Marine Category 3 Vessel Emission Inventories

For the 2005 modeling, the commercial marine category 3 (C3) vessel emissions, a portion of nonroad mobile emissions, were augmented with gridded 2005 emissions from the previous modeling efforts for the rule called “Control of Emissions from New Marine Compression-Ignition Engines at or Above 30 Liters per Cylinder.” Emissions out to 200 nautical miles from the coastline were allocated to states in the proposal modeling. A major comment on the proposal modeling was the following:

Comment: Commenters stated that emissions from commercial marine sources (a component of the nonroad emissions in the summaries that were provided for the NPR) were too high.

Response: EPA reviewed the approach used for commercial marine C₃ emissions in the proposal. In the final modeling, instead of using the boundary of 200 nautical miles from the coast as was used in the proposal, EPA adopted the Mineral Management Service state-federal water boundaries that assign state waters 3-10 nautical miles from the coast. This approach is consistent with the approach used in the 2005 and 2008 National Emission Inventories. In addition, the category 3 commercial marine emissions were adjusted to reflect a coordination between the Emissions Control Area proposal to the International Maritime Organization (EPA-420-F-10-041, August 2010) control strategy; reductions of NO_X, VOC, and CO emissions for new C₃ engines starting in 2011; and fuel sulfur limits that go into effect as early as 2010.

f. Development of Emission Inventories for Other Nonroad Mobile Sources

The nonroad mobile source emissions for sources other than C₃ marine were primarily based on NMIM monthly, county, and process level emissions from the 2005 NEI v2. These emissions were unchanged from proposal modeling, except for PM emissions in California that were updated to correct for missing emissions in a few counties and source categories.

Nonroad mobile emissions were created for future years with NMIM using an approach consistent with that used for 2005. The nonroad emissions for 2012 and 2014 were calculated using NMIM future-year equipment population estimates and control programs. Nonroad mobile emission reductions for 2012 and 2014 include reductions to locomotives, various nonroad engines including diesel engines and various marine engine types, fuel sulfur content, and evaporative emissions standards. A more comprehensive list of control programs included for mobile sources is available in the Emission Inventory Final Rule TSD.

The 2012 and 2014 nonroad mobile emissions for locomotives and category 1 and 2 (C1 and C2) commercial marine vessels were based on emissions published in EPA's Locomotive Marine Rule, Regulatory Impact Assessment, Chapter 3.

g. Development of Nonpoint Emission Inventories

For the proposal Transport Rule modeling, EPA augmented the 2002 NEI nonpoint emission inventory with a non-California Western Regional Air Partnership (WRAP) oil and gas exploration inventory, which includes emissions in several states within the eastern U.S. 12 km modeling domain and additional states within the national 36 km modeling domain. For the final Transport Rule modeling, EPA updated the nonpoint emission estimates for oil and gas sources. EPA continued to use the same WRAP inventory from the proposal, emissions in Texas and Oklahoma were updated but for the final modeling with data from the Texas Commission on Environmental Quality (TCEQ) and the Oklahoma Department of Environmental Quality (DEQ), respectively.

The average-year county-based inventories for wildfire and prescribed burning emissions were unchanged between the proposal and final modeling.

For stationary nonpoint sources, local control programs that may be necessary for areas to attain the annual PM_2.5 NAAQS and the ozone NAAQS are not included in the future base case projections unless specific information about existing enforceable controls was available (e.g., ozone SIP controls from Ozone Transport Commission rules that impact source categories such as Consumer Products, Solvent Cleaning, Adhesives and Sealants). EPA specifically requested comment on local control data as part of the proposal and the October 27 NODA, and incorporated any usable data that was provided into the final inventories.

For stationary nonpoint sources, refueling emissions were projected using the refueling results from the NMIM runs performed for the onroad mobile sector.

Portable fuel container emissions were projected to future years using estimates from previous OTAQ rulemaking inventories. Emissions of ammonia and dust from animal operations were projected based on animal population data from the Department of Agriculture and EPA. Residential wood combustion was projected by replacement of obsolete wood stoves with new wood stoves and a 1 percent annual increase in fireplaces. Landfill emissions were projected using MACT controls. All other nonpoint sources were held constant between 2005 and the future years.

Some specific adjustments to the inventories were made in the final modeling to address comments that were received as described below. Area source MACT programs and controls from the RICE NESHAP were included in the final modeling to address submitted comments, as were fuel sulfur controls that were enforceable and that take effect by 2014.

The major comments that impacted the nonpoint sectors are as follows:

Comment: Commenters stated that the SO₂ emissions from industrial fuel combustion in Nebraska EPA are too high.

Response: EPA reviewed the NEI 2002-based data that had been used for the proposal modeling and determined that emissions from the 2005 inventory compiled for the Central Regional Air Planning Association (CENRAP) were more up to date for this source category and based on more localized data sources. The 2005 CENRAP emissions for industrial fuel combustion were used in the final modeling.

Comment: Commenters stated that EPA should include sulfur rule controls that take effect prior to the future years that were modeled.

Response: EPA included quantifiable sulfur rule controls in 2014 modeling for those states that had implemented the rules (New Jersey and Maine).

Comment: A commenter stated that emissions for Delaware were overestimated for several nonpoint categories in base-year and future-year inventories and provided alternative estimates for these categories.

Response: EPA reviewed the alternative estimates provided and found them to be credible and based on more detailed local scale information than were available in the national inventories. EPA incorporated the alternative emission estimates for Delaware into the final modeling.

Comment: A commenter stated that residual oil is not used as an industrial fuel in South Carolina.

Response: EPA analyzed the emissions from residual oil industrial fuel combustion in South Carolina and all other states, and analyzed preliminary regional planning office inventories and the 2008 NEI submittals. The South Carolina residual oil industrial fuel emissions were determined to be anomalously large in comparison to the near zero emissions in other submittals and were therefore removed from the nonpoint inventory.

2. Air Quality Basis for Identifying Receptors

a. Introduction

In this section, we describe the final approach to identify downwind nonattainment and maintenance receptors. We briefly summarize the modeling platform, the proposed approach to identify receptors, comments received, and the results of the final analysis.

In the Transport Rule, EPA has explicitly given independent meaning to the “interfere with maintenance” prong of section 110(a)(2)(D)(i)(I) by evaluating contributions to identified maintenance receptors as well as contributions to identified nonattainment receptors. EPA identified maintenance receptors as those receptors that would have difficulty maintaining the relevant NAAQS in a scenario that takes into account historic variability in air quality at that receptor. Specifically, EPA projects future air quality design values based on measured data during the period 2003 to 2007. In determining the downwind receptors of concern, EPA does not solely rely on the projection of an average design value based on measured data from the relevant period (in this case 2003 to 2007) to make a determination of “attainment” or “nonattainment.” Instead, EPA also evaluates the maximum future design value at that receptor based on measured data over the relevant period. Receptors for which this latter analysis projects design values higher than the NAAQS are identified as maintenance receptors.

EPA believes it is appropriate and reasonable to use this approach to identify receptors that may have maintenance problems in the future. This approach uses measured data in order to establish potential air quality outcomes at each receptor that take into account the variable meteorological conditions present across the entire period of measured data (2003 to 2007). EPA interprets the maximum future design value to be a potential future air quality outcome consistent with the meteorology that yielded maximum measured concentrations in the ambient data set analyzed for that receptor. In other words, the average design value gives a reasonable projection of future air quality at the receptor under “average” conditions. However, EPA also recognizes that previously experienced meteorological conditions (e.g., dominant wind direction, temperatures, air mass patterns) promoting ozone or fine particle formation that led to maximum concentrations in the measured data may reoccur in the future. The maximum design value gives a reasonable projection of future air quality at the receptor under a scenario in which such conditions do, in fact, reoccur. It also identifies upwind emissions that under those circumstances could interfere with the downwind area's ability to maintain the NAAQS.

Per the court's opinion in North Carolina, it is necessary for the Agency to evaluate “interference with maintenance” separately from “significant contribution to nonattainment” in order to give independent meaning to that phrase in the statute. The approach described above does so and provides a reasonable basis for identifying upwind emissions that interfere with maintenance of the NAAQS at downwind receptors.

Because the methodology is based on actual variations in design values measured at the receptors, EPA believes that the application of this design value methodology for identifying maintenance receptors reasonably anticipates possible future air quality outcomes based on meteorological conditions independent of emission reduction requirements occurring between 2005 (the base year for air quality analysis) and 2012 (the future year for air quality analysis of the base case without CAIR or the Transport Rule in place). EPA uses air quality modeling to properly account for changes in air quality from 2005 to 2012 due to emission control requirements and trends in emission source fleet turnover (such as increasingly cleaner motor vehicle fleets). The air quality modeling process allows EPA to effectively adjust measured data to project design values in 2012 based on the forecast changes in emissions. For a given receptor, the forecast change in emissions from 2005 to 2012 is a constant factor applied across all of the design values from the period 2003 to 2007. Thus, a comparison of the projected (future-year) design values themselves is equivalent to comparing the base period design values from the data set to consider how pollution concentrations are affected by non-modeled factors such as environmental and meteorological variability independent of the forecast emission reductions that stem from successful imposition of emission limitations and controls on various sources between the base and future modeling years. EPA believes it is reasonable to anticipate that these year-to-year meteorological fluctuations may reoccur at any time in the future and are relevant to determining receptors that are at risk of having a problem in the future with maintenance of the NAAQS. Therefore, EPA assesses the relationship of the maximum projected design value for 2012 at each receptor to the relevant NAAQS, and where such a value exceeds the NAAQS, EPA determines that receptor to be a “maintenance” receptor for purposes of defining interference with maintenance under the Transport Rule.

To provide an illustrative example, consider a hypothetical receptor “Y” whose measured data for 2003-2007 yields three design values for annual fine particles: 17 for 2003-05; 14 for 2004-06; and 12 µg/m³ for 2005-07. Thus, the maximum measured design value for this period is 17 and the average design value is 14.3. To determine whether the receptor is a nonattainment or maintenance receptor, EPA projects a corresponding future-year (2012) design value for each measured design value. These projections are based on the results of air quality modeling, which demonstrates predicted changes in pollution concentrations for each receptor from 2005 to 2012. For this example, assume that the projected future-year design values that correspond with the measured design values, are 16 (corresponds with the 2003-05 design value of 17), 13 (corresponds with the 2004-06 design value of 14), and 11 µg/m³ (corresponds with the 2005-07 design value of 12). The average future-year design value is 13.3 (corresponds with the average measured design value from 2003-2007 of 14.3). The projected future design values are all lower than the measured design values because air quality is projected to improve between 2005 and 2012. In this example, the analysis establishes that the average projected future design value is 13.3 and the maximum projected future design value is 16.

The average future (2012) projected design value of 13.3 based on the average design value for the period 2003-07 does not exceed the 1997 annual PM_2.5 NAAQS. For this reason, EPA would conclude that receptor Y will most likely have attainment air quality in the future year. Therefore, it would not be identified as a nonattainment receptor.

However, the future projected design value of 16 based on the maximum design value for the period 2003-07 does exceed the NAAQS. For this reason, EPA would conclude that the receptor may have difficulty maintaining attainment with the NAAQS under future potential meteorological conditions. EPA therefore would identify the receptor as a maintenance receptor and evaluate whether upwind state emissions interfere with maintenance of the NAAQS at that receptor.

EPA's methodology accounts for the range of meteorological conditions reflected by design values from the measured 2003-2007 data at receptor Y and also accounts for the projected changes in emissions from 2005 to 2012 at receptor Y. The range of meteorological conditions is accounted for by using data from three different 3-year periods as described above. The projected changes in emissions are accounted for by applying to the measured design values the forecasted change in PM_2.5 concentrations, as determined through air quality modeling of the 2005 and 2012 emissions. In this example, the maximum measured design value for receptor Y is 17. This design value represents measured data from 2003 to 2005. EPA applies to this design value the modeled 2005-to-2012 change in concentrations at receptor Y to obtain a 2012 maximum design value for that receptor, which is 16. In this way, this maximum 2012 design value takes into consideration the air quality impacts of all known and legally applicable emission limitations taking effect after the 2003 to 2005 base period. Therefore, each of the projected future-year design values provide a fair representation of future air quality at receptor Y under different conditions while accounting for the emissions projected to remain in 2012. EPA thus believes that if one of these future-year design values for a particular receptor exceeds the NAAQS, it is reasonable to conclude that the area may have difficulty maintaining that NAAQS. For this reason, EPA identifies such receptors as maintenance receptors. In this example, EPA would find that while receptor Y's average future-year design value would not exceed the NAAQS, its maximum future-year design value (16) would exceed the NAAQS, and it would thus be designated as a “maintenance” receptor for purposes of the Transport Rule analyses.

In the proposed rule we used air quality modeling to (1) Identify locations where we expected there to be nonattainment and/or maintenance problems for annual average PM_2.5, 24-hour PM_2.5, and/or 8-hour ozone in 2012, (2) quantify the impacts (i.e., air quality contributions) of SO₂ and NO_X emissions from upwind states on downwind annual average and 24-hour PM_2.5 concentrations at monitoring sites projected to be nonattainment or have maintenance problems in 2012 for the 1997 annual and 2006 24-hour PM_2.5 NAAQS, respectively, and (3) quantify the impacts of NO_X emissions from upwind states on downwind 8-hour ozone concentrations at monitoring sites projected to be nonattainment or have maintenance problems in 2012 for the 1997 ozone NAAQS.

To support the proposal, air quality modeling was performed for four emission scenarios: a 2005 base year, a 2012 “no CAIR” base case, a 2014 “no CAIR” base case, and a 2014 control case that reflects the emission reductions expected from the FIPs. The modeling for 2005 was used as the base year for projecting air quality for each of the 3 future-year scenarios. The 2012 base case modeling was used to identify future nonattainment and maintenance locations and to quantify the contributions of emissions in upwind states to annual average and 24-hour PM_2.5 and 8-hour ozone. The 2012 ozone and PM_2.5 concentrations were derived by projecting 2003 through 2007 based ambient ozone and/or PM_2.5 data to the future using the relative (percent) change in modeled concentrations between 2005 and 2012. The 2014 base case and 2014 control case modeling were used to quantify the benefits of this proposal.

In the proposed rule, EPA used the Comprehensive Air Quality Model with Extensions (CAMx) version 5.20 to simulate ozone and PM_2.5 concentrations for the 2005 base year and the 2012 and 2014 future year scenarios. The CAMx model applications were designed to cover states in the central and eastern U.S. using a horizontal resolution of 12 x 12 km.

CAMx contains “source apportionment” tools that are designed to quantify the contribution of emissions from various sources and areas to ozone and PM_2.5 component species in other downwind locations. The source apportionment tools were used to quantify the downwind contributions of ozone and PM_2.5 from upwind states.

In the proposed rule, EPA used a 2005-based air quality modeling platform which included 2005 base year emissions and 2005 meteorology for modeling ozone and PM_2.5 with CAMx.

We received comments related to several aspects of the air quality modeling platform.

Comment: There was wide support from commenters for the use of CAMx as an appropriate, state-of-the science air quality tool for use in the Transport Rule. There were no comments that suggested that EPA should use an alternative model for quantifying interstate transport. Many commenters requested that EPA update the emission inventories used for the Transport Rule and then remodel the 2005 base year and future year emissions using the updated emissions and the most recent version of CAMx to reassess interstate transport for the final rule.

Response: For the final rule we have updated our modeling using the latest public release of CAMx (version 5.30) and associated preprocessors. We have also made numerous improvements to the emission inventories for the 2005 base year as well as the 2012 and 2014 future year base cases in response to public comments. The emissions changes are described in section V.C.1. The projection of future year nonattainment and maintenance sites and the quantification of ozone and PM_2.5 transport for the final rule are based on modeling with CAMx v5.30 using the updated emission inventories. The final rule air quality projections of 2012 nonattainment and maintenance are described below. The final rule interstate contributions are presented in section V.D.

Comment: The performance evaluation of the 2005 base year model predictions for the proposed rule was too cursory and did not provide sufficient detail on model performance. Commenters requested additional analyses and spatial resolution describing how well base year model predictions compare to the corresponding measured values.

Response: For the final rule we have expanded the scope of the model evaluation for 2005 to include a broader suite of statistics to characterize performance for individual subregions of the eastern U.S. modeling domain. The results of the performance evaluation for the final rule 2005 base year air quality modeling are described in the Air Quality Modeling Final Rule TSD.

Comment: The 2005 based modeling platform should be updated to a more recent year. There were several different aspects of this comment. Some commenters stated that EPA should be using a more recent emission inventory as a base year, due to identified changes and updates to the inventories. Other commenters stated that EPA should use a more recent base year, due to a trend of improvement in air quality over the past few years. The commenters claim that the 2005-based EPA modeling does not account for large emission reductions and air quality improvements that have occurred over the last several years.

Response: There are several reasons why the use of a 2005 modeling base case is both reasonable and, in fact, necessary for the Transport Rule. As explained in section V.B, above, because the Transport Rule will replace CAIR, EPA cannot consider reductions associated with CAIR in the analytical baseline emissions scenario. Thus, the base year for the air quality projections should be a year that represents emissions before CAIR was in place (i.e. 2005). We are projecting emissions to a future 2012 “no CAIR” case and therefore want to best represent the air quality change between 2005 and 2012, without CAIR. To do this, we projected emissions that existed before CAIR was in effect and modeled the air quality change that occurs between 2005 and 2012 without CAIR.

A key consideration in our projection methodology is the use of ambient data to anchor the design value projections to the future. The modeling is used in a relative sense by multiplying the modeled percent change in ozone or PM_2.5 species concentrations by the base year ambient data. The ozone and PM_2.5 modeling guidance recommends projecting design values based on 5 years of monitoring data that is centered on the base model year. Using 2005 as a base emissions and meteorological year entailed the use of 2003-2007 ambient air quality data (5 years of data centered about 2005). This was a reasonable choice because the majority of the ambient data from this period was not impacted by CAIR emission reductions.

After 2005, early emission reductions of SO₂ and NO_X in response to CAIR began to impact the measured air quality concentrations. Since the modeling projection methodology uses both modeled and observed data, 2005 is the latest base year that we deemed appropriate (before CAIR emission reductions took place) for use in projecting the measured air quality to a 2012 future year. The early years of the 5 year period (2003, 2004, and 2005) were not impacted by CAIR. The last 2 years in the period (2006 and 2007) were slightly impacted by CAIR emission reductions. But the 5 year average is weighted towards the middle year of the period (2005), so the impact of the years after CAIR promulgation should be minimal.

The 2005 base year was also chosen because it was an appropriate meteorological year. In the eastern U.S. there was relatively high ozone during the summer of 2005 and relatively high PM_2.5 periods during the year. The modeled attainment tests for both ozone and 24-hour PM_2.5 depend on having a sufficient number of “high” modeled days to project to the future. Modeling a year that is not meteorologically conducive to ozone and/or PM_2.5 formation is discouraged by the modeling guidance because a meteorological year that is not conducive to ozone or PM_2.5 formation may be less responsive to changes in emissions in the future. Therefore, projecting the relative change in ozone or PM_2.5 for a non-conducive base year may underestimate the future change in ozone and/or PM_2.5 concentrations.

Additionally, all enforceable emission reductions that occurred between 2005 and 2012 (other than those required under CAIR) are captured by the modeling system. Any enforceable non-EGU emission reductions due to existing rules or the installation of emissions controls after 2005 were included in the 2012 base case inventory. As explained above in section V.B, to capture changes in EGU emissions between 2005 and 2012, EPA did not assume operation of all controls installed during that time period, as many of those controls were built in response to CAIR. EPA used IPM to project 2012 EGU emissions incorporating all non-CAIR enforceable emission constraints; operation of existing pollution controls was taken into account only where non-CAIR constraints made it economic or legally necessary to operate them. We also accounted for permanent source shutdowns that occurred after 2005. Where possible, we incorporated reported emission changes based on comments to the proposed rule and a subsequent emission inventory NODA.

Comment: Several commenters stated that we used a “modeled + monitored” test in CAIR to identify future year nonattainment receptors, but we only used a modeled test in the Transport Rule proposal. They suggest that we should either go back to the “modeled + monitored” test or explain why we should not use monitoring data in the identification of nonattainment and maintenance receptors. They say that we should not base nonattainment and maintenance receptors solely on modeled violations. They also say that we if we had looked at the most recent ambient data we would see that most of the modeled nonattainment and maintenance receptors are already attaining the ozone and/or PM_2.5 NAAQS.

Response: In the identification of future year nonattainment receptors for CAIR, EPA used what was called the “modeled + monitored test”. The most recent ambient data (2001-2003 design values at the time) were examined to further verify that nonattainment was still being measured at potential future year nonattainment receptors. In the proposed Transport Rule, EPA identified future year nonattainment and maintenance receptors based on modeled projections of ambient data from the 2003-2007 time period. The future year receptors were not compared to most recent ambient data to verify that nonattainment still existed.

For the final Transport Rule, there are several reasons that EPA did not examine the most recent ambient data to verify that receptors were still measuring nonattainment. The main reason for dropping the “monitored” part of the modeled + monitored test is the fact that the most recent monitoring data (2007-2009 design values) include large emission reductions from CAIR. As explained in section V.B, above, because the Transport Rule will replace CAIR, we must model a future year base case which does not assume that CAIR is in place (a “no-CAIR” case). It is simply not appropriate to examine the current monitoring data, which represent air quality with CAIR emission reductions in place, and compare the values to 2012 projected air quality that is based on a no-CAIR modeling case. As discussed above, we modeled a 2005 base case with pre-CAIR emissions and a 2012 future “no CAIR” case. The change in modeled air quality is due to the non-CAIR enforceable emission changes between 2005 and 2012 and therefore explicitly does not take CAIR into account. As a consequence, the 2012 projected design values represent a unique case (necessary for analyzing future air quality without either CAIR or its replacement Transport Rule in effect) that cannot be represented by current ambient data.

It is also important to note that all of the projected 2012 design values are based on projections of measured ambient data. They are a combination of measured data and modeled response factors. Therefore, it is inaccurate to imply that future year nonattainment and maintenance receptors are solely based on modeled projections. The future year concentrations are firmly rooted in base year measured ambient data that have been projected to the future using modeled data.

There are additional reasons for not verifying the nonattainment and maintenance receptors against the most recent ambient data. In CAIR we did not explicitly identify maintenance receptors. In the Transport Rule proposal we identified maintenance receptors based on 2012 projections of maximum design values from the 2003-2007 period. Even though receptors may be measuring attainment based on recent data, they may still be at risk for falling back into nonattainment. Therefore, even if commenters argue that recent data show that monitoring sites should not be nonattainment receptors (with which we disagree), the same argument cannot be made regarding maintenance receptors. Clearly, receptors with recent “clean” ambient data may still experience higher PM_2.5 and/or ozone concentrations in the future (based on meteorological and emission variability) and therefore may be appropriate maintenance receptors.

Comment: Several commenters claim that the maintenance receptor methodology overstates actual future design values. They also recommend an alternative methodology which takes into account the downward trend in observed PM_2.5 concentrations over the last 5+ years. The methodology would remove the trend in the data where air quality is improving over the period by applying a linear fit to the data, calculating the residuals and then adding the residuals back to the average of the data. Given a site with a downward trend, this has the effect of decreasing the calculated maximum values from the early years in the period and increasing the values from the end years in the period.

Response: EPA continues to believe that our approach to identify maintenance receptors is reasonable and appropriate. For the final rule, we continue to identify maintenance receptors by projecting the maximum design value from the 2003-2007 period to the future. The methodology assumes that the combination of emissions and meteorology that occurred in the base period (which led to relatively high ambient design values) could happen again in the future (albeit at lower emissions levels). There is no information presented by the commenters which explains why the magnitude of base year design value variability could not occur in the same way in the future. The commenters cite the downward trend in ambient data as the reason why the EPA methodology is not reasonable. However, in most cases, the recent downward trend in ambient data is due to a combination of ongoing emission reductions (which includes CAIR), variability in meteorology, and depressed emissions due to the recession. In fact, the most recent ambient design value period (2007-2009) is heavily influenced by extremely low ozone and PM_2.5 concentrations measured in 2009. The 2009 data are marked by relatively low emissions due to cool summer weather and ongoing effects of the recession. The preliminary 2010 ambient data in the eastern U.S. show that ozone and PM_2.5 values were considerably higher in 2010 compared to 2009. In the states that are included in the final Transport Rule region, there were 158 ozone monitor days that exceeded 84 ppb in 2009 compared to 412 monitor exceedance days in 2010. For PM_2.5, there were 251 monitor days that exceeded 35 μg/m ³ in 2009 compared to 417 monitor exceedance days in 2010. Even though the SO₂ and NO_X emissions were generally lower in 2010, the observed ozone and PM_2.5 concentrations were higher. This shows the important influence of meteorology on ambient concentrations. Clearly, the year to year variability due to meteorology can be large. We acknowledge the downward trend in ambient data over the last few years. But this does not mean that conditions that led to high ozone and/or PM_2.5 in the 2003-2007 period could not occur again in the future. The 2010 ambient data show that meteorology can cause concentrations to go back up, even though there is a downward trend in emissions.

We also believe that the alternate maintenance methodology presented by the commenter is inappropriate. The EPA modeling for 2012 (and 2014) appropriately accounts for emission reductions that occur after 2005 except for those that should not be considered, as explained in section V.B., because they were required only by CAIR. Therefore, the starting point design values used to project to the future should not be lowered to account for emission reduction trends that occur after 2005. Doing so would give “double credit” to the more recent emission reductions and provides an inappropriate downward adjustment to the early design value periods of the 2003-2007 period.

Comment: One commenter claims that EPA did not follow our own modeling guidance by not doing local scale modeling in urban areas with high PM_2.5 concentration gradients. They suggested that the methodology to calculate future year design values should have included dispersion modeling to calculate the change in concentration over time of primary PM_2.5 emissions.

Response: EPA modeling guidance for PM_2.5 attainment demonstrations recommends photochemical grid modeling to examine future year changes in PM_2.5 concentrations. There are several optional aspects of the modeling which are recommended in specific cases. This includes a recommendation for a “local area analysis” using a dispersion model. An area with relatively large local primary PM_2.5 concentration gradients may want to do additional modeling to examine the impacts of local controls on its future year PM_2.5 concentrations. This is particularly important when local controls of primary PM_2.5 are included as part of the attainment demonstration.

As noted above, a “local area analysis” is recommended as part of the local attainment demonstration process in specific situations. It is impractical for EPA to perform this type of analysis for each local area in the regional Transport Rule. National rulemakings are not attainment demonstrations. We are not able to perform fine scale analyses for each area. For the final rule modeling, we have attempted to address all emissions and modeling related comments. We have updated the modeling platform to use the latest version of CAMx and are continuing to model ozone and PM_2.5 at 12km grid resolution, which for PM_2.5 is a more refined grid resolution compared to the CAIR modeling.

Additionally, there is no evidence presented by the commenter that would indicate that the future year PM_2.5 concentrations from the Transport Rule are biased high. In fact, depending on the circumstances, local fine scale grid or dispersion modeling may result in lower or higher future year design values. In a fine scale analysis, the dominant local primary PM_2.5 emissions become a larger percentage of the PM_2.5 concentrations. Therefore, if the local emissions are forecast to decrease, fine scale modeling may lead to lower future design values. However, if the local emissions are forecast to increase or stay the same between the base and future years, local modeling will likely show higher future year design values compared to a regional analysis. This points to the fact that perceived biases in modeling results may not always be correct.

In sum, fine scale modeling of local areas may lead to either higher or lower future year design values. There is no indication that EPA's regional modeling is biased in either direction. EPA's Transport Rule modeling generally followed EPA's modeling guidance and is appropriate for the purpose of this rulemaking.

Comment: One commenter completed and submitted a detailed CAMx based modeling analysis with a 2008 base year and future years of 2014 and 2018. The analysis shows that the majority of the proposed rule 2012 nonattainment and maintenance sites are already attaining based on either 2006-2008 or 2007-2009 ambient data. Based on this, the commenter claims that air quality has improved more rapidly than predicted by EPA's proposed rule modeling. Also, based on the commenter's 2014 modeling of CAIR emissions (including utility consent decrees and state programs), the commenter concludes that no additional controls are needed beyond CAIR to bring most or all sites into attainment by 2014.

Response: As an initial matter, we note that the basic question addressed by the commenter, “whether additional controls beyond CAIR are necessary,” is not on point. As explained previously, the D.C. Circuit remanded CAIR to EPA and it remains in place only temporarily. The question EPA must answer in this rulemaking, therefore, is not what controls in addition to CAIR are necessary but what, if any, restrictions on emissions must be put in place to replace CAIR in order to satisfy the requirements of section 110(a)(2)(D)(i)(I) of the CAA. For this reason, and as explained in greater detail in section V.B of this preamble, any analysis of whether beyond CAIR controls are necessary is irrelevant to this rulemaking. Nonetheless, we have carefully reviewed different aspects of the commenter's analysis. We previously addressed comments related to the use of more recent ambient data to examine future year nonattainment and maintenance receptors. As noted above, the 2006-2008 and 2007-2009 ambient data is heavily influenced by several factors. Among them are the emissions reductions from CAIR, the relatively low recent observed ozone and PM_2.5 concentrations at least partially due to non-conducive meteorology (particularly in 2009), and the atypical suppression of emissions due to the sharp recession. For all of these reasons, we believe it is not possible to directly compare the most recent design values to the predicted future year 2012 and 2014 design values from the Transport Rule. In particular, it is inappropriate to compare current design values to EPA's no-CAIR 2012 future year modeling results. As noted in the comment summary, the commenter's modeling analysis assumed that CAIR was in place in both 2008 and the future years. This is a fundamentally different assumption than the modeling EPA used to define the Transport Rule nonattainment and maintenance receptors in 2012 and is inappropriate for purposes of the Transport Rule for reasons described above and in section V.B.

Additionally, EPA's maintenance methodology chooses the highest of three base year design value periods projected to the future. The commenter only used a single design value period in their analysis and therefore did not fully examine maintenance issues. In fact, the 2014 nonattainment modeling receptors in the final Transport Rule and the commenter's modeling analysis are similar. As documented in section VI.D, in the 2014 final rule remedy case, there is only one remaining nonattainment area for ozone and one remaining nonattainment area for 24-hour PM_2.5. This is similar to the modeling results presented in the comments. However, EPA modeling identifies additional maintenance receptors in 2012 that continue to have maintenance issues in 2014.

EPA also examined our ozone and PM_2.5 projection procedures to see if there might be additional reasons for the relatively lower current ambient design values (and modeled design values in the commenter's analysis) compared to the 2014 remedy modeled values. Upon further analysis of EPA's 24-hour attainment test methodology, we noted certain discrepancies between the methodology and the calculation of the ambient 24-hour design values. In the proposed rule 24-hour attainment test, for each PM_2.5 monitor, we projected the measured 98th percentile concentrations from the 2003-2007 period to the future. A basic assumption in this methodology is that the distribution of high measured days in the base period will be the same in the future. For example, if the observed 98th percentile day is the 3rd high day for a particular year, we assume that the 1st, 2nd, and 3rd high days (and subsequent high days) in the future remain in the same basic distribution. Further examination of the proposed rule modeling found that this is not always the case. In situations where there are large summer PM_2.5 concentration reductions, some of the high days may switch from the summer in the base period to the winter in the future period.

In order to better account for the complicated future response in 24-hour design values, we have updated the 24-hour attainment demonstration methodology to more closely reflect the way 24-hour design values are calculated. In the revised methodology, we do not assume that the temporal distribution of high days in the base and future periods will remain the same. We project a larger set of ambient days from the base period to the future and then re-rank the entire set of days to find the new future 98th percentile value (for each year). More specifically, we project the highest 8 days per quarter (32 days per year) to the future and then re-rank the 32 days to derive the future year 98th percentile concentrations. In the case of the Transport Rule model results, this has the effect of lowering the future year 24-hour design values compared to the old methodology. The 2012 base case design values for all nonattainment and maintenance receptors were either unchanged or lower with the revised methodology.

3. How did EPA project future nonattainment and maintenance for annual PM_2.5, 24-hour PM_2.5, and 8-hour ozone?

Final Rule: In general, the methodology to project ozone and PM_2.5 concentrations to the future year(s) remains the same for the final rule. The proposal modeling followed the modeling guidance procedures for projecting ambient design values to future years. For the final rule, we continue to follow the basic procedures outlined in the guidance. The 8-hour ozone and annual PM_2.5 methodology are unchanged from the proposal. However, the 24-hour PM_2.5 methodology has been updated in the final rule to be more consistent with the calculation of 24-hour PM_2.5 design values. There were also additional minor updates to the ambient data. The methodology to identify maintenance receptors is also unchanged from the proposal. We continue to use the maximum design value (projected from the 5 year base period) to calculate future year maintenance receptors.

As noted in the proposal, EPA considers that the maintenance concept has two components: Year-to-year variability in emissions and air quality, and continued maintenance of the air quality standard over time. The way that EPA defined maintenance based on year-to-year variability (as discussed in detail here) directly affects the requirements of this final rule. EPA also considered whether further reductions were necessary to ensure continued lack of interference with maintenance of the NAAQS over time (e.g., after 2014). EPA concluded that in light of projected emission trends, and also considering the emission reductions from this proposed rule, no further reductions are required solely for this purpose at PM_2.5 and ozone receptors for which we are partially or fully determining significant contribution for the current NAAQS. (See discussion of emission trends in Chapter 7 of TSD entitled “Emission Inventories,” included in the docket for the Transport Rule proposal.)

a. Which ambient ozone and PM_2.5 data did EPA use for the purpose of projecting future year concentrations?

The final rule modeling continues to use a 2005 base case inventory and 2005 meteorology. Therefore, we continue to use ambient data from the 2003-2007 period. For each monitoring site, all valid design values (up to 3) from this period were averaged together. Since 2005 is included in all three design value periods, this has the effect of creating a 5-year weighted average, where the middle year is weighted 3 times, the 2nd and 4th years are weighted twice, and the 1st and 5th years are weighted once. We refer to this as the 5-year weighted average value. The 5-year weighted average values were then projected to the future years that were analyzed for this final rule. The 2003-2005, 2004-2006, and 2005-2007 design values are accessible at http://www.epa.gov/airtrends/values.html. The design values have been updated based on the latest official values. The official values have exceptional events removed from the calculations if they are flagged by states and concurred with by EPA Regional offices.

The procedures for projecting annual average PM_2.5 and 8-hour ozone conform to the methodology in the current attainment demonstration modeling guidance.

b. Projection of Future Annual and 24-Hour PM_2.5 Nonattainment and Maintenance

(1) Methodology for Projecting Future Annual PM_2.5 Nonattainment and Maintenance

For the final rule, annual PM_2.5 modeling was performed for the 2005 base year emissions and for the 2012 base case as part of the approach for projecting which locations are expected to be in nonattainment and/or have difficulty maintaining the PM_2.5 standards in 2012. We refer to these areas as nonattainment sites and maintenance sites respectively.

Concentrations of PM_2.5 in 2012 were estimated by applying the modeled 2005-to-2012 relative change in PM_2.5 species to each of the 3-year ambient monitoring data periods (i.e., 2003-2005, 2004-2006, and 2005-2007) to obtain up to 3 future-year PM_2.5 design values for each monitoring site. We used the highest of these projections at each monitoring site to determine which sites are expected to have maintenance problems in 2012. We used the 5 year weighted average of those projections to determine which monitoring sites are expected to be nonattainment in this future year.

For the analysis of both nonattainment and maintenance, monitoring sites were included in the analysis if they had at least one complete design value in the 2003-2007 period. There were 721 monitoring sites in the 12 km modeling domain which had at least one complete design value period for the annual PM_2.5 NAAQS, and 722 sites which met this criterion for the 24-hour NAAQS.

EPA followed the procedures recommended in the modeling guidance for projecting PM_2.5 by projecting individual PM_2.5 component species and then summing these to calculate the concentration of total PM_2.5. EPA's Modeled Attainment Test Software (MATS) was used to calculate the future year design values. The software (including documentation) is available at: http://www.epa.gov/scram001/modelingapps_mats.htm. Additional details on the annual PM_2.5 nonattainment and maintenance projections methodology can be found in the Air Quality Modeling Final Rule TSD.

The 2012 annual PM_2.5 design values were calculated for each of the 721 sites. The calculated annual PM_2.5 design values are truncated after the second decimal place. This is consistent with the ambient monitoring data truncation and rounding procedures for the annual PM_2.5 NAAQS. Any value that is greater than or equal to 15.05 µg/m³ is rounded to 15.1 µg/m³ and is considered to be violating the NAAQS. Thus, sites with projected 5-year weighted average (“average”) annual PM_2.5 design values of 15.05 µg/m³ or greater are predicted to be nonattainment sites. Sites with projected maximum design values of 15.05 µg/m³ or greater are predicted to be maintenance sites. Note that nonattainment sites are also maintenance sites because the maximum design value is always greater than or equal to the 5-year weighted average. For ease of reference we use the term “nonattainment sites” to refer to those sites that are projected to exceed the NAAQS based on both the average and maximum design values. Those sites that are projected to be attainment based on the average design value, but exceed the NAAQS based on the maximum design value, are referred to as maintenance sites. The monitoring sites that we project to be nonattainment and/or maintenance for the annual PM_2.5 NAAQS in the 2012 base case are the nonattainment/maintenance receptors used for assessing the contribution of emissions in upwind states to downwind nonattainment and maintenance of the annual PM_2.5 NAAQS.

Table V.C-1 contains the 2003-2007 base case period average and maximum annual PM_2.5 design values and the corresponding 2012 base case average and maximum design values for sites projected to be nonattainment of the annual PM_2.5 NAAQS in 2012. Table V.C-2 contains this same information for projected 2012 maintenance sites.

(2) Methodology for Projecting Future 24-Hour PM_2.5 Nonattainment and Maintenance

The procedures for calculating the future year 24-hour PM_2.5 design values have been updated for the final rule. The revised procedures are in response to comments which noted relatively high future year 24-hour PM_2.5 design values in EPA's modeling of the proposed Transport Rule. The updates are intended to make the projection methodology more consistent with the procedures for calculating ambient design values.

As noted above, for the proposed Transport Rule EPA projected for each PM_2.5 monitor the measured 98th percentile concentrations from the 2003-2007 period to the future. As an additional check, we also projected the next highest concentrations from the three calendar quarters in each year when the 98th percentile did not occur in the 2003-2007 base period, to ensure that the future year 98th percentile did not switch seasons in the future year compared to the base year. A basic assumption in this methodology is that the distribution of high measured days in the base period will be the same in the future.

In other words, EPA assumed at proposal that the 98th-percentile day could only be displaced “from below” in the instance that a different day's future concentration exceeded the original 98th-percentile day's future concentration. In that case, the original 98th-percentile day may become the 97th- or 96th-percentile day in the future year; EPA accounted for this possibility at proposal. EPA did not, however, consider that the 98th-percentile day could also be displaced “from above” in the instance that higher-concentration days in the base period were projected to have future concentrations lower than the original 98th-percentile day's future concentration. In that case, the original 98th-percentile day may become the 99th- or 100th-percentile day. Because EPA continued to use that day's future concentration to determine the monitor's future design value at proposal, this sometimes resulted in overstatement of future-year design values for 24-hour PM_2.5 monitoring sites whose seasonal distribution of highest-concentration 24-hour PM_2.5 days changed between the 2003-2007 period and the future year modeling. Examination of the proposed rule remedy modeling (2014 remedy case) showed that many of the highest PM_2.5 days switched from the summer in the base period to the winter in the future period. This is especially true in areas of the upper Midwest which experience both high summer and winter PM_2.5 episodes.

In the revised methodology, we do not assume that the seasonal distribution of high days in the base period years and future years will remain the same. We project a larger set of ambient days from the base period to the future and then re-rank the entire set of days to find the new future 98th percentile value (for each year). More specifically, we project the highest 8 days per quarter (32 days per year) to the future and then re-rank the 32 days to derive the future year 98th percentile concentrations. In the case of the Transport Rule model results, this has the effect of lowering the future year 24-hour design values compared to the old methodology.

The modeling guidance recommendations for state attainment demonstrations have been updated to reflect the changes outlined above. Further details on the 24-hour PM_2.5 design value calculations can be found in the Air Quality Modeling Final Rule TSD. The above procedures for determining future year 24-hour PM_2.5 concentrations were applied for each site. The 24-hour PM_2.5 design values are truncated after the first decimal place. This approach is consistent with the ambient data truncation and rounding procedures for the 24-hour PM_2.5 NAAQS. Any value that is greater than or equal to 35.5 µg/m³ is rounded to 36 μg/m³ and is violating the NAAQS. Sites with future year 5-year weighted average design values of 35.5 μg/m³ or greater, based on the projection of 5-year weighted average concentrations, are predicted to be nonattainment. Sites with future year maximum design values of 35.5 µg/m³ or greater are predicted to be maintenance sites. Note that nonattainment sites for the 24-hour NAAQS are also maintenance sites because the maximum design value is always greater than or equal to the 5-year weighted average. The monitoring sites that we project to be nonattainment and/or maintenance for the 24-hour PM_2.5 NAAQS in the 2012 base case are the nonattainment/maintenance receptors used for assessing the contribution of emissions in upwind states to downwind nonattainment and maintenance of 24-hour PM_2.5 NAAQS as part of this final rule.

Table V.C-3 contains the 2003-2007 base period average and maximum 24-hour PM_2.5 design values and the 2012 base case average and maximum design values for sites projected to be 2012 nonattainment of the 24-hour PM_2.5 NAAQS in 2012. Table V.C-4 contains this same information for projected 2012 24-hour maintenance sites.

(3) Methodology for Projecting Future 8-Hour Ozone Nonattainment and Maintenance

The final rule methodology to calculate 8-hour ozone nonattainment and maintenance receptors is identical to the proposed rule. The May-to-September 24-hour maximum 8-hour average concentrations from the 2005 base case and the 2012 base case were used to project ambient design values to 2012. The following is a brief summary of the future year 8-hour average ozone calculations. Additional details are provided in the Air Quality Modeling Final Rule TSD.

We are using the base period 2003-2007 ambient ozone design value data for projecting future year design values. Relative response factors (RRF) for each monitoring site were calculated as the percent change in ozone on days with modeled ozone greater than 85 ppb.

The maximum future design value is calculated by projecting design values for each of the three base periods (2003-2005, 2004-2006, and 2005-2007) separately. The highest of the three future values is the maximum design value. This maximum value is used to identify the 8-hour ozone maintenance receptors.

The future year design values are truncated to integers in units of ppb. This approach is consistent with the ambient data truncation and rounding procedures for the 8-hour ozone NAAQS. Future year design values that are greater than or equal to 85 ppb are considered to be violating the NAAQS. Sites with future year 5-year weighted average design values of 85 ppb or greater are predicted to be nonattainment. Sites with future year maximum design values of 85 ppb or greater are predicted to be future year maintenance sites. Note that, as described previously for the annual and 24-hour PM_2.5 NAAQS, nonattainment sites for the ozone NAAQS are also maintenance sites because the maximum design value is always greater than or equal to the 5-year weighted average. The monitoring sites that we project to be nonattainment and/or maintenance for the 8-hour ozone NAAQS in the 2012 base case are the nonattainment/maintenance receptors used for assessing the contribution of emissions in upwind states to downwind nonattainment and maintenance of ozone NAAQS.

Table V.C-5 contains the 2003-2007 base period average and maximum 8-hour ozone design values and the 2012 base case average and maximum design values for sites projected to be 2012 nonattainment of the 8-hour ozone NAAQS in 2012. Table V.C-6 contains this same information for projected 2012 8-hour ozone maintenance sites.

D. Pollution Transport From Upwind States

1. Choice of Air Quality Thresholds

a. Thresholds

In this action, EPA uses air quality thresholds to identify linkages between upwind states and downwind nonattainment and maintenance receptors. States whose contributions to a specific receptor meet or exceed the thresholds identified are considered linked to that receptor; those states' emissions (and available emission reductions) are analyzed further in the second step of EPA's significant contribution analysis. States whose contributions are below the thresholds are not included in the Transport Rule for that NAAQS. In other words, we are finding that states whose contributions are below these thresholds do not significantly contribute to nonattainment or interfere with maintenance of the relevant NAAQS.

We use separate air quality thresholds for annual PM_2.5, 24-hour PM_2.5, and 8-hour ozone. Each air quality threshold is calculated as 1 percent of the NAAQS. Specifically, we use an air quality threshold of 0.15 μg/m³ for annual PM_2.5, 0.35 μg/m³ for 24-hour PM_2.5, and 0.8 ppb for 8-hour ozone. These are the same air quality thresholds we proposed.

EPA received a number of comments on the thresholds we proposed, and those comments and EPA's responses are discussed below.

b. General Comments on the Overall Stringency and Use of 1 Percent of the NAAQS

EPA received numerous comments supporting and opposing the proposed thresholds. A number of commenters cited support for EPA's approach. Some commenters believed that use of a 1 percent threshold was too stringent, and recommended that EPA should use a threshold greater than 1 percent. Others believed that 1 percent was not stringent enough, and they recommended using a lower value such as 0.5 percent. EPA believes that for both PM_2.5 and for ozone, it is appropriate to use a threshold of 1 percent of the NAAQS for identifying states whose contributions do not significantly contribute to nonattainment or interfere with maintenance of the relevant NAAQS; therefore, EPA has retained the 1 percent threshold for the reasons described below.

As we found at the time of CAIR, EPA's analysis of base case PM_2.5 transport shows that, in general, PM_2.5 nonattainment problems result from the combined impact of relatively small contributions from many upwind states, along with contributions from in-state sources and, in some cases, substantially larger contributions from a subset of particular upwind states. (See section II of the January 2004 CAIR proposal, 69 FR 4575-87).

In the 1998 NO_X SIP Call (63 FR 57456, October 27, 1998) and in CAIR, EPA also found important contributions from multiple upwind states. As a result of the upwind “collective contributions,” EPA determined that it is appropriate to use a low air quality threshold when analyzing upwind states' contributions to downwind states' attainment and maintenance problems for ozone as well as PM_2.5.

Low threshold values are also warranted, as EPA discussed in the notices for CAIR, due to adverse health impacts associated with ambient PM_2.5 and ozone even at low concentrations (See relevant portions of the CAIR proposal notice (63 FR 4583-84) and the CAIR final rule notice (70 FR 25189-25192)).

To aid in responding to comments, EPA has compiled the contribution modeling results to analyze the impact of different possible thresholds. This analysis demonstrates the reasonableness of using the 1 percent threshold to account for the combined impact of relatively small contributions from many upwind states (see Air Quality Modeling Final Rule TSD). In this analysis, EPA identifies for annual PM_2.5 (sulfate and nitrate), 24-hour PM_2.5 (sulfate and nitrate), and 8-hour ozone receptors: (1) Total upwind state contributions, and (2) the amount of the total upwind state contribution that is captured at thresholds of 1 percent, 5 percent and 0.5 percent of the NAAQS. EPA continues to find that the total “collective contribution” from upwind sources represents a large portion of PM_2.5 and ozone at downwind locations and that the total amount of transport is composed of the individual contribution from numerous upwind states.

The analysis shows that the 1 percent threshold captures a high percentage of the total pollution transport affecting downwind states for both PM_2.5 and ozone. In response to commenters who advocated a higher threshold, EPA observes that higher thresholds would exclude increasingly large percentages of total transport, which we do not believe would be appropriate. For example, a 5 percent threshold would exclude the majority—and for annual PM, more than 80 percent—of interstate pollution transport affecting the downwind state receptors analyzed (based on the average percentage of total interstate transport across all receptors captured at the 5 percent threshold).

In response to commenters who advocated a lower threshold, EPA observes that the analysis shows that a lower threshold such as 0.5 percent would result in relatively modest increases in the overall percentages of PM_2.5 and ozone pollution transport captured relative to the amounts captured at the 1 percent level. A 0.5 percent threshold could lead to emission reduction responsibilities in additional states that individually have a very small impact on those receptors—an indicator that emission controls in those states are likely to have a smaller air quality impact at the downwind receptor. We are not convinced that selecting a threshold below 1 percent is necessary or desirable. A strong indication that the amount of pollution transport being excluded from consideration is not excessive is that the controls required under this rule are projected to eliminate nonattainment and maintenance problems with air quality standards at most downwind state receptors.

Considering the combined downwind impact of multiple upwind states, the health effects of low levels of PM_2.5 and ozone pollution, and EPA's previous use of a 1 percent threshold for PM_2.5 in CAIR, EPA's judgment is that the 1 percent threshold is a reasonable choice.

Some commenters noted that the PM_2.5 thresholds used for this rule are less than the “significant impact levels” (SILs) used for permitting programs. As EPA stated at the time of CAIR, since the thresholds referred to by the commenters serve different purposes than the CAIR threshold for significant contribution, it does not follow that they should be made equivalent (70 FR 25191; May 12, 2005).

c. Comments on the Rounding Conventions for PM_2.5

In the final Transport Rule, EPA is using two-digit values for the PM_2.5 thresholds. Some commenters suggested that EPA should use the same rounding convention for annual PM_2.5 used in CAIR; that is, the threshold should be 0.2 μg/m³ rather than 0.15 μg/m³. The reasons for EPA's decision are below.

The rationale for the single digit value for the final CAIR rule was that a single digit is consistent with the EPA monitoring data reporting requirements in Part 50, Appendix N, section 4.3. These reporting requirements specify that design values for the annual PM_2.5 standard shall be rounded to the tenths place (decimals 0.05 and greater are rounded up to the next 0.1, and any decimal lower than 0.05 is rounded down to the nearest 0.1).

Because the design value is to be reported only to the nearest 0.1 μg/m³, EPA deemed it preferable for the final CAIR to select the threshold value at the nearest 0.1 μg/m³ as well, and hence one percent of the 15 μg/m³, rounded to the nearest 0.1 μg/m³ became 0.2 μg/m³.

The reporting requirements in section Part 50, Appendix N, section 4.3 for the 24-hour PM_2.5 standard state that design values for this standard shall be rounded to the nearest 1 μg/m³ (decimals 0.5 and greater are rounded up to the nearest whole number, and any decimal lower than 0.5 is rounded down to the nearest whole number).

If the approach used in CAIR were to be used to establish an air quality threshold for the 24-hour PM_2.5 NAAQS (which CAIR did not address), the resulting threshold would be zero. One percent of the 24-hour standard is 0.35 μg/m³, and rounding to the nearest whole number would yield an air quality threshold of zero. Thus if we were to apply the same rationale used to develop the annual PM_2.5 threshold for the final CAIR, there would be no air quality threshold for 24-hour PM_2.5, which EPA believes to be counter-intuitive and unworkable as an approach for assessing interstate contributions.

Therefore, for this rule, EPA proposed and is now finalizing an approach that decouples the precision of the air quality thresholds from the monitoring reporting requirements, and uses 2-digit values representing one percent of the PM_2.5 NAAQS; that is, 0.15 μg/m³ for the annual standard, and 0.35 μg/m³ for the 24-hour standard. EPA believes there are a number of considerations favoring this approach. First, it provides for a consistent approach for the annual and 24-hour standards. Second, the approach is readily applicable to any current and future NAAQS and would automatically adjust the stringency of the transport threshold to maintain a constant relationship with the stringency of the relevant NAAQS as they are revised. The CAIR approach would not allow for this continuity: For example, if EPA were to retain the CAIR approach for the annual standard, any future lowering of the PM_2.5 NAAQS to below 15 μg/m³ would reduce the air quality threshold to the same outcome: 0.1 μg/m³. This would occur because any value less than 0.15 μg/m³ would round to 0.1 μg/m³ (assuming EPA would not round down to zero for the reasons described above), which means that the air quality threshold would have a different relative stringency to each possible future NAAQS value. For the above reasons, EPA believes the use of two-digit thresholds for both annual PM_2.5 and 24-hour PM_2.5 in the final rule is both reasonable and appropriate. The departure from the approach used for annual PM_2.5 in CAIR is appropriate given the additional considerations that were not in existence at the time of the final CAIR, and the importance of using a consistent approach to developing air quality thresholds for all NAAQS addressed by this rule as well as future NAAQS considered in future transport-related actions.

Some of these commenters suggested using the CAIR rounding conventions coupled with use of a 1-digit threshold of 0.4 μg/m³ for 24-hour PM_2.5. EPA considered the approach suggested by commenters, but determined that the proposed approach is more appropriate. First, adhering to the rounding conventions used for CAIR for annual PM_2.5 is not workable for the 24-hour standard because the rounding convention would yield a threshold of zero. Rounding alternatively to 0.4 μg/m³ would require EPA to find a basis for rounding the threshold to the nearest 0.1 μg/m³ instead of using a strict application of 1 percent; we do not see any basis for such rounding at this time.

d. Comments Related to the Multi-Factor Test EPA Used for Ozone in CAIR

Some commenters suggested that, for ozone, EPA should use the multiple-metric test we used for CAIR, and not a simple threshold based on 1 percent of the NAAQS. With respect to ozone, EPA proposed in the Transport Rule to take a more straightforward approach to air quality thresholds than the multi-factor approaches used for the NO_X SIP Call and the CAIR. As proposed, EPA is using a contribution metric that is calculated based on the multi-day average contribution. This metric is compared to one percent of the 1997 8-hour ozone standard of 0.08 ppm. Under this approach, one percent of the NAAQS is a value of 0.8 ppb. Contributions of 0.8 ppb and higher are above the threshold; ozone contributions less than 0.8 ppb are below the threshold. In past rulemakings (e.g., CAIR) EPA used multiple ozone metrics, including the average contribution and maximum single day contribution to downwind nonattainment. EPA believes the average contribution (calculated over multiple high ozone days) is a robust metric compared to the maximum contribution on a single day. EPA believes that this approach is preferable because it uses a robust metric, it is consistent with the approach for PM_2.5, and it provides for a consistent approach that takes into account, and is applicable to, any future ozone standards below 0.08 ppm.

One of these commenters suggested that the 0.8 ppb threshold value was substantially more stringent than the 2 ppb screening test which was a part of the approach used for CAIR. The 1 percent threshold (0.8 ppb) is not substantially more stringent than the previous 2 ppb test because of differences in the metrics used to evaluate contributions against these two levels. The 2 ppb test was evaluated using the highest single day absolute model-predicted downwind contribution from an upwind state. The 1 percent threshold is evaluated based on the average relative downwind impact calculated over multiple days. Therefore, it is appropriate to set a lower concentration threshold for use with the average contribution metric calculated for the Transport Rule. More details on the calculation of the contribution metric can be found in the Air Quality Modeling Final Rule TSD. As noted above, EPA believes that the approach used for the proposed rule provides for a simplified, yet robust approach compared to CAIR. Accordingly, for the final rule we have retained the approach used for the proposal.

One commenter suggested that EPA retain the CAIR multiple-factor approach for ozone, and to apply that same approach to 24-hour PM_2.5. As noted above, EPA is not retaining this approach for ozone, and for similar reasons we believe a multi-factor approach is not needed for 24-hour PM_2.5. The approach based on 1 percent of the NAAQS is consistent with the form of the 24-hour standard. In addition, this approach is based on contributions on days with high 24-hour PM_2.5 predictions and therefore is relevant for characterizing transport during short-term high PM_2.5 episodic conditions.

e. Comments on the Relationship to Measurement Precision

Other commenters suggested that, as did commenters on the thresholds used in CAIR, EPA should take into consideration the measurement precision of existing PM_2.5 monitors in setting the thresholds for the Transport Rule. EPA disagrees that monitoring precision is relevant to determining the amount of modeled PM_2.5 or ozone that should be considered to be a “contribution” from upwind states since states are not required to, nor would it be possible for them to, measure their individual state impacts on downwind receptors. The approach for eliminating significant contribution is based on the implementation of enforceable emissions budgets and not on a measurement of ambient air quality. Thus, EPA believes it is a reasonable exercise of its discretion to de-couple monitoring precision from the choice of contribution states.

f. Comments Related to the CAIR Court Decision

Commenters recommended that EPA should have retained the criteria used for CAIR because those values were upheld by the Court. As noted above, EPA could not have used the approach for annual PM_2.5 that was used in CAIR to develop a 24-hour PM_2.5 threshold, as that approach would have yielded a threshold value of zero 24-hour PM_2.5.

Further, nothing in the North Carolina opinion suggests that the thresholds and methods used in CAIR were the only possible approaches EPA could have used, that they were preferable to other approaches, or that other alternatives would not be acceptable. Instead, the Court upheld the 0.2 µg/m³ threshold used for PM_2.5 on the grounds that it was not “wholly unsupported by the record” (North Carolina, 531 F.3d at 915). EPA has determined for reasons explained in the record that the thresholds used in this final rule are both reasonable and appropriate for use in this final rule.

2. Approach for Identifying Contributing Upwind States

This section documents the procedures used by EPA to quantify the contribution of emissions in specific upwind states to air quality concentrations in projected 2012 downwind nonattainment and maintenance locations for annual PM_2.5, 24-hour PM_2.5, and 8-hour ozone. In the proposed rule EPA used CAMx photochemical source apportionment modeling to quantify the impact of emissions in specific upwind states on projected downwind nonattainment and maintenance receptors for both PM_2.5 and 8-hour ozone. In this modeling we tracked the ozone and PM_2.5 formed from 2012 base case emissions from anthropogenic sources in each upwind state in the 12 km modeling domain. The CAMx Particulate Source Apportionment Technique (PSAT) was used to calculate downwind contributions to nonattainment and maintenance of PM_2.5. In the PSAT simulation NO_X emissions are tracked to particulate nitrate concentrations, SO₂ emissions are tracked to particulate sulfate concentrations, and primary particulates (organic carbon, elemental carbon, and other PM_2.5) are tracked as primary particulates. As described earlier in section V.A, the nitrate and sulfate contributions were combined and used to evaluate interstate contributions of PM_2.5.

The CAMx Ozone Source Apportionment Technique (OSAT) was used to calculate downwind 8-hour ozone contributions to nonattainment and maintenance. OSAT tracks the formation of ozone from NO_X and VOC emissions.

Comment: Three commenters stated that the CAMx source apportionment techniques used for the proposed rule reflect state-of-the science technologies and are appropriate for evaluating interstate transport. One commenter asked that EPA do more to demonstrate that the PSAT and OSAT techniques give reliable answers, although no suggestions were provided on how this might be done. Another commenter said that the results of the contribution analyses were consistent with the results of their scientific research.

Response: EPA is not changing its conclusion that the CAMx source apportionment techniques are appropriate for quantifying interstate transport. The strength of the source apportionment technique is that all modeled ozone and/or PM_2.5 mass at a given location in the modeling domain is tracked back to specific sources of emissions and boundary conditions to fully characterize culpable sources. No commenters provided technically valid analyses indicating that EPA's use of CAMx source apportionment techniques are inappropriate for the purposes of the Transport Rule.

Comment: We received comments that certain states included in the proposed rule should be excluded from the final rule because EPA had overstated the 2012 emissions in these states. Commenter requested that we redo the contribution modeling using 2012 base case emission inventories that are revised based on proposed rule comments. Several commenters also asked that EPA update the contribution modeling analyses using the latest version of CAMx.

Response: In response to these comments, we have rerun our source apportionment modeling for PM_2.5 and ozone for the 2012 base case using the updated emission inventories described above in section V.C.1 and the latest version of CAMx, version 5.30.

The states EPA analyzed for interstate contributions for ozone and for PM_2.5 for the final rule are: Alabama, Arkansas, Connecticut, Delaware, Florida, Georgia, Illinois, Indiana, Iowa, Kansas, Kentucky, Louisiana, Maine, Maryland, Massachusetts, Michigan, Minnesota, Mississippi, Missouri, Nebraska, New Hampshire, New Jersey, New York, North Carolina, North Dakota, Ohio, Oklahoma, Pennsylvania, Rhode Island, South Carolina, South Dakota, Tennessee, Texas, Vermont, Virginia, West Virginia, and Wisconsin. These are the same states that EPA analyzed for the proposed rule.

For the proposed rule, we used a relative approach for calculating the contributions to downwind nonattainment and maintenance receptors from the outputs of the source apportionment modeling. As part of this approach, the source apportionment predictions are combined with measurement-based concentrations to calculate the contributions from each state to nonattainment and/or maintenance receptors. This is similar to the approach used to calculate future year design values, as described in section V.C.2.

Comment: One commenter said that using the source apportionment modeling predictions in a relative sense strengthens the determination of contributions and addresses an important source of uncertainty. There were no comments that suggested an alternative approach.

Response: For the final Transport Rule we are applying the relative approach developed for the proposed rule to calculate contributions from each state to downwind nonattainment and maintenance receptors.

As noted above, for the final rule we modeled the updated 2012 base case emissions using CAM_X v5.30 to determine the contributions from emissions in upwind states to nonattainment and maintenance sites in downwind states. Contributions to nonattainment and maintenance receptors are evaluated independently for each state to determine if the contributions are at or above the threshold criteria.

For each upwind state, the maximum contribution to nonattainment is calculated based on the single largest contribution to a future year (2012) downwind nonattainment receptor. The maximum contribution to maintenance is calculated based on the single largest contribution to a future year (2012) downwind maintenance receptor. Since the contributions are calculated independently for each receptor, the upwind contribution to maintenance can sometimes be larger than the contribution to nonattainment, and vice versa. This also means that maximum contributions to nonattainment can be below the threshold while maximum contributions to maintenance may be at or above the threshold, or vice versa.

V.D.2.a. Estimated Interstate Contributions to Annual PM_2.5 and 24-Hour PM_2.5

In this section, we present the interstate contributions from emissions in upwind states to downwind nonattainment and maintenance sites for the annual PM_2.5 NAAQS and the 24-hour PM_2.5 NAAQS based on modeling updated for the final rule. As described previously in section V.D.1, states which contribute 0.15 μg/m³ or more to annual PM_2.5 nonattainment or maintenance in another state are identified as states with contributions large enough to warrant further analysis. For 24-hour PM_2.5, states which contribute 0.35 μg/m ³ or more to 24-hour PM_2.5 nonattainment or maintenance in another state are identified as states with contributions to downwind nonattainment and maintenance sites large enough to warrant further analysis.

For annual PM_2.5, we calculated each state's contribution to each of the 12 monitoring sites that are projected to be nonattainment and each of the 4 sites that are projected to have maintenance problems for the annual PM_2.5 NAAQS in the 2012 base case. A detailed description of the calculations can be found in the Air Quality Modeling Final Rule TSD. The largest contribution from each state to annual PM_2.5 nonattainment in downwind sites is provided in Table V.D-1. The Largest Contribution from Each State to Annual PM_2.5 maintenance in downwind sites is also provided in Table V.D-1. The contributions from each state to all projected 2012 nonattainment and maintenance sites for the annual PM_2.5 NAAQS are provided in the Air Quality Modeling Final Rule TSD.

Based on the state-by-state contribution analysis, there are 18 states which contribute 0.15 μg/m³ or more to downwind annual PM_2.5 nonattainment. These states are: Alabama, Georgia, Illinois, Indiana, Iowa, Kentucky, Maryland, Michigan, Missouri, New York, North Carolina, Ohio, Pennsylvania, South Carolina, Tennessee, Texas, West Virginia, and Wisconsin. In Table V.D-2, we provide a list of the downwind nonattainment sites to which each upwind state contributes 0.15 μg/m³ or more (i.e., the upwind state to downwind nonattainment “linkages”).

There are 12 states which contribute 0.15 μg/m³ or more to downwind annual PM_2.5 maintenance. These states are: Alabama, Illinois, Indiana, Kentucky, Michigan, Missouri, New York, Ohio, Pennsylvania, Tennessee, West Virginia, and Wisconsin. In Table V.D-3, we provide a list of the downwind maintenance sites to which each upwind state contributes 0.15 μg/m³ or more (i.e., the upwind state to downwind maintenance “linkages”).

For 24-hour PM_2.5, we calculated each state's contribution to each of the 20 monitoring sites that are projected to be nonattainment and each of the 21 sites that are projected to have maintenance problems for the 24-hour PM_2.5 NAAQS in the 2012 base case. A detailed description of the calculations can be found in the Air Quality Modeling Final Rule TSD. The largest contribution from each state to 24-hour PM_2.5 nonattainment in downwind sites is provided in Table V.D-4. The largest contribution from each state to 24-hour PM_2.5 maintenance in downwind sites is also provided in Table V.D-4. The contributions from each state to all projected 2012 nonattainment and maintenance sites for the 24-hour PM_2.5 NAAQS are provided in the Air Quality Modeling Final Rule TSD.

Based on the state-by-state contribution analysis, there are 21 states which contribute 0.35 μg/m³ or more to downwind 24-hour PM_2.5 nonattainment. These states are: Alabama, Georgia, Illinois, Indiana, Iowa, Kansas, Kentucky, Maryland, Michigan, Minnesota, Missouri, New Jersey, New York, North Carolina, Ohio, Pennsylvania, Tennessee, Texas, Virginia, West Virginia, and Wisconsin. In Table V.D-5, we provide a list of the downwind nonattainment counties to which each upwind state contributes 0.35 μg/m³ or more (i.e., the upwind state to downwind nonattainment “linkages”).

There are 21 states which contribute 0.35 μg/m³ or more to downwind 24-hour PM_2.5 maintenance. These states are: Alabama, Georgia, Illinois, Indiana, Iowa, Kansas, Kentucky, Maryland, Michigan, Minnesota, Missouri, Nebraska, New Jersey, New York, North Carolina, Ohio, Pennsylvania, Tennessee, Virginia, West Virginia, and Wisconsin. In Table V.D-6, we provide a list of the downwind maintenance sites to which each upwind state contributes 0.35 μg/m³ or more (i.e., the upwind state to downwind maintenance “linkages”).

b. Estimated Interstate Contributions to 8-Hour Ozone

In this section, we present the interstate contributions from emissions in upwind states to downwind nonattainment and maintenance sites for the ozone NAAQS. As described previously in section V.D.1, states which contribute 0.8 ppb or more to 8-hour ozone nonattainment or maintenance in another state are identified as states with contributions to downwind attainment and maintenance sites large enough to warrant further analysis.

We calculated each state's contribution to ozone at each of the 4 monitoring sites that are projected to be nonattainment and each of 6 sites that are projected to have maintenance problems for the 8-hour ozone NAAQS in the 2012 base case. A detailed description of the calculations can be found in the Air Quality Modeling Final Rule TSD. The largest contribution from each state to 8-hour ozone nonattainment in downwind sites is provided in Table V.D-7. The largest contribution from each state to 8-hour ozone maintenance in downwind sites is also provided in Table V.D.2-7. The contributions from each state to all projected 2012 nonattainment and maintenance sites for the 8-hour ozone NAAQS are provided in the Air Quality Modeling Final Rule TSD.

Based on the state-by-state contribution analysis, there are 11 states that contribute 0.8 ppb or more to downwind 8-hour ozone nonattainment. These states are: Alabama, Arkansas, Georgia, Illinois, Indiana, Kentucky, Louisiana, Mississippi, Missouri, Tennessee, and Texas. In Table V.D-8, we provide a list of the downwind nonattainment counties to which each upwind state contributes 0.8 ppb or more (i.e., the upwind state to downwind nonattainment “linkages”).

There are 26 states which contribute 0.8 ppb or more to downwind 8-hour ozone maintenance. These states are: Alabama, Arkansas, Florida, Georgia, Illinois, Indiana, Iowa, Kansas, Kentucky, Louisiana, Maryland, Michigan, Mississippi, Missouri, New Jersey, New York, North Carolina, Ohio, Oklahoma, Pennsylvania, South Carolina, Tennessee, Texas, Virginia, West Virginia, and Wisconsin. In Table V.D.2-9, we provide a list of the downwind nonattainment counties to which each upwind state contributes 0.8 ppb or more (i.e., the upwind state to downwind nonattainment “linkages”).

VI. Quantification of State Emission Reductions Required

A. Cost and Air Quality Structure for Defining Reductions

1. Summary

Section V, above, describes EPA's approach to identifying upwind states with air quality contributions that meet or exceed the air quality thresholds discussed therein for each of the NAAQS addressed in this rule. A state is covered by the Transport Rule if its contributions meet or exceed one of those air quality thresholds and the Agency identifies, using the cost- and air quality-based approach described below, emissions within the state that constitute the state's significant contribution to nonattainment and interference with maintenance with respect to the 1997 ozone, 1997 PM_2.5 or 2006 PM_2.5 NAAQS.

In this section, EPA explains its final cost- and air quality-based approach to quantify the amount of emissions that represent significant contribution to nonattainment and interference with maintenance for each state. EPA then applies that approach for the three different NAAQS being addressed in this rule: The 1997 ozone NAAQS, the 1997 annual PM_2.5 NAAQS and the 2006 24-hour PM_2.5 NAAQS. EPA believes that the methodology finalized could also be used to address transport concerns under other NAAQS, including future revisions to the ozone and PM_2.5 NAAQS.

EPA applies the methodology described herein to fully quantify the emissions that constitute each covered state's significant contribution to nonattainment and interference with maintenance with respect to the 1997 annual PM_2.5 and the 2006 24-hour PM_2.5 NAAQS. The FIPs with respect to the annual and 24-hour PM_2.5 NAAQS that are finalized in this action ensure that all such emissions are prohibited. Each such FIP thus fully satisfies the requirements of 110(a)(2)(D)(i)(I) with respect to the annual and/or 24-hour PM_2.5 NAAQS for the covered state.

EPA also applies the methodology to quantify significant contribution to nonattainment and interference with maintenance with respect to the 1997 ozone NAAQS. However, we have not been able to fully quantify such emissions for all covered states. In this action, EPA fully quantifies the significant contribution to nonattainment and interference with maintenance for 15 states. We finalize FIPs with respect to the 1997 ozone standards for 10 of these 15 states (Florida, Maryland, New Jersey, New York, North Carolina, Ohio, Pennsylvania, South Carolina, Virginia, and West Virginia). We are also publishing a supplemental notice of rulemaking to take comment on whether FIPs should be finalized for the remaining 5 states (Iowa, Kansas, Michigan, Oklahoma, and Wisconsin). The FIPs for these 10 states (and the FIPs for the remaining 5 states, if finalized) fully satisfy the requirements of 110(a)(2)(D)(i)(I) with respect to the 1997 ozone NAAQS for the covered state.

In addition, we apply the methodology described herein to quantify, for 11 additional states, ozone-season NO_X emission reductions that are necessary but may not be sufficient to eliminate all significant contribution to nonattainment and interference with maintenance in other states. We finalize FIPs with respect to the 1997 ozone standards for 10 of these 11 states (Alabama, Arkansas, Georgia, Illinois, Indiana, Kentucky, Louisiana, Mississippi, Tennessee, and Texas). We are also publishing a supplemental notice of rulemaking to take comment on whether FIPs should be finalized for the remaining state (Missouri). The FIPs for these 10 states (and the FIP for the remaining state, if finalized) make measurable progress toward satisfying the requirements of 110(a)(2)(D)(i)(I) with respect to the 1997 ozone NAAQS in each covered state. To the extent that significant contribution to nonattainment and interference with maintenance is not entirely eliminated for the 1997 ozone NAAQS through today's action, EPA will address these instances in a future rulemaking. This is further explained in section VI.D.

With respect to the 1997 annual PM_2.5 NAAQS, this rule finds that 18 states have SO₂ and NO_X emission reduction responsibilities. EPA also finds that 21 states have SO₂ and NO_X emission reduction responsibilities with respect to the 2006 24-hour PM_2.5 NAAQS. There are a total of 23 states that have SO₂ and NO_X emission reduction responsibilities for one or both of the above PM_2.5 NAAQS. We apply the methodology to quantify emission reductions that these states must achieve to eliminate the state's significant contribution to nonattainment and interference with maintenance. The states are listed in Table III-1 in section III of this preamble.

This rule will prohibit all significant contribution to nonattainment and interference with maintenance with respect to the annual and 24-hour PM_2.5. In addition, it will resolve air quality issues at most nonattainment and maintenance receptors identified by EPA. EPA projects that unresolved nonattainment and maintenance issues will remain in only a few downwind states after promulgation and implementation of the Transport Rule. For the annual PM_2.5 standard, EPA projects that this rule will help assure that all areas in the east fully resolve their nonattainment and maintenance concerns. This rule will also help a number of areas achieve the standard earlier than they may have otherwise. For the 2006 24-hour PM_2.5 NAAQS, one area is projected to remain in nonattainment (Liberty-Clairton) and three areas are projected to have remaining maintenance concerns after imposition of the Transport Rule (Chicago, Detroit, and Lancaster County).

The methodology provides similar assistance for ozone, assuring upwind reductions that will assist downwind states in controlling ozone pollution. It reduces ozone concentration levels in 2012 and helps assure that all but two downwind areas fully resolve their nonattainment and maintenance problems with the 1997 ozone NAAQS by 2014. While Houston is projected to still face nonattainment and Baton Rouge is projected to still face maintenance concerns with the 1997 ozone NAAQS, the Transport Rule improves air quality in these two areas and provides both health benefits and assistance for these local areas in meeting the NAAQS requirements. For reasons explained below, EPA will conduct further analysis in a subsequent transport-related rulemaking to determine whether further upwind state reductions are warranted to assist attainment and maintenance of the ozone NAAQS in Houston and Baton Rouge areas.

When EPA proposed this air-quality and cost-based multi-factor approach to identify emissions that constitute significant contribution to nonattainment and interference with maintenance from upwind states with respect to the 1997 ozone, annual PM_2.5, and 2006 24-hour PM_2.5 NAAQS, the Agency indicated that the approach was designed to be applicable to both current and potential future ozone and PM_2.5 NAAQS (75 FR 45214). EPA believes that the final Transport Rule demonstrates the value of this approach for addressing the role of interstate transport of air pollution in communities' ability to comply with current and future NAAQS. EPA believes that the Transport Rule's approach of using air-quality thresholds to determine upwind-to-downwind-state linkages and using the cost- and air quality-based multi-factor approach to quantify significant contribution to nonattainment and interference with maintenance (i.e., to determine the specific amount of emissions that each upwind state must reduce) could serve as a precedent for quantifying upwind state emission reduction responsibilities with respect to potential future NAAQS.

One commenter suggested that the rule could set a flawed precedent for future transport analyses and remedies, as it does not fully eliminate the prohibited emissions in every upwind state. EPA disagrees with this characterization of the Transport Rule. EPA notes that the partial determination of significant contribution to nonattainment and interference with maintenance for certain upwind states in the Transport Rule with respect to the ozone NAAQS is not a function of the multi-factor approach itself, but is instead a function of its limited application in this rulemaking to identify emission reductions from a single source category (EGUs). In fact, the Transport Rule's approach itself allowed EPA to determine for which upwind states we have identified all emissions that constitute significant contribution to nonattainment and interference with maintenance, and for which upwind states we have identified emissions that are necessary but may not be sufficient to eliminate the prohibited emissions. As EPA explained at proposal, developing the additional information needed to consider NO_X emissions from non-EGU source categories in order to fully quantify upwind state responsibility with respect to the 1997 ozone NAAQS would substantially delay promulgation of the Transport Rule. EPA explained that we do not believe that effort should delay the emission reductions and large health benefits this final rule will deliver (75 FR 45213). EPA further explained that we believe it is likely that the Agency can provide the greatest assistance to states in addressing transported pollution by issuing a separate (subsequent) rule to address additional reductions that may be necessary to fully eliminate upwind state responsibility with respect to the 1997 ozone NAAQS (75 FR 45288). Thus, EPA decided to promulgate the Transport Rule as quickly as possible. EPA anticipates that application of this air-quality and cost-based multi-factor approach to a broader set of source categories in a subsequent rulemaking will identify any remaining prohibited emissions in the upwind states for which the Transport Rule may not fully eliminate those emissions with respect to the 1997 ozone NAAQS.

2. Background

After using air quality analysis to identify upwind states that are “linked” to downwind air quality monitoring sites with nonattainment and maintenance problems through contribution of at least one percent of the relevant NAAQS, EPA quantifies the portion of each state's contribution that constitutes its “significant contribution” or “interference with maintenance.”

This section describes the methodology developed by EPA for this analysis and then explains how that methodology is applied to measure significant contribution to nonattainment and interference with maintenance with respect to the NAAQS of concern. For this portion of the analysis, EPA expands upon the methodology used in the NO_X SIP Call and CAIR but modifies it in important respects. In the NO_X SIP Call and CAIR, EPA's methodology defined significant contribution as those emissions that could be removed with the use of “highly cost effective” controls. In the Transport Rule, rather than relying solely on an analysis of what constitutes “highly cost effective” controls, EPA relies on an analysis that accounts for both cost and air quality improvement to identify the portion of a state's contribution that constitutes its significant contribution to nonattainment and interference with maintenance. Furthermore, in response to the Court's opinion in North Carolina, EPA has developed an approach which gives independent meaning to the “interfere with maintenance” prong of section 110(a)(2)(D)(i)(I).

The methodology takes into account both the D.C. Circuit Court's determination that EPA may consider cost when measuring significant contribution, Michigan, 213 F.3d at 679, and its rejection of the manner in which cost was used in the CAIR analysis, North Carolina, 531 F.3d at 917. It also recognizes that the Court accepted—but did not require—EPA's use of a single, uniform cost threshold to measure significant contribution. Michigan, 213 F.3d at 679.

As EPA discussed at length in the Transport Rule proposal, using both air quality and cost factors allows EPA to consider the full range of circumstances and state-specific factors that affect the relationship between upwind emissions and downwind nonattainment and maintenance problems (75 FR 45271). For example, considering cost takes into account the extent to which existing plants are already controlled as well as the potential for, and relative difficulty of, additional emission reductions. Therefore, EPA believes that it is appropriate to consider both cost and air quality metrics when quantifying each state's significant contribution.

This methodology is consistent with the statutory mandate in section 110(a)(2)(D)(i)(I) which requires upwind states to prohibit emissions that significantly contribute to nonattainment or interference with maintenance in another state. As discussed in more detail in the proposal, interpreting significant contribution to nonattainment and interference with maintenance inherently involves a decision on how much emissions control responsibility should be assigned to upwind states, and how much responsibility should be left to downwind states. EPA's methodology is intended to “assign a substantial but reasonable amount of responsibility to upwind states. * * *to control their emissions” (75 FR 45272). EPA believes that upwind states contributing to downwind state air quality degradation should bear substantial responsibility to control their emissions because of the plain language of the good neighbor provision, the health risks and control cost impacts that upwind emissions cause in the downwind state, and the cumulative impact in the downwind state of emissions from multiple upwind states, and the importance of achieving attainment in downwind states as expeditiously as practicable but no later than specific deadlines as required by the Act. EPA's approach does not shift the responsibility for achieving or maintaining the NAAQS to the upwind state. See 75 FR 45272.

The methodology defines each state's significant contribution to nonattainment and interference with maintenance as the emission reductions available at a particular cost threshold in a specific upwind state which effectively address nonattainment and maintenance of the relevant NAAQS in the linked downwind states of concern. Unlike the NO_X SIP Call and CAIR, where EPA's significant contribution analysis had a regional focus, the methodology used in the Transport Rule focuses on state-specific factors. The methodology uses a multi-step process to analyze costs and air quality impacts, identify appropriate cost thresholds, quantify reductions available from EGUs in each state at those thresholds, and consider the impact of variability in EGU operations. There are four steps to this methodology: (1) Identification of each state's emission reductions available at ascending costs per ton as appropriate; (2) assessment of those upwind emission reductions' downwind air quality impacts; (3) identification of upwind “cost thresholds” delivering effective emission reductions and downwind air quality improvement; and (4) enshrinement of the upwind emission reductions available at those cost thresholds in state budgets.

In step one, EPA identifies what emission reductions are available at various cost thresholds, quantifying emission reductions that would occur within each state at ascending costs per ton of emission reductions. In other words, EPA determined for specific cost per ton thresholds, the emission reductions that would be achieved in a state if all EGUs greater than 25 MW in that state used all emission controls and emission reduction measures available at that cost threshold. For purposes of this discussion, we refer to these as “cost curves.”

For this final rule, EPA used updated IPM modeling to conduct a similar cost curve analysis as conducted in the Transport Rule proposal (75 FR 45275). In the proposal, the cost curves only reflected escalating cost for one pollutant while the other pollutant cost was held constant at base case levels (i.e., $0/ton). However, EPA improved the costing analysis for the final rule by identifying upwind emission reductions available as costs were imposed on both SO₂ and NO_X simultaneously for states linked to downwind states on the basis of the PM_2.5 NAAQS. In other words, the cost curves in the proposal depicted state level emissions when only one pollutant was priced (i.e., NO_X at $500/ton). Separate cost curves were done for each pollutant. For the final rule, EPA conducted some preliminary cost curve analysis for identifying NO_X thresholds in this manner. However, for the final cost curve analysis, EPA relied on cost curves that reflected state emissions when pollutants were priced simultaneously (e.g., NO_X at $500/ton and SO₂ at $1,600/ton). For reasons described in section VI.B, EPA was able to conduct this type of analysis because the preliminary cost curves specific to annual and ozone-season NO_X suggested little flexibility in adjusting the $500/ton cost thresholds imposed for each. Therefore, EPA was able to hold the cost threshold constant at $500/ton for these pollutants in its examination of SO₂ at various cost thresholds. EPA believes this approach to cost analysis is a better simulation of the Transport Rule's likely impact on covered sources. Under the final Transport Rule, covered sources in states regulated for PM_2.5 must address compliance requirements for SO₂ and NO_X emissions simultaneously, and this refined approach to cost curve analysis and subsequent air quality analysis better reflects this reality. Section VI.B of this preamble describes the costing analysis in further detail. Also, for more detail on the development of the cost curves, see “Significant Contribution and State Emission Budgets Final Rule TSD” in the docket for this rule.

Although the cost curves presented in this rule only include EGU reductions, EPA also assessed the cost of SO₂ and NO_X emission reductions available for source categories other than EGUs in the proposed rulemaking. This preliminary assessment in the rule proposal suggested that there likely would be very large emission reductions available from EGUs before costs reach the point for which non-EGU sources have available reductions (75 FR 45272). EPA revisited these non-EGU reduction cost levels in this final rulemaking and verified that there are little or no reductions available from non-EGUs at costs lower than the thresholds that EPA has chosen ($500/ton for NO_X, $2,300/ton for SO₂).

Further details on EPA's application of cost curves are provided below, in section VI.B.

In step two, EPA uses an air quality assessment tool to estimate the impact that the combined reductions available from upwind contributing states and the downwind receptor state at different cost-per-ton levels would have on air quality at downwind monitoring sites projected to have nonattainment and/or maintenance problems. While less rigorous than the air quality models used for attainment demonstrations, EPA believes this air quality assessment tool (which has been refined since proposal) is acceptable for assessing the impact of numerous options for upwind emission reductions in the process of defining an upwind state's significant contribution to nonattainment and interference with maintenance. It allows the Agency to anticipate specific air quality impacts of many more potential emission reduction scenarios pertinent to the relevant NAAQS than time- and resource-intensive comprehensive air quality modeling would permit.

Further details on EPA's application of step two in this methodology are provided below, in section VI.C.

In step three, EPA examines cost and air quality information to identify “significant cost thresholds.” EPA considered a significant cost threshold to be a point along the cost curves where a noticeable change occurred in downwind air quality, such as a point where large upwind emission reductions become available because a certain type of emissions control strategy becomes cost-effective.

This methodology allows EPA, where appropriate, to define multiple cost thresholds that vary for a particular pollutant for different upwind states. As explained in the Transport Rule proposal, EPA does not believe it is required to utilize multiple cost thresholds to regulate upwind emissions for purposes of the mandate in CAA section 110(a)(2)(D), but EPA's multi-factor methodology developed for the Transport Rule to define significant contribution to nonattainment and interference with maintenance allows the Agency to consider whether a single cost threshold or multiple cost thresholds are appropriate for meeting the requirements of CAA section 110(a)(2)(D) relevant to a particular NAAQS (75 FR 45274).

In step four, EPA uses the information regarding emission reductions available in each “linked” upwind state at the appropriate cost threshold to form a state “budget,” representing the remaining emissions from covered sources for the state in an average year once significant contribution to nonattainment and interference with maintenance have been eliminated; each budget also allows for the identification of an associated variability limit. These budgets and variability limits are used to develop enforceable requirements under the final remedy. The final rule's methodology for identifying state budgets is derived directly from the cost curves and multi-factor analysis EPA uses to determine each state's significant contribution to nonattainment and interference with maintenance. State emission budgets are discussed in section VI.D and the variability limits are discussed in section V I.E.

B. Cost of Available Emission Reductions (Step 1)

This subsection provides more detail on the cost curves that EPA developed to assess the costs of reducing SO₂ and NO_X emissions to address transport related to ozone and PM_2.5 concentrations (described previously as Step 1). It summarizes the information from the curves and then provides EPA's interpretation of that information. EPA used IPM to develop the EGU cost curves described in this rulemaking. More information can be found regarding EPA's use of IPM for the final Transport Rule in the “Significant Contribution and State Emission Budgets Final Rule TSD”.

The amount of emission reductions that the cost curves suggest are available at various costs are specific to the 2012 and 2014 time periods. These cost estimates factor in the time interval between rule finalization and compliance periods, existing controls already in place, and controls that could potentially come on line by the start of the compliance period. EPA notes that cost curves are a fluid concept and would vary given different compliance dates.

1. Development of Annual NO_X and Ozone-Season NO_X Cost Curves

EPA conducted preliminary cost curve analysis for annual NO_X and ozone-season NO_X in a similar manner to that used in the proposed rulemaking. That is, the impact of various cost thresholds on emissions was examined individually. For example, state level emissions were examined at cost levels for annual NO_X of $500, $1,000, and $2,500/ton while SO₂ was held at base case levels. EPA used this approach to examine NO_X and ozone-season NO_X emission reductions available from EGUs by 2012 and 2014 at various cost levels, reaching to $2,500/ton for annual NO_X and up to $5,000/ton for ozone-season NO_X (in 2007-year dollars). Section VI.D explains why EPA analyzed the $500/ton threshold for annual and ozone-season NO_X. EPA selected two higher cost thresholds to analyze for annual and ozone-season NO_X that provided a reasonable spectrum of emission reduction opportunities from EGUs at higher cost thresholds. Specifically, EPA analyzed these two higher cost thresholds because the first ($1,000/ton) was informative in regards to the additional EGU NO_X emissions reductions available without installation of advanced controls, and the second ($2,500/ton for annual NO_X, $5,000/ton for ozone-season NO_X) was informative in regards to additional EGU reductions available at cost thresholds where advanced NO_X control retrofits are economic for some units. The cost thresholds were only applied to states with air quality contributions that meet or exceed the air quality thresholds as identified in section V.D. For both annual and ozone-season NO_X, EPA did not consider cost thresholds below $500/ton for reasons explained in section VI.D.

EPA observed in the proposal that low-cost NO_X reductions are available at upwind sources with existing pollution control equipment that may not otherwise be operated in the future without the Transport Rule. EPA believes it is appropriate to prohibit any “linked” upwind state from potentially increasing its emissions through a failure to operate these existing pollution controls, which could worsen downwind air quality problems. Thus, EPA reflected operation of these controls in all modeling of different cost thresholds (i.e., the modeling assumes year-round operation of post-combustion NO_X controls in covered PM_2.5 states and ozone-season operation of post-combustion NO_X controls in covered ozone states).

Table VI.B-1 shows the annual NO_X emissions from EGUs at various levels of control cost per ton for 2014. Table VI.B-2 presents the cost curves for ozone-season NO_X emissions from EGUs. As discussed in section VI.D, EPA determined that $500/ton for annual and ozone NO_X was the appropriate cost threshold for this rule (although EPA plans to determine in the future whether a higher cost/ton threshold may be warranted for states contributing to nonattainment or maintenance problems with the 1997 ozone air quality standard projected to remain in two downwind areas).

EPA notes that the cost curves presented here differ somewhat from the cost curves presented in the proposal. The NO_X emissions modeled at a $500/ton cost threshold for the final rule are lower than they were at proposal. In addition, the emission reductions they represent from the updated base case are not as pronounced as was found in modeling for the proposed rule. It is worth emphasizing that the lower emission reductions observed at $500/ton in this final rulemaking are due to a lower starting point in updated base case EGU NO_X emission levels (and thus do not reflect higher NO_X emissions remaining after the reductions made at the $500/ton threshold). While the base case 2012 nationwide annual EGU NO_X emissions were approximately 3 million tons in the proposal, they were only 2.1 million tons in the final rule. This approximately 33 percent reduction in base case EGU NO_X emissions in the final rule modeling relative to the proposal is due to a combination of modeling updates, including lower natural gas prices, reduced electricity demand, newly-modeled consent decrees and state rules, and updated NO_X rates to reflect 2009 emissions data. All of these factors resulted in substantially lower base case Transport Rule NO_X emissions in the final rule modeling.

2. Development of SO₂ Cost Curves

As explained in detail below in section VI.D, EPA determined that a single threshold of $500/ton for ozone-season NO_X control in the states covered for the 1997 ozone NAAQS and a single threshold of $500/ton for annual NO_X control in the states covered for the PM_2.5 NAAQS were appropriate cost thresholds for identifying upwind control under the Transport Rule. With these parameters determined, EPA was able to assess the availability of SO₂ emission reductions from EGUs at various SO₂ cost per ton thresholds with the corresponding NO_X reduction requirements simultaneously represented in the analysis.

This approach of simultaneously modeling cost levels for covered pollutants is different from the approach taken in the proposal. In the proposal, cost curves were developed and examined independently for each pollutant. For example, with the SO₂ cost curves in the proposal, the NO_X cost level was held constant at base case levels as the SO₂ cost threshold was varied from base case levels to $2,400/ton. Commenters noted that this did not accurately reflect a reality where source owners/operators view price signals for all covered pollutants simultaneously and make operation decisions accordingly. For the final rule, EPA included cost thresholds of $500/ton for annual NO_X in PM_2.5 states and $500/ton for ozone-season NO_X in ozone-season states while examining different SO₂ cost thresholds. This allows EPA to develop final cost curves for air quality analysis and budget determination that reflect EGU operation when faced with the appropriate cost thresholds on all covered pollutants. EPA believes this approach of modeling final cost curves is superior to the methodology used in the proposal because it reflects market signals for each pollutant simultaneously, as would be experienced by states and sources regulated under the Transport Rule.

In this manner, EPA examined several SO₂ cost thresholds of $500, $1,600, $2,300, $2,800, $3,300 and $10,000 per ton. EPA selected these cost thresholds for the final rule's analysis as a representative sampling of points along the SO₂ cost curve thoroughly explored at proposal. Modeling of these cost thresholds provided a spectrum of emission reduction opportunities yielding meaningful differences to consider in total costs and air quality improvements at each threshold. The proposal's more detailed analysis using smaller increments between cost thresholds outlined the general form of the sector's SO₂ emission reduction cost curve and therefore allowed EPA to use larger increments between cost thresholds for the final rule's analysis. Each of the cost thresholds examined for the final rule represents a point where there is a significant change in available controls, emission reductions, or costs and economic impacts. EPA believes analysis of these thresholds illustrate a meaningful progression of costs and air quality impacts that enabled the Agency to determine a proper threshold along this cost curve to identify significant contribution to nonattainment and interference with maintenance for this rulemaking.

The cost thresholds above $500/ton were applied starting in 2014. In all modeling, the 2012 cost per ton threshold was held constant at $500/ton as EPA believes that this cost threshold captures all emission reductions feasible by 2012 (see section VI.B.3 below for more discussion). At the higher cost levels (e.g., $2,800/ton and above), the curve does not include all available reductions as they do not include non-EGU reductions. As described above for NO_X, EPA also observed at proposal that substantial low-cost SO₂ reductions are available from the operation of existing scrubbers that may not otherwise operate in the future without the Transport Rule in place. Therefore, all of the final SO₂ cost curves assume operation of existing scrubbers in PM_2.5 states under the Transport Rule. In 2014, approximately 3 million tons of SO₂ reductions can be achieved at the $500/ton cost threshold through operation of existing controls and some fuel switching.

This final cost curve also appropriately reflects the Group 1/Group 2 distinction for states covered for PM_2.5. As discussed in more detail in section VI.D, EPA identified Group 2 states as those that were linked to states where all nonattainment and maintenance issues had been resolved at $500/ton levels. There is no longer any significant contribution to nonattainment or interference with maintenance by these seven Group 2 states at levels above $500/ton. Therefore, in the final curves, these Group 2 states' cost thresholds were held constant at $500/ton as the higher cost thresholds were applied to the remaining Group 1 states starting in 2014. For example, the modeled emissions at the $2,300 per ton cost threshold shown in Table VI.B-3 below reflect each state's emissions when Group 1 states are subjected to a $2,300 per ton SO₂ constraint and Group 2 states are subjected to a $500/ton SO₂ constraint.

Additional reductions can be achieved at the higher cost thresholds. The cost curves demonstrate that sources begin to build significant additional flue gas desulfurization (FGD) retrofits at an SO₂ cost threshold of $1,600 per ton and additional dry sorbent injection (DSI) retrofits at an SO₂ cost threshold of $2,300 per ton.

With these final cost curves in hand, EPA was able to identify the combined reductions available from upwind contributing states and the downwind state, at different cost-per-ton levels. Additionally, EPA was able to examine the economic impacts of imposing such cost constraints on power sector generation. However, this only constitutes a portion of EPA's multi-factor assessment used to determine the amount of emissions that represent significant contribution to nonattainment and interference with maintenance. As noted in the Transport Rule proposal, EPA's multi-factor assessment considered air quality and cost considerations when identifying cost thresholds (75 FR 45271). The air quality portion of the assessment is described in section VI.C of the final Transport Rule preamble.

3. Amount of Reductions That Could Be Achieved by 2012 and 2014

EPA applied escalating SO₂ cost per ton thresholds for Group 1 states to create the cost curves for 2014 and beyond. For 2012 SO₂, the cost per ton was held constant at $500/ton as the cost thresholds in 2014 and beyond were varied. The advanced pollution controls incentivized by these higher cost-per-ton levels can reasonably be installed by 2014. EPA also considered whether any of these emission reductions could be achieved prior to 2014. For the reasons that follow, EPA concluded that significant reductions could be achieved by 2012 and that it is important to require all such reductions by 2012 to ensure that they are achieved as expeditiously as practicable. SO₂ and NO_X reductions come from operating existing controls, installing combustion controls, fuel switching, and increased dispatch of lower-emitting generation which can be achieved by 2012. In general, compliance mechanisms that do not involve post-combustion control installation are feasible before 2014. For this reason, EPA believes it is appropriate to require these emissions to be removed in 2012, consistent with the Act's requirement that downwind states attain the NAAQS as expeditiously as practicable.

Therefore, all of the cost curves presented below include all feasible 2012 reductions up to a threshold of $500/ton for SO₂ and $500/ton for annual NO_X in states linked to receptors for PM_2.5, as well as $500/ton for ozone-season NO_X in states linked to receptors for ozone. These cost per ton levels do not precipitate advanced post-combustion control installation in 2012 (as EPA acknowledges that such installations are not feasible by 2012), but they do promote the compliance options outlined above. The higher cost thresholds for SO₂ Group 1 states were only applied starting in 2014. Therefore, the 2012 state level emissions in the “$2,300 per ton threshold” reflect a cost threshold of only $500/ton for all pollutants (the $2,300 per ton value starts in 2014 for Group 1 states' SO₂).

The table below illustrates the change in state level SO₂ emissions as the higher cost per ton thresholds are applied to Group 1 states.

C. Estimates of Air Quality Impacts (Step 2)

After developing cost curves to show the state-by-state cost-effective emission reductions available, EPA estimates the air quality impacts of these reductions using the air quality assessment tool coupled with full-scale air quality modeling where possible. EPA uses the air quality assessment tool to evaluate the impact on air quality for downwind nonattainment and maintenance receptors from upwind reductions in “linked” states. This section describes the development of the air quality assessment tool and summarizes the results of this evaluation.

1. Development of the Air Quality Assessment Tool and Air Quality Modeling Strategy

In response to comments on the methodology used for the proposed rule, EPA made significant improvements to the air quality assessment tool (AQAT) for the final Transport Rule. Furthermore, EPA relied on CAMx to model the air quality response to NO_X reductions and limited AQAT's role (relative to the Transport Rule proposal) to estimating the relative response of sulfate concentrations from SO₂ reductions. EPA did not use AQAT to address NO_X reductions in the final rule analyses. These and other changes to our approach, as described below and in the “Significant Contribution and State Emission Budgets Final Rule TSD”, address commenter's concerns about the scientific rigor of the design and application of AQAT and commenter's recommendations to rely upon air quality modeling as part of this analysis.

For the final Transport Rule, EPA created an AQAT calibration scenario consisting of full-scale air quality modeling using CAMx of a 2014 control scenario reflecting SO₂ and NO_X emission reductions of similar stringency and from the same geography as the Transport Rule proposal. Modeling of this AQAT calibration scenario reflected all updates made to the air quality modeling platform, as described in the “Air Quality Modeling Final Rule TSD” found in the docket for this rulemaking. CAMx modeling of each receptor's response in this control scenario accounts for complex chemical interactions and covariation of these pollutants. Among the important atmospheric chemical interactions accounted for in CAMx is “nitrate replacement.” Nitrate replacement occurs when SO₂ emission reductions lead to decreases in ammonium sulfate, which in turn, can result in an increase in ammonium nitrate concentrations. As described below, EPA used the CAMx modeling results for this AQAT calibration scenario together with the modeling for the 2012 base case to characterize the response of ozone, nitrate, and sulfate at each nonattainment and maintenance receptor to the mix of upwind NO_X and SO₂ emission reductions at each cost threshold.

As described in section VI.D, EPA determined that the $500/ton threshold for upwind annual and ozone-season NO_X control is appropriate for the final Transport Rule (although EPA plans to determine in the future whether a higher cost/ton threshold may be warranted for states contributing to nonattainment or maintenance problems with the 1997 ozone air quality standard projected to remain at receptors in two downwind areas ). Because this threshold corresponds to the NO_X control strategy modeled in the AQAT calibration scenario described above, EPA is able to rely on this CAMx air quality modeling to assess the response of ozone and nitrate concentrations due to NO_X reductions and does not estimate ozone or nitrate impacts for this final rulemaking using AQAT. Further information on the air quality modeling of this AQAT calibration scenario can be found in the Air Quality Modeling Final Rule TSD and the Significant Contribution and State Emission Budgets Final Rule TSD in the docket for this rulemaking.

In order to estimate 2014 annual and 24-hour PM_2.5 concentrations, AQAT uses the 2012 annual and seasonal contributions which quantify the contribution of SO₂ emissions in specific upwind states to sulfate concentrations at specific downwind receptors. These contributions are described in section V.D.2 and the Air Quality Modeling Final Rule TSD.

EPA utilizes CAMx modeling of the AQAT calibration scenario, described above, to “calibrate” the contribution factors by developing and applying linear sulfate response factors for each downwind receptor. These factors calibrate each receptor's sulfate response to varying levels of upwind SO₂ emissions. These calibration factors are based on the sulfate response modeled by CAMx due to emission changes occurring between the 2012 base case and the 2014 AQAT calibration scenario. Calibration factors were constructed for the annual and 24-hour PM_2.5 AQAT.

To further allow adequate assessment of the seasonal impacts of various levels of upwind SO₂ reductions on each receptor's 24-hour PM_2.5 concentration using AQAT, EPA developed response factors for sulfate on a quarterly basis to capture important air quality differences between summer and winter emissions and concentrations. This process allowed EPA to estimate the air quality values for each season at each cost threshold, and then estimate the air quality design values.

Finally, EPA's air quality assessment accounts for the impact that this differential response in sulfate by quarter can have on the ordering of 24-hour concentrations when calculating the 98th percentile for the 24-hour standard. AQAT estimates quarterly-specific relative response factors that estimate quarterly-specific proportional change in ammonium sulfate resulting from the SO₂ emission reduction from the 2012 base case scenario to the 2014 cost threshold scenario being assessed. These quarterly relative response factors are then applied to each of the maximum 24-hour PM_2.5 concentrations for eight days per quarter per year at each receptor from the 2012 base case. This methodology improvement allows EPA to redetermine the 98th percentile day for each year and recalculate average and maximum design values for the 24-hour PM_2.5 standard.

These improvements for the final rule increase EPA's confidence that the air quality estimates provided by AQAT, now customized for this application, more accurately estimate the results of full-scale air quality modeling of the various levels of upwind SO₂ reductions considered. EPA evaluated the estimates from AQAT using an independent data set, the 2014 base case estimates from CAMx, finding that the results are unbiased with minimal differences. See “Significant Contribution and State Emission Budgets Final Rule TSD” for more details.

As such, EPA believes the revised AQAT provides an appropriate basis for assessing the air quality portion of the multi-factor methodology to define significant contribution to nonattainment and interference with maintenance.

2. Utilization of AQAT To Evaluate Control Scenarios

For the final Transport Rule, EPA performed air quality analysis for each downwind annual and 24-hour PM_2.5 receptor with a nonattainment and/or maintenance problem in the 2012 base case. For each receptor, EPA quantified the sulfate reduction and resulting air quality improvement when a group of states consisting of the upwind states that are “linked” to the downwind receptor (as explained in section V.D) and the downwind state where the receptor is located, all made the SO₂ emission reductions that EPA identified as available at each cost threshold. EPA assumes reductions at each cost threshold from the linked upwind states as well as the downwind receptor state to assess the shared responsibility of these upwind states to address air quality at the identified receptors. Analysis of each receptor did not assume any emission reductions beyond those included in the 2014 base case from upwind states that are not “linked” to that specific downwind receptor (even if the state was “linked” to a different receptor and/or otherwise would have made emission reductions beginning in 2012 due to the Transport Rule).

EPA disagrees with comments suggesting that emission reductions, and resulting decreases in contribution, from upwind states that are not “linked” to a particular downwind receptor should be accounted for in the 2014 AQAT analysis of that receptor. EPA decided to assume reductions only from linked states when analyzing each receptor because EPA is performing a state-specific analysis to support a determination of the amount of each upwind state's responsibility for air quality problems at the downwind receptors that it significantly affects. If the AQAT analysis were to assume emissions reductions in other non-linked states, the AQAT analysis would then contradict the first step of our two-step approach to defining significant contribution to nonattainment and interference with maintenance. Under EPA's two-step approach, only a state that (1) contributes a threshold amount or more to a particular downwind state receptor's air quality problem, and (2) has emission reductions available at the selected cost threshold can be deemed to have responsibility to reduce its emissions to improve air quality at that downwind receptor. EPA believes that the commenters' suggested approach would not qualify as a state-specific approach for determining upwind state responsibility for downwind air quality problems.

Because EPA is relying on the CAMx estimate of nitrate concentrations from the AQAT calibration scenario, the response in nitrate to NO_X reductions at a cost threshold of $500/ton is present in each SO₂ cost threshold scenario analyzed.

EPA determines the cumulative air quality improvement that can be expected at a particular downwind receptor by multiplying each upwind state's percent SO₂ emission reduction by its calibrated receptor specific sulfate response factor and summing the sulfate, nitrate, and other PM_2.5 components (also taken from the 2014 CAMx AQAT calibration scenario).

3. Air Quality Assessment Results

The results of EPA's air quality assessment of the cost threshold scenarios focus on air quality metrics including, but not limited to, average air quality improvement at receptors with 2012 base case nonattainment and maintenance exceedances and an evaluation of estimated receptor design values against annual and 24-hour PM_2.5 standards. See “Significant Contribution and State Emission Budgets Final Rule TSD” for more details.

In EPA's air quality analysis of each downwind receptor, all air quality improvements are measured relative to the “AQAT base case.” This base case reflects AQAT's estimated PM_2.5 concentrations under base case 2014 SO₂ emissions. The AQAT base case itself is not used for any decision points and only serves as an appropriate starting point for comparison of air quality improvements at SO₂ cost thresholds. EPA ensures internal analytic consistency by comparing all air quality improvements at analyzed SO₂ cost thresholds to the AQAT base case.

Regarding average air quality improvement at exceeding 2012 base case receptors, EPA identified 41 receptors with nonattainment or maintenance problems in the 2012 base case. EPA assessed the cumulative reduction in 24-hour PM_2.5 maximum design value at each increasing SO₂ cost threshold from the maximum design value from the AQAT base case, and averaged the reduction across the 41 receptors. The results of this assessment indicate diminishing incremental returns to 24-hour PM_2.5 maximum design value reduction as SO₂ cost threshold levels increase. EPA finds reductions in maximum design value of 4.28 μg/m³ at $500; 4.98 μg/m³ at $1,600; 5.33 μg/m³ at $2,300; 5.46 μg/m³ at $2,800; 5.60 μg/m³ at $3,300; and 6.08 μg/m³ at $10,000. These results are provided in table VI.C-1.

Additionally, EPA evaluated the AQAT estimated 2014 average and maximum design values for these receptors at each cost threshold against the annual and 24-hour PM_2.5 standards. EPA determined the estimated number of receptors with nonattainment or maintenance problems at $500/ton cost threshold of NO_X and each of the cost threshold scenarios assessed for SO₂. These results are provided in table VI.C-2 in terms of the number of receptors and the number of nonattainment areas containing these receptors.

In the proposal, EPA evaluated whether the imposition of the rule's upwind emission reduction requirements could cause changes in operation of electric generating units in states not regulated under the proposal. EPA recognized that such changes could lead to increased emissions in those states, potentially affecting whether they would meet or exceed the 1 percent contribution thresholds used to identify linkages between upwind and downwind states. Such shifting of emissions between states may occur because of the interconnected nature of the country's energy system (including both the electricity grid as well as coal and natural gas supplies).

Using updated emissions and air quality information developed for the final rule, EPA's IPM modeling found that of the states not covered in the final rule for PM_2.5, Arkansas, Colorado, Louisiana, Montana, and Wyoming are all projected to have SO₂ emission increases above 5,000 tons in 2014 with the rule in effect. EPA analysis shows the SO₂ emission increases result from expected shifts to higher sulfur coal in these states. Using AQAT, a state-level assessment of these emission increases relative to the state specific contributions to downwind receptors (where available) indicates that projected increases in the SO₂ emissions would not increase any of these states' contributions to an amount that would meet or exceed the 0.15 μg/m³ or 0.35 μg/m³ thresholds for annual and 24-hour PM_2.5, respectively. For this reason, EPA has determined that it is not necessary to include these additional states in the Transport Rule as a result of the effects of the rule itself on SO₂ emissions in uncovered states. See “Significant Contribution and State Emission Budgets Final Rule TSD” in the docket for this rulemaking for more details.

D. Multi-Factor Analysis and Determination of State Emission Budgets

EPA used the cost, emission, and air quality information described in the previous sections to perform its multi-factor analysis. By looking at different “cost thresholds”—places where there was a noticeable change on the cost curve because emission reductions occur—and examining the corresponding impact on air quality, EPA identified the amount of emissions that represent significant contribution to nonattainment and interference with maintenance within each state. After quantifying this amount of emissions, EPA established state “budgets” which represent the remaining emissions for the state in an average year (step 4).

For states covered by the rule for PM_2.5, EPA calculated annual NO_X and annual SO₂ budgets. For states covered by the rule for ozone, EPA calculated ozone-season NO_X budgets. This section explains the multi-factor assessment and how EPA used this assessment to determine state-specific budgets.

1. Multi-Factor Analysis (Step 3)

a. Overview

As described in section VI.B, EPA examined how different cost thresholds impacted emissions in states with air quality contributions that meet or exceed specific air quality thresholds, as discussed in section V.D of this preamble. Section VI.C summarizes the estimated air quality impacts in 2014 of these emission levels at downwind receptors, including estimates of their nonattainment and maintenance status (see “Significant Contribution and State Emission Budgets Final Rule TSD” for more details). From these two steps, EPA evaluated the interaction between upwind emissions at different cost levels and air quality at downwind receptors to identify “significant cost thresholds.” These cost thresholds are based on air quality considerations (such as the cost at which the air quality assessment analysis projects large numbers of downwind site maintenance and nonattainment problems would be resolved) or cost criteria (such as a cost where large emissions reductions occur because a particular technology is widely implemented at that cost). EPA examined each cost threshold and then used a multi-factor assessment to determine which serve as cost thresholds that eliminate significant contribution to nonattainment and interference with maintenance for upwind states. Air quality considerations in the assessment include, for example, how much air quality improvement in downwind states results from upwind state emission reductions at different levels; whether, considering upwind emission reductions and assumed local (in-state) reductions, the downwind air quality problems would be resolved; and the components of the remaining downwind air quality problem (e.g., whether it is a predominantly local or in-state problem, or whether it still contains a large upwind component). Cost considerations include, for example, how the cost per ton of emission reduction compares with the cost per ton of existing federal and state rules for the same pollutant; whether the cost per ton is consistent with the cost per ton of technologies already widely deployed (similar to the highly-cost-effective criteria used in both the NO_X SIP Call and CAIR); and what cost increase is required to achieve additional meaningful air quality improvement.

The specific cost per ton thresholds selected as a basis for identifying significant contribution to nonattainment and interference with maintenance in this rulemaking apply only to the determinations made in this rule and do not establish any precedent for future EPA actions under section 110(a)(2)(D)(i)(I) or any other section of the CAA. EPA's selection of specific cost thresholds in the context of this rulemaking relies on current analyses of the cost of available emission reductions, the pattern of interstate linkages for pollution transport, and the downwind air quality impacts specifically related to the 1997 ozone NAAQS, the 1997 annual PM_2.5 NAAQS, and the 2006 24-hour PM_2.5 NAAQS. In addition and as explained below, the selection of the threshold for ozone-season NO_X was influenced by the limited scope of this rule. Any or all of these variables used to identify specific cost thresholds are subject to change. Thus, EPA may use different cost thresholds in future actions, even if those actions relate to the same NAAQS addressed in this rule.

b. Cost Thresholds Examined and Selected for Ozone-Season NO_X

In the proposal, EPA examined various cost thresholds for ozone season NO_X and identified a cost threshold with rapidly diminishing returns at $500/ton. EPA observed that moving beyond the $500 cost threshold up to a $2,500 cost threshold would result in only minimal additional ozone season NO_X emission reductions and would likely bypass less expensive non-EGU emission reduction opportunities (75 FR 45281). EPA noted that for greater costs the curves did not include all available reductions as they do not include non-EGU reductions (75 FR 44286). In the proposal, EPA noted the timely promulgation and implementation of this rule is responsive to the Court's remand of CAIR, will accelerate critical air quality improvement, and more effectively address the mandate of CAA section 110(a)(2)(D) to address significant contribution to nonattainment and interference with maintenance as expeditiously as practicable. EPA did not want to risk delaying air quality benefits available from EGU emission reductions, particularly those emission reductions which eliminate significant contribution to nonattainment and interference with maintenance for many receptors, while the Agency conducts additional analysis to support subsequent transport-related rulemakings including coverage of non-EGU sources (75 FR 45285).

EPA received comments suggesting that it consider cost thresholds higher than $500/ton as reductions beyond the proposed $500/ton cost threshold were needed to fully resolve nonattainment and maintenance issues in downwind states analyzed at proposal. Some of these comments suggested EPA should include non-EGUs as they consider the higher cost thresholds, others suggested EPA continue to exclude non-EGU sources in this rulemaking.

In response to those comments that suggested EPA explore higher cost thresholds because nonattainment and maintenance was not fully resolved, EPA first notes that CAA section 110 (a)(2)(D)(i)(I) only requires the elimination of emissions that significantly contribute to nonattainment or interfere with maintenance of the NAAQS in other states. Section 110(a)(2)(D)(i)(I) focuses exclusively on the transport component of nonattainment and maintenance problems. Section 110(a)(2)(D)(i)(I) does not shift to upwind states the responsibility for ensuring that all areas in other states attain the NAAQS. As such, the mandate of section 110(a)(2)(D)(i)(I) is not to ensure that reductions in upwind states are sufficient to bring all downwind areas in to attainment, it is simply to ensure that all significant contribution to nonattainment and interference with maintenance is eliminated. Thus, the presence of residual nonattainment or maintenance areas does not, by itself, signify a failure to satisfy the requirements of 110(a)(2)(D)((i)(I).

Furthermore, as noted in section VI.A, EPA is finalizing coverage only for the EGU emission source-sector category in this rulemaking. EPA has not included non-EGU sources in this final rulemaking. EPA remains convinced that timely promulgation and implementation of this rule is responsive to the Court's remand of CAIR.

To the extent that significant contribution is not eliminated for the 1997 ozone NAAQS standard at the $500/ton cost threshold, EPA is not addressing in this rulemaking whether a cost threshold greater than $500/ton is justified for some upwind states and downwind receptors. EPA believes it can best serve these states where concerns persist regarding projected nonattainment or maintenance of the 1997 ozone NAAQS by quickly finalizing this rule and seeking further non-EGU reductions in subsequent rulemakings. Table VI.B-2 illustrates the small amount of EGU reductions available as cost threshold increases above $500/ton. The ozone-season NO_X reductions available in the Transport Rule states between the $500/ton and $1,000/ton cost thresholds amount to less than 3,000 tons. EPA believes that potentially substantial non-EGU ozone-season NO_X reductions become available approaching the $1,000/ton cost threshold. EPA emphasized this in the proposal, noting that the cost curves for ozone season NO_X did not reflect all available reductions as they do not include non-EGU reductions (75 FR 45286). For these reasons, EPA did not consider cost thresholds greater than $500/ton.

EPA did not consider cost thresholds below $500/ton for ozone-season NO_X. $500/ton is a reasonable threshold representing a significant amount of lowest-cost NO_X emission reductions from EGUs, largely accruing from the installation of combustion controls, such as low-NO_X burners, and constitutes a reasonable cost level for operation of existing NO_X controls such as SCRs. EPA believes it would be inappropriate for a state linked to downwind nonattainment or maintenance areas to stop operating existing pollution control equipment (which would increase their emissions and contribution). This is increasingly likely to occur at cost thresholds lower than $500/ton. Therefore, EPA did not find cost thresholds lower than $500/ton for ozone-season NO_X to be reasonable for development of the Transport Rule cost curves.

As discussed in section III of this preamble, EPA intends to finalize reconsideration of the March 2008 ozone NAAQS in the summer of 2011 and to expeditiously propose a transport-related action to address any necessary upwind state control responsibilities with respect to that reconsidered NAAQS.

c. Cost Thresholds Examined and Selected for Annual NO_X

Following the assessment of the cost curves in section IV.B and the air quality modeling of the AQAT calibration scenario using CAMx, EPA identified a single cost threshold at $500/ton for annual NO_X. Beyond requiring the year-round operation of existing post-combustion NO_X controls and other reductions modeled at $500/ton threshold, EPA observed a limitation in available low-cost annual NO_X reductions from EGUs. Approximately 7,000 tons of annual NO_X reductions were available from EGUs between the $500/ton and the $1,000/ton cost thresholds (See Table VI.B.-1). Furthermore, above the $500/ton threshold, similar to ozone-season NO_X cost curves, the annual NO_X cost curves do not include all available reductions as they do not include non-EGU reductions. EPA analysis suggests that while NO_X emission reductions lead to reductions in PM_2.5, SO₂ reductions are generally more cost-effective than NO_X reductions at reducing PM_2.5 (75 FR 45281). In part, for these reasons, EPA's multi-factor assessment suggested that the $500/ton cost threshold for annual NO_X in concert with the cost thresholds identified for SO₂ were the appropriate cost thresholds for eliminating significant contribution to nonattainment and interference with maintenance. EPA finds in the final Transport Rule that the $500/ton cost threshold for annual NO_X, in concert with the SO₂ cost threshold selected below, successfully eliminates significant contribution to nonattainment and interference with maintenance for the 1997 annual PM_2.5 NAAQS and the 2006 24-hour PM_2.5 NAAQS in the states covered by this Rule for PM_2.5.

The reasons for not considering cost thresholds lower than $500/ton for annual NO_X are the same as those identified for not doing so for ozone-season NO_X. In addition to its PM_2.5 reduction benefits, annual NO_X control at the $500/ton threshold can help to reduce nitrate replacement in the atmosphere. As explained earlier, nitrate replacement happens when SO₂ emissions reductions successfully reduce ammonium sulfate (a component of PM_2.5) but provoke a PM_2.5 rebound effect by freeing up additional ammonia to form ammonium nitrate (another component of PM_2.5).

d. Cost Thresholds Examined and Selected for SO₂

EPA first assessed the downwind air quality impacts of emission reductions modeled at the $500/ton threshold in all states found to be linked to downwind sites for PM_2.5 transport, as well as in the states hosting those downwind sites. The air quality assessment tool projected that those reductions do not fully resolve nonattainment and maintenance problems with the PM_2.5 standards for certain areas to which the following states are linked: Illinois, Indiana, Iowa, Kentucky, Maryland, Michigan, Missouri, New Jersey, New York, North Carolina, Ohio, Pennsylvania, Tennessee, Virginia, West Virginia, and Wisconsin. EPA proceeded to analyze available 2014 emission reductions at higher cost thresholds from these states, collectively referred to as Group 1 states for SO₂ control.

For Group 2 states, the air quality assessment tool projected that the SO₂ reductions at this first cost threshold assessed would resolve the nonattainment and maintenance problems for all of the areas to which the following states are linked: Alabama, Georgia, Kansas, Minnesota, Nebraska, South Carolina, and Texas. EPA thus finds that these states' significant contribution is eliminated at the $500 per ton level in 2014; they are collectively referred to as Group 2 states for SO₂ control. Because their significant contribution is eliminated at this stringency of control, EPA did not analyze higher cost thresholds for Group 2 states.

The states in Group 1 and Group 2 are rationally grouped considering air quality and cost. EPA determined that it would not be appropriate to assign the same cost threshold to Group 2 and Group 1 states because a significantly lower cost threshold was sufficient to resolve air quality problems at all downwind receptors linked to the Group 2 states. Although states are linked to different sets of downwind receptors, EPA analysis indicated that the cost threshold needed to resolve downwind air quality problems varied only to a limited extent among states within Group 1 and among states within Group 2. It did, however, vary greatly between the Group 1 and Group 2 states. The ruling of the DC Circuit in Michigan v. EPA, 213 F.3d 663, 679-80 (D.C. Cir. 2000), accepting EPA's prior use of a transport remedy with uniform controls, supports EPA's decision to use a uniform cost threshold for a group of states.

As discussed in section VI.B, the cost threshold for Group 1 states was examined at escalating levels in 2014 (it remained at $500/ton for Group 2 states). EPA examined emissions at SO₂ cost thresholds of $500, $1,600, $2,300, $2,800, $3,300, and $10,000/ton for Group 1 states in 2014. The higher SO₂ marginal costs were only imposed in Transport Rule states starting in 2014, by which time the advanced pollution control retrofits induced at those higher cost thresholds could be installed. (See section VI.D.2 for EPA's assessment and decisions regarding SO₂ budget formation in Group 1 states in 2014.)

EPA observed some degree of additional air quality benefit at downwind receptors across all of the cost thresholds examined for SO₂, but significant air quality outcomes were achieved at the $2,300/ton cost threshold. The $2,300/ton threshold is projected to resolve the last remaining nonattainment area for the annual PM_2.5 standard (Liberty-Clairton), and it also is projected to resolve the nonattainment and maintenance problems with the 24-hour PM_2.5 standard at 1 monitor in the Detroit area and resolve the maintenance problems in the Cleveland area. There were significant air quality improvements at this level in connection with widespread deployment of pollution control technology, while the cost impacts remained reasonable.

Moving beyond $2,300/ton to the $2,800/ton and $3,300/ton thresholds, EPA projected notably smaller air quality improvements compared to those projected when moving from the $1,600/ton threshold to the $2,300/ton threshold. EPA also projected no ultimate change in the 24-hour PM_2.5 attainment status of the remaining nonattainment area (Liberty-Clairton) or three remaining maintenance areas (Chicago, Detroit, and Lancaster). At the same time, the total program cost continued to increase by about the same interval at each of these thresholds as it had between the $1,600/ton and $2,300/ton thresholds. EPA thus observed a relatively lower cost-effectiveness of downwind PM_2.5 control via upwind SO₂ reductions beyond $2,300/ton for the receptors linked to Group 1 states. Table VI.D-1 and Figure VI.D-1 demonstrate this relationship between cost of EGU SO₂ control and downwind PM_2.5 concentration impacts, showing a sustained diminishing of cost effectiveness beyond the $2,300/ton threshold. The $2,300/ton threshold in this analysis is situated at the “knee-in-the-curve” area of cost-effectiveness for addressing downwind PM_2.5 concentrations with SO₂ reductions, beyond which point the air quality gains per dollar spent on additional reductions are much smaller. This relationship is demonstrative of the economic potency of SO₂ reductions at each cost threshold to address the PM_2.5 concentrations at linked receptors in this analysis.

Furthermore, even at the $10,000/ton cost threshold, AQAT still projects Liberty-Clairton to face maintenance concerns with the annual PM_2.5 standard and is projected to remain in nonattainment of the 24-hour PM_2.5 standard, while the Chicago and Lancaster areas are still projected to have residual maintenance problems with the 24-hour PM_2.5 standard. EPA projected that even total elimination of EGU SO₂ emissions (no matter the cost) would not be able to resolve either nonattainment of the 24-hour PM_2.5 standard in the Liberty-Clairton area or the residual maintenance concerns with that standard in Lancaster County. EPA thus finds that other PM_2.5 strategies, including local reductions of other PM_2.5 precursors, are important to consider for remaining nonattainment and maintenance areas to seek further improvements in PM_2.5 concentrations.

Considering both air quality and cost, EPA's multi-factor analysis indicated $2,300 per ton as an appropriate cost threshold for SO₂ in the Group 1 states. EPA believes the analyzed cost thresholds lower than $2,300/ton were not appropriate for SO₂ control in the Group 1 states under the Transport Rule for the following reasons:

• Downwind air quality impacts up to the $2,300 threshold are significant. Moving up to $2,300/ton successfully resolves all downwind nonattainment of the annual and 24-hour PM_2.5 standards except for the Liberty-Clairton receptor in Allegheny county with respect to 24-hour PM_2.5, which EPA has noted is heavily influenced by a local source of organic carbon (75 FR 45281).
• Upwind emission reductions available up to $2,300/ton are highly cost-effective compared with similar regulations.
• The emission reductions up to this threshold are achievable with widespread deployment of controls that can be installed at power plants by 2014.
• As stated at proposal, EPA finds it reasonable to require a substantial level of control of upwind state emissions that significantly contribute to nonattainment or maintenance problems in another state. The $2,300/ton cost threshold is comparable to EPA's survey of local non-EGU SO₂ reduction opportunities in the PM_2.5 NAAQS RIA, which range in cost from just above $2,300/ton to over $16,000/ton (2007 $). EPA thus finds it reasonable to seek EGU SO₂ reductions up to $2,300/ton (rather than at a lower cost threshold) in the states linked to receptors with ongoing attainment and maintenance concerns with the PM_2.5 NAAQS.

EPA believes the analyzed cost thresholds above $2,300/ton were not appropriate for SO₂ control in the Group 1 states under the Transport Rule for the following reasons:

• As noted above, AQAT suggests reductions up to $2,300/ton were able to resolve all projected downwind nonattainment of the annual and 24-hour PM_2.5 NAAQS, with the sole exception of projected nonattainment of the 24-hour PM_2.5 standard at a receptor in Liberty-Clairton. It is well-established that, in addition to being impacted by regional sources, the Liberty-Clairton area is significantly affected by local emissions from a sizable coke production facility and other nearby sources, leading to high concentrations of organic carbon in this area. EPA finds that the remaining PM_2.5 nonattainment problem is predominantly local and therefore does not believe that it would be appropriate to establish a higher cost threshold solely on the basis of this projected ongoing nonattainment of the 24-hour PM_2.5 standard at the Liberty-Clairton receptor.

• Approximately 70 percent of base case SO₂ emissions from Group 1 states were eliminated at the $2,300/ton cost threshold, leaving a decreasing amount of emission reductions available at each increased cost threshold beyond $2,300/ton.
• Additional EGU SO₂ reductions available from EGUs beyond the $2,300/ton threshold level realize significantly less improvement in downwind PM_2.5 concentrations per dollar spent to impact receptors linked to Group 1 states. In other words, the cost-effectiveness of controlling EGU emissions in Group 1 states to improve downwind PM_2.5 concentrations at the linked receptors is notably diminished beyond the $2,300/ton threshold in this analysis. See Figure VI.D-1.
• EGUs are by far the largest source category for SO₂ emissions. This analysis shows that reductions of EGU SO₂ emissions up to the $2,300/ton cost threshold were significantly more cost-effective for improving downwind PM_2.5 concentrations than further such reductions (beyond the $2,300/ton cost threshold) would be to address the remaining PM_2.5 maintenance concerns. EPA's analysis also shows that these maintenance concerns cannot be fully resolved even with complete elimination of all remaining EGU SO₂ emissions, no matter the cost. EPA finds that other PM_2.5 precursor emission reductions, particularly those from local sources will be critical for states in these remaining areas to consider for controlling PM_2.5 concentrations with respect to maintenance of the 2006 24-hour PM_2.5 NAAQS.

In summary, the appropriate cost thresholds for each state were identified through the multi-factor assessment. This assessment included both cost and air quality considerations. As explained above, the ozone-season NO_X threshold was determined to be $500/ton for all states required to reduce ozone-season NO_X, with residual nonattainment and maintenance concerns to be addressed in a future rulemaking addressing a broader set of source categories for additional cost-effective reductions. For PM_2.5, the appropriate cost threshold for each state was determined to be either the level at which nonattainment and maintenance issues were completely resolved in downwind states to which the state is linked, the level where remaining nonattainment and maintenance issues are primarily local, or where we found greatly diminished improvements in air quality occurring if EPA moved further up the cost curve. This assessment yielded a cost threshold of $2,300/ton on SO₂ for Group 1 states starting in 2014 ($500/ton in 2012), a cost threshold of $500/ton on SO₂ for Group 2 states, and a cost threshold of $500/ton on annual NO_X for all states required to reduce emissions for purposes of the annual or 24-hour PM_2.5 NAAQS in this rule.

As explained above, none of these specific cost thresholds establish any precedent for the cost per ton stringency of reductions EPA may require in future transport-related rulemakings; these specific cost thresholds are based on current analyses of air quality and cost of emission reductions with respect to the NAAQS considered in this rulemaking and thus would not be relevant to future rulemakings (which would consider updated information) or rulemakings with respect to different NAAQS. In particular, EPA acknowledges that additional action EPA will require in a subsequent rulemaking to address significant contribution to nonattainment and interference with maintenance of the 2008 ozone NAAQS (once reconsideration is finalized) is very likely to require a higher cost per ton stringency of ozone-season NO_X control applied to a broader set of source categories from upwind states than found to be appropriate for this rulemaking.

2. State Emission Budgets (Step 4)

a. Budget Methodology

EPA used the multi-factor assessment to identify, for each state, the cost threshold that should be used to quantify that state's significant contribution. As described above, in the context of this rulemaking EPA identified a cost threshold of $500/ton for ozone-season NO_X control for all states required to reduce ozone-season NO_X emissions for purposes of the 1997 ozone NAAQS in this rule. EPA also identified a cost threshold of $500/ton for annual NO_X control for all states required to reduce annual NO_X emissions for purposes of the annual or 24-hour PM_2.5 NAAQS in this rule. Finally, EPA identified a cost threshold of $500/ton of SO₂ starting in 2012 for all states required to reduce SO₂ emissions for purposes of the annual or 24-hour PM_2.5 NAAQS in this rule, and $2,300/ton for the Group 1 states starting in 2014.

EPA used these cost thresholds from the multi-factor analysis to quantify each state's emissions that significantly contribute to nonattainment or interfere with maintenance downwind. For example, for a Group 1 state, EPA modeling of the cost threshold conveys emission reductions available in each covered state from operation of existing pollution controls as well as all emission reductions available at cost thresholds of $500/ton for annual NO_X in 2012 and 2014, $500/ton for SO₂ in 2012, and $2,300/ton for SO₂ in 2014. The total SO₂ and NO_X projected at these cost levels in that state in those years represents that state's emissions once significant contribution to nonattainment or interference with maintenance downwind for the relevant PM_2.5 NAAQS has been eliminated.

EPA's modeling of a state's SO₂ and annual NO_X emission levels (from fossil-fired EGUs 25 MW) at the relevant cost thresholds in each state reflect that state's emissions from covered sources after the removal of significant contribution to nonattainment and interference with maintenance of the PM_2.5 NAAQS considered in this rulemaking. As these state emission levels reflect the removal of significant contribution and interference with maintenance, they are reasonable levels on which to determine state budgets. Consequently, EPA based state budget levels on the state level emissions that remained at the cost threshold. Each state's budget corresponds to its emission level following the elimination of significant contribution to nonattainment and interference with maintenance in an average year (before taking year-to-year variability into account, as discussed in section VI.E below). Therefore, the implementation and realization of these budgeted emission levels leads to the elimination of significant contribution to nonattainment and interference with maintenance and EPA meets the statutory mandate of section 110(a)(2)(D)(i)(I) with respect to the 1997 annual PM_2.5 NAAQS and the 2006 24-hour PM_2.5 NAAQS.

EPA's establishment of state budgets for ozone-season NO_X control follow the same methodology as described above for SO₂ and annual NO_X. Implementation of these ozone-season NO_X budgets reflects the elimination of significant contribution to nonattainment and interference with maintenance of the 1997 ozone NAAQS for 15 states, whereas 11 other states' ozone-season NO_X budgets reflect meaningful progress toward (but may not reflect full completion of) this elimination under the mandate of section 110(a)(2)(D)(i)(I). See section III for lists of states.

This approach to basing budgets on projected state level emissions used in the multi-factor analysis is identical to the approach used in the proposal for determining 2014 SO₂ budgets for Group 1 states. EPA is extending this approach more broadly in the final Transport Rule to create state budgets for ozone-season NO_X, annual NO_X, and SO₂ in all relevant states in both 2012 and 2014. In the proposal EPA used a more complex approach based on a comparison of historic and projected unit-level emissions (further adjusted for operation of existing controls) in each state to create 2012 state budgets for ozone-season NO_X, annual NO_X, and Group 2 SO₂. At the time of proposal, EPA believed that historic 2009 emissions data were in some cases more representative of expected emissions in 2012 than pure modeling projections made at the time (75 FR 45290).

However, following the proposal EPA has made significant updates to the IPM model for projecting EGU emissions, including specifically the adoption of 2009 historic data into its modeling parameters directly. EPA also received substantial public input following the proposal on the model's assumptions and representation of individual units, which allowed EPA to improve its 2012 and 2014 emission projections for states under the cost thresholds considered. These modeling updates diminish the concerns EPA expressed at proposal that 2009 historic data may have offered for some states a better proxy for 2012 emissions than model projections, particularly now that EPA is incorporating 2009 data directly in its updated modeling projections. Given these updates to the model in response to public comment, EPA believes it is more appropriate for the final rule to use a consistent approach based on projected state level emissions for all state budgets, as was done for Group 1 SO₂ budgets in 2014 at proposal. EPA received significant comment supporting the use of the model to project state-level emissions for creating budgets in this manner. EPA also received comments that criticized the proposal's methodology for 2012 budgets for lack of transparency, unnecessary complexity, and inconsistency with the state-level emission projections used in the air quality modeling. EPA's decision for the final Transport Rule to consistently apply across all pollutants the budget methodology originally used for Group 1 SO₂ budgets in 2014 addresses those concerns.

This budget methodology for the final rule uses projected state-level emissions in 2012 and 2014 to set emission budgets for those years on relevant pollutants for that state to control under the Transport Rule. EPA's modeling projects that some states have 2014 emissions that are lower than their 2012 projected emissions even as the same cost threshold (e.g., $500/ton) is applied in both years. This occurs in the annual NO_X, ozone-season NO_X, and Group 2 SO₂ program. As such, EPA's application of this budgeting methodology results in a tightening of budgets in states whose projected emissions of that budgeted pollutant decline from 2012 to 2014 as the cost threshold is held constant.

There are two primary variables that explain the decrease in emissions for some states between 2012 and 2014 as the cost threshold remains constant over both time periods. First, even though the cost threshold is constant between 2012 and 2014 for the programs noted above, the cost threshold for SO₂ Group 1 increases in 2014. This higher cost threshold for Group 1 SO₂ results in obvious reductions in SO₂ emissions in the Group 1 states, but also may lower the cost of certain related NO_X reductions in those states as well such that they become newly available within the $500/ton threshold. For example, if a state increases natural gas generation in response to the higher SO₂ cost threshold, such action also yields additional annual and ozone-season NO_X emission reductions that are cost-effective at the $500/ton NO_X threshold. Where the cost curve modeling shows such additional cost-effective NO_X reductions in tandem with SO₂ control, EPA is therefore reducing those states' 2014 annual NO_X and ozone-season NO_X budgets accordingly, so that those budgets accurately reflect remaining emissions from covered sources in those states after the elimination of all emissions that can be reduced up to the relevant cost thresholds (e.g., $500/ton).

Second, some of these additional reductions are driven by non-Transport Rule variables. These are reductions that occur due to state rules, consent decrees, and other planned changes in generation patterns that occur after 2012, but during or prior to 2014. For example, EPA modeling reflects emission reduction requirements under provisions of a Georgia state rule that go into effect after 2012 but before 2014. These requirements involve the installation and operation of specific advanced pollution controls. These source-specific requirements under a legal authority unrelated to the Transport Rule result in sharp reductions in Georgia's baseline emission projections between 2012 and 2014. Even though the cost threshold for NO_X and for SO₂ in Georgia is $500/ton in both 2012 and 2014, EPA believes it is important to establish separate NO_X and SO₂ budgets that accurately reflect the emissions remaining in Georgia (and other states experiencing similar reductions) after the elimination of emissions that can be reduced up to the Transport Rule remedy's cost thresholds (e.g., $500/ton) (see Table VI.D.3). It illustrates a notable decrease between the 2012 and 2014 state budgets for NO_X and SO₂ in Georgia that is largely driven by state rule requirements. If EPA did not adjust 2014 budgets to account for other emission reductions that would occur even in the baseline, other sources within the state would be allowed to increase their emissions under the unadjusted Transport Rule budgets to offset the emission reductions planned under other requirements such as state rules. Therefore, to prevent the Transport Rule from allowing such offsetting of emission reductions already expected to occur between 2012 and 2014, EPA is establishing separate budgets for 2012 and 2014 in the final Transport Rule to capture emission reductions in each state that would occur for non-Transport Rule-related reasons (i.e., in the base case) during that time.

EPA's modeling also projects that other states would slightly increase emissions from 2012 to 2014 even at the same cost threshold, such as $500/ton. There are two primary variables that explain the increase in emissions for these states between 2012 and 2014. These increases are generally small in magnitude. For annual and ozone season NO_X, they occur as a byproduct of small changes in dispatch related to changes in non-Transport Rule factors (e.g., higher demand in 2014). For SO₂, they primarily occur in Group 2 states and, in addition to the reasons given above, are influenced by some generation shifting from Group 1 to Group 2 states as the Group 1 states begin to face a higher cost threshold in 2014. EPA believes that allowing for such emission growth in covered states beyond 2012 would be inconsistent with the Transport Rule's identification and elimination of significant contribution to nonattainment and interference with maintenance beginning in 2012. Therefore, for any covered state whose emissions of a relevant pollutant are projected to increase from 2012 to 2014 under the relevant cost thresholds selected in the multi-factor analysis described above, EPA is finalizing that state's 2014 emission budget to maintain the same level of the 2012 emission budget, thereby disallowing such an emission increase that is inconsistent with the 110(a)(2)(D)(i)(I) mandate. Tables VI.D-3 and VI.D-4 below list state emission budgets.

The District of Columbia is not covered by the final Transport Rule. As discussed in section V.D of this preamble and as done for the Transport Rule proposal, EPA combined contributions projected in the air quality modeling from Maryland and the District of Columbia to determine whether those jurisdictions collectively contribute to any downwind nonattainment or maintenance receptor in amounts equal to or greater than the 1 percent thresholds. This modeling confirmed that the combined contributions exceed the air quality threshold at downwind receptors for the ozone, annual PM_2.5, and 24-hour PM_2.5 NAAQS considered. Both Maryland and the District of Columbia are therefore linked to these receptors. However, the District of Columbia is not included in the Transport Rule because, in the second step of EPA's significant contribution analysis, we concluded that there are no emission reductions available from EGUs in the District of Columbia at the cost thresholds deemed sufficient to eliminate significant contribution to nonattainment and interference with maintenance of the NAAQS considered at the linked receptors. At the time of this rulemaking, EPA finds only one facility with units meeting the Transport Rule applicability requirements in the District of Columbia. EPA's projections do not show any generation from this facility to be economic under any scenario analyzed (including the base case), and the facility's owners have also announced plans to retire its units in early 2012. Therefore, this unit is projected to have zero emissions in 2012. As such, the total SO₂ and NO_X emissions in the District of Columbia for EGUs that meet the Transport Rule applicability requirements is also projected to be zero. It follows therefore, that EPA did not identify any emission reductions available at any of the cost thresholds considered in the final rule's multi-factor analysis to identify significant contribution to nonattainment and interference with maintenance. For this reason, EPA concludes that no additional limits or reductions are necessary, at this time, in the District of Columbia to satisfy the requirements of section 110(a)(2)(D)(i)(I) with respect to the 1997 ozone, the 1997 PM_2.5 and the 2006 PM_2.5 NAAQS. EPA is therefore neither establishing budgets nor finalizing any FIPs for the District of Columbia in this rule.

EPA notes that the NO_X budgets for five states linked to downwind ozone receptors in the final Transport Rule are equal to their projected 2012 base case emissions. The five states are Arkansas, Indiana, Louisiana, Maryland, and Mississippi. These states are among those found to meet or exceed the 1 percent contribution threshold for the 1997 ozone NAAQS at downwind receptors and are thus “linked” to downwind receptors. EPA therefore evaluates, in the second step of its significant contribution analysis, what emission limits are necessary to ensure that all emissions that constitute the state's significant contribution to nonattainment and interference with maintenance are prohibited. As explained above, EPA decided to require from all such states all reductions available at the $500/ton cost threshold. The five states identified above do not appear to show EGU ozone-season NO_X reductions at the $500/ton cost threshold relative to the 2012 base case projections (which do not take into account reductions to be made in other states as a result of this rule). Therefore, EPA conducted further analysis to evaluate whether such reductions were available in these states and whether emission limits are necessary to prohibit these states from significantly contributing to downwind nonattainment or interfering with maintenance of the 1997 ozone NAAQS in other states. (See the docket to this rulemaking for the IPM run titled TR_uncontrolled_ozone_states_Final.”)

Specifically, EPA projected those states' ozone-season NO_X emissions if all other linked states (but not these five states) were to make all available reductions at the $500/ton threshold. That analysis revealed that if emission limits were not established for these five states, ozone-season NO_X emissions in each of the states would increase (beyond the 2012 base case emission projections), due to interstate shifts in electricity generation that cause “emissions leakage” in uncovered states. These increases would result in each state's emissions being above the level associated with the prohibition of all emissions that can be eliminated at the $500/ton threshold. EPA thus determined that it is necessary to establish emission limits for these states at the $500/ton level. These limits, although equal to the state's 2012 projected base case emissions, are necessary to prohibit all emissions that can be controlled at the $500/ton cost threshold. In other words, the significant contribution to nonattainment and interference with maintenance addressed by the ozone FIPs for these states is the difference between these states' projected emissions if they were not covered under the Transport Rule (but other states were), and their emissions after all emissions that can be eliminated at $500/ton are prohibited.

In addition, EPA notes that four of these five states (Arkansas, Indiana, Louisiana, and Mississippi) are linked to receptors in either the Houston or Baton Rouge areas, which are projected to continue facing nonattainment or maintenance concerns with the 1997 ozone NAAQS, respectively. To allow these states to increase emissions above base case projections would erode the measurable progress toward eliminating significant contribution to nonattainment and interference with maintenance secured by achieving ozone-season NO_X reductions in the other states linked to these receptors. Furthermore, as discussed in section III, EPA may require additional reductions in these states to fully address significant contribution to nonattainment and interference with maintenance with respect to the 1997 ozone NAAQS in a future rulemaking to be proposed after finalizing reconsideration of the 2008 ozone NAAQS.

b. Relationship of Group 1 and Group 2 States for SO₂ Control

In the Proposal, EPA chose not to allow sources in Group 1 states to use Group 2 SO₂ allowances for compliance, and likewise not to allow sources in Group 2 states to use Group 1 SO₂ allowances for compliance at any time. The preamble clearly states, “With regard to interstate trading, the two SO₂ stringency tiers would lead to two exclusive SO₂ trading groups. That is, states in SO₂ Group 1 could not trade with states in SO₂ Group 2” (75 FR 45216). No such distinction or limitation exists for NO_X allowance trading.

EPA received significant public comment both in support and opposition to the two distinct SO₂ trading programs. Those in opposition noted that the variability limits imposed at the state level made the compliance restrictions between the two groups unnecessary. Commenters also noted that it may unfairly penalize sources that are part of the same airshed, but are on opposite sides of a state boundary. Those in favor of the separate SO₂ compliance programs noted that it would reduce the probability of a state exceeding its variability limit. Allowing the use of Group 1 or Group 2 allowances for compliance between the two SO₂ programs would potentially encourage Group 1 states to purchase allowances instead of making reductions necessary to eliminate significant contribution. Group 1 states are states that need continued reductions (beyond the $500/ton threshold) to eliminate their significant contribution to nonattainment and interference with maintenance. Group 2 states have already eliminated their significant contribution to nonattainment and interference with maintenance at the $500/threshold. So to allow Group 1 or Group 2 allowances to be used interchangeably for compliance between the two SO₂ groups would be to allow the shifting of reductions from areas where they are needed to eliminate significant contribution to nonattainment and interference with maintenance to areas where they are not needed to eliminate the prohibited emissions. EPA also agrees that allowing for trading between the two groups in the remedy finalized in this action would increase risk of a state exceeding its variability limit. For these reasons, EPA is finalizing this rulemaking with the same prohibition on SO₂ trading between Group 1 and Group 2 states that was defined in the proposal. Further, EPA clarifies that while trading of allowances (i.e., buying, selling, and banking) is allowed without restriction, it is specifically the surrender of SO₂ allowances for compliance that is limited. As mentioned earlier, a source in a Group 1 state can only use SO₂ allowances allocated to Group 1 states for compliance with the SO₂ trading program. Likewise, a source in a Group 2 state can only use SO₂ allowances allocated to Group 2 states for compliance with the SO₂ trading program.

c. Ozone-Season Budgets

EPA established the ozone-season NO_X budgets in a similar manner to the annual NO_X and SO₂ budgets by using the state level emissions from the cost threshold that reflected the removal of significant contribution to nonattainment and interference with maintenance. Ozone-season budgets were based on the state level emissions from fossil-fuel-fired units greater than 25 MW observed at this cost threshold. As described in section VI.B, all cost thresholds examined reflected the final Transport Rule geography and the marginal costs were applied accordingly. Therefore, for an ozone-only state like Florida, the state level emissions would only reflect an ozone-season cost threshold of $500/ton in the final cost curves for 2012 and 2014. For a state subject to both annual and ozone-season programs, the marginal cost curves would reflect a $500/ton NO_X cost year round, a $500/ton SO₂ cost in 2012 and the $2,300/ton SO₂ cost starting in 2014 if a Group 1 state.

(1) Length of Ozone Season

(a) Proposed Rule. For purposes of determining ozone-season budgets in the proposed rule, EPA defined the ozone season based on a 5 month period (May 1 through September 30). This 5 month ozone season was consistent with the approach taken by the OTAG, the NO_X SIP Call, and CAIR. EPA requested comment on whether EPA should base final rule budgets on a longer season, such as March through October.

(b) Public Comments. Several commenters supported continuing with the May through September time period. One commenter supported continuing with this time period, but argued that EPA should consider lengthening the ozone season for future efforts. One commenter questioned the concept of ozone season budgets and recommended EPA focus on sources with greater emissions on high ozone days.

(c) Final rule. For the final rule, EPA has retained the approach in the proposed rule, as commenters broadly supported the proposal's ozone-season duration and ozone-season NO_X limitations. Notably, many Transport Rule states covered for PM_2.5 reductions will have sources with annual NO_X controls that are likely to keep operating year round to address PM_2.5 and ozone. EPA believes that experience from ozone-season NO_X trading has consistently shown that the emission measures taken to comply with ozone-season budgets provide emission reductions throughout the ozone-season, including the highest ozone days. (See NO_X Budget Trading Program and CAIR Program progress reports in the docket to this rulemaking or at http://www.epa.gov/airmarkets/progress/nbp08.html and http://www.epa.gov/airmarkets/progress/CAIR_09/CAIR09.html.) However, EPA believes that there is merit in future Agency actions addressing ozone transport in considering strategies to target high ozone days more specifically.

d. Summary of Cost Thresholds and Final Budgets for PM_2.5 and Ozone

Summary of methodology. In summary, EPA determined that SO₂ emissions that could be reduced for $2,300/ton in 2014 should be considered a state's significant contribution to nonattainment and interference with maintenance, unless EPA determined that a lesser reduction would fully resolve the nonattainment and/or maintenance problem for all the downwind receptors to which a particular state might be linked. For these Group 2 states EPA is determining that a lesser reduction of SO₂, based on the amount of SO₂ reductions that can be reasonably achieved by 2012 is appropriate. This level is defined by the reductions observed in the $500/ton cost threshold. EPA also determined that all states linked to downwind PM_2.5 nonattainment and maintenance problems should be required to achieve those emission reductions that can be reasonably achieved by 2012. Finally, EPA determined that all states linked to downwind PM_2.5 nonattainment and maintenance problems should, by 2012, remove all NO_X emissions that can be reduced for $500/ton and run all existing controls in 2012.

For ozone-season NO_X, EPA determined that all states linked to downwind ozone and nonattainment and maintenance problems should be required to achieve those ozone-season emission reductions associated with a cost threshold of $500 per ton. Additionally, EPA examined final 2012 and 2014 budgets based on state level emissions at $500 cost threshold.

The budget formation methodology finalized in this action responds to concerns about state budgets expressed by commenters on the Transport Rule proposal. EPA requested comment on the four step approach used to determine significant contribution and determine budgets in the proposal. Some commenters noted that the state level emissions from the cost thresholds used to determine significant contribution to nonattainment and interference with maintenance did not match the state level emissions allowed by the final budgets. The concern was that the state level emissions that reflected the elimination of significant contribution in the AQAT analysis, in particular for NO_X, were less than the emissions allowed by the final budgets. The result would be an implementation that did not quite fully eliminate the significant contribution to nonattainment and interference with maintenance defined in the rule. The proposed budgets not matching the levels reflected in the proposed costing runs were an artifact of the budget formation process that relied on a combination of historic and projected data. While EPA noted this process resulted in state budgets that “reflected” EGU emissions at $500/ton, it was not always consistent with the EGU emissions at $500/ton in the costing runs as the commenters noted. By using the cost curves to determine both significant contribution to nonattainment and interference with maintenance—and state budgets—in the final rule, EPA addresses the commenter's concerns about any inconsistency between the two in the proposal.

Some commenters expressed concern that the Transport Rule would result in state budgets that were in some cases higher than those established in CAIR. Commenters suggested that this would be inconsistent with requirements or the spirit of certain CAA provisions aimed at preventing backsliding, i.e., sections 110(l), 172(e), and 193. However, the DC Court of Appeals rejected the state budgets in CAIR as arbitrary and capricious and not consistent with CAA section 110(a)(2)(D)(i)(I) (North Carolina, 531 F.3d 918 and 921) and remanded CAIR to EPA to promulgate a new rule replacing CAIR and consistent with the Court's decision (North Carolina, 550 F.3d 1178). As discussed elsewhere in this section, on remand EPA developed new, final state budgets that address the Court's concerns and meet section 110(a)(2)(D)(i)(I) requirements.

Although some state budgets under the final rule are higher than those under CAIR, this does not violate either the letter or the spirit of CAA provisions aimed at backsliding. In particular, CAA section 110(l) provides that the Administrator may not approve a plan revision that would “interfere with any * * * applicable requirement” of the CAA. 42 U.S.C. 7410(l). Because the Court reversed and remanded CAIR with instructions to “remedy” the rule's “fundamental flaws” (including specifically the state budgets found to be unlawful (North Carolina, 550 F.3d 1178), it is difficult to see how new state budgets replacing unlawful budgets and meeting section 110(a)(2)(D)(i)(I) requirements could be viewed as interfering with requirements of the CAA. Indeed, the commenters' approach would severely limit EPA's ability to meet the Court's mandate to develop a new rule consistent with section 110(a)(2)(D)(i)(I). See North Carolina, 531 F.3d 921 (explaining that EPA may not require “some states to exceed the mark” of eliminating their significant contribution). Further, the other CAA sections cited by the commenters (section 172(e), addressing circumstances where the Administrator relaxes a NAAQS, and section 193, addressing the treatment of requirements promulgated before the November 15, 1990, enactment date for the 1990 Amendments to the Clean Air Act) are not applicable here.

Additionally, while the CAIR budgets may have been tighter than Transport Rule state budgets for a couple of states, the sum of state budgets that were subject to both CAIR and the Transport Rule is lower under the Transport Rule for the annual programs. Moreover, the carryover of the large Title IV allowance bank in CAIR allowed for a great deal more emissions within any given state than is permitted under the Transport Rule.

E. Approach to Power Sector Emission Variability

1. Introduction to Power Sector Variability

Variability is an inherent aspect of the production and delivery of electricity. It follows that variations in state emissions are not only a result of variations in the level of emission control, but also are caused by the inherent variability in power generation. The state budgets do not account for this latter source of variability at the state level. Emission variability is built into the design of power systems, which use a wide mix of power generation sources with varying use and emission patterns to ensure reliability in electric power generation. Variations in weather, demand due to changes in the level of economic activity, the portion of electric generation that is fossil-fuel-fired, the length and number of outages at power generation units, and other factors, can lead to significant variations in the load levels of different power generation sources. Variations in the load levels of sources in any given state cause variations in the level of emissions in that state. Thus, EPA believes it is appropriate, in this rule, to take into account the variations that are caused by inherent variability in power generation. More specifically, variations in these external variables can cause significant fluctuations in state emissions, even when action has been taken to prohibit all emissions within a state that significantly contribute to nonattainment or interfere with maintenance in another state. For this reason, EPA considers variability when determining the state specific requirements in this rule. EPA does so by developing variability limits and assurance levels for each state, as described in this section, that are consistent with the statutory mandate of CAA section 110(a)(2)(D)(i)(I).

Loads on a power system, and thus on power generation sources in a given state that are on the power system, vary over every time interval, changing not only in the short term and seasonally, but also annually. As noted above, load patterns and levels are determined by a multiplicity of factors, including weather, economic activity, the portion of electric generation that is fossil-fuel-fired, and the length and number of outages at power generation units, which vary over time. In particular, weather obviously varies not just from season-to-season but also from year-to-year, and even small changes in annual weather patterns can affect how the power system and power generation sources on the power system operate during a year. For example, load, and the resulting use of generation sources on an interconnected grid to meet load, depend not only on how hot a summer day is, but also on where a heat wave occurs and how long it lasts. Similarly, a relatively cold winter that drives up winter load may also change what generation sources are used to address the increased demand for heat. Thus, the pattern of generation may shift geographically as a weather pattern moves across the country. Because weather and other factors affecting loads, and the patterns of generation used to meet loads, vary over time and from state to state, the resulting level of emissions also varies over time and from state to state.

This variability in emissions is not a result of variation in emission rates, emission controls, or emission control strategies, but instead is a result of the inherent variability in power generation. Patterns of generation change to ensure demand for electricity is met and to ensure continued reliability of the power system. This results in temporal and geographic fluctuations in emissions. In the final Transport Rule, like the proposed rule, EPA explicitly takes account of these changing patterns of generation and the resultant variability in power sector emissions.

As discussed previously, EPA identified a specific amount of emissions that must be prohibited by each state to meet the requirements of CAA section 110(a)(2)(D)(i)(I). EPA also developed state baseline emissions for power generation sources based on projections of state emissions in an average year before the elimination of prohibited emissions, and state budgets for power generation sources based on projections of state emissions in an average year after the elimination of such emissions. However, because of the inherent variability in state-level baseline emissions—resulting from the inherent variability in loads and power system and power generation source operations—state-level emissions will fluctuate from year-to-year even after all significant contribution to nonattainment and interference with maintenance that EPA identified in this final rule are eliminated. In an above average year, emissions may exceed the state budgets which are based on an analysis of projected emissions in an average year. EPA believes that, because baseline emissions are variable for reasons unrelated to the degree of emission control in a state and emissions after the elimination of all significant contribution to nonattainment and interference with maintenance are therefore also variable, it is appropriate to take this variability into account in developing the remedy for meeting the requirements of CAA section 110(a)(2)(D)(i)(I). The variability limits and assurance levels in the final rule account for this inherent variability, while ensuring that emissions within each state that significantly contribute to nonattainment or interfere with maintenance in another state are prohibited. EPA believes this approach is both reasonable in that it reflects the operation of the power system generation in order to maintain electric reliability and consistent with the statutory mandate of CAA section 110(a)(2)(D)(i)(I). For these reasons, EPA is finalizing variability limits for each state budget to identify the range of emissions that EPA believes is likely to occur in each state following the elimination of all the state's significant contribution to nonattainment and interference with maintenance.

As discussed above, the air quality-assured trading remedy's state-specific budgets represent each state's emissions in an average year after elimination of significant contribution to nonattainment and interference with maintenance. Because actual base case emissions are likely to vary from projected base case emissions, this remedy incorporates provisions that account for such variability. While the primary purpose of this remedy is to eliminate significant contribution and interference with maintenance, EPA believes variability limits also satisfy several other objectives. The remedy provides the flexibility to deal with real-world variability in the operation of the power system through air quality-assured trading and reduces costs of compliance with emission reduction requirements, while still providing assurance for downwind states that significant contribution to nonattainment and interference with maintenance by upwind states will be eliminated. EPA believes the limited fluctuation in state level emissions that this approach permits is consistent with the statutory mandate of section 110(a)(2)(D)(i)(I) because some geographic and temporal shifting of emissions necessarily results from the inherent variability in power generation and is caused by factors unrelated to the degree of emission control, such as weather, economic activity, and unit availability. Far from excusing any state from addressing emissions within the state that significantly contribute to nonattainment or interfere with maintenance in other states, these variability limits ensure that the system can accommodate the inherent variability in the power sector while ensuring that each state eliminates the amount of emissions within the state, in a given year, that must be eliminated to meet the statutory mandate of section 110(a)(2)(D)(i)(I).

Moreover, the structure of the program, which achieves the required emission reductions through limits on the total number of allowances allocated, assurance provisions, and penalty mechanisms, ensures that the variability limits only allow the amount of temporal and geographic shifting of emissions that is likely to result from the inherent variability in power generation, and not from decisions to avoid or delay the installation of necessary controls. Under the remedy, an individual state can have emissions up to its budget plus the variability limit. However, the requirement that all sources hold allowances covering emissions, and the fact that those allowances are allocated based on state-specific budgets without variability, ensure that the total emissions from the states do not exceed the sum of the state budgets. The remedy, therefore, ensures both that total emissions do not exceed the total of the state budgets and that the required emission reductions occur in each state.

This section describes how EPA calculated variability limits for each state to achieve this goal.

2. Transport Rule Variability Limits

EPA performed analyses using historical data to demonstrate that there is year-to-year variability in base case emissions (even when emission rates for all units are held constant) and to quantify the magnitude of this variability.

The focus of the analysis is on quantifying the magnitude of the inherent year-to-year variability in state-level EGU emissions independent of measures taken to control those emissions (and thus due only to changes in electricity generation within each state). EPA used this analysis to set variability limits as part of the remedy to ensure that states are eliminating their significant contribution to nonattainment and interference with maintenance to protect air quality.

As discussed in detail below, EPA is finalizing the Transport Rule with 1-year variability limits calculated using a modified approach from the one described in the proposal. EPA is not including the proposal's 3-year variability limits in the final Transport Rule. EPA received comments that the 3-year variability limits increased program costs and diminished compliance flexibility without delivering any additional air quality benefits. EGU owners and operators expressed concern that 3-year variability limits would be impracticable to implement and that the 1-year variability limits themselves would be adequately stringent to ensure elimination of significant contribution to nonattainment and interference with maintenance in each state.

After further consideration, EPA has concluded that 3-year variability limits would be unnecessary, would be difficult to anticipate, and would not have a measurable impact on air quality benefits. EPA has determined that annual limits are sufficient to eliminate significant contribution to nonattainment and interference with maintenance in all upwind states while accommodating the historically observed year-to-year fluctuation in state-level EGU emissions even at the same rate of emissions control in a given state.

In the proposal, EPA used statistical methods to derive the 3-year variability limit directly from the 1-year variability limit, meaning that the two are statistically equivalent in the long run under certain statistical assumptions. Primarily, these assumptions were that the variation in electric demand around the budget is random from year-to-year and that, when the annual emissions are averaged over a multi-year time period, the average emissions per year will equal the state's budget. The first assumption was also made in the assessment of the historical year-to-year variation in heat input in developing the 1-year limit (see section 2 of the “Power Sector Variability Final Rule TSD” for more details). Regarding the second assumption, since the state-by-state emission budgets are based on the availability of emission reductions at an equal marginal cost level, EPA expects the sources in each of the upwind states to make these cost-effective reductions and to meet the emission budgets each year, on average.

Since the 3-year variability limit was based on average year-to-year variability over a longer time horizon, EPA notes that a random ordering of those years could yield 2 above-average years in a row. If, by chance, a third above-average year were to follow, the state could face violation of the 3-year limit, even if over a time period longer than 3 years, that state would never have exceeded the statistically-equivalent 1-year variability limit and its annual emissions would have averaged to the level of its budget. Effectively, this means that imposing a multi-year variability limit would erode the 1-year variability limit's ability to accommodate historically observed year-to-year variability in state-level EGU emissions (due only to generation changes), and it would do so without providing any additional air quality benefits or protection for downwind areas (since the average emissions over the long time horizon equal the level of the budget).

For more details about the relationship between the 1- and 3-year limits, see the discussions in section 3 of the “Power Sector Variability” TSD from the proposed Transport Rule, which describes the derivation of the 3-year limit from the 1-year variability and section 3 of the “Power Sector Variability Final Rule TSD”, which describes the results of a numerical simulation showing that the 1- and 3-year limits are statistically indistinguishable and, thus, redundant over the course of the program to accommodate year-to-year variability.

While EPA expects the yearly emissions in each state, on average, to equal the level of the budgets, EPA also estimated the air quality impacts of 5, 10, 15, and 20 percent emission variability using the air quality assessment tool, which is presented in section 4 of the “Power Sector Variability Final Rule TSD.” That analysis shows that year-to-year fluctuations of up to 20 percent in SO₂ emissions from upwind states linked to a given downwind receptor do not undermine the ability of the Transport Rule programs to resolve nonattainment or maintenance concerns at that receptor. The analysis presented in the TSD focuses on SO₂ emissions and was designed to examine the sensitivity of downwind air quality to upwind EGU emission levels. The share of total SO₂ emitted by EGUs is significantly larger than the share of total NO_X emitted by EGUs. For example, in the states for which EPA modeled base case contributions of these pollutants, EGUs accounted for 74 percent of total SO₂, 14 percent of total annual NO_X, and 15 percent of total ozone-season NO_X emissions. Therefore, when varying EGU emissions only, downwind air quality would be most sensitive to upwind variations in SO₂, because relative variations in EGU SO₂ emissions have a greater impact on total SO₂ emissions than the same relative variation in EGU NO_X emissions would have on total NO_X emissions affecting downwind air quality. Because the Transport Rule only affects upwind emissions from EGU sources, downwind air quality would be more sensitive to variability in upwind state SO₂ emissions under this rule than variability in upwind state NO_X emissions under this rule (given that the rule affects a smaller scope of total NO_X emissions compared to the scope affected of total SO₂ emissions). Thus, EPA chose to analyze the “worst-case” potential downwind air quality impacts from year-to-year variability above upwind state SO₂ budgets, and EPA therefore believes that its findings from this analysis are valid for ascertaining the potential downwind air quality impacts from variation at those levels in both SO₂ and NO_X under the Transport Rule programs.

Furthermore, because the state budgets are based directly on IPM modeling of electric generation when cost-effective emission reductions have been achieved, sources within each state should have the same incentive to meet that budget, on average, in any given year. Additional EPA analysis supports the claim that states would be no more likely to exceed 1-year variability limits without the 3-year limits than with the 3-year limits. See the “Power Sector Variability Final Rule TSD” for more details on this statistical analysis. Finally, because the state budgets (and thus the total amount of allowances available) are fixed and every covered source must hold allowances covering its emissions, it is not feasible for all, or even many, states to repeatedly exceed their budgets.

The approach calculated the standard deviation in state-level heat input from units expected to be covered by the final Transport Rule over an 11-year time period (2000 through 2010), from which the 95th percent confidence level was calculated. EPA divided this value by the mean to get the percentage variation in heat input. The two-tailed 95th percent confidence level is the equivalent of the 97.5 percent upper (single-tailed) confidence level. This approach yielded an average year-to-year heat input variability for each state, as a proxy for historic year-to-year variability in state-level EGU emissions while holding emission rates constant. The result, expressed as a percentage, conveys the maximum degree to which EGU emissions at the state level may be expected with 95th percent confidence to vary around a given target (i.e., budget) from year-to-year, on average, based on the statistical analysis of historic heat input over the 2000 through 2010 time period.

From the state-by-state variability calculations, EPA identified a single variability level (percentage) for each of the annual and ozone-season programs based on the historic variability measured at units in covered states on an annual basis and an ozone-season basis, respectively. In the proposal, EPA “identified a single set of variability levels * * * to apply to all states in order to make the application of the variability limits straightforward rather than developing state-by-state percentage variability values” (75 FR 45293). In the final rule, EPA is taking the straightforward approach of identifying a single set of variability levels to apply to all states because EPA has determined that it is reasonable to afford all states under the Transport Rule programs the extent of measured historic variability experienced by any Transport Rule state during 2000 through 2010. In the variability analysis for the final rule, EPA identified Tennessee as having the highest measured historic variability of annual heat input of 18 percent, and Virginia as having the highest measured historic variability of ozone-season heat input of 21 percent. Because the percentage of variability in Tennessee on an annual basis and in Virginia on an ozone-season basis are reasonably likely to occur in each of the other states in the future, EPA believes it is appropriate to apply an 18 percent annual variability limit to all states covered by the annual SO₂ and NO_X programs and a 21 percent ozone-season variability limit to all states covered by the ozone-season NO_X program.

EPA's analysis of historic heat input variability in multiple states over the 2000 to 2010 baseline yields a range of potential year-to-year variability values for state-level EGU emissions. As discussed above, any one state's measured variability (in this case, from 2000 to 2010) is due to a multiplicity of factors. These factors include, but are not limited to, variation in weather, variation in demand due to increased or decreased level of economic activity, variation in the portion of electric generation that is fossil-fuel-fired, and variation in the length and number of outages at power generation units, and these individual factors may sometimes act in concert and may other times be offsetting.

The mix and levels of factors present in a state from year-to-year can lead to variation of state-level emissions above and below the level for the state under average conditions. Because the levels of the various factors are difficult to predict on a year-to-year basis for an individual state, the resulting variability in state-level emissions is difficult to predict. Moreover, because the electric generation, transmission, and distribution system in the eastern half of the U.S. is highly integrated, year-to-year variation in these factors in one state can cause year-to-year variability in state-level emissions both in that state and in other states on the system. For example, increased demand due to extreme weather or increased economic activity in one state can be met through increased generation and emissions in a number of states.

Because these factors can vary year-to-year in every state in ways that are difficult to predict and can affect other states, EPA maintains that the maximum variability measured in one state for a discrete period (2000-2010) is reasonably likely to occur in the future in any of the states in the region. Consequently, EPA believes that it is reasonable to use the maximum historic percentage variability figure as a proxy for the percentage variability that any of the states is likely to experience in the future. Although EPA is therefore using a uniform percentage figure for variability, EPA applies that percentage figure to each state-specific budget so that variability in tons of emissions is determined on a state-specific basis. That state-specific number is used in determining whether the assurance provisions and penalty are triggered in the specific state. EPA also believes that it is appropriate to accommodate this potential future variability at the state level if and only if it can be accommodated without undermining the programs' beneficial impacts on downwind air quality that eliminate significant contribution to nonattainment or interference with maintenance of the NAAQS assessed in this rulemaking (see the “Power Sector Variability Final Rule TSD” for more information on this analysis). The Transport Rule identifies and quantifies, on a state-by-state basis, the emissions in each state that significantly contribute to nonattainment or interfere with maintenance in another state. This is done by analyzing specific air pollution linkages between each upwind state and each downwind maintenance or nonattainment receptor. Nonetheless, it is clear from the air quality analyses that the air quality outcome at a given downwind receptor is a function of the cumulative emissions from all upwind states and the receptor's home state. Once the Transport Rule emission reduction requirements are implemented in all states subject to the programs, EPA's analysis shows that the impact on a downwind receptor of any single upwind state's year-to-year fluctuation of up to 20 percent in SO₂ emissions would be so limited as to not disturb that receptor's ability to maintain or attain the NAAQS analyzed in this rulemaking. Therefore, to the extent that such variability has been measured in historic data in any state subject to the Transport Rule programs, it is reasonable to provide for potential future variability in Transport Rule states within the scope of what EPA's analysis shows to preserve downwind air quality gains achieved by the Transport Rule programs.

The approach to establishing variability limits in the final rule modifies the approach from the proposed rule in two ways. First, EPA is applying only a percentage variability limit to each budget in the final rule, whereas the proposed rule applied the greater of a percentage or an absolute tonnage variability limit to each budget. EPA explained in the proposal that it was necessary to impose both a percentage and a tonnage limit due to the inclusion of “states with small numbers of units where expected variability would be more pronounced in percentage terms” (75 FR 45293). However, the states with the smallest numbers of units included at proposal (such as Connecticut and the District of Columbia) are not covered by any of the final Transport Rule's programs. In the final rule's variability analysis, Tennessee has the highest measured annual variability percentage and Virginia has the highest measured ozone-season variability percentage. Both of these states have a sufficient number of units for the percentage variability findings to be representative of variability in all of the Transport Rule states; therefore, it is not necessary to impose a tonnage limitation in the final rule.

Second, EPA has expanded the historic baseline of the variability analysis to consider heat input data from 2000 through 2010, as compared to 2002 through 2008 at proposal, and EPA has also expanded the dataset to include all units expected to be covered by the final Transport Rule's programs. EPA received a number of comments that the proposal's variability limits were too stringent in part because they relied on too short a historical baseline that failed to capture the full extent of long-run year-to-year variability. EPA agrees with these comments and believes that the historic baseline modification described above supports variability limits in the final rule that are a better approximation of future potential year-to-year variability in state-level EGU emissions around the budgets as a function of inherent variability in baseline state-level EGU operations. EPA believes the 2000 through 2010 historic baseline supports a more accurate approximation of year-to-year variability in state-level EGU operations than previously measured on a 2002 through 2008 baseline.

Some commenters expressed the view that allowing variability limits in addition to state budgets undermines the requirements of CAA section 110(a)(2)(D)(i)(I) to eliminate significant contribution to nonattainment and interference with maintenance of the NAAQS in downwind states. EPA disagrees with these comments. As explained above, EPA finds that year-to-year variability is an inherent characteristic of power sector emissions whether or not such emissions are controlled by state budgets; the future year-to-year variability is a component of the sector's emissions baseline before emission reductions are required. As done for proposal, EPA has analyzed the impact of allowing emissions from upwind states in a given year to rise above the budgets but within the variability limits allowed in the final rule. This analysis shows that emission fluctuations around the budgets but within the variability limits will not undermine the downwind air quality gains achieved by the implementation of the Transport Rule budgets, and therefore the variability limits cannot be said to undermine the elimination of significant contribution to nonattainment or interference with maintenance achieved under the Transport Rule programs. Based on historical data and projected air quality impacts, the Agency believes that states will have sufficient flexibility and room to operate within the final rule's variability limits while addressing all emissions identified as significantly contributing to nonattainment or interfering with maintenance in other states.

F. Variability Limits and State Emission Budgets: State Assurance Levels

As explained above, EPA applied the variability levels on a state-by-state basis to calculate specific emission budgets with variability limits. The state budget plus the variability limit is also called the “state assurance level.” Table VI.F-1 shows final state budgets, variability limits, and assurance levels by state for SO₂ emissions. Table VI.F-2 shows final state budgets, variability limits, and assurance levels by state for annual NO_X emissions. Table VI.F-3 shows final state budgets, variability limits, and assurance levels by state for ozone-season NO_X emissions.

See section VII.E for the discussion of how variability limits and state assurance levels are used in the implementation of assurance provisions for the air quality-assured trading programs.

G. How the State Emission Reduction Requirements Are Consistent With Judicial Opinions Interpreting the Clean Air Act

The methodology described in this notice quantifies states' significant contribution to nonattainment and interference with maintenance in a manner that is consistent with the decisions of the DC Circuit. As discussed previously, the DC Circuit has issued two significant decisions addressing the requirements of 110(a)(2)(D)(i)(I). The first opinion largely upheld the NO_X SIP Call, Michigan, 213 F.3d 663, and the second found significant flaws in CAIR, North Carolina, 531 F.3d. 896. In both cases, the Court considered aspects of the methodology used by EPA to identify emissions that, pursuant to section 110(a)(2)(D)(i)(I), must be eliminated due to their impact on air quality in downwind states. EPA believes that the methodology used in this final rule is consistent with both opinions and rectifies the flaws the North Carolina court identified with the methodology used in CAIR. The methodology used for this rule relies on state-specific data to analyze each individual state's significant contribution, uses air quality considerations in addition to cost considerations to identify each state's significant contribution, and gives independent meaning to the “interference with maintenance” prong. This methodology is then applied in a reasonable manner consistent with the relevant judicial opinions.

In North Carolina, the Court held that EPA's approach to evaluating significant contribution was inadequate because, by evaluating only whether emission reductions were highly cost effective “at the regional level assuming a trading program”, it failed to conduct the required state-specific analysis of significant contribution. See id. at 907. EPA, the Court concluded, “never measured the `significant contribution' from sources within an individual state to downwind nonattainment areas.” Id. The Court did not, however, disturb the air-quality-based methodology used by EPA to identify the states with contributions large enough to warrant further consideration.

For this rule, EPA uses a first step similar to that used in CAIR to identify the states with relatively large contributions. However, in contrast to CAIR, it then uses a state-specific analysis. Instead of identifying a single emission level that could be achieved by the application of highly cost effective controls in the region, EPA determines, on a state-by-state basis, what reductions could effectively be achieved by sources in each state. EPA's new approach does not, as the CAIR methodology did, establish a regional cap on emissions that is then divided into state budgets that set the emission reduction requirements for each state. Instead, EPA develops, for each covered state, emission budgets based on the reductions achievable at a particular cost per ton in that particular state, taking into account the need to ensure reliability of the electric generating system. The selected cost/ton levels reflect consideration of both cost factors and air quality factors including the estimated impact of upwind states' emissions on each downwind receptor.

In addition, in developing this approach, EPA was guided by the Court's holdings regarding the use of cost to identify significant contribution. Specifically, the Court held in Michigan that EPA could “in selecting the `significant' level of `contribution' under section 110(a)(2)(D)(i)(I), choose a level corresponding to a certain reduction in cost.” North Carolina, 531 F.3d at 917 (citing Michigan, 213 F.3d at 676-77). This holding also supported the Court's conclusion in Michigan that it was acceptable for EPA to apply a uniform cost-criterion across states. See Michigan, 213 F.3d at 679. In the CAIR case, the Court rejected EPA's analysis, not because it relied on cost considerations to identify significant contribution, but because it found that EPA had failed to draw the significant contribution line at all. See North Carolina, 531 F.3d at 918 (“* * * here EPA did not draw the [significant contribution] line at all. It simply verified sources could meet the SO₂ caps with controls EPA dubbed `highly cost-effective.' ”). The holdings in Michigan regarding the use of cost and a uniform cost-criterion across states were left undisturbed. See, e.g., North Carolina, 531 F.3d at 917 (explaining that in Michigan the Court held that “EPA may `after [a state's] reduction of all [it] could * * * cost-effectively eliminate[],' consider `any remaining contribution insignificant”). In fact, the Court acknowledged that, based on the Michigan holdings, the measurement of a state's significant contribution need not “directly correlate with each state's individualized air quality impact on downwind nonattainment relative to other upwind states.” North Carolina, 531 F.3d at 908.

For these reasons, EPA determined that it was appropriate in this rulemaking to consider the cost of controls to determine what portion of a state's contribution is its “significant contribution.” However, EPA also heeded the North Carolina Court's warning that “EPA can't just pick a cost for a region, and deem `significant' any emissions that sources can eliminate more cheaply.” North Carolina,, 531 F.3d at 918. Thus, in this rulemaking, EPA departs from the practice used in the NO_X SIP Call and in CAIR of evaluating, based solely on the cost of control required in other regulatory environments, what controls would be considered “highly-cost-effective.” Instead, as part of its determination of a reasonable cost per ton for upwind state control, EPA evaluates the air quality impact of reductions at various cost levels and considers the reasonableness of possible cost thresholds as part of a multi-factor analysis.

In addition, the methodology used in this rulemaking gives independent meaning to the interfere with maintenance prong of section 110(a)(2)(D)(i)(I). In North Carolina, the Court concluded that CAIR improperly “gave no independent significance to the `interfere with maintenance' prong of section 110(a)(2)(D)(i)(I) to separately identify upwind sources interfering with downwind maintenance.” North Carolina, 531 F.3d at 910. EPA rectified this flaw in this rulemaking by separately identifying downwind “nonattainment sites” and downwind “maintenance sites.” EPA decided to consider upwind states' contributions not only to sites that EPA projected would be in nonattainment, but also to sites that, based on the historic variability of their emissions, EPA determined may have difficulty maintaining the relevant standards. The specific mechanism EPA used to implement this approach is described in detail in section V.C, previously. For annual PM_2.5, this approach identified 16 maintenance sites in addition to the 32 nonattainment sites identified in the analysis of nonattainment receptors. For 24-hour PM_2.5 this approach identified 38 maintenance sites in addition to the 92 nonattainment sites identified in the analysis of nonattainment receptors. For ozone it identified 16 maintenance sites in addition to the 11 ozone nonattainment sites identified.

EPA applied this methodology using available information and data to measure the emissions from states in the eastern United States that significantly contribute to nonattainment or interfere with maintenance in downwind areas with regard to the 1997 and 2006 PM_2.5 NAAQS and the 1997 ozone NAAQS. Although EPA has not completely quantified the total significant contribution of these states with regard to all existing standards, EPA has determined, on a state-specific basis, that the emissions prohibited in the FIPs are either part of or constitute the state's significant contribution to nonattainment and interference with maintenance. Thus, elimination of these emissions will, at a minimum, make measurable progress towards satisfying the section 110(a)(2)(D)(i)(I) prohibition on significant contribution to nonattainment and interference with maintenance.

VII. FIP Program Structure To Achieve Reductions

A. Overview of Air Quality-Assured Trading Programs

EPA is finalizing an air quality-assured trading remedy that is substantially similar to the preferred trading remedy presented in the proposal. Key differences from the preferred trading remedy in the proposal include:

• Recalculated state budgets and variability limits (i.e., state assurance levels) based on updated modeling;
• Simplified variability limits for 1-year application only;
• Revised allocation methodology for existing and new units and revised new unit set-asides for new units in Transport Rule states and new units potentially locating in Indian country;
• Changed start of assurance provisions to 2012 and increased assurance provision penalties; and
• Removed opt-in provisions.

In the final rule, as in the proposed rule, EPA is promulgating FIPS to require SO₂ and NO_X reductions from power plants in jurisdictions that contribute significantly to nonattainment in, or interfere with maintenance by, a downwind area with respect to the 1997 ozone NAAQS, the 1997 annual PM_2.5 NAAQS, and/or the 2006 24-hour PM_2.5 NAAQS. These FIPs establish state-specific emission control requirements using state budgets starting in 2012, with a second phase of SO₂ reductions in some states in 2014. Section IV explains EPA's authority to issue FIPs.

The air quality-assured trading remedy in the final rule allows interstate trading to account for variability in the electricity sector, but also includes assurance provisions to ensure that the necessary emission reductions occur within each covered state. The assurance provisions restrict EGU emissions within each state to the state's budget plus the variability limit and ensure that every state is making reductions to eliminate the significant contribution to nonattainment and interference with maintenance that EPA has identified. While EPA proposed to impose these assurance provisions starting in 2014, the final rule implements these provisions starting in 2012 (see section VII.E of this preamble). Additionally, the final FIPs include penalty provisions adequate to ensure that the state budget with the variability limit will not be exceeded.

In the final rule, as in the preferred trading remedy discussed in the proposed rule, state-specific emission budgets without the variability limits are used to determine the number of emission allowances allocated to sources in each state. An EGU source is required to hold one SO₂ or one NO_X allowance, respectively, for every ton of SO₂ or NO_X emitted during the control period. Banking of allowances for use or trading in future years is allowed.

The final rule establishes four interstate trading programs, each starting in 2012: two for annual SO₂, one for annual NO_X, and one for ozone-season NO_X. One SO₂ trading program is for sources in states (referred to as SO₂ Group 1) that need to make larger reductions to eliminate their significant contribution, while the second is for sources in states (referred to as SO₂ Group 2) that need to make smaller reductions. A source in a Group 1 state can only use SO₂ allowances allocated to Group 1 states for compliance with the SO₂ trading program. A source in a Group 2 state can only use SO₂ allowances allocated to Group 2 states for compliance with the SO₂ trading program. For compliance in the annual NO_X and ozone-season NO_X trading programs respectively, sources may use annual NO_X and ozone-season NO_X allowances allocated for any state, even if that state is in a different group for SO₂ than the source's state. Four sets of new emission allowances based on the new state-specific budgets without variability are allocated to sources, one set for each of the four trading programs. Each state has the option of replacing these FIPs with state rules. EPA believes that this remedy meets the concerns raised by the Court in the 2008 North Carolina decisions which remanded CAIR to EPA.

In the proposed rule, EPA took comment on all aspects of the preferred trading remedy and on two alternative regulatory options: (1) intrastate trading; and (2) direct control. EPA also took comment on a trading ratios approach.

Comments on the Preferred Trading Remedy: The great majority of public comments supported the preferred trading remedy. Most of these commenters voiced their support for the broadest possible trading mechanism because it allows for the most cost-effective implementation of any emission controls. Commenters noted that flexibility is always needed in the early years of new programs. Further, commenters favoring the preferred remedy agreed with EPA that, by using state-specific control budgets and allowing for interstate trading, the preferred remedy provided electricity generators the flexibility to undertake the most cost-effective reductions while assuring that the resulting reductions occur within the individual states.

Some commenters that supported the preferred remedy felt that, while not ideal, the interstate trading remedy was preferable to the alternative options of intrastate trading or direct control. Many commenters that supported the preferred remedy felt that the intrastate trading remedy and direct control remedy options offer minimal flexibility from a compliance perspective. They stated that this lack of flexibility would unnecessarily increase the cost of emission reductions.

Other commenters who generally support the preferred remedy cited concerns about the level of complexity in the assurance provisions. One commenter surmised that the preferred option creates significant risk where a company could unexpectedly find itself in a noncompliance situation due to the after-the-fact variability analysis. Another said that the rule's features needlessly reduce the system's efficiency and increase complexity. These commenters generally preferred unlimited trading, noting that EPA has proven success with Title IV, the NO_X SIP Call, and CAIR unlimited interstate trading programs and that allowing unrestricted interstate trading would increase flexibility to meet reduction goals and minimize increases in power costs.

EPA is finalizing the preferred trading remedy for the following reasons. EPA believes this approach is the most cost-effective and practical way to comply with the Court decision in North Carolina to ensure that all emissions in a given state that EPA has identified as significantly contributing to downwind nonattainment or interfering with maintenance are eliminated. The vast majority of public commenters agree. In addition, this approach provides the most flexibility for sources while meeting the Clean Air Act requirements and protecting public health. As a result, potential innovations and resulting cost savings are more likely to be found and implemented. Based on historical experience (see the Transport Rule proposal, 75 FR 45315), EPA has shown that the results offered by a flexible trading approach (e.g., flexible compliance choices, incentives to reduce emissions early and in the highest emitting areas, 100 percent compliance with requirements) are substantial. A large number of commenters have corroborated this assessment. As summarized in the proposal, EPA believes that the preferred trading remedy will allow source owners to choose among several compliance options to achieve required emission reductions in the most cost-effective manner, such as installing controls, changing fuels, reducing utilization, buying allowances, or any combination of these actions. Interstate trading with assurance provisions provides additional regulatory flexibility that promotes the power sector's ability to operate as an integrated, interstate system and to provide electric reliability.

Comments on Intrastate Trading: A few commenters favored the first alternative, intrastate trading. One commenter who favored intrastate trading stated that many power plants have avoided investment in pollution controls by buying allowances from other plants, affecting local air quality improvement. EPA notes that this Transport Rule aims to address emissions from one state that significantly contribute to nonattainment or interfere with maintenance of certain NAAQS in other states. Local air quality issues are directly addressed by other provisions in the Clean Air Act.

Several commenters raised concerns about the intrastate trading approach. Some stated, as EPA noted in the proposal, that the intrastate trading option would be more resource intensive, more complex, less flexible, and potentially more susceptible to market manipulation than the other options. In addition, some commenters felt that this alternative would provide less flexibility to ensure electric reliability than the preferred approach, resulting in greater private costs to the power sector and greater social costs for consumers.

EPA is not finalizing the intrastate trading option for the following reasons. As EPA expressed in the proposal and as commenters have agreed, the intrastate trading option would be more resource intensive (both for EPA and for sources), more complex, less flexible, and potentially more susceptible to market manipulation than the preferred trading approach that EPA is finalizing. The intrastate trading option would be more costly and less transparent due to the large number of trading programs that would be operated simultaneously and the large number of annual auctions that would be held every year to address the issues of market power within states. This option would also result in a greater burden for participants operating EGUs in multiple states.

Comments on Direct Control Option: Several commenters favored the second alternative, direct control. One commenter stated that direct control—allowing no trading—was the option best aligned with the 2008 Court decisions. EPA disagrees with this comment for the reasons given below and because, as explained in this rule, EPA believes the air quality-assured trading remedy finalized today is consistent with the decisions of the DC Circuit in North Carolina.

Some commenters, who support direct control, voiced concerns that the other emission trading approaches would disadvantage poor and minority communities or allow increased emission impacts in neighborhoods near power plants. EPA notes that a direct control approach would not require controls on all plants in a state, but only on a sufficient number to address the transport requirements under section 110(a)(2)(d)(i)(I) that this rule addresses, and therefore would not necessarily mandate controls on each neighborhood power plant.

In addition, EPA has conducted an analysis of the effects of the Transport Rule on environmental justice and other vulnerable communities. We concluded that, similar to our experience with the Acid Rain Program, many environmental justice communities are expected to see large health benefits, and none are expected to experience any disbenefits, from implementing an air quality-assured trading program. The results of this analysis are presented in section XII of this preamble and Chapter 5 of the RIA for this rule. In addition, the CAA provides flexibility for state and local authorities to impose stricter limits on sources to address specific local air quality concerns. Such limits are independent of the requirements in this rule, and compliance with Transport Rule requirements in no way excuses a source from complying with other CAA or state law requirements.

Several commenters raised concerns with the direct control approach. One commenter felt that issues with electricity market reliability could occur during high electricity demand periods if sources ceased operations due to approaching their emission rate limitations under a direct control remedy. Another commenter was concerned that applying emission rates under a direct control remedy to small municipal units would cause disproportionate impacts on power plants where pollution control is more expensive. Other commenters cited concerns that EPA's proposed within-state company-wide averaging provision in the direct control proposed alternative (designed to allow some flexibility for sources) would place companies with fewer units at a disadvantage compared to companies with more units. EPA generally agrees with the commenters concerns and has decided not to finalize the direct control remedy for the following reasons. EPA modeling projects that the direct control alternative would result in fewer emission reductions and higher costs compared to the air quality-assured trading remedy. EPA analysis indicates that it is not necessary to implement a direct control approach in order to protect vulnerable and sensitive populations or environmental justice communities. Also, the direct control approach would result in fewer compliance options because a direct control approach would directly regulate individual sources by setting unit-level emission rate limits. This lack of flexibility could lead to potential increases in reliability risks in the electric power system and fewer opportunities for potential technological innovations that reduce emissions further and/or lower costs. For these reasons, EPA believes that this approach is inferior to the air quality-assured trading remedy.

Other Comments: A handful of commenters mentioned the trading ratios approach, though none favored it as a viable alternative. One commenter said the trading ratios approach was not consistent with CAA section 110(a)(2)(D) requirements that reductions in emissions occur in particular geographic locations. Other commenters agreed that it was administratively unworkable and would be difficult to implement due to the complexity and variety of meteorological conditions. EPA generally concurs with the commenters. In the proposal, EPA noted that it would not be possible under this approach, as contemplated, to include enforceable legal requirements to ensure that a specific state's emissions remain below a specified level or to ensure that a specific amount of reductions occur within a particular state. EPA specifically requested comment on whether a ratios trading program could be designed to provide such legal assurances. Of the few comments received, none offered such a solution. For these reasons, EPA is not finalizing this approach.

Some commenters offered additional suggestions, such as: unrestricted trading; using different authorities in the CAA to address interstate transport such as section 110(k)(5) and section 126; and an approach that would replace the assurance provisions by a system using both emission allowances usable (as well as bankable) in any state and assurance allowances usable (but not bankable) in only the state for which they would be issued. While EPA appreciates the thoughtful and constructive comments, we did not find any of these suggestions improved our ability to address interstate transport under CAA section 110(a)(2)(D)(i)(I), in line with the Court decision, in an administratively practical way.

Several commenters liked the idea of establishing unit-by-unit short-term and long-term performance standards/emission rates but suggested adding an overlaid cap and trade program. EPA believes the air quality-assured trading remedy finalized today is consistent with the decisions of the Court in North Carolina and will ensure the reductions necessary to meet statutory requirements.

For the 2012-2013 period, EPA took comment on whether the assurance provisions are needed, since the state-specific budgets would be based on known air pollution controls and the penalty provisions would be adequate to ensure that the budget, including a variability limit, would not be exceeded. Further, EPA proposed to use two variability limits: a 1-year limit, based on the year-to-year variability in emissions relative to the proposed budgets; and a 3-year limit based on the variability in a 3-year average relative to the proposed budget.

Based on comments on the assurance provisions (see section VII.E of this preamble) and variability limits (see section V I.E. 2 of this preamble), EPA is finalizing the Transport Rule with state budgets plus variability limits and assurance provisions starting in 2012 for all of the trading programs. EPA sees an immediate need to ensure that emissions within a state do not exceed the state budget plus the variability limitation in order to comply with the Court's opinion. Further, commenters stated that the 3-year variability limit increased costs and unnecessarily complicated the trading programs. As explained in section VI.E.2, EPA is finalizing the 1-year variability limit starting in 2012, but not the 3-year limit.

B. Applicability

The applicability provisions in the final rule are, except as discussed herein, essentially the same as in the proposed rules and for each of the Transport Rule trading programs.

Under the general applicability provisions of the proposed rule, the Transport Rule trading programs would cover fossil-fuel-fired boilers and combustion turbines serving—on any day starting November 15, 1990 or later—an electrical generator with a nameplate capacity exceeding 25 MWe and producing power for sale, with the exception of certain cogeneration units and solid waste incineration units.

EPA requested comment on whether a more recent year should be used instead. The proposed use of the November 15, 1990 date was consistent with the use of 1990 as the beginning of the historical period for which owners and operators would generally be required to have information about their units for purposes of determining whether the units were covered by the Transport Rule trading programs. Because unit information is generally compiled and retained on a calendar year basis, EPA believes that, for the general applicability provisions, it is preferable to use January 1, rather than November 15. In determining which year should be used as the reference year in the general applicability provisions, EPA considers several factors.

First, in order for owners and operators, and EPA, to be able to determine which units are subject to the Transport Rule trading programs, EPA believes that the reference year should not be so far in the past that the unit information necessary to make applicability determinations is not readily available. This particularly becomes an issue in cases of older units that have changed ownership over time. EPA found, in making some applicability determinations under the CAIR trading programs, that some older units with ownership changes had difficulty obtaining information back as far as twenty or more years. Using January 1, 1990 as the reference date in the general applicability provisions could effectively require some owners and operators to retain unit information going back as far as 20 years. As a point of contrast, under the title V permitting rules, owners and operators are generally required to retain data for 5 years. See 40 CFR 70.6(a)(3)(B).

Second, EPA also believes that the reference year used in the applicability provisions should be far enough in the past that the unit information on which applicability determinations are based provides a full picture of the nature of the unit and its operations over time, such as the types of fuels combusted at the unit and whether the unit has produced electricity for sale.

Third, EPA considers whether selecting a different reference year for the applicability provisions than the one in the proposed rule dramatically changes what units will be covered by the Transport Rule trading programs. In this case, EPA believes, based on available information about the units potentially subject to the Transport Rule, that using a somewhat later year than the one in the proposed rule will likely have little effect on what units are covered. Balancing these factors, EPA concludes that it is reasonable to use January 1, 2005, rather than November 15, 1990, in the general applicability provisions in the final rule.

In the final rule, EPA is taking the same approach with regard to defining whether a boiler or combustion turbine is considered to be “fossil-fuel-fired” as the one used in the proposal. Under the proposed rule, a unit was considered to be “fossil-fuel-fired” if it combusts any amount of fossil fuel at any time in 1990 or later. For the same reasons that EPA decided to use January 1, 2005 in the general applicability provisions, and in order to have a consistent reference year in all applicability-related provisions, the final rule defines a “fossil-fuel-fired” unit as one that combusts any amount of fossil fuel in 2005 or later.

EPA notes that the final Transport Rule allows a state to submit a SIP revision (an abbreviated or full SIP) under which the state may—in addition to making certain types of changes concerning allowance allocations in the Transport Rule trading programs—expand the general applicability provisions of the Transport Rule NO_X Ozone Season Trading Program to cover fossil-fuel-fired boilers and combustion turbines serving—at any time starting January 1, 2005 or later— a generator with a nameplate capacity as low as 15 MWe producing power for sale. The exemptions, discussed below, for cogeneration units and solid waste incineration units still will continue to apply.

Cogeneration unit exemption. Under the final rule (as well as the proposed rule) certain cogeneration units or solid waste incinerators are exempt from the FIP requirements. In particular, the final rule includes an exemption for a unit that qualifies as a cogeneration unit throughout the later of 2005 or the first 12 months during which the unit first produces electricity and continues to qualify through each calendar year ending after the later of 2005 or that 12-month period and that meets the limitation on electricity sales to the grid. In order to meet the definition of “cogeneration unit” in the final rules, a unit (i.e., a fossil-fuel-fired boiler or combustion turbine) must be a topping-cycle or bottoming-cycle that operates as part of a “cogeneration system,” which is defined as an integrated group of equipment at a source (including a boiler, or combustion turbine, and a steam turbine generator) designed to produce useful thermal energy for industrial, commercial, heating, or cooling purposes and electricity through the sequential use of energy. A topping-cycle unit is a unit where the sequential use of energy results in production of useful power first and then, through use of reject heat from such production, in production of useful thermal energy. A bottoming-cycle unit is a unit where the sequential use of energy results in production of useful thermal energy first and then, through use of reject heat from such production, in production of useful power. In order to qualify as a cogeneration unit, a unit also must meet certain efficiency and operating standards.

In the proposed rule, a unit would have to qualify as a cogeneration unit and meet the limitation on electricity sales starting the later of 1990 or the year when the unit begins operating. EPA requested comment on whether a more recent year should be used. For the reasons discussed above concerning the reference year used in the general applicability provisions and in order to have a consistent reference year in all applicability-related provisions, EPA concludes that it is reasonable to use 2005, rather than 1990, in the cogeneration unit exemption provisions in the final rule. Consequently, the final rule provides that the requirements to qualify as a cogeneration unit and to meet the electricity sales limitation start no earlier than 2005.

In the final rule, EPA also clarifies that the electricity sales limitation under the exemption is applied in the same way whether a unit serves only one generator or serves more than one generator. In both cases, the total amount of electricity produced annually by a unit and sold to the grid cannot exceed the greater of one-third of the unit's potential electric output capacity or 219,000 MWhr. This is consistent with the approach taken in the Acid Rain Program (40 CFR 72.7(b)(4)), where the cogeneration unit exemption originated. EPA believes that this clarification is needed to ensure that a unit serving, for example, two generators would not have a limit on sales of electricity to the grid that would be different (i.e., twice as high) from the limit for a unit serving only one generator with the same total nameplate capacity as the first unit's two generators.

EPA also took comment on whether efficiency standards should be applied on a system-wide basis to bottoming-cycle units (where useful thermal energy is produced before useful power is produced), as they are for topping-cycle units (where useful thermal energy is produced after useful power) and whether to exclude, from the requirement to meet the operating and efficiency standards, calendar years during which a cogeneration unit does not operate at all. Several commenters argued EPA should apply efficiency standards to both types of units. EPA agrees that applying efficiency standards on a system-wide basis to both bottoming-cycle and topping-cycle units is reasonable because EPA sees no technical reason to distinguish between the two types of units in this instance. EPA further agrees with commenters that excluding calendar years in which the cogeneration unit does not operate at all, i.e., does not combust any fuel, from the requirements to meet operating and efficiency standards is also reasonable. For such a year, the unit would not produce any useful thermal energy or useful power and therefore could not meet the minimum output requirements in the operating and efficiency standards, but the unit also would not have any emissions. For these reasons, the final rule expressly provides that the operating and efficiency standards do not have to be met for a calendar year throughout which a unit did not operate at all.

Solid waste incineration unit exemption. The final rule also includes an exemption for a unit that qualifies as a solid waste incineration unit during the later of 2005 or the first 12 months during which the unit first produces electricity, that continues to qualify throughout each calendar year ending after the later of 2005 or that 12-month period each year thereafter, and that meets the limitation on fossil-fuel use. In contrast, the exemption for solid waste incineration units in the proposed rule distinguished between units commencing operation before January 1, 1985 and those commencing operation on or after that date. A unit commencing operation before January 1, 1985 would be exempt if it qualified as a solid waste incineration unit starting the later of 1990 or the year when it began producing electricity and its average annual fuel consumption of non-fossil fuels exceeded 80 percent of total heat input during 1985-1987 and during any three consecutive calendar years after 1990. A unit commencing operation on or after January 1, 1985 would be exempt if it qualified as a solid waste incineration unit starting the later of 1990 or the year when it began producing electricity and its average annual fuel consumption of non-fossil fuel exceeded 80 percent of total heat input for the first 3 calendar years of operation and for any 3 consecutive calendar years thereafter.

In the proposal, EPA requested comment on whether it would be problematic to obtain sufficiently detailed information about unit operation potentially as far back as 1985-1987 and 1990, and whether the fuel consumption standard for each unit should be limited to more recent years. For the reasons discussed above concerning the reference year used in the general applicability provisions and in order to have a consistent reference year for all applicability-related provisions, EPA concludes that it is reasonable to use 2005, rather than 1990, in the solid waste incineration unit exemption in the final rule. In particular, EPA notes that the proposed provisions for units commencing operation before January 1, 1985 and for units commencing operation on or after January 1, 1985 could require some owners and operators to retain unit information going back more than 20 years before the promulgation of this final rule. Further, EPA believes that removing the distinction between units commencing operation during these two periods, and referencing somewhat later years as the earliest years for which information on fossil-fuel consumption is required, will result in the exemption still being based on sufficient data to provide a full picture of the nature and operation of the units involved. EPA also believes, based on available information about the units potentially subject to the Transport Rule, that this approach will not significantly change which units qualify for the exemption. Consequently, the final rule removes the distinction based on whether a solid waste incineration unit commences operation before January 1, 1985 or on or after January 1, 1985. In order to be exempt, the unit must qualify as a solid waste incineration unit during the later of 2005 or the first 12 months during which the unit first produces electricity, must continue to qualify throughout each calendar year ending after the later of 2005 or that 12-month period, and must meet the limitation on fossil-fuel use on a 3-year average basis during the first 3 years of operation starting no earlier than 2005 and every 3 years of operation thereafter.

Opt-in units. EPA is not finalizing the opt-in provisions that were discussed in the Transport Rule proposal. EPA proposed opt-in provisions to allow non-covered units to voluntarily opt in to any of the proposed Transport Rule trading programs and receive allocations reflecting 70 percent of the unit's emissions before opting in. These allowances were above the state-specific budgets developed under the Transport Rule to eliminate a state's significant contribution to nonattainment and interference with maintenance. In theory, an opt-in unit that makes reductions below its baseline and sells the freed-up allowances is effectively substituting its new, lower-cost reductions for higher-cost reductions otherwise required by a covered EGU, with the result that the state's significant contribution is still eliminated but at a lower total program cost.

EPA notes that theoretical benefits anticipated from allowing opt-ins did not materialize in prior trading programs with opt-in provisions. The Acid Rain Program has about 23 opt in units; the NO_X Budget Trading Program had five opt-in units; and no units opted into the CAIR programs. As a group, these opt-in units neither eased the achievement of required emission reductions in past trading programs, nor reduced overall program costs.

In the proposal, EPA requested comment on the opt-in provisions, specifically regarding: What are the benefits of and concerns about including opt-in provisions; how to ensure units are not credited for emission reductions the units would have made anyway; whether the proposed 30 percent reduction (i.e., application of the 70 percent multiplier to baseline emissions) or some other percentage reduction, or no reduction, should be applied to the baseline emission rate used in determining allocations; and whether any additional percentage reduction (such as 45 percent) should be applied to SO₂ Group 1 opt-in units in Phase II to reflect the stricter limits for covered units.

Some commenters argued that increasing the Transport Rule budgets for opt-ins would undermine the goal of CAA section 110(a)(2)(D)(i)(I) to eliminate a state's significant contribution to nonattainment and interference with maintenance. One commenter stated that it does not favor allowing sources that are not subject to the emission reduction requirements to be issued allowances that would increase the overall state emission budgets, due to the uncertainty that any reductions made by such units would be surplus, verifiable, permanent and enforceable. This could compromise the integrity of the EGU emission reduction requirements of the Transport Rule and jeopardize assurance that a state's significant contribution would be eliminated, as required by the Court in North Carolina. Other commenters claim that, while no cheap tons are available from non-EGUs and EPA is right not to require non-EGU reductions, EPA should nonetheless allow non-EGUs to choose voluntarily to be covered by opting in.

As mentioned previously, the final Transport Rule does not include any opt-in provisions either in the FIPs or in the provisions allowing modification or replacement of the FIPs through submission of trading program provisions in SIPs. EPA has several reasons for not adopting provisions to allow opt-in units. First, as mentioned above, historically, very few units have opted in. As of 2010, 28 units out of more than 4,700 covered units (23 units out of a total of about 3,600 covered units in the Acid Rain Program and 5 units out of a total of about 2,600 covered units in the NO_X SIP Call) have opted in to EPA trading programs over the past 15 years. In the Acid Rain Program, 3 of the units opted in and then, effective for 2005, opted out. Four of the units opted in, immediately shut down, and continue to receive allowance allocations. Four of the units opted in and continue to operate and receive allowance allocations. Finally, 12 of the units opted in, after CAIR was finalized, in order to receive allowances usable for compliance in the CAIR SO₂ trading program. Because CAIR will be replaced by this Transport Rule, EPA anticipates that these 12 units will opt out of the Acid Rain Program. In the NO_X Budget Trading Program, 3 plants with 5 opt-in units received allocations between 2003 and 2008.

Moreover, EPA has determined that the inclusion of opt-in units in the Transport Rule trading programs would undermine the rule's objective of addressing emissions in each state that significantly contribute to nonattainment or interfere with maintenance in other states. As explained above, EPA has established budgets plus variability limits that states must meet to ensure that the significant contribution to nonattainment and interference with maintenance identified by EPA is addressed. If EPA were to allow opt-ins, and if any opt-in unit were to receive an allocation of allowances for emissions that would be reduced even if the units did not opt in, then the inclusion of that opt-in unit in the program would allow the sources covered by the Transport Rule to emit in excess of the budget plus variability limit with no new, offsetting reduction in emissions. For example, after a unit would opt in, process or fuel changes made for economic reasons (rather than due to any regulatory requirements), or installation of new emission controls or fuel-switching conducted to meet future, non-Transport Rule regulatory requirements, could result in emission reductions that would have occurred “anyway” (i.e., even if the unit had not opted in), and the opt-in unit would be allocated allowances for the portion of its baseline emissions that would be removed by these “anyway” reductions. Allocations above the cap to opt-in units making “anyway” emission reductions would convert these reductions into extra allowances (i.e., authorizations to emit) usable by covered EGUs to meet their requirements to hold allowances for emissions. Because the extra EGU emissions authorized by these extra allowances would not be offset by any new emission reductions by the opt-in units, this could threaten a state's ability to eliminate the significant contribution to nonattainment and interference with maintenance identified by EPA in the final rule. Also, opt-in units, which are allocated allowances outside the state budget for covered units, could increase the possibility that a state's total emissions would exceed the state budget plus variability and thus that the assurance provisions would be triggered.

This problem of allocating allowances for emissions that would have been reduced anyway is illustrated by the recent promulgation of the final rule, National Emission Standards for Hazardous Air Pollutants for Major Sources: Industrial, Commercial, and Institutional Boilers and Process Heaters (76 FR 15608 (March 21, 2011)) (“final Boiler MACT rule”), which requires certain industrial, commercial, and institutional boilers to meet maximum achievable control technology (MACT) standards for emissions of specified hazardous air pollutants, such as hydrogen chloride (HCL) and mercury (Hg). Some of the control technologies that can be used to meet these standards will also provide significant reductions of SO₂ emissions. For example, a boiler may use a wet scrubber or the combination of a dry sorbent injection system and a fabric filter (among other options) to meet the applicable HCL standard or may use a wet scrubber or a combination of activated carbon injection and a fabric filter (among other options) to meet the applicable Hg standard. See 76 FR 15614 (describing testing and compliance requirements when such controls are used to meet these standards); and Memo from Brian Shrager to Amanda Singleton and Graham Gibson, Revised Methodology for Estimating Cost and Emissions Impacts for Industrial, Commercial and Institutional Boilers and Process Heaters National Emissions Standards for Hazardous Air Pollutants—Major Source (February 11, 2011), Document ID EPA-HQ-OAR-2009-0491-4036 (section 3.1, describing control options for HCL and Hg control). In fact, EPA estimated that the new standards would result in emission reductions of not only the hazardous air pollutants directly subject to the standards, but also in other air pollutants such as SO_2. Specifically, EPA projected that compliance with the final Boiler MACT rule standards will result in about 431,000 tons of annual SO₂ reductions from existing boilers subject to the final Boiler MACT rule. This will comprise on average about a 46 percent reduction in SO₂ emissions for this group of boilers. Coal- and oil-fired boilers—which are the boilers likely to have the most uncontrolled SO₂ emissions and so would be the most likely types of units to consider opting into the Transport Rule trading programs if opting-in were allowed—are projected to reduce about 409,000 tons of annual SO₂ as a result of complying with the final Boiler MACT rule, or about a 50 percent reduction in SO₂ emissions. See Memo from Brian Shrager to Amanda Singleton and Graham Gibson, Appendix B-1, (where column CE represents baseline SO₂ emissions and column CH represents SO₂ reductions resulting from the final Boiler MACT rule compliance). The amount of offsetting SO₂ increases projected to result from final Boiler MACT rule compliance, e.g., from additional fuel being combusted to generate electricity to operate emission controls, is minor. See 76 FR 15651 (Table 4) and 15653 (showing projected total SO₂ reductions for all boilers and process heaters of about 442,000 tons and net SO₂ reductions of about 440,000 tons).

Consequently, a boiler subject to the final Boiler MACT rule may install a wet acid gas scrubber or a bag house in order to meet the HCL or Hg standard applicable to boilers under the final Boiler MACT rule and thereby achieve SO₂ emission reductions. If that boiler were to opt in to one of the Transport Rule SO₂ trading programs during the year before installing these controls to comply with the final Boiler MACT rule, then the boiler would be allocated allowances for the unit's current tons of SO₂ emissions and would not need to use these allowances for compliance under the Transport Rule once the final Boiler MACT-related controls were installed. The allowances allocated to the boiler would be additional allowances above the Transport Rule trading budget for the state where the boiler was located. As a result, the boiler would have freed-up allowances above the state trading budget that represent reductions that the boiler would have made anyway (i.e., even if the boiler had not opted in) and that could be sold to EGUs covered by the Transport Rule. In effect, the opting-in of the boiler would result in the conversion of the boiler's SO₂ reductions from the final Boiler MACT rule into increased emissions above the state trading budget from EGUs subject to the Transport Rule.

Commenters addressed this issue. For instance, one commenter suggested that SO₂ reductions made by a boiler under the final Boiler MACT rule should be eligible for opt-in provision allowances under the Transport Rule trading programs. Another commenter stated that, given the uncertainty that reductions made by opt-in units would be surplus, verifiable, permanent, and enforceable, opt-in provisions could compromise the integrity of the EGU emission reductions.

For the reasons explained above, EPA agrees with the latter commenter. Further, EPA notes that none of the commenters supporting adoption of the opt-in provisions suggested any revision to the proposed opt-in provisions that would address this problem. While the proposed opt-in provisions would limit an opt-in unit's allocation for a control period by calculating the allocation using the lesser of the unit's pre-opt-in SO₂ emission rate or the most stringent SO₂ emission rate applicable in that control period, this would not address SO₂ rate reductions that are not directly required by the final Boiler MACT rule but that are a secondary result of using and operating certain emission controls installed to comply with the HCL or Hg standards under the final Boiler MACT rule. Because the secondary SO₂ reductions will vary depending on the type of controls installed and on the extent to which the controls are used, and a boiler may use a combination of emission controls and other approaches to reduce HCL or Hg emissions (such as fuel switching), EPA believes that it is highly unlikely that opt-in provisions could prevent allocation for “anyway” emission reductions resulting from compliance with the final Boiler MACT rule. EPA therefore believes that the final Boiler MACT rule provides a concrete example of why adoption of opt-in provisions could undermine the rule's objective of addressing emissions in each state that significantly contribute to nonattainment or interfere with maintenance in other states. EPA notes that the final Boiler MACT rule, of course, is simply one example of how allocations for “anyway” reductions could occur and undermine the statutory requirements of the Transport Rule.

C. Compliance Deadlines

1. Alignment With NAAQS Attainment Deadlines

The compliance dates in the final Transport Rule are aligned with the attainment deadlines for the relevant NAAQS and consistent with the charges given to EPA by the Court in North Carolina. EPA proposed to require, and the final rule requires, compliance by 2014 with an initial phase of reductions in 2012. Sources are required to comply with annual SO₂ and NO_X requirements by January 1, 2012 and January 1, 2014 for the first and second phases, respectively. Similarly, sources are required to comply with ozone-season NO_X requirements by May 1, 2012, and by May 1, 2014. In selecting these dates, EPA was mindful of the NAAQS attainment deadlines which require reductions as expeditiously as practicable and no later than specified dates (see 42 U.S.C. 7502(a)(2)(A) (general attainment dates); 42 U.S.C. 7511(a)(1) (attainment dates for ozone nonattainment areas)), and also mindful of the court's instruction to “decide what date, whether 2015 or earlier, is as expeditious as practicable for states to eliminate their significant contributions to downwind nonattainment.” North Carolina, 531 F.3d at 930.

1997 PM_2.5 NAAQS Attainment Deadlines. For all areas designated as nonattainment with respect to the 1997 PM_2.5 NAAQS, the deadline for attaining that standard is as expeditious as practicable but no later than April 2010 (5 years after designation), with a possible extension to no later than April 2015 (10 years after designation). Many areas have already come into attainment by the April 2010 deadline due in part to reductions achieved under CAIR. The fact that the 2010 deadline will have passed before the Transport Rule is finalized emphasizes the importance of obtaining reductions as expeditiously as practicable. In addition, reductions achieved in upwind states by the 2014 emissions year will help downwind states demonstrate attainment by the April 2015 deadline.

2006 PM_2.5 NAAQS Attainment Deadlines. For all areas designated as nonattainment with respect to the 2006 24-hour PM_2.5 NAAQS, the attainment deadline must be as expeditious as practicable but no later than December 2014. Areas that fail to meet that deadline can request an extension to as late as December 2019.

Upwind emission reductions achieved by the 2014 emissions year will help meet the December 2014 attainment deadline. In addition, the first phase of reductions in 2012 will help many areas attain in a more expeditious manner.

Further, a deadline of January 1, 2014 also provides adequate and reasonable time for sources to plan for compliance with the Transport Rule and install any necessary controls. EPA believes that this deadline is as expeditious as practicable for the installation of the controls, if any, needed for compliance with the 2014 state emission budgets. (See further discussion in section V.C.2.)

1997 Ozone NAAQS Attainment Deadlines. Ozone nonattainment areas must attain permissible levels of ozone “as expeditiously as practicable,” but no later than the date assigned by EPA in the ozone implementation rule. 40 CFR 51.903. The areas designated nonattainment in 2004 with respect to the 1997 8-hour ozone NAAQS in the eastern United States were assigned maximum attainment dates effectively corresponding to the end of the 2006, 2009, and 2012 ozone seasons. The maximum attainment deadlines for the 1997 standard run from the June 15, 2004 effective date of designation for that standard. The time periods are based on the time periods provided for these classifications in section 181 of the Act, 45 U.S.C. 7511(a). However, instead of running from the 1990 date of enactment of the CAA as specified in section 181, our regulation provides that they run from the date of designation. An area's maximum attainment date is based on its nonattainment classification—that is, whether it is classified as a marginal, moderate, serious, severe, or extreme ozone nonattainment area. Marginal areas have three years from designation to attain the standard. Moderate, serious, severe, and extreme areas have 6, 9, 15, and 20 years, respectively. The maximum attainment deadlines associated with the 1997 ozone standards are June 15, 2007 for marginal areas, June 15, 2010 for moderate areas, and June 15, 2013 for serious areas. Because the actual deadline occurs in the middle of an ozone season, data from that ozone season is not considered when determining whether the area has attained by the deadline. Thus, these maximum attainment deadline dates effectively correspond with the end of the 2006, 2009, and 2012 ozone seasons. Reductions achieved or air quality improvements realized after those dates will not help the areas meet their maximum attainment deadlines.

Many areas have already attained the standard due in part to CAIR, federal mobile source standards, and other local, state, and federal measures. Other areas, however, have been reclassified to a higher classification either because they failed to attain by their attainment date or because the state requested reclassification to avoid missing an attainment date. Those that have not yet attained the standard now have maximum attainment dates ranging from June 2011 (these are the moderate areas that have been granted a 1-year extension due to clean data for the 2009 ozone season) to June 2024. The areas classified as “serious” nonattainment areas have a June 2013 maximum attainment deadline. Areas that missed their earlier deadlines and have been reclassified as “severe” or “extreme” nonattainment areas now have maximum nonattainment deadlines of June 2019 and June 2024 respectively. As explained above, an area with a June 2013 deadline would need to attain based on ozone data from the 2010-2012 ozone seasons, an area with a June 2019 deadline would need to attain based on ozone data from the 2016-2018 ozone seasons, and an area with a June 2024 deadline would need to attain based on ozone data from the 2021-2023 ozone seasons.

The Transport Rule's first phase of reductions in 2012 will help the remaining areas with June 2013 maximum attainment deadlines attain the 1997 8-hour ozone NAAQS by their deadline. If EPA determines that an area failed to attain by the 2013 deadline, the area would be reclassified to severe and would be subject to the more stringent emission control requirements that apply to the severe classification. The reductions will also help areas with later deadlines attain as expeditiously as practicable and improve air quality in those areas.

2012 Interim Compliance Deadline. EPA is requiring an initial phase of reductions starting in 2012. These reductions are necessary to ensure that significant contribution to nonattainment and interference with maintenance are eliminated as expeditiously as practicable and in time to help states meet their attainment deadlines. As the court emphasized in North Carolina, the significant contribution to nonattainment and interference with maintenance from upwind states must be eliminated as expeditiously as practicable to help downwind states to achieve attainment as expeditiously as practicable as required by the CAA. Further, reductions are needed by 2012 to help states attain before the June 2013 maximum attainment date for “serious” ozone nonattainment areas, to ensure states attain as soon after the original April 2010 attainment deadline for the 1997 PM_2.5 NAAQS, and to help states attain before the December 2014 attainment deadline for the 2006 PM_2.5 NAAQS.

In addition, because this final rule will replace CAIR, EPA could not assume that after this rule is finalized, EGUs would continue to emit at the reduced emission levels achieved by CAIR. Instead, it is the emission reduction requirements in the proposed FIPs that will determine the level of EGU emissions in the eastern United States. For this reason also, EPA concludes that it is appropriate to require an initial phase of reductions by 2012 to ensure that existing and planned SO₂ and NO_X controls operate as anticipated.

Addressing the Court's Concern about Timing. As directed by the Court in North Carolina, 531 F.3d 896, and as described previously, EPA established the compliance deadlines in the Transport Rule based on the respective NAAQS attainment requirements and deadlines applicable to the downwind nonattainment and maintenance sites.

The 2012 deadline for compliance with the limits on ozone-season NO_X emissions is necessary to ensure that states with June 2013 maximum attainment deadlines get the assistance needed from upwind states to meet those deadlines. The 2012 deadline for compliance with the limits on annual NO_X and annual SO₂ emissions is necessary to ensure attainment as expeditiously as practicable in areas which failed to attain by the 2010 attainment deadline for the 1997 PM_2.5 NAAQS and had to request an extension to 2015.

Similarly, the 2014 deadline for compliance with the limits on annual NO_X and annual SO₂ emissions is necessary to ensure that downwind states get the benefit of upwind reductions prior to the December 2014 maximum attainment deadline for the 2006 PM_2.5 NAAQS. It is also necessary to ensure reductions occur in time to assist with attainment in downwind areas that received the maximum 5-year extension of the 5-year attainment deadline for the 1997 PM_2.5 NAAQS (taking into account the need for reductions by 2014 to demonstrate attainment by April 2015).

The 2012 compliance deadline for the first-phase of annual NO_X and annual SO₂ emission reductions will assure the reductions are achieved as expeditiously as practicable. A significant amount of the emissions identified as significantly contributing to nonattainment or interfering with maintenance in other states can be eliminated by 2012. EPA believes it is appropriate to do so in light of the court's direction to EPA to ensure states eliminate such emissions as expeditiously as practicable. North Carolina 531, F.3d at 930. Given the time needed to design and construct scrubbers at a large number of facilities, EPA believes the 2014 compliance date is as expeditious as practicable for the full quantity of SO₂ reductions necessary to fully address the significant contribution to nonattainment and interference with maintenance. Requiring reductions in transported pollution as expeditiously as practicable, as well as within maximum deadlines, helps to promote attainment as expeditiously as practicable. This is consistent with statutory provisions that require states to adopt SIPs that provide for attainment as expeditiously as practicable and within the applicable maximum deadlines.

b. Public Comments and EPA Responses

EPA received numerous comments on the proposed compliance dates. A number of commenters supported EPA's compliance schedule and rationale. Other commenters supported extending the compliance deadlines to later dates.

Many commenters questioned the technical feasibility of achieving the required reductions by the 2012 and 2014 dates. EPA's responses to those comments are discussed below in section VII.C.2.

Other commenters provided policy and legal arguments for allowing states to develop SIP alternatives to the FIP, and to build time for that SIP development and review process into the compliance schedule. For example, some commenters asserted that the requirement in the CAA for providing reductions “as expeditiously as practicable” must be balanced with CAA provisions allowing states to develop state implementation plans prior to EPA imposing FIPs. EPA responses to those comments are discussed in section X.

Some commenters suggested that EPA had the ability to leave CAIR in place for a transition period, and by doing this EPA could allow for a longer compliance period for this rule. EPA does not believe it would be appropriate, in light of the Court's decision in North Carolina, to establish a lengthy transition period to the rule that will replace CAIR. Although the Court decided on rehearing to remand CAIR without vacatur, the Court stressed its prior decision that CAIR was deeply flawed and EPA's obligation to remedy those flaws. North Carolina, 550 F.3d 1176. Although the Court did not set a definitive deadline for corrective action, the Court took care to note that the effectiveness of its opinion would not be delayed “indefinitely” and that petitioners could bring a mandamus petition if EPA were to fail to modify CAIR in a manner consistent with its prior opinion. Id. Given the Court's emphasis on remedying CAIR's flaws expeditiously, EPA does not believe it would be appropriate to establish a lengthy transition period to the rule which is to replace CAIR.

As relates to PM_2.5, EPA received a number of comments on its proposal to include a 2012 deadline to ensure that emission reductions needed to reduce PM_2.5 be achieved “as expeditiously as practicable.” Some commenters supported EPA's 2012 deadline. Other commenters believed that it was unnecessary and unwarranted for EPA to impose emission reduction requirements in advance of the 2014 attainment date. In light of the 2014 five-year attainment date for the 2006 PM_2.5 NAAQS (with a possible extension to 2019), and the possible extension to April 2015 for the 1997 PM_2.5 NAAQS, these commenters believed EPA's 2012 emission reduction requirements for annual PM_2.5 and NO_X were not necessary. EPA disagrees with these commenters, for a number of reasons. First, EPA notes (supported by commenters) that there is a clear statutory obligation to attain “as expeditiously as practicable.” Second, EPA notes that there are feasible reductions available by 2012. Third, EPA believes that the substantial health and environmental benefits achieved by the rule underscore the importance of achieving the reductions as soon as possible.

With respect to ozone, some commenters noted that the proposed rule required ozone reductions by 2012 for states impacting areas which EPA's analysis shows will attain the 1997 ozone NAAQS by 2014 without further controls. Those commenters questioned the importance of getting reductions in such states and whether the 2012 deadline is necessary. EPA disagrees with those comments. Except for Houston, all ozone areas within the region addressed by this rule have attainment dates no later than 2013. In effect, this means that emission reductions needed to attain the 1997 ozone NAAQS must be in place by the 2012 ozone season. EPA believes that if there are reductions available by 2012, and those emission reductions have in fact been identified, it is appropriate and necessary to ensure that those reductions are in place.

2. Compliance and Deployment of Pollution Control Technologies

The power industry will undertake a diverse set of actions to comply with the Transport Rule at the start of 2012 and another set of actions when companies in Group 1 states comply with more stringent SO₂ budgets at the start of 2014. In 2012, the industry will largely meet the rule's NO_X requirements by: Operating an extensive existing set of combustion and post-combustion controls on fossil fuel-fired generators; dispatching lower emitting units more often; and installing and operating a limited amount of relatively simple NO_X pollution controls in states not previously subject to CAIR. For the SO₂ requirements, EPA anticipates at a minimum that coal-fired generators will operate the substantial capacity of advanced pollution controls already in place or scheduled for 2012 use; some units will also elect to burn lower-sulfur coals; and the fleet will increase dispatch from lower-sulfur-emitting units as well as from natural gas-fired generators. EPA provides a more detailed explanation below of how fuel switching to lower sulfur coals factored in to the design of the final Transport Rule.

By 2014, EPA's budgets under the Transport Rule will sustain previous NO_X and SO₂ reductions as well as account for reductions from additional advanced NO_X and SO₂ controls that are driven by other state and federal requirements. In addition to these reductions, companies in Group 1 states are also projected to add a limited amount of advanced SO₂ controls in 2014 that will be discussed below.

EPA's expectations are supported by the IPM analysis reported in this rule's RIA (see Chapter 7). Notably, since EPA has established a cap and trade control system for lowering NO_X and SO₂ emissions, individual owners and operators of covered units have some flexibility in meeting the program's requirements as needed and are free to find alternative ways to comply. The RIA clearly shows a viable known pathway for owners and operators to comply at reasonable costs, although it is not the only compliance pathway possible under this flexible regulation that could deliver the emission reductions required under the rule. Notably, by 2014 and beyond, the power industry may also augment the projected compliance efforts with programs aimed at improving energy efficiency.

Table VII.C.2-1—shows EPA's projection of the amount of existing coal-fired generating capacity in gigawatts (GW) that may retrofit various systems for compliance with this rule.

EPA received proposal comments expressing a concern about the feasibility of deploying retrofit air pollution control (APC) technologies in the time frames available between the final date of this rule and the compliance dates. As discussed below, EPA believes that it is feasible for the electric power sector and its APC supply chain to either make most of the projected retrofits in time to meet the 2012 and 2014 compliance deadlines, or to comply by other means.

a. 2012 Power Industry Compliance

EPA's analysis of emission reductions available in 2012 assumes year-round operation of existing post-combustion pollution controls in states covered for PM_2.5 and ozone-season operation of NO_X post-combustion controls in states covered for ozone. EPA also modeled emission reductions available in 2012 at the $500/ton threshold for SO_2, $500/ton for annual NO_X, and $500/ton for ozone-season NO_X.

For SO₂, EPA believes that reductions associated with the following methods of control are available and will be used as compliance strategies to meet the 2012/2013 budgets: (1) Operation of existing controls year-round in PM_2.5 states, (2) operation of scrubbers that are currently scheduled to come online by 2012, (3) some sources switching to lower-sulfur coal (see section VII.C.2.c that follows), and (4) changes in dispatch and generation shifting from higher emitting units to lower emitting units. EPA modeling and selection of a $500/ton cost threshold includes all existing and planned controls operating year round (items 1 and 2). It also reflects an amount of coal switching and generation shifting that can be achieved for $500/ton. This set of expected actions was confirmed in the detailed modeling of EPA's final remedy in the RIA and can be reviewed there.

The power sector is already strongly positioned to achieve the Transport Rule state budgets presented in section VI.D through at least three distinct strategies. First, the sector will optimize its use of the large proportions of advanced pollution controls already present throughout the fleet. Second, the sector will take advantage of the substantial new pollution control technology that is already on the way for deployment by 2012. Third, the remainder of the fleet will flexibly adopt the most economic low-emitting fuel mix available at each unit to deliver cost-effective emission reductions complementing the reductions achieved from optimized use of the fleet's pollution control technology. The state maps in Chapter 7 of this rule's Regulatory Impact Analysis demonstrate how these emission reduction strategies for 2012 will build off of the sector's historic trend toward cleaner generation profiles. Also, the detailed unit-level projection files from EPA's IPM power sector modeling of the Transport Rule remedy (found in the docket for this rulemaking) show how EGUs adopt these strategies to not only reach the 2012 budgets, but in fact in many states overcomply with the budgets and build up a bank of allowances under the programs for future flexibility.

The following paragraphs illustrate the degree to which the existing fleet is already prepared to adopt these emission reductions in 2012 in order to attain the required emission reductions for SO₂, annual NO_X, and ozone-season NO_X under the Transport Rule. More specifically, the illustrative paragraphs demonstrate emission reduction pathways for coal capacity to optimize or increase operation of existing control technology, timely implement existing plans to bring additional control technology on line, and to cost-effectively make use of lower-emitting fuel alternatives.

Of the 240 GW of coal capacity in the Transport Rule region covered for fine particles, approximately 110 GW—more than 45 percent—had existing advanced pollution control for SO₂ already in place in 2010, including scrubbers (FGD), dry sorbent injection (DSI), or circulating fluidized bed boilers. Of this controlled coal capacity, EPA expects a significant portion will improve emission rates through either increased use of control technology and/or additional fuel switching. EPA notes that an additional 39 GW of advanced SO₂ controls in the region are scheduled to come online over the 2010-2012 timeframe and will also assist in meeting 2012 emission reduction requirements. Thus, by 2012 more than half of affected coal capacity—152 GW—will be operating with advanced SO₂ control equipment. Additionally, EPA expects approximately 40 GW of uncontrolled coal capacity in the region to take advantage of the existing coal supply infrastructure, possibly switching coal use or coal blending behaviors to make cost-effective reductions in SO₂ emission rates where economic to respond to the Transport Rule 2012 emission reduction requirements.

EPA notes that approximately 136 GW of the 240 GW—more than 56 percent—of coal capacity in the Transport Rule region covered for fine particles had existing advanced pollution control for NO_X already in place in 2010, including selective catalytic reduction (SCR), selective non-catalytic reduction (SNCR), or circulating fluidized bed boilers. Of this capacity, EPA anticipates a significant portion will improve their NO_X emission rate through increased operation of these existing controls. Additionally, EPA notes that an additional 21 GW of SCR and 4 GW of enhanced combustion controls (including low-NO_X burners and overfire air) are scheduled to come online in the region during the 2010-2012 timeframe, bringing the total region's coal capacity operating with NO_X emission reduction technology to 158 GW (more than 65 percent of total coal capacity in the Transport Rule fine particle region). EPA also projects that approximately 13 GW of coal capacity will make some reduction in their NO_X emission rates by enhancing performance of existing combustion controls or SNCR, or by fuel switching.

In the Transport Rule states covered under the ozone-season program, approximately 145 GW of the 260 GW (more than 55 percent) of coal capacity had existing NO_X control technology in place in 2010. EPA expects a significant portion of that capacity to achieve emission reductions during the 2012 ozone-season through improved operation of SCR. Additionally, in the Transport Rule ozone region there will be approximately 21 GW of additional advanced NO_X control installations and 7 GW of additional combustion control improvements or installations coming online during the 2010 to 2012 time frame. EPA projects that 17 GW of coal capacity in the Transport Rule ozone region will reduce NO_X emission rates by enhancing performance of existing combustion controls or SNCR or by fuel switching.

For NO_X, EPA has also concluded that it is appropriate to require reductions through a limited amount of combustion control improvements, and in some cases, retrofits such as low-NO_X burners (LNB) and/or overfire air (OFA). EPA recognizes that the 6-month time frame between rule finalization and start of the first compliance period would not allow for the installation of a major post-combustion NO_X control such as SCR. Assumed improvements and retrofits for the January 1, 2012 deadline for annual NO_X reductions therefore only involve the much simpler LNB/OFA control modifications or installations. Alternatively, some plant owners might choose to achieve NO_X reductions in a similar time period through an even simpler retrofit—SNCR.

Although the improvements, and in some cases, installation of combustion controls would be an economic means of achieving emission reductions, these specific controls are not required for compliance purposes under the final Transport Rule remedy. Individual sources may comply through other measures (such as purchasing additional allowances) in the event that it takes more than 6 months for installation of a given combustion control. The vast majority of covered sources already have combustion controls installed; therefore, the NO_X reductions associated with these incremental control improvements and installations are small.

Based on the Transport Rule's geography, EPA estimates that approximately 10 GW of coal-fired units may improve, and in some cases, install LNB/OFA specifically in reaction to the Transport Rule NO_X caps. EPA reflects the effects of these installations in the 2012 annual and ozone-season NO_X budgets, which would yield reductions of approximately 28,000 tons of annual NO_X and 14,000 tons of ozone-season NO_X. EPA assumes these controls are cost effective at $500/ton and that they should be incentivized through budgets given the 2013 attainment deadline for ozone areas classified as “serious.” Once installed, LNB/OFA operates any time the boiler is fired and thus yields NO_X reductions beyond the ozone season alone.

In the proposal's LNB technical support document, EPA observes that LNB and/or OFA installations, burner modifications, or other NO_X reduction controls would likely have to be installed during fall 2011 or spring 2012 outages in order to achieve significant reductions for 2012. While this 6-month schedule is aggressive, industry has shown that it can be met. For example, Limestone Electric Generating Station Unit 2, an 820 MW tangentially-fired lignite unit, was retrofitted with Foster Wheeler's Tangential Low NO_X (TLN3) system in less than six months, including engineering, fabrication, delivery and installation. Harlee Branch Unit 4, a 535 MW cell-fired unit, was retrofitted with Riley Power's low-NO_X Dual Air Zone CCV burners on a similar schedule. These are tangentially-fired and wall-fired units, respectively, representative of the unit types that might make LNB/OFA improvements for compliance with this rule. Although such 6-month schedules can be achieved on some units, under favorable circumstances, historical projects suggest a more typical schedule would be 12 to 16 months for the contractor's portion of the work. A plant owner's project planning and procurement work in advance of a contract award would typically involve several additional months. On the other hand, there are other approaches that can also be implemented in a short time frame to achieve significant NO_X reduction. As mentioned above, relatively simple SNCR systems can be installed quickly; and the re-tuning or upgrading of existing combustion control systems can often provide significant NO_X reductions and can be performed quickly.

As stated above, EPA believes that LNB/OFA modifications or retrofits would be possible during the 6-month interim between rule signature and the start of the first compliance period, particularly for those “early movers” who have initiated LNB projects based on the proposed rule. However, as shown in Table VII.C.2-2, below, even if all LNB modifications or installations are delayed until the beginning of the 2012 ozone season, the reductions only represent 1 percent of most covered states' annual NO_X budgets, and no more than 11 percent of any affected state's annual NO_X budget. Under such a scenario, these delayed reductions would still be well within the 18 percent variability limit applied to each state's annual NO_X budget. In light of this limited consequence and the supporting material above, EPA includes LNB-driven NO_X reductions in both annual and ozone-season NO_X budgets for 2012.

b. 2014 Power Industry Compliance

EPA projects that compliance with 2014 requirements for NO_X will result largely from operation of existing and future controls required by state and other federal requirements, as well as the appropriate dispatch of the electric generation fleet. EPA does not project additional NO_X pollution control retrofits aside from about 10 GWs of combustion control improvements or retrofits projected for the 2012 compliance period. To comply with the rule's SO₂ requirements, EPA projects that the power industry will rely on existing controls, operate newly installed advanced controls necessary for other binding state and federal requirements, rely more on relatively lower sulfur coals, and dispatch lower-emitting generation units. In Group 1 states, industry is projected to increase switching to lower sulfur coals and install a limited amount of additional scrubbers and other advanced pollution control technology. EPA's assessment of the industry's ability to install SO₂ pollution controls in 2014 and undertake the projected coal switching follows below.

EPA's modeling of least-cost compliance with the state budgets under the Transport Rule projects approximately 5.9 GW of FGD systems and lesser amounts of other technologies will be retrofitted by 2014 for compliance with the Transport Rule. EPA's schedule assumptions for these larger more complex projects were developed in an earlier study and mentioned in the proposal: 27 months for retrofitted wet FGD and 21 months for SCR. Note that a dry FGD system, due to its relatively simpler configuration and lesser cost, would typically take somewhat less time to retrofit than wet FGD.

As discussed below, EPA believes that its schedule assumptions remain reasonable expectations for sources that have completed most of their preliminary project planning and can quickly make commitments to proceed. These schedules do not include the extensive time that some plant owners might spend in making a decision on whether or not to retrofit. They do include the time needed to make a final confirmation of the type of technology to be used at a particular site, to prepare bid requests, award contracts, perform engineering, obtain construction and operating permits (in parallel with project activities), perform construction, tie-in to the existing plant systems, and perform integrated systems testing.

EPA received comments on the proposed rule indicating that some past single-unit APC retrofits had considerably longer schedules, with a few exceeding 48 months. EPA engineering staff have extensive experience with power plant and APC system design, construction, and operation. Based on that experience, EPA can observe that in the absence of a compelling deadline or major economic incentive, many large project schedules are considerably longer than necessary. Given further observations as explained below, EPA believes it is reasonable to expect that almost all future APC retrofits can be completed far more quickly than they were in recent history. EPA's perspective on this matter derives in part from a comparison of longer APC schedules (as provided by some commenters) to the project schedule for an entire new coal-fired unit, including its APC systems. Springerville Unit 3, for example, is a new 400 MW subbituminous coal-fired unit with SCR and dry FGD that became operational in July 2006, some 33 months after the turnkey engineering-construction contractor was given a notice to proceed with engineering. Springerville was clearly on an accelerated schedule, as its original planned schedule was about 38 months. Another example is Dallman Unit 4, a high-sulfur bituminous coal-fired 200 MW unit with SCR, fabric filter, wet FGD, and wet ESP. Dallman Unit 4 was first synchronized in May 2009, several months ahead of schedule, and about 36 months after its turnkey contractor placed initial major equipment orders. The main point here is that recent APC project schedules, and those of large complex power projects, can be significantly accelerated. Because the scope and complexity of the work involved for an entire new coal unit and its APC systems is perhaps five times greater than that of a retrofit wet FGD system alone, EPA believes it is reasonable to expect that even the most complex retrofit APC project can be significantly accelerated as well. Additional factors are discussed below that further support the feasibility of installing by 2014 the 5.9 GW of FGD retrofits projected for this rule.

Although IPM modeling provides reliable estimates on a regional basis, and cannot be as accurate at the level of individual plants or units, it is informative and relevant to consider IPM's plant level projections in this case. Although the IPM-projected retrofits named below may not actually occur, IPM projects that they would be economic and would allow industry to meet the tighter SO₂ emission standards in Group 1 states in 2014. EPA notes that the owners of the particular plants mentioned below (Duke Energy, AEP, Edison International) are large, experienced, versatile utilities that have done considerable advance planning and should also have above-average flexibility to comply with state budgets across their fleets. EPA would expect such owners to have relatively little difficulty in permitting and financing FGD retrofits.

Of the Transport Rule-related FGD retrofits, 0.2 GW is projected to use dry FGD, which EPA expects to be simpler and quicker to install than wet FGD. Half of the 5.9 GW (Muskingum, Rockport) has already been committed under consent decrees to add controls or retire; and EPA reasonably believes that significant preliminary project planning work has already been done for those projects. An additional 1,200 MW (Homer City) had completed project planning and was ready to proceed in 2007, before putting the project on hold. The latter plant is now facing EPA legal action and the possibility of a required expeditious FGD retrofit. Thus, of the 5.9 GW of projected FGD retrofits resulting from this rule, nearly 75 percent appears to be in good position for an early start of construction, and over 3 GW of that would be bringing forward already committed compliance start dates.

Any of the above mentioned potential retrofits or any other unit that might choose to retrofit FGD for a January 2014 compliance date will likely have to use various methods to accelerate the project schedule. Such methods could include the use of parallel permitting, overtime and/or two-shift work schedules during construction, and 5- or 6-day work weeks instead of the 4-day × 10-hour schedules often used to minimize cost when time is not of the essence. Increased use of offsite modularization and pre-fabrication of APC components could also shorten schedules and reduce job hours.

EPA believes that the January 1, 2014 compliance date is as expeditious as practicable for the sources installing large, complex control systems. The following additional observations support EPA's expectation that the limited 5.9 GW of FGD retrofits can be realized in the 30 month interim between rule signature and the start of 2014:

• There are documented instances of large, complex wet FGD retrofits being deployed in less than 30-months (excluding the time for owners' project planning). Examples are Killen Station Unit 2, and Asheville Unit 1.⁷⁵

• In 2009 the APC supply chain deployed more than six times more GW capacity of FGD and SCR controls than the 5.9 GW of FGD that would be deployed by 2014 under this Rule.
• The APC supply chain has seen a 2-year decline in deployments since its peak in 2009, but in 2011 is nonetheless putting into service about three times more GW capacity of FGD and SCR controls than the 5.9 of FGD that would be deployed under this Rule.
• Because the supply chain has been in decline, but remains quite active, there are now adequate supply chain resources available that can be quickly reengaged to support a rapid deployment of 5.9 GW of FGD.

EPA recognizes that the installation of any amount of scrubbers in this short time frame will require aggressive action by plant owners and that the owners who can meet this schedule will already have done their project planning and will be ready to place orders. An example of such “early movers” was seen in the power sector's anticipation of CAIR. EPA data indicate that solely CAIR-driven FGD and SCR deployments of about 6 GW occurred within two and one-half years after CAIR's finalization in mid-2005, showing that at least 20 percent of the total CAIR-only controls effort through a 2010 compliance date was sufficiently planned for installation to start before or immediately upon finalization of the rule. EPA reasonably expects that similar advance planning has already been done for units that would retrofit under this rule.

In the event that a particular control installation requires additional time into 2014 to come online, EPA believes compliance would not be jeopardized given the ability of sources to purchase allowances during that time. This approach could be supported by some sources with FGD that have the ability to increase their SO₂ removal above historic rates, perhaps through relatively low cost upgrades to improve scrubber effectiveness, or by operating scrubbers at higher chemistry ratios. The ability of sources to temporarily or permanently substitute dry DSI for FGD serves as another backstop for any feasibility issues regarding FGD. Note that the updated modeling for this rule projects the addition by 2014 of about 3 GW of DSI for SO₂ control using trona or other sorbent. DSI is a relatively low capital cost technology that readily can be installed in the time frame available for compliance.

It should also be noted that most APC retrofits will involve a source outage for final “tie-in” of retrofitted systems to existing systems, during which time emissions from the affected units are zero. For some sources, the duration of this tie-in outage may effectively extend the deadline by which all of the projected emission reductions need to occur.

Although EPA believes that installation of 5.9 GW of FGD at facilities by January 1, 2014 is feasible, EPA also conducted an IPM sensitivity analysis to examine a scenario in which FGD retrofitting by 2014 is not allowed. Results of EPA's “no FGD build in 2014” analysis indicate that if the power industry were subjected to the requirements of this rule without an FGD retrofit option for compliance until after 2014, covered units would still be able to meet the Transport Rule requirements in every state while respecting each state's assurance level. (See the docket to this rulemaking for the IPM run titled “TR_No_FGD_ in2014_Scenario_Final.”)

In this scenario without the availability of new FGD by 2014, sources in covered states complied with the Transport Rule budgets by using moderate additional amounts of DSI retrofits, switching to larger shares of sub-bituminous coal, and dispatching larger amounts of natural gas-fired generation in lieu of the FGD retrofits that are projected as being most economic under modeling of the Transport Rule remedy. Because new FGD capacity is included in EPA's projection of the least-cost set of SO₂ emission reductions required in Group 1 states, the “no FGD” sensitivity scenario did project higher system costs, although these costs were still substantially lower than the remedy EPA modeled in the Transport Rule proposal.

The “no FGD” analysis indicates that while the ability of Group 1 states to meet their 2014 SO₂ budgets is facilitated by FGD retrofits, they are by no means required, nor is Transport Rule compliance jeopardized by their absence. Even under a scenario in which sources fail to complete FGD retrofits by 2014, sources in the affected states would have other compliance options available at reasonable cost to meet the state's budget requirements. This analysis shows that Group 1 states would be able to comply with their 2014 SO₂ budgets by relying on other emission reduction opportunities that do not require FGD retrofits. EPA analysis confirms that those alternatives are feasible both in terms of cost and timing.

Finally, EPA recognizes that, when finalized later this year as currently scheduled, the Mercury and Air Toxics Standards (MATS) will require significant retrofit activity at covered sources in the power sector with a 2015 compliance date for that rule. EPA's projections of retrofit activity under the final Transport Rule are highly compatible with its projections of retrofit activity under the proposed MATS (which included the proposed Transport Rule in its baseline). EPA therefore anticipates that the Transport Rule's projected retrofit activity will not only be the least-cost compliance pathway to meeting state budgets in 2014 but will also accelerate emission reductions subsequently required by the effective date of MATS. The final Transport Rule's projected 2014 retrofit installations will also further incentivize the power sector to ramp up its retrofit installation capabilities to achieve broader deployment of the projected pollution control retrofits under the proposed MATS.

Considering all the reasons given above, EPA has concluded that the 2014 requirements for SO₂ emissions in the states covered by the Transport Rule are reasonable and can be met by the power industry by a variety of means.

c. Coal Switching for SO₂ Compliance in 2012 and 2014

Coal switching is another mechanism which can be used along with operating pollution controls in 2012 for compliance. It will be a complementary activity by many coal-fired units alongside of operating pollution controls and the addition of more scrubbers and DSI in 2014.

In the proposal, EPA noted that coal switching could serve as a compliance mechanism for 2012. EPA requested comment on the reasonableness of EPA's assumption that coal switching will have relatively little cost or schedule impact on most units. EPA received substantial comment suggesting that the coal switching and coal blending projected by EPA modeling are not feasible for all units, and that, if feasible, would often incur a cost through the derating of the unit associated with the switch to a lower sulfur coal or coal blend. Additionally, sources indicated that coal switching by 2012 would not always be possible in the six month window between final rule signature and start of compliance. These feasibility concerns stemmed from restrictions included in existing coal supply contracts and from boiler design constraints that may hinder coal switching within a 6 month window.

EPA agrees with these concerns and revised its IPM modeling to limit coal switching capability in 2012 for particular units that may have trouble switching coals or coal blends in a six month time frame. A cost adder was also included in the IPM modeling for coal switching to capture the potential cost burden of deratings that might accompany switching to a very low sulfur subbituminous coal or coal blend.

A particular commenter concern regarding switching to lower sulfur within the eastern bituminous coals related to a possible impact on the performance of a cold-side electrostatic precipitator (ESP). Some ESPs that operate at acceptably high collection efficiency when using a high- or medium-sulfur bituminous coal may experience some loss in collection efficiency when a lower sulfur coal is used. Whether this occurs on a specific unit, and the extent to which it occurs, would depend on the design margins built into the existing ESP, the percentage change in coal sulfur content, and other factors. In any case, industry experience indicates that relatively inexpensive practices to maintain high ESP performance on lower sulfur bituminous coals are available and can be used successfully where necessary. These include a range of upgrades to ESP components and flue gas conditioning. EPA therefore assumes that it will not be necessary for units that switch from higher to lower sulfur bituminous to make a costly replacement of the ESP.

Coal switching as a SO₂ compliance option might also include switching from bituminous to subbituminous coal. EPA's analysis does not assume that a unit designed for bituminous can switch to (very low sulfur) subbituminous coal unless the unit's historical data demonstrate that capability in the past. EPA assumes that units with that demonstrated capability have already made any investments needed to handle a switch back to the use of subbituminous coal at a similar percentage of its heat input as in the past. For IPM analysis in the final rule EPA also introduced a coal switching option that assumes that units can increase a historically low percentage use of subbituminous to a “maximum” level, if economic. This option includes an appropriate derate in output, increase in heat rate, and additional capital and operating costs. Details of this and other IPM updates for this rule are provided in the IPM Modeling Documentation in the docket for this rulemaking (“Documentation Supplement for EPA Base Case v.4.10_FTransport—Updates for Final Transport Rule”).

Some commenters also expressed concern with the assumption that coal-switching from lignite to subbituminous is a cost-effective or feasible emission reduction strategy, particularly at Texas EGUs. EPA carefully considered these comments and adjusted its modeling of cost-effective reductions to address this concern. Specifically, EPA made adjustment in the model so that it assumes coal-switching is not a compliance option at the specific units where commenters identified technical barriers to subbituminous coal consumption. The Transport Rule emission budgets are based on this adjusted modeling which does not assume any infeasible coal-switching from lignite to subbituminous. In addition, EPA's analysis of cost-effective reductions in each state presented in section VI.B shows that Texas is capable of cost-effectively meeting its Transport Rule emission budgets; however, EPA also conducted sensitivity analysis that shows Texas can also achieve the required cost-effective emission reductions even while maintaining current levels of lignite consumption at affected EGUs. More details regarding this analysis, including a table comparing key parameters between the main Transport Rule remedy analysis and this Texas lignite sensitivity, can be found in the response to comments document and the IPM model output files included in the docket for this rulemaking.

D. Allocation of Emission Allowances

Under the final rule, EPA distributes a number of SO₂, annual NO_X, and ozone-season NO_X emission allowances to covered units in each state equal to the SO₂, annual NO_X, and ozone-season NO_X budgets for those states. These budgets are addressed in section VI.D of this preamble. This section discusses the methodology EPA uses to allocate allowances to covered units in each state.

As discussed later in section VII.D.2, EPA is setting aside a base 2 percent of each state's budgets for allowance allocations for new units, with 5 percent of that 2 percent, or 0.1 percent of the total state budget being set aside for new units located in Indian country. To this base 2 percent, EPA is setting aside an additional percentage on a state-by-state basis, ranging from 0 to 6 percent (yielding total set asides of 2 percent to 8 percent), for units planned to be built. The remainder of the state budget is allocated to existing units. Tables VI.D.-3 and VI.D.-4 in this preamble show the SO₂, annual NO_X, and ozone-season NO_X budgets for each covered state (without the variability limits). In allocating allowances to existing and new units, EPA distributes four discrete types of emission allowances for four separate programs: SO₂ Group 1 allowances, SO₂ Group 2 allowances, annual NO_X allowances, and ozone-season NO_X allowances.

In the SO₂ Group 1 and SO₂ Group 2 programs, each SO₂ allowance authorizes the emission of one ton of SO₂ in that vintage year or earlier and is usable for compliance only in the program for which the allowance was issued. In the annual NO_X program, each annual NO_X allowance authorizes the emission of one ton of NO_X in that vintage year or earlier in that program. In the ozone-season NO_X program, each ozone-season NO_X allowance authorizes the emission of one ton of NO_X during the regulatory ozone season (May through September for this final rule) in that vintage year or earlier for that program.

In each of the four trading programs, a covered source is required to hold sufficient allowances (issued in the respective trading program) to cover the emissions from all covered units at the source during the control period. EPA assesses compliance with these allowance-holding requirements at the source (i.e., facility) level.

This section explains how, in this final rule, EPA allocates a state's budget to existing units and new units in that state. This section also describes the new unit set-asides and Indian country new unit set-asides in each state, allocations to units that are not operating, and the recordation of allowance allocations in source compliance accounts.

1. Allocations to Existing Units

This subsection describes the methodology EPA will use in the FIPs finalized in this action to allocate to existing units. The same methodology will be used to allocate allowances to existing units for all four trading programs.

For the reasons explained below, EPA has decided to base allocations made under the FIPs on historic heat input, subject to a maximum allocation limit to any individual unit based on that unit's maximum historic emissions. This methodology gives each existing unit an allocation equal to its share of the state's historic heat input for all the covered units in the program, except where that allocation would exceed its maximum historic emissions; this methodology constrains the heat input-based allocations from exceeding any unit's maximum historic emissions. Further detail on the implementation of this approach is provided in section VII.D.1.c below as well as in the Allowance Allocation Final Rule TSD in the docket for this rulemaking. All existing-unit allocations for 2012 will be made pursuant to the FIPs. However, as described in section X, states may submit SIPs or abbreviated SIPs to use different allocation methodologies for allowances of vintage year 2013 and later.

a. Summary of Allocation Methodologies and Comments

EPA took comment on three distinct allocation methodologies for existing units. The first—an emissions-based option—was presented in the original Transport Rule proposal (75 FR 45309). The second and third—heat input option 1 and heat input option 2—were presented in a Notice of Data Availability (76 FR 1113). EPA received numerous comments on all three options.

i. Emission-Based Allocation Methodology

The emission-based option presented in the original Transport Rule proposal would base allowance allocations to existing units on each covered unit's calculated emission “share” of that state's budget for a given pollutant under the Transport Rule. The proposed rule stated that “for 2012, each existing unit in a given state receives allowances commensurate with the unit's emissions reflected in whichever total emissions amount is lower for the state, 2009 emissions or 2012 base case emissions projections. In either case, the allocation is adjusted downward, if the unit has additional pollution controls projected to be online by 2012. * * * For states with lower SO₂ budgets in 2014 (SO₂ Group 1 states), each unit's allocation for 2014 and later is determined in proportion to its share of the 2014 state budget, as projected by IPM” (75 FR 45309).

Many commenters objected to this projected emission allocation methodology. Commenters offered two principle objections. First, they argued EPA should not use unit-level model projections to allocate allowances. Second, they argued the use of any emission-based allowance methodology is improper. Many of these commenters argued that instead of an emission-based allocation methodology, EPA should use a heat-input-based allocation methodology.

Commenters' objections to the use of unit level model projections focused primarily on the accuracy of such projections. While many commenters supported the use of modeling projections in determining state emission budgets, they argued that the unit-level model projections were not sufficiently accurate to use as a basis for allocating allowances to individual units. Among other things, they argued that the modeling used for the proposal did not recognize certain non-economic factors that may cause individual units to operate differently than the model projects. Commenters also argued that EPA's modeling does not capture all up-to-date contracts and other economic arrangements made at the unit-level which may affect operational decision-making. Some of these commenters continued to support the use of an emission-based allocation approach, but urged EPA to use more up-to-date and specific unit-level data in its modeling projections. Others opposed the use of any emission-based allocation approach.

EPA acknowledges that the model may not, at this time, capture all relevant operational decision factors for each individual unit. EPA also recognizes that there are unit-level details of operational decision-making and economic arrangements (such as certain contracts for electricity sales) that are private and thus unavailable to EPA on an ongoing basis for modeling purposes. EPA believes these potential omissions would not have a significant impact on EPA's determination of significant contribution at the state level; however, EPA recognizes they could conceivably have a significant impact on projections at the individual unit level. EPA thus agrees with commenters that the unit-level emission projections from its modeling may not reflect all possible operational decisions at a given unit and are therefore not an appropriate proxy measure to use as a basis for allocating allowances to individual units.

Many commenters also argued that, even if the emission projections could be adjusted to capture all known and up-to-date unit-level operational factors, EPA should not use any emission-based allocation approach. They argued that an emission-based approach should not be used because it is not fuel-neutral. That is to say, the type of fuel consumed significantly affects the emissions from, and therefore the allocation to, a given unit under an emission-based approach. Commenters argued that an approach that is not fuel-neutral effectively awards higher-emitting units. Commenters also argued that a projected emission-based approach should not be used because it is not control-neutral. In other words, whether or not a unit has installed controls would significantly affect the allocation for a given unit under an emission-based approach. Under an emission-based approach, controlled units receive significantly fewer allowances than uncontrolled units. Such an approach, commenters pointed out, effectively penalizes sources who have taken action to reduce emissions.

EPA acknowledges that an emission-based approach would not be fuel-neutral or control-neutral. EPA notes that the DC Circuit rejected the fuel adjustment factors that were used in CAIR to adjust state budgets based on the type of fuel burned at each covered unit. North Carolina, 531 F.3d 918-21 (rejecting use of fuel adjustments in setting state NO_X budgets). While the proposal's allocation methodology did not explicitly adopt “fuel adjustment factors” for allocation purposes, EPA recognizes that an emission-based allocation methodology effectively advantages or disadvantages units based on the type of fuel they combust.

In addition, several commenters argued that the proposal's emission-based methodology would inappropriately reward the highest emitters under the program with more allowances than their lower-emitting counterparts would receive. EPA acknowledges that such a methodology would allocate more allowances to units whose emissions make up a larger share of the proposed Transport Rule programs' state budgets. EPA notes that because any allocation patterns under the Transport Rule FIPs would be established in advance of covered sources' compliance decisions (i.e., decisions regarding how much to emit under the programs), covered sources cannot be “rewarded” by adjusting their future emissions. However, EPA notes commenters' observations that the proposal's methodology would reduce allocations to units that previously installed pollution control technology or invested in cleaner forms of generation in anticipation of CAIR. EPA concluded in review of these comments that the proposed Transport Rule's allocation methodology unintentionally yielded this distributional outcome. EPA therefore considered alternative allocation methodologies described below.

A substantial portion of the commenters who objected to the proposal's emission-based allocation option urged EPA to consider historic heat input based approaches. EPA agreed it should accept comment on the use of historic heat input-based approaches and published a NODA to provide an opportunity for comment on two specific heat input options and the allocations that would result from application of those options to the proposed Transport Rule state budgets.

ii. Heat Input Allocation Option 1

The first heat input option presented by EPA in the NODA (“Option 1”) allocates allowances to units based solely on their historic heat input. Under this option, EPA would establish a 5-year historic heat input baseline for each covered unit and allocate allowances to sources at levels proportional to the each unit's share of the total historic heat input at all covered units in that state.

Numerous commenters supported the use of a heat-input based allocation methodology. These commenters stated that basing allocations on historic heat input has the following advantages over the proposal's emission-based allocation methodology:

(A) For certain types of units, historic heat input data may offer a better representation of unit-level operation than model projections of unit-level emissions; furthermore, for all units, historic heat input is typically represented by quality-assured data reported by sources from continuous emission monitoring systems, which strengthens its accuracy.

(B) Historic heat input data are generally fuel-neutral in that they do not generally yield higher allocations for units burning or projected to burn higher emitting fuels.

(C) Historic heat input data are generally emission-control-neutral in that they do not generally yield reduced allocations for units that installed or are projected to install pollution control technology.

Many commenters also argued that a heat input-based allocation methodology should be used because, unlike the proposal's emission-based methodology, a heat-input based methodology would be generally fuel-neutral and control-neutral and would rely on unit-level quality-assured data instead of on modeling projections.

Several commenters expressed support for specific aspects of heat input option number one. From a technical standpoint, commenters noted that heat input option 1 relied on the highest-quality and most transparent data EPA had provided as a basis for allocating allowances under the Transport Rule programs. They argued that the calculation methodology for heat input option 1 is more readily re-created and understood by sources than either the proposal's methodology or EPA's application of the “reasonable upper-bound capacity utilization factor and a well-controlled emission rate” in heat input option 2 (described in greater detail below). They also pointed out that it is similar to methodologies used in previous trading programs, such as the NO_X Budget Trading Program (see 40 CFR 96.42(a) (b) (calculating each existing EGU's allocation by multiplying each unit's historic heat input by 0.15 lb/mmBtu)). In addition, commenters supported the reliance of heat input option 1 on continuous emission monitoring system (CEMS) data that are reported to EPA and certified by the source's designated representative (DR) as accurate and complete. In addition, many commenters supported EPA's use of historic data without further transformation by any calculation factors created by EPA.

From a policy perspective, commenters highlighted the fuel neutrality and emission-control neutrality aspects of heat input option 1. They noted that this option does not, in contrast to the proposal's emission-based methodology, penalize a source, through a reduced allowance allocation, for having chosen a generation technology or emission control technology that was more favorable to public health and the environment. EPA agrees with these observations. The allocation pattern associated with this option does not advantage or disadvantage units based on either the fuel consumed or the presence or absence of a pollution control technology. In this respect, it is a neutral approach that does not “reward” high-emitting units or “penalize” low-emitting units, including, for example, those units on which pollution control technology was installed in anticipation of CAIR.

EPA agrees with the aforementioned arguments from these commenters regarding the technical and policy merits of this heat input-based allocation methodology. EPA believes that the quality-assured heat input data reported by EGUs under its programs are among the most detailed and sound unit-level data accessible by EPA. EPA believes the calculation of any individual unit's share of this historic heat input data is a straightforward, clear, and simple calculation to perform, such that EPA's calculated allowance allocations under this approach can be relatively easily replicated.

EPA also agrees with commenters that such data has previously supported allowance allocation procedures for highly successful program implementation of the ARP and the NO_X Budget Trading Program (NBP). Notably, Congress chose a heat input-based allocation approach when authorizing the ARP in title IV of the Clean Air Act, suggesting that Congress viewed heat input as a reasonable basis for allocation. Additionally, EPA's selection of a heat input-based approach for the NBP was not legally challenged, implying that stakeholders generally saw a heat input-based approach as reasonable.

EPA also agrees with comments observing that allocations made under this heat input approach do not advantage or disadvantage units based on their choice of fuel combustion or pollution control technology, and that allocations under this approach would thus be “fuel-neutral” and “control-neutral.” EPA also agrees with commenters that unlike the proposed rule's emission-based methodology, this heat input methodology does not yield lower allocation to units that reduced emissions in advance of the Transport Rule relative to units that did not make such emission reductions.

Other commenters objected to the use of a heat-input based allocation methodology. These commenters argued that the allocation pattern associated with a heat-input allocation methodology would yield “windfall profits”—in the form of allowance allocations greatly in excess of likely emissions—for certain units, particularly with regard to SO₂ allowance allocations for units combusting natural gas. EPA disagrees with the characterization of the excess allowances as “windfall profits.” Allocations based on heat-input alone are fuel-neutral and control-neutral. The characterization of the heat-input allocation methodology as creating “windfall profits” for any unit is based on the assumption that all units should be allocated allowances based on emissions, not heat input. In arguing the heat-input approach creates a “windfall” for some units, commenters are assuming that the allocation of allowances above a unit's projected emissions constitutes a “windfall”—a conclusion EPA does not accept. EPA believes that under market-based regulatory programs, it is appropriate to base initial allowance allocations on a neutral factor and allow the market to determine the least-cost pattern of emission reductions in each state to achieve the reductions that address the state's significant contribution and interference with maintenance under the final Transport Rule programs. EPA disagrees that future allowance transactions (following a neutral-factor initial allocation) in response to these market forces can be characterized as “windfall profits.” As explained above, EPA believes it is appropriate to allocate allowances based on a neutral factor. Commenters appear to ask EPA, instead of allocating based on a neutral factor, to consider the unit-level distributional impacts of each allocation methodology and to select an allocation methodology on the basis of equity. EPA does not believe it would be appropriate for the agency to pick an allocation methodology to achieve any particular distributional outcome as such considerations are not related to the statutory mandate of CAA section 110(a)(2)(D)(i)(I). Instead, EPA believes it is appropriate to allocate allowances to sources covered by its trading programs based on a neutral factor. Furthermore, CAA section 110(a)(2)(D)(i)(I) requires prohibition of certain emissions within a state (i.e., a state's significant contribution and interference with maintenance). It does not direct EPA to use any particular methodology for allocating allowances under a trading program designed to ensure all such emissions are prohibited. As such, EPA believes it is appropriate to allocate allowances based on a neutral factor representing fossil energy content used to produce electricity. Detailed considerations of equity, as the DC Circuit reminded EPA, are not related to the statutory mandate of section 110(a)(2)(D)(i)(I). North Carolina, 531 F.3d 921.

Some commenters objected to the use of a heat input-based approach by arguing that higher-emitting units would not receive an initial allocation sufficient to cover their emissions. EPA does not believe it is reasonable to expect initial allocations to cover each unit's emissions under a trading program aimed at producing meaningful emission reductions. In its administration of prior trading programs such as the ARP and the NBP, EPA has made initial allowance allocations using a heat input-based approach, and virtually all covered sources have successfully complied at the end of each compliance period by making cost-effective emission reductions, purchasing additional allowances through robust markets to cover emissions, or undertaking both types of activities. EPA disagrees with commenters' arguments that allowance allocations should be used to compensate units with higher emissions.

iii. Heat Input Allocation Methodology Option 2

The second heat input option presented by EPA for public comment also would use historic heat input but would apply a constraint to unit-level allocations under certain circumstances. Specifically, under this option unit-level allocations would not be allowed to exceed what EPA determines, based on historic emissions and other factors, to be the units' “reasonably foreseeable maximum emissions.”

To apply this constraint, EPA first would determine whether the allocation to a unit under an unconstrained heat-input methodology would exceed that unit's maximum historic emissions of the relevant pollutant since 2003 “in order to reflect unit-level emissions before and after the promulgation of the CAIR” (76 FR 1115). Using this baseline would enhance the neutrality of the maximum historic emissions data because it would capture the highest emissions of the unit during that period regardless of what fuels it combusted or what pollution control devices were installed and used at any particular time during that period. In other words, a unit's allocation would not be reduced due to a recent decision to switch fuels or install pollution controls.

Second, for this option, EPA then would adjust that maximum historic emissions data by applying a “well-controlled rate maximum,” designed to place “a reasonably foreseeable maximum emissions level reflecting a reasonable upper-bound capacity utilization factor and a well-controlled emission rate that all units (regardless of the type of fuel they combust) can meet for the pollutant” (76 FR 1115). This option would constrain certain units' allocations that, if based solely on historic heat input, would be determined by EPA to be “in excess of their reasonably foreseeable maximum emissions” under the Transport Rule programs (76 FR 1115).

As noted above, commenters offered numerous arguments in favor of using a historic heat input approach. These arguments apply equally to heat input option 1 and heat input option 2. EPA also received numerous comments comparing the two heat input options presented.

Many commenters preferred heat input option 1's reliance purely on historic data as compared with heat input option 2's reliance on that data modified by the application of EPA-determined “reasonable upper bound capacity factors” and “well-controlled emission rates.” Commenters also criticized the complexity of these modification factors in heat input option 2. While EPA believes both options represent viable approaches, the Agency agrees with commenters that the application of these factors increase the complexity of allocation determinations and would adjust unit-specific historic data by applying EPA-created factors generically determined for broad categories of units.

Some commenters suggested that EPA's application of these modification factors could also represent legal vulnerabilities for the Transport Rule. In particular, they were concerned that the capacity factors and well controlled emission rates presented as part of heat input option 2 could be perceived as arbitrary. While EPA does not agree that these modification factors are arbitrary, the Agency does recognize that application of such EPA-created generic factors in determining unit-specific allocations increases the complexity of the allocation approach and raises issues regarding whether such generic factors are appropriately applied to each individual unit.

iv. General Comments on EPA's Authority To Allocate Allowances

Numerous commenters also noted that EPA has generally broad authority in selecting an allocation methodology under CAA sections 110(a)(2)(D)(i)(I) and 302(y). EPA agrees with commenters that the Agency has broad discretion in this area. Neither the CAA nor the D.C. Circuit Court's opinion in North Carolina specifies a particular methodology that EPA must use to allocate allowances to individual units. CAA section 110(a)(2)(D)(i)(I) requires prohibition of emissions “within the state” that significantly contribute to nonattainment or interfere with maintenance and gives states broad discretion to develop a control program in a SIP that achieves this objective. EPA has similarly broad discretion when issuing a FIP to realize this objective. Moreover, while the definition of FIP in CAA section 302(y) clarifies that a FIP may include “enforceable emission limitations or other control measures, means or techniques (including economic incentives, such as marketable permits or auctions of emissions allowances),” this section does not require EPA to use any particular methodology to allocate allowances under a FIP trading program. In light of this lack of direction in the CAA concerning allowance allocation, EPA has broad discretion to select an allocation methodology that is reasonable and consistent with the goals of CAA section 110(a)(2)(D)(i)(I).

The body of public comment makes it clear that no allocation option could be deemed satisfactory from the perspective of all stakeholders. Public comments from most states and industrial stakeholders with a substantial interest in how EPA allocates allowances under the Transport Rule FIPs expressed support for an historical heat input-based approach as opposed to the proposal's emission-based approach. Most commenters favored this historical heat input data basis as the most sound and offered technical data corrections, which EPA considered and generally used in the final rule. EPA believes it is reasonable to select a heat input-based approach for the final Transport Rule because this approach is consistent with the rule's statutory objectives and has been found, when implemented in prior trading programs, to be a credible, workable allocation approach.

b. Final FIP Allocation Methodology

After consideration of all comments, EPA decided to allocate allowances to individual units based on that units' share of the state's historic heat-input, but to ensure that no unit's allocations exceed that unit's historic emissions. EPA decided to use the allocation methodology originally presented as heat input option 2, modified in response to public comments. EPA decided to use heat input option 2 but without the application of the “reasonable upper-bound capacity utilization factor and a well-controlled emission rate” factors. This allocation approach reflects the Agency's response to extensive public comment on the options presented in the proposed Transport Rule and subsequent NODAs and is a logical outgrowth of those actions. EPA is using this approach to allocate allowances under the FIPs for all four trading programs. Further details on the calculation and implementation of this approach are provided below in section VII.D.1.c and can also be found in the Allowance Allocation Final Rule TSD in the docket for this rulemaking.

The principal reasons for this decision are:

• EPA believes that existing-unit allowance allocation under the Transport Rule should not generally advantage or disadvantage units based on the selection of fuels consumed or of pollution controls installed at a given unit in anticipation of either the Clean Air Interstate Rule or the Transport Rule, i.e., fuel or control decisions taken from 2003 onward. An approach that does not advantage or disadvantage units in this way avoids allocating in a way that would effectively penalize units that have already invested in cleaner fuels or other pollution reduction measures that will continue to deliver important emission reductions under this rulemaking. The approach selected in the final rule generally does not penalize such units and is thus generally fuel-neutral and control-neutral in its allocation determinations.
• EPA finds that the selected approach maximizes transparency and clarity of allowance allocations. EPA has already made public the historic heat input and historic emissions data on which this approach is based, and its application to calculate unit-level allocations in each state under that state's emission budgets finalized in this Transport Rule can be relatively easily replicated.
• EPA finds that quality-assured historic CEMS-quality data used to implement this approach represent the most technically superior data available to EPA at the time of this rulemaking for calculating unit-level allocations. The selected approach relies on unmodified historic data reported directly by the vast majority of covered sources, whose designated representatives have already attested to the validity and accuracy of this data. EPA agrees with commenters that allowance allocations should be based on quality-assured data to the maximum extent possible. This approach uses the most accurate data currently available to EPA.
• Heat-input based approaches were used to allocate allowances under both the NO_X Budget Trading Program and the Acid Rain Program. Allocation under these programs was readily and easily administered, and the programs achieved or exceeded their environmental goals. The selected approach's use of heat input as a basis for allocations builds on prior legislative and administrative approaches to allowance allocations for trading programs.
• EPA also finds that the selected approach's addition of a constraint to heat input-based allocations where such allocations would otherwise exceed a unit's maximum historic emissions is a reasonable extension of a heat input-based allocation approach. The Transport Rule trading programs are established to achieve overall emission reductions in each covered state. As a group, covered sources within each state must make the necessary reductions under these programs. In light of each program's goal to reduce each state's overall emissions, it is logical and consistent with that goal that the starting point for each source under these programs—i.e., the initial allocations of shares of the state budget to covered units—be an amount of allowances no greater than each unit's maximum historic emissions. Under the trading programs, any source may emit a ton of SO₂ or NO_X for which it holds a corresponding allowance, which it may acquire either by initial allocation or by subsequent purchase, to the extent consistent with the assurance provisions (discussed elsewhere in this preamble) that ensure achievement of the requisite overall reductions in each state. Consequently, the initial allocations to the units at each source are the starting point for each source's efforts to comply with the allowance-holding and assurance provision requirements, but do not determine the source's strategies for compliance and ultimate level of emissions. EPA believes that a starting point of unit-level heat input-based allocations constrained not to exceed each specific units' maximum historic emissions is reasonable and consistent with the program goals of reducing overall emissions in each state: Each existing unit is allocated an amount that either reflects reduced unit emissions or does not exceed historic emissions, and, from that starting point, the units, as a group, reduce overall emissions to the level required for each state. Conversely, EPA believes that a starting point allocating some units more than they have ever emitted would be illogical in programs aimed at reducing overall emissions.

EPA believes that this selected allocation methodology for the final Transport Rule FIPs is within its authority under the Clean Air Act. Section 110(a)(2)(D)(i)(I) of the CAA requires that emissions “within a state” that significantly contribute to nonattainment or interfere with maintenance in another state be prohibited. In the final Transport Rule, EPA analyzed each individual state's significant contribution and interference with maintenance and calculated budgets that represent each state's emissions after the elimination of prohibited emissions in an average year. The methodology used to allocate allowances in a state budget to individual units in the state has no impact on that state's budget or on the requirement that the state's emissions not exceed that budget plus variability. Regardless of the allocation methodology used, the state's responsibility for eliminating its significant contribution and interference with maintenance remains unchanged. This is reflected by the fact that allocations under each state's budget, regardless of how they are made, cannot change that state's budget. In sum, the allocation methodology has no impact on the final rule's ability to satisfy the statutory mandate of CAA section 110(a)(2)(D)(i)(I) to eliminate significant contribution to nonattainment and interference with maintenance.

Consistent with its broad authority in CAA sections 110(a)(2)(D)(i)(II) and 302(y), EPA believes that data quality, fuel-neutrality, control-neutrality, transparency, clarity, consistency with program goals, and successful experience in previous trading programs are reasonable factors on which to base the selection of an allowance allocation methodology for existing units for the final Transport Rule. EPA believes that the transparency and clarity of this allocation approach builds credibility with the public that the government is distributing a public resource—i.e., allowances—precisely as stated in this rulemaking, with clear execution that can be relatively easily verified.

EPA also believes that the final Transport Rule's heat input-based approach for existing units is consistent with the goals of the Clean Air Act because it allocates allowances to existing units on the basis of a neutral factor that does not advantage or disadvantage a unit based on what fuel the unit burns or whether or not a unit has installed controls in anticipation of these regulations. In contrast, allocations under the proposal's emission-based methodology would give a greater share of allowances to units with higher emission rates, which are generally responsible for a greater share of a state's total emissions. Because these higher-emitting rate units are generally responsible for a greater share of emissions, it follows that they are also responsible for a greater share of a state's significant contribution to nonattainment and interference with maintenance. The proposal's emission-based allocation methodology would disadvantage one of two otherwise identical existing units if it invested in emission reductions in anticipation of the Clean Air Interstate Rule or this final Transport Rule.

The heat-input allocation methodology selected for the final Transport Rule does not have this flaw. In contrast to the proposal's emission-based allocation approach, the heat input allocation methodology selected by EPA yields a smaller proportion of allowances relative to emissions to higher-emission-rate units and a higher proportion of allowances relative to emissions to lower-emission-rate units. For example, assume that in a state with two units and in a baseline year, Unit A combusts 100 mmBtu of heat input and emits 1,000 tons while Unit B combusts 100 mmBtu of heat input and emits only 500 tons. Assume also that this state's future Transport Rule emissions budget for this pollutant is only 500 tons. Because Units A and B each make up an even share of historic heat input for the state, the final rule's heat input-based approach would allocate the same share of allowances (250 tons) to each unit. In this example, Unit A's initial allocation of 250 is a smaller proportion of its historic emissions (25 percent of its baseline 1,000-ton emissions), while Unit B's initial allocation of 250 is a larger proportion of its historic emissions (50 percent of its baseline 500-ton emissions). Therefore, Unit B's ability to emit fewer tons per mmBtu of heat content used for generating electricity (as compared with Unit A) results in Unit B receiving a larger proportion of its historic emissions as an initial allocation share than Unit A receives.

This relative distributional pattern yielded is consistent with the goals of CAA section 110(a)(2)(D)(i)(I) because under this distribution, higher-emitting units, which are responsible for a greater share of the state's significant contribution to nonattainment and interference with maintenance, would require relatively more allowances in order to cover their pre-existing emissions than would lower-emitting units. EPA believes this initial allocation pattern is an appropriate reflection of the goals of CAA section 110(a)(2)(D)(i)(I).

The heat input-based allowance methodology selected by EPA is fuel-neutral, control-neutral, transparent, based on reliable data, and similar to the allocation methodologies used in the NO_X SIP Call and Acid Rain Program. For all these reasons, EPA determined that it is appropriate to use a heat input-based allocation methodology in this rule.

In addition, this allocation methodology is similar to an output-based allocation approach, which would base allocations on the quantity of electricity generated (rather than energy content combusted) and would also be fuel-neutral, control-neutral, and able to reward generation units that operate the most efficiently. Many state and industry commenters advocated using an output-based approach due to its reported strong value in promoting efficiency. However, at this time EPA does not have access to unit-level output data that is as quality-assured or comprehensive as its data sets on heat input across the units considered. Therefore, EPA is using a heat input-based approach under the Transport Rule in part due to its ability to serve as a reasonable proxy for an output-based standard using the most quality-assured data that EPA has to date.

In the NODA, EPA noted that final state budgets and allocations may differ from the proposed budgets and allocations because EPA was still in the process of updating its emission inventories and modeling in response to public comments, including comments on IPM. Thus, unit-level allocations in the NODA provided an indication of the proportional share of a state's budget that would be allocated to individual existing units if the alternative methodologies were used. The allocations made final today are based on budgets that reflect the updated modeling and comments received during the comment period.

c. Calculation of Existing Unit Allocations Under the Final Transport Rule FIPs

Allocations under this final methodology for each existing unit are determined by applying the following steps.

1. For each unit in the list of potential existing Transport Rule units, annual heat input values for the baseline period of 2006 through 2010 are identified using data reported to EPA or, where EPA data is unavailable, using data reported to the Energy Information Administration (EIA). For a baseline year for which a unit has no data on heat input (e.g., for a baseline year before the year when a unit started operating), the unit is assigned a zero value. (Step 2 explains how such zero values are treated in the calculations.) The allocation method uses a 5-year baseline to approximate a unit's normal operating conditions over time.

2. For each unit, the three highest, non-zero annual heat input values within the 5-year baseline are selected and averaged. Selecting the three highest, non-zero annual heat input values within the five-year baseline reduces the likelihood that any particular single year's operations (which might be negatively affected by outages or other unusual events) would determine a unit's allocation. If a unit does not have three non-zero heat input values during the 5-year baseline period, EPA averages only those years for which a unit does have non-zero heat input values. For example, if a unit has only reported data for 2008 and 2009 among the baseline years and the reported heat input values are 2 and 4 mmBtus, respectively, then the unit's average heat input used to determine its pro-rata share of the state budget is (2+4)/2 = 3.

3. Each unit is assigned a baseline heat input value calculated as described in step 2, above, referred to as the “3-year average heat input.”

4. The 3-year average heat inputs of all covered existing units in a state are summed to obtain that state's total “3-year average heat input.”

5. Each unit's 3-year average heat input is divided by the state's total 3-year average heat input to determine that unit's share of the state's total 3-year average heat input.

6. Each unit's share of the state's total 3-year average heat input is multiplied by the existing-unit portion of the state budget (i.e., the state budget minus the state's new unit set-aside and, if applicable, minus the Indian country new unit set-aside) to determine that unit's initial allocation.

7. An 8-year (2003-2010) historic emissions baseline is established for SO₂, NO_X, and ozone-season NO_X based on data reported to EPA or, where EPA data is unavailable, based on EIA data. This approach uses this 8-year historic emissions baseline in order to capture the unit-level emissions before and after the promulgation of CAIR.

8. For each unit, the maximum annual historic SO₂ and NO_X emissions are identified within the 8-year baseline. Similarly, the maximum ozone season NO_X emissions from the 8-year baseline for each unit are identified. These values are referred to as the “maximum historic baseline emissions” for each unit.

9. If a unit has an initial historic heat-input based allocation (as determined in step 6) that exceeds its maximum historic baseline emissions (as determined in step 8), then its allocation equals the maximum historic baseline emissions for that unit.

10. The difference (if positive) under step 9 between a unit's historic heat-input-based allocation and its “maximum historic baseline emissions” is reapportioned on the same basis as described in steps 1 through 6 to units whose historic heat-input-based allocation does not exceed its maximum historic baseline emissions. Steps 7, 8, and 9 are repeated with each revised allocation distribution until the entire existing-unit portion of the state budget is allocated. The resulting allocation value is rounded to the nearest whole ton using conventional rounding.

Table VI.D-1 below provides an illustrative application of the steps 1-10 in a hypothetical state.

2. Allocations to New Units

EPA is finalizing—similar to the proposal (75 FR 45310)—an approach to allocate emission allowances to new units from new unit set-asides in each state. A “new unit” may be any of the following: (1) A covered unit commencing commercial operation on or after January 1, 2010; (2) any unit that becomes a covered unit by meeting applicability criteria subsequent to January 1, 2010; (3) any unit that relocates into a different state covered by the Transport Rule; and (4) any existing covered unit that stopped operating for 2 consecutive years but resumes commercial operation at some point thereafter.

The proposed Transport Rule would have required that owners and operators initially request allowances from the new unit set-aside when the unit first became eligible for an allocation. EPA now believes that it can identify which units become eligible and when they become eligible, based on information provided in other submissions (e.g., certificates of representation, monitoring system certifications, and quarterly emissions reports) that the final rule already requires such units to make to EPA. EPA concludes that requiring owners and operators to submit requests of new unit set-aside allocations would impose an unnecessary burden on the owners and operators, as well as on EPA, and therefore EPA has removed this requirement in the final rule.

The following sections describe the methodology in the final Transport Rule for allocating to new units, how EPA determined the size of new unit set-asides in the final rule, and how EPA has provided for allocations to new units that locate in Indian Country.

a. New Unit Allocation Methodology

The proposal's new unit allocation methodology did not provide any allocation for a new unit's first control period of commercial operation. Some commenters expressed concern about the lack of new unit allocations the first year of commercial operation. In order to address this concern, EPA is modifying the new unit allocation methodology in this final rule to include allocations to new units for the first control period in which the units are in commercial operation, as well as for control periods in subsequent years.

The final rule's allocation to new units is performed in two “rounds.” The first round is the same as the new unit allocation procedures in the proposal (except for elimination of the requirements that owners and operators request the allocations) and occurs during the control period for which the allocations are made. These first round allocations are based on new unit emissions during the prior control period and are recorded in allowance accounts in the Allowance Management System for the units by August 1 of each control period. For example, for the 2012 vintage year, “first-round” allocations would be made to new units by August 1, 2012 based on their emissions in the 2011 control period (as monitored and reported in accordance with Part 75 of the Acid Rain Program regulations). If the new unit set-aside is insufficient to accommodate first round allocations reflecting all new units' prior control period emissions, the first round allocations are made pro rata to new units based on their share of total new unit emissions in the prior control period.

The second round of allocations accommodates new units that come online during the control period for which the allocations are made and did not therefore receive any allocation in the first round. The second round also accommodates new units that come online partway into the prior control period and therefore received an allocation in the first round that did not extend to cover operations in a full control period. This second round of new unit allocation is therefore applicable only to new units coming online either during the control period of the allocation or during the control period immediately prior. New units coming online earlier than the previous control period only receive first-round allocations from the new unit set-asides, as first-round allocations to those units are based on operational data spanning an entire control period.

Second-round allocations are based on new unit emissions during the same control period as the vintage year of the allowances allocated. For example, for the 2012 vintage year, “second-round” allocations are based on the difference between the new unit's emissions in the 2012 control period and the new unit allocation (if any) that the unit received in the first round of allocations. For a unit coming online in 2012, this amount equals its total emissions during the 2012 control period. For a unit coming online in 2011, this amount equals its incremental emissions in 2012 beyond its emissions in 2011, as such a unit would have already received a first-round allocation from the new unit set-aside based on its emissions in 2011. Second-round allocations are recorded in allowance accounts by November 15 for the NO_X ozone season trading program (ahead of the December 1 compliance deadline) and by February 15 of the following calendar year for NO_X and SO₂ annual trading programs (ahead of the March 1 compliance deadline).

This methodology only allocates in the second round whatever allowances remain in the new unit set-asides after the first-round allocations have been recorded. If the new unit set-aside available for second round allocations is insufficient to accommodate allocations based on the difference between control period emissions and any first round allocations for the units involved, then the second round allocations are made pro rate to the new units based on their share of the total of such differences.

b. Determination of New Unit Set-Asides

The proposed Transport Rule identified new units using a threshold online date of January 1, 2012, whereas the final Transport Rule uses a threshold online date of January 1, 2010. As explained above, EPA adjusted this cutoff date because the final Transport Rule's allocation methodology for existing units requires that EPA possess at least 1 full year of historic data in order to calculate allocations. As a consequence, EPA recognizes that the proposal's methodology to determine the size of the new unit set-asides based only on new EGUs forecast by the model would fail to account for known EGUs that have come online, or are planned to come online, after January 1, 2010. Therefore, EPA has modified its approach to determining the size of the new unit set-asides in the final rule to account for both “potential” units (i.e., those that are not yet planned or under construction but are projected by modeling to be built) and ”planned” units (i.e., those that are known units with planned online dates after January 1, 2010). EPA uses the distinction between “potential” and “planned” new units to determine the ultimate size of each state's new unit set-aside (as a percentage of that state's budgets for each pollutant covered); however, the new unit allocation methodology described above applies the same to “potential” and “planned” new units.

The first step of EPA's analysis to determine the new unit set-asides accounts for likely future emissions from potential units, and its methodology is taken directly from the Transport Rule proposal but reflects updated modeling (see “Allowance Allocation to Existing and New Units Under the Transport Rule Federal Implementation Plans” TSD for detailed findings). This analysis informed EPA's decision to establish a minimum new unit set-aside size of 2 percent of each state's budget for each pollutant that is configured to accommodate future emissions from potential units.

For the final rule, EPA augmented its new unit set-aside determination to account for “planned” units through an additional step. Because the location of these “planned” units is known and identified in EPA modeling, this second step is a state-specific modification of the size of the new unit set-asides. That is, EPA only increased new unit set-asides above the 2 percent minimum established in the first step for states that had additional known units coming online between January 1, 2010, and January 1, 2012.

The increases made to the new unit set-asides for these planned units reflect the projected emissions from these units. Therefore, if the expected emissions of a given pollutant from all “planned” new units in a given state were equal to 3 percent of that state's budget for that pollutant, then EPA added that amount to the base 2 percent new unit set-aside (creating a hypothetical new unit set-aside of 5 percent for that pollutant in that state). See “Allowance Allocation to Existing and New Units Under the Transport Rule Federal Implementation Plans” TSD for detailed results showing how EPA determined the size of each new unit set-aside reflecting the application of both of the steps described above. This approach to determining the size of state new unit set-asides is a logical outgrowth of the proposal, the NODA on allowance allocations, and updated modeling results. In fact, EPA received comments that using a January 1, 2010 cutoff date for distinguishing between existing and new units would result in the new unit set-aside, as proposed, being insufficient to meet the needs of units already under construction. EPA believes that the approach adopted in the final rule results in new unit set-asides that reasonably accommodate the foreseeable emissions from both planned and potential new units in each state.

The new unit allocation percentages for each state are shown in Table VII.D.2-1.

c. Procedures for Allocating New Unit Set-Asides

For the first round of new unit set-aside allocations, the Administrator will promulgate a notice of data availability informing the public of the specific new unit allocations and provide an opportunity for submission of objections on the grounds that the allocations are not consistent with the requirements of the relevant final rule provisions. A second notice of data availability will subsequently be promulgated in order to make any necessary corrections in the specific new unit allocations. As discussed elsewhere in this preamble, the final rule establishes a different schedule for promulgation of these notices of data availability than the proposed rule. In particular, a single set of deadlines (i.e., for the first notice in the first round of allocations, June 1 of the year for which the new unit allocations are described in the notice and, for the second notice of the first round, August 1 of that year) for promulgation of the notices is established for all of the Transport Rule trading programs. EPA believes that these deadlines will provide sufficient time for EPA to obtain final emissions data for the prior year for the units involved and to calculate the allocations and promulgate the notices. Further, the approach of using the same deadline for all of the Transport Rule trading programs will simplify EPA's implementation and reduce the complexity of the process for source owners and operators.

For the second round of new unit set-aside allocations, the Administrator will also promulgate two notices of data availability. However, the deadlines for the notices differ for the NO_X ozone season trading program and for the SO₂ and NO_X annual trading programs because control period emissions data (used in making second round allocations) become available sooner, and the compliance deadline for holding allowances covering emissions is sooner, in the NO_X ozone season trading program. The control period in the NO_X ozone season program ends on September 30, and fourth quarter emissions reports must be submitted to EPA by October 30, while the control periods in the SO₂ and NO_X annual programs end on December 31 and fourth quarter emission reports are due by January 30. Further, in order for the second round allocations to be available to be used for compliance with the allowance-holding requirement, the second round needs to be completed before the compliance dates, which are December 1 in the NO_X ozone season program and March 1 in the SO₂ and NO_X annual programs. Consequently, for the NO_X ozone season program the Administrator will promulgate by September 15 a notice of data availability identifying the units eligible for second round allocations and by November 15 a second NODA of the list of eligible units and their second round allocations, which will also be recorded in the allowance accounts by that date. The comparable deadlines for the SO₂ and NO_X annual programs are December 15 and February 15. EPA believes that these deadlines will provide sufficient time for EPA to identify the units and obtain their needed emissions data and to calculate the allocations and promulgate the notices.

d. Addition of Allowances to New Unit Set-Asides

As discussed elsewhere in this preamble, EPA proposed that, if a unit with an existing-unit allocation does not operate for 3 consecutive years, the allowances that would otherwise have been allocated to that unit, starting in the seventh year after the first year of non-operation, would be allocated to the new unit set-aside for the state in which the retired unit is located. EPA is retaining this provision in the final rule but is changing the time of non-operation to 2 years and the time of allowance allocation to a non-operating unit to 4 years. Starting in the fifth year of non-operation, allowances will be allocated to the new unit set-aside for the state in which the non-operating unit is located.

EPA received comments that the new unit set-asides were not sufficient to encourage the operation of new units. One commenter suggested that allowance allocations should cease after 3 years of non-operation because the financial incentive gained from receiving allowances beyond the 3-year period is insignificant relative to operating and fuel costs. Another commenter said that providing allowances to non-operating units is unnecessary and distorts the market.

In addition to increasing the size of the new unit set-aside in this final rule, as described above, EPA is terminating existing unit allocations starting in the fifth year after the unit does not operate for 2 consecutive years and reallocating to the new unit set-aside the allowances that the unit otherwise would have received for the fifth and subsequent years in order to make them available for new units in the state. This approach allows the new unit set-asides to grow over time.

e. Allocations to New Units Locating in Indian Country

EPA received several comments on the proposed rule that it did not explicitly address the distribution of allowances to potential new units built in Indian country. EPA recognized this concern and requested comment on this topic in the January 7, 2011 NODA.

In the final rule, EPA is providing a mechanism to make allowances available in the future for new units built in Indian country. The final rule establishes an Indian country new unit set-aside for each pollutant in each state whose borders encompass Indian country (i.e., Florida, Iowa, Kansas, Louisiana, Michigan, Minnesota, Mississippi, Nebraska, New York, North Carolina, South Carolina, Texas, and Wisconsin). EPA will retain administration of these Indian country new unit set-asides as part of the Transport Rule trading programs whether or not a Transport Rule state elects to modify or replace the Transport Rule FIPs through approved SIP revisions. EPA does not create Indian country new unit set-asides for states lacking Indian country within their borders.

EPA determined the size of each Indian country new unit set-aside by calculating the ratio of square mileage of Indian country to the square mileage of the state within whose borders Indian country is located. This calculation yielded a maximum percentage of 5 percent when assessing all of the states encompassing Indian country subject to the final Transport Rule; this is referred to as the “5 percent Indian country factor” below. To determine the maximum percentage, EPA used the American Indian Reservations/Federally Recognized Tribal Entities dataset, which contains data for the 562 federally recognized tribal entities in the contiguous U.S. and Alaska. EPA accessed the data to analyze the Transport Rule region and compare the square miles of Indian country with the square miles of the Transport Rule state that includes the Indian country. EPA then took the highest percentage as the number to be applied across all states with Indian country to determine the size of the Indian country new unit set-aside pertinent to that state's budgets under the Transport Rule. EPA chose to use the maximum percentage (5 percent) from the Indian country analysis to determine the Indian country set-aside for each state on the basis that this approach would reserve a reasonable number of allowances from each state's budget for potential allocation to new units that may locate in Indian country within that state's borders. Any allowances from the Indian country new unit set-aside that are not allocated in a given control period are redistributed into the state's new unit set-aside. As discussed above, any allowances not allocated from that new unit set-aside are redistributed to existing units based on the existing units' share of the total existing unit allocations.

To calculate the size of each tribal new unit set-aside, EPA applied this 5 percent Indian country factor to the portion of the state's new unit set-aside originally determined by accounting for “potential” new units, which as described above was set at 2 percent of each pollutant's budget in each state. Therefore, the Indian country new unit set-aside is 5 percent of 2 percent of a state's budget, or 0.1 percent of that total state budget. EPA did not apply the 5 percent Indian country factor to the state-specific planned unit portion of each state's new unit set-aside because the planned unit portion is determined using projected emissions from specific, known units coming online after January 1, 2010, and none of these known units are located in Indian country.

The Indian country new unit set-asides in the following Transport Rule states with Indian Country are shown in Table VII.D.2-2.

Under the FIPs, EPA allocates allowances from Indian country new unit set-asides in essentially the same manner as it allocates allowances from state new unit set-asides. The approach for identifying, and determining the number of allowances allocated to, new units in Indian country is the same as the approach for identifying and determining allocations for non-Indian country new units covered by the state new unit set-aside, and allocations are made in two rounds using the same schedules for promulgation of notices of data availability. However, as discussed above, unallocated allowances in the Indian country set-asides are handled differently from unallocated allowances in the state new unit set-asides in that unallocated Indian country new unit set-aside allowances are first transferred back into the state new unit set-aside and then, if still not allocated to new units, are distributed to existing units in the state. EPA believes that the above-described approach in establishing and handling the Indian country new unit set-asides and state new unit set-asides is a reasonable way of making a sufficient amount of allowances available for new units in the state and Indian country located in the state and ensuring that the entire state budget is available to either new or existing units in the state and Indian country. EPA retains administration of these Indian country new unit set-asides (and, of course, the portions of state budgets that comprise these set-asides) as part of the Transport Rule trading programs even if a state elects to modify or replace the Transport Rule FIPs through approved SIP revisions. EPA continues to manage and distribute the Indian country new unit set-aside allowances in the same manner as under the FIPs. Unallocated allowances in the Indian country new unit set-aside will be returned to the portion of the state budget allocated under the approved SIP's allocation provisions. EPA believes that this approach is reasonable because EPA, rather than the states, has the authority and responsibility of administering the Transport Rule with regard to new units that locate in Indian country.

E. Assurance Provisions

To ensure that the FIPs require the elimination of all emissions that EPA has identified that significantly contribute to nonattainment or interfere with maintenance within each individual state, the Agency is adopting assurance provisions in addition to the requirement that sources hold allowances sufficient to cover their emissions. These assurance provisions limit emissions from each state to an amount equal to that state's trading budget plus the variability limit for that state (i.e., the state assurance level). As discussed in section VI of this preamble, this variability limit takes into account the inherent variability in baseline EGU emissions and recognizes that state emissions may vary somewhat after all significant contribution to nonattainment and interference with maintenance are eliminated. This approach also provides sources with flexibility to manage growth and electric reliability requirements, thereby ensuring the country's electric demand will be met, while meeting the statutory requirement of eliminating significant contribution to nonattainment and interference with maintenance.

Starting in 2012, EPA is establishing, as part of the FIPs, limits on the total emissions that may be emitted from EGUs at sources in each state. For any single year, the state's emissions must not exceed the state budget with the variability limit allowed for any single year for that state (i.e., the state's 1-year variability limit). In other words, in addition to covered sources being required to hold allowances sufficient to cover their emissions, the total sum of EGU emissions in a particular state cannot exceed the state budget with the state's 1-year variability limit in any 1 year (i.e., the state's assurance level). EPA is not finalizing 3-year variability limits that were included in the proposal for the reasons explained previously in section VI.E of this preamble. The state budgets, variability limits, and state assurance levels for each state are shown in Tables VI.F-1, VI.F-2 and VI.F-3 in section VI.F of this preamble. The basis for the variability limits is also described in section VI.E of this preamble. Additional details may be found in the Power Sector Variability Final Rule TSD in the docket to this rule.

To implement this requirement, EPA first evaluates whether any state's total EGU emissions in a control period exceeded the state's assurance level. If any state's EGU emissions in a control period exceed the state assurance level, then EPA applies additional criteria to determine which owners and operators of units in the state will be subject to an allowance surrender requirement. In applying the additional criteria, EPA evaluates which groups of units at the common designated representative (DR) level had emissions exceeding the respective common DR's share of the state assurance level (regardless of whether the source had enough allowances to cover its emissions) during the control period.

The requirement that owners and operators surrender allowances under the assurance provisions will be triggered only if two criteria are met: (1) The group of sources and units with a common DR are located in a state where the total state EGU emissions for a control period exceed the state assurance level; and (2) that group with the common DR had emissions exceeding the respective DR's share of the state assurance level. The share of the assurance penalty borne by the owners and operators will be based on the amount by which the total emissions for the units in the group exceed the common DR's share of the state assurance level as a percentage of the total calculated for all such groups of sources and units in the state. Thus, the owners and operators of each such group of sources and units must surrender an amount of allowances equal to the excess of state EGU emissions over the state assurance level multiplied by the owners' and operators' percentage and multiplied by two (to reflect the penalty of two allowances for each ton of the state's excess EGU emissions). See Table VII.E-1 below for an illustrative example.

This approach in the final rule of implementing the assurance provisions on a common designated representative basis contrasts with the approach in the proposed rule of implementing the assurance provisions on an owner basis. In the January 7, 2011 NODA, EPA requested comment on the alternative of basing the assurance provision penalty using common designated representatives, and some commenters supported this alternative. The common designated representative approach is simpler and avoids the need to collect information on percentage ownership (which information is not used in any other provisions of the Transport Rule trading programs).

In addition, the common designated representative approach provides additional flexibility to owners and operators who have only one or a few units in a given state but have the option of selecting a common designated representative with owners and operators of other units in the state. EPA expects companies in various states will readily be able to manage their emissions to stay collectively below their state's assurance levels as they track emissions quarterly throughout the year and manage their generation units and pollution control efforts accordingly. However, if the state appears to be approaching its assurance level, this final rule also gives companies the ability to further ensure that they will not have excess emissions by combining multiple units under a common DR. This flexibility allows utilities to re-balance allowances and emissions to mitigate penalty risk if the state violates its assurance level. In a state that does not appear to risk violating its assurance level in a given period, utilities would not need to consider the assurance aspect of selecting DRs. However, EPA anticipates that in the event utilities desire additional certainty or mitigation of assurance penalty risk, they will take advantage of this common DR provision or pursue similar private arrangements with each other to cover their emissions at the lowest possible cost.

While the DR provision could benefit utilities by allowing them to pool their penalty risk, the utilities would still be subject to the antitrust laws. As with any joint venture between competitors, the efficiency benefits of pooling risk would be weighed against any anticompetitive harm associated with DRs.

This new feature in the final rule, in conjunction with the simplifications to the final rule's variability limits described in section VI.E, will give companies under the air quality-assured trading program greater flexibility in each state to determine the most cost-effective pattern of emission reductions while EPA ensures each state meets its assurance level needed to address the significant contribution in each state.

In the January 7, 2011 NODA, EPA also requested comment on continuing to link allocations to assurance provision allowance surrender requirements. Even though the final rule uses a different allowance allocation methodology than the allocation methodology that was proposed, the final rule continues to treat the groups of units with greater emissions than their allocations plus share of state variability as responsible for the state's excess of emissions over the state assurance level. EPA believes that this approach is reasonable because any state that exceeds its state assurance level likely does so because not all units have made the reductions necessary to eliminate the state's contribution to nonattainment or interference with maintenance. Moreover, the groups of units with emissions exceeding their allocations plus share of variability are the units most likely to have contributed to the state's exceedance of its state assurance level and thus to the state's triggering of the assurance provisions. Consequently, EPA concludes that it is reasonable to penalize owners and operators of those sources and units (grouped by common DR) for the state's exceedance through application of the assurance provision allowance surrender requirement. Some commenters stated that this is a reasonable approach.

While a few commenters suggested alternative approaches to the assurance provisions, EPA believes that the suggested alternatives are not workable and are likely to create implementation problems. These commenters suggested variations of approaches that would have created state-specific and vintage year-specific allowances that would have been traded independently of compliance allowances. These differentiated allowances would have fragmented the allowance markets and made the programs resemble the intrastate trading option that EPA rejected because of market power and other concerns described in the proposal.

The existence of the assurance provisions with significant penalties imposed if a state's emissions exceed the state budget with the variability limit, along with other features of the Transport Rule trading programs discussed below, will ensure that state emissions stay below the level of the budget with the variability limit. In making compliance decisions and determining to what extent to rely on purchased or banked allowances, owners and operators will have to take into account the risk of triggering the assurance provisions in the state involved and of incurring significant assurance provision penalties. The greater the extent to which units sharing a common DR have emissions exceeding the DR units' allocations plus share of the state variability limit, the greater the risk of being subject to the assurance provision penalties.

As discussed previously in section VII.D.2, EPA allocates allowances to a new unit for the control period during which the unit commences commercial operation from the new unit set-aside based on its emissions. In the case where assurance provisions for a state are triggered in the year that a new unit commences operation, the unit's share of the state assurance level is calculated using the unit's allocation from the new unit set-aside plus its proportional share of the variability limit. There is the possibility that a new unit would receive no allocation for the control period during which the unit commences commercial operation. EPA sees no reasonable basis for disadvantaging owners and operators because they started up a new unit and EPA had no emissions data on which to base an allocation from the new unit set-aside or no allowances were available for the unit in the state's new unit set-aside. For these new units, EPA would use a specific surrogate number to calculate the maximum amount of emissions that the unit would likely have had during that year. The surrogate emission number applies only if the state's assurance provisions are triggered and only in the first year of the new unit's commercial operation for a new unit that did not receive an allocation from the set-aside. The methodology for calculating the surrogate emission number is essentially unchanged from the proposal (75 FR 45313). For more details on capacity factors for new units, see “Capacity Factors Analysis for New Units Final Rule TSD.”

These assurance provisions are above and beyond the fundamental requirement for each source to hold enough allowances to cover its emissions in the control period. Failure to hold enough allowances to cover emissions is a violation of the CAA, subject to an automatic penalty and discretionary civil penalties, as described in section VII.F of this preamble.

Several features of the air quality-assured trading programs work in conjunction with the assurance provisions to ensure state emissions do not exceed state assurance levels. The air quality-assured trading programs have: State-specific budgets that do not include the variability limits and that are the basis for allocating allowances in each state so that total allocations in a state cannot exceed the state budget; a requirement that owners and operators of each source hold enough allowances to cover source emissions for each control period; assurance provisions that require owners and operators to hold a significant amount of additional allowances in a state if the assurance provisions are triggered; and additional penalties for failing to hold sufficient allowances under the assurance provisions. The underlying mechanism of cap and trade—with a cap on allowances issued and a requirement to hold allowances covering emissions—has succeeded, even without assurance provisions, in broadly reducing emissions below allowance allocation levels. The accumulated data, history, and experience from cap and trade programs underscore that emission reduction requirements and environmental and public health goals of the programs have been met and, in many instances, exceeded. Additionally, EPA has now added assurance provisions to ensure that emissions within a state do not exceed the state budget with the variability limitation that eliminates the state's significant contribution to nonattainment and interference with maintenance in downwind states.

Emissions from a common DR's group of units in excess of the DR's share of the state budget with the variability limit are not a violation of the rule or the CAA, but do lead to strict allowance surrender requirements. Specifically, the owners and operators with a common DR will be required to surrender two allowances for each ton of their proportional share of the exceedance of the state budget with the variability limit. Failing to hold sufficient allowances to meet the allowance surrender requirement will be a violation of the regulations and the CAA and subject to discretionary civil penalties under CAA section 113. Allowances surrendered to meet an assurance provision penalty may be from the year immediately following the control period in which the state assurance level was exceeded (i.e., the year during which the penalty is assessed) or any prior year. Any future vintage allowances beyond the year in which the penalty is assessed may not be used to meet an assurance provision penalty.

This penalty level is a change from the proposal, in which one allowance was to be surrendered for each ton of emissions over the state assurance level. EPA ran an IPM modeling scenario in order to assess the level of penalty that would be sufficient to deter sources from exceeding state assurance levels. According to the model, no state would exceed its assurance level and incur the two-for-one allowance penalty in either 2012 or 2014, although some states emit up to the assurance level. The two-for-one allowance surrender requirement is significant, and EPA believes that this penalty—along with the other elements of the Transport Rule discussed above—will be sufficient to ensure that the state emissions will not exceed the budgets plus the variability limits. See the Assurance Penalty Level Analysis Final Rule TSD for further details of the analysis.

Below are examples of how the penalty will be assessed for four common designated representatives in the same state if the assurance provisions are triggered. In the first case, DR1's combined units were allowed to emit up to 71 tons of SO₂ (60 * 118 percent), but actually emitted 75 tons during the control period, or 4 more than their share of the state assurance level. Since the state, as a whole exceeded the state assurance level by 15 tons, DR1's share of the penalty is 25 percent of the total penalty, or 8 allowances (25 percent of 30).

In the proposal, EPA took comment on whether assurance provisions should be implemented starting in 2012 or 2014. While a number of commenters supported the proposal to start in 2014, EPA received several comments making the case that starting assurance provisions in 2012 would be more compatible with the Court's opinion in North Carolina, which emphasized EPA's obligation to require elimination of emissions within the states that significantly contribute to nonattainment or interfere with maintenance. In this final rule, EPA makes the assurance provisions effective starting in 2012 because this approach provides even further assurance, consistent with North Carolina, that each state's prohibited emissions will be eliminated from the start of the Transport Rule trading programs.

F. Penalties

Under the final Transport Rule FIPs (like under the proposed rule), the owners and operators of each covered source must hold, as of the allowance transfer deadline, an allowance for each ton of SO₂ or NO_X emitted by the source and are subject to penalties if they fail to comply with this allowance-holding requirement.

In particular, the owners and operators must hold in the source's compliance account in the Allowance Management System enough allowances issued for the respective Transport Rule annual trading program (SO₂ Group 1, SO₂ Group 2, or annual NO_X program) to cover the annual emissions of the relevant pollutant from all covered units at the source. The allowances must have been issued for the year in which the emissions occurred or a prior year. If the owners and operators fail to meet this allowance-holding requirement, they must provide—for deduction by the Administrator from the source's compliance account—one allowance as an offset, and one allowance as an excess emissions penalty, for each ton of emissions (i.e., excess emissions) in excess of the amount of allowances held. The allowances surrendered for the excess emissions penalty must be allocated for the control period in the year immediately following the year when the excess emissions occurred or for a control period in any prior year. The offset and the excess emissions penalty are automatic requirements in that they must be met without any further action by EPA (e.g., any additional proceedings) regardless of the reason for the occurrence of the excess emissions. In addition, each ton of excess emissions, as well as each day in the averaging period (i.e., the control period of one calendar year), constitute a violation of the CAA, and the maximum discretionary civil penalty is $25,000 (inflation-adjusted to $37,500 for 2010) per violation under CAA section 113. This means that, if a source has emissions in excess of allowances held for the source as of the allowance transfer deadline for a control period, the number of tons of excess emissions multiplied by the total number of days in that control period and multiplied by $25,000 (inflation adjusted) equals the maximum discretionary civil penalty for that occurrence of excess emissions.

For the ozone-season NO_X trading program, the same provisions apply as for an annual program, except that the averaging period (i.e., the control period) is the ozone season, not a calendar year. Consequently, the relevant emissions are for an ozone season, the allowances usable to meet the allowance-holding requirement are allowances issued for Transport Rule ozone-season NO_X trading program for the ozone season involved or a prior ozone season, and the number of days used in calculating the maximum civil penalty is the number in the ozone season.

Commenters expressed concern that the proposed FIPs expressly stated that, for purposes of determining the maximum discretionary civil penalty for failure to meet the allowance-holding requirement, each ton of emissions lacking a held allowance would be a violation and each day in the averaging period involved would be a violation. Some commenters compared the proposed penalty provisions for excess emissions with the excess emissions penalty provisions under the Acid Rain Program and claimed that the proposed penalty provisions differed from the Acid Rain Program provisions and were excessive.

In fact, however, the final FIP provisions concerning discretionary civil penalties are essentially the same as those under the Acid Rain Program, as well as those under the NO_X Budget Trading Program and the CAIR trading programs. In particular, the Acid Rain Program regulations state that each ton of SO₂ excess emissions constitutes “a separate violation” of the CAA. 40 CFR 72.9(c)(2). Moreover, while the Acid Rain Program regulations do not expressly address that each day in the averaging period (i.e., a calendar year control period under the Acid Rain Program) constitutes a separate violation when a unit has excess emissions for the calendar year, the courts have addressed this question. In decisions applying the discretionary civil penalty provisions in section 309(d) of the Clean Water Act, which are analogous to the civil penalty provisions in CAA section 113, the courts have interpreted the provisions to mean that, when a source violates the emission limitation for a multi-day control period, the source has a violation for each day in the control period, as well as for each ton of excess emissions on each such day. See, e.g., Chesapeake Bay Foun. v. Gwaltney of Smithfield, 791 F.2d 304, 313-15 (4th Cir. 1986), Atlantic States Legal Foun. v. Tyson Foods, 897 F.2d 1128, 1139-40 (11th Cir. 1990), and U.S. v. Allegheny Ludlum Corp., 366 F.3d 164, 169 (3d. Cir. 2004). As noted by the courts, the treatment of each ton and each day as a separate violation is used for purposes of setting the maximum discretionary civil penalty. Because CAA section 113 sets the maximum civil penalty, EPA, of course, has the discretion to tailor the penalty amount that it seeks in any specific occurrence of excess emissions to reflect the circumstances of that excess emission occurrence. See 42 U.S.C. 7413(b) (stating that the Administrator may commence a civil action “to assess and recover a civil penalty of not more than $25,000 per day for each violation”). Moreover, when a district court imposes a civil penalty, the court “retains discretion to assess a penalty much smaller than the maximum, as the situation requires.” Chesapeake Bay, 791 F.2d at 316. In addition, the Acid Rain Program regulations state that any allowance deduction, excess emission penalty, or interest under the Acid Rain Program regulations “shall not affect liability” of the owners and operators “for any additional fine, penalty, or assessment, or their obligation to comply with any other remedy, for the same violation, as ordered under the [CAA],” including under CAA section 113 providing for discretionary civil penalties. 40 CFR 77.1(b). In summary, under the Acid Rain Program, each ton of excess emissions and each day in the averaging period (i.e., the calendar year) constitute a violation, the resulting number of violations times $2,000 is the maximum civil penalty for violating owners and operators, and EPA has the discretion to impose a civil penalty at or below such maximum, in addition to the automatic requirement to surrender one allowance and pay $2,000 (inflation adjusted) for each ton of excess emissions.

The final FIPs take an analogous approach to that under the Acid Rain Program. Specifically, the final FIPs state both that each ton of excess emissions is a violation of the CAA and that each day in the averaging period (i.e., a calendar year under the annual programs and the ozone season under the ozone-season program) is a violation. Moreover, the imposition of civil penalties at or below the maximum amount resulting from the maximum penalty calculation is in addition to the automatic allowance surrender and penalty totaling 2 allowances per ton of excess emissions. Thus, commenters' assertion that the approach in the final FIPs is inconsistent with the approach in the Acid Rain Program is incorrect. Moreover, EPA has taken this same general approach in two other trading programs (i.e., the NO_X Budget Trading Program and the CAIR trading programs), whose regulations explicitly state that each ton and each day of the averaging period constitute a violation. See 40 CFR 96.54(d)(3) (NO_X Budget Trading Program); and 40 CFR 96.106(d) (CAIR).

In any event, EPA maintains that the approach of treating each excess emission ton and each day in the averaging period as a violation for purposes of calculating the maximum discretionary civil penalty is reasonable. Some commenters suggested that only the days on which a source's cumulative control period emissions exceed the amount of allowances that the source then holds for that control period should be treated as a violation. However, this suggested approach makes little sense in the context of the Transport Rule trading programs.

In order to provide owners and operators compliance flexibility, the Transport Rule trading programs do not require source owners and operators to hold any amount of allowances to cover emissions until the allowance transfer deadline, no matter what the source's cumulative control period emissions are before that deadline. The commenters' approach of comparing—each day, cumulative emissions and allowances held—for purposes of calculating maximum civil penalties would be inconsistent with the flexibility that EPA intends to provide owners and operators. For example, under the commenters' suggested approach, owners and operators that buy or sell allowances in the allowance market or hold allowances in a company-wide account, do not transfer allowances into their source's compliance account until just before the allowance transfer deadline, and end up with some excess emissions for the calendar year would face a significantly higher maximum civil penalty than owners and operators that every day increase the amount of allowances held in their source's compliance account as the source's cumulative emissions increase and end up with the same amount of excess emissions for the calendar year. In short, the commenters' approach would penalize owners and operators that use some of the compliance flexibility that the trading programs are intended to provide.

EPA also maintains that it is reasonable to both impose the automatic allowance surrender and penalty provisions and to retain the discretion to impose civil penalties for the same occurrence of excess emissions. This approach encourages compliance with the allowance-holding requirement by ensuring that violating owners and operators are penalized automatically (i.e., without any further administrative or judicial proceedings, except for appeals) and that EPA can seek additional penalties where the circumstances warrant discretionary civil penalties. In fact, the Acid Rain Program, for which CAA Title IV mandated this approach, has achieved a very high level of compliance with the requirement to hold allowances covering SO₂ emissions and therefore resulted in major reductions in utility SO₂ emissions. See 42 U.S.C.7651j(a). Similarly, the NO_X Budget Trading Program and CAIR trading programs, which took the same approach, also have achieved very high compliance levels and major utility emission reductions.

EPA notes that, in calculating maximum civil penalties when owners and operators fail to hold allowances required under the assurance provisions in the final FIPs, EPA takes a similar approach in determining the number of violations. Each ton for which an allowance is not held as required and each day in the control period involved constitute a violation of the CAA. As discussed elsewhere in this preamble, EPA believes that this calculation approach is also reasonable in the context of the assurance provisions and that taking an approach like the commenters' suggested approach described above would be inconsistent with some of the flexibility that the Transport Rule trading programs are intended to provide.

G. Allowance Management System

The final Transport Rule trading programs, like the proposed preferred remedy, utilize EPA's allowance management system (AMS), which currently supports allowance surrender, transfer, and tracking activity under the Acid Rain Program and CAIR. EPA received no adverse comment on this aspect of the proposed rule.

The primary role of AMS is to provide an efficient, automated means for covered sources to comply and for EPA to determine whether covered sources are complying, with the emissions-related provisions of the Transport Rule trading programs. As was proposed, each of the final SO₂ trading programs and final NO_X trading programs is separately handled in the AMS, which is used to track Transport Rule trading program SO₂ and NO_X allowances held by covered sources, as well as such allowances held by other entities or individuals.

In addition, the AMS tracks: The allocation of all SO₂ and NO_X allowances; holdings of SO₂ and NO_X allowances in compliance accounts (i.e., accounts for individual covered sources), general accounts (i.e., accounts for other entities such as companies and brokers), and assurance accounts (i.e., accounts for allowance surrender by owners and operators of groups of sources and units with common designated representatives under the assurance provisions); deduction of SO₂ and NO_X allowances for compliance purposes (including deductions from assurance accounts where necessary); and transfers of allowances between accounts. The AMS also allows the public to see whether each source is in compliance and provides information to the allowance market and the public in general, including information on ownership of allowances, dates of allowance transfers, buyer and seller information, and the serial numbers of allowances transferred.

H. Emissions Monitoring and Reporting

Under the proposed rule, units subject to the Transport Rule trading programs would monitor and report NO_X and SO₂ mass emissions in accordance with 40 CFR part 75, as incorporated in the proposed rule, and with certain other specified requirements, such as compliance deadlines.

In the final rule, like the proposed rule, covered units must comply with emissions monitoring and reporting requirements that are largely incorporated from Part 75 monitoring and reporting requirements.

Under the final rule and under Part 75, a unit has several options for monitoring and reporting, namely the use of: a CEMS; an excepted monitoring methodology (NO_X mass monitoring for certain peaking units and SO₂ mass monitoring for certain oil- and gas-fired units); low mass emissions monitoring for certain non-coal-fired, low emitting units; or an alternative monitoring system approved by the Administrator through a petition process. In addition, the Administrator can approve petitions for alternatives to Transport Rule and Part 75 monitoring, recordkeeping, and reporting requirements.

Further, the final rule and Part 75 specify that each CEMS must undergo rigorous initial certification testing and periodic quality assurance testing thereafter, including the use of relative accuracy test audits (RATAs) and 24-hour calibrations. In addition, when a monitoring system is not operating properly, standard substitute data procedures are applied and result in a conservative estimate of emissions for the period involved.

In addition, the final rule and Part 75 require electronic submission, to the Administrator and in a format prescribed by the Administrator, of a quarterly emissions report. The report must contain all of the data required concerning NO_X annual and ozone-season and SO₂ annual emissions.

Most Transport Rule units are in states subject to CAIR and are already monitoring and reporting NO_X and/or SO₂ under CAIR and the Acid Rain Program, which programs also use Part 75 monitoring and reporting. Units under the Transport Rule annual trading programs and in states subject to CAIR generally have no changes to their monitoring and reporting requirements. These units must continue to monitor and submit reports on a year-round basis as they have under CAIR. Therefore, units in the following states must monitor and report both SO₂ and NO_X year-round under the Transport Rule: Alabama, Georgia, Illinois, Indiana, Iowa, Kansas, Kentucky, Maryland, Michigan, Minnesota, Missouri, Nebraska, New Jersey, New York, North Carolina, Ohio, Pennsylvania, South Carolina, Tennessee, Texas, Virginia, West Virginia and Wisconsin.

Some states (Kansas, Minnesota, and Nebraska) subject to the Transport Rule annual trading programs were not subject to CAIR. Transport Rule units in those states must meet monitoring and reporting requirements that are new except to the extent the units were subject to Part 75 under some other program (such as the Acid Rain Program).

Further, some states (Florida, Louisiana, and Mississippi) subject to the Transport Rule ozone-season trading program but not the Transport Rule annual trading programs were subject to the annual and ozone-season trading programs under CAIR. Transport Rule units in those states must continue to monitor and report in accordance with Part 75 but have the option of monitoring and reporting on a year-round or ozone-season-only basis.

In addition, one state (Arkansas) subject to the Transport Rule ozone-season trading program but not to the Transport Rule annual trading program was similarly subject to only the ozone-season trading program in CAIR. Transport Rule units in that state continue to have the option of monitoring and reporting NO_X on a year-round or ozone-season-only basis.

Finally, some states (Connecticut, Delaware, District of Columbia, and Massachusetts) that were subject to CAIR are not subject to the Transport Rule. Electric generating units in those states must continue to meet monitoring and reporting requirements only to the extent the units are subject to Part 75 under some other program (such as the Acid Rain Program or a state adopted program requiring such monitoring and reporting).

EPA is finalizing requirements for existing Transport Rule units in states covered by the Transport Rule annual trading programs to monitor and report SO₂ and NO_X emissions by January 1, 2012 programs and for existing Transport Rule units in states covered by the Transport Rule ozone-season trading program to monitor NO_X emissions by May 1, 2012. The use of Part 75 certified monitoring methodologies is required in both cases. As discussed previously, most covered existing units will generally have no changes to their monitoring and reporting requirements and will continue to monitor and submit reports under Part 75 as they have under CAIR. Existing units that have not been subject to Part 75 monitoring and reporting requirements in the past have less than 1 year to install, certify, and operate the required monitoring systems. EPA believes that these units will be able to comply with this requirement because the monitoring equipment needed is not extensive or is largely in place already for the purpose of meeting other requirements. Quality assurance and reporting provisions and data system upgrades may be necessary, but EPA believes that there is sufficient time to accomplish this by the deadline for existing units in the final rule.

In the proposed rule, the compliance deadline for installing, certifying, and operating the required monitoring systems at new units was based upon the date of commencement of commercial operation. A new unit would have to install and certify its monitoring system within 180 days of the commencement of commercial operation. The final rule adopts this deadline, which is consistent with the approach recently adopted in Part 75 under the Acid Rain Program. See 76 FR 17288, 17289 (March 28, 2011).

Using this deadline (rather than a deadline, used previously in Part 75, of the earlier of the unit's 90th operating day or 180 days after the unit's commencement of commercial operation) ensures that new units have sufficient time to complete installation and certification of monitoring systems and facilitates units' compliance. Because of unit shakedown problems, some new units have had difficulty meeting a deadline earlier than 180 days after commencement of commercial operation. Further, using this deadline facilitates owners' and operators, and EPA's, ability to track important dates related to monitoring, reporting, and allowance holding. Under the final rule, the requirement that a unit hold enough allowances to cover its emissions starts on the later of the commencement of the Transport Rule trading program involved or the deadline for installation and certification of the monitoring system. Having a simple, easily determined deadline (180 days after the commencement of commercial operation) makes it easier for owners and operators and EPA to determine when allowance-holding requirements begin, as well as when monitoring and reporting requirements begin. In contrast, using a deadline involving determination of a unit's 90th operating day required keeping track of any days on which the unit did not operate (e.g., due to problems associated with shakedown of the unit). EPA found that owners and operators have had more difficulty reporting the 90th operating day than in reporting the commencement of commercial operation, and once the latter date is reported, EPA can independently determine the 180th calendar day after the reported date.

I. Permitting

1. Title V Permitting

The final Transport Rule (like the proposed rule) does not establish any permitting requirements independent of those under Title V of the CAA and the regulations implementing Title V, 40 CFR Parts 70 and 71. All major stationary sources of air pollution and certain other sources are required to apply for title V operating permits that include emission limitations and other conditions as necessary to assure compliance with applicable requirements of the CAA, including the requirements of the applicable State Implementation Plan. CAA §§ 502(a) and 504(a), 42 U.S.C. 7661a(a) and 7661c(a). The “applicable requirements,” that must be addressed in title V permits are defined in the Title V regulations (40 CFR 70.2 and 71.2 (definition of “applicable requirement”)).