Skip to Content

Rule

Approval, Disapproval and Promulgation of Implementation Plans; State of Wyoming; Regional Haze State Implementation Plan; Federal Implementation Plan for Regional Haze

Document Details

Information about this document as published in the Federal Register.

Published Document

This document has been published in the Federal Register. Use the PDF linked in the document sidebar for the official electronic format.

Start Preamble Start Printed Page 5032

AGENCY:

Environmental Protection Agency.

ACTION:

Final rule.

SUMMARY:

The Environmental Protection Agency (EPA) is partially approving and partially disapproving a State Implementation Plan (SIP) submitted by the State of Wyoming on January 12, 2011, that addresses regional haze. This SIP was submitted to address the requirements of the Clean Air Act (CAA or “the Act”) and rules that require states to address in specific ways any existing anthropogenic impairment of visibility in mandatory Class I areas caused by emissions of air pollutants from numerous sources located over a wide geographic area (also referred to as the “regional haze program”). States are required to assure reasonable progress toward the national goal of achieving natural visibility conditions in Class I areas. EPA is approving several aspects of Wyoming's regional haze SIP that we had proposed to disapprove in our June 10, 2013 proposed rule in light of public comments and newly available information indicating the adequacy of the SIP with respect to those aspects. EPA is also approving some aspects of the State's SIP that we proposed to approve. EPA is promulgating a Federal Implementation Plan (FIP) to address some of the deficiencies identified in our proposed partial disapproval of Wyoming's regional haze SIP issued on June 10, 2013. EPA is taking this action pursuant to sections 110 and 169A of the CAA.

DATES:

This final rule is effective March 3, 2014.

ADDRESSES:

EPA has established a docket for this action under Docket ID No. EPA-R08-OAR-2012-0026. All documents in the docket are listed on the www.regulations.gov Web site.

Publicly available docket materials are available either electronically through www.regulations.gov, or in hard copy at the Air Program, Environmental Protection Agency (EPA), Region 8, 1595 Wynkoop Street, Denver, Colorado 80202-1129. EPA requests that if, at all possible, you contact the individual listed in the FOR FURTHER INFORMATION CONTACT section to view the hard copy of the docket. You may view the hard copy of the docket Monday through Friday, 8 a.m. to 4 p.m., excluding Federal holidays.

Start Further Info

FOR FURTHER INFORMATION CONTACT:

Laurel Dygowski, Air Program, Mailcode 8P-AR, Environmental Protection Agency, Region 8, 1595 Wynkoop Street, Denver, Colorado 80202-1129, (303) 312-6144, dygowski.laurel@epa.gov.

End Further Info End Preamble Start Supplemental Information

SUPPLEMENTARY INFORMATION:

Definitions

For the purpose of this document, we are giving meaning to certain words or initials as follows:

i. The words or initials Act or CAA mean or refer to the Clean Air Act, unless the context indicates otherwise.

ii. The initials AFUDC mean or refer to Allowance for Funds Utilized During Construction.

iii. The initials APA mean or refer to the Administrative Procedures Act.

iv. The initials AQRV mean or refer to Air Quality Related Value.

v. The initials BACT mean or refer to Best Available Control Technology.

vi. The initials BART mean or refer to Best Available Retrofit Technology.

vii. The initials CAMD mean or refer to Clean Air Markets Division.

viii. The initials CAMx mean or refer to Comprehensive Air Quality Model.

ix. The initials CCM mean or refer to EPA's Control Cost Manual.

x. The initials CLRC mean or refer to the Construction Labor Research Council.

xi. The initials CMAQ mean or refer to Community Multi-Scale Air Quality modeling system.

xii. The initials CSAPR mean or refer to the Cross-State Air Pollution Rule.

xiii. The initial DEQ mean or refer to the Wyoming Department of Environmental Quality.

xiv. The initials EGUs mean or refer to Electric Generating Units.

xv. The initials EIS mean or refer to Environmental Impact Statement.

xvi. The words EPA, we, us or our mean or refer to the United States Environmental Protection Agency.

xvii. The initials ESP mean or refer to electrostatic precipitator.

xviii. The initials FIP mean or refer to Federal Implementation Plan.

xix. The initials FLM mean or refer to Federal Land Managers.

xx. The initials FR mean or refer to the Federal Register.

xxi. The initials GAQM mean or refer to Guidance on Air Quality Models.

xxii. The initials IMPROVE mean or refer to Interagency Monitoring of Protected Visual Environments monitoring network.

xxiii. The initials IPM mean or refer to Integrated Planning Model.

xxiv. The initials IWAQM mean or refer to Interagency Workgroup on Air Quality Modeling.

xxv. The initials LNB mean or refer to low NOX burners.

xxvi. The initials LRS mean or refer to Laramie River Station.

xxvii. The initials LTS mean or refer to long term strategy.

xxviii. The initials MATS mean or refer to the Mercury and Air Toxics Standard.

xxix. The initials MW mean or refer to megawatts.

xxx. The initials NAAQS mean or refer to National Ambient Air Quality Standards.

xxxi. The initials NEPA mean or refer to National Environmental Policy Act.

xxxii. The initials NH3 mean or refer to ammonia.

xxxiii. The initials NOX mean or refer to nitrogen oxides.

xxxiv. The initials OFA mean or refer to overfire air.

xxxv. The initials PM mean or refer to particulate matter.

xxxvi. The initials PM2.5 mean or refer to particulate matter with an aerodynamic diameter of less than 2.5 micrometers.

xxxvii. The initials PM10 mean or refer to particulate matter with an aerodynamic diameter of less than 10 micrometers.

xxxviii. The initials PTE mean or refer to potential to emit.

xxxix. The initials RAVI mean or refer to reasonably attributable visibility impairment.

xl. The initials RHR mean or refer to the Regional Haze Rule.

xli. The initials RIS mean or refer to Regulatory Impact Statement.

xlii. The initials RPG mean or refer to reasonable progress goals.

xliii. The initials RPO mean or refer to Regional Planning Organization.

xliv. The initials SCR mean or refer to selective catalytic reduction.

xlv. The initials SIP mean or refer to State Implementation Plan.

xlvi. The initials SNCR mean or refer to selective non-catalytic reduction.

xlvii. The initials SO2 mean or refer to sulfur dioxide.

xlviii. The initials SOFA mean or refer to separated overfire air.

xlix. The initials UMRA mean or refer to the Unfunded Mandates Reform Act.

l. The initials URP mean or refer to Uniform Rate of Progress.

li. The initials VOC mean or refer to volatile organic compounds.

lii. The initials WAQSR mean or refer to the Wyoming Air Quality Standards and Regulations.

liii. The initials WRAP mean or refer to the Western Regional Air Partnership.

liv. The words Wyoming and State mean the State of Wyoming.

Table of Contents

I. Background

A. Regional Haze

i. Requirements of the CAA and EPA's Regional Haze Rule (RHR)

ii. Roles of Agencies in Addressing Regional Haze

B. Requirements for the Regional Haze SIPs

i. The CAA and the Regional Haze Rule

ii. Determination of Baseline, Natural, and Current Visibility Conditions

iii. Determination of Reasonable Progress GoalsStart Printed Page 5033

iv. Best Available Retrofit Technology

v. Long-Term Strategy

vi. Coordinating Regional Haze and Reasonably Attributable Visibility Impairment

vii. Monitoring Strategy and Other Implementation Plan Requirements

viii. Consultation With States and Federal Land Managers (FLMs)

C. Our Proposal

D. Public Participation

II. Final Action

III. Changes From Proposed Rule and Reasons for Changes

A. Changes to Proposed Costs and Visibility Improvements

B. Changes to Our Proposed Determinations

1. Dave Johnston Unit 3

2. Dave Johnston Unit 4

3. Naughton Units 1 and 2

4. Naughton Unit 3

5. Wyodak

6. Jim Bridger

7. Dave Johnston Units 1 and 2

IV. Basis for Our Final Action

A. Laramie River

B. Jim Bridger

C. Dave Johnston Units 3 and 4

D. Naughton

E. Wyodak

F. Dave Johnston Units 1 and 2 (Reasonable Progress)

V. Issues Raised by Commenters and EPA's Responses

A. Legal Issues

1. EPA Authority and State Discretion

2. Compliance With Section 307(d)

3. Compliance With Section 169A(d)

4. Public Hearings

5. RHR and BART Guidelines

6. Reasonableness Standard

7. Reliance on Emission Reductions

8. Presumptive Limits

9. Compliance With 40 CFR 51.308

10. Legal Analysis

11. Consideration of Existing Controls

12. Consent Decree

13. Monitoring, Recordkeeping and Reporting

B. Modeling

1. General Comments

2. EPA Modeling

a. Description of Revised EPA Modeling

b. Comments on EPA Modeling

C. Overarching Comments on BART

1. BART-Eligible Sources

2. Cost of Controls

3. Consideration of the Five Factors

4. Visibility Improvement

5. PM BART Determinations

6. Incremental Costs and Visibility

7. Other Comments on BART

D. BART Sources

1. Basin Electric Laramie River Station Units 1-3

a. General Comments

b. NOX BART Determination

2. Jim Bridger Units 1-4

a. NOX BART Determination

b. PM BART Determination

3. Dave Johnston Unit 3 and Unit 4

a. NOX BART Determination

b. Alternative Control Technology Proposal

4. Naughton Units 1-3

a. NOX BART Determination

b. Alternative Control Technology Proposal

5. Wyodak

6. Trona Mines

a. FMC Westvaco and General Chemical Green River

b. FMC Granger Trona Mine

E. Reasonable Progress

1. RPGs

2. Reasonable Progress Sources

a. Oil and Gas Sources

b. Dave Johnston Unit 1 and Unit 2

F. General Comments

1. Replacement of FIP Elements With SIP

2. Public Comment

3. Economic Concerns

4. National Ambient Air Quality Standards (NAAQS)

5. Other

VI. Non-Relevant Comments From EPA's Original June 4, 2012 Proposal

A. General Comments

B. Basin Electric Laramie River

C. Jim Bridger Units 1-4

D. Dave Johnston Units 3 and 4

E. Naughton Units 1-3

F. Wyodak

G. Dave Johnston Units 1 and 2

H. Modeling

VII. Statutory and Executive Order Reviews

I. Background

The CAA requires each state to develop plans, referred to as SIPs, to meet various air quality requirements. A state must submit its SIP and SIP revisions to us for approval. Once approved, a SIP is enforceable by EPA and citizens under the CAA, also known as being federally enforceable. If a state fails to make a required SIP submittal or if we find that a state's required submittal is incomplete or unapprovable, then we must promulgate a FIP to fill this regulatory gap. CAA section 110(c)(1). This action involves the requirement that states have SIPs that address regional haze.

Few states submitted a regional haze SIP prior to the December 17, 2007 deadline, and on January 15, 2009, EPA found that 37 states, including Wyoming,[1] the District of Columbia, and the Virgin Islands, had failed to submit SIPs addressing the regional haze requirements. 74 FR 2392. Once EPA has found that a state has failed to make a required submission, EPA is required to promulgate a FIP within two years unless the state submits a SIP and the Agency approves it within the two-year period. CAA section 110(c)(1). Wyoming subsequently submitted a SIP addressing regional haze on January 12, 2011.

States in the west were given the option to meet the requirements of the RHR either under 40 CFR 51.309 or 40 CFR 51.308. Wyoming chose to adopt the requirements of 40 CFR 51.309. Section 309 requires states to adopt regional haze strategies that are based on recommendations from the Grand Canyon Visibility Transport Commission for protecting the 16 Class I areas in the Colorado Plateau area, including a sulfur dioxide (SO2) backstop cap and trade program, SO2 milestones, and other requirements such as smoke management, a program to address mobile sources, and pollution prevention. Also, section 309(g) includes requirements for SIP provisions, including NOX BART, to address visibility impairment at other Class I areas. On December 12, 2012, we finalized approval of Wyoming's 309 regional haze SIP for the requirements relating to the SO2 backstop cap and trade program, milestones and the other requirements.[2] Today's action addresses the remaining portion of Wyoming's SIP, including the Best Available Retrofit Technology (BART) determinations for nitrogen oxides (NOX) and particulate matter (PM).

In a lawsuit in the U.S. District Court for the District of Colorado, environmental groups sued EPA for our failure to take timely action with respect to the regional haze requirements of the CAA and our regulations.[3] In particular, the lawsuits alleged that we had failed to promulgate FIPs for these requirements within the two-year period allowed by CAA section 110(c) or, in the alternative, fully approve SIPs addressing these requirements.

As a result of these lawsuits, we entered into a consent decree. The consent decree requires that we sign a notice of final rulemaking addressing the regional haze requirements for Wyoming by January 10, 2014.[4] We are meeting that requirement with the signing of this final rule

A. Regional Haze

Regional haze is visibility impairment that is produced by a multitude of sources and activities which are located across a broad geographic area and emit fine particles (PM2.5) (e.g., sulfates, nitrates, organic carbon (OC), elemental carbon (EC), and soil dust), and their precursors (e.g., sulfur dioxide (SO2), Start Printed Page 5034NOX, and in some cases, ammonia (NH3) and volatile organic compounds (VOC)). Fine particle precursors react in the atmosphere to form PM2.5, which impairs visibility by scattering and absorbing light. Visibility impairment reduces the clarity, color, and visible distance that one can see. PM2.5 can also cause serious health effects and mortality in humans and contributes to environmental effects such as acid deposition and eutrophication.

Data from the existing visibility monitoring network, the “Interagency Monitoring of Protected Visual Environments” (IMPROVE) monitoring network, show that visibility impairment caused by air pollution occurs virtually all the time at most national park and wilderness areas. The average visual range [5] in many Class I areas (i.e., national parks and memorial parks, wilderness areas, and international parks meeting certain size criteria) in the western United States is 100-150 kilometers, or about one-half to two-thirds of the visual range that would exist without anthropogenic air pollution. In most of the eastern Class I areas of the United States, the average visual range is less than 30 kilometers, or about one-fifth of the visual range that would exist under estimated natural conditions. 64 FR 35715 (July 1, 1999).

i. Requirements of the CAA and EPA's Regional Haze Rule (RHR)

In section 169A of the 1977 Amendments to the CAA, Congress created a program for protecting visibility in the nation's national parks and wilderness areas. This section of the CAA establishes as a national goal the “prevention of any future, and the remedying of any existing, impairment of visibility in mandatory Class I Federal areas [6] which impairment results from manmade air pollution.” On December 2, 1980, EPA promulgated regulations to address visibility impairment in Class I areas that is “reasonably attributable” to a single source or small group of sources, i.e., “reasonably attributable visibility impairment.” 45 FR 80084. These regulations represented the first phase in addressing visibility impairment. EPA deferred action on regional haze that emanates from a variety of sources until monitoring, modeling and scientific knowledge about the relationships between pollutants and visibility impairment were improved.

Congress added section 169B to the CAA in 1990 to address regional haze issues. EPA promulgated a rule to address regional haze on July 1, 1999. 64 FR 35714 (July 1, 1999), codified at 40 CFR part 51, subpart P. The RHR revised the existing visibility regulations to integrate into the regulation provisions addressing regional haze impairment and established a comprehensive visibility protection program for Class I areas. The requirements for regional haze, found at 40 CFR 51.308 and 51.309, are included in EPA's visibility protection regulations at 40 CFR 51.300-51.309. Some of the main elements of the regional haze requirements are summarized in section III of this preamble. The requirement to submit a regional haze SIP applies to all 50 states, the District of Columbia and the Virgin Islands. 40 CFR 51.308(b) requires states to submit the first implementation plan addressing regional haze visibility impairment no later than December 17, 2007.[7]

Few states submitted a regional haze SIP prior to the December 17, 2007 deadline, and on January 15, 2009, EPA found that 37 states (including Wyoming), the District of Columbia, and the Virgin Islands, had failed to submit SIPs addressing the regional haze requirements. 74 FR 2392. Once EPA has found that a state has failed to make a required submission, EPA is required to promulgate a FIP within two years unless the state submits a SIP and the Agency approves it within the two-year period. CAA section110(c)(1).

ii. Roles of Agencies in Addressing Regional Haze

Successful implementation of the regional haze program will require long-term regional coordination among states, tribal governments, and various federal agencies. As noted above, pollution affecting the air quality in Class I areas can be transported over long distances, even hundreds of kilometers. Therefore, to effectively address the problem of visibility impairment in Class I areas, states need to develop strategies in coordination with one another, taking into account the effect of emissions from one jurisdiction on the air quality in another.

Because the pollutants that lead to regional haze can originate from sources located across broad geographic areas, EPA has encouraged the states and tribes across the United States to address visibility impairment from a regional perspective. Five regional planning organizations (RPOs) were developed to address regional haze and related issues. The RPOs first evaluated technical information to better understand how their states and tribes impact Class I areas across the country, and then pursued the development of regional strategies to reduce emissions of pollutants that lead to regional haze.

The Western Regional Air Partnership (WRAP) RPO is a collaborative effort of state governments, tribal governments, and various federal agencies established to initiate and coordinate activities associated with the management of regional haze, visibility and other air quality issues in the western United States. WRAP member state governments include: Alaska, Arizona, California, Colorado, Idaho, Montana, New Mexico, North Dakota, Oregon, South Dakota, Utah, Washington, and Wyoming. Tribal members include Campo Band of Kumeyaay Indians, Confederated Salish and Kootenai Tribes, Cortina Indian Rancheria, Hopi Tribe, Hualapai Nation of the Grand Canyon, Native Village of Shungnak, Nez Perce Tribe, Northern Cheyenne Tribe, Pueblo of Acoma, Pueblo of San Felipe, and Shoshone-Bannock Tribes of Fort Hall.

B. Requirements for Regional Haze SIPs

The following is a summary of the requirements of the RHR. See 40 CFR 51.308 for further detail regarding the requirements of the rule.

i. The CAA and the Regional Haze Rule

Regional haze SIPs must assure reasonable progress towards the national goal of achieving natural visibility conditions in Class I areas. Section 169A of the CAA and EPA's implementing regulations require states to establish long-term strategies for making reasonable progress toward meeting this goal. Implementation plans must also give specific attention to certain stationary sources that were in Start Printed Page 5035existence on August 7, 1977, but were not in operation before August 7, 1962, and require these sources, where appropriate, to install BART controls for the purpose of eliminating or reducing visibility impairment. The specific regional haze SIP requirements are discussed in further detail below.

ii. Determination of Baseline, Natural, and Current Visibility Conditions

The RHR establishes the deciview as the principal metric or unit for expressing visibility. See 70 FR 39104, 39118. This visibility metric expresses uniform changes in the degree of haze in terms of common increments across the entire range of visibility conditions, from pristine to extremely hazy conditions. Visibility expressed in deciviews is determined by using air quality measurements to estimate light extinction and then transforming the value of light extinction using a logarithmic function. The deciview is a more useful measure for tracking progress in improving visibility than light extinction itself because each deciview change is an equal incremental change in visibility perceived by the human eye. Most people can detect a change in visibility at one deciview.[8]

The deciview is used in expressing RPGs (which are interim visibility goals towards meeting the national visibility goal), defining baseline, current, and natural conditions, and tracking changes in visibility. The regional haze SIPs must contain measures that ensure “reasonable progress” toward the national goal of preventing and remedying visibility impairment in Class I areas caused by anthropogenic air pollution by reducing anthropogenic emissions that cause regional haze. The national goal is a return to natural conditions, i.e., anthropogenic sources of air pollution would no longer impair visibility in Class I areas.

To track changes in visibility over time at each of the 156 Class I areas covered by the visibility program (40 CFR 81.401-437), and as part of the process for determining reasonable progress, states must calculate the degree of existing visibility impairment at each Class I area at the time of each regional haze SIP submittal and periodically review progress every five years midway through each 10-year implementation period. To do this, the RHR requires states to determine the degree of impairment (in deciviews) for the average of the 20 percent least impaired (“best”) and 20 percent most impaired (“worst”) visibility days over a specified time period at each of their Class I areas. In addition, states must also develop an estimate of natural visibility conditions for the purpose of comparing progress toward the national goal. Natural visibility is determined by estimating the natural concentrations of pollutants that cause visibility impairment and then calculating total light extinction based on those estimates. We have provided guidance to states regarding how to calculate baseline, natural and current visibility conditions.[9]

For the first regional haze SIPs that were due by December 17, 2007, “baseline visibility conditions” were the starting points for assessing “current” visibility impairment. Baseline visibility conditions represent the degree of visibility impairment for the 20 percent least impaired days and 20 percent most impaired days for each calendar year from 2000 to 2004. Using monitoring data for 2000 through 2004, states are required to calculate the average degree of visibility impairment for each Class I area, based on the average of annual values over the five-year period. The comparison of initial baseline visibility conditions to natural visibility conditions indicates the amount of improvement necessary to attain natural visibility, while the future comparison of baseline conditions to the then current conditions will indicate the amount of progress made. In general, the 2000-2004 baseline period is considered the time from which improvement in visibility is measured.

iii. Determination of Reasonable Progress Goals

The vehicle for ensuring continuing progress towards achieving the natural visibility goal is the submission of a series of regional haze SIPs from the states that establish two RPGs (i.e., two distinct goals, one for the “best” and one for the “worst” days) for every Class I area for each (approximately) 10-year implementation period. See 40 CFR 51.308(d), (f). The RHR does not mandate specific milestones or rates of progress, but instead calls for states to establish goals that provide for “reasonable progress” toward achieving natural visibility conditions. In setting RPGs, states must provide for an improvement in visibility for the most impaired days over the (approximately) 10-year period of the SIP, and ensure no degradation in visibility for the least impaired days over the same period. Id.

In establishing RPGs, states are required to consider the following factors established in section 169A of the CAA and in our RHR at 40 CFR 51.308(d)(1)(i)(A): (1) The costs of compliance; (2) the time necessary for compliance; (3) the energy and non-air quality environmental impacts of compliance; and (4) the remaining useful life of any potentially affected sources. States must demonstrate in their SIPs how these factors are considered when selecting the RPGs for the best and worst days for each applicable Class I area. In setting the RPGs, states must also consider the rate of progress needed to reach natural visibility conditions by 2064 (referred to as the “uniform rate of progress” (URP) or the “glidepath”) and the emission reduction measures needed to achieve that rate of progress over the 10-year period of the SIP. Uniform progress towards achievement of natural conditions by the year 2064 represents a rate of progress, which states are to use for analytical comparison to the amount of progress they expect to achieve. In setting RPGs, each state with one or more Class I areas (“Class I state”) must also consult with potentially “contributing states,” i.e., other nearby states with emission sources that may be affecting visibility impairment at the state's Class I areas. 40 CFR 51.308(d)(1)(iv). In determining whether a state's goals for visibility improvement provide for reasonable progress toward natural visibility conditions, EPA is required to evaluate the demonstrations developed by the state pursuant to paragraphs 40 CFR 51.308(d)(1)(i) and (d)(1)(ii). 40 CFR 51.308(d)(1)(iii).

iv. Best Available Retrofit Technology

Section 169A of the CAA directs states to evaluate the use of retrofit controls at certain larger, often uncontrolled, older stationary sources in order to address visibility impacts from these sources. Specifically, section 169A(b)(2)(A) of the CAA requires states to revise their SIPs to contain such measures as may be necessary to make reasonable progress towards the natural visibility goal, including a requirement that certain categories of existing major Start Printed Page 5036stationary sources[10] built between 1962 and 1977 procure, install, and operate the “Best Available Retrofit Technology” as determined by the state. Under the RHR, states are directed to conduct BART determinations for such “BART-eligible” sources that may be anticipated to cause or contribute to any visibility impairment in a Class I area. Rather than requiring source-specific BART controls, states also have the flexibility to adopt an emissions trading program or other alternative program as long as the alternative provides greater reasonable progress towards improving visibility than BART.

On July 6, 2005, EPA published the Guidelines for BART Determinations Under the Regional Haze Rule at appendix Y to 40 CFR part 51 (hereinafter referred to as the “BART Guidelines”) to assist states in determining which of their sources should be subject to the BART requirements and in determining appropriate emission limits for each applicable source. 70 FR 39104. In making a BART determination for a fossil fuel-fired electric generating plant with a total generating capacity in excess of 750 megawatts (MW), a state must use the approach set forth in the BART Guidelines. Generally, a state is encouraged, but not required, to follow the BART Guidelines in making BART determinations for other types of sources. Regardless of source size or type, a state must meet the requirements of the CAA and our regulations for selection of BART, and the state's BART analysis and determination must be reasonable in light of the overarching purpose of the regional haze program.

The process of establishing BART emission limitations can be logically broken down into three steps: First, states identify those sources which meet the definition of “BART-eligible source” set forth in 40 CFR 51.301; [11] second, states determine which of such sources “emits any air pollutant which may reasonably be anticipated to cause or contribute to any impairment of visibility in any such area” (a source which fits this description is “subject to BART”); and third, for each source subject-to-BART, states then identify the best available type and level of control for reducing emissions.

States must address all visibility-impairing pollutants emitted by a source in the BART determination process. The most significant visibility impairing pollutants are SO2, NOX, and PM. EPA has stated that states should use their best judgment in determining whether VOC or NH3 emissions impair visibility in Class I areas.

Under the BART Guidelines, states may select an exemption threshold value for their BART modeling, below which a BART-eligible source would not be expected to cause or contribute to visibility impairment in any Class I area. The state must document this exemption threshold value in the SIP and must state the basis for its selection of that value. Any source with emissions that model above the threshold value would be subject to a BART determination review. The BART Guidelines acknowledge varying circumstances affecting different Class I areas. States should consider the number of emission sources affecting the Class I areas at issue and the magnitude of the individual sources' impacts. Any exemption threshold set by the state should not be higher than 0.5 deciview. 40 CFR part 51, appendix Y, section III.A.1.

In their SIPs, states must identify the sources that are subject-to-BART and document their BART control determination analyses for such sources. In making their BART determinations, section 169A(g)(2) of the CAA requires that states consider the following factors when evaluating potential control technologies: (1) The costs of compliance; (2) the energy and non-air quality environmental impacts of compliance; (3) any existing pollution control technology in use at the source; (4) the remaining useful life of the source; and (5) the degree of improvement in visibility which may reasonably be anticipated to result from the use of such technology.

A regional haze SIP must include source-specific BART emission limits and compliance schedules for each source subject-to-BART. Once a state has made its BART determination, the BART controls must be installed and in operation as expeditiously as practicable, but no later than five years after the date of EPA approval of the regional haze SIP. CAA section 169(g)(4) and 40 CFR 51.308(e)(1)(iv). In addition to what is required by the RHR, general SIP requirements mandate that the SIP must also include all regulatory requirements related to monitoring, recordkeeping, and reporting for the BART controls on the source. See e.g. CAA section 110(a). As noted above, the RHR allows states to implement an alternative program in lieu of BART so long as the alternative program can be demonstrated to achieve greater reasonable progress toward the national visibility goal than would BART.

v. Long-Term Strategy

Consistent with the requirement in section 169A(b) of the CAA that states include in their regional haze SIP a 10 to 15-year strategy for making reasonable progress, section 51.308(d)(3) of the RHR requires that states include a LTS in their regional haze SIPs. The LTS is the compilation of all control measures a state will use during the implementation period of the specific SIP submittal to meet applicable RPGs. The LTS must include “enforceable emissions limitations, compliance schedules, and other measures as necessary to achieve the reasonable progress goals” for all Class I areas within, or affected by emissions from, the state. 40 CFR 51.308(d)(3).

When a state's emissions are reasonably anticipated to cause or contribute to visibility impairment in a Class I area located in another state, the RHR requires the impacted state to coordinate with the contributing states in order to develop coordinated emissions management strategies. 40 CFR 51.308(d)(3)(i). In such cases, the contributing state must demonstrate that it has included, in its SIP, all measures necessary to obtain its share of the emission reductions needed to meet the RPGs for the Class I area. Id. at (d)(3)(ii). The RPOs have provided forums for significant interstate consultation, but additional consultations between states may be required to sufficiently address interstate visibility issues. This is especially true where two states belong to different RPOs.

States should consider all types of anthropogenic sources of visibility impairment in developing their long-term strategy, including stationary, minor, mobile, and area sources. At a minimum, states must describe how each of the following seven factors listed below are taken into account in developing their LTS: (1) Emission reductions due to ongoing air pollution control programs, including measures to address RAVI; (2) measures to mitigate the impacts of construction activities; (3) emissions limitations and schedules for compliance to achieve the RPG; (4) source retirement and replacement schedules; (5) smoke management techniques for agricultural and forestry management purposes including plans as currently exist within the state for these purposes; (6) enforceability of emissions limitations and control measures; and (7) the anticipated net Start Printed Page 5037effect on visibility due to projected changes in point, area, and mobile source emissions over the period addressed by the LTS. 40 CFR 51.308(d)(3)(v).

vi. Coordinating Regional Haze and Reasonably Attributable Visibility Impairment

As part of the RHR, EPA revised 40 CFR 51.306(c) regarding the LTS for RAVI to require that the RAVI plan must provide for a periodic review and SIP revision not less frequently than every three years until the date of submission of the state's first plan addressing regional haze visibility impairment, which was due December 17, 2007, in accordance with 40 CFR 51.308(b) and (c). On or before this date, the state must revise its plan to provide for review and revision of a coordinated LTS for addressing RAVI and regional haze, and the state must submit the first such coordinated LTS with its first regional haze SIP. Future coordinated LTS's, and periodic progress reports evaluating progress towards RPGs, must be submitted consistent with the schedule for SIP submission and periodic progress reports set forth in 40 CFR 51.308(f) and 51.308(g), respectively. The periodic review of a state's LTS must report on both regional haze and RAVI impairment and must be submitted to EPA as a SIP revision.

vii. Monitoring Strategy and Other Implementation Plan Requirements

Section 51.308(d)(4) of the RHR includes the requirement for a monitoring strategy for measuring, characterizing, and reporting of regional haze visibility impairment that is representative of all mandatory Class I Federal areas within the state. The strategy must be coordinated with the monitoring strategy required in section 51.305 for RAVI. Compliance with this requirement may be met through “participation” in the IMPROVE network, i.e., review and use of monitoring data from the network. The monitoring strategy is due with the first regional haze SIP, and it must be reviewed every five years. The monitoring strategy must also provide for additional monitoring sites if the IMPROVE network is not sufficient to determine whether RPGs will be met.

The SIP must also provide for the following:

  • Procedures for using monitoring data and other information in a state with mandatory Class I areas to determine the contribution of emissions from within the state to regional haze visibility impairment at Class I areas both within and outside the state;
  • Procedures for using monitoring data and other information in a state with no mandatory Class I areas to determine the contribution of emissions from within the state to regional haze visibility impairment at Class I areas in other states;
  • Reporting of all visibility monitoring data to the Administrator at least annually for each Class I area in the state, and where possible, in electronic format;
  • Developing a statewide inventory of emissions of pollutants that are reasonably anticipated to cause or contribute to visibility impairment in any Class I area. The inventory must include emissions for a baseline year, emissions for the most recent year for which data are available, and estimates of future projected emissions. A state must also make a commitment to update the inventory periodically; and
  • Other elements, including reporting, recordkeeping, and other measures necessary to assess and report on visibility.

The RHR requires control strategies to cover an initial implementation period extending to the year 2018, with a comprehensive reassessment and revision of those strategies, as appropriate, every 10 years thereafter. Periodic SIP revisions must meet the core requirements of section 51.308(d) with the exception of BART. The requirement to evaluate sources for BART applies only to the first regional haze SIP. Facilities subject-to-BART must continue to comply with the BART provisions of section 51.308(e), as noted above. Periodic SIP revisions will assure that the statutory requirement of reasonable progress will continue to be met.

viii. Consultation With States and Federal Land Managers (FLMs)

The RHR requires that states consult with FLMs before adopting and submitting their SIPs. 40 CFR 51.308(i). States must provide FLMs an opportunity for consultation, in person and at least 60 days prior to holding any public hearing on the SIP. This consultation must include the opportunity for the FLMs to discuss their assessment of impairment of visibility in any Class I area and to offer recommendations on the development of the RPGs and on the development and implementation of strategies to address visibility impairment. Further, a state must include in its SIP a description of how it addressed any comments provided by the FLMs. Finally, a SIP must provide procedures for continuing consultation between the state and FLMs regarding the state's visibility protection program, including development and review of SIP revisions, five-year progress reports, and the implementation of other programs having the potential to contribute to impairment of visibility in Class I areas.

C. Our Proposal

We signed our notice of proposed rulemaking on May 23, 2013,[12] and it was published in the Federal Register on June 10, 2013 (78 FR 34738). In our 2013 proposal, we proposed to approve many of Wyoming's regional haze SIP, including the State's identification of its BART sources, its identification of those BART sources that may be anticipated to cause or contribute to visibility impairment, and the State's BART determinations for PM. Because of deficiencies in Wyoming's NOX BART analyses, however, we proposed to disapprove the NOX BART emissions limitations for a number of sources, as well as the reasonable progress goals and long-term strategy. We proposed to address the NOX BART requirements for these sources and the other deficiencies in the Wyoming plan in a FIP, based on our analysis of the relevant factors. For several BART sources we also asked in the proposed rulemaking if interested parties had additional information regarding the BART factors and EPA's proposed determinations, for example our weighing of average costs, incremental costs, visibility improvement, and timing of installation of such controls, and in light of such information, whether the interested parties thought the Agency should consider another BART control technology option that could be finalized either instead of, or in conjunction with, BART as proposed.[13]

In our 2013 proposal we proposed to disapprove the following:Start Printed Page 5038

  • The State's nitrogen oxides (NOX) best available retrofit technology (BART) determinations for PacifiCorp Dave Johnston Units 3 and 4, PacifiCorp Naughton Units 1 and 2, PacifiCorp Wyodak Unit 1, and Basin Electric Laramie River Units 1, 2, and 3.
  • The State's NOX reasonable progress determinations for PacifiCorp Dave Johnston Units 1 and 2.
  • Wyoming's reasonable progress goals (RPGs).
  • The State's monitoring, recordkeeping, and reporting requirements in Chapter 6.4 of the SIP.
  • Portions of the State's long-term strategy (LTS) that rely on or reflect other aspects of the regional haze SIP that we are disapproving.
  • The provisions necessary to meet the requirements for the coordination of the review of the reasonably attributable visibility impairment (RAVI) and the regional haze LTS.

We proposed the promulgation of a FIP to address the deficiencies in the Wyoming regional haze SIP that we identified in the proposed notice. The proposed FIP included the following elements:

  • NOX BART determinations and limits for PacifiCorp Dave Johnston Units 3 and 4, PacifiCorp Naughton Units 1 and 2, PacifiCorp Wyodak Unit 1, and Basin Electric Laramie River Units 1, 2, and 3.
  • NOX reasonable progress determinations and limits for PacifiCorp Dave Johnston Units 1 and 2.
  • RPGs consistent with the SIP limits proposed for approval and the proposed FIP limits.
  • Monitoring, recordkeeping, and reporting requirements applicable to all BART and reasonable progress sources for which there is a SIP or FIP emissions limit.
  • LTS elements pertaining to emission limits and compliance schedules for the proposed BART and reasonable progress FIP emission limits.
  • Provisions to ensure the coordination of the RAVI and regional haze LTS.

We also requested comment on an alternative proposal, related to the State's NOX BART determinations, for PacifiCorp Jim Bridger Units 1 and 2, that would involve disapproval and the promulgation of a FIP.

D. Public Participation

We requested comments on all aspects of our proposed action. In our proposed rulemaking, we provided a 60-day comment period, with the comment period closing on August 9, 2013. We also held a public hearing on June 24, 2013, in Cheyenne, Wyoming. We received requests from Wyoming's governor, congressional delegation, and Department of Environmental Quality (DEQ), among others, for additional public hearings and an extended public comment period. As a result, we held two more public hearings. We held a hearing on July 17, 2013, in Cheyenne, Wyoming, and on July 26, 2013, in Casper, Wyoming. We also extended the comment period to August 26, 2013. We provided public notice of the additional hearings and extension of the public comment period on July 8, 2013. 78 FR 40654.

II. Final Action

Based upon comments received on our proposed action, in this final action we are partially approving and partially disapproving Wyoming's regional haze SIP submitted on January 12, 2011. We are approving the majority of the State's regional haze determinations. For the fifteen coal fired power plant units in Wyoming subject to the regional haze requirements, we are approving the State's NOX emission control technology decisions for 10 of those units. We are also approving the State's plan for the non-power plant facilities subject to regional haze requirements and the State's plan for control of PM. We are approving all aspects of Wyoming's SIP, except for the following elements which we are disapproving:

  • The State's NOX BART determinations for PacifiCorp Dave Johnston Unit 3, PacifiCorp Wyodak Unit 1, and Basin Electric Laramie River Units 1, 2, and 3.
  • Wyoming's RPGs.
  • The State's monitoring, recordkeeping, and reporting requirements in Chapter 6.4 of the SIP.
  • Portions of the State's LTS that rely on or reflect other aspects of the regional haze SIP that we are disapproving.
  • The provisions necessary to meet the requirements for the coordination of the review of the RAVI and the regional haze LTS.

The final FIP includes the following elements:

  • NOX BART determinations and emission limits for PacifiCorp Dave Johnston Unit 3, Wyodak Unit 1, and Basin Electric Laramie River Units 1, 2, and 3.
  • RPGs consistent with the SIP emission limits finalized for approval and the finalized FIP emission limits.
  • Monitoring, recordkeeping, and reporting requirements applicable to all BART sources for which there is a SIP or FIP emissions limit.
  • LTS elements pertaining to emission limits and compliance schedules for the finalized FIP emission limits.
  • Provisions to ensure the coordination of the RAVI and regional haze LTS.

Although we are promulgating a Federal plan, a state may always submit a new regional haze SIP to EPA for review and we would welcome such a submission. The CAA requires EPA to take action on such a SIP submittal that is determined to be complete within 12 months. If the State were to submit a revision meeting the requirements of the CAA and the regional haze regulations, we would propose approval of the State's plan as expeditiously as practicable. We are mindful of the costs of our final action but have considered the costs and visibility improvement that other states and EPA have required for BART controls.

Table 1 shows the NOX BART control technologies, associated cost, and emission reductions for each source that is subject to the FIP.

Table 1—Control Technologies, Costs, Emission Limits, and Cost Effectiveness for Sources Subject to the FIP

SourceTechnology *Emission limit—lb/MMBtu (30-day rolling average)Total capital cost ($)Total annualized cost ($)Average cost- effectiveness ($/ton)
Dave Johnston Unit 3New low-NOX burners (LNBs) with overfire air (OFA) and shut down in 2027; or new LNBs with OFA and selective catalytic reduction (SCR) **0.28 (for LNBs with OFA)$15,976,696 (for LNBs with OFA)$1,828,137 (for LNBs with OFA)$644 (for LNBs with OFA).
Start Printed Page 5039
Laramie River Unit 1New LNBs/OFA and SCR0.07$180,254,572$21,770,134$4,461.
Laramie River Unit 2New LNBs with OFA and SCR0.07$188,826,333$22,691,467$4,424.
Laramie River Unit 3New LNBs with OFA and SCR0.07$188,437,953$22,666,982$4,375.
Wyodak Unit 1New LNBs with OFA and SCR0.07$119,501,862$12,714,153$4,036.
* The technology listed is the technology evaluated as BART, but sources can choose to use another technology or combination of technologies to meet established limits.
** As used in this and the following tables, “new” means replacing the control technology that was in place at the time of the State's BART analyses in May 2009 with new control technology, most of which was installed post-2009.

III. Changes From Proposed Rule and Reasons for Changes

A. Changes to Proposed Costs and Visibility Improvements

As described in this section and elsewhere in today's final rule, we have revised our cost of compliance analysis and visibility improvement modeling from our June 10, 2013 proposed action for all of the BART and reasonable progress electric generating units (EGUs).

EPA revised the cost analyses from those found in the proposed rule based upon input from various commenters. Some of factors that caused us to revise our cost estimates included accounting for site elevation in the SCR capital cost, change in SCR reagent to anhydrous ammonia from urea, change in auxiliary electrical cost from market price to generating cost, change in urea SNCR chemical utilization for some units due to high furnace temperatures, and consideration of shorter plant lifetimes in some instances. In addition, EPA incorporated some of the costs provided by commenters in their site specific cost estimates where we found those costs to be sufficiently supported. Per EPA's Control Cost Manual (CCM), use of site specific cost estimates is preferable to the use of generalized costs where those site specific costs can be supported and are appropriate.

EPA addressed comments on the visibility improvement modeling in the proposed rule by developing a new protocol that makes several improvements in the modeling, including the use of the current regulatory version of the CALPUFF model (version 5.8), the use of an improved method to assess the effects of pollutants on light scattering and visibility impairment (Method 8), the use of lower background ammonia concentrations, and the use of an ammonia limiting correction for BART sources with multiple units. In particular, we have used new values for ammonia background that reflect robust monitoring data and the appropriate default concentrations for the geography in the state.

The results of our revised cost analysis, along with the revised visibility impacts, are presented in Tables 2 through 17 below and summarized for each source below the set of tables for that source. Details regarding our revised cost analysis and visibility improvement modeling can be found in the docket.[14 15]

Table 2—Summary of EPA's Laramie River Unit 1 NOX BART Analysis

Control technologyEmission rate (lb/MMBtu; annual average)Emission reduction (tpy)Annualized costsAverage cost effectiveness ($/ton)Incremental cost effectiveness ($/ton)Visibility improvement (Delta deciview for the maximum 98th percentile impact at Badlands National Park)
New LNBs with OFA0.191,556$2,268,806$1,4580.18
New LNBs with OFA and selective non-catalytic reduction (SNCR)0.152,4458,554,8963,485$6,9930.28
New LNBs with OFA and SCR0.054,88021,770,1344,4615,4490.57
Start Printed Page 5040

Table 3—Summary of EPA's Laramie River Unit 2 NOX BART Analysis

Control technologyEmission rate (lb/MMBtu; annual average)Emission reduction (tpy)Annualized costsAverage cost effectiveness ($/ton)Incremental cost effectiveness ($/ton)Visibility improvement (Delta deciview for the maximum 98th percentile impact at Badlands National Park)
New LNBs with OFA0.191823$2,268,806$1,2440.18
New LNBs with OFA and SNCR0.152,7178,531,6313,140$7,0060.27
New LNBs with OFA and SCR0.055,12922,691,4674,4245,8710.53

Table 4—Summary of EPA's Laramie River Unit 3 NOX BART Analysis

Control technologyEmission rate (lb/MMBtu; annual average)Emission reduction (tpy)Annualized costsAverage cost effectiveness ($/ton)Incremental cost effectiveness ($/ton)Visibility improvement (Delta deciview for the maximum 98th percentile impact at Badlands National Park)
New LNBs with OFA0.191789$2,268,806$1,2680.18
New LNBs with OFA and SNCR0.152,7068,643,8393,194$6,9510.27
New LNBs with OFA and SCR0.055,18122,666,9824,3755,6670.52

EPA's January 2014 modeling protocol, Appendix H, shows the model predicted visibility improvement for each emissions control technology at each of the Class I areas that we modeled in our analysis. For Laramie River we modeled visibility impairment at Badlands National Park, Wind Cave National Park, Rawah Wilderness Area, and Rocky Mountain National Park. At Laramie River Unit 1 the model visibility improvements with LNB/OFA/SCR were 0.57 deciviews at Badlands National Park, 0.47 deciviews at Wind Cave National Park, 0.25 deciviews at Rawah Wilderness Area, and 0.39 at Rocky Mountain National Park. At Laramie River Unit 2 the model visibility improvements with LNB/OFA/SCR were 0.53 deciviews at Badlands, 0.43 deciviews at Wind Cave, 0.26 deciviews at Rawah, and 0.31 at Rocky Mountain. At Laramie River Unit 3 the model visibility improvements with LNB/OFA/SCR were 0.52 deciviews at Badlands, 0.44 deciviews at Wind Cave, 0.23 deciviews at Rawah, and 0.28 at Rocky Mountain.

Table 5—Summary of EPA's Jim Bridger Unit 1 NOX BART Analysis

Control technologyEmission rate (lb/MMBtu; annual average)Emission reduction (tpy)Annualized costsAverage cost effectiveness ($/ton)Incremental cost effectiveness ($/ton)Visibility improvement (Delta deciview for the maximum 98th percentile impact at Bridger Wilderness Area) **
New LNBs with SOFA0.184,558$1,167,297$2560.17/0.23
New LNBs with SOFA and SNCR0.145,3324,330,052812$4,0880.20/0.27
New LNBs with SOFA and SCR0.057,35219,372,1052,6357,4470.27/0.37
* Values shown are model results using ammonia based on monitoring and default data; values in straight font are model results using a monitored monthly varying concentration/values in italics are model results using an IWAQM default 0.5 ppb background ammonia.

Table 6—Summary of EPA's Jim Bridger Unit 2 NOX BART Analysis

Control technologyEmission rate (lb/MMBtu; annual average)Emission reduction (tpy)Annualized costsAverage cost effectiveness ($/ton)Incremental cost effectiveness ($/ton)Visibility improvement (Delta deciview for the maximum 98th percentile impact at Bridger Wilderness Area) *
New LNBs with SOFA0.193,787$1,167,297$3080.16 /0.21
New LNBs with SOFA and SNCR0.154,5454,291,184944$4,1220.19 /0.25
New LNBs with SOFA and SCR0.056,55422,307,4923,4038,9680.27 /0.36
* Values shown are model results using ammonia based on monitoring and default data; values in straight font are model results using a monitored monthly varying concentration/values in italics are model results using an IWAQM default 0.5 ppb background ammonia.
Start Printed Page 5041

Table 7—Summary of EPA's Jim Bridger Unit 3 NOX BART Analysis

Control technologyEmission rate (lb/MMBtu; annual average)Emission reduction (tpy)Annualized costsAverage cost effectiveness ($/ton)Incremental cost effectiveness ($/ton)Visibility improvement (Delta deciview for the maximum 98th percentile impact at Bridger Wilderness Area) *
New LNBs with SOFA0.203,710$1,167,297$3150.14 /0.19
New LNBs with SOFA and SNCR0.164,5394,458,776982$3,9720.17 /0.23
New LNBs with SOFA and SCR0.056,79922,573,9203,3208,0150.26 /0.35
* Values shown are model results using ammonia based on monitoring and default data; values in straight font are model results using a monitored monthly varying concentration/values in italics are model results using an IWAQM default 0.5 ppb background ammonia.

Table 8—Summary of EPA's Jim Bridger Unit 4 NOX BART Analysis

Control technologyEmission rate (lb/MMBtu; annual average)Emission reduction (tpy)Annualized costsAverage cost effectiveness ($/ton)Incremental cost effectiveness ($/ton)Visibility improvement (Delta deciview for the maximum 98th percentile impact at Rawah Wilderness Area) *
New LNBs with SOFA0.194,161$1,167,297$2810.25 /0.23
New LNBs with SOFA and SNCR0.154,9564,372,457882$4,0350.30 /0.28
New LNBs with SOFA and SCR0.057,10819,494,4172,7437,0270.45 /0.42
* Values shown are model results using ammonia based on monitoring and default data; values in straight font are model results using a monitored monthly varying concentration/values in italics are model results using an IWAQM default 0.5 ppb background ammonia.

EPA's January 2014 modeling protocol, Appendix H, shows the model predicted visibility improvement for each emissions control technology at each of the Class I areas that we modeled in our analysis of Jim Bridger. Model simulations were performed using a monthly varying background ammonia concentration and using the IWAQM default concentration for forested areas of 0.5 ppb. For Jim Bridger we modeled visibility impairment at Bridger Wilderness Area, Fitzpatrick Wilderness Area, Mt Zirkel Wilderness Area, Rawah Wilderness Area, Rocky Mountain National Park, Grand Teton National Park, Teton Wilderness Area, Washakie Wilderness Area and Yellowstone National Park. Under the State's LTS, LNB/OFA/SCR would be required on Jim Bridger Units 1 and 2 in 2022 and 2021. Under the State's LTS, LNB/OFA/SCR would be required on Jim Bridger Units 3 and 4 in 2015 and 2016.

For Jim Bridger Unit 1, using monthly varying ammonia concentrations, model visibility improvements with LNB/OFA/SCR were: 0.37 deciviews at Bridger; 0.26 deciviews at Fitzpatrick; 0.29 deciviews at Mt Zirkel; 0.35deciviews at Rawah; 0.36 deciviews at Rocky Mountain; 0.17 deciviews at Grand Teton; 0.14 deciviews at Teton; 0.19 deciviews at Washakie; and 0.15 deciviews at Yellowstone.

For Jim Bridger Unit 1, using a constant 0.5 ppb ammonia concentration, model visibility improvements with LNB/OFA/SCR were: 0.37 deciviews at Bridger; 0.26 deciviews at Fitzpatrick; 0.29 deciviews at Mt Zirkel; 0.35 deciviews at Rawah; 0.36 deciviews at Rocky Mountain; 0.17 deciviews at Grand Teton; 0.14 deciviews at Teton; 0.19 deciviews at Washakie; and 0.15 deciviews at Yellowstone.

For Jim Bridger Unit 2, using monthly varying ammonia concentrations, model visibility improvements with LNB/OFA/SCR were: 0.36 deciviews at Bridger; 0.26 deciviews at Fitzpatrick; 0.28 deciviews at Mt Zirkel; 0.35 deciviews at Rawah; 0.36 deciviews at Rocky Mountain; 0.16 deciviews at Grand Teton; 0.14 deciviews at Teton; 0.19 deciviews at Washakie; and 0.14 deciviews at Yellowstone.

For Jim Bridger Unit 2, using a constant 0.5 ppb ammonia concentration, model visibility improvements with LNB/OFA/SCR were: 0.36 deciviews at Bridger; 0.26 deciviews at Fitzpatrick; 0.28 deciviews at Mt Zirkel; 0.35 deciviews at Rawah; 0.36 deciviews at Rocky Mountain; 0.16 deciviews at Grand Teton; 0.14 deciviews at Teton; 0.19 deciviews at Washakie; and 0.14 deciviews at Yellowstone.

For Jim Bridger Unit 3, using monthly varying ammonia concentrations, model visibility improvements with LNB/OFA/SCR were: 0.35 deciviews at Bridger; 0.25 deciviews at Fitzpatrick; 0.28 deciviews at Mt Zirkel; 0.33 deciviews at Rawah; 0.34 deciviews at Rocky Mountain; 0.16 deciviews at Grand Teton; 0.14 deciviews at Teton; 0.18 deciviews at Washakie; and 0.14 deciviews at Yellowstone.

For Jim Bridger Unit 3, using a constant 0.5 ppb ammonia concentration, model visibility improvements with LNB/OFA/SCR were: 0.35 deciviews at Bridger; 0.25 deciviews at Fitzpatrick; 0.28 deciviews at Mt Zirkel; 0.33 deciviews at Rawah; 0.34 deciviews at Rocky Mountain; 0.16 deciviews at Grand Teton; 0.14 deciviews at Teton; 0.18 deciviews at Washakie; and 0.14 deciviews at Yellowstone.

For Jim Bridger Unit 4, using monthly varying ammonia concentrations, model visibility improvements with LNB/OFA/SCR were: 0.38 deciviews at Bridger; 0.28 deciviews at Fitzpatrick; 0.19 deciviews at Mt Zirkel; 0.42 deciviews at Rawah; 0.38 deciviews at Rocky Mountain; 0.32 deciviews at Grand Teton; 0.15 deciviews at Teton; 0.30 deciviews at Washakie; and 0.16 deciviews at Yellowstone.

For Jim Bridger Unit 4, using a constant 0.5 ppb ammonia concentration, model visibility improvements with LNB/OFA/SCR were: 0.38 deciviews at Bridger; 0.28 Start Printed Page 5042deciviews at Fitzpatrick; 0.27 deciviews at Mt Zirkel; 0.42 deciviews at Rawah; 0.38 deciviews at Rocky Mountain; 0.32 deciviews at Grand Teton; 0.15 deciviews at Teton; 0.30 deciviews at Washakie; and 0.16 deciviews at Yellowstone.

Table 9—Summary of EPA's Dave Johnston Unit 3 NOX BART Analysis

[9 Year remaining useful life]

Control technologyEmission rate (lb/MMBtu; annual average)Emission reduction (tpy)Annualized costsAverage cost effectiveness ($/ton)Incremental cost effectiveness ($/ton)Visibility improvement (Delta deciview for the maximum 98th percentile impact at Wind Cave National Park)
New LNBs with OFA0.222,837$1,828,137$6440.33
New LNBs with OFA and SNCR0.163,3563,898,9301,162$3,9880.39
New LNBs with OFA and SCR0.054,43316,591,0063,74211,7810.51

Table 10—Summary of EPA's Dave Johnston Unit 3 NOX BART Analysis

[20 Year remaining useful life]

Control technologyEmission rate (lb/MMBtu; annual average)Emission reduction (tpy)Annualized costsAverage cost effectiveness ($/ton)Incremental cost effectiveness ($/ton)Visibility improvement (Delta deciview for the maximum 98th percentile impact at Wind Cave National Park)
New LNBs with OFA0.222,837$1,699,807$5990.33
New LNBs with OFA and SNCR0.163,3563,510,5891,046$3,4880.39
New LNBs with OFA and SCR0.054,43311,680,1442,6357,5830.51

Table 11—Summary of EPA's Dave Johnston Unit 4 NOX BART Analysis

Control technologyEmission rate (lb/MMBtu; annual average)Emission reduction (tpy)Annualized costsAverage cost effectiveness ($/ton)Incremental cost effectiveness ($/ton)Visibility improvement (Delta deciview for the maximum 98th percentile impact at Wind Cave National Park)
New LNBs with OFA0.143,114$767,342$2460.41
New LNBs with OFA and SNCR0.113,5052,541,600725$4,5350.46
New LNBs with OFA and SCR0.054,37714,158,8993,23513,3120.57

EPA's January 2014 modeling protocol, Appendix H, shows the model predicted visibility improvement for each emissions control technology at each of the Class I areas that we modeled in our analysis of Dave Johnston. For Dave Johnston we modeled visibility impairment at Badlands National Park, Wind Cave National Park, Mt Zirkel Wilderness Area, Rawah Wilderness Area, and Rocky Mountain National Park. At Dave Johnston Unit 3 the model visibility improvements with LNB/OFA/SCR were 0.47 deciviews at Badlands National Park, 0.51 deciviews at Wind Cave National Park, 0.20 deciviews at Mt Zirkel Wilderness Area, 0.40 deciviews at Rawah Wilderness Area, and 0.28 at Rocky Mountain National Park. At Dave Johnston Unit 4 the model visibility improvements with LNB/OFA were 0.55 deciviews at Badlands National Park, 0.57 deciviews at Wind Cave National Park, 0.24 deciviews at Mt Zirkel Wilderness Area, 0.34 deciviews at Rawah Wilderness Area, and 0.33 deciviews at Rocky Mountain National Park.Start Printed Page 5043

Table 12—Summary of EPA's Naughton Unit 1 NOX BART Analysis

Control technologyEmission rate (lb/MMBtu; annual average)Emission reduction (tpy)Annualized costsAverage cost effectiveness ($/ton)Incremental cost effectiveness ($/ton)Visibility improvement (Delta deciview for the Maximum 98th percentile impact at Bridger Wilderness Area) *
New LNBs with OFA0.212,100$932,466$4440.22/0.26
New LNBs with OFA and SNCR0.162,4632,234,827907$3,5840.26/0.30
New LNBs with OFA and SCR0.053,2099,974,6163,10910,3840.33/0.39
* Values shown are model results using ammonia based on monitoring and default data; values in straight font are model results using a monitored monthly varying concentration/values in italics are model results using an IWAQM default 0.5 ppb background ammonia.

Table 13—Summary of EPA's Naughton Unit 2 NOX BART Analysis

Control technologyEmission rate (lb/MMBtu; annual average)Emission reduction (tpy)Annualized costsAverage cost effectiveness ($/ton)Incremental cost effectiveness ($/ton)Visibility improvement (Delta deciview for the Maximum 98th percentile impact at Bridger Wilderness Area) *
New LNBs with OFA0.212,586$883,900$3420.28/0.32
New LNBs with OFA and SNCR0.163,0242,480,832820$3,6470.34/0.38
New LNBs with OFA and SCR0.053,92210,062,7502,5668,4400.42/0.46
* Values shown are model results using ammonia based on monitoring and default data; values in straight font are model results using a monitored monthly varying concentration/values in italics are model results using an IWAQM default 0.5 ppb background ammonia.

Table 14—Summary of EPA's Naughton Unit 3 NOX BART Analysis

[In lieu of conversion of Naughton Unit 3 to natural gas per PacifiCorp request]

Control technologyEmission rate (lb/MMBtu; annual average)Emission reduction (tpy)Annualized costsAverage cost effectiveness ($/ton)Incremental cost effectiveness ($/ton)Visibility improvement (Delta deciview for the Maximum 98th percentile impact at Bridger Wilderness Area) *
Existing LNBs with OFA**0.33442$106,393$2400.05/0.07
Existing LNBs with OFA and SNCR0.231,6733,852,3772,303$3,0450.20/0.29
Existing LNBs with OFA and SCR0.053,92213,604,7023,4694,3350.49/0.60
* Values shown are model results using ammonia based on monitoring and default data; values in straight font are model results using a monitored monthly varying concentration/values in italics are model results using an IWAQM default 0.5 ppb background ammonia.
** As used in this table, “existing” means the control technology that was in place at the time of the State's BART analyses in May 2009.

EPA's January 2014 modeling protocol, Appendix H, shows the model predicted visibility improvement for each emissions control technology at each of the Class I areas that we modeled in our analysis of Naughton. For Naughton we modeled visibility impairment at Bridger Wilderness Area, Fitzpatrick Wilderness Area, North Absaroka Wilderness Area, Washakie Wilderness Area, Teton Wilderness Area, Grand Teton National Park and Yellowstone National Park. Model simulations were performed using a monthly varying background ammonia concentration and using the IWAQM default concentration for forested areas of 0.5 ppb.

For Naughton Unit 1 model visibility improvements, using monthly varying ammonia concentrations, with LNB/OFA and LNB/OFA/SCR were, respectively: 0.22 and 0.33 deciviews at Bridger; 0.19 and 0.29 deciviews at Fitzpatrick; 0.10 and 0.14 at North Absaroka; 0.10 and 0.15 deciviews at Washakie; 0.10 and 0.16 deciviews at Teton; 0.15 and 0.23 deciviews at Grand Teton; and 0.12 and 0.18 deciviews at Yellowstone.

For Naughton Unit 1 model visibility improvements, using a constant 0.5 ppb ammonia concentration, with LNB/OFA and LNB/OFA/SCR were, respectively: 0.26 and 0.39 deciviews at Bridger; 0.22 and 0.30 deciviews at Fitzpatrick; 0.10 and 0.14 at North Absaroka; 0.12 and 0.17 deciviews at Washakie; 0.13 and 0.19 deciviews at Teton; 0.19 and 0.29 deciviews at Grand Teton; and 0.13 and 0.19 deciviews at Yellowstone.Start Printed Page 5044

For Naughton Unit 2 model visibility improvements, using monthly varying ammonia concentrations, with LNB/OFA and LNB/OFA/SCR were, respectively: 0.28 and 0.42 deciviews at Bridger; 0.25 and 0.36 deciviews at Fitzpatrick; 0.12 and 0.17 at North Absaroka; 0.15 and 0.22 deciviews at Washakie; 0.14 and 0.21 deciviews at Teton; 0.18 and 0.28 deciviews at Grand Teton; and 0.16 and 0.22 deciviews at Yellowstone.

For Naughton Unit 2 model visibility improvements, using a constant 0.5 ppb ammonia concentration, with LNB/OFA and LNB/OFA/SCR were, respectively: 0.32 and 0.46 deciviews at Bridger; 0.26 and 0.38 deciviews at Fitzpatrick; 0.12 and 0.17 at North Absaroka; 0.16 and 0.22 deciviews at Washakie; 0.17 and 0.25 deciviews at Teton; 0.25 and 0.38 deciviews at Grand Teton; and 0.17 and 0.24 deciviews at Yellowstone.

For Naughton Unit 3 model visibility improvements, using monthly varying ammonia concentrations, with LNB/OFA and LNB/OFA/SCR were, respectively: 0.05 and 0.49 deciviews at Bridger; 0.05 and 0.42 deciviews at Fitzpatrick; 0.03 and 0.24 at North Absaroka; 0.05 and 0.37 deciviews at Washakie; 0.04 and 0.38 deciviews at Teton; 0.04 and 0.38 deciviews at Grand Teton; and 0.04 and 0.39 deciviews at Yellowstone.

For Naughton Unit 3 model visibility improvements, using a constant 0.5 ppb ammonia concentration, with LNB/OFA and LNB/OFA/SCR were, respectively: 0.07 and 0.60 deciviews at Bridger; 0.05 and 0.44 deciviews at Fitzpatrick; 0.03 and 0.24 at North Absaroka; 0. and 0. deciviews at Washakie; 0.05 and 0.39 deciviews at Teton; 0.06 and 0.41 deciviews at Grand Teton; and 0.05 and 0.40 deciviews at Yellowstone.

Table 15—Summary of EPA's Wyodak NOX BART Analysis

Control technologyEmission rate (lb/MMBtu; annual average)Emission reduction (tpy)Annualized costsAverage cost effectiveness ($/ton)Incremental cost effectiveness ($/ton)Visibility improvement (delta deciview for the maximum 98th percentile impact at Wind Cave National Park)
New LNBs with OFA0.191,239$1,272,427$1,0270.21
New LNBs with OFA and SNCR0.151,9143,726,5731,9473,6350.32
New LNBs with OFA and SCR0.053,73515,073,5024,0366,2330.61

EPA's January 2014 modeling protocol, Appendix H, shows the model predicted visibility improvement for each emissions control technology at each of the Class I areas that we modeled in our analysis of Wyodak . For Wyodak we modeled visibility impairment at Badlands National Park and Wind Cave National Park. At Wyodak Unit 1 the model visibility improvements with LNB/OFA/SCR were 0.61 deciviews at Wind Cave and 0.38 deciviews at Badlands National Park.

Table 16—Summary of EPA's Dave Johnston Unit 1 NOX Reasonable Progress Analysis

Control technologyEmission rate (lb/MMBtu; annual average)Emission reduction (tpy)Annualized costsAverage cost effectiveness ($/ton)Incremental cost effectiveness ($/ton)Visibility improvement (delta deciview for the maximum 98th percentile impact at Wind Cave National Park)
LNBs with OFA *0.201,226$1,214,000$9900.12
LNBs with OFA and SNCR0.151,4662,096,4301,4303,6700.14
LNBs with OFA and SCR0.051,9476,808,3743,4969,7980.18
* As used in this and the following tables, control technology that is not preceded by either “new” or “existing” (as in the above tables) means the control technology will be installed for the first time.

Table 17—Summary of EPA's Dave Johnston Unit 2 NOX Reasonable Progress Analysis

Control technologyEmission rate (lb/MMBtu; annual average)Emission reduction (tpy)Annualized costsAverage cost effectiveness ($/ton)Incremental cost effectiveness ($/ton)Visibility improvement (delta deciview for the maximum 98th percentile impact at Wind Cave National Park)
LNBs with OFA0.201,180$1,441,146$1,2210.11
LNBs with OFA and SNCR0.151,4252,335,0221,6383,6450.14
LNBs with OFA and SCR0.051,9167,037,9693,6739,5880.18
Start Printed Page 5045

B. Changes to Our Proposed Determinations

1. Dave Johnston Unit 3

We proposed to require PacifiCorp Dave Johnston Unit 3 to meet a FIP emission limit of 0.07 lb/MMBtu (30-day rolling average) for NOX BART (assumes the installation of LNBs/OFA plus SCR). Based on our revised costs of compliance and visibility impacts, we would still conclude that NOX BART is an emission limit of 0.07 lb/MMBtu (30-day rolling average). PacifiCorp submitted comments on our proposed rulemaking on August 26, 2013. In those comments, PacifiCorp indicated in various places (e.g., page 37) that instead of installing SCR, it would shut down Dave Johnston Unit 3 in 2027. Our regulatory language now provides PacifiCorp two alternative paths to compliance with the FIP. The first path includes a requirement for Dave Johnston Unit 3 to cease operation by December 31, 2027. For this path, we are requiring Dave Johnston Unit 3 to meet a FIP limit of 0.28 lb/MMBtu (30-day rolling average) no later than five years after the date of our final action. This emission limit assumes the installation of LNBs/OFA. The second compliance path gives PacifiCorp the option to instead meet a 0.07 lb/MMBtu emission limit (assumes installation of SCR) within five years of our final action with no requirement for shut down.

EPA met with PacifiCorp on October 31, 2013, to clarify the comments submitted by PacifiCorp (see October 31, 2013 memo to docket). Specifically, EPA asked if, in lieu of a requirement for SCR, PacifiCorp was asking for EPA to include an enforceable requirement in the FIP for Dave Johnston Unit 3 to shut down in 2027, and for EPA to make a BART determination based on that limited remaining useful life. PacifiCorp confirmed that it did want EPA to include an enforceable requirement in the FIP for PacifiCorp to shut down Dave Johnston Unit 3 by December 31, 2027, and to make a BART determination accordingly. As detailed in the following section, we determined that if the unit shuts down by December 31, 2027, SCR would no longer be NOX BART.

Generally, EPA does not interpret the regional haze rule to provide us with authority to make a BART determination that requires the shutdown of a source. In other states, we have approved state-adopted requirements for the shutdown of a source, which have usually been negotiated between the source operator and the state, and we have accordingly approved BART determinations that took into account the resulting shorter useful life of the affected source. In the case of Dave Johnson Unit 3, the State has not submitted a SIP revision to require the shutdown that PacifiCorp intends to implement, so there is no enforceable shutdown commitment that we can approve. We believe that without an enforceable requirement for the shutdown, we cannot make a BART determination that reflects the shorter planned useful life of the unit. Therefore, we are incorporating the shutdown requirement into one of the two compliance paths available to PacifiCorp, in order to allow it to only be required to install and maintain the less expensive LNBs/OFA emission controls rather than the more expensive SCR controls. We welcome a SIP revision that would make the shutdown requirement State law, and we would withdraw the shutdown requirement from the SIP upon approving such a SIP revision.

2. Dave Johnston Unit 4

We proposed to require PacifiCorp Dave Johnston Unit 4 to meet a FIP emission limit of 0.12 lb/MMBtu (30-day rolling average) for NOX BART (assuming the installation of LNBs/OFA with SNCR). Based on our revised costs of compliance and visibility impacts, we no longer conclude that NOX BART is an emission limit of 0.12 lb/MMBtu (30-day rolling average). Based on our new cost and visibility improvement numbers, we conclude that NOX BART is represented by the SIP emission limit of 0.15 lb/MMBtu (30-day rolling average) for this unit. This emission limit assumes the installation of LNBs/OFA. As such, we are approving Wyoming's NOX BART determination for Dave Johnston Unit 4.

3. Naughton Units 1 and 2

We proposed to require PacifiCorp Naughton Units 1 and 2 to meet a FIP emission limit of 0.07 lb/MMBtu (30-day rolling average) for NOX BART (assuming the installation of LNBs/OFA with SCR). As detailed in the next section, based on our revised costs of compliance and visibility impacts, we no longer conclude that NOX BART is an emission limit of 0.07 lb/MMBtu (30-day rolling average). Based on our new cost and visibility improvement numbers, we conclude that NOX BART is represented by the SIP emission limit of 0.26 lb/MMBtu (30-day rolling average) for each unit. This emission limit assumes the installation of LNBs/OFA. As such, we are approving Wyoming's NOX BART determination for Naughton Units 1 and 2.

4. Naughton Unit 3

We proposed to approve the State's NOX BART determination for Naughton Unit 3, which was an emission limit of 0.07 lb/MMBtu (30-day rolling average) (assumes the installation of LNBs/OFA with SCR). PacifiCorp submitted comments on our proposed rulemaking on August 26, 2013. In those comments, PacifiCorp indicated (page 72) that instead of installing SCR as required by the SIP, it plans to convert Naughton Unit 3 to natural gas in 2018 without installation of any post-combustion control of NOX emissions. Conversion to natural gas in this manner can be expected to result in NOX emissions that are higher than the 0.07 lb/MMBtu limit in the SIP combined with much lower SO2 and PM emissions, with a substantially lower overall remaining impact on visibility. On July 5, 2013, Wyoming issued Air Quality permit MD-14506 to PacifiCorp that reflects the conversion of Naughton Unit 3 to natural gas in June of 2018. EPA met with PacifiCorp on October 31, 2013, to clarify the comments submitted by PacifiCorp (see October 31, 2013 memo to docket). PacifiCorp requested that EPA include in its final action the emission limits for SO2, PM, and NOX that the State had in its permit MD-14506 that it issued to PacifiCorp. EPA supports PacifiCorp's conversion of Naughton Unit 3 to natural gas. However, we have the authority and obligation to take action on the SIP as submitted by the State, and there is no basis to disapprove the SIP. Since we are approving the SIP, we do not have authority to impose FIP limits even if independently requested by a source. Therefore, we cannot use the FIP to relieve Naughton Unit 3 of the obligation to achieve the 0.07 lb/MMBtu NOX emission limit in the SIP nor to impose emission limits for SO2 and PM that reflect the planned conversion to natural gas. Under the terms of the SIP, the compliance deadlines for the emission limits in the SIP for Naughton Unit 3 do not become effective until five years after our final action. We understand that Wyoming intends to submit a revision to their regional haze SIP for Naughton Unit 3 that reflects the BART NOX emission limits in its permit MD-14506 as soon as practicable. EPA intends to act on this SIP revision in an expedited timeframe to reflect the conversion of Naughton Unit 3 to natural gas and a revised BART NOX limit. In our final action we are approving Wyoming's NOX BART determination for Naughton Unit 3. Our regulatory language reflects the following emission limit for Naughton Start Printed Page 5046Unit 3 for NOX: 0.07 lb/MMBtu (30-day rolling average).

5. Wyodak

We proposed to require PacifiCorp Wyodak Unit 1 to meet a FIP emission limit of 0.17 lb/MMBtu (30-day rolling average) for NOX BART (assuming the installation of LNBs/OFA with SNCR). Based on our revised costs of compliance and visibility impacts, as well as comments received during the public comment period (see section V), we no longer conclude that NOX BART is an emission limit of 0.17 lb/MMBtu (30-day rolling average). Based on our new cost and visibility improvement numbers, we conclude that NOX BART is a FIP emission limit of 0.07 lb/MMBtu (30-day rolling average) for this unit. This emission limit assumes the installation of LNBs/OFA with SCR. As detailed in the next section, based on our weighing of the five factors, we find that the average cost-effectiveness of SCR ($4,036/ton) and the incremental cost-effectiveness ($6,233/ton), combined with a visibility improvement of 0.61 deciviews at the most impacted Class I area, makes the selection of SCR for BART reasonable.

6. Jim Bridger

In our proposal, we proposed to approve the State's NOX BART and LTS determinations for Jim Bridger Units 1 and 2. The State's BART determination required each unit to meet an emissions limit of 0.26 lb/MMBtu (30-day rolling average) within five years of our approval of the SIP, based on new LNB plus OFA. The LTS determination required each unit to meet an emission limit of 0.07 lb/MMBtu (30-day rolling average) by December 31, 2022, and December 31, 2021, respectively. EPA proposed to approve these compliance dates for numerous reasons as discussed in detail in our proposed rulemaking. 78 FR 34755. We also proposed an alternative FIP BART determination that would require Jim Bridger Units 1 and 2 to meet an emission limit of 0.07 lb/MMBtu (30-day rolling average) within five years of our final rulemaking. 78 FR 34780. We are finalizing our proposed approval of the State's BART and LTS determinations for Jim Bridger Units 1 and 2, although the reasons for our final action on Jim Bridger Units 1 and 2 have changed from our proposed action.

In our proposed rulemaking, we stated:

EPA is proposing to determine that BART for all units at Jim Bridger would be SCR if the units were considered individually, based on the five factors, without regard for the controls being required at other units in the PacifiCorp system. However, when the cost of BART controls at other PacifiCorp owned EGUs is considered as part of the cost factor for the Jim Bridger Units, EPA is proposing that Wyoming's determination that NOX BART for these units is new LNB plus OFA for is reasonable. Considering costs broadly, it would be unreasonable to require any further retrofits at this source within five years of our final action. We note that the CAA establishes five years at the longest period that can be allowed for compliance with BART emission limits.” 78 FR 34756. However, as discussed in detail in section V.D.2 below, we do not think PacifiCorp has presented ample evidence to show that it would be unreasonable or not feasible for them to install numerous SCRs within the five year BART period. Nonetheless, we are approving the State's BART determination and LTS for Jim Bridger Units 1 and 2 based on our consideration of the five factors, as detailed in the next section.

We are approving the State's SIP requirement that Jim Bridger Units 1 and 2 meet an emission limit of 0.07 lb/MMBtu (30-day rolling average) by 2022 and 2021, respectively. We are also approving the State's BART determination that requires Jim Bridger Units 1 and 2 to meet a NOX emission limit of 0.26 lb/MMBtu (30-day rolling average) within five years of our final action.

For Jim Bridger Units 3 and 4 we proposed to approve the SIP with regard to the State's determination that the appropriate level of NOX control for Units 3 and 4 for purposes of reasonable progress is the SCR-based emission limit in the SIP of 0.07 lb/MMBtu, with compliance dates of December 31, 2015 for Unit 3 and December 31, 2016 for Unit 4. In our proposal we noted that since the State is requiring PacifiCorp to install the LTS controls within the timeline that BART controls would have to be installed pursuant to 40 CFR 51.308(e)(iv), we proposed to approve the State's compliance schedule and emission limit of 0.07 lb/MMBtu for Jim Bridger Units 3 and 4 as meeting the BART requirements.

We are finalizing our proposed approval of the State's BART and LTS determinations for Jim Bridger Units 3 and 4, although, similar to Units 1 and 2, the reasons for our final action on Units 3 and 4 have changed from our proposed action.

7. Dave Johnston Units 1 and 2

We proposed to require PacifiCorp Dave Johnston Units 1 and 2 to meet a FIP emission limit of 0.22 lb/MMBtu (30-day rolling average) for NOX under reasonable progress (assuming the installation of LNBs/OFA). As detailed in the next section, based on our revised costs and visibility impacts, we no longer conclude that an emission limit of 0.22 lb/MMBtu (30-day rolling average) is warranted. We are approving Wyoming's NOX reasonable progress determinations for Dave Johnston Units 1 and 2 (i.e., no controls).

IV. Basis for Our Final Action

We have fully considered all significant comments on our proposal and have concluded that no changes from our proposal other than those discussed in detail above are warranted. Our action is based on an evaluation of Wyoming's regional haze SIP against the regional haze requirements at 40 CFR 51.300-51.309 and CAA sections 169A and 169B. All general SIP requirements contained in CAA section 110, other provisions of the CAA, and our regulations applicable to this action were also evaluated. The purpose of this action is to ensure compliance with these requirements. Our authority for action on Wyoming's SIP submittal is based on CAA section 110(k). Our authority to promulgate a FIP is based on CAA section 110(c).

In our proposal, EPA asked interested parties to provide additional information on both our evaluation of the BART factors and our proposed determinations. 78 FR 38745. We provided notice that any supplemental information we received could lead us to select BART control technologies or compliance deadlines that differed from our proposal. In response to this request, we received extensive comments on the visibility modeling and cost estimates that we provided in the proposal for NOX BART control technologies. As a result of these comments, we have revised our visibility modeling and cost estimates. The details of these changes and our reasons for making them are provided elsewhere in this document and in our responses to the comments. Based on these changes, we have reassessed our proposed action on the State's NOX BART determinations for each of the subject-to-BART sources by re-evaluating the five statutory factors.[16] We have also reassessed our proposed action on the State's NOX reasonable progress determination for Dave Johnston Units 1 and 2. In this section, we describe in detail our reassessment of the statutory factors for these sources based on our revised visibility modeling and cost estimates. For two sources—Jim Bridger and Wyodak—we also received additional comments, explained below, that caused us to Start Printed Page 5047reconsider certain aspects of our decision for those sources.

EPA notes that, in considering the visibility improvements reflected in our revised modeling, EPA interprets the BART Guidelines to require consideration of the visibility improvement from BART applied to the entire BART-eligible source. The BART Guidelines explain that, “[i]f the emissions from the list of emissions units at a stationary source exceed a potential to emit of 250 tons per year for any visibility-impairing pollutant, then that collection of emissions units is a BART-eligible source.” In other words, the BART-eligible source (the list of BART emissions units at a source) is the collection of units for which one must make a BART determination. The BART Guidelines state “you must conduct a visibility improvement determination for the source(s) as part of the BART determination.” This requires consideration of the visibility improvement from BART applied to the BART-eligible source as a whole.

We note, however, that while our regulations require states and EPA to assess visibility improvement on a source-wide basis, they provide flexibility to also consider unit-specific visibility improvement in order to more fully inform the reasonableness of a BART determination, but that does not replace the consideration of visibility benefit from the source (facility) as a whole. In making the BART determinations in this final action we have considered visibility improvements at the source, and then also at the units that comprise the source.

As explained in more detail later in this decision, we received during the comment period significant input on expected costs associated with different control technologies. We discuss in the section above and in our response to comments, the changes we made in response to comments received on costs of different control technologies. As discussed above and in our response to comments, we have revised our modeling analysis in light of the input we received during the public comment period. This additional information and analysis result in different costs and visibility benefits, two of the five BART factors. In some cases this leads us to finalize our proposal, and in other cases to reach a different conclusion.

This decision, which addresses multiple facilities in a state where numerous Class 1 areas are impacted to a greater or lesser degree, illustrates clearly the case-by-case nature of the BART determination process. The interplay among the five factors, and in particular the cost and visibility factors, is highly significant and determinative of the outcome. In considering this information, as we have noted in prior decisions, our first assessment is whether the state's determination is reasonable in light of the facts and consistent with the requirements of the Clean Air Act and implementing regulations. If we determine that it is, even if we might have reached a different outcome if it were our decision to make in the first instance, we will approve the SIP.

Below is a more specific discussion of our determinations in the final decision. As stated above more detailed information on our determinations can be found in the response to comments sections of this rulemaking.

A. Laramie River

The State's regional haze SIP determined that NOX BART for Laramie River Units 1, 2, and 3 is new LNB/SOFA. We proposed to disapprove the State's determination because the State neglected to reasonably assess the costs of compliance and visibility improvement in accordance with the BART Guidelines. 78 FR 34766. After revising the State's costs and modeling and re-evaluating the statutory factors, we proposed to determine that NOX BART is LNB/SOFA + SCR, with an emissions limit of 0.07 lb/MMBtu for each unit. We sought comment generally on the BART factors and our control determinations and indicated that we could revise our control determinations depending on any new information that we received.

As the result of the comments received on our proposal, we have further revised our calculation of the costs of compliance and visibility modeling. We have considered any comments on the other BART factors but we have not changed our assessment of the other BART factors. The revised visibility modeling for the most impacted Class I area (Badlands) is presented in the following table.

Table 18—Visibility Modeling for Laramie River Station

Laramie River StationLNB/SOFALNB/SOFA + SNCRLNB/SOFA + SCR
Unit 10.18 deciviews0.28 deciviews0.57 deciviews
Unit 20.18 deciviews0.27 deciviews0.53 deciviews
Unit 30.18 deciviews0.27 deciviews0.52 deciviews
Total *0.54 deciviews0.82 deciviews1.62 deciviews
* The total visibility improvement was estimated as the sum of the visibility improvement from each unit.

We also considered the visibility improvement at other impacted Class I areas (Wind Cave, Rawah, and Rocky Mountain), which range from 0.25 to 0.47 deciviews, 0.26 to 0.43 deciviews, and 0.23 to 0.44 deciviews, for Units 1, 2, and 3, respectively. Further details regarding our revised visibility modeling and cost estimates were provided in section III.A.

After re-evaluating the BART factors, we continue to find that LNB/SOFA + SCR is reasonable as BART and are therefore finalizing our proposal. The visibility improvement associated with LNB/SOFA + SCR at the most impacted Class I area is significant on both a source-wide (1.62 deciviews) and unit-specific (0.52-0.57 deciviews) basis. The significant visibility improvement at three other impacted Class I areas also supports the selection of this option. Finally, we believe that the incremental visibility improvement at the most impacted Class I area of SCR over SNCR (nearly double in all cases) warrants the selection of the most stringent control.

In regards to the costs of compliance, we found that the revised average and incremental cost-effectiveness of LNB/SOFA + SCR is in line with what we have found to be acceptable in our other FIPs. The average cost-effectiveness per unit ranges from $4,375 to $4,461/ton, while the incremental cost-effectiveness ranges from $5,449 to $5,871/ton. We believe that these costs are reasonable, especially in light of the significant visibility improvement associated with LNB/SOFA + SCR. As a result, we are finalizing our proposed disapproval of the State's NOX BART determination for Laramie River Station and finalizing our proposed FIP that includes a NOX BART determination of LNB/SOFA + SCR, Start Printed Page 5048with an emission limit of 0.07 lb/ MMBtu (30-day rolling average).

B. Jim Bridger

The State's regional haze SIP determined that NOX BART for Jim Bridger Units 1-4 is new LNBs with SOFA. The State also determined that SCR should be installed at each unit as part of the State's long-term strategy to achieve reasonable progress at several Class I areas, and set compliance dates of December 31, 2022, December 31, 2021, December 31, 2015, and December 31, 2016 for Units 1-4, respectively.

In our proposal, we indicated that the State had neglected to reasonably assess the costs of compliance and visibility improvement for Jim Bridger in accordance with the BART Guidelines. We nonetheless proposed to approve the State's BART and reasonable progress determinations for Units 3 and 4 because the compliance deadlines to install SCR on these units were sufficient to meet the requirements of BART. We are now finalizing our proposed action for Units 3 and 4.

We also proposed to approve the State's BART and reasonable progress determinations for Units 1 and 2, but on a different basis. There, we indicated that given the number of SCR retrofits PacifiCorp had to perform in Wyoming and in other states, it might not be affordable for PacifiCorp to install two additional SCRs on Jim Bridger Units 1 and 2 within the five-year BART compliance period. We requested additional information from commenters regarding whether the affordability provisions of the BART Guidelines should be applied to Units 1 and 2. In the alternative, we proposed to find that NOX BART for Units 1 and 2 was an emission limit of 0.07 lb/ MMBtu (30-day rolling average) based on the installation of LNB/SOFA + SCR with a compliance deadline of five years. Under this scenario, we acknowledged that the cost-effectiveness of LNB/SOFA + SCR at Units 1 and 2 was within the range of what EPA and the State itself had found reasonable in other BART determinations. We also considered the significant visibility improvement demonstrated by the State's modeling to warrant LNB/SOFA + SCR as BART. Finally, we sought comment generally on the BART factors and our control determinations and indicated that we could revise our control determinations depending on any new information that we received.

In response to our proposal, we received both supportive and adverse comments regarding whether the affordability provisions of the BART Guidelines should apply to Units 1 and 2. As explained in more detail in our responses to these comments, we agree that PacifiCorp did not make a sufficient showing that it could not afford to install LNB/SOFA + SCR on Units 1 and 2 within the five-year compliance period. Nevertheless, we also received new information regarding the costs of compliance and visibility benefits associated with Jim Bridger and have revised our cost estimates and visibility modeling for all four units accordingly. We have considered any comments on the other BART factors but we have not changed our assessment of the other BART factors.

The revised visibility modeling for the most impacted Class I area (Bridger) is presented in the following table (with straight font representing modeled results using an ammonia background based on a monitored monthly varying concentration, italicized font representing modeled results using IWAQM default 0.5 ppb background ammonia).[17]

Table 19—Visibility Modeling for Jim Bridger

Jim BridgerLNB/SOFALNB/SOFA + SNCRLNB/SOFA + SCR
Unit 10.17/0.23 deciviews0.20/0.27 deciviews0.27/0.37 deciviews
Unit 20.16/0.21 deciviews0.19/0.25 deciviews0.27/0.36 deciviews
Unit 30.14/0.19 deciviews0.17/0.23 deciviews0.26/0.35 deciviews
Unit 40.25/0.23 deciviews0.30/0.28 deciviews0.45/0.42 deciviews
Total *0.72/0.86 deciviews0.86/1.03 deciviews1.25/1.5 deciviews
* The total visibility improvement was estimated as the sum of the visibility improvement from each unit.

We also considered the visibility improvements at other impacted Class I areas (Bridger, Fitzpatrick, Rawah, Rocky Mountain, Grand Teton, Teton, Washakie, and Yellowstone), which range from 0.26 to 0.91 deciviews, 0.26 to 0.89 deciviews, 0.24 to 0.87 deciviews, and 0.27 to 1.0 deciviews, for Units 1-4, respectively. Further details regarding our revised visibility modeling and cost estimates are provided in section III.A.

After re-evaluating the BART factors, we are approving the State's determination that LNB/SOFA is NOX BART for Units 1-4. The visibility improvement associated with LNB/SOFA + SCR at the most impacted Class I area is significant on a source-wide basis (1.25 to 1.5 deciviews). The fact that Jim Bridger Station affects a number of other Class I areas, which also would see appreciable visibility improvement with the installation of LNB/SOFA + SCR, also weighs in favor of selecting this option as BART. The unit-specific benefits for Units 1 and 2 are somewhat more modest (0.27-0.37 deciviews), however, especially considering the low incremental improvement over SNCR (0.07-0.11 deciviews). The incremental visibility improvement of SNCR over LNB/SOFA is even smaller (0.03-0.04 deciviews).

In regards to the costs of compliance, we found that the revised average cost-effectiveness of LNB/SOFA + SCR is in line with what we have found to be acceptable in our other FIPs. The average cost-effectiveness is $4,088 and $4,461/ton at Units 1 and 2, respectively. The incremental cost-effectiveness, on the other hand, is on the high end of what we have found to be reasonable in our other FIPs. The incremental cost-effectiveness is $7,477 and $8,986/ ton at Units 1 and 2, respectively.

Ultimately however, while we believe that these costs and visibility improvements could potentially justify LNB/SOFA + SCR as BART, because this is a close call and because the State has chosen to require SCR as a reasonable progress control, we believe deference to the State is appropriate in this instance. We are therefore finalizing Start Printed Page 5049our approval of the State's determination to require SCR at Jim Bridger Units 1-4, with an emission limit of 0.07 lb/MMBtu (30-day rolling average), as part of its long-term strategy. We are also finalizing our approval of the compliance dates of December 31, 2022, December 31, 2021, December 31, 2015, and December 31, 2016 for Units 1- 4 respectively.

C. Dave Johnston Units 3 and 4

The State's regional haze SIP determined that NOX BART for Dave Johnston Units 3 and 4 is LNB/OFA. We proposed to disapprove the State's determination because the State neglected to reasonably assess the costs of compliance and visibility improvement in accordance with the BART Guidelines. 78 FR 34778. After revising the State's costs and modeling and re-evaluating the statutory factors, we proposed to determine that NOX BART for Unit 3 is LNB/SOFA + SCR, with an emission limit of 0.07 lb/ MMBtu (30-day rolling average). We proposed that NOX BART for Unit 4 is LNB/SOFA + SNCR, with an emission limit of 0.12 lb/ MMBtu. We sought comment generally on the BART factors and our control determinations and indicated that we could revise our control determinations depending on any new information that we received.

As the result of the comments received on our proposal, we have further revised our calculation of the costs of compliance and visibility modeling. We have considered any comments on the other BART factors but we have not changed our assessment of the other BART factors. The revised visibility modeling for the most impacted Class I area (Wind Cave) is presented in the following table.

 Table 20—Visibility Modeling for Dave Johnston (BART Units)

Dave JohnstonLNB/OFALNB/OFA + SNCRLNB/OFA + SCR
Unit 30.33 deciviews0.39 deciviews0.51 deciviews
Unit 40.41 deciviews0.46 deciviews0.57 deciviews
Total *0.74 deciviews0.85 deciviews1.08 deciviews
* The total visibility improvement was estimated as the sum of the visibility improvement from each unit.

We also considered the visibility improvement at other impacted Class I areas (Badlands, Mt Zirkel, Rawah, and Rocky Mountain), which range from 0.20 to 0.47 deciviews and 0.24 to 0.55 deciviews, for Units 3 and 4, respectively. Further details regarding our revised visibility modeling and cost estimates were provided in section III.A.

After re-evaluating the BART factors, we no longer believe that LNB/OFA + SNCR is NOX BART for Dave Johnston Unit 4. As we explained in the proposal, the incremental cost-effectiveness of LNB/OFA + SCR was and continues to be excessive ($13,312), so we have eliminated this control option. While the revised average and incremental costs of LNB/OFA + SNCR continue to be reasonable, the incremental visibility improvement of SNCR over LNB/OFA is now only 0.05 deciviews. In light of this new visibility information, we believe that the State's determination that LNB/OFA is NOX BART for Unit 4 was reasonable and are approving it accordingly.

In regards to Dave Johnston Unit 3, we continue to believe that LNB/OFA + SCR is NOX BART. The visibility improvement associated with LNB/SOFA + SCR at the most impacted Class I area is significant (0.51 deciviews). The visibility improvement at several other impacted Class I areas also supports the selection of this option. Finally, we do not believe that the incremental visibility improvement at the most impacted Class I area of SCR over SNCR (0.12 deciviews) is sufficiently insignificant to warrant the elimination of the most stringent control in this instance.

In regards to the costs of compliance, we found that the revised average and incremental cost-effectiveness of LNB/SOFA + SCR is in line with what we have found to be acceptable in our other FIPs. The average cost-effectiveness is $2,635/ton, while the incremental cost-effectiveness is $7,583/ton. We believe that these costs are reasonable, especially in light of the significant visibility improvement associated with LNB/SOFA + SCR.

In response to other comments we received, we also considered an alternative BART analysis for Unit 3 based on PacifiCorp's commitment to retire Unit 3 by 2027 in lieu of installing SCR. Using a 9-year remaining useful life as the amortization period for Unit 3, the incremental cost-effectiveness of LNB/OFA + SCR becomes excessive ($11,781). Furthermore, the incremental visibility improvement at the most impacted Class I area from use of LNB/OFA to use of LNB/OFA+ SNCR is only 0.06 deciviews. Thus, taking all five factors into account, including the remaining useful life of nine years, we conclude that the NOX BART would be LNB/OFA in this scenario.

To provide flexibility, we are finalizing both scenarios in a FIP for Dave Johnston Unit 3. Under the first scenario, we are finalizing a NOX BART determination of LNB/OFA + SCR, with an emission limit of 0.07 lbs/ MMBtu (30-day rolling average). Under the alternative scenario, based on a commitment to retire Unit 3 by 2027, we are finalizing a NOX BART determination of LNB/OFA, with an emission limit of 0.28 lbs/ MMBtu (30-day rolling average).

D. Naughton

The State's regional haze SIP determined that NOX BART is new LNB/OFA for Naughton Units 1 and 2 and LNB/OFA + SCR for Naughton Unit 3. We proposed to approve the State's determination for Unit 3, but proposed to disapprove the State's determination for Units 1 and 2 because the State neglected to reasonably assess the costs of compliance and visibility improvement in accordance with the BART Guidelines. 78 FR 34748. After revising the State's costs and modeling and re-evaluating the statutory factors, we proposed to determine that NOX BART for Units 1 and 2 is LNB/SOFA + SCR, with an emissions limit of 0.07 lb/MMBtu for each unit. We sought comment generally on the BART factors and our control determinations and indicated that we could revise our control determinations depending on any new information that we received.

As the result of the comments received on our proposal, we have further revised our calculation of the costs of compliance and visibility modeling. We have considered any comments on the other BART factors but we have not changed our assessment of the other BART factors. The revised visibility modeling for the most impacted Class I area (Bridger) is presented in the following table (with straight font representing modeled Start Printed Page 5050results using an ammonia background based on a monitored monthly varying concentration, italicized font representing modeled results using IWAQM default 0.5 ppb background ammonia).

Table 21—Visibility Modeling for Naughton

NaughtonLNB/OFALNB/OFA + SNCRLNB/OFA + SCR
Unit 10.22/0.26 deciviews0.26/0.30 deciviews0.33/0.39 deciviews.
Unit 20.28/0.32 deciviews0.34/0.38 deciviews0.42/0.46 deciviews.
Unit 30.05/0.07 deciviews0.20/0.29 deciviews0.49/0.60 deciviews.
Total *0.55/0.65 deciviews0.80/0.97 deciviews1.24/1.45 deciviews
* The total visibility improvement was estimated as the sum of the visibility improvement from each unit.

We also considered the visibility improvement at other impacted Class I areas (Fitzpatrick, North Absaroka, Washakie, Teton, Grand Teton, and Yellowstone), which range from 0.10 to 0.30 deciviews, 0.08 to 0.42 deciviews, and 0.13 to 0.49 deciviews, for Units 1, 2, and 3, respectively. Further details regarding our revised visibility modeling and cost estimates were provided in section III.A.

After re-evaluating the BART factors, we no longer believe that LNB/OFA + SCR is NOX BART for Naughton Units 1 and 2. The visibility improvement associated with LNB/SOFA + SCR at the most impacted Class I area remains significant on a source-wide basis (1.24-1.45 deciviews) but more modest on a unit-specific basis (0.33-0.46 deciviews). The visibility improvement at six other impacted Class I areas continues to support the selection of this option as well. In regards to the costs of compliance, however, we found that while the revised average cost-effectiveness values for LNB/OFA + SCR were acceptable, the revised incremental cost-effectiveness values were beyond the upper end of the range (higher even than Jim Bridger) of what we have found to be acceptable in our other FIPs. For Units 1 and 2, respectively, the average cost-effectiveness per unit is $3,109 and $2,566/ ton, while the incremental cost-effectiveness is $10,384 and $8,440/ ton. Consequently, we believe that it was not unreasonable for the State to reject LNB/OFA + SCR as BART. Furthermore, we cannot say the State acted unreasonably in rejecting LNB/OFA + SNCR at Units 1 and 2 because the incremental visibility improvement of SNCR over LNB/OFA, while possibly appreciable, is very low at just 0.10 deciviews across both units. Therefore, based on our analysis we believe that the State's determination that LNB/OFA is NOX BART for Units 1 and 2, with an emission limit of 0.28 lbs/ MMBtu, was ultimately reasonable and are approving it accordingly.

E. Wyodak

The State's regional haze SIP determined that NOX BART for Wyodak Unit 1 is new LNBs with OFA. We proposed to disapprove the State's determination because the State neglected to reasonably assess the costs of compliance and visibility improvement in accordance with the BART Guidelines. 78 FR 34784-34785. As a result, we also proposed a FIP for NOX BART. After considering the BART factors, we noted that the cost-effectiveness and visibility improvement of the most stringent control option, LNB/OFA + SCR, were within the range of values that EPA had found reasonable in other FIPs. However, we proposed not to require LNB/OFA + SCR as NOX BART for Wyodak Unit 1. Instead, we proposed to require LNB/OFA + SNCR based on the reasoning that the cumulative visibility improvement of SCR across all Class I areas was low when compared to the cumulative visibility improvement associated with SCR at Dave Johnston Unit 3, Laramie River Units 1-3, and Naughton Units 1 and 2. We sought comment generally on the BART factors and our control determinations and indicated that we could revise our control determinations depending on any new information that we received. Based on our discussion of LNB/OFA + SCR at Wyodak, that control option was among those that we invited comment on.

In response to our proposal for Wyodak, we received comments that cumulative visibility improvement should not be used as a basis to reject a control option that has already been deemed reasonable based on visibility improvement at the most impacted Class I area. The commenters pointed out that such an approach would have the illogical effect of allowing an added benefit (visibility improvement at multiple Class I areas) to weigh in favor of less stringent controls. We agree with this criticism and want to make clear today that where a control is warranted as BART based on the costs of controls and visibility benefits at the most impacted area alone, cumulative visibility benefits can only strengthen the case for that control, not suggest that it is unwarranted. Similarly, where a control might not be warranted as BART based on the improvement at a single Class I area, significant cumulative benefits are an additional consideration that could warrant that the control be selected as BART.

In addition, we have further revised our calculation of the costs of compliance and visibility modeling for Wyodak Unit 1. We have not changed our assessment of the other BART factors. The revised visibility modeling for the most impacted Class I area (Wind Cave) is presented in the following table.

Table 22—Visibility Modeling for Wyodak

WyodakLNB/SOFALNB/SOFA + SNCRLNB/SOFA + SCR
Unit 10.21 deciviews0.32 deciviews0.61 deciviews.

We also considered the visibility improvement at a second impacted Class I area (Badlands), which is a maximum of 0.38 deciviews for LNB/SOFA + SCR. Further details regarding our revised visibility modeling and cost estimates were provided in the previous section.Start Printed Page 5051

After re-evaluating the BART factors and dismissing our earlier rationale for rejecting an otherwise reasonable control, we find that LNB/SOFA + SCR is reasonable as BART. As the BART-eligible source in this case is a single unit, the source-wide and unit-specific visibility improvements associated with the various control options are the same. The visibility improvement associated with LNB/SOFA + SCR at the most impacted Class I area (0.61 deciviews) is significant. There is also a more modest visibility improvement (0.38 deciviews) at a second impacted Class I area that supports the selection of this option. Finally, we believe that the incremental visibility improvement at the most impacted Class I area of SCR over SNCR (nearly double) warrants the selection of the most stringent control.

In regards to the costs of compliance, we found that the revised average and incremental cost-effectiveness of LNB/SOFA + SCR is in line with what we have found to be acceptable in our other FIPs. The average cost-effectiveness is $4,036/ton, while the incremental cost-effectiveness of SCR over SNCR is $6,223/ton. We believe that these costs are reasonable, especially in light of the significant visibility improvement associated with LNB/SOFA + SCR at Wind Cave. As a result, we are finalizing our proposed disapproval of the State's NOX BART determination for Wyodak Unit 1. Additionally, after carefully considering adverse comments, we have decided not to finalize our proposed NOX determination of LNB/SOFA + SNCR, but rather are finalizing a NOX BART determination of LNB/SOFA + SCR, with an emission limit of 0.07 lb/ MMBtu (30-day rolling average).

F. Dave Johnston Units 1 and 2 (Reasonable Progress)

We proposed to disapprove the State's determination to not impose LNB/OFA as reasonable progress controls for NOX at Dave Johnston Units 1 and 2. Based on our original cost estimates and visibility modeling, we also proposed to require PacifiCorp Dave Johnston Units 1 and 2 to meet a FIP emission limit of 0.22 lb/MMBtu (30-day rolling average) (assuming the installation of LNB/OFA). Based on our revised cost estimates and visibility modeling that we developed in response to comments, however, we no longer conclude that reasonable progress controls are warranted this planning period. While we continue to disagree with the State's reasoning for not imposing controls (as detailed in our response to comments), we are not prepared to say the State's ultimate decision was unreasonable. In evaluating the four reasonable progress factors and the visibility improvement associated with potential controls, we found that the average and incremental cost-effectiveness of LNB/OFA ($990/ton and $1,221/ton, respectively), while reasonable if viewed in isolation, was not necessarily justified this planning period in light of the relatively modest visibility improvement predicted by the revised modeling (0.11 deciviews—0.12 deciviews at the most impacted Class I area). As a result, we are approving the State's reasonable progress determination of no new controls for Dave Johnston Units 1 and 2, but we expect the State to revisit the issue during the next planning period.

V. Issues Raised by Commenters and EPA's Responses

A. Legal Issues

1. EPA Authority and State Discretion

Comment: Multiple commenters stated that CAA Section 169A and the Regional Haze Rule (RHR) give the states the lead in developing their regional haze SIPs. Some commenters went further in stating that Wyoming is given almost complete discretion in creating its regional haze SIP. These commenters argued that, because Wyoming is given such discretion, EPA lacks the statutory authority to disapprove the State's regional haze SIP. Specifically, some commenters pointed to the flexibility the State is granted in developing its BART determinations and other RHR requirements. The commenters stated that the CAA anticipates that EPA will create guidance and that the states, using their discretion, will use this guidance to develop regional haze SIPs. The State of Wyoming and other parties argued that each factor in the five-factor analysis used to make its BART determinations was appropriately weighed based on the State's own discretion. The commenters therefore argue that EPA has no basis on which to disapprove the five-factor analysis and that EPA does not have authority to reject a state's BART determination solely because EPA would have conducted the analysis in a different way or reached a different conclusion. The commenters went on to say that the State, after considering all statutory factors, made BART determinations for all subject-to-BART sources in a manner consistent with 40 CFR Part 51 Appendix Y, the established CAA requirements, and the interests of the State of Wyoming.

Numerous commenters went on to say that the U.S. Court of Appeals for the D.C. Circuit has affirmed that EPA's role in determining BART is limited and that a state's role is paramount. The court found that the CAA “calls for states to play the lead role in designing and implementing regional haze programs.” Am. Corn Growers Ass'n v. EPA, 291 F.3d 1, 2 (D.C. Cir. 2002). The commenters stated that the court also reversed a portion of EPA's original RHR because it found that EPA's method of analyzing visibility improvements distorted the statutory BART factors and was “inconsistent with the Act's provisions giving the states broad authority over BART determinations.” Id., see also Utility Air Regulatory Group v. EPA, 471 F.3d 1333, 1336 (D.C. Cir. 2006) (The second step in a BART determination “requires states to determine the particular technology that an individual source `subject to BART' must install.”).

The commenters asserted that states have the primary responsibility for preventing air pollution under the CAA. CAA section 101(a)(3), 42 U.S.C. 7401(a)(3). Pursuant to this principle, states, not EPA, have always had primary control over decisions to impose specific emission limits (and therefore specific pollution control technologies) for individual facilities. By congressional design, EPA “is relegated . . . to a secondary role in the process of determining and enforcing the specific, source-by-source emission limitations which are necessary [to meet] national standards.” Train v. NRDC, 421 U.S. 60, 79 (1975). This basic division of responsibilities between EPA and the states remained unchanged when Congress amended the Act in 1977 and again in 1990. See Virginia v. EPA, 108 F.3d 1397, 1408-09 (D.C. Cir. 1997).

Response: Congress crafted the CAA to provide for states to take the lead in developing SIPs, but balanced that decision by requiring EPA to review the SIPs to determine whether they meet the requirements of the CAA. EPA's review of SIPs is not limited to a ministerial type of automatic approval of a state's decisions. See North Dakota v. EPA, 730 F.3d 750, 760-61 (8th Cir. 2013) (“Although the CAA grants states the primary role of determining the appropriate pollution controls within their borders, EPA is left with more than the ministerial task of routinely approving SIP submissions.”) (hereinafter “North Dakota”). EPA must consider not only whether the State considered the appropriate factors, but whether the State acted reasonably in doing so. In undertaking such a review, EPA does not “usurp” the State's authority, but ensures that such authority is reasonably exercised. EPA has the authority to issue a FIP either Start Printed Page 5052when EPA has made a finding that the state has failed to timely submit a SIP or when EPA has found a SIP deficient. Here, EPA has authority on both grounds, and we have approved as much of the Wyoming regional haze SIP as possible, while promulgating a FIP only to fill the remaining gaps. Our action today is consistent with the statute.

Our action does not contradict the Supreme Court's decision in Train. States have significant responsibilities in the implementation of the CAA and meeting the requirements of the RHR. We recognize that states have the primary responsibility of drafting a SIP to address the requirements of the CAA's visibility program. We also recognize that we have the responsibility of ensuring that SIPs, including regional haze SIPs, conform to CAA requirements. We cannot approve a regional haze SIP that fails to address BART with a reasoned consideration of the statutory and regulatory requirements of the CAA and the RHR. See Oklahoma v. EPA, 723 F.3d 1201, 1207 (10th Cir. 2013) (“We agree with the EPA that the statute provides the agency with the power to review Oklahoma's BART determination for these four units.”) (hereinafter “Oklahoma”).

Contrary to the commenters' assertions, we recognize the State's primary responsibility in drafting a SIP. In fact, we have approved many of the State's determinations, including the entirety of Wyoming's Section 309 BART alternative for SO2 emissions. We are disapproving the State's NOX BART determinations, as the CAA requires, because the State neglected to properly consider the costs of compliance and the visibility benefits associated with several of the available control options.

We also disagree that our proposal is inconsistent with the American Corn Growers and Utility Air Regulatory Group decisions. These cases dealt with EPA's authority to issue broad regulations that prescribed how states must conduct their BART determinations. They did not address EPA's authority to review regional haze SIPs for compliance with the mandates of the CAA or EPA's now finalized implementing regulations. The Tenth Circuit, in concluding that EPA had authority to disapprove a BART determination that did not follow the BART Guidelines, stated that the American Corn Growers opinion “does not alter this conclusion.” Oklahoma v. EPA, 723 F.3d 1201, 1208 (10th Cir. 2013).

Because the CAA sets certain mandatory statutory deadlines and provides for citizen suits when the Administrator fails to perform a mandatory duty, we are required by the terms of a consent decree to ensure that Wyoming's CAA requirements for regional haze are finalized by January 10, 2014. Because we have found that the State's regional haze SIP did not satisfy CAA and RHR requirements in full and because we have previously found that Wyoming failed to timely submit its regional haze SIP, we have not only the authority, but a statutory duty to promulgate a FIP that meets those requirements. We have reviewed this decision in light of other decisions made by us, as well as decisions made in other states SIPs. Our action today in large part approves the regional haze SIP submitted by Wyoming. Our disapproval of Wyoming's NOX BART and reasonable progress determinations and imposition of a FIP is not intended to encroach on State authority. Rather, our action today is required by the CAA to ensure that the State has a complete plan in place to address the CAA's visibility requirements.

Comment: The fact that Congress gave states primacy in making BART determinations is noteworthy and related to the fact that the regional haze program is focused on an aesthetic benefit, not a public health standard. Under other sections of the CAA, primarily those dealing with health-based standards, Congress directed EPA to establish standards that do not take costs into consideration. States then develop plans to meet those health-based standards. Under the New Source Performance Standards program (section 111 of the CAA) and National Emission Standards for Hazardous Air Pollutants program (section 112), EPA routinely establishes specific emission limits for large industrial sources. The regional haze program, which deals with an aesthetic standard, was clearly laid out by Congress to be different in its approach, to avoid establishing emission limits, to give states authority to decide appropriate controls, and allow states to weigh the costs against the benefits.

Response: We do not agree with this commenter's characterization of the regional haze program or the CAA's visibility requirements. While it is true that the goal of CAA sections 169A and 169B is to improve visibility in national parks and wilderness areas rather than to prevent adverse human health effects, Congress structured the program so that states' decisions had to be made in the form of SIPs, which EPA has the authority to review for compliance with all CAA requirements. Furthermore, Congress did not create an approach that would allow states to avoid establishing emission limits. On the contrary, Congress specifically directed EPA's regulations to require states to devise “emission limits . . . necessary to make reasonable progress,” CAA section 169A(b)(2), including the requirement to establish BART, which the RHR defines as “an emission limitation.” 40 CFR 51.301.

Comment: EPA's actions leave nothing under the CAA's framework by which Wyoming could make an approvable BART determination. EPA has overreached and exceeded its statutory authority by proposing a FIP that replaces Wyoming's considered judgment with EPA's priorities and policy choices.

Response: We disagree with this comment. EPA is not substituting its judgment for that of the State of Wyoming or issuing a FIP merely to advance priorities and policy choices. Rather, we have determined that Wyoming did not properly follow the BART Guidelines or the CCM in conducting its BART analyses and, therefore, did not correctly consider the costs of compliance or the visibility benefits associated with available control technologies as the CAA requires. Consequently, we are finalizing a FIP in today's action to remedy the gaps left by these inadequacies. We note, however, that the CAA's framework provides Wyoming with the opportunity to submit a SIP revision at any time that could replace all or a portion of EPA's FIP, and we encourage Wyoming to do so.

Comment: EPA clearly gave the states more discretion through rulemaking when it split the universe of BART sources impacted by the BART Guidelines into power plants greater than 750 megawatts (MW) and all others. States were merely encouraged to follow the BART Guidelines for the smaller BART sources. EPA says in the preamble “that states should view the guidelines as helpful guidance for these other categories.” In saying this, EPA is affording even more discretion to the states in making BART determinations for the smaller BART sources. EPA has proposed disapproval of Wyoming's BART determination and proposed a FIP for one of these smaller sources, the Wyodak Unit 1 335 MW power plant. The State believes that the EPA is again overreaching in its action by proposing a FIP for Wyodak Unit 1, where Wyoming was not even required to follow the BART Guidelines in arriving at its BART determination.

Response: We agree that the BART Guidelines are only mandatory for Start Printed Page 5053“fossil-fuel fired power plants having a total generating capacity greater than 750 megawatts.” 40 CFR 51.308(e)(1)(ii)(B). However, the fact that a state may deviate from the procedures in the BART Guidelines when selecting BART for smaller EGUs does not mean that a state has unfettered discretion to act unreasonably or inconsistently with the CAA or the RHR. Ultimately, a state must still adopt the “best available retrofit technology,” CAA section 169A(b)(2)B); 40 CFR 51.308(e)(1)(ii)(A), while reasonably considering the five statutory factors.

The RHR further defines BART to mean “an emission limitation based on the degree of reduction achievable through the application of the best system of continuous emission reduction for each pollutant which is emitted by an existing stationary facility.” 40 CFR 51.301 (emphasis added). We do not interpret this requirement to allow a state to dismiss the best system of continuous emission reduction under the mantle of unlimited state discretion. As we discuss elsewhere in this document, Wyoming erroneously evaluated costs and visibility benefits when analyzing the various control options available for Wyodak, and thereby did not reasonably consider the statutory factors and select the best system of control.

Comment: EPA's RHR gave states the flexibility to choose alternatives to the BART process, such as participation in a trading program. EPA spells out in the preamble that this “substantial flexibility” provides the “states the ability to choose the least costly and least burdensome alternative.” EPA and 28 states on the east coast took advantage of this flexibility when it declared that the cap and trade program for ozone nonattainment would, for the most part, satisfy the requirements of BART. The important point here is that EPA wanted and pushed for flexible, cost-savings approaches to address regional haze. EPA is still pushing for approval of the Cross States Air Pollution Rule (CSAPR) as a solution to regional haze problems on the east coast.

There appears to be a consistency issue within the EPA over the application of flexibility. Wyoming does not think EPA meant for an approach to promote costs savings and less burdensome solutions to be restricted to one area of the country or certain types of solutions. However, EPA's proposal to partially disapprove Wyoming's regional haze SIP and impose more costly and burdensome FIP requirements for seven BART units in the State of Wyoming appear to be inconsistent with EPA's purported “substantial flexibility.” EPA's failure to recognize Wyoming's discretion in these areas is arbitrary and capricious.

Response: Wyoming had the opportunity to submit better-than-BART alternatives in lieu of source-specific NOX BART determinations. Wyoming did not do so. Because Wyoming did not take advantage of the flexibility afforded by better-than-BART alternatives, we must review Wyoming's BART determinations for compliance with the applicable requirements of the CAA, RHR, and BART Guidelines. Our proposal clearly laid out the bases for our proposed disapproval of the State's NOX BART determinations, and we have relied on the standards contained in our regulations and the authority that Congress granted us to review and determine whether Wyoming's regional haze SIP complied with the minimum statutory and regulatory requirements. To the extent a cost analysis relies on values that are inaccurate, a state has not considered cost in a reasoned or reasonable fashion. To the extent a state has considered visibility improvement from potential emissions controls in a way that substantially understates the improvement or does so in a way that is not consistent with the CAA, the state has not considered visibility improvement in a reasoned or reasonable fashion. In these circumstances—as discussed in more detail in the proposed notice and this final notice—EPA is required to disapprove the relevant aspects of the SIP. In determining SIP adequacy, we must exercise our judgment and expertise regarding complex technical issues, and it is entirely appropriate that we do so. Courts have recognized this necessity and deferred to our exercise of discretion when reviewing SIPs. See, e.g., Connecticut Fund for the Env't., Inc. v. EPA, 696 F.2d 169 (2nd Cir. 1982); Michigan Dep't. of Envtl. Quality v. Browner, 230 F.3d 181 (6th Cir. 2000); Mont. Sulphur & Chem. Co. v. EPA, 2012 U.S. App. LEXIS 1056 (9th Cir. Jan. 19, 2012).

Comment: One commenter asserted that the U.S. Court of Appeals for the D.C. Circuit's decision to vacate CSAPR is relevant to the Wyoming FIP. EME Homer City Generation, L.P. v. EPA., 696 F.3d 7 (D.C. Cir. 2012), cert. granted 570 U.S. (June 24, 2013) (No. 12-1182) (CSPAR Decision), and stated that EPA's proposed Wyoming FIP exceeds EPA's statutory authority. The commenter also states that in vacating CSAPR, the D.C. Circuit held that EPA's “FIP-first” approach exceeds EPA's authority because EPA issued a FIP at the same time it determined the emission reduction parameters that the states were supposed to implement. The commenter stated that EPA's theory was that EPA can define the end goals and simultaneously issue federal plans to implement them, upending that process and placing the Federal Government firmly in the driver's seat at both steps.

Other commenters stated that the D.C. Circuit's rejection of the CSAPR rule is irrelevant to EPA's regional haze rulemaking for Wyoming. They asserted that the regional haze program differs from the CAA's good-neighbor provision in fundamental ways that make the court's rejection of CSAPR irrelevant to EPA's action on Wyoming's regional haze plan. The commenters stated that the CAA's visibility provisions establish a technology-based standard for eligible major sources, including PacifiCorp's coal-fired power plants in Wyoming. See 42 U.S.C. 7491(b)(2)(A). To help achieve “reasonable progress” toward the national visibility goal, eligible sources must install BART for haze-causing pollutants. Id. BART is defined as: “an emission limitation based on the degree of reduction achievable through the application of the best system of continuous emission reduction for each pollutant which is emitted by an existing stationary facility.” 40 CFR 51.301. The emission limitation must be established on a case-by-case basis, taking into consideration the technology available, the costs of compliance, the energy and non-air quality environmental impacts of compliance, any pollution control equipment in use or in existence at the source, the remaining useful life of the source, and the degree of improvement in visibility which may reasonably be anticipated to result from the use of such technology. Unlike the D.C. Circuit's interpretation of the good-neighbor provision, the BART definition establishes a floor for emissions reductions, but no ceiling. States must ensure that eligible sources install the best pollution control devices.

These commenters also argued that when a SIP fails to establish a program that meets CAA requirements, then EPA has an obligation to promulgate a FIP. Here, they argued, EPA carried out its statutory duty in proposing a partial FIP for Wyoming. EPA's role is not mere “rubber-stamping” of poor SIPs. EPA “has a duty to evaluate the adequacy of the existing SIP as a whole when approving SIP revisions.” Ass'n of Irritated Residents v. EPA, 632 F.3d 584, 591 (9th Cir. 2011). A FIP “fill[s] all or a portion of a gap or otherwise correct[s] all or a portion of an inadequacy in a State implementation plan.” 42 U.S.C. Start Printed Page 50547602(y) (emphasis added). In proposing to reject many of Wyoming's inadequate BART determinations, and proposing a partial FIP, EPA is merely acting to fulfill its own regulatory obligations under the Act.

Response: With respect to the comment that we lacked authority to promulgate a FIP due to the D.C. Circuit's decision in EME Homer City, we disagree. In EME Homer City, the D.C. Circuit vacated CSAPR, which was promulgated by EPA to address interstate transport of SO2 and NOX under CAA section 110(a)(2)(D). The court found that CSAPR exceeded EPA's authority under section 110 because the rule had the potential to require upwind States to reduce emissions by more than their own significant contributions to downwind nonattainment and because EPA had not given states an opportunity to submit SIPs after EPA had quantified their obligations for emissions reductions.

In the regional haze context, by contrast, EPA defined states' obligations under the RHR and the BART Guidelines well in advance of its findings of failure to submit and subsequent SIP disapprovals. EPA promulgated the original RHR on July 1, 1999 (64 FR 35714). Following the D.C. Circuit's decision in American Corn Growers, EPA revised the RHR and issued the final BART Guidelines on July 6, 2005. (70 FR 39104). The revised RHR and the BART Guidelines were upheld by the D.C. Circuit in Utility Air Regulatory Group v. EPA, 471 F.3d 1333 (D.C. Cir. 2006).[18] As explained in our proposal and elsewhere in this document, the BART Guidelines provide detailed instructions to states on how to determine which sources are subject to BART and how to analyze the five statutory factors in order to set emissions limits representing BART for each subject-to-BART source.[19] In 2006, responding to specific questions from various states and Regional Planning Organizations (RPOs), EPA issued further guidance to help states implement the RHR and BART Guidelines.[20]

As noted in prior responses, EPA issued a finding of failure to submit for regional haze SIPs on January 15, 2009 (74 FR 2392), triggering a FIP clock under CAA section 110(c). By this time, states already had more than three years since issuance of the final BART Guidelines to develop their regional haze SIPs. By the time the FIP clock actually ran out in January 2011, EPA had received regional haze SIPs from nearly every state. EPA has since proposed and approved, in part or in whole, the vast majority of these SIPs.[21 22] This stands in contrast to the situation in EME Homer City, where the court noted that, “every Transport Rule State that submitted a good neighbor SIP for the 2006 24-hour PM2.5 NAAQS was disapproved.” Thus, it is clear that states had ample opportunity to submit approvable regional haze SIPs before EPA was obligated to promulgate regional haze FIPs under CAA section 110(c).

One commenter also pointed to the D.C. Circuit's general statements concerning state and federal roles under the CAA and argues that EPA has exceeded its statutorily mandated role in proposing to disapprove portions of Wyoming's regional haze SIP and promulgate a FIP. While we agree that the general principles concerning state and federal roles under Title I of the CAA apply to our action here, we do not agree that our action is inconsistent with those principles. In this action, we are fulfilling our statutory duty to review Wyoming's regional haze SIP, including its BART determinations, for compliance with the applicable requirements of the CAA and the RHR, and to disapprove any portions of the plan that do not meet those requirements. Based on our review of the SIP, we proposed to determine that certain elements of Wyoming's regional haze SIP did meet the requirements of the CAA and the RHR, and we proposed to approve those elements. However, for the reasons explained in detail in our proposed notices and elsewhere in this document, we have concluded that five of Wyoming's BART determinations [23] and four elements of the regional haze SIP [24] did not comply with the requirements of the CAA and the RHR. Based on these findings, we are required to disapprove these portions of Wyoming's regional haze SIP. As discussed in detail in several below responses, the CAA provides EPA with the authority to review and reject an inadequate regional haze SIP. Oklahoma v. EPA, 723 F.3d 1201, 1207 (10th Cir. 2013).

Comment: One commenter stated that the limits on EPA's authority to reject a SIP were affirmed by the Fifth Circuit in Texas v. EPA, 690 F.3d 670 (5th Cir. 2012), vacating EPA's rejection of a Texas SIP revision implementing its minor new source review program (i.e., the Texas Flexible Permit Program). In the Texas decision, the court reaffirmed the principle that if a SIP or SIP revision meets the statutory criteria of the CAA, then EPA must approve it. The Wyoming regional haze SIP meets the statutory criteria of the CAA. Therefore, EPA's disapproval of the Wyoming regional haze SIP exceeds EPA's statutory authority.

Response: In Texas, the Fifth Circuit found that EPA had failed to tie its disapproval to any specific requirement in the CAA or EPA's implementing regulations.[25] In this action, our disapproval is based explicitly and squarely on the SIP's failure to comply with the CAA section 169A(b)(2)(A), as implemented through the RHR and the Start Printed Page 5055BART Guidelines. Just because a court found EPA's disapproval invalid in one case does not mean that finding applies in all cases. This situation involves a very different program under the CAA and a very different state submittal and review. The Texas case does not involve BART or the CAA's regional haze provisions at all. Rather, it involved EPA's disapproval of SIP revisions involving Texas's minor new source review program. There are a limited number of specific requirements in EPA rules for minor source review programs. In contrast, regional haze SIPs and BART determinations are subject to the detailed requirements set forth in CAA section 169A, the RHR, and the BART Guidelines.

Comment: One commenter stated that the CSAPR decision criticized the CSAPR's FIP-first approach because it forces states to “take a stab in the dark” on their compliance obligations only to be judged later whether they hit the mark. As the D.C. Circuit explained in the CSAPR decision, a “SIP logically cannot be deemed to lack a required submission or deemed to be deficient for failure to meet . . . [an] obligation before EPA quantifies the . . . obligation.” EME Homer City Generation, L.P. v. EPA., 696 F.3d 7, 49 (D.C. Cir. 2012), cert. granted 570 U.S. (June 24, 2013) (No. 12-1182) (hereinafter “CSPAR Decision”).

Other commenters reject this assertion, explaining that Wyoming was not forced to take a “stab in the dark” in developing its regional haze SIP. In EME Homer City, the D.C. Circuit accepted the state petitioners' argument that they had no obligation to submit SIPs until after EPA defined each state`s contribution to interstate pollution and the necessary emissions reductions to address that contribution. EME Homer City, 2012 WL 3570721, at *18 (“[L]ogically, a SIP cannot be deemed to lack a required submission . . . until after EPA has defined the State`s good neighbor obligation.”; “There is no way for an upwind State to know its obligation . . . until EPA defines it.”).

Response: We do not agree that Wyoming was forced to take a “stab in the dark” in developing its regional haze SIP. The regional haze program and the interstate transport obligations under the CAA are quite different. The states' regional haze obligations have been clearly defined. EPA issued BART Guidelines establishing detailed parameters for state BART determinations in 2005. Commenter's charge that EPA may never issue a FIP in such circumstances is incorrect. We explain in detail above how the CAA's visibility provisions and EPA's implementing regulations differ from the good-neighbor provision at issue in EME Homer City. Wyoming was well aware of these requirements as it developed its regional haze SIP, through EPA comment letters and meetings between EPA and the State. Finally, unlike the petitioners in EME Homer City, none of the commenters here dispute that Wyoming's regional haze SIP and BART determinations were “required submission[s].”

Comment: One commenter stated that the CSAPR decision also made clear that any FIP issued by EPA must be related to the “end goal of the statute.” The D.C. Circuit stated in the CSAPR decision: “[T]he end goal of the statute is attainment in the downwind state. EPA's authority to force reductions on upwind states ends at the point where the affected downwind State achieves attainment.” CSAPR Decision at p. 25.

The “end goal” of the regional haze statutory requirements is to gradually achieve “natural visibility” conditions by the year 2064 under an emission reduction approach known as reasonable progress as determined by the states. EPA's rush in the proposed Wyoming FIP to front-load as many emission reductions as possible in the first five years of this decades-long program is a clear indication that EPA has lost sight of the “end goal” of the regional haze program. Likewise, EPA's failure to account for, and properly address, other causes of visibility impairment in its FIP, such as natural causes (forest fires), out of state sources, oil and gas sources, etc., demonstrates that EPA has lost focus on the “end goal” of the regional haze program. EPA's proposed Wyoming FIP violates this “end goal” principle espoused by the CSAPR decision.

Response: EPA is required to evaluate BART factors included in state SIPs (e.g., ultimately rejecting methodological flaws and data flaws in estimating costs of compliance and visibility, as we have done in this final action), where the flaws in the analysis prevented the State of Wyoming from conducting meaningful consideration of the BART factors, as required by the BART Guidelines, and moored to the CAA's BART and SIP provisions. North Dakota v. EPA, 730 F.3d 750, 761 (8th Cir. 2013).

Furthermore, we do not agree that one provision of the CAA should be read and applied in isolation. The commenter's position would ignore the rest of the CAA's statutory requirements and violate the “fundamental canon of statutory construction that the words of a statute must be read in their context and with a view to their place in the overall statutory scheme.” A court must therefore interpret the statute “as a symmetrical and coherent regulatory scheme,” and “fit, if possible, all parts into an harmonious whole.” FDA v. Brown & Williamson Tobacco Corp., 529 U.S. 120, 133 (2000) (quoting Davis v. Michigan Depart of Treasury, 489 U.S. 803, 809 (1989); Gustafson v. Alloyd Co., 513 U.S. 561, 569 (1995); and FTC v. Mandel Brothers, Inc., 359 U.S. 385, 389 (1959)). The commenter's claim that one provision in the CAA overrides all other statutory provisions is unfounded and not supported by the CAA. In particular, the statutory requirements for BART are separate and distinct from the statutory requirements for reasonable progress.

Moreover, as explained elsewhere in this document, EPA's action fully accounts for other causes of visibility impairment. With respect to wildfires, we explain in detail elsewhere in this document the role that fires play in determining natural background conditions. With respect to oil and gas sources, we are approving the State's determination to not impose controls on this source category during this planning period, in part because the State already applies minor source BACT to many of them through the State's SIP-approved minor NSR program, and in part because controls on these sources are not so cost-effective that we are prepared to say the State was unreasonable. With respect to accounting for out-of-state sources, we cited sources outside the Western Regional Air Partnership (WRAP) domain as one factor that made it reasonable for our RPGs to fall short of the uniform rate of progress (URP) and unreasonable to achieve the URP. Finally, we note that we are approving some of Wyoming's BART determinations and all of Wyoming's reasonable progress determinations. Additionally, BART is required in the first planning period, which ends in 2018, and is required to be installed as expeditiously as practicable, but in no event later than five years after the effective date of this final notice. In light of the fact that many of Wyoming's Class I areas are not even expected to meet the URP this planning period, the notion that EPA has required “front-loading” of controls is utterly without merit.

Comment: Some commenters stated that the CSAPR decision considered, and then rejected, a “reasonableness” standard put forth by EPA as the only limit on its authority to impose emission reductions under the CSAPR. CSAPR Decision at p. 37, ftnt. 23. EPA likewise purports to impose a Start Printed Page 5056reasonableness standard as adequate justification for rejecting the Wyoming regional haze SIP and imposing a FIP. The CSAPR decision makes clear that such a reasonableness standard, not included in the CAA itself, does not have a place in justifying EPA's actions in issuing a FIP. For this added reason, the CSAPR decision makes clear the FIP exceeds EPA's statutory authority.

Response: EPA disagrees with this comment. First, the commenters misunderstand the cited footnote in the CSAPR decision. In the D.C. Circuit's view, EPA ignored statutory limits on its authority and instead claimed that reasonableness was the only bound on EPA's authority. Here, EPA makes no such claim. EPA, of course, has the authority and the duty to review Wyoming's SIP for compliance with the CAA and the RHR.

In reviewing the Wyoming regional haze SIP, EPA has determined that a “reasonableness” standard is in fact harmonious with the CAA and the RHR, and the courts have agreed. Oklahoma v. EPA, 723 F.3d 1201, 1207 (10th Cir. 2013) (“The EPA therefore had a reasonable basis for rejecting the 2008 Cost Estimates [that were based on the overnight costing method] as not complying with the guidelines.”); see also North Dakota v. EPA, 730 F.3d 750, 761 (8th Cir. 2013) (explaining EPA is not required to “approve a BART determination that is based upon an analysis that is neither reasoned nor moored to the CAA's provisions”).

The CAA requires states to submit SIPs that contain such measures as may be necessary to make reasonable progress toward achieving natural visibility conditions, including BART. The CAA accordingly requires the states to submit a regional haze SIP that includes BART as one necessary measure for achieving natural visibility conditions. See Alaska Dep't of Envtl. Conservation v. EPA, 540 U.S. 461, 500 (2004) (in a related context, holding that EPA validly issued stop work orders because the state's BACT determination “simply did not qualify as reasonable in light of the statutory guides.” (emphasis added)) (hereinafter “ADEC”). Thus we are not establishing a new reasonableness standard, as the commenter asserts.

Comment: In the CSPAR decision, the D.C. Circuit found it “inconceivable” that Congress would bury in the CAA “an open-ended authorization for EPA to effectively force every power plant in the upwind States to install every emissions control technology EPA deems `cost effective'.” CSAPR Decision at p. 40. In so finding, the court refused to transform a “narrow” provision into a “broad and unusual” authority that would overtake other core provisions of the Act.” Id. Similarly, it is inconceivable in the regional haze context that Congress would bury an open-ended authorization allowing EPA to ignore its own BART Guidelines, overrun carefully crafted state regional haze SIPs and BART determinations, and require the installation of expensive emission controls which result in minimal regional haze improvements. This principle espoused in the CSAPR decision is particularly applicable in the regional haze context where, just like in the CSAPR, EPA's BART determinations in the Wyoming FIP are “not a clear numerical target—far from it—until EPA defines the target.” CSAPR Decision at p. 48. And in spite of EPA initially helping to define “the target” by issuing its BART Guidelines (which EPA subsequently ignored), EPA did not begin to redefine the target until it began to issue various determinations around the country in reaction to various state regional haze SIPs. Even then, EPA's “target” is not clear and certainly is impossible to determine, on a state or source-by-source basis, until EPA sets the target in a state-specific FIP.

Like the upwind states in the CSAPR decision, it was “impossible” for Wyoming to determine its regional haze obligation “until EPA defined it.” Id. This process effectively allows EPA to impose any standard it wants with little ability for the states (or sources) to achieve the redefined target through a state-led process because of the tight deadlines imposed by EPA as a result of negotiated consent decree deadlines.

Response: We do not agree that we have ignored the CAA and BART Guidelines. As explained in our proposed notice and elsewhere in this document, our decisions are firmly grounded on the CAA provisions and BART Guidelines, and Wyoming was well aware of these requirements as it developed its SIP. In addition, the comparison of BART determinations and the CSAPR decision is not appropriate. In contrast to CSPAR, the CAA and RHR do not set specific numerical targets for BART determinations. Instead, they require states to reasonably consider the five statutory factors, which, as we have detailed in our proposal and in our response to comments, Wyoming did not do. Furthermore, EPA provided extensive comments to the State on the proposed regional haze SIP and met with the State on numerous occasions, so the State was aware of EPA's concerns regarding approvability before the SIP was submitted to EPA. As explained below in greater detail, the Consent Decree that covers this action has not hindered Wyoming's ability to develop and submit an approvable SIP. Wyoming can submit new SIP revisions, and request that EPA review and approve them, to replace the FIP elements at any time.

To the extent that the comment argues that the RHR itself is invalid for similar reasons to those for which the D.C. Circuit vacated CSAPR, the time to make those arguments has passed.

Finally, in establishing the BART requirements, Congress was addressing a category of large sources that pre-dated the modern NSR affected sources, which were determined to significantly contribute to regional haze and set an expectation that included consideration of cost, feasibility, and effect on regional haze (as well as the other five factors) for those sources, many of which did not have modern pollution controls because of their age and because they hadn't been addressed through ozone SIPs the way so many eastern sources had. This is one of the reasons why the western regional haze SIPs are seeing emission controls.

Comment: One of the commenters stated that one of the key conclusions of the CSAPR decision was that EPA exceeded the scope of its authority by requiring emission reductions beyond the statutory or regulatory requirements. In the CSAPR decision, the court looked at the fact that once EPA had determined that an upwind emission source contributed “significantly” to nonattainment or maintenance of the standard in a downwind state; it was “in” for purposes of requiring emission reductions. The emission reduction requirements were then based on cost-effectiveness thresholds that were applied uniformly throughout the CSAPR region. In other words, all emissions that could be reduced, for example, for a cost between $1 and $500 per ton were effectively required. The court held that this approach resulted in a situation where some sources had to bear a disproportionate amount of costs, based on their relative contribution to the nonattainment or maintenance problem.

Similarly in the regional haze context, EPA established an “in or out” criteria of a 0.5 deciview impact. Sources with modeling results that suggested the impact was greater than 0.5 were “in” and required further analysis. If, under EPA's FIP approach, the facilities could cost-effectively (as determined by EPA, not the states) control emissions, they were required to do so. Oftentimes, EPA has required the controls notwithstanding the negligible contribution the emission reductions Start Printed Page 5057will have towards meeting the requirement of the RHR. EPA's conclusions requiring individual sources to reduce emissions under its subjective cost-effectiveness criteria have no relationship to visibility impacts or improvements, and EPA failing to conduct that modeling, but supporting a determination of reasonableness of controls “based on the high cost effectiveness at each of the units.” 77 FR at 33034, 33038 and 33055.

EPA's conclusions regarding emission reductions that are based on the cost-effectiveness of controls without an appropriate linkage to visibility improvement and meeting the goals and objectives of the RHR exceed EPA's statutory authority as suggested by the CSAPR decision.

Response: We agree with some of this comment and disagree with other portions. As an initial matter, as we explained in our proposed notice, we note that:

Wyoming used a contribution threshold of 0.5 deciviews for determining which sources are subject-to-BART. By using a contribution threshold of 0.5 deciviews, Wyoming exempted seven of the fourteen BART-eligible sources in the State from further review under the BART requirements. Based on the modeling results, the State determined that P4 Production, FMC Granger, and OCI Wyoming had an impact of .07 deciview, 0.39 deciview, and 0.07 deciview, respectively, at Bridger Wilderness. Black Hills Neil Simpson 1, Sinclair Casper Refinery, and Sinclair—Sinclair Refinery have an impact of 0.27 deciview, 0.06 deciview, and 0.12 deciview, respectively, at Wind Cave. Dyno-Nobel had an impact of 0.22 deciview at Rocky Mountain National Park. These sources' modeled visibility impacts fell below the State's threshold of 0.5 deciview and were determined not to be subject-to-BART. 78 FR 34747

Since the State's approach is consistent with the BART Guidelines [26] and given the relatively limited impact on visibility from these seven sources, as explained earlier in this document and in our proposals, we are finalizing our proposal to approve Wyoming's threshold of 0.5 deciviews as reasonable for determining whether its BART-eligible sources are subject-to-BART. 78 FR 34734, 34747

We do not agree that our decision exceeds our statutory authority and the goals and objectives of the RHR. CAA section 110(a)(2)(J) requires each plan submitted by a state to “meet the applicable requirements” of Part C of Title I of the CAA, including those for “visibility protection.” In the case of a regional haze SIP submittal, the “applicable requirements” include the requirement that each source found subject-to-BART, “procure, install, and operate, as expeditiously as practicable (and maintain thereafter) the best available retrofit technology . . .” [27] Section 169A(g)(2) further provides that:

In determining best available retrofit technology the State (or the Administrator in determining emission limitations which reflect such technology) shall take into consideration the costs of compliance, the energy and non-air quality environmental impacts of compliance, any existing pollution control technology in use at the source, the remaining useful life of the source, and the degree of improvement in visibility which may reasonably be anticipated to result from the use of such technology.[28]

Similarly, the RHR provides that:

The determination of BART must be based on an analysis of the best system of continuous emission control technology available and associated emission reductions achievable for each BART-eligible source that is subject to BART within the State. In this analysis, the State must take into consideration the technology available, the costs of compliance, the energy and non-air quality environmental impacts of compliance, any pollution control equipment in use at the source, the remaining useful life of the source, and the degree of improvement in visibility which may reasonably be anticipated to result from the use of such technology.[29]

Wyoming's BART determinations for NOX at five BART units fall short of these requirements in several respects.

First, Wyoming did not analyze the “best system of continuous emission control technology available and associated emission reductions achievable.” This is explained in detail in our proposed rulemaking, the docket for this action, and elsewhere in this document. Therefore, Wyoming has not demonstrated that its BART determinations were “based on an analysis of the best system of continuous emission control technology available and associated emission reductions achievable.”

For example, as we explained in our proposed notices and elsewhere in this final action, Wyoming did not appropriately consider the “degree of improvement in visibility which may reasonably be anticipated” from installation of BART because it did not provide visibility improvement modeling from which the benefits of individual NOX controls could be ascertained. Thus Wyoming's BART determinations for NOX do not meet the requirements of CAA section 169A(g)(2) or 40 CFR 51.308(e)(1)(ii)(A).

Additionally, as explained in our proposed notices and elsewhere in the modeling section of this final action, it was not possible to ascertain the visibility improvement from the NOX control options as the State modeled emission reductions for multiple pollutants together. For this reason, in the modeling conducted by EPA, we held SO2 and PM emission rates constant (reflecting the “committed controls” for those pollutants identified by Wyoming), and varied only the NOX emission rate. This allowed us to isolate the degree of visibility improvement attributable to the NOX control option.

In addition, 40 CFR 51.308(e)(1)(ii)(B) provides that the determination of BART for fossil-fuel fired power plants having a total generating capacity greater than 750 megawatts must be made pursuant to the guidelines in appendix Y of part 51 (Guidelines for BART Determinations under the Regional Haze Rule).

All of the Wyoming BART sources, except Wyodak, each have a generating capacity greater than 750 megawatts. Therefore, the BART determinations for these BART sources must be made pursuant to the BART Guidelines. However, Wyoming's BART determinations for these sources did not fully comply with the BART Guidelines. In particular, as explained more fully elsewhere in this document, contrary to the Guidelines' admonition that “cost estimates should be based on the CCM, where possible,” the control cost calculations supplied by the utilities and relied upon by Wyoming included costs not allowed by the CCM, such as owner's costs and Allowance for Funds Utilized During Construction (AFUDC). Thus, Wyoming's consideration of the “cost of compliance” for these units was not consistent with the Guidelines. Furthermore, as explained elsewhere in this document, Wyoming's consideration of visibility benefits was inconsistent with the Guidelines because the State did not provide visibility modeling from which the visibility improvement from individual controls could be ascertained. Finally, for all pollutants at all units covered by today's action, Wyoming's regional haze SIP does not meet the requirements of 40 CFR 51.308(e)(1)(iv) and (v) because it lacks the following elements:

  • A requirement that each source subject to BART be required to install and operate BART as expeditiously as practicable, but in no event later than 5 years after approval of the implementation plan revision.Start Printed Page 5058
  • A requirement that each source subject to BART maintain the control equipment required by this subpart and establish procedures to ensure such equipment is properly operated and maintained.

These two requirements are mandatory elements of the RHR and are necessary to ensure that BART is procured, installed, and operated as expeditiously as practicable and maintained thereafter, as required under CAA section 169A(b)(2)(A).

Moreover, the CAA and regional haze rule require that SIPs contain provisions that make emissions limits, including BART limits, practically enforceable. CAA section 110(a)(2)(A)-(B) require that emissions limits such as BART be “practically enforceable” and SIPs provide for establishment, methods and procedures necessary to monitor, compile, and analyze data. CAA section 302(k) requires emissions limits to be met on a continuous basis. Additionally, CAA section 169A(b)(2) requires that regional haze SIPs include “such emission limits, schedules of compliance and other reasonable measures” necessary to meet the goals of the regional haze program.” As discussed in our proposed notices and elsewhere in this final notice, Wyoming's regional haze SIP lacks requirements for monitoring, recordkeeping, and reporting sufficient to ensure that the BART limits are enforceable and are met on a continuous basis.

Therefore, Wyoming's BART determinations for these five units covered by the FIP do not meet the BART requirements of the CAA, the RHR and the BART Guidelines. Additionally, Wyoming's SIP requirements do not ensure the BART limits are enforceable for all BART sources for which there is a SIP or FIP emissions limit, and therefore do not meet the requirements of the CAA and RHR. Accordingly, we are compelled to partially approve and partially disapprove Wyoming's regional haze SIP.

Comment: EPA cannot invoke its Section 110 SIP approval authority as grounds for rejecting state BART determinations with which it disagrees. The CAA does not require any specific degree of visibility improvement in the determination and only requires BART for the purpose of eliminating or reducing impairment to visibility. See CAA Section 169A, 42 U.S.C. 7491.

Consistent with the long-recognized principle that EPA may not “condition approval of the plan of any State, on the State's adoption of a specific control measure,” Virginia, 108 F.3d at 1408, EPA has no statutory authority to disapprove a SIP that contains a BART determination for an individual facility that complies with the statutory BART factors. Any other result would allow EPA to employ its generalized SIP approval authority to “run roughshod over the procedural prerogatives that the Act has reserved to the States.” Bethlehem Steel Corp., 742 F.2d at 1036.

The fact that states must propose SIP revisions “as may be necessary” to achieve reasonable progress does not mean EPA has authority to countermand the textual commitment of specific BART decisions to the states. The D.C. Circuit interpreted similar language in Section 110(k)(5) to constrain EPA's authority over SIP approval and disapproval. See Virginia, 108 F.3d at 1409. The SIP call provisions of Section 110(k)(5) state that when a SIP is inadequate “the Administrator shall require the State to revise the plan as necessary to correct such inadequacies.” But the Virginia court rejected the agency's expansive view of this phrase as authority to impose specific control measures for specific emission sources.

Response: States are required by the CAA to address the BART requirements in their SIP. Our disapproval of the NOX BART determinations in the Wyoming regional haze SIP is authorized under the CAA because the State's NOX BART determinations for the five units do not satisfy the statutory criteria. The State's analysis of the cost effectiveness of controls and visibility analyses were flawed due to reasons discussed elsewhere in the proposed and final notices. While states have authority to exercise different choices in determining BART, the determinations must be reasonably supported. Wyoming's errors in taking into consideration the costs of compliance were significant enough that we cannot conclude the State determined BART according to CAA standards. The cases cited by the commenters stress important limits on EPA authority in reviewing SIP submissions, but our disapproval of these NOX BART determinations for the five units has an appropriate basis in our CAA authority. We did not require Wyoming to adopt specific control measures for specific emission sources. Instead, we disapproved some of Wyoming's BART determinations for reasons described in detail in our proposal and elsewhere in our response to comments. To promulgate our FIP, EPA then had both the authority and the duty to determine specific control measures for specific sources.

Finally, contrary to the commenter's assertion, the Bethlehem Steel case is inapplicable here. We are promulgating BART emission limitations and other FIP elements described elsewhere in this document under the authority of CAA section 110(c), not through our action on Wyoming's SIP. We have authority to promulgate our FIP under 110(c) on two separate grounds: first, based on our January 2009 finding of failure to submit the regional haze plan elements required by 40 CFR 51.309(g), the reasonable progress requirements for areas other than the 16 Class I areas covered by the Grand Canyon Visibility Transport Commission Report; and second, based on our partial disapproval of the regional haze SIP.

Comment: We received comments that EPA does not have the authority under the CAA to issue a regional haze FIP in this instance. Commenters contend that EPA's role under Section 110 in reviewing states' regional haze SIPs is narrow and that the CAA confines EPA to the ministerial function of reviewing SIPs for consistency with the CAA's requirements. Commenters assert that Wyoming submitted a regional haze SIP that met the requirements of Section 51.309 and included all the required elements and that EPA admits that Wyoming has considered all five BART factors. Therefore, commenters go on to say that EPA's sole function was to review whether Wyoming followed the regional haze requirements, including Appendix Y, in preparing the Wyoming regional haze SIP, and Congress did not authorize EPA to “second guess” Wyoming's BART decision making, or to substitute its own judgment, simply because EPA would prefer different BART and reasonable progress NOX controls. Commenters go on to point out that courts have consistently held that states are primarily responsible for SIP development; EPA's role is ministerial. Commenters cite that the Supreme Court has recognized the states' primary role in developing SIPs, holding “so long as the ultimate effect of a State's choice of emission limitations is in compliance with the national standards for ambient air, the State is at liberty to adopt whatever mix of emission limitations it deems best suited to its particular situation.” Train v. NRDC, 421 U.S. 60, 79 (1975). Commenters argue that EPA is going beyond its ministerial function of reviewing Wyoming's regional haze SIP for consistency with the CAA's requirements; it is attempting to design Wyoming's SIP by establishing new NOX emission limits, contrary to its promulgated BART regulations. Commenters go on to say that EPA Start Printed Page 5059should follow the structure of the CAA and give deference to the State's judgment in determining BART in Wyoming's regional haze SIP.

Response: States are required by the CAA to address the BART requirements in their SIP. Our disapproval of the NOX BART determinations in the Wyoming regional haze SIP is authorized under the CAA because the State's NOX BART determinations for the five units do not satisfy the statutory criteria. The State's analyses of the cost effectiveness of controls and visibility analyses were flawed due to reasons discussed in the introduction and BART sections of this document. While states have the authority to exercise different choices in determining BART, the determinations must be reasonably supported. Wyoming's errors in taking into consideration the costs of compliance and visibility analyses were significant enough that we cannot conclude the State determined BART according to CAA standards. The cases cited by the commenters stress important limits on EPA authority in reviewing SIP submissions, but our disapproval of these NOX BART determinations for the five units has an appropriate basis in our CAA authority.

Comment: Under the CAA, both the federal government and the states have responsibilities for maintaining and improving air quality. The federal government has the authority to set specific emissions targets, but the states have the authority to develop and impose their own regulatory structure to meet those. As long as the State meets its specific criteria, which Wyoming can and will show that it has done, the fact that EPA does not share the State's opinion regarding the best course of action is immaterial.

This reading of the CAA is the opinion of the Congress that passed the regional haze program in 1977. Committee and floor debate in Congress at the time makes clear that Congress fully intended for the states to possess a high degree of primacy in regional haze decisions. The primary sponsor of the CAA and 1977 amendments in the Senate was the late Senator Edmund Muskie, a Democrat from Maine. In his opening address to the Senate on the Conference Report to the 1977 amendments, Senator Muskie said, “under this legislation, the administrator of the EPA will be more reliant on local and state capabilities to create the institutional and infrastructural changes necessary to achieve clean air. And perhaps this is as it should be. We have learned that there is little political support for inartfully conceived national measures. We have learned that where change can be made, it must be made with the full understanding and support of the people who are affected by that change.”

While the courts in some instances may not give adequate weight to the intent of Congress in drafting legislation, Congress's intent in passing the nation's law is something that Congress itself takes very seriously. Some courts have honored Congressional intent and upheld the CAA as a cooperative statute. In Appalachia Power Company v. EPA [sic], the courts determined that the CAA includes a cooperative standard they call a federalism bar. In Train and Luminant Generation Co., LLC v. EPA, 675 F.3d 917 (5th Cir. 2012) (hereinafter “Luminant”), the courts held that the EPA had no authority to overturn the decisions of the states so long as the basic requirements of Section 110 are met.

EPA does not have the authority under the CAA to issue a regional haze FIP in this instance. EPA contends its review of the Wyoming SIP is “pursuant to section 110 of the CAA.” 78 FR 34738. Section 110(a)(2) provides the general requirements that a SIP must contain. Importantly, EPA's role under Section 110 in reviewing states' regional haze SIPs is narrow: “With regard to implementation, the (CAA) confines the EPA to the ministerial function of reviewing SIPs for consistency with the (CAA)'s requirements .” Luminant Generation Co., LLC v. EPA, 675 F.3d 917, 921 (5th Cir. 2012) (citing section 110(k)(3)). As the court in Luminant explained, if the state's submissions “satisfy those basic requirements (found in section 110), the EPA must approve them,” and “(t)hat is the full extent of the EPA's authority in the SIP-approval process because that is all the authority that the CAA confers.” Id. at 932. Here, Wyoming submitted a regional haze SIP that met the requirements of Section 309 and included all the required elements. The Wyoming SIP submittals are well developed and comprehensive. EPA admits that Wyoming considered all five BART factors. 78 FR 34748. Therefore, EPA's role was to review whether Wyoming followed the regional haze requirements, including Appendix Y, and provided factual support for the Wyoming regional haze SIP. Congress did not authorize EPA to “second guess” Wyoming's BART decision making, or to substitute its own judgment, simply because EPA would prefer different BART and reasonable progress NOX controls.

More recently, the D.C. Court vacated the CSAPR. The court's 2012 opinion in the CSAPR case is illustrative for our purposes because the EPA used very similar arguments to justify their authority in CSAPR as they're using today for regional haze. In vacating the CSAPR rule, the D.C. Circuit Court writes “under the CAA, the federal government sets air quality standards, but states retain the primary responsibility for choosing how to attain those standards within their borders. The Act thus leaves it to the individual states to determine, in the first instance, the particular restrictions that will be imposed on particular emitters within their borders.” The court goes on to write that “. . .the statutory federalism bar prohibits the EPA from using the SIP process to force states to adopt specific control measures.”

Response: We responded to similar comments above.[30] With respect to EPA's supposed admission that Wyoming considered the five BART factors, the precise language in the proposal notice is: “We find that Wyoming considered all five steps above in its BART determinations, but we propose to find that its consideration of the costs of compliance and visibility improvement for the EGUs was inadequate and did not properly follow the requirements in the BART Guidelines and statutory requirements, as explained below.” 78 FR 34748. With respect to the legislative history quoted, the comment does not provide any connection between the general remarks of Senator Muskie regarding the 1977 Amendments and EPA's interpretation of the visibility provisions in the Act.

Comment: We received numerous general comments that EPA has overstepped its authority and that states have the responsibility of determining what controls are necessary for regional haze.

Response: As explained earlier, the states have the responsibility to draft the regional haze SIP and EPA has the responsibility of ensuring state plans, including regional haze SIPs, conform to the CAA. As the drafter of the regional haze SIP, the State generally has the authority to decide how each of the BART factors are taken into account and weighed. EPA is not disapproving Wyoming's BART determinations because we disagree with how Wyoming weighed the relevant factors, such as the cost of controls or the degree of visibility improvement resulting from Start Printed Page 5060the use of controls. EPA is disapproving certain Wyoming BART determinations because the State did not consider these factors in its BART determinations in accordance with the RHR and the Act.

Comment: EPA's regional haze FIP failed to afford the required deference to the technical, policy and other discretion granted to Wyoming under the CAA and regional haze program. Congress added section 169A to the CAA in order to address the “impairment of visibility” in Class I areas that “results from man-made air pollution.” This provision of the CAA, in turn, describes separate roles for EPA, the states, and major sources such as PacifiCorp's BART Units.

EPA's roles are to create a report, see CAA section 169A(a)(2)-(3), create regional haze regulations, see CAA section 169A(a)(4), provide guidelines for the states, see CAA section 169A(b)(1), and determine whether regional haze SIPs submitted by the states follow the regulations and guidelines, and contain the required elements. CAA section 110. The states' roles, which are central to the regional haze program, are intended to be accomplished using substantial discretion which, in turn, requires significant deference from EPA. States are required to submit a regional haze SIP that contains “emission limits, schedules of compliance and other measures as may be necessary to make reasonable progress toward meeting the national goal.” CAA section 169A(b)(2). States also must “determine[*thnsp;]” BART for “each major stationary source.” CAA 169A(b)(2)(A). BART sources, such as PacifiCorp's BART units, are required to “procure, install, and operate (BART) as expeditiously as practicable.” CAA section 169A(b)(2)(A).

Thus, the CAA mandates that states have the primary role in developing regional haze SIPs to protect visibility in Class I areas. Likewise, the RHR makes clear that states have the responsibility to create and implement regional haze SIPs. In contrast, EPA's role is to develop “guidelines” for the states to use in implementing regional haze SIPs and to determine whether states followed those guidelines. CAA section 169A(b)(1). In short, the CAA anticipates that states, using their discretion, develop regional haze SIPs using EPA guidelines. This is exactly what Wyoming did in issuing BART permits and developing the Wyoming regional haze SIP.

In issuing regional haze guidelines, EPA recognized the broad discretion granted to the states by the CAA. Specifically, EPA adopted guidance to address BART determinations for certain large electrical generating facilities, referred to as “Appendix Y.” EPA created further guidance in the Federal Register responding to comments concerning the then-proposed Appendix Y, referred to as the “Preamble.” EPA recognized in the Preamble that “how states make BART determinations or how they determine which sources are subject to BART” are among the issues “where the Act and legislative history indicate that Congress evinced a special concern with insuring that states would be the decision makers.” 70 FR 39104, 39137 (July 6, 2005).

Likewise, in analyzing the applicability of certain executive orders, EPA stated that “ultimately states will determine the sources subject to BART and the appropriate level of control for such sources” and that “states will accordingly exercise substantial intervening discretion in implementing the final rule.” Id. at 39155. The U.S. Court of Appeals for the D.C. Circuit has affirmed that EPA's role regarding regional haze programs is limited and that a state's role is paramount. Indeed, the Court found that the CAA “calls for states to play the lead role in designing and implementing regional haze programs.” American Corn Growers Ass'n v. E.P.A., 291 F.3d 1, 2 (D.C. Cir. 2002). The court also reversed a portion of EPA's original RHR because it found that EPA's method of analyzing visibility improvements distorted the statutory BART factors and was “inconsistent with the Act's provisions giving the states broad authority over BART determinations.” Id. at 8; (see also Utility Air Regulatory Group v. EPA, 471 F.3d 1333, 1336 (D.C. Cir. 2006) (The second step in a BART determination “requires states to determine the particular technology that an individual source `subject to BART' must install.”)). The court in American Corn Growers emphasized that Congress specifically entrusted states with making BART five-factor analysis decisions: “[t]o treat one of the five statutory factors in such a dramatically different fashion distorts the judgment Congress directed the states to make for each BART-eligible source.” American Corn Growers, 291 F.3d at 6.

The court in American Corn Growers also outlined the relevant legislative history that recounts a specific agreement reached in Congress which granted this authority to the states: “The `agreement' to which the Conference Report refers was an agreement to reject the House bill's provisions giving EPA the power to determine whether a source contributes to visibility impairment and, if so, what BART controls should be applied to that source. Pursuant to the agreement, language was inserted to make it clear that the states—not EPA—would make these BART determinations. The Conference Report thus confirms that Congress intended the states to decide which sources impair visibility and what BART controls should apply to those sources. The RHR attempts to deprive the states of some of this statutory authority, in contravention of the Act.” Id. at 8. EPA's FIP action makes the same mistake and, if finalized, will be similarly reversible.

In sum, based on the language in the CAA, the RHR, EPA's own guidelines, and case law, the states have significant discretion when creating regional haze SIPs.

Response: We responded to similar comments above and elsewhere in this document.

Comment: EPA failed to properly account for that discretion in analyzing the Wyoming regional haze SIP. EPA should have acknowledged that the Wyoming regional haze SIP followed the law and was supported by the facts. Examples of EPA ignoring Wyoming's discretion include: Visibility improvement; cost effectiveness analysis; modeling; application of the five BART factors; and reasonable progress analyses.

Response: We responded to similar comments above and elsewhere in this document.

Comment: EPA's proposed action ignores the congressional commitment to have local decisions under the CAA—particularly those relating to BART—made by the states. States have the primary responsibility for preventing air pollution under the CAA. CAA section 101(a)(3), 42 U.S.C. 7401(a)(3). Pursuant to this principle, states, not EPA, have always had primary control over decisions to impose specific emission limits (and therefore specific pollution control technologies) for individual facilities. By congressional design, under the CAA EPA “is relegated . . . to a secondary role in the process of determining and enforcing the specific, source-by-source emission limitations which are necessary [to meet] national standards.” Train v. NRDC, 421 U.S. 60, 79 (1975) (hereinafter “Train”). This basic division of responsibilities between EPA and the States remained unchanged when Congress amended the Act in 1977 and again in 1990. See Virginia v. EPA, 108 F.3d 1397, 1408-10 (D.C. Cir. 1997).

Congress took this principle a step further under the regional haze program, specifically directing that BART is to be Start Printed Page 5061“determined by the State.” CAA section 169A(b)(2)(A), 42 U.S.C. section 7491(b)(2)(A). Congress adopted the BART provisions to address visibility, rather than health concerns. See H.R. Rep. 95-294, at 529 (1977) (“It should be made clear at the outset that this provision [concerning BART] is totally unrelated to any question involving public health.”) (separate views of Messrs. Devine, Krueger, Broyhill, Gammage, Clarence J. Brown, Collins, Moore and Stockman). Congress therefore sensibly left decisions relating to the imposition of costly visibility control technologies on certain existing sources entirely to the states, where local factors could be properly considered and implemented:

The agreement clarifies that the state, rather than the Administrator, identifies the source that impairs visibility in the Federal class I areas. “. . . In establishing emission limitations for any source which impairs visibility, the State shall determine what constitutes `best available retrofit technology' . . .” H.R. Conf. Rep. 95-564, at 155 (1977). While the original House bill would have given EPA the power to determine what BART controls should be applied to individual sources, Congress eventually inserted the current statutory language to make it clear that the States, rather than EPA, would make BART determinations. See id.; 5 Leg. History of CAA Amendments 1997 P.L. 95-95, H8663 (1997) (“The provision [in the original bill] was modified to give States a greater role in identifying sources which are contributing (or may in the future contribute) to visibility problems and in establishing control requirements for those sources.”). Senator Muskie confirmed during the floor debate that “the State, not the Administrator, identifies a source that may impair visibility” and that “it is the State which determines what constitutes `Best Available Retrofit Technology.' ” 123 Cong. Rec. 26,854 (1977).

The federal courts have enforced this legislative intent. In American Corn Growers, the D.C. Circuit quoted at length from the legislative history of section 169A to conclude that it was “clear that the States—not EPA—would make these BART determinations.” 291 F.3d at 8; see also id. at 8 (“The Conference Report . . . confirms that Congress intended the States to decide which sources impair visibility and what BART controls should apply to those sources.”). American Corn Growers reaffirms that the states have “broad authority” to make their own BART determinations. Id. It also reaffirms that EPA cannot “deprive the states of some of this statutory authority,” nor can EPA “constrain[ ] authority Congress conferred on the states” with respect to BART determinations. Id. at 8-9. It was for this reason that the court struck EPA's first attempt at the Regional Haze Rule: it purported to tell the states how to make BART determinations. Id. at 6-7. The same court later reiterated that BART “requires States to determine the particular technology that an individual source `subject to BART' must install.” Utility Air Regulatory Grp. v. EPA, 471 F.3d 1333, 1336 (D.C. Cir. 2006).

Other federal courts have recognized the cooperative federalism policies on which the CAA in general—and the regional haze provisions in particular—are based. See, e.g., Texas v. EPA, 690 F.3d 670, 684 (5th Cir. 2012); Ellis v. Gallatin Steel Co., 390 F.3d 461, 467 (6th Cir. 2004); Sierra Club v. EPA, 315 F.3d 1295, 1300 (11th Cir. 2002); Am. Lung Ass'n of N.J. v. Kean, 871 F.2d 319, 322 (3d Cir. 1989). Under cooperative federalism, states retain the discretion and flexibility to make their own choices based on local conditions, histories, and policies. See, e.g., Budget Prepay, Inc. v. AT&T Corp., 605 F.3d 273, 281 (5th Cir. 2010) (“ `cooperative federalism' . . . necessarily implies that states may reach differing conclusions on specific issues relating to the implementation of the [statute]”); Global NAPs, Inc. v. Mass. Dep't of Telecom. & Energy, 427 F.3d 34, 46 (1st Cir. 2005) (cooperative federalism has “the intended effect of leaving state commissions free, where warranted, to reflect the policy choices made by their states” and to implement statutory provisions “fairly and with due regard to . . . local conditions . . . and . . . historical circumstances”); Taylor v. Vt. Dep't of Educ., 313 F.3d 768, 777 (2d Cir. 2002) (“`[c]ooperative federalism . . . allows some substantive differentiation among the states in the determination of which . . . theories, practices, and approaches will be utilized'”) (citation omitted).

In sum, Congress directed that BART determinations are to be made by the states, allowing the states to make their own BART choices based on local conditions and other considerations. Because EPA may not exercise authority “in a manner that is inconsistent with the administrative structure that Congress enacted into law,” ETSI Pipeline Project v. Missouri, 484 U.S. 495, 517 (1998), EPA may not disapprove a state BART determination that complies with the CAA, whether or not EPA agrees with the state's decision. Here, EPA has not demonstrated that Wyoming's BART determination violates the CAA, and for that reason EPA must approve the BART determination in the SIP even if it “disagrees” with it. Instead, just as in its rulemaking at issue in Texas, EPA's Proposed Rule “transgresses the CAA's delineated boundaries of [the] cooperative relationship” between EPA and the states. 690 F.3d at 686. Response: EPA disagrees with this comment. First, the legislative history of the 1977 Amendments cited by the commenter is incomplete. The complete legislative history, when fairly read, contradicts the commenter and confirms EPA's supervisory role in reviewing state regional haze SIP submittals, including the state's initial BART determinations.

The 1977 Amendments resulted from a conference agreement that reconciled the House bill, H.R. 6161, and the Senate bill, S. 252. The conference committee agreed to adopt the visibility protection provisions of section 116 of the House bill, with certain modifications. With respect to the BART provision in what is now section 169A(b)(2)(A) of the Act, the conference agreement inserted the phrase “as determined by the State (or the Administrator in the case of a plan promulgated under [section 110(c) of the Act])” in the two places it now appears in that section.[31] The conference agreement inserted similar language into the definition of BART in section 169A(g)(2). The 1977 Amendments also added section 110(a)(2)(J) to the Act, which makes (among other things) a regional haze SIP that meets the requirements of part C relating to visibility protection a required part of a state's SIP.

Thus, H.R. 6161 required states to submit regional haze SIPs containing BART determinations, but did not explicitly specify that the BART determinations should, in the first instance, be made by the state. The conference agreement language clarified that states should make BART determinations as part of their SIP submittals, as explained in the conference report:

The agreement clarifies that the State, rather than the Administrator, identifies the source that impairs visibility in the Federal class I areas identified and thereby fall within the requirements of this section. . . . In establishing emission limitations for any source which impairs visibility, the State shall determine what constitutes “best available retrofit technology” (as defined in this section) in establishing emission limitations on a source-by-source basis to be included in the State implementation plan so as to carry out the requirements of this section.

Start Printed Page 5062

H.R. Conf. Rep. 95-564, at 155 (1977) (emphasis added). In other words, BART determinations are a required element (“the State shall determine”) of a state's regional haze SIP submittal (“to be included in the State implementation plan”). However, the conference report does not say that the state's determination is final. For example, it does not say: “The State shall determine, and EPA shall abide by . . .” Thus, all the conference report says is that states must provide BART determinations as part of the state's required regional haze submittal. As the Tenth Circuit Court of Appeals stated, “All the conference agreement referenced by the D.C. Circuit did was shift the initial responsibility for making BART determinations from the EPA to the state. But that does not differ from other parts of the CAA—states have the ability to create SIPs, but they are subject to EPA review.” Oklahoma v. EPA, 723 F.3d 1201, 1209 (10th Cir. 2013).

Another portion of the legislative history, only partially quoted by the commenter, confirms EPA's supervisory role. Congressman Rogers inserted into the Congressional Record a Clean Air Conference Report (1977): Statement of Intent; Clarification of Select Principles. 123 Cong. Rec. 27070 (daily ed. Aug. 4, 1977) (statement of Cong. Rogers). The Statement of Intent clarified “some important points on the intention and effect of the conferees action [that] may have been overlooked or may be unclear in the text of the conference bill or the accompanying statement of managers.” Id. Under section “D. Visibility protection,” the first full paragraph states:

The conferees essentially agreed to the House provision for visibility protection. The provision was modified to give States a greater role in identifying sources which are contributing (or may in the future contribute) to visibility problems and in establishing control requirements for those sources. However, the conferees rejected a motion to delete the national goal. The conferees also rejected a motion to delete EPA's supervisory role under section 110 to assure that the required progress toward that goal will be achieved by the revised State plan. If a State visibility protection plan is not adequate to assure such progress, then the Administrator must disapprove that portion of the SIP and promulgate a visibility protection plan under section 110(c). Thus, visibility protection in most mandatory federal Class I areas remains a national commitment, which is nationally enforceable.

Id. (emphasis added). Thus, the Statement of Intent, instead of supporting the commenter's arguments, confirms EPA's supervisory role over states' regional haze SIPs, as the conferees deliberately rejected a proposal to remove that supervisory role.[32] The Statement of Intent also only describes states as having a “greater role” in determining BART; it does not describe that role as exclusive.

With respect to Senator Muskie's statements, the comment omits a portion of the legislative history regarding application of the BART Guidelines. Oklahoma v. EPA, 723 F.3d 1201, 1209-10 (10th Cir. 2013). The Tenth Circuit considered those statements in context and confirmed EPA's authority to ensure that state BART determinations for fossil-fuel fired power plants having a total generating capacity greater than 750 MW complied with the BART Guidelines. Id. With respect to the separate views of several Representatives regarding visibility protection as unrelated to public health, those views are of a small minority that opposed any provisions for visibility protection whatsoever. H.R. Rep. 95-294, at 530 (1977). Their views did not carry the day and, in any case, are irrelevant to the question of EPA's supervisory role.

With respect to the remainder of the comment regarding various court opinions, we have responded to similar comments elsewhere. EPA's action here violates neither the holdings in American Corn Growers and UARG regarding the RHR, nor the generic remarks regarding cooperative federalism in the other cited cases.

Comment: Although EPA cites “errors” made by Wyoming in its BART determination for Laramie River Station, EPA has not—and cannot—demonstrate that any of these alleged “errors” represents a violation of the CAA. These are technical disagreements over judgments committed by Congress to the states—not grounds for EPA to step in and dictate a technology choice. Section 169A does not confer any authority upon EPA to make a BART determination when the state has made one. Once the state makes a BART determination, EPA's authority to review it in the SIP review process is very limited. Section 110 mandates that “[EPA] shall approve such [SIP] submittal as a whole if it meets all of the applicable requirements of this chapter.” 42 U.S.C. 7410(k)(3). See also Forest Guardians v. Babbitt, 174 F.3d 1178, 1187 (10th Cir. 1999) (“The Supreme Court and this circuit have made clear that when a statute uses the word `shall,' Congress has imposed a mandatory duty upon the subject of the command.”).

As the Fifth Circuit recently expressed, “the Act confines the EPA to the ministerial function of reviewing SIPs for consistency with the Act's requirements,” and “[t]h[e] statutory imperative [of section 110(k)(3)] leaves the agency no discretion to do anything other than ensure that a state's submission meets the CAA's requirements and, if it does, approve it.” Luminant, 675 F.3d at 921, 926. See also id. at 932 (“If [the State's] regulations satisfy th[e] basic requirements [of the CAA], the EPA must approve them, as section 7410(k)(3) requires. That is the full extent of the EPA's authority in the SIP-approval process because that is all the authority that the CAA confers.”) Texas, 690 F.3d at 676 (“[I]f a SIP or a revised SIP meets the statutory criteria of the CAA, then the EPA must approve it.”); Bethlehem Steel Corp. v. Gorsuch, 742 F.2d 1028, 1036 (7th Cir. 1984) (EPA's SIP disapproval power is “constrained by the substantive criteria in 42 U.S.C. 7410(a)(2)(A)-(K)”); Fla. Power & Light Co. v. Costle, 650 F.2d 579, 581 (5th Cir. 1981) (“If a SIP or a revised SIP meets the statutory criteria. . . the EPA must approve it.”) (citations omitted).

Since Wyoming's BART decision for Laramie River Station, along with its associated SIP revision, meets the requirements set forth in the CAA, EPA has no discretion and must approve it in its entirety. As the Supreme Court explained in the NAAQS context: The Act gives the Agency no authority to question the wisdom of a state's choices of emission limitations if they are part of a plan which satisfies the standards of section 110(a)(2), and the Agency may devise and promulgate a specific plan of its own only if a state fails to submit an implementation plan which satisfies those standards. Section 110(c). Thus, so long as the ultimate effect of a state's choice of emission limitations is compliance with the national standards for ambient air, the state is at liberty to adopt whatever mix of emission limitations it deems best suited to its particular situation. Train, 421 U.S. at 79; see also Virginia, 108 F.3d at 1408-10 (confirming that the 1977 Amendments to section 110 did not alter the division of responsibilities recognized in Train). Accord Union Elec. Co. v. EPA, 427 U.S. 246, 267 (1976) (“[T]he State has virtually absolute power in allocating emission Start Printed Page 5063limitations so long as national standards are met.”).

The fact that states must propose SIP revisions “as may be necessary” to achieve reasonable progress does not mean that EPA has authority to countermand the textual commitment to leave BART decisions to the states. The D.C. Circuit interpreted similar language in Section 110(k)(5) to constrain EPA's authority over SIP approval and disapproval. See Virginia, 108 F.3d at 1409. The SIP call provisions of Section 110(k)(5) similarly state that when a SIP is inadequate “[EPA] shall require the State to revise the plan as necessary to correct such inadequacies.” But the Virginia court rejected the agency's expansive view of this phrase as authority to impose specific control measures for specific emission sources: EPA apparently thinks the “as necessary” language in section 110(k)(5) altered the division of responsibilities between the states and the agency. We suppose the idea is that because section 110(k)(5) empowers EPA to “require the State to revise the plan as necessary to correct” inadequacies, it empowers EPA to require the state to include particular control measures in the revised plan.

There is nothing to this. Id. at 1409. Instead, the court concluded that this phrase “keep[s] EPA within bounds.” Id. at 1410. Imposition of a FIP is intended to be a drastic penalty, imposed only where a state fails to provide the air pollution reductions required by the CAA, as “it rescinds state authority to make the many sensitive and policy choices that a pollution control regime demands.'” Id. at 1406-07 (citation omitted). The court also expressed, in rejecting EPA's interpretation of Section 110(k)(5), that “[w]e would have to see much clearer language to believe a statute allowed a federal agency to intrude so deeply into state political processes.” Id. at 1410.

EPA must therefore approve the Wyoming SIP as it relates to BART at Laramie River Station, as compliance with the law is all that is required. See Luminant, 675 F.3d at 926 (EPA's reliance on factors other than compliance with the CAA in disapproving a SIP violated the Administrative Procedures Act (APA), as it was “in excess of statutory authority,” and was arbitrary and capricious, as it considered “a `factor[ ] which Congress has not intended [the EPA] to consider' ”) (quoting 5 U.S.C. 706(2)(C) and State Farm, 463 U.S. at 43) (alteration in original).

Response: EPA is not substituting its judgment on required technology for the State's in this decision. Rather, we have determined that Wyoming's analysis and determinations were not performed consistent with the CAA and implementing regulations. EPA considered the State's SIP as well as the most recent information submitted by Basin Electric and others for the Laramie River BART units. As explained in detail in our response to similar comments in the BART section of this document, we found Basin Electric's estimates of SCR capital cost deficient in a number of respects, specifically: (1) Inadequate explanation for the high labor rates that were assumed when compared to published labor rates; (2) High overtime and per diem costs without sufficient explanation; (3) Apparent duplication of costs associated with General Facilities; (4) Inclusion of AFUDC; (5) Apparent duplication of contingencies and other cost adders; and (6) Addition of unnecessary SO3 mitigation system. All of these contributed to excessively high capital cost. Sargent & Lundy also assumed excessively high cost for replacement catalyst, which contributes to high operating cost. As we explain elsewhere, these deficiencies are inconsistent with the CAA and RHR.

We responded to similar comments regarding the remaining comments above and elsewhere in this document.

Comment: To the extent that the Supreme Court in ADEC suggested it was adopting a “reasonableness” standard, and did not expressly state that what it was doing was adopting an “arbitrary and capricious” standard, the Supreme Court and other federal courts have confirmed that these two standards are nearly interchangeable. Moreover, to the extent that there is any perceivable difference between the two standards, these cases confirm that “reasonable” means something more like “not arbitrary and capricious” than “not what EPA would prefer.” See, e.g., Marsh v. Ore. Nat. Res. Council, 490 U.S. 360, 377 n.23 (1989) (“as some of the[ ] courts have recognized, the difference between the `arbitrary and capricious' and `reasonableness' standards is not of great pragmatic consequence”) (citing cases); Ridenour v. Kaiser-Hill Co., 397 F.3d 925, 939 (10th Cir. 2005) (“When a party challenges agency action as arbitrary and capricious the reasonableness of the agency's action is judged in accordance with its stated reasons.”) (citation omitted); Amisub (PSL), Inc. v. Colo. Dep't of Social Servs., 879 F.2d 789, 800 (10th Cir. 1989) (the court's role in applying the arbitrary and capricious review standard is “to determine if there was a reasonable factual basis to support” the agency's findings); United States v. Minnkota Power Co-Op Inc., 831 F. Supp.2d 1109, 1119 (D.N.D. 2001) (expressing that the “reasonableness” standard employed by the ADEC Court is the same as the “arbitrary and capricious” standard).

Under the APA's arbitrary and capricious review standard, administrative action is presumed valid, and review of that action is “ `narrow in scope.' ” Copar Pumice Co. v. Tidwell, 603 F.3d 780, 793 (10th Cir. 2010) (citation omitted). “Agency action is arbitrary and capricious only if the agency `has relied on factors which Congress has not intended it to consider, entirely failed to consider an important aspect of the problem, offered an explanation for its decision that runs counter to the evidence before the agency,' or if the agency action `is so implausible that it could not be ascribed to a difference in view or the product of agency expertise.' ” Id. (quoting State Farm, 463 U.S. at 43). A court will not “substitute [its] judgment for that of the agency,” but will only consider whether the agency provided a “reasoned basis” for its action. Id. at 793-94 (quoting State Farm, 463 U.S. at 43). The courts also have developed a series of related standards designed to ensure that courts afford appropriate deference to an agency's technical and policy choices, and refrain from substituting the courts' judgment for that of the agency. For the same reasons that arbitrary and capricious review should apply to EPA's review of a state BART determination, these related standards also should apply: (1) The State's BART decision is presumed valid, and EPA bears the burden of proving otherwise, see Hillsdale Envt'l Loss Prevention, Inc. v. U.S. Army Corps of Eng'rs, 702 F.3d 1156, 1165 (10th Cir. 2012); (2) the State's decision may be set aside “ `only for substantial procedural or substantive reasons,' ” id. (citation omitted); and (3) where experts might disagree about a technical issue, EPA must defer to the “reasonable opinions” of the States' experts, see Colo. Wild v. U.S. Forest Serv., 435 F.3d 1204, 1214 (10th Cir. 2011). See also Minnkota Power, 831 F. Supp.2d at 1119-20 (the same principles that apply to court review of agency action under the APA apply to EPA challenges to state BACT determinations).

EPA's proposal does not formulate or apply these standards, and thus does not establish grounds to overrule the State's BART determination for Basin Electric's Laramie River Station. EPA has not found that Wyoming “entirely failed to consider an important aspect of the problem,” considered factors Congress did not intend it to consider, or reached a decision “so implausible” Start Printed Page 5064as to be arbitrary. Nor has EPA found that Wyoming's explanation for its decision runs counter to the evidence that was before it. Instead, EPA complains of minor alleged deviations from broadly worded and highly flexible guidelines deliberately designed to be consulted but not rigidly adhered to in any event. EPA therefore must approve the State's BART decision for Laramie River, as any other result represents EPA's substitution of its judgment over Wyoming's, which EPA has no statutory authority to do.

Response: EPA disagrees with this comment, which is based on a fundamental misunderstanding of EPA's role. In acting on a state's SIP submittal, EPA does not sit in the position of a reviewing federal court. Instead, EPA is the agency entrusted by Congress with administering the CAA. Thus Congress has “vested EPA with explicit and sweeping authority to enforce CAA requirements” and requires that “EPA step in to ensure that the statutory requirements are honored.” Alaska Dep't of Envtl. Conservation v. EPA, 540 U.S. 461, 490 (2004). Reviewing courts, on the other hand, “are not experts in the field” and thus defer to decisions by “the agency charged with the administration of the statute.” Chevron, Inc. v. Natural Res. Def. Council, 467 U.S. 837, 866 (1984).

In the context of acting on a regional haze SIP, EPA must assure that it meets the requirements of the Act and the RHR, including requirements regarding BART. EPA—unlike a reviewing court—is not required to defer to the state's technical judgments. Instead, EPA is not only authorized, but required to exercise independent technical judgment in evaluating the adequacy of a state's regional haze SIP, including its BART determinations, just as EPA must exercise such judgment in evaluating other SIPs. In evaluating other SIPs, EPA's role is always to make a judgment about SIP adequacy, not just to meet and maintain the NAAQS, but also to meet other requirements that do not have a numeric value. In this case, Congress did not establish NAAQS by which to measure visibility improvement; instead, it established a reasonable progress standard and required that EPA assure that such progress be achieved. Here, contrary to the commenter's assertion, we are exercising judgment within the parameters laid out in the CAA and our regulations. Our interpretation of our regulations and of the CAA, and our technical judgments, are entitled to deference. See, e.g., Michigan Dep't. of Envtl. Quality v. Browner, 230 F.3d 181 (6th Cir. 2000); Connecticut Fund for the Env't., Inc. v. EPA, 696 F.2d 169 (2nd Cir. 1982); Voyageurs Nat'l Park Ass'n v. Norton, 381 F.3d 759 (8th Cir. 2004); Mont. Sulphur & Chem. Co. v. EPA, 666 F.3d 1174 (9th Cir. 2012).

The comment does not cite to anything in the ADEC opinion (or, for that matter, in the CAA itself) that suggests EPA must, in reviewing a SIP submittal, adopt the APA standards of review. Instead, in ADEC the Supreme Court upheld EPA's position that the State permitting agency's BACT determination “did not qualify as reasonable in light of the statutory guides.” Alaska Dep't of Envtl. Conservation v. EPA, 540 U.S. 461, 484 (2004). The mere coincidence that some courts have described the APA standards of review as essentially a “reasonableness” standard does not compel EPA to adopt the APA standards of review; nor did the ADEC opinion suggest EPA must do so. As explained above, a fundamental difference between EPA and a reviewing court is that courts lack technical expertise and so generally defer to agency technical judgments; on the other hand, EPA is the expert agency entrusted by Congress with administering the CAA and exercising its best technical judgment in doing so. Another fundamental difference is that a reviewing court is limited to the record compiled by the administrative agency, but EPA in its review of a SIP submittal is not limited just to the record compiled by the state agency, and may supplement the record with (among other things) EPA's own expert reports and analyses. In fact, if the cases cited by the commenter discussing the APA standard of review stand for anything, it is the proposition that if and when EPA's action on this SIP submittal is subject to judicial review, the court will base its decision on the record compiled by EPA and give appropriate deference to EPA's technical judgments and interpretations of the Act and the RHR. Accordingly, the Eighth and Tenth Circuit Court of Appeals have applied the APA standard of review to EPA's actions on other regional haze SIP submittals. See Oklahoma v. EPA, 723 F.3d 1201 (10th Cir. 2013), North Dakota v. EPA, 730 F.3d 750 (8th Cir. 2013).

The discussion of the standard of review in the district court's order and opinion in United States v. Minnkota Power Co-op., Inc., 831 F. Supp. 2d 1109 (D.N.D. 2011), cited by commenter, is not to the contrary. The district court's opinion first quotes the ADEC opinion for the proposition that the question presented is whether “the state agency's BACT determination was reasonable, in light of the statutory guides and the state administrative record.” Id. at 1119 (emphasis added). The district court's opinion then again quotes the ADEC opinion: “We apply the familiar default standard of the Administrative Procedure Act . . . and ask whether the Agency's action was `arbitrary, capricious, an abuse of discretion, or otherwise not in accordance with law.' ” Id. (emphasis added). In the context of the ADEC opinion, the Agency referred to by the Supreme Court in the second quote is EPA, not the State agency. The district court's opinion then continues by quoting a separate Supreme Court opinion discussing the similarities of the arbitrary and capricious standard and the reasonableness standard. This fails to establish any sort of connection between the APA standard and EPA's review of a state determination. In addition, Minnkota Power took place in the context of an enforcement action, not action on a SIP submittal. The EPA had entered into a consent decree that (among other things) “establishe[d] the standard of review governing the EPA's challenge to the North Dakota NOX BACT Determination.” Id. at 1112. The consent decree provided that “[t]he disputing Party shall bear the burden of proof throughout the dispute resolution process.” Thus, Minnkota Power has nothing to say about use of the APA standard in EPA's review of a state's BART determination.

Comment: In applying the arbitrary and capricious standard, EPA should accord the same deference to a state's BART determination that courts accord to an agency decision under the National Environmental Policy Act (NEPA), which, like section 169A, “does not mandate particular results, but simply prescribes the necessary process.” Robertson v. Methow Valley Citizens Council, 490 U.S. 332, 350 (1989). See also 42 U.S.C. 4332(2)(C) (any agency contemplating a “major Federal action [that] significantly affect[s] the quality of the human environment” must prepare an environmental impact statement [EIS] analyzing the action's environmental effects). Under NEPA, “[t]he role of the courts is simply to ensure that the agency has adequately considered and disclosed the environmental impact of its actions and that its decision is not arbitrary or capricious.” Baltimore Gas & Elec. Co. v. NRDC, 462 U.S. 87, 97-98 (1983).

The purpose of this deferential review standard under NEPA is to prevent a court from “substitut[ing] its judgment for that of the agency.” Kleppe v. Sierra Club, 427 U.S. 390, 410 n.21 (1976). As the Supreme Court explained in Kleppe, Start Printed Page 5065“[t]he only role for a court is to insure that the agency has taken a `hard look' at environmental consequences; it cannot `interject itself within the area of discretion of the executive as to the choice of the action to be taken.'” Id. (citing NRDC v. Morton, 458 F.2d 827, 838 (D.C. Cir. 1972)).

Under this review standard, “even if [the reviewing court] would have made a different choice had the matter been before [the court] de novo,” the court “cannot displace the agencies' choice” between conflicting views, evidence, data, and scientific opinions. Custer Cnty. Action Ass'n v. Garvey, 256 F.3d 1024, 1036 (10th Cir. 2001). Thus, even in the face of technical objections, a court will uphold the agency's action so long as it is supported by substantial evidence in the administrative record, is adequate to foster informed public participation and decision making, and is not otherwise arbitrary or capricious. Id.

Moreover, as the courts have repeatedly recognized, “[d]eficiencies in an EIS that are mere `flyspecks' and do not defeat NEPA's goals of informed decision making and informed public comment will not lead to reversal.'” WildEarth Guardians v. NPS, 703 F.3d 1178, 1183 (10th Cir. 2013) (quoting New Mexico v. BLM, 565 F.3d 683, 704 (10th Cir. 2009)). See also Custer Cnty, 256 F.3d at 1035 (“Our objective is not to `fly speck' the [EIS], but rather, to make a `pragmatic judgment whether the [EIS]'s form, content and preparation foster both informed decision-making and informed public participation.'”) (citation omitted).

The same principles apply here, where Congress has expressly delegated the BART decision to the states, did not mandate the states to reach a specific outcome, and established only a decision making process for the states to follow—not a required outcome. If the state considered all five statutory factors to arrive at a result that improves visibility, and its decision is not arbitrary, capricious, an abuse of discretion, or otherwise not in accordance with law, EPA must affirm the BART selection—even if EPA would or could have made a different selection.

Response: EPA disagrees with this comment. The comment does not identify anything in the NEPA court decisions that demonstrates that those decisions are applicable to EPA's review of a SIP submittal. In fact, Section 7(c) of the Energy Supply and Environmental Coordination Act of 1974 (15 U.S.C. 793(c)(1)) exempts actions under the CAA from the requirements of NEPA. Specifically, this section states that “[n]o action taken under the CAA [42 U.S.C. 7401 et seq.] shall be deemed a major Federal action significantly affecting the quality of the human environment within the meaning of the National Environmental Policy Act of 1969 [42 U.S.C. 4321 et seq.].” While the standard of review for EPA's SIP and FIP decisions may be similar to that under NEPA,[33] the NEPA decisions simply are not applicable in the CAA context.

Furthermore, NEPA relies solely on “procedural mechanisms—as opposed to substantive, result-based standards.” Robertson v. Methow Valley Citizens Council, 490 U.S. 332, 353 (1989). Unlike NEPA, the CAA's regional haze program has specific substantive requirements, and EPA must ensure that SIP submittals meet the requirements of the Act, including the substantive provisions of the regional haze program. See CAA Section 110(a)(2)(J) (SIP submittals must meet applicable requirements of Part C of title I, including visibility protection). As the Eighth Circuit Court of Appeals stated: “EPA is left with more than the ministerial task of routinely approving SIP submissions.” North Dakota v. EPA, 730 F.3d 750 (8th Cir. 2013) (emphasis added).

Comment: One commenter asserted that the U.S. Supreme Court and the lower federal courts have long recognized and applied the principle of “harmless error” where an agency may have committed an error, but that error did not affect the outcome of its decision. See, e.g., Nat'l Ass'n of Home Builders v. Defenders of Wildlife, 551 U.S. 644, 659 (2007); Hillsdale, 702 F.3d at 1165. See generally 5 U.S.C. 706 (“[D]ue account shall be taken of the rule of prejudicial error.”).

The commenter argued that the courts also have long recognized the related principle that agencies may “overlook circumstances that in context may fairly be considered de minimis,” as part of the broad notion that “the law does not concern itself with trifling matters.” Alabama Power Co. v. Costle, 636 F.2d 323, 360 (D.C. Cir. 1979). Thus, for instance, the D.C. Circuit rejected a challenge to a Federal Aviation Administration rule where the agency had used “inappropriate guidelines for measuring the effects of noise” in its determination that a proposed airport site would not result in any “use” of a nearby wildlife refuge. Allison v. Dep't of Transp., 908 F.2d 1024, 1026 (D.C. Cir. 1990). The court cited the APA provision requiring consideration of “prejudicial error,” and expressed that “[a] court should not upset a decision because of errors that are not material.” Id. at 1029 (citations omitted). See also Grunman Data Sys. Corp. v. Widnall, 15 F.3d 1044, 1048 (Fed. Cir. 1994) (rejecting bid protest although agency may have violated accounting principles in its analysis of the best value bid, as any accounting errors were “de minimis,” and stating that “overturning awards on de minimis errors wastes resources and time, and is needlessly disruptive of procurement activities and governmental programs and operations”) (citation omitted).

Finally, the commenter argued, the courts have repeatedly held that agency action should not be reversed due to mere calculation errors that do not render a rule arbitrary and capricious. See, e.g., Michigan v. EPA, 213 F.3d 663, 691 (D.C. Cir. 2000) (rejecting challenge to EPA decision despite error in calculation); Chem. Mfrs. Ass'n v. EPA, 870 F.2d 177, 241, clarified on reh'g, 885 F.2d 253 (5th Cir. 1989) (same); CPC Int'l, Inc. v. Train, 540 F.2d 1329, 1343-44 (8th Cir. 1976) (same). The commenter stated that these same principles should apply to EPA's review of the State's BART determinations, such that EPA has no authority to disapprove the State's decisions if a deviation from the BART Guidelines and CCM was merely de minimis and at most harmless error that did not affect the State's selection of BART. Indeed, EPA's approach itself suggests that the BART Guidelines and CCM were intended to be flexible, and that EPA's review of compliance with their provisions is subject to a materiality standard. For instance, in the Proposed Rule, EPA proposes to disapprove certain BART determinations based on purported deviations from the BART Guidelines and CCM in assessing cost and visibility, yet it also proposes to approve other BART determinations “because [it has] determined that the State's conclusions were reasonable despite the cost and visibility errors” identified by EPA. 78 FR 34750. And, while the Tenth Circuit's decision in Oklahoma v. EPA is not yet final, as petitions for rehearing may yet be filed, that court similarly suggested that there Start Printed Page 5066was a materiality element to a state's compliance with the BART Guidelines, noting, in particular, that the State's cost estimates were “more than ten times EPA's stated average costs per ton for th[e] technology, and nearly five times as much as the upper limit of EPA's expected cost range.” —F.3d—, 2013 U.S. App. LEXIS 14634, at *25 (10th Cir. July 19, 2013). Notably, that case did not involve SCR technology, which the CCM affords a greater amount of flexibility in assessing, and the State had failed to note and explain its deviations from the CCM.

By applying these principles here, the commenter asserted, any deviation from the BART Guidelines and CCM was de minimis, and mere harmless error. Certainly, EPA has not shown that the State would have made a different BART selection had it assessed the cost and visibility factors in the manner EPA suggests—particularly as the selection of BART must be made by weighing all five factors, and as the differences between the State's and EPA's assessments of cost and visibility are not so substantial as to necessitate a different result. In other states, EPA has acknowledged that a state's BART determination may be disapproved on account of a claimed error only if the error would have changed the BART determination. In approving Colorado's regional haze SIP, EPA did not disapprove the BART determination for the Martin Drake power plant, despite EPA's disagreement regarding the control efficiency of SCR because the discrepancy would not have changed the outcome. 77 FR 76871, 76875-76 (Dec. 31, 2012) (“[We] find that it was not unreasonable for Colorado to use 0.07 lb/MMBtu to model the predicted visibility improvement from SCR. Moreover, while we do agree that assuming a control efficiency of 0.05 lb/MMBtu would have resulted in greater modeled visibility benefits, we do not agree that the difference in visibility benefits would have led Colorado to a different conclusion given the magnitude of the benefits associated with SCR.”). The commenter advocated that EPA should take a similar approach in Wyoming.

The commenter finished by stating that if there is a question as to whether the State might have made a different BART selection had it assessed cost and visibility in the manner suggested by EPA, EPA should return the issue to the State to reweigh the BART factors with that information. See SKF USA Inc. v. United States, 254 F.3d 1022, 1029 (Fed. Cir. 2001) (courts may remand matters to the agency upon request to correct “clerical errors, transcription errors, or erroneous calculations”).

Response: The cases cited to by the commenter all concern standards by which courts evaluate agency action, not standards by which EPA, an administrative agency, evaluates SIP submissions for compliance with the requirements of the CAA. The cases are therefore inapposite. Nevertheless, in situations where a state's SIP reaches a reasonable result overall despite violations of certain statutory or regulatory requirements, EPA believes that approving the SIP is sometimes a better use of scarce administrative resources and more in line with principles of cooperative federalism than promulgating a FIP. This approach is arguably similar to the principle of “harmless error” that courts adhere to in the context of judicial review.

In this situation, however, the errors committed by Wyoming in its regional haze SIP were neither harmless nor de minimis. As we have explained previously, because Wyoming did not properly calculate the costs of the various control options or accurately estimate the visibility improvement associated with these controls, the State's ultimate selection of BART for several EGUs did not represent the best system of continuous emission reduction. As the Eighth and Tenth Circuits have recently held, EPA acts within its power under section 169A of the CAA when it rejects a BART determination on the basis that a state did not properly take into consideration the costs of compliance as a result of methodological or data flaws. See Oklahoma v. EPA, 723 F.3d 1201, 1212 (10th Cir. 2013); North Dakota v. EPA, 730 F.3d 750 (8th Cir. 2013). This same reasoning applies equally to the other statutory BART factors, such as visibility improvement.

We also disagree with the commenter that our action on the Colorado regional haze SIP implies that a similar outcome is warranted here. In that action, we stated that “it was not unreasonable for Colorado to use 0.07 lb/MMBtu to model the predicted visibility improvement from SCR.” 77 FR 76871, 76875 (Dec. 31, 2012). Thus, we did not disagree with Colorado's choice of control efficiency, as the commenter claims, and the situation bears no relationship to this one, where we have carefully explained our disagreement with multiple aspects of Wyoming's NOX BART determinations.

Finally, we decline to “return the issue to the State,” as the commenter proposes. At this time, the Wyoming regional haze SIP is many years overdue, and the deadline for EPA to issue a FIP has long since passed. We note, however, that Wyoming is free to submit a SIP revision at any time that, if approved, could replace all or a portion of EPA's FIP.

Comment: EPA's proposal to disapprove Wyoming's BART determination for Laramie River not only overrides the State's technical judgment but also renders moot with a stroke of a pen the extensive judicial, administrative, and political processes developed by the State to implement its obligations under the CAA as a separate sovereign. Wyoming has enacted a robust and independent set of administrative and judicial procedures to review and potentially overturn BART decisions made by the State. These procedures are part of the State's SIP expressly approved by EPA, 40 CFR 52.2620, making them federally enforceable.

Wyoming's air quality regulations require a source subject to BART to apply for and obtain a BART permit. In this case, Laramie River Station's BART permit was issued pursuant to Wyoming Air Quality Standards and Regulations (WAQSR) Chapter 6, Sections 2 and 9. The rules requiring BART permits in Wyoming were adopted on October 9, 2006 as a new section to meet the requirements of EPA's RHR. Chapter 6 requires facilities seeking permits to comply with all the rules and regulations of Wyoming. Chapter 6, Section 9 of the Air Quality Division's rules and regulations govern BART permits. Section 9(e)(iv) requires that the opportunity for public comment on BART permits follow the procedures specified in Chapter 6, Section 2(m). That section, in turn, establishes a notice and comment procedure that specifically requires a copy of the public notice to be sent to EPA. Thus, EPA approved Wyoming's plan that specifically contemplates EPA's inclusion in State administrative review proceedings. See 40 CFR 52.2620; see also US Magnesium, LLC v. EPA, 690 F.3d 1157, 1159 (10th Cir. 2012) (EPA's approval of a State's SIP gave the SIP the force and effect of federal law).

Here, EPA received the required notice at every step of the proceedings. EPA, however, chose to participate to only a limited extent. After submitting August 3, 2009 comments to the State's BART Application Analysis and proposed permit and October 26, 2009 comments to Wyoming's draft regional haze SIP, EPA excised itself from the process. Despite its prior comments on Basin Electric's BART permit and the regional haze SIP, EPA did not seek to intervene in Basin Electric's administrative appeal to the Start Printed Page 5067Environmental Quality Council or comment on Basin Electric's settlement agreement with the Environmental Quality Council. EPA could have advised the Environmental Quality Council that it believed the proposed settlement violated the CAA or was otherwise arbitrary and capricious, but it did not. Instead, illustrating its disregard for State primacy, EPA now proposes to disapprove the NOX BART emissions limits in the settlement agreement and final SIP, years after the administrative process concluded.

As the dissenters in ADEC described, EPA should not be permitted to avoid a “more painstaking state process by a mere stroke of the pen under the agency's letterhead.” 540 U.S. at 509 (Kennedy, J., dissenting) (discussing an analogous process for BACT determinations). The CAA's “strict” division of authority creates a “statutory federalism bar [that] prohibits EPA from using the SIP process to force States to adopt specific control measures.” EME Homer City Generation, L.P. v. EPA, 696 F.3d 7, 29 (D.C. Cir. 2012) (citing Virginia, 108 F.3d at 1410). But that is precisely what EPA seeks to do here. EPA's approach both confuses the CAA “with a general administrative law statute like the [APA]” and upsets “the balance between State and Federal Governments.” See ADEC, 540 U.S. at 507-17 (Kennedy, J., dissenting). Simply put, it is inappropriate for EPA to dodge the administrative and judicial review process established in the State of Wyoming through overturning of Wyoming's BART decision by administrative fiat. See id. at 510 (Kennedy, J., dissenting). It was only after Wyoming submitted its regional haze SIP to EPA that EPA announced it found the settlement “unreasonable” and something with which it “disagreed.” Based upon these assertions, and without demonstrating that the BART permit actually violates the CAA, EPA now proposes to void all the extensive administrative proceedings, processes, comment periods, and permit finality accorded under State law.

This improperly impinges upon state authority. Under the regional haze program, deference to state authority is far more compelling than issues related to public health under the BACT program, and so the Supreme Court's holding in ADEC that EPA may not require “recourse to state processes” is inapplicable to BART decisions. ADEC, 541 U.S. at 492. EPA should conduct itself in accordance with the spirit of its representation to the Supreme Court that it has never sought to override a state court judgment, and should not seek to override a state BART decision that has been litigated to administrative conclusion under state law, particularly where, as here, EPA never advised the State adjudicators or the parties to the State proceedings that it considered the permit to be invalid under the CAA. EPA could have participated in the State administrative appeal proceeding or, at a minimum, appeared in the proceeding to register an objection to the settlement agreement. Having elected not to do so, EPA should respect the result of the State's process. Alternatively, EPA is precluded from overruling the Laramie River BART permit decision that resulted from that process. ADEC, 540 U.S. at 491 n.14. EPA had notice and ample opportunity to contest the appropriateness and legality of the BART permit in Wyoming, but simply chose not to do so.

EPA is not free to let parties like Basin Electric spend thousands of dollars and years of effort resolving the terms of a BART permit, only to find the process wasted because EPA disagrees yet chose to ignore multiple notices of the State proceedings. Absent application of claim preclusion under these circumstances, EPA could effectively “rescind[ ] state authority to make the many sensitive and policy choices that a pollution control regime demands.” Virginia, 108 F.3d at 1406-07 (citation omitted). Here, EPA does not intrude upon state political processes; it ignores them, upsetting “the balance between State and Federal Governments.” See ADEC, 540 U.S. at 507-17 (Kennedy, J., dissenting).

EPA's interference with State's prerogatives also violates the Tenth Amendment to the United States Constitution. “[T]he Tenth Amendment confirms that the power of the Federal Government is subject to limits that may, in a given instance, reserve power to the States.” New York v. United States, 505 U.S. 144, 157 (1992). See also U.S. Const. amend. X (“The powers not delegated to the United States by the Constitution, nor prohibited by it to the States, are reserved to the States respectively, or to the people.”). Here, EPA's rejection of Wyoming's BART decision and imposition of its own not only overrides Congress' resolution to leave localized BART analyses in the hands of the states, but also infringes on Wyoming's (and its citizens') Tenth Amendment right to have those decisions made and adjudicated by the State. See Arlington, 133 S.Ct. at 1874 (although Chevron deference generally applies to an agency's interpretation of the scope of its authority, “[w]here Congress has established a clear line, the agency cannot go beyond it; and where Congress has established an ambiguous line, the agency can go no further than the ambiguity will fairly allow”); Hodel v. Va. Surface Min. & Reclamation Ass'n, 452 U.S. 264, 289 (1981) (statute survived Tenth Amendment scrutiny because it “establishes a program of cooperative federalism that allows the States, within limits established by federal minimum standards, to enact and administer their own regulatory programs, structured to meet their own particular needs,” instead of “commandeer[ing] the legislative processes of the States by directly compelling them to enact and enforce a federal regulatory program”).

Earlier comments provided similar arguments, by noting that Wyoming issued its BART Application Analysis and proposed permit on May 28, 2009, and accepted public comments on its analysis and proposed permit for a period of 60 days, followed by a public hearing on August 6, 2009. Numerous comments were received, including comments from EPA dated August 3, 2009. EPA did not comment that Wyoming's proposed BART determination violated the CAA. Nor did EPA identify any action taken by Wyoming in connection with the permit that was arbitrary or capricious. While EPA regularly encouraged Wyoming to consider both SNCR and SCR technologies, at no point did EPA advise Wyoming that BART controls of LNBs and OFA for the Laramie River Station would violate the CAA or otherwise be arbitrary and capricious. Basin Electric appealed its BART permit to the Environmental Quality Council, arguing that Wyoming's imposition of additional technology requirements in 2018 as part of its long term goals exceeded its authority for terms contained in a BART permit. In its appeal, Basin Electric accepted LNB and OFA as BART but objected to the additional permit condition related to long term strategies.

Basin Electric served its Petition for Review before the Environmental Quality Council on EPA, and EPA received this notice of appeal, as indicated by its acceptance of the certified mail forwarding the appeal. Thereafter, EPA chose not to comment or otherwise participate in Basin Electric's appeal and never informed the parties or the Environmental Quality Council that EPA considered Wyoming's BART decision to violate the CAA. In fact, no contention was made, by any person or entity, that the BART permit issued by Wyoming violated the CAA.

After litigation, Basin Electric's appeal was settled. Wyoming agreed to remove the provision related to future Start Printed Page 5068control strategies in exchange for Basin Electric's agreement to reduce emission levels further than those proposed in the original permit and provide even further reductions by the end of 2017. This proposed settlement was presented to the Environmental Quality Council for approval. No persons or entities objected to the proposed settlement, including EPA.

Only after Wyoming's regional haze SIP was submitted to EPA did EPA announce that it found the settlement “unreasonable” and something with which it “disagreed.” Based upon these assertions, and without demonstrating that the BART permit actually violates the CAA, EPA now proposes to void all of the extensive administrative proceedings, processes, comment periods and permit finality accorded under state law.

This violates the explicit representations EPA made to the United States Supreme Court that decisions to over-ride state technology choices are rarely undertaken and therefore do not pose a threat to state adjudicative processes. In footnote 14 of the ADEC decision, the Court quoted EPA for the proposition that EPA has engaged in “restrained and moderate” use of its authority to overrule specific technology choices and has never “asserted authority to override a state-court judgment.” Based upon this understanding, the majority in ADEC dismissed concerns expressed by the dissent about state/federal relations, stating that “[e]xperience . . . affords no grounding for the dissent's predictions that EPA oversight . . . will `rewor[k] . . . the balance between State and Federal Governments' and threaten state courts' independence.” ADEC, 540 U.S. at 493 n. 16. With its proposed action here, however, EPA is doing precisely what the dissent in ADEC predicted, ignoring the extended contested case process afforded under state law and the final administrative litigation resolution reached under state law.

While Basin Electric's appeal ended short of a court proceeding, the distinction between a litigated judgment in an administrative appeal and a judgment in a state court proceeding is not significant. In both cases, EPA's proposed action fails to respect the cooperative federalism that underlies the CAA in general. Under the RHR deference to state authority is far more compelling than issues related to public health under the BACT program, and so the Supreme Court's holding in ADEC that EPA may not require “recourse to state processes” is inapplicable to BART decisions. ADEC, 541 U.S. at 492. EPA should conduct itself in accordance with the spirit of its representation to the Supreme Court that it has never sought to override a state-court judgment, and should not attempt to override a state BART decision that has been litigated to an administrative conclusion under state law particularly where, as here, EPA never advised the state adjudicators or the parties to the state proceedings that it considered the permit to be invalid under the CAA. EPA could have participated in the State administrative appeal proceeding or at a minimum appeared therein to register an objection to the settlement agreement. Having elected not to do so, EPA should respect the result of the State's process.

Response: EPA disagrees with this comment. As an initial matter, as provided in detail elsewhere in this section and in the docket for this action, we provided feedback to the State in our comment letters on the proposed SIP and in meeting with State and company officials; therefore, the State and companies were aware of our expectations.

That WAQSR Chapter 6, Section 2 has been approved into the SIP does not somehow commit EPA to participate in Wyoming's BART permit process. The Act and the RHR do not require that BART be determined through a permit process that is subject to administrative appeal or through a permit process at all. The SIP-approved provision in Chapter 6, Section 2 for notice to EPA of permit actions meets the requirements of 40 CFR 51.161(d), regarding public procedures for review of new or modified sources, not BART sources. Furthermore, nothing in Chapter 6, Section 2 suggests that notice to EPA of a permit process somehow binds EPA to participate in that process.

The commenter provides no statutory, regulatory, or judicial authority to support the proposition that EPA must participate in state administrative or judicial procedures. With respect to state judicial procedures, the Supreme Court has stated: “[i]t would be unusual, to say the least, for Congress to remit a federal agency enforcing federal law solely to state court.” Alaska Dep't of Envtl. Conservation v. EPA, 540 U.S. 461, 493 (2004). Thus the Court “decline[d] to read such an uncommon regime into the [CAA].” Id. The commenter's notion that the ADEC opinion (which concerned a BACT determination under the PSD program) is inapplicable to BART determinations, merely because BART determination are part of a program to improve visibility rather than public health, finds no support in the ADEC opinion or anywhere in the CAA. We elsewhere respond to comments that argue that the language of the CAA itself requires a greater level of deference to states BART determinations.

With respect to the dissent in ADEC, that dissent of course does not represent the opinion of the Supreme Court. Nonetheless, EPA is not undoing the State's process through the “mere stroke of a pen on the Agency's letterhead,” but instead is acting on the State's regional haze submittal through notice-and-comment rulemaking that is potentially subject to judicial review. Furthermore, EPA is not confusing the CAA with the APA; our authority and duty to review the State's regional haze SIP for compliance with the CAA and the RHR stems from the CAA itself. As we discuss elsewhere, EPA's role in reviewing SIPs differs in many key aspects from that of a court reviewing agency action under the APA.

Under the CAA, states are required to submit SIPS that contain emissions limits necessary to protect visibility, and EPA is required to disapprove of any inadequate SIPs and promulgate FIPs in their place. 42 U.S.C. 7491(b)(2); Section 7410(c)(1)(A). The CAA does not require EPA to participate in state proceedings related to its SIP submission, nor does it preclude EPA from carrying out its statutory duty to disapprove an inadequate SIP if EPA does not participate in state proceedings. The notion that BART determinations are insulated from EPA review simply because the State has an administrative appeal process not only has no support in the Act, it is contrary to the purposes of the Act and EPA's express obligation to approve only SIP submittals that meet the requirements of the Act.

Moreover, any state BART decisions made under an unapproved SIP are not federally enforceable because any SIP “shall not be treated as meeting the requirements of this chapter until the Administrator approves the entire plan revision as complying with the applicable requirements.” 42 U.S.C. 7410(k)(3); see also Gen. Motors Corp. v. United States, 496 U.S. 530, 540 (1990) (holding EPA may bring enforcement action under an existing SIP while a SIP proposal is pending).

Finally, this action does not violate the Tenth Amendment. The Supreme Court has explained that “where Congress has the authority to regulate private activity under the Commerce Clause, we have recognized Congress' power to offer States the choice of regulating that activity according to federal standards or having state law pre-empted by federal regulation.” New Start Printed Page 5069York v. United States, 505 U.S. 144, 167 (1992); see also U.S. Const. Art. I, Section 8, cl. 3 (commerce clause); id. Art. VI, cl. 2 (supremacy clause). The commenter does not argue that the CAA is outside of Congress' Commerce Clause authority. Through the SIP/FIP mechanism, the Act offered Wyoming the choice of regulating sources in the State in accordance with the regional haze provisions in the CAA and with rules promulgated by EPA under its CAA authority; thus the Act itself does not violate the Tenth Amendment. With respect to this particular action, our disapproval of Wyoming's regional haze SIP and our FIP compel no action on the part of the State and are not coercive vis-a-vis the State. As explained elsewhere in these responses, EPA has not required Wyoming to adopt specific control measures. Instead, our FIP contains requirements applicable only to some private companies. The Tenth Amendment is not implicated by our action.

Comment: Even if EPA can contravene the state process, it should still require compelling circumstances demonstrating a plain and unambiguous violation of the CAA before it countermands a state proceeding. Such a showing is necessary to preserve the balance between Federal and state governments under the CAA. EPA is undermining the significance and integrity of the State appeals process as well as the State's authority to determine BART. EPA is also making it possible for interested parties, including environmental groups, to ignore their procedural obligation to voice objections under State law because they can wait to raise them when EPA acts on a proposed SIP. EPA chose not to participate in the BART permit process and the resulting appeals, despite knowing that the very NOX control equipment at issue in the regional haze FIP was being determined. Under the principles of comity, EPA should be barred from now addressing these issues at this late period. Under these circumstances, EPA should not be allowed to raise complaints with a BART permit for the first time in the federal proceeding. Failure to do so diminishes State law and puts parties like Basin Electric into a position where they must pursue State remedies to avoid finality under State law but find that such actions mean nothing in the end under the federal process.

Response: EPA disagrees with this comment. Nothing in the CAA sets some sort of “compelling circumstance” standard for disapproval of a SIP. Instead, we have the duty to ensure that regional haze SIP submittals meet the requirements of the Act and the RHR. See CAA Section 110(a)(2)(J) (SIP submittals must meet applicable requirements of Part C of title I, including visibility protection). We do not agree that we are prohibited from identifying deficiencies in the Wyoming SIP after the State rulemaking process is complete, and the commenter cites nothing in the Act to the contrary. Furthermore, many of the concerns raised in this action were communicated to the State in our comment letters and in numerous meetings with State officials. With respect to comments we have received from environmental organizations on our proposed action on Wyoming's SIP, the CAA does not require those organizations to participate in state processes.[34] EPA is taking actions specified under the CAA in partially approving and partially disapproving the Wyoming SIP. The CAA also specifies the responsibility of EPA to issue a FIP when states have not met their requirements under the CAA. EPA is promulgating this FIP to fill the regulatory gap created by the partial disapproval. Under the FIP, the State retains its authority to submit future regional haze SIPs consistent with CAA and RHR requirements; we do not discount the possibility of a future, approvable SIP submission that results in the modification or withdrawal of the FIP. This rulemaking does not change the distribution of power between the states and EPA.

Comment: BART applies to specific emission sources and requires consideration of facts applicable to specific source locations. Unlike a rule, or a SIP generally, a BART determination effectively adjudicates the specific rights and legal obligations of individual emissions sources. This typically entitles individual source owners to substantive procedural rights and remedies under state law when a BART determination is made. In Wyoming, for example, each individual source is required to apply for a BART Permit. Wyoming law affords the source being regulated with special opportunities to be heard, both as part of the public review of a permit application and, in the case of a permit, in an adjudicative hearing with opportunities to challenge factual determinations, call and question witnesses, and present evidence. When an applicant applies for a BART construction permit, the applicant is afforded the opportunity to present its own views and responses to comments to the state agency. If a permit is issued or denied, the applicant can appeal the permit decision to the Environmental Quality Council, which has statutory authority to amend, grant, modify, or deny the permit. Wyo. Stat. Section 35-11-802. This proceeding is conducted as a contested case, affording the applicant the right to cross-examine the Environmental Quality Council's technical experts regarding their BART assumptions and conclusions.

The applicant also can call its own experts and witnesses. Wyo. Dep't of Envtl. Quality Rules and Regulations, Wyo. Admin. Code ENV PP Ch. 2 Sections 1-14. With these procedures, BART permit applicants can challenge the cost estimates and assumptions underlying a BART permit decision, including making a showing, as Basin Electric does here by comment only, that EPA's consultants have ignored critical site-specific conditions.

EPA's effort to impose BART determinations by federal rulemaking impermissibly deprives source owners of these substantive procedural rights afforded under State law. This is one reason courts have taken a strong stance against EPA imposing specific control technologies through partial approval of a SIP. Leaving site-specific decisions in the hands of the states provides state-sponsored procedural rights for the individually regulated sources. See Virginia, 108 F.3d at 1406-10; Michigan v. Thomas, 805 F.2d 176, 186 (6th Cir. 1986); Bethlehem Steel Corp., 742 F.2d at 1035-37 (all holding that EPA may not render a state SIP more stringent than intended by the state by partial SIP approval or imposition of control technologies). A BART determination requires consideration of complex, case-specific control technologies and makes fact-dependent determinations for individual named sources, which effectively makes the federal BART determination an administrative order directed specifically at Basin Electric rather than a rule generally applicable to the public. Under these circumstances, EPA cannot order specific emission limits and consequent expensive control technologies without affording Basin Electric a hearing at which it can cross examine EPA's consultants. Basin Electric must also be given an opportunity to challenge EPA's interpretation of the facts. When EPA moves from a quasi-legislative function to a quasi-judicial function, as it has by making fact-based determinations for specific, named sources, it must provide the required procedural protections for those affected by its actions. See Start Printed Page 5070Londoner v. City & Cnty. of Denver, 210 U.S. 373, 386 (1908) (requiring an agency to provide notice and an adjudicative hearing for individuals suffering specific injury from an agency rule); compare Amoco Oil Co. v. EPA, 501 F.2d 722, 734-35 (D.C. Cir. 1974) (agency action was quasi-legislative because it did not rely on “findings of fact” and evidence to make determinations for a single source).

One administrative law expert designated the distinction between rule making and adjudication as “perhaps the most critical distinction in all of administrative law.” Gary Lawson, Federal Administrative Law 10 (American Casebook Series, Thomson-West 4th ed. 2007). It is an important distinction because it separates agency decisions that function as policy from those that make situational determinations. “A plain[ ] instance of administrative adjudication occurs where an administrative agency at one and the same time makes a rule and applies it to a concrete situation . . . The essential difference between legislation and adjudication is not that one looks to the future and the other to the past . . . What distinguishes legislation from adjudication is that the former affects the rights of individuals in the abstract and must be applied in a further proceeding before the legal position of any particular individual will be definitely touched by it; while adjudication operates concretely upon individuals in their individual capacity.” John Dickinson, Administrative Justice and the Supremacy of Law in the United States 16-21 (Harvard University Press 1927), quoted in Gary Lawson, Federal Administrative Law 10-11(American Casebook Series, Thomson-West 4th ed. 2007).

In the Proposed Rule, EPA makes specific factual findings about individual sources. EPA relies on its expert consultant Andover to draw specific factual conclusions about retrofit construction costs for Laramie River, yet it affords Basin Electric no opportunity to confront its expert over the Andover Report's error-filled findings. In order to provide due process, a specific party like Basin Electric who is singled out and subjected to EPA's fact-based determinations must be allowed “the right to support his allegations by argument however brief[,] and, if need be, by proof, however informal.” Londoner, 210 U.S. at 386. In the case of Laramie River, the requirement for a hearing is especially strong because “[t]he extent to which procedural due process must be afforded the recipient is influenced by the extent to which he may be `condemned to suffer grievous loss.' ” Goldberg v. Kelly, 397 U.S. 254, 262-63 (1970) (citing Joint Anti-Fascist Refugee Comm. v. McGrath, 341 U.S. 123, 168 (1951)).

EPA must afford these procedural rights to Basin Electric if EPA is going to assume control over site-specific BART determinations, rather than leave them to the states as Congress intended. Section 169A's directive that BART be determined by the states permits states to afford individual emissions sources the procedural and other rights that due process requires for site-specific regulation, and EPA must afford these same rights to source owners if it is going to federalize the BART program by rejecting all state determinations with which its technical consultants disagree.

Response: EPA disagrees with this comment. EPA's procedures did not deprive Basin Electric of due process. First, the comment confuses the issues by arguing that under State law Basin Electric has “substantive procedural rights” and that EPA's procedures somehow deprived Basin Electric of these. But due process under the Fifth Amendment does not require EPA to give exactly the same process that the State gave. The commenter provides no authority for the existence of something called a state “substantive procedural right” that the United States is bound by the Fifth Amendment to respect.[35] Instead, federal due process protects substantive fundamental rights and procedural rights if the claimant has a constitutionally protected life, liberty, or property interest. See U.S. Const., Amend. V (“nor be deprived of life, liberty, or property, without due process of law”). That the comment attempts to make a state procedure into a constitutionally protected interest by calling it a “substantive procedural right” is of no avail; the comment identifies no attribute of the state procedure that makes it into a constitutionally protected “life, liberty, or property” interest under either the text of the Fifth Amendment or the case law interpreting that Amendment. See Cleveland Bd. Of Educ. v. Loudermill, 470 U.S. 532, 541 (1985) (“[T]he Due Process Clause provides that certain substantive rights—life, liberty, and property—cannot be deprived except pursuant to constitutionally adequate procedures. The categories of substance and procedure are distinct.”). Nor does Basin Electric have a protected interest in the outcome of the State BART permit process. There is no “legitimate claim of entitlement” to that outcome, Board of Regents of State Colleges v. Roth, 408 U.S. 564, 577 (1972), as the State's BART determination was always subject to review by EPA under the CAA. In the end, what the Fifth Amendment does potentially protect is Basin Electric's property interest itself, not the State procedure. As we now explain, EPA's procedures were sufficient to satisfy the requirements of due process with respect to Basin Electric's property interest.

CAA section 307(d) specifies the procedures that EPA is required to follow in promulgating a FIP. Section 307(d) does not require adjudicatory hearings, nor does it require EPA to allow for cross-examination of EPA's consultants.[36] Additionally, the Administrative Procedure Act only requires adjudicatory hearings if a particular statute specifies that a rule must be made “on the record after an opportunity for an agency hearing.” [37] No such requirement is contained in section 307(d).[38] The Supreme Court has explained that courts face an extremely high burden in order to impose additional procedures beyond those specifically required by statute because “unwarranted judicial examination of perceived procedural shortcomings of a rulemaking proceeding can do nothing but seriously interfere with that process prescribed by Congress.” [39] EPA followed the Start Printed Page 5071procedures required by Congress in the CAA and EPA believes that no additional proceedings are warranted.

Moreover, Congress specifically contemplated and rejected a cross-examination requirement for public hearings in section 307.[40] The House bill contained an opportunity to cross-examine those who made oral presentations at the public hearing. During Conference Committee, this was deleted and replaced with a requirement that the rulemaking record remain open for thirty days after public hearing to allow interested parties to submit rebuttal and supplemental information.[41]

The comment cites Goldberg v. Kelly, 397 U.S. 254, 262-63 (1970) and argues that Basin Electric, like the welfare recipient in Goldberg, has an especially strong claim to an evidentiary hearing prior to EPA's final rulemaking because Basin Electric may be “condemned to suffer grievous loss.” The comment fails to explain why the private interest of Basin Electric here is identical to the Goldberg welfare recipient's private interest in an evidentiary hearing before the termination of welfare benefits. The comment also does not examine the factors set out in Mathew v. Eldridge, 424 U.S. 319 (1976),[42] for determining what due process requires, and so does not provide any reason for EPA to think that the procedures here were inadequate. In particular, the comment provides no basis to think that EPA's procedures created a serious “risk of an erroneous deprivation” of Basin Electric's interest and that there would be any “probable value” to cross-examination. With respect to the alleged errors referred to in the comment, Basin Electric has made its arguments as to why they are errors and EPA has responded why they are not. If Basin Electric thinks EPA's responses are inadequate, then Basin Electric may seek judicial review of EPA's action under section 307(b) of the Act. The risk of erroneous deprivation appears small, and Basin Electric's comment gives no reason to think otherwise. Basin Electric's comment also does not identify any particular value to cross-examination in this context. As the comment admits, the matters here are ones of technical judgment; they are not (for example) eyewitness accounts that might benefit from cross-examination.

EPA also notes that the comment fails to discuss “the Government's interest, including . . . the fiscal and administrative burdens” that cross-examination would entail. Eldridge alternatively identified this third factor as “the public interest.” Eldridge, 424 U.S. at 347. In considering the burdens imposed by a full adjudicatory hearing on the Government and the public, the Tenth Circuit Court of Appeals stated (albeit before Eldridge, so not in the context of applying the Eldridge factors):

Unending procedure could be produced by an adjudicatory hearing. This could bring about unending delay which would not only impede but completely stifle congressional policy. We do not, of course, condemn the trial court's concern for the rights of [the petitioner]. Those rights are important and the court should be sensitive to them, but those rights are not of such magnitude as to overcome congressional policy and the rights of the remainder of the community.

Anaconda Co. v. Ruckelshaus, 482 F.2d 1301, 1307 (10th Cir. 1973). The comment gives EPA no reason to think otherwise.

With respect to the comment's invocation of the BiMetallic-Londoner distinction between rulemaking and adjudication, it is not clear that Londoner applies here, where the interests of many parties are at stake. See Anaconda, 482 F.2d at 1306 (“The fact that Anaconda alone is involved is not conclusive on the question as to whether the hearing should be adjudicatory, for there are many other interested parties and groups who are affected and are entitled to be heard. So the guidelines enunciated by Mr. Justice Holmes in Bi-Metallic Investment Co. v. State Board of Equalization are not applicable.”) (citation omitted). Even if the distinction does apply, due process does not per se require a full adjudicatory hearing. As the comment admits, what due process does require is that a person “have the right to support his allegations by argument, however brief: and, if need be, by proof, however informal.” Londoner v. City & Cnty. of Denver, 210 U.S. 373, 386 (1908). Thus the “core of due process is the right to notice and a meaningful opportunity to be heard.” [43] With respect to whether a full evidentiary hearing is required, “differences in the origin and function of administrative agencies preclude wholesale transplantation of the rules of procedure, trial, and review which have evolved from the history and experience of courts. The judicial model of an evidentiary hearing is neither a required, nor even the most effective, method of decision making in all circumstances.” Eldridge, 424 U.S. at 348 (citations and quotations omitted).

EPA believes Basin Electric was afforded a meaningful opportunity to be heard and present evidence to EPA in support of its position. EPA notified the public of its proposed rule, held a public hearing, and accepted public comments for a period of 60 days.[44] In an effort to provide a greater opportunity for public comment on the proposed rule, EPA held two additional public hearings and extended the comment period to 75 days, which goes beyond the procedures required by the CAA. [45] Basin Electric submitted extensive comments prior to the first comment deadline, participated in two public hearings, and submitted additional comments during the extended public comment period.[46] Basin Electric took full advantage of its opportunity to be heard and was not denied due process.

Comment: Section 169A requires the State to take into consideration five different factors when making its BART determination. 43 U.S.C. 7491(g)(2). But these factors “were meant to be considered together” to arrive at a single judgment committed to the State: A BART emission limit. American Corn Growers, 291 F.3d at 6. Moreover, only Wyoming—not EPA—is entitled to determine the weight and significance to assign costs, feasibility, and visibility improvements. 70 FR 39123 (“The State makes a BART determination based on the estimates available for each criterion, and as the CAA does not Start Printed Page 5072specify how the State should take these factors into account, the States are free to determine the weight and significance to be assigned to each factor.”); see also 40 CFR Part 51, App. Y, Section IV.D.5.

By applying a different assessment of costs and visibility than those employed by Wyoming in its BART determination, and assuming that these assessments mandate a different BART outcome, EPA's proposed FIP rejects the State's determinations on cost, feasibility, and visibility improvement without considering whether, taken together, the five statutory factors would compel a different result than the one reached by Wyoming. The net result is a decision imposing a different BART choice than that selected by the State by splitting the statutory factors and giving them separate and independent determinative significance—the same legal error EPA made in American Corn Growers. The “splitting of the statutory factors is consistent with neither the text nor the structure of the statute.” 291 F.3d at 6.

Wyoming must therefore be afforded an opportunity to reconsider its BART determination before EPA imposes a FIP. This is necessary to preserve State primacy in the BART determination. States “determine what is too costly (and what is not) for a particular source.” Am. Corn Growers, 291 F.3d at 6-7. The actual BART determination flows not from any one of the statutory factors, but instead from consideration of all of them together. That is why it is erroneous for EPA to impose its own BART choice without explaining how it reached that choice upon consideration of all five statutory factors. If EPA acts to correct alleged errors in the State's cost assessment or visibility modeling, EPA must remand the statutory evaluation back to the State. Section 110(c) contemplates that States should be given an opportunity to correct any “deficiencies,” and this statutory opportunity should not be taken from the State as a result of self-imposed consent decree deadlines. Doing so destroys State primacy in the BART determination.

It also results in a BART determination from EPA that is not informed and explained by an independent assessment of the five statutory factors. EPA's failure to remand the BART determination back to the State therefore results in neither the State nor EPA making a BART assessment that considers all of the statutory factors together. While Basin Electric acknowledges that the Tenth Circuit Court of Appeals recently reached a different conclusion in Oklahoma v. EPA, 723 F.3d 1201 (10th Cir. 2013), that case is not yet final and that Court was not presented with, and did not consider, the fundamental problem associated with EPA's effort to make one of the five statutory factors outcome determinative. EPA cannot cause an outcome in which no agency has actually complied with the statute, which is what happens when EPA simultaneously disapproves the State's BART assessment on one or two statutory factors and then imposes a different BART assessment based upon cost and visibility factors combined with the State's prior consideration of the other factors, as EPA does here. This is not a procedural error, but rather an error that results in no agency—neither the State nor EPA—actually complying with the statute by considering all five statutory factors together before arriving at a BART emission limit.

Response: EPA does not agree with this comment. The RHR and the BART Guidelines allow the reviewing authority (State, Tribe, or EPA) the discretion to determine how to weigh and in what order to evaluate the statutory factors (cost of compliance, the energy and non-air quality environmental impacts of compliance, any existing pollution control technology in use at the source, the remaining useful life of the source, and the degree of improvement in visibility which may reasonably be anticipated to result from the use of such technology), as long as the reviewing authority justifies its selection of the “best” level of control and explains the CAA factors that led the reviewing authority to choose that option over other control levels.[47] In this action, having disapproved the State's BART determinations for NOX at five units, “all of the rights and duties that would otherwise fall to the State accrue instead to EPA.” [48] This includes a significant degree of discretion in deciding how to weigh the five factors, so long as that weighing is accompanied by reasoned explanation for adopting the technology selected as BART, based on the five factors, and in accordance with the BART Guidelines. EPA has provided a detailed explanation of our BART evaluation process and five-factor analyses in our proposal, and elsewhere in this final notice. We have weighed the potential energy and non-air environmental quality impacts of the various control options along with the other statutory factors in our BART analyses. We have not, as the commenter surmises, approved the State's assessment of certain factors and disapproved the assessment of others, replacing just the factors we have disapproved. Instead, for those NOX BART determinations we are disapproving, we have disapproved them in their entirety. Then EPA independently assessed and weighed the five factors. That we adopted the State's assessment of certain factors as our own does not change this. Thus the split in authority that the commenter suggests simply has not occurred.

We also disagree that our proposal is inconsistent with the American Corn Growers decision. In American Corn Growers, the petitioners challenged the original RHR because, among other things, the RHR treated one of the five statutory factors differently than the others by requiring states to consider the degree of visibility improvement from imposing BART on a group of sources rather than on a source-specific basis.[49] The court concluded that such a requirement could force states to apply BART controls at sources without evidence that the individual sources contributed to visibility impairment at a Class I area, which encroached on states' primary authority under the regional haze provisions to determine which individual sources are subject to BART and what BART controls are appropriate for each source.[50] Therefore, the court vacated the visibility improvement part of the original RHR as contrary to the statute.[51] Contrary to some commenters' suggestions, however, the American Corn Growers decision did not address EPA's authority to reject a state's BART determinations for failure to conform to the CAA, the RHR, or the BART Guidelines.

Finally, as explained elsewhere in this final rule, we have the authority to promulgate a FIP concurrently with a disapproval action.

Comment: EPA's FIP is subject to APA review. Accordingly, it cannot withstand judicial scrutiny if it is arbitrary, capricious, an abuse of discretion, or not in accordance with the law. See 5 U.S.C. 706(2)(A); Olenhouse, 42 F.3d at 1574. More generally, a court will set it aside “if the agency relied on factors which Congress has not intended for it to consider, entirely failed to Start Printed Page 5073consider an important aspect of the problem, offered an explanation for its decision that runs counter to the evidence before the agency, or is so implausible that it could not be ascribed to a difference in view or the product of agency expertise.” State Farm, 463 U.S. at 43.

A court reviewing agency action under the APA must “ascertain whether the agency examined the relevant data and articulated a rational connection between the facts found and the decision made.” Olenhouse, 42 F.3d at 1574 (citing State Farm, 463 U.S. at 43) (footnote omitted). A reviewing court also must review the agency's explanation to “determine whether the agency considered all relevant factors and whether there has been a clear error of judgment.” Id. (citing, inter alia, Citizens to Preserve Overton Park, Inc. v. Volpe, 401 U.S. 402, 416 (1971)). The court “`should not attempt itself to make up for . . . deficiencies'” in the agency's reasoning and “may not supply a reasoned basis for the agency's action that the agency itself has not given.” Id. at 1574-75 (quoting State Farm, 463 U.S. at 43) (emphasis removed).

As a result, “`an agency's action must be upheld, if at all, on the basis articulated by the agency itself,” and “the grounds upon which the agency acted must be clearly disclosed in, and sustained by, the record.” Id. at 1575 (quoting State Farm, 463 U.S. at 50). In its decision, “[t]he agency must make plain its course of inquiry, its analysis and its reasoning.” Id. Moreover, its action must be “supported by the facts in the record.” Id. This means the action must be supported by “substantial evidence,” i.e., “`enough to justify, if the trial were to a jury, a refusal to direct a verdict when the conclusion to be drawn is one of fact.'” Id. (citation omitted). In addition to providing a basis for invalidating the agency action, an agency's failure to fully explain and support its reasoning warrants a court's grant of less deference to the agency's decisions. See, e.g., Achernar Broad. Co. v. FCC, 62 F.3d 1441, 1447 (D.C. Cir. 1995) (“no deference is due when the agency has stopped shy of carefully considering the disputed facts”); NLRB v. P*I*E Nationwide, Inc., 923 F.2d 506, 518 n.16 (7th Cir. 1991) (“deference given to an agency is not granted freely, it is purchased; the agency must exercise its touted expertise and “explain the rationale and factual basis for its decision”) (citation omitted).

Although a court generally will defer to an agency's experts when the agency acts within its area of expertise, a court will not do so and will invalidate the agency's action where its expert's decisions were arbitrary and capricious. See, e.g., Garvey, 256 F.3d at 1036 (agencies can rely on their own experts only “so long as their decisions are not arbitrary and capricious”) (citation omitted). See also NetCoalition v. SEC, 615 F.3d 525, 539 (D.C. Cir. 2010) (“[W]e do not defer to the agency's conclusory or unsupported suppositions.”) (citation omitted); Brower v. Evans, 257 F.3d 1058, 1067 (9th Cir. 2001) (“The deference accorded to an agency's scientific or technical expertise is not unlimited. The presumption of agency expertise can be rebutted when its decisions, while relying on scientific expertise, are not reasoned.”) (citation omitted); Nat. Resources Defense Council, 725 F.2d at 768, 771 (the court owed EPA no deference where the agency “complete[ly] fail[ed] to consider the criteria that should inform [its decision]”). Similarly, an agency can rely on a model “only so long as it `explains the assumptions and methodology used in preparing the model' and `provides a complete analytical defense' should the model be challenged.” Appalachian Power Co. v. EPA, 249 F.3d 1032, 1052 (D.C. Cir. 2001) (citation and brackets omitted). See also Sierra Club v. Costle, 657 F.2d 298, 333 (D.C. Cir. 1981) (although computer modeling undoubtedly “is a useful and often essential tool,” an “agency must sufficiently explain the assumptions and methodology used in preparing the model” and must “provide a complete analytic defense of its model (and) respond to each objection with a reasoned presentation”) (internal quotation marks omitted), rev'd on other grounds, 463 U.S. 680 (1983); id. (there must be “a rational connection between the factual inputs, modeling assumptions, modeling results and conclusions drawn from these results”). Here, in promulgating its FIP, EPA was required to do the same thing Wyoming did: determine BART by “tak[ing] into consideration” the five statutory factors, including the costs of compliance, the energy and non-air quality environmental impacts of compliance, any existing pollution control technology in use at the source, the remaining useful life of the source, and the degree of improvement in visibility that may reasonably be anticipated to result from the use of the technology. CAA Section 169A(g)(2), 42 U.S.C. 7491(g)(2). As the D.C. Circuit explained in American Corn Growers, “the factors were meant to be considered together” in determining BART, as “[t]he language of section 169A(g)(2) can be read in no other way.” 291 F.3d at 6.

Accordingly, in order to comply with the CAA and withstand APA review, EPA must fully explain how it assessed and weighed the five BART factors together, and it must support that explanation with record facts. EPA has failed to do so. Additionally, the same regulations EPA promulgates for state BART determinations must also apply to BART determinations made by EPA. See CAA Section 169A(b)(2)(A), 42 U.S.C. 7491(b)(2)(A). Indeed, it would be arbitrary and capricious for EPA to require a state to follow certain specific guidelines in making a BART determination, yet to not itself follow those same guidelines in making that same determination after taking it out of the state's hands. Moreover, EPA has suggested that the BART Guidelines and Cost Manual are mandatory provisions that must be followed in order to comply with the CAA.

Response: We disagree with this comment. As detailed elsewhere in this document and documented in the supporting record, EPA applied the BART statutory factors and BART Guidelines to each and every BART unit that is covered under this rulemaking; fully considered all significant comments submitted on the proposed notices and incorporated those comments as appropriate; provided basis for the decisions; applied models that are specified in the BART Guidelines (thus, the opportunity for commenters to challenge the specified models has long passed); developed and provided detailed explanations regarding EPA's model inputs and settings; and rationally applied the modeling results to the final determinations in applying the BART and reasonable progress factors. The comment does not identify any deficiency in any portion of this.

Comment: Wyoming developed a SIP that established reasonable progress toward meeting the national goal for regional haze as required under the CAA Section 169A(a)(1). EPA's establishment of a 2064 goal and glide path requires incremental visibility improvement for successive planning periods. EPA also clearly explains in these requirements that the glide path and 2064 target date are not binding. This provides considerable latitude to the individual states that are responsible to develop a regional haze SIP that makes reasonable progress in a way that works to achieve the visibility goals over time.

The State developed and submitted a plan that would make substantial progress in reducing haze at the affected Class I areas. The State followed the process in the EPA's Regional Haze Start Printed Page 5074Guidelines, yet because it came to a different conclusion than EPA, the plan was rejected and replaced with EPA's FIP.

By rejecting the State's reasonable approach, EPA has ignored its own requirements and guidance. EPA's issuance of a FIP not only ignores the flexibility and authority granted the State, it also ignores EPA's guidance for establishing reasonable control requirements.

Response: EPA disagrees with this comment. While the RHR does not require states to achieve the URP, when a state's selected RPGs do not meet the URP, the state must demonstrate, based on the four reasonable progress factors, that meeting the URP is not reasonable and that the selected RPGs are reasonable. 40 CFR 51.308(d)(1)(ii). As discussed elsewhere, the State did not appropriately consider the four reasonable progress factors for Dave Johnston Units 1 and 2, and to the extent that the State relied on its BART determinations to show reasonable progress for those sources, we have disapproved some of those BART determinations. While the comment states that EPA “ignored its own requirements and guidance,” the comment does not cite any particular requirement that EPA purportedly violated.

Comment: The EPA proposal is deficient in large measure because the EPA has identified what it views as deficiencies in the Wyoming SIP and, rather than ordering reconsideration of all relevant factors with improved data, has created a FIP that suffers from analytical errors and arrogates the EPA's role in development and review of SIPs. If the EPA was convinced Wyoming's cost estimates were in error, it should have directed corrections, rather than substituting other flawed data and its own judgment. Indeed, it is apparent the EPA is not committed to maintaining the CAA's deference to states' authority to formulate workable haze plans. Otherwise, the EPA would have required Wyoming to correct perceived cost estimate errors and subsequently reevaluate BART factors. The EPA instead, substituted its own errors and performed its own evaluation in pursuit of its own goals.

Another commenter argued that EPA should not impose a FIP until it has issued a final rule disapproving the Wyoming regional haze SIP. 42 U.S.C. 7410(c)(1)(B). EPA should first conduct a rulemaking and take public comment on the Wyoming regional haze SIP submission, issue its determination on the regional haze SIP, and then seek input from the State. (See 42 U.S.C. 7410(c)(1)(B); see also 42 U.S.C. 7607(d)(B) (rulemaking provisions apply to “the promulgation or revision of an implementation plan by the Administrator under section 7410(c)”) Otherwise, EPA removes the State from its assigned role as the one determining BART.

The facts here illustrate this problem. EPA initially agreed with Wyoming's BART determinations for Naughton Units 1 and 2, and Dave Johnston Unit 3. EPA then reversed itself, supposedly on the basis of new cost and visibility information. Without offering Wyoming any chance to review the new information and issue a new BART determination, EPA disapproved Wyoming's BART determination for these units, and instituted new BART determinations for these units through a regional haze FIP. EPA's failure to provide Wyoming an opportunity to review this new information, and address it through a revised BART determination, violates the applicable CAA statutes.

The CAA defines a FIP as a plan (or portion thereof) promulgated by the (EPA) Administrator to fill all or a portion of a gap or otherwise correct all or a portion of an inadequacy in a SIP. 42 U.S.C. 7602(y). Until EPA first assesses the Wyoming regional haze SIP, develops a proposed rule to approve or disapprove the Wyoming regional haze SIP, solicits and receives public comment on that proposed rule, considers the comments and information, and takes final action on whether (and to what extent) to approve the Wyoming SIP, EPA cannot know whether there is a “gap” in the Wyoming regional haze SIP that needs to be filled or whether (and to what extent) there is an “inadequacy” in the Wyoming regional haze SIP that needs to be corrected. Id. Moreover, EPA's failure to obtain public comments prior to proposing a regional haze FIP deprives Wyoming of an opportunity to correct any “deficiencies” identified by EPA. Here, where EPA claims to have obtained new cost and visibility information but did not allow Wyoming an opportunity to review and act on the new information, EPA's final determination regarding the Wyoming regional haze SIP ignores the State's authority under the CAA (including the regulatory programs implicated by CAA Section169A) to design and implement plans to control air pollution control within its borders. (See 42 U.S.C. 7401(a)(3).) Therefore, EPA illegally seeks to impose its regional haze FIP and should withdraw the same.

Earlier comments argued that EPA cannot impose a regional haze FIP until it has issued a final rule disapproving Wyoming's regional haze SIP. 42 U.S.C. 7410(c)(1)(B) mandates that disapproval of all or part of a SIP is a prerequisite to promulgation of a FIP. EPA must first conduct a rulemaking and take public comment on Wyoming's regional haze SIP submission, issue its determination on the regional haze SIP, and then proceed, or not, with promulgation of a regional haze FIP. (See 42 U.S.C. 7410(c)(1)(B); see also 42 U.S.C. 7607(d)(B) (rulemaking provisions apply to “the promulgation or revision of an implementation plan by the Administrator under section 7410(c)”)

Response: We disagree with this comment. We have the authority to promulgate a FIP concurrently with a disapproval action. Nowhere in the CAA is there language that limits EPA's authority to simultaneously propose a FIP and propose disapproval of a state's SIP where there has been a prior finding of a failure to submit. This timing for FIP promulgation is authorized under CAA section 110(c)(1). As has been noted in past FIP promulgation actions, the language of CAA section 110(c)(1), by its terms, establishes a two-year period within which we must promulgate the FIP, and provides no further constraints on timing. See, e.g., 76 FR 25178, at 25202. Wyoming failed to submit the 40 CFR 51.309(g) plan elements by December 17, 2007, as required under the CAA and our implementing regulations. Two years later, Wyoming still had not submitted these required plan elements. When we made the finding in 2009 that Wyoming had failed to submit these regional haze SIP elements (see 74 FR 2392), that created an obligation for us to promulgate a FIP by January 2011. We are exercising our discretion to promulgate the FIP concurrently with our disapproval action because of the applicable statutory deadlines requiring us at this time to promulgate regional haze BART determinations to the extent Wyoming's BART determinations are not approvable. In these concurrent SIP/FIP actions, if comments or other information cause us to reconsider portions of our proposed disapproval, and instead approve additional portions of Wyoming's SIP, we can readily adjust our FIP accordingly by not finalizing the FIP portions that are no longer needed, as, indeed we are doing in this case. Thus, the supposed procedural problem the comment identifies simply does not exist.

With respect to the argument that the CAA requires EPA, before promulgating a FIP, to give additional opportunities to Wyoming to address the deficiencies that EPA has identified, in fact the Start Printed Page 5075opposite is true. Under section 110(c)(1) of the CAA, EPA must promulgate a FIP within 2 years of a finding of failure to submit a required SIP submittal. As explained above, the requirement for a FIP promulgation in today's action was triggered by a finding published on January 15, 2009 (74 FR 2392), that Wyoming (among other states) had failed to make a submittal to address the requirements of 40 CFR 51.309(g). Thus, EPA had an obligation to promulgate a FIP for the requirements of 40 CFR 51.309(g) by January 15, 2011, unless the State submitted and EPA approved a SIP addressing the deficiency. Although we are approving portions of Wyoming's SIP that meet the requirements of 51.309(g), we are disapproving other portions and, therefore, are still under an obligation to promulgate a FIP for those portions. In considering a similar argument to that made by the commenter, the Tenth Circuit Court of Appeals has stated:

Once the EPA issued findings that Oklahoma failed to submit the required SIP under the Regional Haze Rule, the EPA had an obligation to promulgate a FIP. The statute itself makes clear that the mere filing of a SIP by Oklahoma does not relieve the EPA of its duty. And the petitioners do not point to any language that requires the EPA to delay its promulgation of a FIP until it rules on a proposed SIP. As the EPA points out, such a rule would essentially nullify any time limits the EPA placed on states. States could forestall the promulgation of a FIP by submitting one inadequate SIP after another.

Oklahoma v. EPA, 723 F.3d 1201, 1223 (10th Cir. 2013).

Finally, as explained elsewhere, under the FIP, the State retains its authority to submit future regional haze SIPs consistent with CAA and RHR requirements; which may result in the modification or withdrawal of the FIP.

Comment: The CAA and the RHR provide substantial discretion to states to determine how best to make reasonable progress toward achieving natural visibility conditions in designated areas. Reasonable progress—the touchstone of the regional haze program—is a flexible benchmark. See 42 U.S.C. 7491(g)(1). In recognition of this overarching flexibility and the need to account for local conditions, Congress directed EPA to allow states discretion in how they determine the BART for improving visibility. Id. Section 7491(b)(2)(A); Am. Corn Grower Ass'n v. EPA, 291 F.3d 1, 8 (D.C. Cir. 2002) (“Congress intended the states to decide which sources impair visibility and what BART controls should apply to those source.”); see also 40 CFR 51.308(e)(1)(ii)(A).

Against this backdrop of state discretion, the CAA requires SIPs to include: generally, “such emission limits, schedules of compliance and other measures as may be necessary to make reasonable progress toward meeting the national goal [of natural visibility conditions in national parks and wilderness areas],” 42 U.S.C. 7491(b)(2); “a long-term (ten to fifteen years) strategy for making reasonable progress toward meeting the national goal,” id. Section 7491(b)(2)(B); and more specifically, a plan for particular sources to “procure, install, and operate, as expeditiously as practicable (and maintain thereafter) the best available retrofit technology,” id. Section 7491(b)(2)(A).

Response: The CAA gives states substantial but not unfettered discretion in determining BART and reasonable progress. We have already largely addressed the assertions in this comment in our responses to comments on our legal authority. Furthermore, as a hypothetical example, EPA would not defer to a state determination that the remaining useful life of a source is one year if relevant evidence indicates the remaining useful life is 20 years. Limits on state discretion are inherent in the CAA and our regulations; otherwise, states would be free to reach decisions that are arbitrary and capricious or inconsistent with the purpose behind the CAA and EPA's regulations. As we have stated, while we have approved much of Wyoming's SIP submittal, those elements which we have disapproved and for which we are finalizing a FIP thwart the goals stated by Congress in CAA section 169A and underlying the RHR. Those statutory and regulatory provisions cannot be simply dismissed under the mantle of state discretion.

Comment: On May 28, 2009, Wyoming published its BART application analyses for the PacifiCorp and Basin Electric facilities subject to BART. Wyoming solicited public comments on the analyses and to that end held public hearings. EPA commented on Wyoming's analyses on August 3, 2009. EPA was fully aware of Wyoming's BART proposals, but, at that time EPA gave no indication that Wyoming's BART proposals violated the CAA or were unreasonable.

Both PacifiCorp and Basin Electric ultimately challenged Wyoming's BART determinations before the Wyoming Environmental Quality Council. See Appeal & Pet. for Review of BART Permits, In re BART Permit Nos. MD-6040 and MD-6042, No. 10-2801 (Wyo. Envtl. Quality Council Feb. 26, 2010) (PacifiCorp Petition); Appeal & Pet. for Review, In re Basin Electric Power Coop., No. 10-2802 (Wyo. Envtl. Quality Council March 8, 2010) (Basin Petition). The Environmental Quality Council is an independent administrative body charged with adjudicating issues arising under Wyoming environmental law, including BART determinations. See Wyo. Stat. Ann. Sections 35-11-111, 112.

Both Basin Electric and PacifiCorp served their petitions for review on EPA Region 8. EPA was again fully apprised of Wyoming's final BART decisions, as well as the appeals of those decisions. EPA elected not to participate in those proceedings, and, again, provided no indication that EPA viewed Wyoming's BART decisions as invalid.

After filing motions for summary judgment, PacifiCorp and Basin Electric both ultimately settled their litigation with the State. The Environmental Quality Council approved the settlements after providing an opportunity for public comment. EPA did not comment on the settlement agreements. Because no aggrieved person appealed the Council's decision approving the settlements, the permit decisions became final by operation of law. Wyoming therefore incorporated the BART permits into its SIP.

Years later, when EPA proposed action on Wyoming's SIP, EPA raised for the first time its disagreement with the BART decisions that PacifiCorp, Basin, and Wyoming had already litigated to conclusion. Because EPA had the opportunity to participate in the litigation and elected not to, EPA is now precluded from collaterally attacking those permit decisions. See, e.g., ADEC, 540 U.S. at 490 n.14. To conclude otherwise—that EPA can forgo participation in state adjudications only to later attack the conclusions of those state processes—is to give EPA the power to nullify state court judgments. Id. at 1015 (Kennedy, J., dissenting). Congress did not intend to so empower EPA to turn federalism on its head through the regional haze program.

Response: EPA disagrees with this comment. First, the comment does not identify any way in which EPA is precluded from exercising its authority and duty under the CAA to ensure that SIP submittals meet the requirements of the Act. The notion that a state BART determination is insulated from the requirements of the Act merely because the state has an administrative appeal process is contrary to the Act itself as well as the Supremacy Clause of the U.S. Constitution. Had Congress wanted to require EPA to participate in state rulemaking or permit processes, Congress would have explicitly stated this in the Act. With respect to the Start Printed Page 5076 ADEC dissent, it is just that, a dissent. Even if the dissent were somehow relevant, EPA is not nullifying a state court judgment. The Wyoming Environmental Quality Council is not within the State judicial branch. It is an executive agency. The members are appointed by the Governor and serve at the Governor's pleasure. See Wyo. Stat. Ann. Section 35-11-111(a) (“Council members shall be appointed by the governor with the advice and consent of the senate. The governor may remove any council member as provided in W.S. 9-1-202.”); Section 9-1-202(a) (“[A]ny person may be removed by the governor, at the governor's pleasure, if appointed by the governor to serve . . . as a member of a state board or commission.”).

Furthermore, EPA's comments to Wyoming on its proposed SIP and BART permits, which are in the docket for this action, emphasized that we would only come to a final conclusion regarding the adequacy of Wyoming's BART determinations when we acted on Wyoming's regional haze SIP revision, through public notice and comment rulemaking. While we may have been silent on some issues, silence from the EPA does not signify implicit approval. Any lack of participation by the EPA in the state administrative appeal proceeding or failure to register an objection to the settlement agreement is not an indication that a state's proposed BART determination will be approved following its submittal as part of a larger regional haze SIP, as discussed in greater detail elsewhere in this document. Wyoming is required to adopt a final BART determination as part of its regional haze SIP. As explained elsewhere in this document, once a state submits a SIP to the EPA, we are authorized to approve, partially approve, or disapprove the SIP, and we have the duty to assure that the SIP submittal complies with the requirements of the Act. The statutory scheme explicitly provides for this.

Alaska Depart of Environmental Conservation v. Environmental Protection Agency, 540 U.S. 461 (2004) concerned EPA's response to ADEC's issuance of a permit to a mine that provided, as BACT, unreasonably low NOX controls. Accordingly, EPA issued three orders prohibiting ADEC from granting the permit unless it satisfactorily documented its reasoning behind its BACT determination. The Ninth Circuit held the three orders were a proper exercise of EPA's authority and discretion. The Supreme Court affirmed. EPA agrees with the commenter that EPA made representations to the Court stating the need to accord “appropriate deference” to states' determinations. EPA also agrees that we made the representation that we have never asserted our authority to override a state-court judgment, and therefore, the fear that EPA will threaten state courts' independence is unfounded.

While EPA did make these representations, these representations are not inconsistent with EPA's decision to disapprove Wyoming's BART determination for Laramie River Station. As explained above, we are not overriding a state-court judgment. Furthermore, the notion that a state administrative appeal process can insulate a BART determination from federal requirements itself “turns federalism on its head.” See U.S. Constitution, Art. VI, cl. 2 (supremacy clause).

In this instance, some of Wyoming's BART determinations were unreasonable in terms of cost effectiveness and other factors as detailed elsewhere in this document (detailed descriptions of the cost assumption are described in the comments specific to the units elsewhere in this document).[52] Finding Wyoming's BART determinations to be unreasonable is a “restrained and moderate” use of EPA's statutory authority. See 540 U.S. at n.14. Following EPA's issuance of orders to ADEC for failing to establish a reasonable BACT, the Court noted, “Only when a state agency's BACT determination is `not based on a reasoned analysis' . . . may EPA step in to ensure that the statutory requirements are honored.” 540 U.S. 461, 490. In the case of Wyoming's BART determinations, EPA adhered to a similar role. Upon finding some of Wyoming's BART determinations unreasonable, EPA disapproved those determinations and proposed an alternative standard.

EPA continues to acknowledge the importance of significant deference to state authorities regarding their BART determinations since they are in the best position to make these determinations given their close familiarity with the unique characteristics of their particular area. This structure encourages cooperative federalism, a principle that underlies the CAA. However, this “initial responsibility” does not permit the state to make unreasonable BART determinations. See 540 U.S. at 464. EPA is not using its authority to disapprove part of a state's SIP as a way to override legitimate administrative litigation reached under state law. Rather, we are enforcing a requirement of the CAA concerning anthropogenic impairment of visibility by ensuring that reasonable BART controls are considered. State adjudicative processes are not threatened because states are free to use these processes to reach their own BART determination, provided that this determination is reasonable and consistent with the CAA.

Comment: Nowhere does the Act command national consistency in BART cost estimates and, to the contrary, by allowing states to make individualized BART determinations, Congress demonstrated that consistency was not intended to be a component of the regional haze program, save for the uniform objective of attaining natural visibility conditions. The commenter indicated that the RHR takes the same approach, allowing states wide discretion to conduct BART analyses, and that the BART Guidelines encourage states to take into account site-specific conditions that impact costs. In light of these authorities, the commenter believes that the EPA cannot disapprove the State's cost analyses simply because they do not fit within the EPA's preferred vision of national uniformity.

Another comment argued that EPA claimed that the State failed to follow the CCM, and the EPA supported this claim by quoting the CCM as saying that the EPA prefers consistency in control cost estimates (78 FR 34749). The CAA, the RHR, the BART Guidelines, and the fact that different sources have vastly different designs belie the EPA's preference for “consistency.” Nowhere does the Act command national consistency in BART cost estimates and, to the contrary, by allowing states to make individualized BART determinations, Congress demonstrated that consistency was not intended to be a component of the regional haze program, save for the uniform objective of attaining natural visibility conditions. The commenter indicated that the RHR takes the same approach, allowing states wide discretion to conduct BART analyses, and that the BART Guidelines encourage states to take into account site-specific conditions that impact costs. In light of these authorities, the commenter believes that the EPA cannot disapprove the State's cost analyses simply because they do not fit within the EPA's preferred vision of national uniformity.

Response: As we explain in our response to other comments in the legal issue section, we have authority to assess the reasonableness of a state's Start Printed Page 5077analysis of costs; and a state's discretion must be reasonably exercised in compliance with the applicable requirements. While we agree that site-specific challenges must be identified and factored into the cost effectiveness analysis, the SIP elements disapproved elsewhere in this document items are not “site-specific conditions,” but rather use of the wrong costing methodology and improper categorization of costs, as well as other issues. An erroneous analysis of costs, whether due to methodological or to data flaws, prevents a state from conducting a meaningful consideration of the cost of compliance factor. North Dakota v. U.S. EPA, 730 F.3d 750, 761 (8th Cir. 2013).

EPA is not relegated to a ministerial role. Id. We have not replaced cost estimates, modeling analyses and other SIP elements submitted by the State solely for the purpose of ensuring consistency across states. When a state or source puts forward costs estimates that are atypical, it is reasonable for us to scrutinize such estimates more closely to determine whether they are reasonable or inflated. Also, given that the assessment of costs is necessarily a comparative analysis and one marker of reasonableness, it is reasonable to insist that certain standardized and accepted costing practices be followed absent unique circumstances. Such consistency is particularly relevant for BART determinations at fossil-fuel fired power plants having a capacity in excess of 750 MW, which must be made pursuant to the BART Guidelines.[53] To the extent a BART determination for such a power plant is plainly inconsistent with EPA-approved determinations for similar sources, it is more likely to be inconsistent with the RHR and the BART Guidelines and therefore to warrant greater scrutiny for compliance with the applicable requirements.

Comment: Basin Electric submits with these comments an updated cost estimate for SNCR and SCR emission controls at Laramie River Station. That report states that in Sergeant & Lundy's opinion SNCR would likely achieve a 48% reduction from EPA's input emission rate. However, when it made its BART determination the State did not have the benefit of this report and made its judgment based on the best information available at the time. EPA, in its August 3, 2009 comments on Wyoming's BART permit for Laramie River Station, stated that it estimated that “SNCR can reduce NOX by 40%-50% for most large boilers (EPA Air Pollutions Control Cost Manual, 2002, Sixth ed., EPA-452-02-001. Section 4.2, Chapter 1, pg. 1-3.).” States are entitled to rely on information available at the time they make BART determinations, and EPA may not disapprove a state's BART based on information that becomes available later. This principle seems particularly appropriate when at the time EPA itself asserts the bona fides of information similar to that relied upon by the State.

Response: We disagree with this comment. EPA is required to take new information submitted as part of this rulemaking into consideration. Indeed, EPA has taken into consideration the updated cost estimate information submitted by Basin Electric for SNCR and SCR at Laramie River Station, which was not available to Wyoming. See Sierra Club v. EPA, 671 F.3d 955, 967 (9th Cir. 2012) (“if new information indicates to EPA that an existing SIP or SIP awaiting approval is inaccurate or not current, then, viewing air quality and scope of emissions with public interest in mind, EPA should properly evaluate the new information and may not simply ignore it without reasoned explanation of its choice”); see also 42 USC 7607(d)(6)(B) (“The promulgated rule shall also be accompanied by a response to each of the significant comments, criticisms, and new data submitted . . . during the comment period.”) (emphasis added). Thus, EPA is required, at a minimum, to take new information into account during the SIP approval process and, if necessary, alter its final decision accordingly. As explained in detail elsewhere, section 307(d) of the Act explicitly provides for the consideration of information developed after the proposed rule is published.

EPA considered this new cost information and the assessment of our evaluation regarding this information appears elsewhere in this document.

Comment: EPA is again overstepping its role in this process. Wyoming completed its BART analysis in 2009, more than three years ago, and it would have been impossible to incorporate the alleged urea price increases in that analysis. Simply put, Wyoming's BART determination is hardly arbitrary and capricious simply because it failed to take into account alleged urea price increases some three years after Wyoming completed its BART analysis. Wyoming did precisely what the Guidelines instruct: made a BART determination based on information available before the close of its public comment period. 40 CFR Part 51, App. Y., Section IV(D)(2)(3). To disapprove Wyoming's cost analysis based on information that was not available to the State would be to employ a “gotcha” approach that runs contrary to EPA's own regulations and counter to EPA's commitment to do its job fairly and objectively. If the urea issue is truly material, EPA should, at a minimum, allow Wyoming to consider whether this new information would affect its BART determination before disapproving that determination.

Another commenter suggests that urea prices are relevant to operating costs for SNCR but are not relevant to SCR. If the State's urea prices were too low, that would mean the State had underestimated the cost of SNCR, which is what EPA claims in its proposal. 78 FR 34748. Such an underestimate would have no material impact on the State's BART determination and thus provides no basis for EPA's disapproval. Once again, this is a fact that in retrospect supports the State's BART decision, rather than demonstrating it to be arbitrary. If Wyoming's estimate of the cost of SNCR should have been higher, as EPA maintains, the higher cost would tend to add further support for rejecting SNCR—the more expensive a control technology, the stronger the reason to reject it as BART. So if EPA is correct in claiming the State's assumed urea price was too low, it is incorrect in claiming this made a difference in the State's BART determination. A mistake in a cost assumption, if there was a mistake, is not a per se reason to reject a BART determination. Such a mistake would help support disapproval of a cost analysis and resulting BART determination only if it overstated costs in a material way and thus tended to make a technology appear significantly more costly than it actually would be.

Response: We disagree with portions of these comments. As we explained in responses to similar comments below in the section on Overarching Comments on BART, we agree that a change in the market price of urea, in and of itself, may have not provided EPA sufficient grounds for rejecting the State's SNCR analysis. However, we identified a number of deficiencies in our proposed rule, that when taken collectively, led EPA to conclude that Wyoming's consideration of the costs of compliance and visibility improvement for the EGUs was inadequate and did not properly follow the requirements in the BART Guidelines and statutory requirements. 78 FR 34748. Therefore, regardless of the market price of urea, EPA would have reached the same conclusion.

Additionally, EPA is required to take into account the urea price information and we have taken that technical information into account as detailed elsewhere in this final notice and the Start Printed Page 5078docket. As explained in detail above, while this information was not available to the State, EPA nonetheless had a duty to consider any new information submitted during public comment when reviewing the states' SIPs. See Sierra Club v. EPA, 671 F.3d 955, 967 (9th Cir. 2012).

Therefore, while the new urea cost information was not available to the State, EPA was nonetheless obligated to consider any new information submitted during public comment when reviewing the states' SIPs. Thus, EPA is required, at a minimum, to take new information into account during the SIP approval process and, if necessary, alter its final decision accordingly. Regarding the comment that Wyoming should get an opportunity to consider this information before EPA takes final action, see responses to similar comments above.

Comment: EPA relies on its consultant's report as a basis for rejecting Wyoming's cost analysis for SNCR and proposing to disapprove the State's NOX BART for Laramie River Station. 78 FR 34748. EPA may not reject the State's estimate of the NOX reduction achievable with SNCR just because EPA's consultant disagrees with the State. Under the appropriate legal standard, EPA must defer to the State's technical assessment absent demonstration it is arbitrary and capricious—which EPA has not attempted to prove. Nor can EPA mount a credible argument that its consultant's report is superior to the State's. The report does not comply with EPA's own Guidelines, as interpreted by EPA, and ignores site-specific conditions that have a huge impact on the cost of NOX emission controls. Given the flaws in the report, EPA's reliance on it is not only arbitrary and capricious, but downright astonishing.

Response: We disagree with the commenter's assertion that we have rejected the State's estimate of cost analysis for SNCR and the NOX reduction achievable with SNCR just because we disagree with the State. During the public comment period on our proposed rulemaking, Basin Electric, as well as other parties, submitted information concerning cost estimates. We have placed this information to the docket and as explained elsewhere in this document, taken it into account as part of this final rulemaking. This final action clearly explains the basis for our disapproval of State's NOX BART for Laramie River Station, based on comments received and our cost and visibility analysis, we are disapproving others. We also disagree that we are required to defer to the State's technical judgments and to apply an arbitrary and capricious standard in reviewing the State's SIP submittal. We respond in detail to those arguments elsewhere.

Comment: This commenter stated that even if the Wyoming's cost analyses were revised to reflect the EPA's high urea prices, the average cost effectiveness of SNCR would still be consistent with the State's original analyses. The commenter noted that the EPA's average and incremental cost effectiveness numbers for SNCR fall well below the values considered by the State to be cost effective and therefore are consistent with the State's original conclusion that the costs of compliance from the application of SNCR to the EGUs were reasonable. The commenter added that even if the State-analyzed urea costs are adjusted to reflect EPA's urea costs, the average cost effectiveness values remain below $2,600 dollars per ton of NOX reduced and with incremental cost effectiveness values below $5,000 dollars per ton of NOX reduced (citing commenter's Exhibit 10), and those values are consistent with the State's original conclusion. The commenter believes that it is clear that the EPA does not take issue with Wyoming's cost analyses, but rather Wyoming's BART conclusions. The commenter contended that the EPA's allegation that Wyoming incorrectly analyzed costs is simply an excuse for EPA to override Wyoming's BART determinations because EPA does not like the result. The commenter asserted that the EPA must explain why Wyoming's ultimate BART determinations run afoul of the law, rather than hold up allegations of technical deficiencies as window dressing for EPA to take over the role Congress gave to states to make BART determinations.

Response: We disagree with this comment. As we explained earlier in this final notice, Congress crafted the CAA to provide for states to take the lead in developing implementation plans, but balanced that decision by requiring EPA to review the plans to determine whether a SIP meets the requirements of the CAA. EPA's review of SIPs is not limited to a ministerial type of automatic approval of a state's decisions. EPA must consider not only whether the State considered the appropriate factors but acted reasonably in doing so. EPA has the authority to issue a FIP either when EPA has made a finding that the State has failed to timely submit a SIP or where EPA has found a SIP deficient. Here, EPA has authority on both grounds, and we have chosen to approve as much of the Wyoming SIP as possible and to adopt a FIP only to fill the remaining gap. Our action today is consistent with the statute. We disagree that technical deficiencies are mere “window dressing”; instead, appropriate technical analyses are fundamental to a reasoned BART determination. Finally, details of technical issues regarding urea costs are addressed elsewhere in this rule.

Comment: No single factor justifies disapproval of the State's BART. The authority to determine BART belongs to states, and BART determinations must be based on all five BART factors weighted together. States are responsible for balancing those factors and deciding how much weight to give to each factor. 70 FR 39123, 39130, 39170. To show that Wyoming had been arbitrary and capricious in making a BART determination, EPA would bear a heavy burden—a burden that it does not even begin to meet based on a disagreement that the State's cost analysis for SCR was in error. EPA's own incremental cost effectiveness for SCR is more than $5000/ton, which is a high cost even if lower than the State's. EPA makes no attempt to argue that the difference between its incremental cost effectiveness and the State's would have changed the State's selection of BART or rendered the State's BART arbitrary or illegal.

Response: We responded to similar comments elsewhere. First, as we explain in detail elsewhere, we disagree that EPA's review of a state's SIP submittal is limited to an arbitrary and capricious standard. Second, as we explain in detail elsewhere, we disagree that states have the sole authority to determine BART. Third, as we explain in detail elsewhere, we disagree that a “harmless error” standard should be applied.

Comment: In June of 2012, EPA issued a proposal that analyzed the cost effectiveness of various NOX control technologies at Laramie River Station. 77 FR 33051. Although EPA disagreed with the State's NOX BART determination for Laramie River Station, EPA accepted and relied on the State's cost analysis for NOX controls, which concluded that SCR would cost $3305 per ton of NOX removed, while SNCR would cost $2036 per ton of NOX removed. 77 FR 33051, Table 30 (These values are for Unit 3. The State's conclusions for Units 1 and 2 were similar.) In light of these estimates, EPA eliminated SCR from consideration at Laramie River Station “because the cost effectiveness value is significantly higher than LNBs with OFA and there is a comparatively small incremental visibility improvement over LNBs with Start Printed Page 5079OFA.” Id. EPA now expressly disavows its earlier finding, apparently as a result of comments that raised questions with the State's analysis and a cost analysis prepared by Andover. 78 FR 34740, 34748. Yet EPA's own cost analysis—based entirely on the findings of a technically infirm and legally indefensible contractor analysis of the costs of SNCR and SCR at Laramie River Station—concludes that the cost effectiveness of SCR at Laramie River Station ranges from $3,589 to $3,903, which exceed Wyoming's cost effectiveness demonstrations. Id. at 34774-34775. For EPA to take the position SCR is now cost effective, based on a higher estimate of tons NOX removed that is inconsistent with its earlier position and without any further explanation, is arbitrary and capricious. Cf. W. States Petroleum, 87F.3d at 284 (EPA “may not depart, sub silentio, from its usual rules of decision to reach a different, unexplained result in a single case”).

Response: We disagree with this comment. EPA's June 2012 Federal Register notice was a “proposal,” not a final agency action. Based on additional information and analyses, on June 10, 2013 we reproposed to partially approve and partially disapprove the Wyoming SIP. Therefore, contrary to commenter's assertions, we had not taken a final agency action in June 2012 and the Western States Petroleum case in not applicable here. In addition, we fully explained the reasons for the changes in our proposed action. We note that adjustments in cost-effectiveness of SCR were not the only factor in our proposed changes. We also revised modeling of visibility benefits of SNCR and SCR and cost-effectiveness of SNCR, which played a role in our reproposed BART determination.

2. Compliance With Section 307(d)

Comment: EPA cannot adopt a FIP using a procedure that simultaneously proposes both disapproval of a SIP BART determination and a different BART determination as a FIP. Doing so results in a violation of Section 307(d), which requires EPA to first announce the “statement of basis and purpose” that accompanies the FIP, including a summary of “the factual data on which the . . . rule is based” and “the major legal interpretations and policy considerations underlying the . . . rule.” 42 U.S.C. 7607(d)(1)(B), (d)(3)(A) & (C), (d)(6)(A). The reason is simple. BART determinations are inherently technical evaluations that consider costs, feasibility, potential plant shut-downs, etc. The same requirement would apply to any BART determination undertaken by EPA as part of a FIP. Thus, any response by EPA to comments that Basin Electric and others submit in support of Wyoming's BART determination will necessarily have to deal with new detailed technical information and data, particularly when, as here, EPA has initially proposed to reject a BART determination as inadequately supported and thus has invited extensive comments. EPA's responses to comments will then necessarily become part of the grounds supporting any new BART determination in a FIP, but will not have been publicly disclosed until EPA's response to comments on the SIP. Thus, EPA will be unable to provide a substantive statement of basis and purpose for the FIP in the same proposal to disapprove the SIP unless it intends to ignore comments. Yet this violates EPA's statutory obligation to announce all the facts and grounds supporting a FIP before adoption. It also wholly undermines the underlying purposes of the APA's notice and comment obligations. See, e.g., United States v. Cain, 583 F.3d 408, 420 (6th Cir. 2009) (these obligations are intended to “ensure fair treatment for persons to be affected by regulation” and to “ensure that affected parties may participate in decision making at an early stage'”) (citations omitted); NRDC v. Thomas, 805 F.2d 410, 437 (D.C. Cir. 1986) (the purposes of these obligations include that “notice improves the quality of agency rulemaking by ensuring that agency regulations will be tested by exposure to diverse public comment,” that “notice and the opportunity to be heard are an essential component of fairness to affected parties,” and that “by giving affected parties an opportunity to develop evidence in the record to support their objections to a rule, notice enhances the quality of judicial review”) (quoting Small Refiner Lead Phase-Down Task Force v. EPA, 705 F.2d 506, 547 (D.C. Cir. 1983).

This must be true, unless EPA's proposed course of action has already been determined, meaning that EPA has already decided to reject the SIP BART determinations and replace them with its own regardless of the comments submitted. Such prejudgment would be contrary to law. See, e.g., Davis v. Mineta, 302 F.3d 1104, 1112 (10th Cir. 2002) (plaintiffs were likely to prevail in showing agency acted arbitrarily and capriciously, in part because the agency “prejudged the NEPA issues”); Metcalf v. Daley, 214 F.3d 1135, 1146 (9th Cir. 2000) (agencies' environmental assessment prepared under NEPA was “demonstrably suspect” and “fatally defective” because the agencies “were predisposed” to a particular finding; agencies must conduct “an objective evaluation free of the previous taint”). Yet that is plainly what EPA is suggesting by its effort to simultaneously disapprove one BART determination while proposing another. Either EPA must ignore the comments so as not to establish new grounds for the FIP, or it must reject the comments on substantive grounds that become justification for the FIP but have never been publicly disclosed. Either way, its action violates APA standards.

This is a consequence of the procedural posture into which EPA has put itself by taking no action on the SIP until the end of the Sections 110(c) FIP clock. To follow the requirements of Sections 307(d), EPA must first propose to disapprove a SIP, take comment, and then make a decision after full and fair consideration of the comments. If, after open-minded consideration of the comments, EPA continues to believe the SIP must be disapproved, then and only then can EPA lawfully propose a different BART determination in a FIP, articulating for public comment why the proposed federal BART determination is legal and the State BART determination is not.

Failure to follow this procedure necessarily results in a violation of the law, one way or another. Nor does the existence of a Consent Decree excuse EPA's failure to follow the correct procedure. A court-fashioned decree may not foreclose the total range of procedural options available to an agency. See Watt v. Energy Action Educ. Found., 454 U.S. 151, 168-69 (1981) (refusing to limit the procedural options within the discretion of an agency); Marina T. Larson, Consent Decrees and the EPA: Are They Really Enforceable Against the Agency?, 1 Pace Envt'l L. Rev. 147, 160-63 (1983) (arguing that consent decrees may not limit agency procedural options). EPA waited until compelled by Court Order to propose disapproval of the State BART determination, but could have done so much earlier. In any event, the obligations EPA negotiated for itself in the Consent Decree cannot be used to deprive Wyoming or Basin Electric the substantive procedural rights afforded by the CAA.

Response: EPA disagrees with this comment, which fundamentally misunderstands the nature of notice-and-comment rulemaking. As the Ninth Circuit stated in another context:

Nothing prohibits the Agency from adding supporting documentation for a final rule in response to public comments. In fact, adherence to the [petitioners'] view might Start Printed Page 5080result in the EPA's never being able to issue a final rule capable of standing up to review: every time the Agency responded to public comments, such as those in this rulemaking, it would trigger a new comment period. Thus, either the comment period would continue in a never-ending circle, or, if the EPA chose not to respond to the last set of public comments, any final rule could be struck down for lack of support in the record.

Rybachek v. U.S. EPA, 904 F.2d 1276, 1286 (9th Cir. 1990).

In the context of the CAA, the specific rulemaking provisions in section 307(d) are in accord with this. Under section 307(d)(3), the notice for the proposed rule must be accompanied by a statement of basis and purpose, including “a summary of (A) the factual data on which the proposed rule is based; (B) the methodology used in obtaining the data and in analyzing the data; and (C) the major legal interpretations and policy considerations underlying the proposed rule.” 42 USC 7607(d)(3) (emphasis added). “All data, information, and documents referred to in [section 307(d)(3)] on which the proposed rule relies shall be included in the docket on the date of publication of the proposed rule.” Id. (emphasis added). Then, under section 307(d)(6), the promulgated rule must “be accompanied by (i) a statement of basis and purpose like that referred to in [section 307(d)(3)] with respect to a proposed rule.” 42 USC 7607(d)(6)(A) (emphasis added). In other words, the statement of basis and purpose must provide a summary of (among other things) the factual data and methodologies on which the promulgated rule is based. In addition, section 307(d)(6) specifically requires a “response to each of the significant comments, criticisms, and new data submitted . . . during the comment period.” 42 USC 7607(d)(6)(B) (emphasis added). And finally, “the promulgated rule may not be based . . . on any information or data which has not been placed in the docket as of the date of such promulgation,” id. 7607(d)(C), which by implication allows EPA to base the promulgated rule on information and data that is placed in the docket before the date of promulgation. Thus, section 307(d)(6) specifically contemplates that the Agency can in its promulgated rule rely on additional information and data that EPA develops after the proposed rule has been published.

In this instance, our FIP proposal was in accord with the requirements of section 307(d) of the Act. In particular, before the proposed rule was published, we included in the docket all the factual data, such as cost estimates and visibility modeling, on which the proposed rule was based. The comment identifies no deficiency in this regard. Instead, according to the comment the supposed deficiency is the failure to include in the docket for the proposal the data and information that EPA will develop to respond to comments. But, as discussed above, this is no deficiency; instead section 307(d) specifically contemplates that this will happen.

The argument in the comment regarding EPA's alleged prejudgment of its decision also belies a misunderstanding of notice-and-comment rulemaking. Under the comment's theory, in order to not have “prejudged” the outcome, EPA would have to avoid proposing any particular outcome in its notice of proposed rulemaking. However, under section 307(d)(3), “the notice of proposed rulemaking shall be published in the Federal Register, as provided under section 553(b) [of the APA].” Under section 553(b) of the APA, the “notice shall include” (among other things) “either the terms or substance of the proposed rule or a description of the subjects and issues involved.” 5 USC 553(b)(3) (emphasis added). Thus it is of course explicitly permitted under the CAA and the APA for a proposal notice to contain EPA's proposed disapproval of the State's BART determinations and EPA's proposed FIP BART determinations. This does not indicate prejudgment at all; indeed in this action EPA is adjusting certain determinations in response to certain comments, and in fact EPA previously reproposed its action on Wyoming's SIP based upon new information submitted by the public (77 FR 3302). The cases cited by the comment regarding prejudgment concern NEPA analysis and are not on point.

As the commenter noted, regional haze requirements apply both to our action on Wyoming's SIP submittal and our FIP. EPA disagrees that the BART determinations in its FIP, which must meet the same regional haze requirements as the BART determinations in Wyoming's SIP, must be published in a separate rulemaking procedure. To the extent that a comment on our proposed disapproval was identified as also relevant to our proposed FIP, we have responded to it. The commenter was not deprived of procedural rights merely because the commenter could not submit information twice in two separate rulemakings. All affected parties had ample opportunity to submit any pertinent information to EPA.

Regarding the consent decree, we have elsewhere explained that it did not limit or modify EPA's substantive discretion. With respect to the comment's argument that it improperly limited EPA's procedural discretion, any such limits are found in the statutory deadlines and mandatory duties in the Act itself. The case cited in the comment, Watt v. Energy Action Educ. Found., 454 U.S. 151 (1981), did not concern a consent decree and is not on point. In it, the Supreme Court was “unable to find anything, either in the legislative history or in the 1978 Amendments [to the Outer Continental Shelf Lands Act] themselves, that compels the conclusion that the Congress as a whole intended to limit the Secretary of the Interior's discretion” with respect to choice of bidding systems for oil and gas leases. Id. at 168. By contrast, the CAA sets certain statutory deadlines for EPA's action on SIP submittals and FIP promulgations and thereby explicitly limits the Administrator's discretion for final action. We elsewhere respond to comments that EPA's promulgation of its FIP was outside EPA's authority under 110(c) of the Act. Finally, the cited law review article, Marina T. Larson, Consent Decrees and the EPA: Are They Really Enforceable Against the Agency?, 1 Pace Envt'l L. Rev. 147 (1983), is also not on point. It discusses a settlement agreement which “set[ ] forth specific methods and formalized criteria for the [A]dministrator to use in assessing the need for regulation. These rules [would] control the nature of the data collected and its subsequent interpretation, and [would] have a significant influence on the substantive decisions reached.” Id. at 162. No such constraints have been placed on our methods and use of data in the aforementioned consent decree. We respond elsewhere to comments about procedural due process rights.

3. Compliance With Section 169A(d)

Comment: One commenter argued that section 169A(d) of the CAA requires that before holding a hearing on a proposed regional haze plan, “the State (or the Administrator, in the case of a [FIP]), shall consult in person with the appropriate federal land manager (FLM) or managers and shall include a summary of the conclusions and recommendations of the FLMs in the notice to the public.” 42 U.S.C. 7491(d). In its proposed action, EPA recites this land manager consultation requirement as it applies to SIPs, 78 FR 34744, but, EPA notably ignores that this requirement applies equally to FIPs.Start Printed Page 5081

The commenter asserted that not once in any of EPA's public notices of the hearings EPA held on its proposed FIP did EPA include a summary of the conclusions and recommendations of the FLMs in the notice to the public. See 78 FR 34738 (June 10, 2013); 78 FR 40654 (July 8, 2013). Consequently, the commenter argued that EPA cannot rely on the State's public notices because the State held its public hearings years before EPA proposed its FIP and because the SIP differs substantially from the FIP.

The commenter argued that EPA's failure to comply with Section 169A(d) can be understood only as arbitrary and capricious. The CAA has required consultation with FLMs, which oversee the Class I areas the regional haze program aims to protect, from the very beginning of the regional haze program, see 42 U.S.C. 749l(a)(2), and continuously through the development of each implementation plan, id. Sections 749l(d). Congress therefore understood the importance of working closely with FLMs in regional haze planning.

In 1999, EPA plainly understood the significance of consulting the FLMs when it promulgated the RHR. See 64 FR 35714, 35747 (July 1, 1999) (describing land manager consultation as “important and necessary”). Both times EPA proposed action on Wyoming's SIP—in 2012 and again in 2013—EPA reiterated the need to consult with FLMs when developing a regional haze implementation plan. 77 FR 33022, 33028 (June 4, 2012); 78 FR 34738, 34744-45 (June 10, 2013).

Against this backdrop, the commenter explained, EPA's failure to explain why EPA believed it did not have to consult with the FLMs when promulgating its FIP for Wyoming, let alone comply with the simple consultation process set forth in Section 169A(d), is plainly arbitrary and capricious. Because FLMs play a critical statutory role in the regional haze program, there is a substantial likelihood that EPA's proposed FIP would be significantly different if EPA had complied with Section 169A(d).

Response: EPA agrees that consultation with the FLMs is an important aspect of the regional haze program. EPA has engaged with the appropriate FLMs on all of its regional haze actions, including its proposed actions on the Wyoming regional haze SIP. While EPA did not include a summary of the FLMs' conclusions and recommendations on the proposed FIP in the public hearing notices, those conclusions and recommendations are readily available to the public in the online docket for this rulemaking.[54]

EPA also disagrees with the commenter that the consultation materials contained in the State's public notices are irrelevant just because the State conducted its public hearings many years ago. The FLMs concluded at that time that the Wyoming regional haze SIP did not adequately protect the State's Class I areas, and these conclusions and recommendations informed EPA when we proposed to disapprove portions of the Wyoming regional haze SIP and issue a FIP.

Finally, there is no basis to the commenter's claim that EPA's proposed FIP would be significantly different if we had included the FLMs' conclusions and recommendations in the public hearing notices. We carefully considered the comments of the FLMs and have responded to them elsewhere throughout this document. As those responses explain in more detail, we have chosen not to change our proposed NOX BART determinations in all of the ways in which the FLMs requested. We point out, however, that had EPA adopted the FLMs' recommendations, we would be requiring SCR on all of the BART-eligible EGUs in Wyoming, a result that this particular commenter has vigorously opposed.

Comment: The processes Congress required EPA to follow under the regional haze program were circumvented. For example, the CAA requires both states and EPA to consult with FLMs on regional haze implementation plans. Public notice of the FLMs' conclusions and recommendations is to occur before holding a hearing on the plan. While EPA recites this requirement in its proposed action, it utterly failed to include any FLM consultation on behalf of its agency. EPA held three hearings and not once in any hearing did the EPA indicate it had consulted the FLMs in Wyoming and no conclusions or recommendations of any consultations were provided.

Response: See above response.

4. Public Hearings

Comment: EPA's regional haze plan promulgation regulations require EPA to provide public notice at least thirty days in advance of a hearing on a proposed implementation plan. 40 CFR 51.102(d) (a plan hearing “will be held only after reasonable notice, which will be considered to include, at least 30 days prior to the hearing(s)”); see also 40 CFR 51.100(i). Although EPA held three public hearings on its proposed FIP for Wyoming, not once did EPA provide the public at least thirty days advance notice of the hearing. EPA proposed its FIP on June 10, 2013 and provided only fourteen days notice of its hearing on the proposal. 78 FR 34738, 34738. After Governor Mead, Wyoming's Congressional Delegation, and the Wyoming Department of Environmental Quality (DEQ) pointed out to EPA that fourteen days provided far too inadequate notice for the public to understand the proposed FIP and therefore meaningfully participate in the public hearing, EPA agreed to hold two additional hearings. On July 8, 2013, EPA publicly noticed its plans to hold the additional hearings on July 17, 2013 and July 26, 2013. 78 FR 40654, 40654. Thus, although EPA had the opportunity to correct its errors, it failed to do so by again providing less than thirty days notice of its hearings.

Here again, EPA's noncompliance with its own regulatory processes is arbitrary and capricious. EPA cannot ignore the law for its own benefit without at least providing a reasoned justification for doing so. In this case EPA has provided no such explanation, thereby rendering its failure an arbitrary abuse of power. And by shortcutting public participation, EPA undermined the central democratic purposes of notice-and-comment rule-making. Had EPA honored the law and held itself to the same standards it holds states, the public could have more meaningfully commented on EPA's proposal. As a result of that public input, EPA's proposed FIP might be considerably different, assuming, as we must, that EPA would have considered those comments with an open mind.

DEQ understands that EPA rushed its FIP promulgation process in order to meet the deadlines it consensually established with a third party in litigation to which Wyoming was not a party. But, EPA's outside arrangements do not excuse it from complying with the law, or allow it to shortcut public participation in the promulgation of a rule, especially one that will harm Wyoming. DEQ discourages EPA from imposing its illegally promulgated FIP on Wyoming. But, in the event EPA decides nevertheless to do so, DEQ encourages EPA to re-propose its FIP in a manner that complies with the statutory and regulatory plan development processes. To do otherwise is to arbitrarily hold states to a different plan promulgation standard than EPA itself adheres to, even though the CAA makes no such distinction. Such irrationally unequal treatment is the essence of arbitrary regulation.Start Printed Page 5082

Response: EPA disagrees with this comment. First, 40 CFR 51.102(d) implements the requirement in section 110(a)(2) that state plans “be adopted by the State after reasonable notice and hearing.” See 72 FR 38787 (July 16, 2007). When EPA—which is not a state—promulgates a FIP, EPA instead is bound by the requirements in section 307(d) of the Act. EPA has not promulgated specific regulations governing EPA's processes under section 307(d); however, EPA complied with the public hearing requirements in 307(d) as explained below. The definition of “State agency” in 51.100(i) does not contradict this; indeed the commenter elsewhere protests vigorously elsewhere that states, not EPA, are “primarily responsible for development and implementation of a plan under the Act.” 40 CFR 51.100(i). Thus, EPA does not fall under the definition of “State agency.” We also note that EPA initially provided a 60-day comment period for this action and then extended it 15 more days; under 40 CFR 51.102. States need only provide a 30-day period for written comments. See 72 FR at 38788 (“Whether or not a public hearing is held, the State is required to provide a 30-day period for the written submission of comments from the public.”).

In promulgating a FIP under CAA section 110(c), EPA is required to: “give interested persons an opportunity for the oral presentation of data, views, or arguments, in addition to an opportunity to make written submissions; keep a transcript of any oral presentation; and keep the record of such proceeding open for thirty days after completion of the proceeding to provide an opportunity for submission of rebuttal and supplementary information.” [55] In this rulemaking, EPA held three public hearings on its proposed FIP. In addition to the public hearing initially scheduled on June 24, 2013 in Cheyenne, Wyoming, additional public hearings were held on July 17, 2013 in Cheyenne, Wyoming and on July 26, 2013 in Casper, Wyoming. The transcripts for those hearings consisted of 321 pages. These hearings were announced in the Federal Register on June 10, 2013 and July 8, 2013,[56] and a pre-publication version of the proposal was posted on EPA's Web site prior to publication in the Federal Register. The proposal was published in the Federal Register on June 10, 2013 and was initially scheduled to close on August 9, 2013. The public comment period was extended in response to letters received from the Governor and Congressional delegation, which are in the docket for this action, and public comments were accepted through August 26, 2013, 30 days after the last hearing, as required. EPA received over 1900 comments on the reproposal, including over 130 unique comments submitted from organizations, companies, and individuals. The major comments consisted of over 1130 pages, including attachments. The commenters have not explained how their ability to comment was impaired in any way by the opportunities for public comment that EPA provided, including three public hearings and the 75-day comment period.

Comment: EPA failed to follow its own rules for providing public notice of hearings on regional haze implementation plans. Those rules require a minimum of 30 days advance public notice of hearings on implementation plans. The first notice in the Federal Register of a public hearing was issued on June 10, 2013, for a public hearing to be held on June 24, 2013. EPA issued a second notice for additional public hearings on July 8, 2013 in the Federal Register. The notice identified July 17, 2013 and July 26, 2013 as dates set. This provided the public nine and eighteen days notice of the respective hearings.

Response: We disagree with this comment, see above response.

5. RHR and BART Guidelines

Comment: Regardless of the effect of AFUDC on cost effectiveness as demonstrated by the Sargent & Lundy sensitivity analyses, EPA has no authority, as part of its interpretation of a non-binding guidance document, to impose restrictions on the categories of costs that states can include when assessing the “costs of compliance” in a BART determination. EPA has failed to make a showing that Wyoming's compliance with Sections 169A(g)(2) or otherwise violates governing law. Including AFUDC is not a lawful ground for disapproving Laramie River Station BART, and it is improper to exclude AFUDC in EPA's FIP analysis for Laramie River.

Response: EPA disagrees with this comment. EPA's revised cost-effectiveness values are consistent with EPA's regulations and the parameters set forth in the Control Cost Manual. EPA explained in promulgating the BART Guidelines that “[s]tates have flexibility in how they calculate costs. “See 70 FR at 39127 (July 6, 2005). A state may deviate from the Control Cost Manual provided its analysis is reasonable. EPA independently evaluated Sargent & Lundy cost-effectiveness calculation, explaining elsewhere in this document that the CCM explicitly excludes AFUDC from control costs, and EPA's estimates were correct in excluding AFUDC. See Oklahoma v. U.S. EPA, 723 F.3d 1201, 1212 (10th Cir. 2013) (“The EPA therefore had a reasonable basis for rejecting the 2008 Cost Estimates [that were based on the overnight costing method] as not complying with the guidelines.”)

Furthermore, as Region 9 explained in responding to similar comments: [57]

EPA disagrees “with commenters' assertions that AFUDC is a cost that should be incorporated into our cost analysis, as it is inconsistent with CCM methodology. The utility industry uses a method known as “levelized costing” to conduct its internal comparisons, which is different from the methods specified by the CCM. Utilities use “levelized costing” to allow them to recover project costs over a period of several years and, as a result, realize a reasonable return on their investment. The CCM uses an approach sometimes referred to as overnight costing, which treats the costs of a project as if the project were completed “overnight”, with no construction period and no interest accrual. Since assets under construction do not provide service to current customers, utilities cannot charge the interest and allowed return on equity associated with these assets to customers while under construction. Under the “levelized costing” methodology, AFUDC capitalizes the interest and return on equity that would accrue over the construction period and adds them to the rate base when construction is completed and the assets are used. Although it is included in capital costs, AFUDC primarily represents a tool for utilities to capture their cost of borrowing and return on equity during construction periods. AFUDC is not allowed as a capitalized cost associated with a pollution control device under CCM's overnight costing methodology, and is specifically disallowed for SCRs (i.e., set to zero) in the CCM.[58] Therefore, in reviewing other BART determinations, EPA has consistently excluded AFUDC.[59]

Comment: EPA claims that Wyoming should have used actual emissions during the baseline period instead of calculating baseline emissions from the actual average heat input and actual average emission rate. EPA apparently claims that this deviated from the BART Guidelines. 78 FR 34773-34774. Start Printed Page 5083However, the Guidelines do not mandate EPA's approach. They say, rather, that the baseline emissions rate “should represent a realistic depiction of anticipated annual emissions for the source” and “in general” states should estimate anticipated emissions based on actual baseline emissions. 70 FR 39167. Nothing in the text of the Guidelines requires states to use any particular approach to estimate future emissions. The Guidelines were constructed to assist the states in making cost assessments, not to mandate the same assessment and the same results in every case by use of mandatory checklists. The word “should” in the Guidelines makes clear there is no mandatory action required. See Aragon v. United States, 146 F.3d 819, 826 (10th Cir. 1998) (describing Air Force Manual 85-14's use of the word “should” as “suggestive, rather than mandatory language” in a Federal Tort Claims Act case); In re Glacier Bay, 71 F.3d 1447, 1452-53 (9th Cir. 1995) (interpreting the National Oceanic and Atmospheric Administration's use of the word “should” in manuals and instructions as “suggestive” language conferring hydrographers with discretion); Culbert v. Young, 834 F.2d 624, 628 (7th Cir. 1987) (holding that use of the word “should” in a Wisconsin Administrative Code provision governing inmate discipline “only advises the security director on what criteria to consider but does not require him to consider them,” and explaining that “[t]he word `should,' unlike the words `shall,' `will,' or `must,' is permissive rather than mandatory”). See also Dickson v. Sec'y of Defense, 68 F.3d 1396, 1401 (D.C. Cir. 1995) (“When a statute uses a permissive term such as `may' rather than a mandatory term such as `shall,' this choice of language suggests that Congress intends to confer some discretion on the agency, and that courts should accordingly show deference to the agency's determination.”) (emphasis omitted).

EPA is therefore merely disagreeing with a judgment call made by the State, not pointing to violation of a mandatory methodology. And, even though not required to do so, Wyoming did follow the recommendation in the Guidelines. Although EPA contends that the State used a baseline based on annual average heat input for 2001-2003 and an emission rate of 0.27 rather than the “actual annual average” emissions, 78 FR 34773-34774, the State's May 28, 2009 BART Analysis actually says “[b]aseline emissions [are] based on continuous emissions monitoring (CEM) annual averages for 2001-2003.”

But even if EPA were correct, EPA would still be wrong in asserting that the State failed to follow the BART Guidelines. The approach that EPA objects to would be an appropriate method to realistically depict anticipated annual emissions. Certainly it would be reasonable to multiply the actual annual amount of heat in Laramie River coal during the baseline period by the same baseline emission rate of 0.27 lb/MMBtu that was used by EPA's own consultant. 78 FR at 34773; Review of Estimated Compliance Costs for Wyoming Electric Generating (EGUs)—Revision of Previous Memo, memo from Jim Staudt, Andover Technology Partners, to Doug Grano, EC/R, Inc., Feb. 7, 2013 (“Andover Report”) at 15 Table 4, EPA docket cite EPA-R08-OAR-2012-0026-0086. Any estimate of anticipated emissions is necessarily a projection, and by definition cannot require exclusive reliance on past actual emissions.

That the State's approach to baseline emissions was a realistic projection is borne out by the fact that the annual baseline emissions the State used to calculate cost effectiveness for Laramie River differs from EPA's baseline by only the following de minimis amounts: 269 tons higher than EPA's 6051 tons for Unit 1, a difference of only 4%; 8 tons lower than EPA's 6285 tons for Unit 2, a difference of only 0.1%; and 73 tons higher than EPA's 6375 tons for Unit 3, a difference of only 1%. No fair assessment could conclude that such de minimis differences violate the Guidelines or yield an “implausible” result so extreme as to be arbitrary and capricious. 78 FR 34773-34776.

If EPA's values are realistic, the State's values are realistic. There is no material difference between them. The objective of a BART determination is to arrive at a technology selection that weighs and takes into account the five BART factors. The negligible difference between EPA's baseline emissions and the State's is not material and therefore is not a valid ground for disapproving the State's NOX BART for Laramie River, and EPA has made no effort to show otherwise. EPA's role is not to fly speck each and every aspect of the BART process in a search for reasons to disapprove the State's determination.

In fact, EPA proposes to approve other BART determinations made by Wyoming despite the same alleged “errors,” unequivocally demonstrating that its disagreement with Wyoming's approach to baseline calculations does not amount to proof of a legal violation by the State. EPA claims that for several Wyoming sources subject to BART, Wyoming committed the same “cost and visibility errors” that EPA claims for Laramie River, but proposes nonetheless to approve the BART determinations for these sources “because we have determined that the State's conclusions were reasonable despite the cost and visibility errors.” 78 FR 34750. EPA contradicts itself when it overlooks errors for other sources and yet claims those same “errors” as per se reasons to disapprove BART for Laramie River Station. Such inconsistent treatment is erroneous. See W. States Petroleum v. EPA, 87 F.3d 280, 282 (9th Cir. 1996). EPA's own behavior therefore demonstrates that the baseline used for Laramie River is not a material departure from any requirement and is not a basis for disapproval of the State's BART determination. EPA is stretching to find any excuse to impose its own technology preferences, contrary to law.

Wyoming's choice of baseline emissions is neither inconsistent with the BART Guidelines nor materially different from EPA's allegedly correct baseline emissions, and therefore is not a valid ground for disapproving Wyoming's NOX BART for Laramie River.

Response: We disagree with some aspects of this comment, but agree with others. First, we disagree with the commenter's characterization of the BART Guidelines as other than mandatory in the case of Laramie River Station, including in regard to how baseline emissions are calculated. The generating capacity of Laramie River Station of 1,705 MW surpasses the threshold of 750 MW used to determine whether the BART Guidelines must be applied. As stated in the RHR: “The determination of BART for fossil-fuel fired power plants having a total generating capacity greater than 750 megawatts must be made pursuant to the guidelines in appendix Y of this part (Guidelines for BART Determinations Under the Regional Haze Rule).” [60] Moreover, the commenter's attempts to turn “should” into “may” are of no avail. Because the BART Guidelines are mandatory for EGUs larger than 750 MW, EPA's use of the word “should” indicates a mandate, not a suggestion. Elsewhere in the Guidelines, EPA uses “may” when EPA means “may.” See, e.g. 40 CFR Part 51, App'x Y, II.A.4 (“In order to simplify BART determinations, States may choose to identify de minimis levels of pollutants at BART-eligible sources (but are not required to do so).”) (emphasis added). Furthermore, the Tenth Circuit Court of Appeals has interpreted “should” in the Guidelines to mean “required.” See Start Printed Page 5084Oklahoma v. U.S. EPA, 723 F.3d 1201, 1213 (10th Cir. 2013) (“The guidelines require that states provide support for any site-specific costs that depart from the generic numbers in the Control Cost Manual. See 40 CFR part 51 app. Y(IV)(D)(4)(a) n.15 (“You should include documentation for any additional information you used for the cost calculations, including any information supplied by vendors that affects your assumptions regarding purchased equipment costs, equipment life, replacement of major components, and any other element of the calculation that differs from the Control Cost Manual.”)”) (emphasis added).

Notwithstanding that the BART Guidelines are mandatory for Laramie River Station, we agree that Wyoming's approach, having used both the actual NOX emission rate and the actual heat input from the baseline period, resulted in a realistic depiction of anticipated annual emissions consistent with the BART Guidelines, that these emissions differed only slightly from baseline emissions estimated by EPA and that, therefore, Wyoming's treatment of baseline emissions by itself was not a basis for EPA to disapprove NOX BART for Laramie River Station. Nonetheless, as discussed in response to other comments, we maintain that there were other deficiencies in Wyoming's BART analysis for Laramie River Station that remain a valid basis for our disapproval. Most notably, Wyoming did not consider the visibility impacts of SNCR as required by the CAA and BART Guidelines.

Comment: Against its longstanding 30-year history of interpreting and applying the RHR and Guidelines, EPA has now embarked on a spate of BART disapprovals demonstrating that the agency is now interpreting and applying the Guidelines and CCM very differently than it did in the past, and signaling that EPA has actually decided to reinterpret the statute and Guidelines without notice and comment to the states.

EPA is manufacturing requirements in the Guidelines that do not exist, for the purpose of abandoning the administrative structure conferring state primacy that Congress created with both the CAA generally and the Regional Haze Statute in particular. EPA is doing so by interpreting the BART Guidelines and CCM as setting forth detailed, mandatory regulatory requirements that are not actually in the text, and by seeking to make any deviation from the recommendations in the Guidelines or CCM grounds for voiding states' BART choices.

EPA is attempting to convert recommendations into mandates. This new interpretation of the Guidelines and Cost Manual is erroneous, contrary to their statutory role, unannounced, and calculated to federalize BART decisions by making them all follow identical paths whether or not local considerations and costs warrant separate treatment in control decisions.

Response: Our proposal clearly laid out the bases for our proposed approval and disapproval of the State's BART and reasonable progress determinations, as well as other SIP elements. We have relied on the standards contained in our regional haze regulations and the authority that Congress granted us to review and determine whether SIPs comply with the minimum statutory and regulatory requirements.[61] To the extent we have found that the State's cost analysis relies on values that do not conform to applicable requirements of the Act and regulations, we have disapproved those elements of the analysis. To the extent the state has considered visibility improvement from potential emissions controls in a way that is inconsistent with the CAA and regulations, we have disapproved those elements of the analysis.

Where, as explained in our proposed notice and final notice, a state determines that a less stringent control technology is the “best available,” as was the case here with regard to NOX emissions, the state must justify its decision by explaining how the BART factors led it to choose that level of control over more stringent options. See 70 FR 39170-71. While a state has significant discretion regarding how to conduct its BART analysis, EPA must ultimately ensure that the state has demonstrated it has a reasoned basis, consistent with the Act's requirements, for determining that a given emissions control technology is “the best available” for each source. See Oklahoma, 723 F.3d at 1208 (“[W[hile it is undoubtedly true that the statute gives states discretion in balancing the five BART factors, it also mandates that the state adhere to certain requirements when conducting a BART analysis.”).

In determining SIP adequacy, we inevitably exercise our judgment and expertise regarding technical issues, and it is entirely appropriate that we do so. Courts have recognized this necessity and deferred to our exercise of discretion when reviewing SIPs. See, e.g., Connecticut Fund for the Env't., Inc. v. EPA, 696 F.2d 169 (2nd Cir. 1982); Michigan Dep't. of Envtl. Quality v. Browner, 230 F.3d 181 (6th Cir. 2000); Mont. Sulphur & Chem. Co. v. EPA, 666 F.3d 1174, 1190 (9th Cir. 2012) cert. denied, 133 S. Ct. 409, (2012). Contrary to the commenter's assertion, we have not abandoned the State's primacy. In fact, we have approved the vast majority of the State's determinations. We are only disapproving the State's analyses and decisions that do not conform to the CAA and regulations. We are authorized to do so.

Comment: As early as 1979, EPA recognized that the regional haze program is organized around “goals” and “reasonable progress,” and not hard objective requirements: Section 169A of the CAA provides for consideration of the degree or significance of visibility improvement, costs, energy, and other factors in applying retrofit controls to major sources and in making “reasonable” progress toward the national goal. These provisions indicate that some flexibility can be allowed in implementing control programs for remedying existing impairment and that priorities can be established.

Thus, while the BART analysis may include consideration of factors similar to those applied in a BACT analysis, BART does not require any threshold level of control. As EPA acknowledged in its 2004 re-proposal of the BART Guidelines, “for the BART analysis, there is no minimum level of control required.” 69 FR. 25184, 25219 (May 5, 2004). The RHR's “national goal” is not a mandate but, rather, a foundation for analytical tools to be used by the states in setting RPGs. The BART Guidelines were therefore developed to assist states in making their own BART determinations by providing analytical tools. They were not designed or intended by Congress to impose inflexible mandates that become tripwires for EPA to use as a means of federalizing BART decisions with set criteria. EPA's current effort to convert the Guidelines into something they were not intended to be is improper and calculated to shift to EPA authority over BART determinations that Congress reserved to the states. “[A]n agency cannot create regulations which are beyond the scope of its delegated authority.” Nagahi v. INS, 219 F.3d 1166, 1169 (10th Cir. 2000). Nor can an agency reinterpret regulations for that purpose.Start Printed Page 5085

Congress authorized EPA to provide guidelines only as to limited aspects of a state's BART decision-making process, and left the majority of that process to the states' discretion. Specifically, in the subsection immediately preceding the reference to the Guidelines, Congress directed EPA to conduct a study on available methods for implementing the national goal and provide recommendations to Congress for (1) “methods for identifying, characterizing, determining, quantifying, and measuring visibility impairment in Federal areas”; (2) “modeling techniques (or other methods) for determining the extent to which manmade air pollution may reasonably be anticipated to cause or contribute to such impairment”; and (3) “methods for preventing and remedying such manmade air pollution and resulting visibility impairment.” CAA Sections 169A(a)(3)(A)-(C), 42 U.S.C. 7491(a)(3)(A)-(C).

In the next subsection, Congress directed EPA to promulgate regulations—but with any regulation of the states' BART determinations confined to those limited areas on which EPA had been directed to conduct studies and make a report to Congress. Specifically, CAA Section 169(b) provides, in pertinent part, that the regulations “shall—(1) provide guidelines to the States, taking into account the recommendations under subsection (a)(3) of this section on appropriate techniques and methods for implementing this section (as provided in subparagraphs (A) through (C) of such subsection (a)(3)), and (2) require each applicable implementation plan for a State . . . to contain such emission limits, schedules of compliance and other measures as may be necessary to make reasonable progress toward meeting the national goal.” Id. Sections 7491(b)(1)-(2).

Accordingly, Congress only authorized EPA to promulgate regulations or guidelines on the identification and measurement of visibility impairment, the methods for measuring and predicting future visibility impairment, the methods for preventing and remedying air pollution and resulting visibility impairment, and the CAA's general requirement that states develop SIPs to include the BART and reasonable progress determinations required by the RHR. Congress did not authorize EPA to promulgate regulations or guidelines mandating exactly how the states should conduct their BART analyses, and made clear that the purpose of the guidelines was to provide “recommendations” to the states.

Consistent with the statute and regulations, the BART Guidelines contemplate a two-step process: (1) the “Attribution Step,” which consists of analyzing which sources are appropriately subject to BART controls; and (2) the “Determination Step,” which consists of determining, based on the five statutory BART factors, an appropriate level of control. 70 FR 39108, 39126; see also Utility Air Regulatory Group, 471 F.3d at 1335-36 (discussing two-step process). The Guidelines for the Determination Step are designed as a “step-by-step guide” for states to identify the “best system of continuous emissions control technology,” taking into account the five BART factors. 70 FR 39127. See also id. at 39158 (the Guidelines describe a “process for making BART determinations”). They are merely “helpful guidance” for sources other than power plants with a capacity greater than 750 MW. Id. at 39108; Utility Air Regulatory Group, 471 F.3d at 1339. Yet, even for larger power plants, the Guidelines are procedural in nature, setting forth criteria for evaluating control alternatives, but not mandating a substantive result. As EPA acknowledges, to mandate a choice of technology would infringe on “those areas where the Act and legislative history indicate that Congress evinced a special concern with insuring that States would be the decision makers.” 70 FR 39137. See also id. at 39107 (“The State must determine the appropriate level of BART control”).

The flexibility afforded by the Guidelines is critical to ensuring that states maintain primacy in making BART determinations. When EPA re-proposed the Guidelines in 2004, for example, EPA requested comment on a sequential process—similar to a BACT analysis—for considering the five statutory BART factors. 69 FR 25197-25198. In the final rule, however, EPA concluded that “States should retain the discretion to evaluate control options in whatever order they choose, so long as the State explains its analysis of the CAA factors.” 70 FR 39130. EPA also expressed that the Guidelines confer authority on the state to make “a BART determination based on the estimates available for each criterion, and as the CAA does not specify how the state should take these factors into account, the states are free to determine the weight and significance to be assigned to each factor.” Id. at 39123.

EPA further emphasized the flexibility inherent in each step of the BART determination: “States have flexibility in how they calculate costs,” id. at 39127, and “have the flexibility to develop their own methods to evaluate model results,” id. at 39108. EPA points out that “States should have flexibility when evaluating the fifth [visibility] statutory factor.” Id. at 39129. See also id. (“Because each Class I area is unique, we believe States should have flexibility to assess visibility improvements due to BART controls by one or more methods, or by a combination of methods . . .”). Even the presumptive emission limits for power plants greater than 750 MW “are presumptions only; in making a BART determination, states have the ability to consider the specific characteristics of the source at issue and to find that the presumptive limits would not be appropriate for that source.” Id. at 39134.

Response: EPA agrees that states play an important role in the regional haze program. However, EPA disagrees that this action conflicts with the State's statutory role or that this rule is beyond EPA's authority. First, the regional haze program explains that EPA “shall . . . require each applicable implementation plan for a State . . . to contain such emission limits, schedules of compliance, and other measures as may be necessary to make reasonable progress toward meeting the national goal.” 42 U.S.C. 7491(b)(2). The CAA makes clear that EPA is statutorily obligated to reject a SIP that would “interfere with any applicable requirement concerning attainment and reasonable further progress . . . or any other applicable requirement of this chapter.” 42 U.S.C. 7410(l). Thus the CAA provides EPA with the authority to review and reject an inadequate regional haze SIP. Oklahoma v. EPA, 723 F.3d 1201, 1207 (10th Cir. 2013); North Dakota v. EPA, 730 F.3d 750 (8th Cir. 2013).

Second, EPA is required to establish guidelines to ensure that states achieve the visibility goals set forth in the Act. 42 U.S.C. 7491(b)(1). EPA agrees that states have some flexibility in BART determinations, but that flexibility is limited and states must provide EPA with reasoned analysis for their SIP decisions. Oklahoma v. EPA, 723 F.3d 1201, 1207 (10th Cir. 2013) (noting that while “it is undoubtedly true that the statute gives states discretion in balancing the five BART factors, it also mandates that the state adhere to certain requirements when conducting a BART analysis”); North Dakota v. EPA, 730 F.3d 750 (8th Cir. 2013) (explaining EPA is not required to “approve a BART determination that is based upon an analysis that is neither reasoned nor moored to the CAA's provisions”). The regional haze guidelines provide states Start Printed Page 5086with methods to determine BART that EPA considers reasonable, although states may consider methods not provided for in the guidelines in certain circumstances. For example, in explaining a state's flexibility to determine costs, the guidelines note that “if there are elements or sources that are not addressed by the Control Cost Manual or there are additional cost methods that could be used, we believe that these could serve as useful supplemental information.” 70 FR No. 128 39127. (July 6, 2005). A state, however, must demonstrate that any methods it has used to determine BART that are not found within the guidelines are reasonable.

EPA may, and has, approved state BART determinations that do not rigidly follow the BART guidelines, so long as the state's determinations are reasonable. Here, however, Wyoming's methods were inconsistent with the BART guidelines, unreasonable, and inconsistent with the CAA's statutory and regulatory requirements, as explained elsewhere in these comments. Nothing in this rule displaces a state's discretion to balance the five factors, if the state calculates the factors using reasonable methods that are consistent with the regulatory and statutory requirements of the CAA.

Comment: EPA is now construing the BART Guidelines to treat “recommendations” as “mandates” such that states no longer have the authority to vary from the recommendations, however insignificantly, without finding EPA disapproving their BART determinations. Such an interpretation violates both the plain language of the CAA and its underlying cooperative federalism structure. First, Section 169A(b)(2)(A) provides that BART shall “be determined by the State.” 42 U.S.C. 7491(b)(2)(A). Section 169A(g)(2) provides that states are to determine the “costs of compliance” and the “degree of improvement in visibility.” Id. Section 7491(g)(2). Any interpretation and application of the BART Guidelines and CCM that has the effect, whether directly or indirectly, of mandating particular outcomes or approaches to reaching a BART determination invades state authority. States do the cost of compliance and visibility assessments, not EPA. Treating recommendations as mandates has the effect of forcing all states to follow each recommendation precisely the same way, effectively federalizing the BART determination by affording EPA the authority to employ the SIP approval process as a means of forcing all states to take the same approach required by EPA in all cases or find their independent decisions overruled. This violates the structure and design by Congress, and conflicts with the congressional commitment of the BART decision to the States. American Corn Growers, 291 F.3d at 7-10. This problem did not exist when EPA historically construed the “recommendations” in the Guidelines to be “recommendations” rather than mandates, but EPA's current approach of identifying deviations from the CCM or from the “recommendations” of the Guidelines as “errors of law” destroys state primacy and thus conflicts with the plain language of the statute and is unreasonable and not entitled to deference.

EPA's interpretation of the BART Guidelines violates Section 169A of the CAA because it also restricts state discretion in the decision-making process. It is the states, not EPA, that are authorized to determine BART. 42 U.S.C. 7491(b). In doing so they are directed to take into consideration the five BART factors—costs of compliance, energy and non-air quality environmental impacts of compliance, any existing pollution control technology in use at the source, the remaining useful life of the source, and the improvement in visibility that would be achieved by the use of control technology. Id. Section 7491(g)(2). The states must determine how to balance these factors, and how much weight to give each of the factors, on a case-by-case basis.

However, EPA interprets the BART Guidelines as authorizing it to disapprove the State's BART determination based on alleged technical failures to follow each and every paragraph and recommendation in the Guidelines. By relying on isolated instances of alleged deviation from the Guidelines, such an interpretation totally undermines the State's prerogative to determine how to weigh and balance all factors and therefore conflicts directly with the statutory grant of authority to the states to make BART determinations in accordance with all five BART factors. Section 169A does not tell the states how to take the factors into account, nor does it describe how each of the factors must be treated. The provision directing EPA to provide guidelines to the states, id. Section 7491(b)(1), must be read in concert with the broad grant of authority and discretion to states, and does not change the fundamental thrust of the statute. EPA's interpretation that states are constrained to dot every “i” and cross every “t” the way EPA insists directly conflicts with the statute's grant of BART decision making authority to the states. If the BART Guidelines mean what EPA claims they mean, the Guidelines violate the CAA.

Response: As explained elsewhere in this document, we disagree with the commenter's assertions. The CAA does not give states unlimited discretion to determine BART; EPA retains the same supervisory role it has with respect to any SIP submission. We also disagree that our proposal is inconsistent with the American Corn Growers decision. We have determined that Wyoming utilized flawed cost assessments and incorrectly estimated the visibility impacts of controls. We have determined these issues resulted in non-approvable BART determinations for the units for which we proposed a FIP. We recognize the State's broad authority over BART determinations, and recognize the State's authority to attribute weight and significance to the statutory factors in making BART determinations. As a separate matter, however, a state's BART determination must be reasoned and based on an adequate record. Although we have largely approved the State's regional haze SIP, we cannot agree that CAA requirements are satisfied with respect to certain specific BART determinations and other necessary FIP elements.[62]

Comment: The BART Guidelines provide that the “basis for equipment costs estimates” should be documented. Id. at 39166. The Guidelines give states the option of using “data supplied by an equipment vendor (i.e., budget estimates or bids) or by a referenced source (such as the Cost Manual, fifth Edition, February 1996, EPA 453/B-96-001).” Id. 3.

In footnote language, the Guidelines reiterate that costs should be documented, including “any information supplied by vendors that affects your assumptions regarding purchased equipment costs, equipment life, replacement of major components, and any other element of the calculation that differs from the Control Cost Manual.” Id. at 39167 n.15. EPA relies heavily on this footnote to assert that states, including Wyoming, have failed to comply with the Guidelines because they have not adequately documented strict compliance with the CCM. This is an erroneous and unreasonable interpretation of the Guidelines. When read in conjunction with the CAA- which bestows substantial discretion on the states in making BART Start Printed Page 5087determinations—and other statements made in the BART Guidelines and the preamble, this footnote language does not require states to supply vendor quotes or other specific information documenting every single deviation from the CCM, nor does it confer authority on EPA to reject a state's BART determination when the state fails to do so. Cf. United Savings Ass'n v. Timbers of Inwood Forest Assocs., 484 U.S. 365, 371 (1988) (a provision read in isolation “is often clarified by the remainder of the statutory scheme . . . because only one of the permissible meanings produces a substantive effect that is compatible with the rest of the law”); United States v. Boisdore's Heirs, 49 U.S. 113, 122 (1850) (“[W]e must not be guided by a single sentence or member of a sentence, but look to the provisions of the whole law, and to its object and policy.”).

Treating the CCM as a binding checklist conflicts with the CAA, both in a general sense, by attempting to mandate exactly how a state must evaluate and apply the five BART factors, and in a specific sense, by excluding certain costs from consideration in a BART analysis in the face of statutory language mandating that BART be determined based on the actual “costs of compliance,” not some artificial costs of compliance. As to the first issue, EPA itself has recognized that the CCM is “a good reference tool,” which can be supplemented “if there are elements or sources that are not addressed by the Control Cost Manual or there are additional cost methods that could be used.” 70 FR at 39127. “States have flexibility in how they calculate costs,” which is not appropriately circumscribed by recommendations set out in a non-binding manual. See id. See also id. at 39153 (States retain discretion in considering “a number of the factors set forth in section 169A(g)(2), including the costs of compliance”). As to the second issue, EPA cannot cite to or rely upon the CCM to challenge any decision by the states taking into account actual rather than theoretical costs, because the statute requires that real costs be considered. CAA Section 169A(g)(2), 42 U.S.C. 7491(g)(2). The CCM does not impose binding obligations on states undertaking BART determinations, and failure to comply with its overly general and non-source specific recommendations is not grounds for rejection of a state's analysis of the costs of compliance.

Additionally, the CCM has not been subject to notice and comment under the APA, 5 U.S.C. 701-706; it has not been published in the Code of Federal Regulations (CFR); and it is not formally incorporated by reference into the BART Guidelines. Therefore, it is merely a policy statement that is not binding on the states. Furthermore, simply referencing the CCM in the BART Guidelines is not adequate to make that non-binding guidance document legally enforceable. “Agency statements `having general applicability and legal effect' are to be published in the Code of Federal Regulations.” NRDC v. EPA, 559 F.3d 561, 564 (D.C. Cir. 2009), citing 1 CFR 8.1(a). See also Brock v. Cathedral Bluffs Shale Co., 796 F.2d 533, 539 (D.C. Cir. 1986) (“The real dividing point between regulations and general statements of policy is publication in the Code of Federal Regulations, which the statute authorizes to contains only documents `having general applicability and legal effect . . .' ”) (emphasis in original). Accordingly, EPA's assertion that a state has failed to comply with the BART Guidelines by using costing methodology other than that set forth in the CCM is contrary to federal law and is arbitrary and capricious.

Federal regulations require that in order for material to be formally incorporated by reference into the Federal Register and the CFR, EPA must seek approval from the Director of the Federal Register. 1 CFR 51.1. Documents are eligible for incorporation only if they meet certain criteria; incorporation of a document “produced by the same agency that is seeking its approval” is generally inappropriate unless the Director of the Federal Register finds that the document also “possess[es] other unique or highly unusual qualities.” Id. Section 51.7(a)-(b). Furthermore, language incorporating a publication by reference must be “as precise and complete as possible,” including a statement that the document is “incorporated by reference” and “[i]nform[ing] the user that the incorporated publication is a requirement.” Id. Section 51.9(b)(1), (3). Finally, dynamic incorporations into the CFR are prohibited. Id. Section 51.1(f) (“Incorporation by reference of a publication is limited to the edition of the publication that is approved. Future amendments or revisions of the publication are not included.”). See also 76 FR 33590, 33593 (June 8, 2011) (OSHA noting that “it cannot incorporate by reference the latest editions of consensus standards without undertaking new rulemaking because such action would . . . deprive the public of the notice-and-comment period required by law”).

EPA has not complied with the requirements for incorporating the CCM into the regulations directing states to undertake BART Determinations or into the BART Guidelines. The regulations make no mention of the CCM. The BART Guidelines reference the CCM, but do not indicate that EPA was seeking approval for incorporation by reference; and, in any event, it is unlikely that the CCM meets the requirements for incorporation by reference. Additionally, the Guidelines reference the 5th edition of the CCM but direct states to use the most recent version of the CCM, 70 FR 39167 n.14, and dynamic incorporation is expressly prohibited by the regulations governing incorporation by reference, 1 CFR 51.1(f). Where EPA has failed to comply with the requirements for incorporation by reference, the referenced material is “ineffective to impose obligations upon, or to adversely affect” third parties. NRDC v. Train, 566 F.2d 451, 457 (D.C. Cir. 1977). Therefore, the CCM does not constitute binding law, and EPA has no authority to reject Wyoming's BART determinations on grounds the State allegedly strayed from the CCM's cost methodology.

Response: EPA disagrees with this comment. First, with regards to notice-and-comment procedures, the BART Guidelines, including the references within them to the Control Cost Manual, have gone through appropriate public comment procedures and the time to challenge the BART Guidelines' references to the CCM has passed. If the commenter believes the BART Guidelines improperly incorporated by reference the CCM, the commenter could have requested judicial review within 60 days of the publication of the BART Guidelines in the Federal Register. We note that the BART Guidelines have indeed been published in the Code of Federal Regulations, in Appendix Y to Part 51 of Title 40. In addition, the reference to the CCM in Appendix Y provides adequate notice to the public that EPA intended the most recent version of the CCM to be used, and provides a link to the CCM itself.

Moreover, the very action that we are completing today has gone through notice-and-comment procedures. Thus, the public has had full opportunity to comment on our application of the CCM. Furthermore, the commenter's arguments that incorporation by reference is necessary for anything with binding legal effect miss the mark. The BART Guidelines do not contain a legally binding requirement to use the CCM, because as we explain next, the Guidelines clearly state that states may deviate from the CCM.

Commenter mischaracterizes EPA's use and application of the Control Cost Manual. EPA's revised cost-effectiveness values are consistent with Start Printed Page 5088CAA and RHR requirements. EPA explained in issuing the BART Guidelines that “[s]tates have flexibility in how they calculate costs.” See 70 FR at 39127 (July 6, 2005). A state may deviate from the Control Cost Manual provided its analysis is reasonable and the deviations are documented. Here, as discussed elsewhere in this document, Wyoming's cost-effectiveness values were not reasonable. We disagree with commenter's view that our cost analysis is improper, but we agree that the CCM is not the only source of information for the BART analysis. For instance, the reference to the CCM in the BART Guidelines clearly recognizes the potential limitations of the CCM and the need to consider additional information sources:

The basis for equipment cost estimates also should be documented, either with data supplied by an equipment vendor (i.e., budget estimates or bids) or by a referenced source (such as the OAQPS Control Cost Manual, Fifth Edition, February 1996, EPA 453/B-96-001). In order to maintain and improve consistency, cost estimates should be based on the OAQPS Control Cost Manual, where possible. The Control Cost Manual addresses most control technologies in sufficient detail for a BART analysis. The cost analysis should also take into account any site-specific design or other conditions identified above that affect the cost of a particular BART technology option.[63]

As to unusual circumstances, the BART Guidelines call for “documentation” to be provided for “any unusual circumstances that exist for the source that would lead to cost-effectiveness estimates that would exceed that for recent retrofits,[64] which as discussed elsewhere in this final notice were not provided.

Comment: If EPA is making a BART determination as part of a FIP, it must comply with the RHR. Section 169A(g)(2) requires the BART determination to take into consideration five statutory factors. These factors “were meant to be considered together” to arrive at a single judgment: a BART emission limit. Am. Corn Growers, 291 F.3d at 6. EPA's proposed FIP, however, does not present a discussion, finding, or evaluation of the five statutory factors taken together. Instead, EPA merely states that it proposes to find that Wyoming's BART analysis fulfills all of the BART requirements except as to cost-effectiveness and visibility benefits. EPA then proposes to engraft onto Wyoming's consideration of the five statutory BART factors its own cost-effectiveness and visibility analysis, to arrive at the conclusion that SCR is BART. This fails to comply with the statute. The selection of the BART emission limit is arrived at by considering all five BART factors taken together. This requires, for example, that the selection of SCR as BART represents an acceptable balancing of energy and non-air quality environmental factors. When Wyoming made this assessment, however, it was considering LNBs and OFA, and thus its conclusion—which EPA proposes to approve—noted that “combustion control using LNB with OFA does not require non-air quality environmental mitigation for the use of chemical reagents (i.e., ammonia or urea) and there is a minimal energy impact.” This weighing of statutory factors does not discuss or apply SCR, and therefore cannot be adopted by EPA to support its own BART emissions limit in its FIP. EPA is therefore proposing a BART emission limit without independently considering the five statutory BART factors, in violation of Section 169A(g)(2).

Nor does EPA articulate any reasoning supporting its proposed BART emission limit that applies all of the statutory factors. This violates EPA's obligation to cogently explain and articulate each step in its reasoning for proposed action. State Farm, 463 U.S. at 48 (“[A]n agency must cogently explain why it has exercised its discretion in a given manner.”). In fact, even as to the cost-effectiveness and visibility improvements EPA relies upon for its BART emission limit, EPA states that they are adopted because they are “in the range of what EPA has found reasonable for BART in other SIP and FIP actions.” 78 FR 34776. But EPA does not identify which “actions” it is talking about, EPA does not show how the five factors considered in those other “actions” make those “actions” comparable this action, and EPA does not pay even minimal lip service to the statutory requirement that emission limits must be based upon local considerations arrived at by a careful weighing of statutory factors unique in each case. EPA is just selecting a preferred technology (SCR) because it considers the cost of such technology to be acceptable to impose upon Basin Electric, without regard to whether, when considered for its impacts locally in Wyoming as Congress intended, it is the “best” control option for all of the circumstances fully considered. This violates five-factor decision-making process required by the CAA.

Response: We disagree with this comment. Contrary to commenter's assertions, EPA selected the BART emission limits by considering all five BART factors taken together and has complied with CAA and RHR requirements. As discussed in our proposal (see for example discussion starting at 78 FR 34774) and in our response to comments in this action (see sections V.B, V.C, and V.D), we clearly consider all five factors.

6. Reasonableness Standard

Comment: EPA cannot sidestep the CAA's mandate for state discretion by developing and applying a new “reasonableness” standard for evaluating and rejecting that discretion. EPA's regional haze FIP action, however, does just that. For example, EPA incorrectly declared “the state's BART analysis and determination must be reasonable in light of the overarching purpose of the regional haze program.” (See 78 FR 34743) This overly broad and illegal “reasonableness” standard allows EPA to reject any BART determination that EPA dislikes by merely arguing that a state's BART determination is “unreasonable” and without comparing the state's determination to any firm or fixed standards. EPA's “reasonableness” standard requires statutory and regulatory limitations on EPA's authority to disapprove a reasoned RH SIP. The fallacy of EPA's improper reasonableness standard is made even more apparent in its application by EPA, which simply rejects as “unreasonable” many of Wyoming's BART-related decisions without offering sufficient justification of why that is the case.

In creating and employing its reasonableness standard, EPA goes to an even greater extreme by defining “reasonable” in the most self-serving manner imaginable. In short, EPA defines “reasonable” to mean that EPA agrees with the state's exercise of discretion, and it defines “unreasonable” to mean EPA does not agree with the state. (See e.g., 78 FR 34,767, where EPA substitutes its consideration of costs and visibility improvement for Wyoming's). In this way, EPA attempts to bootstrap itself into the role of the sole decision-maker of what is BART and what is not. The CAA does not countenance such overreaching by EPA. For all of the criticism that EPA makes concerning the state's analyses, the reality is that the results of the analyses of both agencies are very similar. In some cases, EPA's numbers (such as the cost of SNCR at Wyodak) provide less of a justification for EPA's chosen BART controls than Wyoming's numbers did in its analyses. However, EPA has used its broad and unjustified criticisms of the State's work to discredit the State's studies and Start Printed Page 5089usurp the discretion the State has applied to its BART determinations.

We also received numerous earlier comments pertaining to EPA's use of a “reasonableness” standard for evaluating BART determinations. For example, commenters pointed out that EPA incorrectly declared “the State's BART analysis and determination must be reasonable in light of the overarching purpose of the regional haze program.” Commenters asserted that the fallacy of this improper reasonableness standard is apparent in its application by EPA, which simply rejects as “unreasonable” many of Wyoming's BART-related decisions without offering a sufficient explanation of why that is the case. Commenters state that EPA makes no attempt to explain how any of Wyoming's BART determinations are “unreasonable,” but simply decrees that they are unsupported by any comparison to any standards, regulations, or statutes.

Commenters argued that the reasonableness standard employed by EPA is not found in the CAA, the RHR, its Preamble, or Appendix Y. Commenters go on to point out that nowhere does EPA define or explain what constitutes “reasonable in light of the overarching purpose of the regional haze program”, and that this standard has not been defined or subjected to notice and comment rulemaking. Commenters pointed out that the CAA does not authorize EPA to adopt and employ “a reasonable in light of the overarching purpose of the regional haze program” criterion for approving or disapproving a state BART determination as CAA Section169A(b)(2)(A) only requires the State to consider five statutory factors. Commenters asserted that the CAA does not impose an additional requirement that the final BART determination is “reasonable in light of the overarching purpose of the regional haze program” as determined by EPA and as such EPA's imposition of this additional criterion is therefore lacking in statutory authority. One commenter stated that there are no numerical minimums that emission rates much achieve in a BART determination and there are no statutory minimum “visibility improvement” obligations.

One commenter went on to point that the failure to define how it will determine reasonableness leads to inconsistent and subjective agency action, as illustrated by EPA's inconsistent treatment of BART decisions around the country. The commenter pointed to BART decisions in Oklahoma, North Dakota, and Nevada as examples where EPA's failure to define reasonableness has led to inconsistent BART decisions.

Another commenter argued that throughout its proposal, EPA claims to have reviewed Wyoming's SIP under a “reasonableness” standard. See, e.g., 78 FR 34776 (“we do not consider Wyoming's analyses . . . to be reasonable”); see also id. at 34778. EPA apparently believes that this standard allows EPA to substitute its judgment for the State's whenever EPA generally alleges that the State's conclusions or methods are not reasonable. Yet EPA cites no statutory or regulatory authority to support its malleable application of this “reasonableness” standard of review. EPA appears to have crafted its flexible reasonableness standard from Alaska Department of Environmental Conservation v. EPA, 540 U.S. 461 (2004). That case stands for the proposition that EPA has authority to reject a state decision that “is not based on a reasoned analysis[.]” Id. at 490 (internal quotation omitted). EPA has misapplied that standard in its proposal to disapprove Wyoming's SIP.

The commenter further argued that the ADEC standard does not allow EPA to disapprove SIPs whenever, in EPA's opinion, some element of the SIP is not reasonable. Instead, EPA must provide SIPs “considerable leeway” and may not “second guess” state decisions[.]” ADEC, at 490 (internal citation omitted). Accordingly, EPA may disapprove a SIP under ADEC only by showing that the SIP is arbitrary. See id. at 490-91. EPA therefore must defer to the Wyoming's determinations in the SIP, and may not simply substitute its judgment for the State's. And, of course, EPA carries the burdens of production and persuasion to show that the State acted unreasonably in light of the statutes and administrative record. Id. at 494.

The commenter asserted that EPA has failed to carry those burdens in its proposed partial disapproval of Wyoming's regional haze SIP. The administrative record demonstrates that Wyoming's SIP will achieve the statutory goal of reasonable progress. EPA has not shown otherwise. EPA has shown only that if it had crafted the implementation plan in the first instance, it would have done so differently than Wyoming did. But the law does not allow EPA to simply substitute EPA's preferences for the State's. Before EPA can disapprove the SIP, it must show that the SIP is arbitrary, in light of the statutes and the record, and with consideration for the deference owed the State's determinations. For example, with respect to Jim Bridger Units 1 and 2 the only meaningful difference in outcomes between EPA's proposed FIP and the SIP is a roughly five-year period in which EPA's proposed controls will result in lesser emissions, though without a perceptible visibility improvement. Save for this distinction, the SIP and FIP create essentially equal improvements in visibility. EPA does not explain why a reduction in NOX emissions that is more expensive but not more effective at improving visibility is more reasonable than the SIP. That lack of explanation renders EPA's proposal arbitrary, and decidedly “unreasonable.”

Response: EPA disagrees with this comment. The CAA requires states to submit SIPs that contain such measures as may be necessary to make reasonable progress toward achieving natural visibility conditions, including BART. The CAA accordingly requires the states to submit a regional haze SIP that includes BART as one necessary measure for achieving natural visibility conditions. In view of the statutory language, it is logical that the reasonableness of the State's BART analysis and determination would be evaluated in light of the purpose of the regional haze program. In addition, our regional haze regulations, at 40 CFR 51.308(d)(ii), provide that when a state has established a RPG that provides for a slower rate of improvement in visibility than the URP (as has Wyoming), the state must demonstrate, based on the reasonable progress factors—i.e., costs of compliance, time necessary for compliance, energy and non-air quality environmental impacts of compliance, and remaining useful life of affected sources—that the URP to attain natural visibility conditions by 2064 is not reasonable and that the progress goal adopted by the state is reasonable. 40 CFR 51.308(d)(iii) provides that, “in determining whether the State's goal for visibility improvement provides for reasonable progress towards natural visibility conditions, the Administrator will evaluate” the state's demonstrations under section 51.308(d)(ii). It is clear that our regulations and the CAA require that we review the reasonableness of the State's BART determinations in light of the goal of achieving natural visibility conditions. This approach is also inherent in our role as the administrative agency empowered to review and approve SIPs. Thus, we are not establishing a new reasonableness standard, as the commenter asserts. As we discuss elsewhere, ADEC supports the use of this standard, and does not require EPA to apply a sort of “arbitrary and Start Printed Page 5090capricious” standard in reviewing Wyoming's SIP submittal. The language regarding the burdens of production and persuasion in ADEC are inapplicable, as they refer to a litigation context that is not present here.

Furthermore, this is a SIP review action, and we believe that EPA is not only authorized, but required to exercise independent technical judgment in evaluating the adequacy of the State's regional haze SIP, including its BART determinations, just as EPA must exercise such judgment in evaluating other SIPs. In evaluating other SIPs, EPA is constantly exercising judgment about SIP adequacy, not just to meet and maintain the NAAQS, but also to meet other requirements that do not have a numeric value. In this case, Congress did not establish NAAQS by which to measure visibility improvement; instead, it established a reasonable progress standard and required that EPA assure that such progress be achieved. Here, contrary to the commenter's assertion, we are exercising judgment within the parameters laid out in the CAA and our regulations. Our interpretation of our regulations and of the CAA, and our technical judgments, are entitled to deference. See, e.g., Michigan Dep't. of Envtl. Quality v. Browner, 230 F.3d 181 (6th Cir. 2000); Connecticut Fund for the Env't., Inc. v. EPA, 696 F.2d 169 (2nd Cir. 1982); Voyageurs Nat'l Park Ass'n v. Norton, 381 F.3d 759 (8th Cir. 2004); Mont. Sulphur & Chem. Co. v. United States EPA, 2012 U.S. App. LEXIS 1056 (9th Cir. Jan. 19, 2012).

Finally, regarding commenters' assertions that we are being inconsistent, because the comment is not specific about what aspect of our proposed disapproval is believed to be inconsistent with other EPA decisions, it is not possible for EPA to address in this response any specific concerns. As articulated in our proposed rulemaking and further explained in our responses to other comments, EPA's partial approval and partial disapproval of the Wyoming regional haze SIP is consistent with the CAA, the RHR, BART Rule, and EPA guidance.

Comment: In the absence of criteria or standards by which “reasonableness” may be assessed, EPA's claim that the State's BART for Laramie River Station is unreasonable is by definition a mere subjective conclusion without basis or foundation. EPA must instead articulate a standard grounded in the statute by which it evaluates and disapproves a SIP and then must support its decision with a plausible explanation connecting the facts to its standard.

Response: The CAA requires states to submit SIPs that contain such measures as may be necessary to make reasonable progress toward achieving natural visibility conditions, including BART. The CAA accordingly requires the states to submit a regional haze SIP that includes BART as one necessary measure for achieving natural visibility conditions. In view of the statutory language, it is reasonable for the State's BART analysis and determination to be evaluated in light of the purpose of the regional haze program.

In addition, our regional haze regulations, at 40 CFR 51.308(d)(ii), provide that when a state has established a RPG that provides for a slower rate of improvement in visibility than the URP (as has Wyoming), the state must demonstrate, based on the reasonable progress factors—i.e., costs of compliance, time necessary for compliance, energy and non-air quality environmental impacts of compliance, and remaining useful life of affected sources—that the rate of progress to attain natural visibility conditions by 2064 is not reasonable and that the progress goal adopted by the state is reasonable. 40 CFR 51.308(d)(iii) provides that, “in determining whether the State's goal for visibility improvement provides for reasonable progress towards natural visibility conditions, the Administrator will evaluate” the state's demonstrations under section 51.308(d)(ii). Therefore, it is clear that our regulations and the CAA require that we review the reasonableness of the State's BART determinations in light of the goal of achieving natural visibility conditions. This approach is also inherent in our role as the administrative agency empowered to review and approve SIPs. Thus, we are not establishing a new reasonableness standard, as the commenter asserts.

Here, Wyoming concluded that a limit of 0.21lb/MMBtu for Laramie River Station could be achieved with operation of LNBs with OFA. As presented in the Introduction section and elsewhere in the notice, the State's regional haze SIP determined that NOX BART for Laramie River Units 1, 2, and 3 is new LNB/SOFA. We proposed to disapprove the State's determination because the State did not reasonably assess the costs of compliance and visibility improvement in accordance with the BART Guidelines. 78 FR 34766. After revising the State's costs and modeling and re-evaluating the statutory factors, we proposed to determine that NOX BART is LNB/SOFA + SCR, with an emissions limit of 0.07 lb/MMBtu for each unit. As the result of the comments received on our proposal, we have further revised our calculation of the costs of compliance and visibility modeling. For example, as explained in the BART section of this document, we corrected cost estimates for elevation and provided detailed comments regarding how site characteristics were addressed using available satellite imagery and why this is a valid approach for providing estimates that are acceptable for BART analysis and consistent with CAA and regulations. While we accepted some of the revised costs, again as explained in the BART section of this document, we did not accept others. For example, we did not accept cost assumptions where the necessary supporting documentation was not provided. After re-evaluating the BART factors, we continue to find that LNB/SOFA + SCR is reasonable as BART and are therefore finalizing our proposal. As a result, we are finalizing our proposed disapproval of the State's NOX BART determination for Laramie River Station and finalizing our proposed FIP that includes a NOX BART determination of LNB/SOFA + SCR, with an emission limit of 0.07 lb/MMBtu (30-day rolling average). The facts presented here and elsewhere in our final notice, provided a basis and foundation, grounded on the CAA and regulations, for the EPA to reach its decision regarding the unreasonableness of Wyoming's BART for Laramie River Station.

Comment: EPA attempted to use post-hoc, immaterial changes that it calculated in costs and visibility improvements to justify usurping Wyoming's BART decision-making authority. This runs counter to the vast discretion EPA has given to other states' regional haze SIPs. In Oregon, for example, despite EPA and Oregon differing in how each calculated BART costs that resulted in cost variance of over $700 per ton, EPA stated that such difference between the two estimates would not materially affect Oregon's evaluation. The difference between the cost analyses under EPA's FIP action and the Wyoming regional haze SIP similarly is immaterial. Similarly, in Colorado, the State's plan included a cost analysis that, according to EPA, was not conducted in accordance with EPA's Control Cost Manual. In addition, EPA explained that Colorado should have more thoroughly considered the visibility impacts of controlling emissions from one BART unit on the various impacted Class I areas and not focused on just the most impacted Class I area. Nevertheless, EPA approved the State's SIP, explaining that “Colorado's Start Printed Page 5091plan achieves a reasonable result overall.” EPA should afford Wyoming the same degree of deference it afforded Colorado and Oregon. As demonstrated by the impacts of the Wyoming SIP, it “achieves a reasonable result overall.”

EPA's inconsistency is not just limited to its disparate actions between states. In Wyoming, EPA acted inconsistently in its BART determinations between sources within the state. For example, EPA accepted Wyoming's cost and visibility BART analyses for FMC Westvaco and General Chemical, along with the PM BART analyses for PacifiCorp's and Basin Electric's BART units. At the same time, EPA rejected the NOX BART cost and visibility analyses for PacifiCorp's and Basin Electric's BART units. Wyoming, however, used the same BART analysis methodology for those BART units at which EPA accepted the Wyoming BART analysis as it did at those BART units for which EPA did not. By rejecting some cost and visibility analyses on the basis that they were improperly performed, while accepting others that were performed in the same manner, EPA acted arbitrarily and capriciously.

Response: We disagree with this comment. In evaluating a State's BART determination, EPA has the discretion to develop additional information, such as cost and visibility analyses. In the end, this additional information, may confirm the State's BART determination as reasonable, or it may lead EPA to disapprove the State's BART determination as unreasonable. However, EPA is not required to develop additional information for all BART determinations in order to review the State's BART determination. If a State's BART determination appears to have reached a reasonable conclusion, taking into account existing information and the potential magnitude or effect of technical flaws in cost or visibility analyses, EPA may approve the BART determination. However, if the potential technical flaws in analyses make it possible that the State's BART determination would be unreasonable, then EPA may develop additional information to try to determine whether the State's BART determination would fall within the range of reasonable outcomes using proper technical analyses. For example, as we explain elsewhere in responding to comments on modeling, in this action EPA was unable to ascertain the visibility benefits of individual NOX controls for the PacifiCorp units from the State's modeling because the emission reductions for multiple pollutants were modeled together, and therefore we were unable to assess the reasonableness of the State's BART determinations.[65] Similarly, for the Basin Electric units, we were unable to ascertain the visibility benefits of SNCR. For that reason, we developed additional modeling. In some cases, the additional modeling confirmed the reasonableness of the State's decisions while in others it did not.

With respect to the State's PM BART determinations, the dollar per ton costs for higher-than-current levels of control were generally high (regardless of potential flaws in determining those costs), so existing information was adequate to find that the PM BART determinations were reasonable. With respect to FMC Westvaco and General Chemical, the State's modeling (which as we discuss elsewhere used a conservative estimate of background ammonia which would tend to result in an overestimation of visibility impacts) showed fairly low visibility benefits from NOX controls. Based on consideration of the five BART factors, the State selected combustion controls for these BART sources. EPA also finds these determinations reasonable, and EPA has no reason (nor does the commenter provide one) to think otherwise.

With respect to the comments regarding Oregon and Colorado, although consistency with similar determinations is one hallmark of reasonableness, the BART determinations are very fact-specific and cannot be easily compared across states. For example, in the Oregon action, EPA noted that (among other things) the source would shutdown in 2020, so “it [was] reasonable for the state to consider the sizable capital cost difference between [two technologies], and the relatively small incremental visibility improvement between the two technologies.” 76 FR 38900. Thus, EPA could assess on the basis of existing information that the State's BART determination was reasonable. With respect to the Colorado SIP, we disagree with the commenter that the Wyoming and Colorado SIPs would achieve comparable visibility improvement.

With respect to consistency generally, in this action we have considered the five factors in the context of each facility. Although one factor (such as visibility improvement or costs of compliance) may be similar for a unit in another state, each factor must be weighed in the context of the other source-specific BART factors.

Comment: Section 169A(g)(2) of the CAA requires states, in determining BART, to “take into consideration the costs of compliance, the energy and non-air quality environmental impacts of compliance, any existing air pollution control technology in use at the source, the remaining useful life of the source, and the degree of visibility improvement which may reasonably be anticipated to result from the use of such technology.” 42 U.S.C. 7491(g)(2). The CAA does not require the achievement of any specific degree of visibility improvement, and only requires that a BART determination eliminate or reduce impairment to visibility. See id. Section 7491. If the state's determination does so, the state has complied with the statute and nothing authorizes EPA to propose or impose its own BART decision.

EPA's proposed action, however, articulates a number of additional grounds that must be met for a SIP to be “approvable.” These additional grounds are not found in the text of the CAA and have never been defined or promulgated with notice and comment rulemaking. For example, EPA's proposed action articulates a two pronged test for BART SIP approval: First, “a state must meet the requirements of the CAA and our regulations for selection of BART”; and then second, “the state's BART analysis and determination must be reasonable in light of the overarching purpose of the regional haze program.” 78 FR 34743.

Basin Electric has no problem with the first prong of this test, i.e., that a state's SIP must “meet the requirements of the CAA” and “any [applicable] regulations”—so long as those regulations are confined to the areas Congress allowed EPA to regulate. However, the second prong, i.e., that “the State's BART analysis and determination must be reasonable in light of the overarching purpose of the regional haze program,” sets out a new “reasonableness” obligation that is neither defined in nor separately set forth in the Act. Essentially, EPA is proposing to measure a BART determination not just against the statutory criteria but also against EPA's own subjective view whether the result reached is reasonable enough to meet Start Printed Page 5092the “overarching goal” of the Act. But since EPA acknowledges that neither the Act nor the regulations “mandate specific milestones or rates of progress,” 76 FR 58577, EPA's subjective reasonable enough requirement imposes a new legislative standard that either goes beyond or, for the first time, purports to define “the requirements of the Act.” This empowers EPA to disapprove a state BART determination and replace it with its own on reasonableness grounds that have never been defined or vetted through public notice and comment.

The same is true with EPA's assertion that Wyoming did not provide “sufficient documentation.” 78 FR 34749. EPA is asserting the existence of, and then a failure to meet, a “sufficient documentation” requirement that is both undefined and entirely of EPA's own creation. This allows EPA to extend its regulatory reach to determine and impose its own view of BART when a state's reasoning, according to EPA, fails to meet unannounced and undefined legislative criteria. Such an expansion of EPA's substantive powers is illegal. EPA may not employ evaluative criteria that effectively extend or define the reach of the CAA without first subjecting those criteria to public notice and comment. See, e.g., Syncor Int'l Corp. v. Shalala, 127 F.3d 90, 95-96 (D.C. Cir. 1997) (requiring the FDA to subject a rule that extended its regulatory reach to notice and comment before applying it); U.S. Tel. Ass'n v. FCC, 28 F.3d 1232, 1233-34 (D.C. Cir. 1994) (finding that the FCC's application of a new standard was a substantive rule requiring notice and comment); Am. Mining Congress v. Mine Safety & Health Admin., 995 F.2d 1106, 1112 (D.C. Cir. 1993) (outlining the factors to apply in determining when a rule is substantive and thus requires notice and comment).

As the D.C. Circuit Court has explained, when an agency implements a substantive change to its regulations that alters the boundaries of what the agency can regulate, the change must be subject to public notice and comment so that an agency does not expand its power without public involvement. Appalachian Power Co. v. EPA, 208 F.3d 1015, 1024 (D.C. Cir. 2000). The same is true when EPA purports, for the first time, to vet a state SIP revision against criteria of its own making not set forth in the governing statute or the existing regulations. Here, EPA is effectively stating that: (1) The most cost-effective (on a dollar per ton basis) control technology must be selected as BART; (2) a state BART determination must be “adequately justified,” “sufficiently documented,” and “properly made”; and (3) the state's determination must meet EPA's subjective view of “reasonableness” in reaching the non-binding goal of the regional haze Program. Each of these new criteria is outcome determinative, according to EPA, and each must be met for the State to be considered in compliance with Section 169A. As such, these are new legislative rules that cannot be adopted and imposed without first being submitted to notice and comment rulemaking as required by CAA Section 307(d), 4207 U.S.C. 7607.

An important indicator of when public notice is required is that the change would allow the agency to extend its own power: “[A] substantive rule modifies or adds to a legal norm based on the agency's own authority . . . And, it is because the agency is engaged in lawmaking that the APA requires it to comply with notice and comment.” Syncor, 127 F.3d at 95 (emphasis in original). EPA's current proposal to disapprove Wyoming's BART determination does exactly that. EPA uses its own authority to modify the legal norm for reviewing State BART decisions to give itself the ultimate authority to impose its own favored BART standards.

The need for advance rulemaking is particularly acute when EPA interprets and applies a statute that itself establishes no concrete, objective requirements. No specific rates of progress, technologies, or visibility improvements are mandated by the RHR. Unlike review of a SIP, where EPA applies specifically defined ambient concentrations to determine if the SIP should be approved, there are no objective criteria against which to measure the “reasonableness” of any state BART determination with respect to cost and visibility judgments.

Under EPA's self-defined standards, EPA is left with unfettered discretion to disapprove any decision with which it disagrees on the grounds that it is not “reasonable” enough to meet EPA's preferences. This is why the law requires EPA to first define and promulgate rules explaining what is “reasonable” enough, or what is “sufficiently documented” enough, to support a BART determination. Otherwise, EPA can trump state discretion on the basis of internally conceived and unexamined evaluative criteria that extend EPA's reach without public involvement.

Response: EPA disagrees with this comment. First, even assuming that EPA's proposed action on the Wyoming regional haze SIP articulated new grounds for evaluating a regional haze SIP, the proposed action provides the public with the opportunity to comment. As evidenced by the commenter's submission, the commenter had the opportunity to comment on EPA's approach to evaluating the Wyoming regional haze SIP and to identify any concerns associated with the statement at issue from our proposal and other aspects of our action.

Second, the CAA requires states to submit SIPs that contain such measures as may be necessary to make reasonable progress toward achieving natural visibility conditions, including BART. The CAA accordingly requires the states to submit a regional haze SIP that includes BART as one necessary measure for achieving natural visibility conditions. In view of the statutory language, it is reasonable that the State's BART analysis and determination would be evaluated in light of the purpose of the regional haze program. In addition, our regional haze regulations, at 40 CFR 51.308(d)(ii), provide that when a state has established a RPG that provides for a slower rate of improvement in visibility than the URP (as has Wyoming), the state must demonstrate, based on the reasonable progress factors—i.e., costs of compliance, time necessary for compliance, energy and non-air quality environmental impacts of compliance, and remaining useful life of affected sources—that the rate of progress to attain natural visibility conditions by 2064 is not reasonable and that the progress goal adopted by the state is reasonable. 40 CFR 51.308(d)(iii) provides that, “in determining whether the State's goal for visibility improvement provides for reasonable progress towards natural visibility conditions, the Administrator will evaluate” the state's demonstrations under section 51.308(d)(ii). It is clear that our regulations and the CAA require that we review the reasonableness of the State's BART determinations in light of the goal of achieving natural visibility conditions. This approach is also inherent in our role as the administrative agency empowered to review and approve SIPs. Thus, we are not establishing a new reasonableness standard, as the commenter asserts.

As explained above, our proposal clearly laid out the bases for our proposed disapproval of the State's BART and reasonable progress determinations, and we have relied on the standards contained in our regional haze regulations and the authority that Congress granted us to review and determine whether SIPs comply with Start Printed Page 5093the minimum statutory and regulatory requirements. In determining SIP adequacy, we inevitably exercise our judgment and expertise regarding technical issues, and it is entirely appropriate that we do so. Courts have recognized this necessity and deferred to our exercise of discretion when reviewing SIPs.

Finally, we disagree with the argument that we must approve a BART determination where the SIP reflects consideration of the five factors and the BART selection will result in some improvement in visibility. We think Congress expected more when it required the application of “best available retrofit technology.”

Comment: In 2004, EPA represented to the United States Supreme Court that it would act only very rarely to overrule a state decision selecting control technology for specific sources. ADEC. Relying upon this representation to rebut doubts expressed by the dissent, the Supreme Court affirmed EPA's decision to overrule a BACT decision made by the State of Alaska on the grounds that the State's decision was not “reasonable” because the record lacked the information necessary to support the State's cost assessment. The ADEC Court held that EPA could review state BACT determinations to ascertain whether they were “reasonable in light of the statutory guides and the state administrative record.” Id. at 494.

EPA now relies upon the Supreme Court's use of the word “reasonable” in the BACT context to assume authority to judge the “reasonableness” of state BART decisions when reviewing SIP revisions under Section 110, and thus to disapprove any BART determination it considers “unreasonable” “in light of the over-arching purpose of the regional haze program.” 78 FR 34743. This formulation seriously misconstrues and misstates the Supreme Court's holding and runs counter to the CAA's conferral of authority on the State in selecting BART. “Reasonableness in EPA's subjective view” cannot be applied as a rubric for approving state BART decisions, as it allows EPA to impose its own BART preferences. Rather than adhere to the core principles of cooperative federalism codified in the RHR by only rarely overruling state technology choices, EPA instead does exactly what it represented to the Supreme Court it would not do—routinely overrule state determinations—and it does so under the rubric of authority to evaluate “reasonableness” on a subjective basis.

Examination of EPA's action in this and related BART proceedings around the country demonstrates that EPA is not using the “reasonableness” standard that was actually approved in ADEC. Far from endorsing a generic “reasonableness in EPA's view” standard, the ADEC Court echoed the language of APA arbitrary and capricious review and upheld EPA's rejection of a State BACT determination on grounds that the State's determination was not supported by the administrative record. The Court stated that “[o]nly when a state agency's BACT determination is `not based on a reasoned analysis' may EPA step in to ensure that the statutory requirements are honored,” and that the Act “authorizes EPA to act in the unusual case in which a state permitting authority has determined BACT arbitrarily.” 540 U.S. at 490-91; citation omitted). The Court added that “EPA adhered to that limited role here, explaining why ADEC's BACT determination was `arbitrary' and contrary to [the State]'s own findings.” Id. The Court thus held that EPA had properly exercised its authority to reject the State's BACT determination when the State switched from an initial finding that a certain technology was economically feasible to finding that the same technology was economically infeasible with “no factual basis in the record” to support the change. Id. at 496-500.

Here, EPA makes no effort to formulate and apply a “reasonableness” standard that appropriately preserves for EPA only the “limited role” of insuring that a state decision is not arbitrary and capricious and lacking in record support. Instead, EPA scours the record for inconsequential actions taken by states which it can portray as “inconsistent with” the massively complex, out-dated, and non-binding CCM or with the largely advisory Guidelines so that EPA can declare the state's decision to be “unreasonable” and take over the choice of BART technology. EPA does not demonstrate any arbitrary or capricious conduct, any lack of reasoned decision making, or any other documented failure by the State to follow the requirements of the statute, as contemplated by the standard actually approved in ADEC. As a result, EPA is not employing the “reasonableness” test properly, and with that error is arrogating power Congress left to the States, precisely as predicted by the ADEC dissent. In state after state, EPA is now striking down state BART decisions and cloaking its disregard for state primacy by adjudging those decisions as “unreasonable,” purportedly in reliance upon authority granted by ADEC. But it strains credulity for EPA to assert that state after state is making essentially the same repeated arbitrary and capricious decisions, the remedy for which is almost always mandatory imposition by EPA of its preferred technology choice: SCR. EPA's “reasonableness” test is therefore fundamentally erroneous. EPA may not exercise authority “in a manner that is inconsistent with the administrative structure that Congress enacted into law,” ETSI Pipeline Project, 484 U.S. at 517, by applying a subjective reasonableness standard to federalize BART decisions.

Response: We responded to similar comments elsewhere in this document.

Comment: Because EPA's proposed disapproval of Wyoming's BART determination for Laramie River Station is inconsistent with EPA's prior approval of other state BART choices, EPA's proposed decision is an abuse of discretion and not entitled to deference from a reviewing court.

For instance, in a CAA case involving EPA approval of state Title V programs, the Ninth Circuit reversed EPA's disapproval of one state's program where EPA's decision “conflict[ed] substantially with numerous EPA decisions in other states and localities.” W. States Petroleum, 87 F.3d at 282. In that case, EPA had conditioned final approval of Washington's proposed Title V program on the State's repeal of certain insignificant emissions units (“IEU”) exemptions. EPA eventually approved the State's Title V program, but disapproved the IEUs exemptions as inconsistent with the applicable regulations. Id. at 283. Industry members and the State challenged EPA's disapproval on the basis that EPA's decision was inconsistent with its prior interpretation and application of the regulations in other states. Id. at 282-83. Specifically, EPA had condoned the exemption of IEUs from the permit content requirements of the regulations in at least eight other state and local programs. Id. at 283. Based on this evidence, the Ninth Circuit held that EPA's rejection of Washington's IEU rules was “undeniably a change in agency interpretation.” Id. at 284. Accordingly, EPA was required to support its change by “reasoned analysis,” which it did not do. Id. (EPA “may not depart, sub silentio, from its usual rules of decision to reach a different, unexplained result in a single case”). The court held that EPA abuses its discretion where it approves numerous state programs that include the very same aspects forming the basis for EPA's denial of another state's program. Id. at 285. Start Printed Page 5094

Other courts have similarly expressed that an agency acts arbitrarily and capriciously when it departs from prior interpretations or precedent without adequately explaining the reasons for its departure. See, e.g., Cnty. of Los Angeles v. Shalala, 192 F.3d 1005, 1022 (D.C. Cir. 1999) (“A long line of precedent has established that an agency action is arbitrary when the agency offer[s] insufficient reasons for treating similar situations differently.”); Shaw's Supermarkets, Inc. v. N.L.R.B., 884 F.2d 34, 41 (1st Cir. 1989) (“Unless an agency either follows or consciously changes the rules developed in its precedent, those subject to the agency's authority cannot use its precedent as a guide for their conduct; nor will that precedent check arbitrary agency action.”); Puerto Rican Cement Co. v. EPA, 889 F.2d 292, 298 (1st Cir. 1989) (noting “the well-established legal doctrine that an agency `must either follow its own precedents or explain why it departs from them' ”) (citation omitted); Int'l Internship Programs v. Napolitano, 853 F. Supp.2d 86, 94 (D.D.C. 2012) (“[I]f an agency adopts `a new position inconsistent with' an existing regulation, or effects `a substantive change in the regulation,' the agency must comply with the notice and comment requirements of the APA.”) (citation omitted). Moreover, consistency is a factor to be weighed in determining how much deference an agency's interpretation is entitled to receive. Good Samaritan Hosp. v. Shalala, 508 U.S. 402, 417 (1993) (“[T]he consistency of an agency's position is a factor in assessing the weight that position is due.”). When an “Agency's regulations reflect the Agency's own longstanding interpretation,” a court “will normally accord particular deference” to such “interpretation of `longstanding' duration.” Barnhart v. Walton, 535 U.S. 212, 219-20 (2002). But “the case for judicial deference is less compelling with respect to agency positions that are inconsistent with previously held views.” Pauley, 501 U.S. at 698. “An agency interpretation of a relevant provision which conflicts with the agency's earlier interpretation is `entitled to considerably less deference' than a consistently held agency view.” I.N.S. v. Cardoza-Fonseca, 480 U.S. 421, 446 n.30 (1987). See also Watt v. Alaska, 451 U.S. 259, 273 (1981) (“The Department [of Interior]'s current interpretation, being in conflict with its initial position, is entitled to considerably less deference.”); W. States Petroleum, 87 F.3d at 285 (the court “need not defer to the EPA because the EPA has abused its discretion in departing from its own prior standards”).

Here, EPA has taken an inconsistent approach in interpreting the RHR, the Guidelines, and the CCM. In particular, EPA's current interpretation of its role and the states' role under these provisions conflicts with its prior, long-held understanding that states serve the primary role in determining BART and that EPA should not interfere with the many judgments that go into making BART determinations.

More specifically, EPA's application of its improper and subjective “reasonableness” standard when reviewing BART determinations in the SIP approval process has yielded inconsistent, and therefore arbitrary and capricious, results. Here, EPA identifies what it describes as “cost and visibility errors for EGUs” in Wyoming's SIP sufficient to permit EPA to disapprove the BART determination for Laramie River, yet EPA proceeds to approve other Wyoming BART decisions as “reasonable” “despite the[se] . . . errors.” 78 FR 34750. Either EPA is applying the law arbitrarily and capriciously, or it is simply approving as “reasonable” only those choices with which it agrees, either of which is erroneous. EPA must be reasonably consistent in reviewing state BART determinations.

Response: We responded to similar comments above.

Comment: EPA's implementation of the RHR does not satisfy the CAA's requirements of consistency. The conclusions reached by EPA on similar issues vary from case to case in ways that cannot be explained by statute, regulation, or guiding principle. EPA seems to act on BART determinations with an eye towards achieving its desired outcome rather than implementing the CAA even-handedly. This is the definition of caprice. States, regulated entities, and the public are left guessing as to what will be required in any given case. Because EPA has been so inconsistent in the current case and in its overall administration of the RHR, its proposal to disapprove Wyoming's BART determinations for Laramie River and to impose a FIP is arbitrary and capricious and must be abandoned.

Response: We responded to similar comments above.

7. Reliance on Emission Reductions

Comment: EPA's regional haze FIP action is also illegal, arbitrary, and capricious because it relies upon factors outside of the BART five-factor analysis. Nowhere in the five-factor analysis, or anywhere in the Appendix Y Guidelines, is there any support for EPA using an “emissions reduction” factor. But this is exactly what EPA has done in its FIP. For example, EPA cited “emission reductions” as the basis for the FIP BART NOX decisions for Dave Johnston Unit 3 (See 77 FR 33052), Wyodak (See 77 FR 33055) and Laramie River (See 77 FR 33001), among others. In doing so, however, EPA failed to account for the fact that the regional haze program is not an emissions reduction program per se, but is a visibility improvement program.

Additionally, it is improper for EPA to reject Wyoming's BART determinations, which relied upon the proper balancing of all five BART factors, and replace those BART determinations with EPA's analysis, which relied upon factors outside the five-factor analysis, such as emissions reductions. (See e.g., 77 FR at 33,052.) Courts have held that when an agency relies on factors “which Congress has not intended it to consider,” then such action is arbitrary and capricious. Arizona Public Service Co. v. US EPA, 562 F.3d 1116, 1123 (10th Cir. 2009).

Earlier comments asserted that EPA's regional haze FIP is also illegal, arbitrary, and capricious because it relies upon factors outside of the BART five-factor analysis. Nowhere in the five-factor analysis, or anywhere in Appendix Y, is there any support for EPA using an “emissions reduction” factor. But this is exactly what EPA has done. For example, EPA cites “emission reductions” as the basis for the regional haze FIP BART NOX decisions for Dave Johnston Unit 3, Wyodak, and Laramie River Station, among others. In doing so, however, EPA fails to account for the fact that the regional haze program is not an emissions reduction program per se, but is a visibility improvement program.

EPA's over-reliance on “emissions reductions” outside of the mandated BART factors has caused EPA to overstep the boundaries of the regional haze program. This is evidenced by the virtually non-existent visibility improvements associated with SNCR at Wyodak and Dave Johnston that EPA approved because of the associated emission reductions. Additionally, it is improper for EPA to reject Wyoming's BART determinations, which relied upon the proper balancing of all five BART factors, and replace those BART determinations with EPA's analysis, which relied upon factors outside the five-factor analysis. Courts have held that when an agency relies on factors “which Congress has not intended it to consider,” then such action is arbitrary and capricious. Arizona Public Service Start Printed Page 5095Co. v. US EPA, 562 F.3d 1116, 1123 (10th Cir. 2009).

Response: We disagree with the commenter's characterization of the role of emission reductions in the BART analyses. The RHR provides that:

The determination of BART must be based on an analysis of the best system of continuous emission control technology available and associated emission reductions achievable for each BART-eligible source that is subject to BART * * * [66]

Thus, the BART Guidelines clearly contemplate the assessment of emission reductions.

Emission reductions are a consideration in calculating both average and incremental cost effectiveness in order to evaluate the cost of compliance (one of the five factors). 70 FR 39167 and 39168. Contrary to the commenters' assertions, however, our disapproval of Wyoming's DEQ's BART analyses is not “based” on emission reductions, rather the analyses was based on the best system of continuous emission control technology and associated emission reductions achievable, as used in developing the BART factor information. For example, as discussed elsewhere in this section and final notice, contrary to the Guidelines' admonition that “cost estimates should be based on the CCM, where possible,” the control cost calculations supplied by the utilities and relied upon by Wyoming included costs not allowed by the CCM, such as owner's costs and Allowance for Funds Utilized During Construction (AFUDC). Thus, Wyoming's consideration of the “cost of compliance” for these units was not consistent with the Guidelines. Furthermore, as explained elsewhere in this document, Wyoming's consideration of visibility benefits was inconsistent with the Guidelines because the State did not provide visibility modeling from which the visibility improvement from individual controls could be ascertained. EPA's analyses comports with the CAA and RHR requirements; therefore, we did not consider factors outside the Agency's authority.

In regard to EPA's disapproval of Wyoming's BART decisions on five units, EPA's decision was based on a careful weighing of the five factors, including cost of compliance (average and incremental) and visibility improvement. Just because EPA points out the emission reductions does not mean that it has cited “emission reductions” as the only basis for the regional haze FIP BART NOX decisions for these units.

8. Presumptive Limits

Comment: EPA's regional haze FIP is improper because it requires post-combustion NOX controls as BART, when EPA guidelines make clear that only combustion controls for NOX are contemplated. (See e.g. 77 FR at 33,053.) EPA's Preamble and other guidance confirm that the combustion controls of LNBs and OFA (in some form) are “BART technology” for the BART units. In the Preamble and the RHR, EPA stated that, except for cyclone boilers, the “types of current combustion control technology options assumed include low NOX burners, over-fire air, and coal reburning.” 70 FR 39134; see also 39,144 (“For all other coal-fired units, our analysis assumed these units will install current combustion control technology.”) (emphasis added). In fact, in the Technical Support Document used to develop the presumptive BART NOX emissions limits, EPA explained that the “methodology EPA used in applying current combustion control technology to BART-eligible EGUs” included applying “a complete set of combustion controls. A complete set of combustion controls for most units includes a low NOX burner and over-fire air.” (“Technical Support Document, Methodology for Developing NOX Presumptive Limits,” EPA Clean Air Markets Division, pg. 1 (dated June 15, 2005)).

EPA's Preamble and Appendix Y identify post-combustion controls for NOX, such as SCR and SNCR, as “BART technology” for only “cyclone” units. EPA made it clear that for “other units, we are not establishing presumptive limits based on the installation of SCR.” 70 FR 39136. Therefore, EPA's presumptive “BART technology” is LNBs and some type of OFA. EPA further elaborated in the Preamble on SCR costs, stating that although “States may in specific cases find that the use of SCR is appropriate, we have not determined that SCR is generally cost-effective for BART across unit types.” Id.; see also 40 CFR Part 51, Appendix Y, Section IV.E.5.[67] Because EPA improperly requires post-combustion controls in its regional haze FIP, EPA should withdraw this requirement and approve the Wyoming SIP. If EPA desires to impose post-combustion controls as BART NOX, it must first amend Appendix Y through a proper rulemaking procedure.

Commenters further assert that, when EPA issued the RHR, it established presumptive NOX BART limits for power plants based on EPA's conclusions about the cost effectiveness of certain emissions control technologies, including SCR and combustion controls. 70 FR at 39131, 39134-39136. These limits are based on EPA's acknowledgment that NOX controls vary considerably and only in “relatively rare cases” would SCR be appropriate. 69 FR 25184, 25202 (May 5, 2004). EPA's own pronouncement on the cost effectiveness of SCR belie its finding that SCR is cost effective at Laramie River Station.

The presumptive limits for NOX are differentiated by boiler design and type of coal because NOX control technologies are not “one size fits all” and cost effectiveness is variable. Id. at 39134. As EPA noted in proposing presumptive NOX BART limits, “the removal efficiencies and costs associated” with NOX controls “vary considerably, depending upon the design and operating parameters of the particular boiler being analyzed.” 69 FR at 25202. For that reason, EPA proposed (and ultimately finalized) presumptive NOX BART limits that would not require post-combustion controls: “States should require the lowest emission rate that can be achieved without the installation of post-combustion controls” because they are “applicable to most EGUs, are relatively inexpensive, and are already widely Start Printed Page 5096applied.” Id. Indeed, EPA “recognize[d] that a small number of the largest power plants may need to install an SCR unit to meet this control level. In such relatively rare cases, a State, at its discretion, may find SCR to be appropriate if the source causes visibility impacts sufficiently large to warrant the additional capital cost.” Id. (emphasis added).

EPA's presumptive BART determinations for coal-fired EGUs of various boiler configurations demonstrate that emissions control devices with an average cost effectiveness greater than $1,350 per ton are not cost effective. Sargent & Lundy analyzed the presumptive BART limits in EPA's “Technical Support Document for BART NOX Limits for Electric Generating Units Excel Spreadsheet” and EPA's “Technical Support Document—Methodology for Developing BART NOX Presumptive Limits,” and compiled EPA's cost effectiveness thresholds for each boiler design and coal type. Sargent & Lundy, “BART Cost Effectiveness Thresholds” (Jan. 6, 2010). Exhibit 17 to commenter 0148. The report was prepared to supplement North Dakota's BART determination for Basin Electric's Leland Olds Station Units 1 and 2, but is equally applicable to any BART determination for coal-fired utility boilers, including Laramie River Station. Sargent & Lundy concludes that based on EPA's own assumptions about acceptable cost effective levels, “a threshold of $1,350/ton should be used to establish the cost-effectiveness of NOX retrofit control technologies.” Id. at 12.

Sargent & Lundy's report demonstrates that EPA consistently found control technologies to be cost effective if the cost of NOX removal was less than $1,350/ton, and not to be cost effective if the cost of NOX removal was greater than $1,350/ton. Id. at Figure 3 and accompanying text. For example, for all boiler categories other than cyclone units, SCR had an overall average cost effectiveness of $1,749/ton NOX removed and was considered not to be cost effective. Id. at 11. Combustion controls at non-cyclone boilers had an overall average cost effectiveness of $535/ton NOX removed and were found to be cost effective. Id.

SCR is not cost effective at Laramie River Station because it greatly exceeds the $1,350/ton threshold used by EPA in its presumptive BART determinations. EPA's own flawed cost effectiveness analysis estimates that installation of SCR at Laramie River Station would range from between $3,589 and $3,903 per ton of NOX removed—far above the $1,350/ton threshold used in its presumptive BART determination. 78 FR at 34775-34776. EPA does not mention its presumptive BART limits in its proposed disapproval of Wyoming's BART determinations, and offers no explanation for departing from the presumptive levels and the associated use of combustion controls. 78 FR at 34772-34777. Moreover, when Sargent & Lundy estimated costs of SCR at Laramie River Station based on a detailed scoping-level study, it found that costs per ton of NOX removed would range from $8,531 to $9,048, an amount seven times greater than the threshold used in the presumptive BART determination. Sargent & Lundy Evaluation, Table 7. See also Section XVIII.A.

We received numerous comments earlier that EPA's regional haze FIP is improper because the BART units are meeting the presumptive limits in the BART guidelines based on the installation of combustion controls. Commenters go on to assert that the BART Guidelines only require the installation of LNBs/OFA and that EPA determined in the guidelines that SCR was generally not cost-effective for BART. One commenter noted that EPA has completely ignored the presumptive BART limits in the proposed action and that this is contrary to the express requirements in both the RHR and the BART Guidelines. The commenter goes on to say that EPA's attempt to completely ignore the BART limits makes the presumptive BART limits meaningless and this is contrary to the requirements of the CAA and the clear intent of the BART Guidelines. One commenter asserted that the BART Guidelines show that an alternative analysis is required only when a source cannot meet the presumptive limits, and that while a state may choose to establish a limit that is more stringent than the BART limit, there is nothing in the BART Guidelines that would require a state to do so.

Commenters asserted that EPA adopted the presumptive BART limits to establish the specific control levels required for EGUs. Commenters point out that EPA has not repealed the presumptive limits from the promulgated BART Guidelines, but in this action EPA does not even deign to acknowledge the existence of the presumptive limits, as if the presumptive BART limits were no longer a binding regulation. Commenters argued that unless and until EPA goes through notice and comment rulemaking to remove the presumptive emissions limits and establish other requirements consistent with the CAA, then EPA must approve a state's BART determination that meets the presumptive regulatory limits.

Response: We disagree with the commenters. The CAA states the following regarding emission limits for fossil-fuel fired generating power plants having a total generating capacity in excess of 750 MW:

In the case of a fossil-fuel fired generating power plant having a total generating capacity in excess of 750 megawatts, the emissions limitations required under this paragraph shall be determined pursuant to guidelines, promulgated by the Administrator under paragraph (1).

EPA disagrees that the CAA mandates specific control levels (i.e., presumptive emission limits) for power plants with a total generating capacity of 750 MW or greater. Rather, the CAA directed EPA to develop guidelines for states to establish BART emission limits, and required that power plants having a total generating capacity in excess of 750 MW follow the guidelines when establishing BART emission limits. In response, in 2005 EPA promulgated the BART Guidelines, which provide a detailed description of how a state must approach the BART determination process for certain large EGUs, and required that the determination of fossil-fuel fired power plants having a total generating capacity greater than 750 MW must be made pursuant to the BART Guidelines. As such, the plain reading of the CAA language makes it clear the intent was to make the BART Guidelines mandatory for EGUs larger than 750 MW, as opposed to presumptive limits.

Compliance with EPA's “presumptive” NOX emission limits does not excuse a state from performing such an analysis, because the presumptive limits serve as a floor, not a ceiling, for BART. Furthermore, the presumptive limits in the Guidelines do not supplant the Act's mandate to consider the five statutory factors, as codified in the RHR.[68] Additionally, commenters provide no showing that the assumptions underlying EPA's older, generic calculations representative of hundreds of plants in fact represent BART, under current circumstances, at these particular plants. Moreover, far from rendering the Start Printed Page 5097presumptive limits “meaningless,” EPA's interpretation is in fact necessary to effectuate the purpose of the Haze Rule. The fundamental purpose of the BART requirement is to determine the “best system of continuous emission control technology available and associated emission reductions achievable for each BART-eligible source.” 40 CFR 51.308(e)(1)(ii)(A) (emphasis added). To allow states to adopt the presumptive limits without any assessment of whether those limits represent the “best” control for a particular EGU at the time of the determination would be unreasonable in light of the overarching purpose of the Haze Rule and the CAA's visibility requirements. The presumptive limits ensure that states aim to achieve, at a minimum, the level of emissions reduction that was available and cost-effective at the time the BART Guidelines were adopted. EPA elaborated in the BART Guidelines themselves, clarifying that the Agency expected states to not only meet, but in appropriate cases exceed the presumptive limits: “While these [presumptive] levels may represent current control capabilities, we expect that scrubber technology will continue to improve and control costs continue to decline. You should be sure to consider the level of control that is currently best achievable at the time that you are conducting your BART analysis.” 40 CFR part 51, App. Y, at IV.E.4 (emphasis added). Therefore, EPA's proposed rulemaking on the Wyoming regional haze SIP is not contrary to the requirements of the CAA and regulations.

Additionally, for each source subject to BART, the RHR, at 40 CFR 51.308(e)(1)(ii)(A), requires that states identify the level of control representing BART after considering the factors set out in CAA section 169A(g), as follows: “[s]tates must identify the best system of continuous emission control technology for each source subject to BART taking into account the technology available, the costs of compliance, the energy and non-air quality environmental impacts of compliance, any pollution control equipment in use at the source, the remaining useful life of the source, and the degree of visibility improvement that may be expected from available control technology.” 70 FR 39158. In other words, the presumptive limits do not obviate the need to identify the best system of continuous emission control technology on a case-by-case basis considering the five factors. A state may not simply “stop” its evaluation of potential control levels at the presumptive level of control if more stringent control technologies or limits are technically feasible. We do not read the BART guidelines in appendix Y to contradict the requirement in our regulations to determine “the degree of reduction achievable through the application of the best system of continuous emission reduction” “on a case-by-case basis,” considering the five factors. 40 CFR 51.301 (definition of Best Available Retrofit Technology); 40 CFR 51.308(e).

Also, our interpretation is supported by the following language in our BART guidelines: “While these levels may represent current control capabilities, we expect that scrubber technology will continue to improve and control costs continue to decline. You should be sure to consider the level of control that is currently best achievable at the time that you are conducting your BART analysis.” 70 FR 39171.

The presumptive limits are meaningful as indicating a level of control that EPA generally considered achievable and cost effective at the time it adopted the BART guidelines in 2005, but not a value that a state could adopt without conducting a five factor analysis considering more stringent, technically feasible levels of control.

Commenters focus on narrow passages of the BART guidelines to support their view that the presumptive limits represent the most stringent BART controls that EPA can require for regional haze. However, these passages must be reconciled with the language of the RHR cited above, as well as other passages of the BART guidelines and associated preamble. A central concept expressed in the guidelines is that a state is not required to consider the five factors if it has selected the most stringent level of control; otherwise, a state must fully consider the five factors in determining BART. 40 CFR part 51, appendix Y, section IV.D.1, step 1.9.

Undoubtedly, as the commenters note, the presumptive limits for NOX represent cost effective controls, but it is well-understood that limits based on combustion controls do not represent the most stringent level of control for NOX. Thus, a state which selects combustion controls and the associated presumptive limit for NOX as BART may only do so after rejecting more stringent control technologies based on full consideration of the five factors. Our interpretation reasonably reconciles the various provisions of our regulations. We have clearly communicated our views on this subject in other states, and, following our interpretation, Wyoming conducted an analysis of control technologies that would achieve a more stringent limit than combustion controls.

In promulgating a FIP for the Wyoming BART sources, we arrived at an emission limit based on consideration of the five factors. Contrary to the commenter's suggestion, EPA's BART guidelines do not establish a presumptive cost effectiveness level that is a “safe harbor” or “shield” for state BART determinations, or that EPA, when promulgating a FIP, may not exceed in determining BART. Once a FIP is required, we stand in the state's shoes. This is not EPA establishing a new presumptive limit or national rule; it is EPA, acting in the State's shoes, conducting a reasonable source-specific consideration of cost and the other regulatory factors.

9. Compliance With 40 CFR 51.308

Comment: EPA should have judged Wyoming's BART determinations on the basis of whether or not the Wyoming BART determinations are “necessary” to make “reasonable progress.” EPA's RHRs provide two regulatory paths to address regional haze. (See 77 FR 30953, 30957 (May 24, 2012).) “One is 40 CFR 51.308, requiring states to perform individual point source BART determinations and evaluate the need for other control strategies.” Id. “The other method for addressing regional haze is through 40 CFR 51.309, and is an option for nine states termed the `Transport Region States' which include: . . . Wyoming, . . . By meeting the requirements under 40 CFR 51.309, states are making reasonable progress toward the national goal of achieving natural visibility conditions for the 16 Class I areas on the Colorado Plateau.” Id. Wyoming submitted the Wyoming regional haze SIPs under Section 309. Therefore, the requirements of Section 308 only apply to the extent required by Section 309. Importantly, NOX emissions and controls under Section 309 are treated differently than NOX emissions and controls under Section 308. This is because Congress and EPA purposefully focused Section 309 on addressing the issue of SO2 emissions, the predominant cause of regional haze on the Colorado Plateau in the western US. By contrast, Section 309 recognizes that NOX emissions have a significantly smaller impact on visibility on the Colorado Plateau. In fact, the WRAP report estimated that “stationary source NOX emissions result in nitrates that probably cause about 2 to 5 percent of the impairment on the Colorado Plateau.” Several illustrations in the WRAP NOX report show that nitrate emissions have very little impact on Class I areas in or near Utah and Start Printed Page 5098Wyoming. The WRAP report also explains that “NOX controls will have a relatively small impact on PM and visibility in the West.”

The Wyoming SIP, including BART determinations for NOX, is consistent with the WRAP's NOX information, and also properly acknowledges the relatively small impact nitrates from stationary sources like PacifiCorp's BART units have on visibility impairment in Wyoming. Wyoming's SIP, page 62, states that “the majority of nitrate stems from mobile sources.” The SIP also explains that in all but one Class I area “contributions from other states and Canada are much larger than contributions from inside Wyoming.” Id. Wyoming correctly determined, consistent with the WRAP reports and other data, that controlling NOX emissions from stationary sources like PacifiCorp's BART units would yield very little visibility improvement in Wyoming. EPA's own regional haze visibility map shows that visibility in Wyoming is among the best in the country.

In light of the above information, it is understandable that Section 309 focuses on addressing SO2 emissions. Indeed, WRAP focused their efforts primarily on SO2 emissions because the research indicated this pollutant had the greatest impact on visibility. “Recommendations for Improving Western Vistas,” authored by the Grand Canyon Visibility Transport Commission, (June 10, 1996) at page 32 (identifying sulfates as “the most significant contributor to visibility impairment” from stationary sources). In a separate action, EPA acknowledged that Wyoming has complied with the Section 309's SO2 requirements and made great progress towards improving and protecting visibility as a result. For all of these reasons, Section 309 takes a different approach to NOX emissions than does Section 308, placing much less emphasis on the need for significant reductions in NOX emissions and instead focusing almost all attention and resources in the western U.S. on reducing SO2 emissions. EPA's FIP, with its incredibly expensive and unneeded NOX control equipment, ignored the focus and intent of Section 309 and refused to acknowledge the discretion available to Wyoming to balance this information in making its BART determinations.

Additionally, as a result of the lesser emphasis in Section 309 on NOX emissions, Section 51.309(d)(4)(vii) requires a regional haze SIP to “contain any necessary long term strategies and BART requirements for stationary source . . . NOX emissions.” Section 308, by contrast, does not include a similar “necessary to achieve reasonable progress” threshold for BART. The difference between the two requirements is both intentional and meaningful. If a state like Wyoming finds that a particular BART requirement is not “necessary” to make “reasonable progress,” then that BART requirement should not be required as part of the regional haze SIP. This interpretation is supported by EPA's own position in Central Arizona Water Conservancy District v. United States, 990 F.2d 1531 (9th Cir. 1993). There, “EPA chose not to adopt the emission control limits indicated by the BART analysis, but instead to adopt an emissions limitations standard that would produce greater visibility improvement at a lower cost.” Id. at 1543. The court agreed with EPA, stating that “Congress's use of the term `including' in Section 7491(b)(2) prior to its listing BART as a method of attaining `reasonable progress' supports EPA's position that it has the discretion to adopt implementation plan provisions other than those provided by BART analyses in situations where the agency reasonably concludes that more `reasonable progress' will thereby be attained.” Id. This same rationale applies to the term “necessary” in Section 309. Therefore, in rejecting Wyoming's regional haze SIP and adopting a FIP, EPA is required to show that the Wyoming SIP will not achieve “necessary reasonable progress” towards the visibility goal, and EPA's FIP will. EPA has failed to provide any support for such a position.

Other comments suggest that by meeting the requirements under 40 CFR 51.309, states are making reasonable progress toward the national goal of achieving natural visibility conditions for the 16 Class I areas on the Colorado Plateau. Wyoming submitted its regional haze SIPs under section 51.309. Therefore, the requirements of section 51.308 only apply to the extent required by section 51.309.

Wyoming's regional haze SIP is consistent with WRAP's NOX information, and also emphasizes the relatively small impact nitrates that stationary sources have on visibility issues in Wyoming. Wyoming correctly determined, consistent with the WRAP reports and other data, that controlling NOX emissions from stationary sources like PacifiCorp's units would yield very little visibility improvement in Wyoming. Section 51.309 understandably is intended to focus on SO2 due to the greater visibility impact from SO2. In a separate action, EPA acknowledged that Wyoming has complied with the section 51.309's SO2 requirements and made great progress towards improving and protecting visibility as a result.

As a result of the lesser emphasis in section 51.309 on NOX emissions, 40 CFR 51.309(d)(4)(vii) requires a regional haze SIP to “contain any necessary long term strategies and BART requirements for stationary source . . . NOX emissions.” Section 51.308, by contrast, does not contain a similar “necessary” threshold for BART. If a BART requirement is not “necessary” for a section 51.309 state, such as Wyoming, to make “reasonable progress,” then it is not required as part of the regional haze SIP. In other words, section 51.309 allows a state even more discretion because of this “necessary” requirement than would otherwise be allowed under section 51.308. Wyoming has authority to adopt those regional haze SIP provisions that it believes provide for “reasonable progress,” even when those plan provisions do not align directly with BART as that may be determined under Section 51.308.

40 CFR 51.309(d)(4)(vii) provides that “[a]ny such BART provisions may be submitted pursuant to either 51.308(e)(1) or 51.308(e)(2).” By using the permissive term “may,” EPA makes clear that such a submission, under either subsection, is voluntary and not mandatory for section 51.309 states. For this reason, Wyoming, as a WRAP state, was never required to comply with 40 CFR 51.308(e)(1)(ii)(A) and is only required to include whatever BART NOX determinations are “necessary,” as determined by the State. If Wyoming's section 51.309's SO2 controls already provide the adequate level of visibility improvement and protection, then, by definition, little or no BART NOX controls would be “necessary.” EPA has failed to show how any “necessary” NOX controls were excluded from the Wyoming regional haze SIP; therefore it should approve Wyoming's regional haze SIP.

Response: We disagree with these comments. As explained in our proposed rulemaking for section 51.309(d)(4)(viii), we explained that the provision “is intended to clarify that if EPA determines that the SO2 emission reductions milestones and backstop trading program submitted in the section 51.309 SIP makes greater reasonable progress than BART for SO2, this will not constitute a determination that BART for PM or NOX is satisfied for any sources which would otherwise be subject to BART for those pollutants” (emphasis added). 70 FR 44169 (Aug. 1, 2005). EPA does not interpret this rule to mean that there are different BART requirements for section 308 and 309 Start Printed Page 5099regional haze SIPs. EPA's rulemaking made no finding that BART determinations conducted for a state submitting a SIP under section 51.309 should be conducted any differently than a state submitting a FIP under only section 308. The use of the word “necessary” in section 51.309(d)(4)(viii) was to explain that some states may have BART NOX emission limitations, while others may not. As already explained elsewhere in proposal and our response to other comments, Wyoming did not conduct a proper evaluation of the five statutory factors, as required by 40 CFR 51.308(e)(1)(ii)(A) and section 169A(g) of the CAA.

EPA also disagrees with commenter's assertion that a BART submission is discretionary. 40 CFR 51.309(d)(4)(viii) is clear in that the implementation plan “must” contain BART requirements. The proposed rulemaking explained that the provision that provides that “[a]ny such BART provisions may be submitted pursuant to either Section 51.308(e)(1) or 51.308(e)(2),” was included to “allow States the flexibility to address these BART provisions either on a source-by-source basis under Section 51.308(e)(1), or through an alternative strategy under Section 51.308(e)(2).” 70 FR 44169 (Aug. 1, 2005).

Moreover, EPA's proposal made clear that “[i]n limited circumstances, it may be possible for a State to demonstrate that an alternative program which controls only emissions from SO2 could achieve greater visibility improvement than application of source-specific BART controls on emissions of SO2, NOX and/or PM. We nevertheless believe that such a showing will be quite difficult to make in most geographic areas, given that controls on SO2 emissions alone in most cases will result in increased formation of ammonium nitrate particles.” 70 FR 44169 (Aug. 1, 2005). Wyoming's RH SIP does not include a demonstration that the backstop SO2 trading program under Section 51.309 achieves greater visibility improvement than application of source-specific PM BART controls. Therefore, Wyoming's Section 51.309 SIP does not provide the adequate level of visibility improvement to meet the BART requirements.

With respect to the relationship of BART and requirements for reasonable progress under 40 CFR 51.308, EPA interprets the reasonable progress requirements to apply to BART sources. As explained in our guidance, due to the similarity of the BART and reasonable progress factors, states may reasonably rely on their BART determinations to show reasonable progress for those sources for the first planning period. However, BART is an independent requirement of the statute and the RHR. We have disapproved certain BART determinations by Wyoming not due to a failure to make reasonable progress, but due to a failure to consider the BART factors appropriately.

10. Legal Analysis

Comment: We received comments that the proposed rule is costly and that preliminary calculations by the State of Wyoming showed that the BART and long-term strategies under the proposed rule will cost over $170 million on an annualized basis; with total capital cost will be over $1 billion, and annual operating costs of nearly $600 million. Commenters went on to say that since the rulemaking action will exceed $100 million dollars in annual costs it should be reviewed according to the standards established in Executive Orders 12866 and 13563. * * *

Another commenter notes that EPA has also failed to conduct any analysis of the impacts under the Unfunded Mandates Reform Act (UMRA). In addition to the capital costs of nearly $750,000,000 for Laramie River Station alone, the annual operating costs of an SCR system at Laramie River Station are over $ 15,000,000. The commenter asserts that this amount is nearly double that projected by EPAs expert Andover of just under ($5,000,000), using generalized information. These annual operating costs, on top of the capital costs, for the three units at Laramie River Station alone, are significant, and when coupled with the impacts for the remaining five PacifiCorp units, far exceed the thresholds of the UMRA.

The UMRA is designed to ensure that Congress and federal agencies analyze the impact of proposed statutes and regulations on local governments and other entities before taking action. Where the estimates indicate at least a $50 million per fiscal year direct cost of all intergovernmental mandates, or a $100 million per fiscal year direct cost of private sector mandates, an analysis is required to evaluate the impact on local governments and private entities, and if necessary, the mandate must be funded. Western Minnesota, Missouri River Energy Services, and the governmental entities they serve—together with the others served by the remaining co-owners of Laramie River Station—will feel an annual impact in excess of $50 million per year should the EPA's FIP become final. Failure of EPA to conduct any analysis of the impact of imposing an unfunded mandate on the small governmental entities served by Laramie River Station, and the other BART units in Wyoming shows a blatant disregard for the regulatory process and protections that are to be accorded such significant rulemakings.

Earlier comments argued that the UMRA has been applied to EPA actions where the costs to regulated entities in numerous states have been aggregated. Based upon this precedent, PacifiCorp believes that EPA should aggregate all regional haze compliance costs across Wyoming, Utah, Colorado and Arizona for PacifiCorp, which would easily exceed the $100 million threshold. At a minimum, EPA should aggregate costs for EPA's FIPs in Wyoming and Arizona, which would also exceed the $100 million threshold.

Additional earlier comments suggested that the UMRA requires federal agencies to identify unfunded federal mandates. For rules that contain federal mandates, such as EPA's regional haze FIP action requiring expensive pollution controls, title II of UMRA requires the agencies to prepare written statements, or “regulatory impact statements,” (RIS) containing specific descriptions and estimates, including a qualitative and quantitative assessment of the anticipated costs and benefits of the mandate. This requirement is triggered by any rule that “may result in the expenditure by state, local, and tribal governments, in the aggregate, or by the private sector, of $100,000,000 or more (adjusted annually for inflation) in any 1 year. . .” 2 U.S.C. 1532(a).

When a RIS is required, the agency is also required to “identify and consider a reasonable number of regulatory alternatives and from those alternatives select the least costly, most cost effective, or least burdensome alternative that achieves the objectives of the rule” or explain why that alternative was not selected. 2 USCA Section 1535.

Here, EPA has failed to comply with the UMRA, arguing that the regional haze FIP “does not contain a federal mandate that may result in expenditures that exceed the inflation adjusted UMRA threshold of $100 million.” EPA is wrong. PacifiCorp currently anticipates spending more than $100 million dollars in at least 2014 ($104 million), 2015 ($175 million), and 2016 ($154 million) to comply with EPA's regional haze FIP for Wyoming (based on alternative “one” for the Jim Bridger plant). If the regional haze compliance costs imposed by EPA's proposed FIP in Arizona and EPA's approval of the Colorado regional haze SIP are factored in, the costs to PacifiCorp in a given Start Printed Page 5100year would be significantly higher. Also, when the BART NOX and PM determinations are approved by EPA for Utah, these costs to PacifiCorp in a given year could be much, much higher.

Response: We disagree with this comment. Under section 202 of the UMRA, before promulgating any final rule for which a general notice of proposed rulemaking was published, EPA must prepare a written statement, including a cost-benefit analysis, if that rule includes any “Federal mandates” that may result in expenditures to State, local, and Tribal governments, in the aggregate, or to the private sector, of $100 million or more (adjusted for inflation) in any one year. EPA has determined that this rule does not contain a Federal mandate that may result in expenditures that exceed the inflation-adjusted UMRA threshold of $100 million (in 1996 dollars) by State, local, or Tribal governments or the private sector in any one year. We estimate that the total annual costs in the aggregate are approximately $93 million (see Table 1).

Comment: EPA's regional haze FIP states that EPA's proposed action is not subject to Executive Order 13211, “Actions Concerning Regulations that Significantly Affect Energy Supply, Distribution, or Use” (66 FR 28,355 (May 22, 2001)), because the proposed action “is not a significant regulatory action under Executive Order 12866.” 78 FR 34790. EPA further claims the proposed regional haze FIP is not a “significant regulatory action” under Executive Order 12866 because the “proposed FIP applies to only five facilities” and is “therefore not a rule of general applicability.” EPA is incorrect, and should withdraw its regional haze FIP.

Executive Order 13211 provides that agencies shall submit a statement of energy effects for matters “identified as significant energy actions.” A “significant energy action” is defined as “any action by an agency . . . that promulgates or is expected to lead to the promulgation of a final rule or regulation . . . that is a significant regulatory action under Executive Order 12866 or any successor order” and “likely to have a significant adverse effect on the supply, distribution, or use of energy”; or is “designated by the Administrator of the Office of Information and regulatory Affairs as a significant energy action.” Id. Section 4(b) (emphasis added). Executive Order 12866, in turn, which concerns Regulatory Planning and Review, defines a “significant regulatory action” as any regulatory action that is likely to result in a rule that may: (1) Have an annual effect on the economy of $100 million or more or adversely affect in a material way the economy, a sector of the economy, productivity, competition, jobs, the environment, public health or safety, or State, local, or tribal governments or communities. 58 FR 51735, 51738 (Oct. 4, 1993)

According to PacifiCorp's current estimates (excluding allowance for funds used during construction (AFUDC)), it will spend more than $100 million dollars in capital costs alone in 2014 ($225 million), 2015 ($139 million), 2017 ($146 million) and 2018 ($118 million) to comply with EPA's regional haze FIP for Wyoming (based on alternative “one” for the Jim Bridger plant). If regional haze compliance costs currently imposed or approved by EPA on PacifiCorp's BART Units in Arizona and Colorado are factored in, the total capital cost impacts to PacifiCorp in any given year would be significantly higher; increasing to approximately $246 million in 2014, $190 million in 2015, $168 million in 2016, $181 million in 2017, and $118 million in 2018. Also, because the BART NOX and PM determinations have not yet been approved by EPA for PacifiCorp's BART Units in Utah, EPA's ultimate BART requirements in Utah likely will add even more costs in overlapping installation and compliance years, with total project costs for SCR installations on PacifiCorp's Utah units currently estimated to cost in excess of $150 million per unit to install (again, excluding AFUDC). Based upon these basic costs alone, there is no doubt that EPA's FIP meets the definition of a “significant regulatory action.” Other large costs, including those related to EPA's BART determinations for Basin Electric, also should be factored into this analysis together with PacifiCorp's costs because they are part of the same “sector of the economy.” Also, as demonstrated by PacifiCorp's July 12, 2012, submittal in this docket, EPA's regional haze FIP will have an adverse effect on the supply and distribution of electricity within PacifiCorp's system. Therefore, EPA's determination that Executive Order 13211 did not apply is incorrect, and arbitrary and capricious.

Moreover, EPA has admitted in the proposed rule that system-wide “affordability” costs should be part of the BART analysis. 78 FR 34756. Because EPA's FIP is a “significant regulatory action,” EPA must prepare a “Statement of Energy Effects” for the Administrator of the Office of Information and Regulatory Affairs, Office of Management and Budget. (See Executive Order 13211, Section 2.) Because EPA did not do so, the regional haze FIP is improper.

Executive Order 12866, in turn, which concerns Regulatory Planning and Review, defines a “significant regulatory action” as any regulatory action that is likely to result in a rule that may: (1) Have an annual effect on the economy of $100 million or more or adversely affect in a material way the economy, a sector of the economy, productivity, competition, jobs, the environment, public health or safety, or State, local, or tribal governments or communities.”

Also, as demonstrated by PacifiCorp's July 12, 2012 submittal in this docket, EPA's regional haze FIP action will have an adverse effect on the supply and distribution of electricity within PacifiCorp's system. Therefore, EPA's determination that Executive Order 13211 did not apply is incorrect, and arbitrary and capricious.

Moreover, because EPA's regional haze FIP action is a “significant regulatory action,” before imposing its regional haze FIP EPA must first prepare a “Statement of Energy Effects” for the Administrator of the Office of Information and Regulatory Affairs, Office of Management and Budget. Such a statement must include a “detailed statement” by the agency concerning “any adverse effects on energy supply, distribution, or use (including a shortfall in supply, price increases . . .) should the proposal be implemented,” and “reasonable alternatives to the action with adverse energy effects and the expected effects of such alternatives on energy supply, distribution, and use.” Accordingly, based on an analysis of the relevant factors, EPA's regional haze FIP is improper because EPA failed to conduct the required regulatory analysis and failed to prepare the required documentation.

Executive Order 12866, in turn, which concerns Regulatory Planning and Review, defines a “significant regulatory action” as any regulatory action that is likely to result in a rule that may: (1) Have an annual effect on the economy of $100 million or more or adversely affect in a material way the economy, a sector of the economy, productivity, competition, jobs, the environment, public health or safety, or State, local, or tribal governments or communities.”

Response: EPA disagrees that Executive Order 13211 applies to this action. Order 13211 is only applicable to an agency regulation that is a “significant regulatory action” under Executive Order 12866. Executive Order 13211(4)(b). Order 13211 also explicitly adopts the definitions of “regulation” and “rule” as explained in Executive Order 12866. Executive Order Start Printed Page 510113211(4)(a). However, this action does not fit within the definition of Executive Order 12866, which defines a “regulation” or “rule” as an “agency statement of general applicability.” Executive Order 12866(3)(d). Here, EPA's action was individually tailored for a limited number of BART eligible sources in Wyoming and is not generally applicable. Thus this action is not governed by Order 12866 and, therefore, is also not governed by Executive Order 13211. As discussed in more detail in Statutory and Executive Orders Review section of this document, the costs for this action are about $93 million annually.

Moreover, as explained in more detail elsewhere, EPA took the cost of compliance into consideration when making BART determinations to ensure this rule's requirements are beneficial and not unduly burdensome. The commenter is correct that EPA may, in its discretion, consider system affordability costs beyond the direct compliance costs on an individual facility in extraordinary circumstances. As explained in the Basis for Final Action section and elsewhere in the proposed and final actions, we proposed to approve the State's BART and reasonable progress determinations for Jim Bridger Units 1 and 2, but on a different basis.[69] In response to our proposal, we received both supportive and adverse comments regarding whether the affordability provisions of the BART Guidelines should apply to Units 1 and 2. As explained in more detail in our responses to these comments, we agree that PacifiCorp did not make a sufficient showing that it could not afford to install LNB/SOFA + SCR on Units 1 and 2 within the five-year compliance period. Nevertheless, we also received new information regarding the costs of compliance and visibility benefits associated with Jim Bridger and have revised our cost estimates and visibility modeling for all four units accordingly. Using this revised information, we re-evaluated the five BART factors. Ultimately however, while we believe that these costs and visibility improvements could potentially justify LNB/SOFA + SCR as BART, because this is a close call and because the State has chosen to require SCR as a reasonable progress control, we believe deference to the State is appropriate in this instance. We are therefore finalizing our approval of the State's determination to require SCR at Jim Bridger Units 1-4, with an emission limit of 0.07 lb/MMBtu (30-day rolling average), as part of its LTS. We also note that, neither the CAA nor the regional haze regulations require EPA to consider costs beyond an individual facility's direct compliance costs. 42 U.S.C.A. Section 7491(g)(1), (2); 40 CFR 51.301. Further, nothing in the Order is to be construed to impair or otherwise affect the authority granted by law to EPA, nor does it create any right or benefit enforceable at law.[70]

Comment: The EPA's FIP fails to account for the significant economic impacts on small entities as required by the Regulatory Flexibility Act. The estimated capital cost alone to install SCRs at Laramie River Station only is nearly $750,000,000. For Western Minnesota and its members the total impact would result in an increase in wholesale electric rates of twelve percent, which includes a ten percent increase due to the capital costs for installation of the SCRs and an additional increase of two percent annually for operating expenses. The members of Western Minnesota and Missouri River Energy Services are small governmental units, which must be given consideration under the Regulatory Flexibility Act.

Under the Regulatory Flexibility Act, EPA is required to analyze the economic impact of proposed regulations when there is likely to be a significant economic impact on a substantial number of small entities, and to consider regulatory alternatives that will achieve the agency's goal while minimizing the burden on small entities. The certification that EPA has provided with this proposed rule is perfunctory at best, and does not seek to analyze the actual elements required by the Regulatory Flexibility Act.

The EPA has wholly failed to conduct any regulatory flexibility analysis pursuant to the Regulatory Flexibility Act, which further demonstrates the arbitrariness of this proposed FIP. If it had, it would acknowledge that the Wyoming SIP for NOX provides a reasonable alternative that has a far less significant economic impact on small entities while providing virtually the same improvement in visibility.

Response: EPA disagrees with this comment. Courts have interpreted the Regulator Flexibility Act to require a regulatory flexibility analysis only when a substantial number of small entities will be subject to the requirements of the agency's action. See, e.g., Mid-Tex Elec. Co-op, Inc. v. FERC, 773 F.2d 327, 342 (D.C. Cir. 1985). The economic analysis described in the Regulatory Flexibility Act is not required; however, if the head of an agency certifies that the rule will not have a significant economic impact on a significant number of small entities. 5 U.S.C. 605(b). As the commenter noted, such certification was made by EPA and published in the Federal Register as required by the act. No other action is required by EPA because the agency is not imposing any requirements on small entities. Here, only a limited number of entities have incurred compliance obligations under this action, and none of those entities are “small entities.”

EPA still seeks to minimize the impact of its actions have on small entities. EPA sought comments regarding the economic impact from all entities affected by this action and carefully considered all relevant information. As described elsewhere, EPA believes that this action is necessary to achieve the objectives of the CAA and that the visibility improvements justify the costs of this rule, as established in the Act and implementing regulations.

Comment: The underlying purpose of Executive Order 12866 (Order) is to foster a regulatory regime that respects the role of local government, recognizes that the private sector is the “best engine for economic growth,” and appreciates the need to develop regulations that do not impose “unacceptable or unreasonable costs on society.” Exec. Order No. 12,866, 58 FR 51735 (Oct. 4, 1993). The Order requires agencies that propose a significant regulatory action to consider a multitude of quantitative and qualitative factors during the rulemaking process. Id.

A “significant regulatory action” is one in which the resulting rule is likely to “[h]ave an annual effect on the economy of $100 million or more or Start Printed Page 5102adversely affect in a material way the economy, a sector of the economy, productivity, competition, jobs . . . or State, local, or tribal governments or communities.” 58 FR 51735.

Allow us, by way of example, the opportunity to outline the significance that the proposed rule will have on one of Wyoming's counties. The Laramie River Station (Station), one of the five targeted EGUs under the proposed FIP, is the largest employer in Platte County. These are good paying jobs that include health and retirement benefits for the Station's employees. In a rural county with 8,800 people, where the average annual household income is $46,916, there is concern that the use of a control regime as proposed in the FIP may make operation of the Station substantially cost-prohibitive and therefore jeopardize some of the best jobs in Platte County. Moreover, the retrofitting outlined in the proposed rule will likely increase the electric rates of some of Wyoming's most vulnerable citizens.

The Station also provides a significant source of revenue for Platte County. In the 2012 Fiscal Year, the Station provided over $3.7 million in state assessed taxes. This is a significant source of revenue for Platte County, revenue that is needed to sustain essential government functions, such as operation of the county jail, maintaining county roads and bridges, and county health services. Platte County is but one example. Each of the other potentially affected counties (Converse, Lincoln, and Campbell) share those three principal concerns: (1) The targeted EGUs provide a significant source of employment to county residents, (2) volatility of electric rates on some of Wyoming's most vulnerable citizens, and (3) the EGUs provide a significant source of revenue in order to sustain essential county services.

For these reasons, Wyoming's County Commissioners cannot accept EPA's conclusion that the proposed rule is not a significant regulatory action. Accordingly, because the proposed rule is a significant regulatory action, it should be subject to review in accordance with Executive Order 12866 and, by extension, Executive Order 13563.

Response: Executive Order 13563 is supplemental to and reaffirms the principles, structures, and definitions governing contemporary regulatory review that were established in Executive Order 12866 of September 30, 1993. In general, the Order seeks to ensure the regulatory process is based on the best available science; allows for public participation and an open exchange of ideas; promotes predictability and reduces uncertainty; identifies and uses the best, most innovative, and least burdensome tools for achieving regulatory ends; and takes into account benefits and costs, both quantitative and qualitative. However, nothing in the Order shall be construed to impair or otherwise affect the authority granted by law to the Agency. In our review process the cost of compliance was one of the elements addressed to ensure that the requirements to achieve the goals stated in the CAA were beneficial and not burdensome to the regulated entity. Please refer elsewhere in our response to comments (e.g., Introduction and BART sections) for a detailed analysis of the elements required by the CAA and RHR for BART determinations.

Comment: EPA, through this proposed rule, fails to recognize, or even to make an effort to understand, the burden imposed upon Wyoming and its local governmental entities. If it had done so, it would have acknowledged the fundamental value and attainable progress derived from Wyoming's regional haze SIP. Instead, what EPA proposes is a disingenuous and poorly crafted rule that ignores fundamental realities existing in the counties of Wyoming, that they are rural, traditionally low-income, and the economic drivers are typically limited to tourism, agriculture, or industry. As such, the proposed rule would create a disproportionate impact on those communities playing host to industry.

“Not in my backyard” is simply not an option for many rural communities. It therefore poses a fundamental question of equity, a concern reiterated in Executive Order 13563. Executive Order No. 13,563, 76 FR 3821 (Jan. 18, 2011). Concerns regarding equity require the EPA to consider who is bearing the cost of the proposed rule. The bottom line is that increased energy costs that will result from the proposed FIP will disproportionately hurt our local economies.

Additional comments argue that EPA is required to seek views of appropriate local officials' before imposing regulatory requirements that might significantly or uniquely affect a particular governmental entity. EPA must then seek to minimize any burdens that significantly or uniquely affect the local governmental entity in a manner that is consistent with achieving the underlying regulatory objective.

Response: EPA disagrees with this comment. Executive Order 13563 is supplemental to and reaffirms the principles, structures, and definitions governing contemporary regulatory review that were established in Executive Order 12866. Executive Order 13563 Section 1(b). In general, the Order seeks to ensure the regulatory process is based on the best available science; allows for public participation and an open exchange of ideas; promotes predictability and reduces uncertainty; identifies and uses the best, most innovative, and least burdensome tools for achieving regulatory ends; and takes into account benefits and costs, both quantitative and qualitative. However, nothing in the Order shall be construed to impair or otherwise affect the authority granted by law to EPA, nor does it create any right or benefit enforceable at law. Executive Order 13563 Section 7(b), (d). Each BART source was examined and the cost of compliance was one of the factors addressed to ensure the rule's requirements are beneficial and not unduly burdensome to the regulated entities. We also note the following: (1) There will likely be beneficial effects on tourism due to improved visibility at the Class I areas; [71] (2) emission controls can be installed over a period of time; and (3) in this final action we are mostly approving the provisions of the State's SIP. Moreover, as explained above, courts have interpreted the Regulatory Flexibility Act to require a regulatory flexibility analysis only when a substantial number of small entities will be subject to the requirements of the agency's action. While EPA has not made a determination that a substantial number of small entities will be subject to the requirements of this final action, we nevertheless seek to minimize the impact our actions have on small entities. EPA sought comments regarding the economic impact from all Start Printed Page 5103entities affected by this action and carefully considered all relevant information. As described elsewhere, EPA believes that this action is necessary to achieve the objectives of the CAA and that the visibility improvements justify the costs of this rule, as established in the Act and implementing regulations. Please refer elsewhere for a detailed analysis of the elements required by our regulations for BART determinations.

Comment: In imposing these additional costs the proposed action will unnecessarily impact power generation in Wyoming and lead ultimately to increased utility costs for Wyoming residents. Along these lines, we believe the proposed action fails to comply with Executive Order 13132. The notice of proposed action states: “Under Executive Order 13132, EPA may not issue a regulation that has federalism implications, that imposes substantial direct compliance costs, and that is not required by statute, unless the federal government provides the funds necessary to pay the direct compliance costs incurred by state and local governments, or EPA consults with state and local officials early in the process of developing the proposed regulation.” The notice switches that standard by concluding: “This rule will not have substantial direct effects on the states, on the relationship between the national government and the states, or on the distribution of power and responsibilities among the various levels of government, as specified in Executive Order 13132, because it merely addresses the State not fully meeting its obligation to prohibit emissions from interfering with other states measures to protect visibility established in the CAA. Thus, Executive Order 13132 does not apply to this action.” (78 FR 34790).

The conclusion that Executive Order 13132 does not apply to this action appears inconsistent with the standard of the Order. The regulation will impose substantial direct compliance cost on local governments and there is no provision for funding those costs by the federal government. The Wyoming Municipal Power Agency (WMPA) is a joint powers board created by eight Wyoming municipalities to generate and transmit electricity for the residents of those municipalities. Through a partnership, WMPA and thus each of those municipalities, own a substantial interest in the Laramie River Station. WMPA estimates that the EPA's proposal would cost an estimated $600 million for the Laramie River Station. When costs are imposed upon a facility, owners of that facility must initially bear those costs. Whether the generating facility is owned 100% by a single municipality, or shared through a partnership or cooperative, as a result of that ownership interest a substantial direct compliance cost is imposed on the municipal owners.

Response: EPA disagrees that Executive Order 13132 applies to this action. The Order only applies to agency actions that fit within the Order's definition of “policies that have federalism implications.” The Order defines such actions as rules “that have substantial direct effects on states, on the relationship between the national government and the States, or on the distribution of power and responsibilities among the various levels of government.” Executive Order 13132 Section (1)(a). In contrast, this action merely addresses Wyoming's existing obligations under the CAA and thus does not impose any additional burdens beyond that which the law already requires. Because this rule does not fit within the definition of “policies that have federalism implications,” the Order does not apply to this action.

Moreover, the additional elements of the standard urged by the commenter do not apply to this action. EPA must consult a state or provide funding only if a regulation that has federalism implications (1) imposes substantial direct compliance costs on state and local governments, and (2) is not required by statute. Executive Order 13132 Section (6)(b). First, as the commenter noted, the regulation imposes compliance costs on Laramie River Station, not directly on state or local governments. Municipalities may possibly be indirectly impacted because of their membership in WMPA, which in turn retains a 1.37% ownership interest in Laramie River Station through a partnership with the Missouri Basin Power Project. However, this action does not impose any direct compliance costs on local governments. Second, the CAA requires that states promulgate adequate SIPs to achieve the CAA's visibility goals, and further requires EPA to promulgate FIPs if a state fails to meet its obligations. 42 U.S.C. 7492(3)(2), Section 7410(c)(1); see also WildEarth Guardians v. Jackson, No. 11-CV-00001-CMA-MEH, 2011 WL 4485964 at *6 (D. Colo. Separt 27, 2011) (finding EPA's duty to either approve a state regional haze SIP or promulgate a FIP is nondiscretionary). Because this action does not have federal implications, does not impose direct compliance costs on local governments, and is required by statute, Executive Order 13132 does not apply to this action.

Comment: EPA did not properly vet its proposed FIP against a number of requirements detailed throughout Presidential Executive Orders and within the CAA. Though Presidential Executive Orders are not binding by law, they foster an open, transparent rule-making process. For example, Executive Order 12866 states, “The American people deserve a regulatory system that works for them, not against them: a regulatory system that . . . improves the performance of the economy without imposing unacceptable or unreasonable costs on society.” Accordingly, the Order requires federal agencies, including EPA, to develop regulations “in the most cost effective manner” and to “adopt a regulation only upon a reasoned determination that the benefits of the intended regulation justify its costs.” An additional $1.2 billion in costs under EPA's proposal in exchange for no perceptible change in visibility does not qualify as “a reasoned determination that the benefits of the intended regulation justify its costs.”

EPA's proposed action will result in over $170 million annual adverse economic impact, well over the $100 million annual threshold identified in Executive Order 12866. EPA failed to properly consider material effects its proposed action will have on the economy, productivity, competition, and jobs. By forcing unnecessarily expensive technologies, electricity rates will rise even further, putting additional strain on businesses and millions of customers that receive electricity from the generating stations in Wyoming.

Response: EPA disagrees with this comment. As explained elsewhere, this action does not fit within the definition of Executive Order 12866. The Order defines a “regulation” or “rule” as an “agency statement of general applicability.” Executive Order 12866(3)(d). Here, EPA's action was individually tailored for a limited number of BART eligible sources in Wyoming, and thus is not generally applicable and not governed by Order 12866. Moreover, as explained in more detail in the BART section, EPA took the cost of compliance into consideration when making its BART determinations to ensure the rule's requirements are beneficial and not unduly burdensome.

11. Consideration of Existing Controls

Comment: Several commenters asserted that EPA did not properly take into account the existing pollution control technology in use at the BART-eligible EGUs, as required by CAA section 169A(g)(2) and the BART Start Printed Page 5104Guidelines. These commenters alleged that EPA was required to consider updated combustion controls, which were installed to comply with Wyoming's regional haze SIP, by adjusting the baseline emissions rate for each facility to account for any emissions reductions that have been achieved since the 2001-2003 baseline period. The commenters suggested that had EPA relied on available 2011-2012 emissions data, which reflect the NOX reductions achieved by some of these newly installed controls, the cost-effectiveness values for SNCR and SCR would have been higher, while the visibility improvement associated with SNCR and SCR would have been lower.

For example, one commenter stated that the baseline emission rate for NOX in 2001-2003 was 0.27 lbs/MMBtu at Laramie River, but that the emission rate had dropped to 0.19 lb/MMBtu after the installation of over-fire air and low NOX burners. This commenter asserted that, had EPA adjusted the baseline to the latter emission rate, the average cost-effectiveness for SNCR would be between $6,967/ton and $7,014/ton, while the average cost-effectiveness for SCR would be between $8,531/ton and $9,048/ton. Based on these values, the commenter argued that neither SNCR nor SCR is cost-effective and therefore both technologies should be eliminated as NOX BART for Laramie River.

Another commenter pointed to other EPA regional haze actions where EPA adjusted baseline emissions to account for recently installed controls, such as EPA's final actions on the Arizona regional haze SIP, 77 FR 72512, and Montana regional haze FIP, 77 FR 57864. This commenter argued that because EPA had adjusted baseline emissions for some Arizona and Montana EGUs to account for controls recently installed to satisfy consent decrees obligations or CAA requirements unrelated to regional haze, EPA was required to do so for Wyoming's EGUs as well.

One commenter submitted additional comments, after the close of the public comment period, in response to the decision of the U.S. Court of Appeals for the Eighth Circuit in North Dakota v. EPA, 730 F.3d 750 (8th Cir. 2013). This commenter again asserted that EPA had failed to consider the low NOX burners and over-fire air the commenter had installed at its facilities to comply with the Wyoming regional haze SIP. The commenter argued that EPA's alleged failure to consider these controls violates the holding in North Dakota, that “any existing pollution control technology” includes all existing controls, including those that are voluntarily installed by the source. Moreover, the commenter explained, the low NOX burners and over-fire air at its facilities were not voluntary controls because they were installed to meet CAA requirements, were federally enforceable, and were incorporated into the Wyoming regional haze SIP.

A final commenter also submitted late comments in response to the decision in North Dakota and the previous commenter's letter. This commenter argued that the North Dakota decision does not require EPA to account for existing controls by factoring their associated emissions reductions into baseline emissions. The commenter explained that using a consistent baseline prevents certain sources from claiming credit for minor emission reduction measures taken in the midst of the ongoing regional haze planning process, thereby making more stringent controls appear less beneficial. The commenter also explained that the fixed baseline period of 2001-2003 allows EPA to make “apples-to-apples” comparisons of the cost-effectiveness and visibility benefits of evaluated technologies across all BART sources. The commenter pointed out that EPA's method of evaluating combustion controls as a BART option, not as part of the emission baseline, was identical to the approach that Wyoming and the sources themselves had taken in their own BART analyses. The commenter argued that this approach is correct because it ensures that the emissions reductions associated with existing controls installed after the baseline period are evaluated in the BART analysis. It also factors the cost of such controls into the source's compliance costs, rather than assuming that such costs are zero simply because they already have been incurred. Finally, the commenter asserted that even if EPA were to adjust the baseline emissions for the EGUs in question, SCR would still be BART for all of the EGUs.

Response: One of the statutory factors EPA is to consider is “any existing pollution control technology in use at the source.” 42 U.S.C. 7491(g)(2). The CAA does not specify how states or EPA must “take into consideration” this factor. The BART Guidelines provide little additional guidance, stating only that “[f]or emission units subject to a BART review, there will often be control measures or devices already in place. For such emission units, it is important to include control options that involve improvements to existing controls and not to limit the control options only to those measures that involve a complete replacement of control devices.” 40 CFR part 51, app. Y, at IV.D.1.6. Consequently, we believe that states and EPA have considerable discretion in how they consider existing controls in use at a source, so long as that consideration is explained and reasonable. Ultimately, states or EPA should consider the totality of the circumstances (e.g., the purpose of any existing controls, when and why they were installed, compatibility with other control options, enforceability, and other pertinent factors) in determining how they will evaluate existing controls in a BART analysis.

For example, one way in which a state or EPA can consider existing controls, as contemplated by the BART Guidelines, is by evaluating whether additional control options are technologically compatible with a source's existing controls, or whether the presence of the existing controls would render the installation of some additional control options infeasible. In the case of NOX, the presence of existing combustion controls at a source, such as low NOX burners or over-fire air, does not impact the feasibility of installing post-combustion controls, such as SNCR or SCR. Consequently, EPA reasonably determined in this instance that the presence of existing combustion controls at several of the BART-eligible EGUs would not preclude the installation of either SNCR or SCR.

Pointing to our regional haze actions in Arizona and Montana, several of the commenters asserted that EPA was required to consider existing controls by adjusting the baseline emissions of several sources to account for reductions achieved after the baseline period. We disagree. The BART Guidelines recommend that baseline emissions should be “based upon actual emissions from a baseline period.” 40 CFR part 51, app. Y, at IV.D.4.d.1. While the BART Guidelines allow states or EPA to adjust baseline emissions to take into account projections of “future operating parameters” by making such assumptions into enforceable limits, id. at IV.D.4.d.2, the BART Guidelines are silent as to how reductions resulting from the post-baseline installation of controls should be treated. One way to take account of such reductions is to update the baseline, as we did in our regional haze actions for Arizona and Montana. In those rulemakings, we determined that updating the baseline was appropriate because several sources had recently installed combustion controls to comply with consent-decree obligations and acid-rain requirements, while another source had changed its coal supply. The fact that these controls were installed to comply with other CAA requirements heavily informed Start Printed Page 5105EPA's decision to update the sources' baseline emissions. As we explained in our Arizona action, however, “an `updated baseline' might not be appropriate in all instances. For instance, if it appeared that controls had been installed early in order to avoid a more stringent BART determination, it would presumably not be appropriate to use a baseline representing these new controls.” 72 FR 72526.

We believe that this is one such instance. First, unlike in Arizona and Montana, the sources did not install the combustion controls in question to comply with other CAA requirements. Rather, as stated above in the comment, the sources installed the controls to comply with Wyoming's selection of BART in its regional haze SIP. This distinction is important because, by their very nature, baseline emissions should be “a realistic depiction of anticipated annual emissions” before the installation of BART. 40 CFR part 51, app. Y, at IV.D.4.d. Thus, while baseline emissions can take into account newly installed controls and in some cases future operating parameters, baseline emissions still must represent a pre-BART scenario so that the cost-effectiveness and visibility benefits of all potential BART control options can be evaluated from a consistent benchmark. Indeed, it would be passing strange for EPA to update the commenters' baseline emissions to incorporate emission reductions that they admittedly achieved to comply with BART. Doing so would bias EPA's analysis of additional control options by giving the commenters credit for emissions reductions attributable to BART, but treating the costs they incurred to achieve those reductions as if they had never occurred.

Second, we note that the Wyoming regional haze SIP did not require compliance with BART until five years after EPA's approval of the SIP. At the time the sources installed the combustion controls, EPA had not yet acted upon Wyoming's regional haze SIP, and the sources had no way of knowing whether EPA would ultimately approve or disapprove Wyoming's BART determinations. Thus, it appears that the sources' decision to install the combustion controls early may have been motivated by an intent to avoid the possibility of a more stringent BART determination by EPA under the theory now advanced in the comment. To be consistent with our statements in the Arizona regional haze action, we believe that it would have been inappropriate for EPA to have “taken into consideration” the newly installed combustion controls at the commenters' facilities by updating the baseline in this case.

Nonetheless, EPA recognizes that we must “take into consideration” all existing controls to comply with the CAA, and have therefore taken the sources' existing combustion controls into consideration in other ways. For example, in addition to considering whether the source's existing combustion controls were compatible with the installation of post-combustion controls, we also used the source's current NOX emission rates when we evaluated the size, design, and reagent/catalyst cost of SNCR and SCR. For example, in the case of Laramie River, we did not use the baseline emission rate of 0.27 lbs/MMBtu, but rather the current emission rate of 0.19 lb/MMBtu that appropriately reflects the installation of over-fire air and low NOX burners. Due to the lower NOX emission rate, the size of the SNCR and SCR systems and the amount of reagent/catalyst necessary to operate them are lower than if we had simply assumed the baseline emission rate.

Moreover, we do not believe that our action is inconsistent with the Eighth Circuit's decision in North Dakota. In our regional haze action for North Dakota, we refused to consider the DryFiningTM control technology in use at Coal Creek Station when we promulgated a FIP. We argued to the court that the CAA did not require states or EPA to consider controls that were voluntarily installed after the baseline period. The court rejected this position, holding that “EPA's refusal to consider the existing pollution control technology in use at the Coal Creek Station because it had been voluntarily installed was arbitrary and capricious.” North Dakota, 2013 U.S. App. LEXIS 19442, at*30. The court explained that “any existing pollution control technology” included even voluntarily installed controls. However, the court did not opine as to how existing controls must be considered. Here, EPA reasonably considered the existing controls at the BART-eligible sources in the several ways described above. North Dakota does not require us to “take into consideration” existing controls by adjusting baseline emissions, which would have been inappropriate in this instance.

Finally, while we acknowledge the supportive comments from the final commenter on this issue and agree with many of the points that were made, we decline to require SCR at all of the BART-eligible EGUs, for reasons explained elsewhere in this document.

12. Consent Decree

Comment: As it had on other SIPs, EPA neglected to act on Wyoming's SIP, and as a result exposed itself to liability for violating Section 110(k) of the CAA. See 42 U.S.C. 7410(k)(2), (3) (setting deadlines for EPA action on SIPs). Wyoming could have sued EPA for failing to take action on Wyoming's SIP, but in the spirit of cooperation, elected not to. Instead, special interest groups sued EPA for its failure to comply with the Act. See Compl., WildEarth Guardians v. Jackson, No. 1:11-cv-00001-CMA-MEH (D. Colo. Jan. 2, 2011). Wyoming did not participate in this litigation for two reasons: First, Wyoming was not aware of the litigation until EPA published the proposed consent decree, 76 FR 34983 (June 15, 2011); and, second, EPA has repeatedly opposed state attempts to participate in litigation that impacts the processing of SIPs, see, e.g., Def. Opp. to North Dakota's Motion to Intervene, WildEarth Guardians v. Jackson, No.4:09-CV-02453-CW (N.D. Cal. Oct. 20, 2011).

The special interest groups' litigation, in turn, has driven EPA's approach to Wyoming's SIP. The litigation has established arbitrary deadlines for EPA to act on Wyoming's SIP, which EPA and the special interest groups have repeatedly extended for their convenience. Not once has EPA consulted the State on these deadlines. More troubling, through settlement of that litigation, EPA has committed to particular courses of action on Wyoming's SIP. EPA has cut Wyoming out of the cooperative federalism Congress intended to guide the regional haze program.

This dubious approach to implementing the CAA harms states. The unprecedented influence the special interest groups have exerted over EPA's treatment of Wyoming's SIP, coupled with EPA's effort to conceal its communications with those groups, lead a reasonable observer to seriously question the objectivity of EPA's proposed action on Wyoming's SIP.

The commenter also asserted that the EPA let sue-and-settle tactics pervert what is typically a cooperative process. Nongovernmental groups should not be allowed to coerce an agency into setting policy as a result of litigation. Wyoming considers this an attack on states' rights, which does nothing to further the partnership between EPA and Wyoming, especially when Wyoming can't participate in those discussions.

Response: We disagree with commenter's assertions. The Act provides citizens with the right to sue EPA when EPA fails to meet a statutory Start Printed Page 5106deadline, 42 U.S.C. 7604(a)(2), and courts have the authority to establish a remedy that sets new deadlines and compels EPA to do what Congress required. Id. Section 7604(a). When EPA is confronted with such suits, it is reasonable and proper for EPA, working with the Department of Justice, to decide that it is in the public interest, and a more efficient use of executive and judicial branch resources, to settle such cases rather than litigate them. Congress recognized that EPA has authority to settle cases, and directed EPA to provide public notice and an opportunity to comment on proposed consent decrees before finalizing them. Id. Section 7413(g).

As explained in the Introduction section of this document, the consent decree of which the commenter complains was the result of a citizen suit that sought to compel EPA to approve SIPs or promulgate FIPs to meet long overdue regional haze requirements.[72] The State of Wyoming did not submit its regional haze implementation plan in a timely fashion as required by the CAA and EPA's regulations. Facing substantial legal risk, EPA reasonably negotiated a settlement resulting in a consent decree that set new deadlines for EPA to take actions required by the Act.[73] The Consent Decree was published in the Federal Register as is required under the CAA section 113(g) and provided 30 days for public comment. See 76 FR 34983 (June 6, 2011). For Wyoming, EPA's obligations to fully approve SIPs or promulgate a FIP were now due six years after the original 2007 deadline for the submission of regional haze SIPs. The consent decree was also subject to district court review before its entry. Neither the commenters nor any other party objected to the deadlines established for EPA's action on the Wyoming regional haze SIP. The United States District Court for Colorado found the terms of the consent decree reasonable.

The commenter's argument that EPA used these consent decree deadlines and “has committed to particular courses of action on Wyoming's SIP” with regard to the Final Rule is without merit. The consent decree did not limit or change EPA's substantive rulemaking authority or discretion in any way. Rather, the consent decree specifically permitted EPA to satisfy its obligations either by approving the States' regional haze SIPs or by promulgating a FIP. EPA also provided more than 70 days from the date of publication in the Federal Register for interested parties to submit comments on the proposed rule, longer than the 30-day public comment period required by the Act. 42 U.S.C. 7607(h). EPA obtained several extensions of the consent decree deadline for Wyoming to provide the Agency with adequate time to conduct the rulemaking.[74] For all these reasons, neither the consent decree nor the deadlines it imposed rendered EPA's Final Rule arbitrary, capricious, or contrary to law.

Finally, EPA did not rely on information that was not in the docket for this rule. Therefore, contrary to commenter's assertions, all information relied upon has been disclosed.

Comment: EPA quickly entered into a settlement agreement to resolve the special interest groups' litigation, rather than defend its actions and honor Wyoming's patience with EPA's inaction. In settling the litigation, EPA agreed to take final action on Wyoming's SIP by April 15, 2012. Consent Decree, WildEarth Guardians v. Jackson, No. 1: 11-cv-00001-CMA-MEH, at 4, '1]6 (D. Colo. Separt 27, 2011) (WildEarth Guardians). Recognizing that it still could not meet its statutory obligation to act on Wyoming's SIP, EPA persuaded the special interest groups to extend that deadline thirty days to May 15, 2012. Stip. to Extend Four Deadlines in Consent Decree at 3, ~ 6, WildEarth Guardians, (D. Colo. Jan. 10, 2012).

On June 2, 2012, eighteen months after Wyoming submitted its SIP, EPA proposed to partially approve and partially disapprove the SIP. 77 FR 33022. But, as a result of EPA's unlawfully delayed action, Wyoming's SIP became complete by operation of law. See 42 U.S.C. 7410(k)(l)(B). Accordingly, EPA cannot now propose to disapprove Wyoming's SIP on the grounds that it lacks information. To do otherwise is to render Section 110(k)(l)(B) meaningless.

Response: EPA disagrees with this comment. First, the commenter offers no grounds on which EPA could have defended the cited litigation, which involved mandatory statutory deadlines under the Act.

Second, EPA disagrees with the commenter's interpretation of CAA section 110(k)(l)(B). Under the CAA, EPA's SIP review is a two-step process. See CAA Section 110(k). First, within six months of a SIP submission, EPA must make a threshold “completeness determination” to determine whether the SIP contains certain “minimum criteria” designated by EPA as “the information necessary to . . . determine whether the plan submission complies with the provisions of [the CAA].” See id. Section 110(k)(1)(A), (B). These minimum criteria are listed in Appendix V to 40 CFR. Part 51 and include a relatively short list of eight “Administrative Materials” and nine “Technical Support” requirements, such as evidence that the state properly adopted the SIP and technical demonstrations that allow EPA to evaluate compliance with the substantive requirements of the CAA. See 40 CFR. part 51, App. V. If EPA fails to make the completeness determination within six months, the SIP is deemed complete by operation of law. See CAA Section 110(k)(1)(C).

Importantly, however, a determination of completeness, either by EPA or by operation of law, does not mean that the SIP has been approved as compliant with the substantive requirements of the CAA. Indeed, Appendix V does not include any substantive requirements, such as the requirement that regional haze SIPs include a five-factor BART analysis. These requirements are included elsewhere in the CAA, the Haze Rule, and the BART Guidelines.

Instead, EPA evaluates SIPs for compliance with the substantive requirements in the second step of EPA's review, which EPA must complete within one year after the SIP is determined to be complete by EPA or deemed complete by operation of law. See CAA Section 110(k)(2), (3), & (l) (providing a one-year deadline by which EPA must determine whether the SIP “meets all of the applicable requirements” or “interferes with any applicable requirement” of the Act); see also NRDC v. Browner, 57 F.3d 1122, 1126 (D.C. Cir. 1995) (“Under the two-stage procedure established in [section] 110(k), EPA first makes an essentially ministerial finding of completeness, a process taking at most six months. By contrast, the plan approval process may take up to twelve months due to the more extensive technical analyses necessary to ensure that the SIP meets Start Printed Page 5107the Act's substantive requirements.” (emphasis added)). Accordingly, a completeness determination in Step 1 does not deprive EPA of authority to disapprove a SIP in Step 2 for failure to comply with substantive requirements of the CAA, the Haze Rule, and the BART Guidelines. Instead, a completeness determination merely triggers EPA's duty to evaluate the substance of a SIP in the first instance and either approve or disapprove the SIP as necessary within one year. As explained above, EPA has authority to substantively review states' five-factor BART analyses in Step 2, and must disapprove a SIP if its analysis fails to comply with the requirements of the CAA, the Haze Rule, or the BART Guidelines. See Oklahoma, 723 F.3d at 1207-10. Thus, the comment is incorrect in stating that EPA's action here renders section 110(k)(1)(B) meaningless.

Moreover, courts have repeatedly stated that EPA does not lose its statutory authority to act under the CAA for a failure to meet its statutory deadline and that the proper remedy in the case of delay is for a party to seek an order to compel action. Oklahoma v. EPA., 723 F.3d 1201, 1224 (10th Cir. 2013) (explaining that although the CAA “undoubtedly requires that the EPA promulgate a FIP within two years, it does not stand to reason that it loses its ability to do so after this two-year period expires”); Montana Sulphur & Chem. Co. v. EPA., 666 F.3d 1174, 1190 (9th Cir. 2012) cert. denied, 133 S. Ct. 409, (2012) (explaining that although CAA has an “explicit deadline . . . it does not follow that the agency loses authority to act if it fails to meet that deadline”); Gen. Motors Corp. v. United States, 496 U.S. 530, 541 (1990) (holding EPA does not lose authority under CAA because “other statutory remedies are available when EPA delays action on a SIP revision”); see also Brock v. Pierce Cnty., 476 U.S. 253, 260 (1986) (holding that when “there are less drastic remedies available for failure to meet a statutory deadline . . . courts should not assume that Congress intended the agency to lose its power to act”).

Comment: Two months after the period for commenting on EPA's 2012 proposal closed, EPA and the special interest groups again modified the consent decree to allow EPA additional time to take action on Wyoming's SIP. See Stip. To Extend Deadline in Consent Decree., WildEarth Guardians (D. Colo. Oct. 3, 2012). Then, two months after extending the deadline for action on Wyoming's SIP, EPA asked the court to again extend EPA's deadline, this time until September 27, 2013. Def. Unopposed Mot. To Modify Two Deadlines in Consent Decree, at 1, WildEarth Guardians (Dec. 10, 2012).

As grounds for the request, EPA cited the special interest groups' comments, which EPA asserted “necessitate[d] re-proposal of the rule.” Id. at 3-4. The court, in turn, granted EPA's request. Order To Modify Consent Decree, WildEarth Guardians (Dec. 13, 2012). Even after extending its deadline to take action on Wyoming's SIP three times, EPA still needed more time. So, on March 25, 2013, EPA and the special interest groups again agreed to extend EPA's deadline for action on Wyoming's SIP. Stip. To Extend Deadlines in Consent Decree, WildEarth Guardians (March 25, 2013) (extending deadline until Nov. 21, 2013). Seemingly as a condition for obtaining the special interests groups' consent to the extension, EPA ostensibly agreed to a timetable for Wyoming sources to install emission controls faster than what Wyoming proposed. Compare id. at 2, ~ 6 (“EPA will propose to determine, for each source subject to BART, the period of time for BART compliance that is as expeditious as practicable”), with 78 FR 34778 (“We propose that PacifiCorp meet our proposed emission limit . . . as expeditiously as practicable, but no later than five years after EPA finalizes action”). Had Wyoming known when EPA proposed the consent decree in 2011 that EPA would commit to a particular course action on Wyoming's SIP, rather than just a date for taking some unspecified action, Wyoming would have sought to intervene in the litigation.

Response: EPA disagrees that it committed to any particular course of action in the Consent Decree. The Consent Decree only specified a timetable for EPA to promulgate rules consistent with its statutory obligations under the CAA, but did not commit the EPA to any particular course of action not already required by law. In fact, the Consent Decree specifically states: “Nothing in this Consent Decree shall be construed to limit or modify any discretion accorded EPA by the CAA or by general principles of administrative law in taking the actions which are the subject of this Consent Decree, including the discretion to alter, amend, or revise any final actions contemplated by this Consent Decree.”

We disagree with the commenter's allegations that EPA agreed in the Consent Decree to a timetable for Wyoming sources to install BART controls faster than what Wyoming proposed in its SIP. Paragraph 6 of the Stipulation To Extend Deadlines in Consent Decree, WildEarth Guardians (March 25, 2013) specifies that:

By May 23, 2013, EPA shall sign a notice of re-proposed rulemaking in which it proposes approval of a SIP, promulgation of a FIP, partial approval of a SIP and promulgation of a partial FIP, or approval of a SIP or promulgation of a FIP in the alternative, for the State of Wyoming, to meet the regional haze implementation plan requirements that were due by December 17, 2007, under 40 CFR 51.309(g). In its re-proposal, EPA will propose to determine, for each source subject to BART, the period of time for BART compliance that is as expeditious as practicable, as required by 42 U.S.C. Section 7491. (emphasis added).

Commenter neglects to include the last phrase in this provision in its comment—“as required by 42 U.S.C. Section 7491.” The Consent Decree required EPA to meet that CAA requirement.[75] Therefore, the Consent Decree mirrors and is consistent with the CAA BART requirements.

Finally, we are acting consistently with the Act and the RHR, as we discuss in detail elsewhere throughout this final action.

13. Monitoring, Recordkeeping and Reporting

Comment: EPA proposed a FIP for all monitoring, recordkeeping, and reporting requirements related to BART and reasonable progress sources for which there is a SIP or FIP emissions limit. EPA notes that the State's monitoring, recordkeeping, and reporting language in their SIP includes references to WAQSR chapters that EPA has not approved as part of the SIP and erroneously concludes that this means Wyoming's requirements are not federally enforceable.

Wyoming does not concur with EPA's findings. The monitoring, recordkeeping and reporting language in the State's regional haze SIP is taken directly from air quality permits issued under the SIP-approved permitting provisions in Chapter 6, Section 2 of the WAQSR, and are therefore federally enforceable. See 40 CFR 52.2620(c)(1). When drafting the monitoring, recordkeeping, and reporting requirements, the State incorporated the requirements of 40 CFR part 60 for trona facilities and for EGUs. Specifically, 40 CFR part 60 subparts D and Da were incorporated into the monitoring conditions for each Start Printed Page 5108BART permit. In the case of EGUs, by relying upon subpart D and Da, the State is also incorporating the requirements of 40 CFR part 75, since the monitoring provisions of subpart Da refer back to the continuous emissions monitor requirements under the Acid Rain Program codified in 40 CFR part 75.

While Wyoming allows for data substitution using the methodology prescribed in 40 CFR part 75, this is only applicable to annual emissions to account for periods when the continuous emissions monitor is down and the emissions unit continues to operate. Substituting data for these operating periods is more conservative than removing them altogether. EPA asserts that there are numerous clarifications and rewording needed; however, these monitoring, recordkeeping, and reporting requirements are currently in effect for PacifiCorp and Basin Electric units, and the companies are able to demonstrate compliance using them. Furthermore, the recordkeeping and reporting requirements contained in Chapter 5, Section 2 of the WAQSR are modeled after 40 CFR part 60, as these provisions are delegated to the State by EPA. See 40 CFR 60.4(c). While WAQSR Chapter 5, Section 2 is not part of Wyoming's SIP, the requirements therein are analogous to the federal New Source Performance Standards requirements and are made federally enforceable through incorporation by a Chapter 6, Section 2 BART permit and EPA's delegation to Wyoming.

Since these monitoring, recordkeeping, and reporting requirements are contained in federally enforceable permits and the affected companies are already able to demonstrate compliance with the BART emissions limits using them, Wyoming concludes that it did include appropriate and adequate monitoring, recordkeeping, and reporting requirements in the SIP.

Response: EPA disagrees with this comment. As discussed above in this section, EPA's approach in this action is entirely consistent with section 169A(b)(2) which, as we wrote when we promulgated the BART Guidelines, “provides that EPA must require SIPs to contain emission limits, schedules of compliance, and other measures as may be necessary to make reasonable progress towards meeting the goal” (emphasis added). 70 FR 39120. The regulations require that the states “must submit an implementation plan containing emission limitations representing BART.” 40 CPR 51.308(e). The Guidelines require that states “must establish an enforceable emission limit for each subject emission unit at the source and for each pollutant subject to review that is emitted from the source.” 70 FR 39172. CAA section 110(a)(2)(A) also requires that SIPs shall “include enforceable emission limitations . . . as may be necessary or appropriate to meet the applicable requirements of [the Act].”

Therefore, EPA disagrees that the use of BART permits to implement the monitoring, recordkeeping, and reporting necessary to ensure compliance with BART emission limitations is adequate under the CAA. In addition, in response to another comment, we are removing the requirements for annual emission limits for BART and reasonable progress sources. (See section IV.C.3 of this rulemaking). Thus, the point raised by the commenter pertaining to data substitution no longer applies to our final action.

B. Modeling

1. General Comments

Comment: One commenter stated that EPA must re-evaluate its method for assessing visibility impacts from wildfires or states will never be able to achieve natural background goals. The commenter went on to say that EPA should (1) eliminate the impacts from fire from the annual contribution to the deciview analysis or (2) properly incorporate it into the natural background equation to establish a glide path states can achieve. The commenter provided graphical data from the Interagency Monitoring of Protected Visual Environments (IMPROVE) network to show the contributions to light extinction from organic carbon, elemental carbon, and nitrate.

Response: EPA disagrees that we must re-evaluate our methods in this action. However, EPA agrees that wildfires can be an important source of visibility impairment, especially in the western states during the summer wildfire season. EPA recognized that variability in natural sources of visibility impairment causes variability in natural haze levels as described in its “Guidance for Estimating Natural Visibility Conditions Under the Regional Haze Rule.” [76] The preamble to the BART guidelines (70 FR 39124) describes an approach used to measure progress toward natural visibility in Mandatory Class I areas that includes a URP toward natural conditions for the 20 percent worst days and no degradation of visibility on the 20 percent best days. The use of the 20 percent worst natural conditions days in the calculation of the URP takes into consideration visibility impairment from wild fires, windblown dust and other natural sources of haze. The “Guidance for Estimating Natural Visibility” also discusses the use of the 20 percent best days and the 20 percent worst days estimates of natural visibility, provides for revisions to these estimates as better data becomes available, and discusses possible approaches for refining natural conditions estimates.[77] The commenter does not identify any way in which EPA's action was inconsistent with this guidance or the RHR.

Comment: At the hearing, Governor Mead and representatives of Wyoming, as well as industry representatives, argued that worsening haze has been caused by wildfires.

Response: We acknowledge the commenter's points on wildfires,[78] and that they can be an important source of visibility impairment, especially in the western states during the summer wildfire season. As discussed in more detail above and in our proposed notice, EPA recognizes that variability in natural sources of visibility impairment cause variability in natural haze levels and provided approaches to address this in the preamble to the BART guidelines (70 FR 39124). However, while we acknowledge that wildfires contribute to regional haze, the BART CALPUFF Start Printed Page 5109modeling has demonstrated that Wyoming's BART sources are also significant contributors to regional haze.

Comment: Although the various BART application analyses conducted by Wyoming for PacifiCorp's BART units note that Wyoming conducted a “comprehensive visibility analysis covering all three visibility impairing pollutants,” the analyses also state: “While visibility impacts were addressed in a cumulative analysis of all three pollutants, Post-Control Scenario B is directly comparable to Post-Control Scenario A as the only difference is directly attributable to the installation of SCR. Subtracting the modeled values from each other yield the incremental visibility improvement from SCR.” In other words, Wyoming clearly considered—and made available to EPA—the very specific NOX information that EPA claims it “was not possible for EPA, or any other party, to ascertain.” Simply claiming it “was not possible for EPA” to ascertain results from available information does not justify EPA in rejecting Wyoming's NOX BART determinations. Wyoming had considered SCR-specific visibility information. EPA cannot use the alleged lack of this information to justify requiring SCR as BART.

Response: We disagree with all aspects of this comment. Although a state is not required by EPA's regulations to model the visibility impacts from all possible control alternatives if the state selects the most stringent controls available as BART that is not what happened here. Wyoming rejected SNCR and SCR as BART without adequately assessing the visibility benefits of these control strategies. Given the cost effectiveness of these controls, the State's failure to consider visibility impacts was not reasonable and was inconsistent with the CAA and regulations. EPA was compelled to perform additional CALPUFF modeling for NOX BART determinations to allow for consideration of visibility impacts. For example, while Wyoming took into consideration the degree of visibility improvement for other BART NOX control options for the PacifiCorp EGUs, such as SCR, the State did not do so for SNCR. The visibility improvement for SNCR was neither provided in the State's SIP nor made available to the EPA. Wyoming did not assess the visibility improvement of SNCR despite having found it to be a technically feasible control option, and having considered a number of the other statutory factors for SNCR, such as costs of compliance and energy impacts. Given that nothing in the State's analysis suggested that SNCR was inappropriate, Wyoming's failure to consider the visibility improvement of SNCR is clearly in conflict with the statutory requirements set forth in section 169A(g)(2) of the CAA, which require that states take into consideration “the degree of improvement in visibility which may reasonably be anticipated to result from the use of such technology.” Since Wyoming did not do so, EPA conducted additional CALPUFF modeling to fill this gap in the State's visibility analysis.

In addition, as stated in our 2012 proposed rule and in our 2013 re-proposal, it was not possible for EPA, or any other party, to ascertain the visibility improvement from the NOX control options as emission reductions for multiple pollutants were modeled together. That is, since the visibility improvement for each of the State's control scenarios was due to the combined emission reductions associated with SO2, NOX, and PM controls, it was not possible to isolate what portion of the improvement was attributable to the NOX controls alone. For this reason, in the modeling conducted by EPA, we held SO2 and PM emission rates constant (reflecting the “committed controls” for those pollutants identified by Wyoming), and varied only the NOX emission rate. This allowed us to isolate the degree of visibility improvement attributable to the NOX control option.

We do agree that Wyoming's analysis allows for the incremental comparison between two post-control options (Post-Control Scenario A and Post-Control Scenario B). However, the BART Guidelines require more than this, stating that you should “[a]ssess the visibility improvement based on the modeled change in visibility impacts for the pre-control and post-control emission scenarios.” [79] That is, it is not sufficient to assess only the incremental visibility between control options, you must also assess the visibility improvement of each control option relative to the pre-control scenario. Therefore, Wyoming clearly did not assess visibility improvement in a manner consistent with that explicitly prescribed by the BART Guidelines.

In summary, while States may have some discretion in how to determine visibility impacts, Wyoming did not fulfill the basic statutory requirement to consider the visibility improvement of each of the NOX control options they identified as technically feasible. They also did not assess visibility improvement in a manner consistent with the approach prescribed by the BART Guidelines. As a result, EPA concluded it would be appropriate to conduct additional CALPUFF modeling.

Comment: The egregiousness of EPA's actions becomes even more apparent when comparing EPA's conclusions regarding cost and visibility impacts for certain of PacifiCorp's BART units against the cost and visibility impact conclusions reached by Wyoming for the same units. Table 2 in our comments provides a comparison between Wyoming's modeled delta deciview improvements and EPA's delta deciview improvements based on the “new information” EPA claims it has developed. Recognizing EPA's conclusion that one deciview is barely perceptible to the human eye and considering the inaccuracies and limitations of the model inputs and versions of the visibility models being used, there is no significant difference between Wyoming's results and EPA's results. Additionally, without any “bright line” test regarding the amount of visibility improvement that justifies a given control device, EPA cannot show that these insignificant differences would have any impact on the BART determinations for PacifiCorp's BART units.

Response: With respect to the “bright line” test, EPA's task in conducting modeling for developing our FIP is to conduct modeling in accordance with the CAA and RHR using our best scientific and technical judgment. We then consider the modeling results, along with the other BART factors, in making the BART determinations. While it is permissible to establish a “bright line,” we have not done so. Furthermore, as we discuss in detail in section IV of this final notice and elsewhere in response to comments, we disagree with commenter's assertions that there are no significant differences between Wyoming's results and EPA's results. We have addressed the issue of perceptibility elsewhere in our response to comments.

Comment: We received comments that by the year 2022, EPA's plan and Wyoming's plan achieve essentially identical results for visibility, and therefore, the proposed FIP would have no net visibility benefit over the SIP.

Response: We disagree. Our proposed FIP, by merit of requiring more stringent controls than those proposed in the SIP for some sources shown to cause or contribute to visibility impairment, results in greater visibility benefit. Although, based on our revised analyses for visibility impacts and costs of compliance and considering the five BART factors, we have revised some of Start Printed Page 5110our proposed control determinations, this assessment remains true for today's final rule. In particular, our final rule results in greater visibility improvement than the SIP for PacifiCorp Dave Johnston Unit 3 and Wyodak and Basin Electric Laramie River Units 1-3. The improvement in visibility stemming from the FIP, as compared to the SIP, can easily be discerned by reviewing relevant control options as found in Tables 2 through 17 of section III.A of this action.

Comment: On average, the 2000 baseline level for Class I areas in Wyoming is 11 deciviews. The 2064 natural background goal is 6 deciviews. IMPROVE data suggests that there is not this amount of nitrate improvement to be obtained. It appears EPA is trying to achieve a greater reduction for nitrates than is required at this time to reach the 2064 natural background goal. For Bridger, the total amount of deciview reduction for controls (6.08) exceeds the entire deciview reduction (all pollutants) from baseline to 2064. The value (6.08 deciviews) is also 10 fold higher than Wyoming's contribution to nitrate levels (0.62 deciviews—see Table 23 of EPA's proposed rulemaking) at the Class I area for the 2000-2004 baseline year. This value was calculated by multiplying 6.2% times the Class I area's 2000 baseline deciview value of 11.1. The same discussion is valid for Yellowstone as well, where the modeled nitrate reductions equal 2.27 deciviews and Wyoming's total reduction potential is only 0.82 deciviews.

Response: EPA disagrees with this comment. The commenter appears to be referring to annual mean estimates of visibility impairment at Class I areas, and comparing these estimates with the original CALPUFF modeled visibility improvements in our 2012 proposal. The values referred to in Table 23 of EPA's proposed rulemaking are mean estimates for the 20% worst visibility days. The BART Guidelines recommend that visibility impairment be evaluated for the 98th percentile contribution for each BART source. It is likely that 98th percentile visibility impacts will differ significantly from annual mean impacts, so it is not possible to directly compare our modeled impacts on the 98th percentile day to seasonal mean or the mean of the worst 20% days. However, we also note that in the revised final modeling included in this action, the CALPUFF modeled visibility improvements are less than the values cited above by the commenter from the original proposal, and the commenter's comparisons are no longer relevant.

Comment: The measured visibility impairment at IMPROVE stations offers a more certain and reliable quantification of the actual cause of visibility impairment than the CALPUFF model approved by the EPA for BART visibility assessments. CALPUFF is an EPA-approved model for long-range transport, as described in the EPA's “Guideline on Air Quality Models” 40 CFR Part 51 Appendix W, but only for the modeling of PSD increments. The treatment of chemical transformation, which is a crucial aspect for any model that is used for visibility assessments, is considered to be inadequate within CALPUFF. In fact, the lead modelers at the EPA's Office of Air Quality Planning and Standards have initiated a formal regulatory process to more fully evaluate (and very possibly replace) CALPUFF as an EPA-preferred model for long-range transport.

Response: EPA disagrees that the use of monitoring data from the IMPROVE network would provide a more accurate assessment of the predicted degree of visibility from the use of controls at a source than does CALPUFF. The commenter has not explained how monitoring data could be used to assess the actual current contributions to visibility impairment, and in any case, models are needed to estimate the potential future visibility impacts from the use of a range of controls at a specific source. In recommending the use of CALPUFF for assessing source specific visibility impacts, EPA recognized that the model had certain limitations but concluded that “[f]or purposes of the regional haze rule's BART provisions . . . CALPUFF is sufficiently reliable to inform the decision-making process.” [80] EPA accordingly appropriately used CALPUFF in this action. We further note that the requirements of 40 CFR 51.112 and 40 CFR part 51, Appendix W, Guideline on Air Quality Models (GAQM) and the BART Guidelines which refers to GAQM as the authority for using CALPUFF, provide the framework for determining the appropriate model platforms and versions and inputs to be used. The use of CALPUFF is subject to GAQM requirements in section 3.0(b), 4, and 6.2.1(e) which includes an approved modeling protocol to use the current 5.8 version.[81]

In promulgating the BART guidelines, EPA addressed concerns with CALPUFF's treatment of chemical transformations by recommending that states use the 98th percentile of modeled visibility impacts,[82] an approach that EPA followed here, to address the possibility that the model could overestimate impacts. EPA's discussion of CALPUFF in this 2005 rulemaking addresses these issues at length.[83]

EPA's modeling in this action was consistent with the BART Guidelines and Appendix W. As explained in more detail above, in recommending the use of CALPUFF for assessing source specific visibility impacts, EPA recognized that the model had certain limitations but concluded that “[f]or purposes of the regional haze rule's BART provisions . . . CALPUFF is sufficiently reliable to inform the decision-making process.” [84] To the extent that the comment takes issue with the provisions in the BART Guidelines for use of CALPUFF as described above, the legal deadline for challenging the use of CALPUFF has passed. In addition we encourage the commenter to provide input in the event that EPA develops any new future visibility guidelines and predictive models.

With respect to the comment on the IMPROVE data, we have addressed this in the response to another comment.

Comment: Regional haze is affected by global geologic, atmospheric and anthropogenic sources. None of the sources are controllable to the extent of achieving “natural visibility conditions” at the targeted time frame. The quantification of “natural visibility” at any geographic point is irrational. Natural visibility is a temporal quantity and therefore any quantified value is subjective and not scientific. Regional haze is subject to global atmospheric conditions which provide dilutive action to the identified sources of haze (anthropogenic or otherwise). Atmospheric conditions are directly related to the baseline eleven year solar cycle. To have any form of validity, the collection period would necessarily encompass at least one full solar cycle, arguably two full cycles. Furthermore, the dilutive effects of atmospheric conditions (and therefore, the quantification of visibility) are directly related to the known periodic oceanic events commonly referred to as “el Niño” and “la Niña”. These events have been monitored and quantified since 1950 and occurrences are sub-categorized as weak, moderate and strong. The periodicity of strong events Start Printed Page 5111for both “el Niño” and “la Niña” is every 9-11 years. The last strong “el Niño” occurred in 1997. The last two strong “la Niña” events occurred in 2010 and 1999. The baseline data collection for regional haze is ignorant of these significant atmospheric events, which makes the data collection period irrelevant and therefore the “baseline visibility” invalid.

Response: EPA disagrees with the characterizations in this comment. EPA recognizes that a variety of global scale, natural emissions sources affect natural visibility levels at Class I areas, and we described methods used to assess natural haze levels.[85] We disagree that it is necessary to model visibility impairment for one or two full solar cycles. The formation of fine particulate matter, and subsequent impacts on regional haze, depend on variations in local meteorological conditions. Variability in meteorological conditions is primarily determined by seasonal weather conditions, and the modeling period of calendar years 2001-2003 used in our analysis includes substantial variability in weather conditions. While phenomena such as el Niño and la Niña can affect the frequency of extreme events, our modeling analysis is based on the 98th percentile visibility impacts and is therefore designed to exclude extreme events. El Niño and la Niña events may also affect annual total precipitation, temperature and other meteorological parameters, however, the commenter has not provided any evidence that the 98th percentile visibility impacts would differ significantly during an el Niño and la Niña year. We believe that it is sufficient to model visibility impairment for a 3 year period. In the preamble to the BART Guidelines, we discussed meteorological variability and explained how use of the 98th percentile would minimize the likelihood that the highest modeled visibility impacts would be caused by unusual meteorological conditions. 70 FR 39121. As explained above, our use of the 98th percentile is consistent with the BART Guidelines and in recommending the use of CALPUFF for assessing source specific visibility impacts, EPA recognized that the model had certain limitations but concluded that “[f]or purposes of the regional haze rule's BART provisions . . . CALPUFF is sufficiently reliable to inform the decision-making process.” [86] Thus to the extent that the comment suggests that the BART Guidelines should have used a different percentile to account for solar variability in solar cycles, the legal deadline for challenging the use of CALPUFF has passed. In addition we encourage the commenter to provide input in the event that EPA develops any new future visibility guidelines and predictive models.

2. EPA Modeling

a. Comments on EPA Modeling

Comment: Several commenters have argued that EPA should have used updated models and procedures for its revised modeling. In addition, several commenters have argued that the State's Protocol was overly conservative in its treatment of background ammonia concentrations, and that monitoring data show that background ammonia levels are significantly lower than the 2 ppb concentration specified in the Protocol. Commenters in particular directed EPA's attention to ambient monitoring data for ammonia and particulate ammonium at a monitoring site in Boulder in western Wyoming and at several Class I areas. Ambient monitoring at the Boulder site was performed from 2006 to 2011 and these data were recently published by Li et al.,[87] while the monitoring data at the Class I areas for an 8 month period from April 2011 to January 2012 and were presented at a conference in 2012.[88]

Response: We agree with the comments that we should perform new modeling using updated model versions and different background ammonia concentrations. In this response we explain why we originally used the same modeling approach used by Wyoming and why we have updated the modeling for this action. In 2006, the State adopted its “BART Air Modeling Protocol” (Protocol) [89] that specified the approach for using the CALPUFF modeling system to evaluate BART visibility impacts, and the State and several BART sources performed modeling studies that were consistent with that protocol. For our original proposal in 2012, EPA performed additional modeling using the State's Protocol to evaluate a limited number of emissions scenarios that the State had not evaluated in its modeling. EPA recognized that there had been updates to CALPUFF modeling guidance and model versions after 2006 when the State adopted the Protocol, however, in our original proposal in 2012, which included a limited, gap-filling analysis, we proposed that it was preferable to maintain consistency with the modeling approach that the State had adopted in its Protocol. In our re-proposal on June 10, 2013, EPA recognized that some of the options used from the State's Protocol were inconsistent with BART Guidelines, such as the approach for determining baseline emissions. As a result, for the re-proposal EPA performed new modeling using updated emissions data for the baseline period and for all emissions control technologies, however, EPA continued to use the State's Protocol for EPA's re-proposed modeling analysis.

After evaluating comments on the re-proposal, EPA determined that it was necessary to remodel all of the baseline and control technology scenarios using different background ammonia concentrations for the BART sources that we reconsidered for this action, including Naughton, Jim Bridger, Laramie River, Dave Johnston and Wyodak. Because this approach represents a significant change from State's original Protocol, we believed that it was appropriate to develop a new modeling protocol that also adopts the current model version approved for regulatory use, CALPUFF version 5.8, and current regulatory default options. In making this decision, we considered the merits of continuing to use the State's original protocol versus the benefits of using the updated CALPUFF model that became available after the State's Protocol was adopted, and different background ammonia concentrations based in part on data that have also become available since then, and we concluded that it was necessary to adopt an updated Protocol [90] to respond fully to a number of issues raised by various commenters. The new EPA Protocol for modeling of Wyoming BART uses the same CALPUFF model version 5.8 as did the protocol that we previously adopted for modeling BART sources visibility impacts in Montana.[91]

Start Printed Page 5112

EPA evaluated the comments and the ambient ammonia monitoring data submitted by commenters. EPA understands that there is no single accepted method for estimating the background concentration of ammonia, and that any method will have advantages and disadvantages. The lack of consensus on a method was a factor in EPA's decision to set aside the 2 ppb concentration value specified in the State's Protocol and instead to rely in part on the default values in Interagency Workgroup on Air Quality Modeling (IWAQM) Phase 2 report [92] and in part on monitoring data. Specifically, for BART sources in western Wyoming we performed two modeling runs, one relying on an IWAQM default value and the other relying only on monitoring data. As presented below, EPA's two sets of modeling results for this BART source support our final BART determinations, as they both show similar visibility benefits. As explained below, we relied only on an IWAQM default value for BART sources in eastern Wyoming.

The 1998 IWAQM report is the only guidance available for choosing ammonia background concentrations. The IWAQM Phase 2 report relied on a 1992 review of ambient monitoring data for ammonia by Langford et al.[93] and explains that: “. . . the formation of particulate nitrate is dependent on the ambient concentration of ammonia, which preferentially reacts with sulfate. The ambient ammonia concentration is an input to the model. Accurate specification of this parameter is critical to the accurate estimation of particulate nitrate concentrations. Based on a review of available data, Langford et al. suggest that typical (within a factor of 2) background values of ammonia are: 10 ppb for grasslands, 0.5 ppb for forest, and 1 ppb for arid lands at 20 degrees Celsius. Langford et al. (1992) provide strong evidence that background levels of ammonia show strong dependence with ambient temperature (variations of a factor of 3 or 4) and a strong dependence on the soil pH. However, given all the uncertainties in ammonia data, IWAQM recommends use of the background levels provided above, unless specific data are available for the modeling domain that would discredit the values cited. It should be noted, however, that in areas where there are high ambient levels of sulfate, values such as 10 ppb might overestimate the formation of particulate nitrate from a given source, for these polluted conditions. Furthermore, areas in the vicinity of strong point sources of ammonia, such as feed lots or other agricultural areas may experience locally high levels of background ammonia.” [94]

The IWAQM Phase 2 report also states that “[i]n a refined analysis, “the background concentrations of ozone and ammonia are allowed to vary in time and space.” [95] In summary, given numerous uncertainties in ammonia data, the IWAQM Phase 2 report recommends use of the background values it provides for different land use categories, unless specific data is available in the modeling domain as a more accurate substitute for its recommended default values, and allows for the consideration of background ammonia concentrations that vary seasonally or spatially.

EPA has reviewed monitoring data for ammonia and ammonium that have been collected at one site in western Wyoming since 2006.[96] We have determined that the monitoring data from this site are the most representative monitoring data available for characterizing ammonia and ammonium background levels in the modeling domains used for western Wyoming as explained in detail below. Based on this analysis, EPA has concluded that the constant 2 ppb background concentration used by the State is substantially higher than the observed combined ammonia and particulate ammonium concentrations at this monitoring site in western Wyoming, especially during the winter season when the observed sum of ammonia and particulate ammonium concentration were typically much lower than 2 ppb.[97] Therefore, for two BART sources in western Wyoming (PacifiCorp's Naughton and Jim Bridger) in one of our modeling runs we modeled using monthly varying ammonia background concentrations based on the combined observed concentration of ammonia and particulate ammonium at this monitoring site in western Wyoming, as described in the EPA Protocol.[98] In a second modeling run for these two BART sources, we modeled using the default IWAQM ammonia concentration of 0.5 ppb for forested areas. Although western Wyoming includes a mixture of arid grasslands and forested areas, we used the IWAQM default value of 0.5 ppb for forested areas because the Class I areas in the modeling domain are primarily forested, and because the monitoring data more nearly matched the IWAQM recommendation for forests than the default for the other land types.

When Wyoming adopted its Protocol in 2006, the State explained that: “A constant background ammonia concentration of 2.0 ppb is specified. This value is based upon monitoring data from nearby states and IWAQM guidance. Experience suggests that 2.0 ppb is conservative in that it is unlikely to significantly limit nitrate formation in the model computations.” [99]

The Wyoming Protocol specified a constant 2 ppb background ammonia concentration, but the Protocol (addressing source owners performing their own BART analyses) states that: “[i]f you believe that ammonia limiting is appropriate for a specific BART analysis, justification should be discussed with the Division prior to its use.” [100] The Wyoming Protocol in the text quoted here refers to a method for correcting CALPUFF for ammonia limiting conditions, which indicates that the State recognized the possibility that its protocol could be overly conservative in its treatment of ammonia. Therefore, we believe it is appropriate and consistent with the IWAQM Phase 2 report and the intention of the State's Protocol to model a BART source in western Wyoming using both the newly available monitoring data, and the default concentration recommended in the IWAQM Phase 2 report, to represent background ammonia concentrations more accurately than would be the case if we retained the 2 ppb value specified in the State Protocol.

We describe here in more detail the ambient monitoring data from the site in western Wyoming and our use of those Start Printed Page 5113data. Li et al (2013) [101] report on an analysis of ambient monitoring data conducted from 2006 to 2011 at the Pinedale site in western Wyoming in an area with significant oil and gas production. The monitoring site included measurements of gaseous ammonia (NH3) and particulate ammonium (NH4) and a complete set of acidic species including gaseous nitric acid and particulate nitrate and sulfate. The complete set of measurements makes it possible to determine the total ammonia and ammonium concentration (NHX=NH3+NH4) and to determine if ammonium nitrate formation is limited by the availability of excess NH3. Li et al found significant seasonal variability in NH3 and NH4 concentrations at the site and concluded that excess nitric acid was present in winter, while NH3 concentrations were close to zero in winter, indicating that formation of ammonium nitrate was limited by the availability of ammonia at this site in winter.[102] Because ammonia at this monitoring site may have been affected by nearby sources of sulfuric acid and nitric acid, which would deplete the gaseous ammonia concentration locally, we used the combined gaseous and particulate measurement of NHX to estimate monthly average background ammonia concentrations, with a low concentration in January of 0.36 ppb and a peak concentration in 1.12 ppb in August.[103] The monitor is located in an area that includes nearby sources of ammonia emissions from livestock and other anthropogenic sources, including a nearby area of oil and gas production activity, which could result in locally elevated ammonia compared to the area more immediate to the BART source and to the nearest Class I areas. Moreover, some of the particulate ammonium observed at the site was irreversibly bound with sulfate and may have had a non-local origin due to long range transport of ammonium sulfate. These factors mean that this estimate of local background may tend to overestimate the regional background ammonia concentration and thus also overestimate the visibility benefit due to NOX reductions at sources. There may be other unknown factors also working in the same direction or in the other direction. For example, monitoring methods for ammonia and quality systems for characterizing monitoring accuracy have not been standardized to the extent that methods for other ambient compounds have been, resulting in uncertainty as to whether there is bias in the measurements.

As discussed above, we also modeled the PacifiCorp Naughton and Jim Bridger BART sources in western Wyoming using a constant background ammonia concentration of 0.5 ppb, which is the IWAQM Phase 2 report default recommendation for forested areas. Model results using either approach to determine an ammonia background concentration support our final BART determinations, as they both show similar visibility benefits.

We also remodeled the BART sources in eastern Wyoming including Wyodak, Dave Johnston and Laramie River. The Class I areas most impacted by these BART sources are Badlands and Wind Cave National Parks in western South Dakota. The closest long-term ammonia monitoring site to these Class I areas is at site at Beulah, North Dakota operated by the State of North Dakota. This site is about 280 miles from Badlands National Park and about 300 miles away from Wind Cave NP. The area around Beaulah site includes a mix of agricultural lands and grassland. Measured monthly average gaseous ammonia concentrations at the Beulah site in central North Dakota vary from about 1 to 2 ppb throughout the year, with the lowest values in fall and winter.[104] Additionally, combined ammonia and particulate ammonium measurements have been reported at Wind Cave National Park for an 8 month period by Chen et al.[105] This study measured NHX daily average concentrations in the range of about 0.05 to 4 ppb, with an annual average concentration of 0.75 ppb. We considered these monitoring data sufficient to put into serious doubt the 2 ppb concentration specified in the State's Protocol, but insufficient to support either a single or an alternative modeling run based on monitoring data. Therefore, we chose to rely on the IWAQM Phase 2 report for a single set of modeling runs for the BART sources in eastern Wyoming. The area around Wind Cave National Park includes forested areas, including Black Hills National Forest, while the area around Badlands National Park includes a mix of arid and grass lands. While there is uncertainty in the appropriate background ammonia level in this region, we used the IWAQM Phase 2 report recommended value of 1 ppb for arid lands because it falls within the range of the limited monitoring data available in nearby regions and because it is represents an intermediate level for the different land use types within the region.

For both the eastern and western Wyoming modeling domains and runs, we corrected for ammonia limiting conditions. The correction for ammonia limiting conditions is a post-processing step in POSTUTIL, one component of the CALPUFF modeling system. Because CALPUFF simulates each BART unit individually, the background ammonia concentration is assumed by the model to be fully available to react with emissions from each unit. In reality, the total emissions from the combined units compete for the available ammonia. Also, because CALPUFF simulates multiple parcels of air originating at each unit, there is the possibility that different parcels can overlap at a Class I area. The ammonia limiting correction in POSTUTIL is designed to repartition the available ammonia to react with emissions from all of the units and overlapping air parcels, thereby avoiding double counting of the background ammonia. We used the same ammonia limiting correction in our modeling for Montana BART [106] sources, and this is a standard configuration in the CALPUFF modeling system.

In summary, we concluded that it was more reasonable to model visibility impacts in western Wyoming using both the default IWAQM Phase 2 report recommendation for forested areas and using a seasonally varying NH3 concentration in western Wyoming that was based on long term monitoring of NH3 and NH4 from one site, than to use the 2 ppb concentration specified in the State's Protocol. We found that the visibility impacts predicted for the various control levels at the PacifiCorp Naughton and Jim Bridger BART sources were very similar with these two approaches and that either set of results supports the same BART determination for these sources. Therefore, we did not have to make a Start Printed Page 5114determination that one approach was superior to the other, and we have not done so. We also determined that it was appropriate to use the default IWAQM Phase 2 report recommendation of 1 ppb for eastern Wyoming.

Comment: A commenter believes that the improvements from the addition of SCR technology are based on highly conservative models which overestimate the deciview reduction as compared to actual monitored data collected at the IMPROVE stations. Subsequently, the improvement in visibility provided by SCR is not supported by the escalated costs of $600 million above the cost to install Wyoming's proposed control technologies.

Response: EPA disagrees with this comment to the extent that it takes issue with EPA's modeling. As discussed in response to other commenters, EPA recognized that the State's original modeling protocol specified a fixed value of 2 ppb for background ammonia. EPA has performed new modeling using lower background concentrations and using a correction for ammonia limiting conditions when modeling multiple units from a single BART source.

EPA also used the most current regulatory approved versions of the models in the updated modeling. EPA has recognized that the CALPUFF model can be conservative in estimating visibility impairment, and therefore, EPA has used the 98th percentile model results instead of the maximum modeled visibility impairment to address the possibility of model overpredictions. In recommending the use of CALPUFF for assessing source specific visibility impacts, EPA recognized that the model had certain limitations but concluded that “[f]or purposes of the RHR's BART provisions, CALPUFF is sufficiently reliable to inform the decision making process, e.g., see 77 FR 39123.[107] As discussed above, to the extent that the comment takes issue with the use of the 98th percentile, the legal deadline for challenging the use of CALPUFF has passed, but we encourage the commenter to provide input in the event that EPA develops any new future visibility guidelines and predictive models.

Comment: The commenter states that EPA made five errors in its visibility modeling, including (1) given the general inaccuracy in CALPUFF unit-specific modeling, not allowing Wyoming the deference accorded it under the CAA; (2) relying upon an outdated CALPUFF method of visibility modeling, contrary to EPA precedent; (3) violating the applicable modeling guidance, Appendix W, by not using the “best” science; (4) violating the Data Quality Act by not using the “best” science; and (5) failing to recognize the gross overestimations and internal inconsistencies in EPA's modeling approach.

Response: EPA disagrees with this comment. In response to item (1): In promulgating the BART guidelines we made the decision in the final BART Guideline to recommend that the model be used to estimate the 98th percentile visibility impairment rather than the highest daily impact value as proposed. We made the decision because “there are other features of our recommended modeling approach that are likely to overstate the actual visibility effects of an individual source. Most important, the simplified chemistry in the model tends to magnify the actual visibility effects of that source. Because of these features and the uncertainties associated with the model, we believe it is appropriate to use the 98th percentile, a more robust approach that does not give undue weight to the extreme tail of the distribution.” 77 FR 39121. In regard to deference to the state, as discussed in more detail elsewhere in this document, Congress crafted the CAA to provide for states to take the lead for implementing plans, but balanced that decision by requiring EPA to approve the plans or prescribe a federal plan should the state plan be inadequate to meet CAA and regulatory requirements. Our action today is consistent with the statute and regulations. In response to item (2): We initially relied on the State's modeling Protocol, adopted in 2006, that specified model versions available at that time, but that have since been updated. In our original proposal we performed limited gap filling modeling that was consistent with the State's Protocol and that used the same model versions as the State. In this final action, as presented in more detail in the Protocol in the docket, due to a number of other changes in the protocol, we also updated the protocol to use the current regulatory version of the CALPUFF modeling system. In response to items (3) and (4): In recommending the use of CALPUFF for assessing source specific visibility impacts, EPA recognized that the model had certain limitations but concluded that “[f]or purposes of the regional haze rule's BART provisions . . . CALPUFF is sufficiently reliable to inform the decision-making process.” [108] EPA accordingly appropriately used CALPUFF in this action. EPA recognized that there were uncertainties in the science of the CALPUFF modeling system, and therefore used the less conservative 98th percentile value to model results to address the possibility that the model was overly conservative. We address concerns about the Data Quality Act (also referred to as the Information Quality Act), elsewhere in this document. In response to item (5): EPA recognized that the State's original Protocol was inconsistent with the IWAQM report and monitoring data because of the use of a constant 2 ppb ammonia concentration, and our modeling in this action relied on ambient monitoring data and the default values consistent with IWAQM Phase 2 report, to specify two alternatives for more realistic background ammonia concentrations in western Wyoming. We also reduced the background ammonia concentration from 2 to 1 ppb in eastern Wyoming, as discussed previously. A seasonal ammonia concentration was not adopted due to the lack of high quality monitoring data in eastern Wyoming; however, as discussed previously, the 1 ppb background estimate is consistent with IWAQM Phase 2 report. As discussed elsewhere, we also used the ammonia limiting correction for modeling multiple units from a single BART source to address concerns with the model being overly conservative.

Comment: Contrary to its own guidance, EPA failed to use the most realistic model. 40 CFR part 51, Appendix W, EPA's modeling guidance, demands that the “best” model should always be used. EPA failed to use the “best” model in Wyoming, which is CALPUFF 6.4. Therefore, EPA failed to follow Appendix W's requirements.

Response: We disagree with this comment. As explained above, EPA followed the RHR. Specifically, in recommending the use of CALPUFF for assessing source specific visibility impacts, EPA recognized that the model had certain limitations but concluded Start Printed Page 5115that “[f]or purposes of the regional haze rule's BART provisions . . . CALPUFF is sufficiently reliable to inform the decision-making process.” [109] EPA accordingly appropriately used CALPUFF in this action. The use of CALPUFF is subject to GAQM requirements in section 3.0(b), 4, and 6.2.1(e) which includes an approved protocol to use the current 5.8 version, which is the version we used for EPA's final modeling analysis. We did not use CALPUFF Version 6.4 because this version of the model has not been approved by EPA for regulatory use.

We made the decision in the final BART Guidelines to use less than the highest daily impact value for assessment of visibility impacts. We made this decision in response to comments we received expressing concern that the chemistry modules in the CALPUFF model are less advanced and that use of the 1st High was conservative and the knowledge that CALPUFF's simplified chemistry could lead to model overpredictions and thus be conservative. We decided in the BART Guideline to use the 98 percentile for CALPUFF based modeling results. We also received comments opposed to using the day with the worst meteorology, but the primary reason we changed to using a less stringent metric than the day with the highest visibility impact was because of concerns about overestimations in CALPUFF's simplified chemistry. As a result, we determined that it was appropriate to use the 98th percentile or 8th High value when modeling all days of the year instead of the 1st High value, also described as the Highest Daily impact level for each year modeled: “Most important, the simplified chemistry in the model tends to magnify the actual visibility effects of that source. Because of these features and the uncertainty associated with the model, we believe it is appropriate to use the 98th percentile—a more robust approach that does not give undue weight to the extreme tail of the distribution.” 110

Comment: EPA's modeling for its regional haze FIP action was inadequate and incomplete. (EPA failed to re-run WRAP regional modeling due to “time and resource constraints”). Therefore, EPA's regional haze FIP action violates the “Guideline on Air Quality Models,” 40 CFR part 51 Appendix W, the Information Quality Act and the implementing guidelines issued, respectively, by the U.S. Office of Management and Budget (OMB) and the EPA which require information disseminated by EPA to be accurate, complete, reliable and unbiased. The Information Quality Act and EPA's “Information Quality Guidelines” place a heightened standard on “influential” information, including scientific information regarding health, safety, or environmental risk assessments. EPA's inaccurate and incomplete visibility modeling is by definition “influential,” because EPA could reasonably determine that dissemination of the information will have or does have a clear and substantial impact on important public policies or important private sector decisions, such as the BART NOX determinations in EPA's regional haze FIP.

Therefore, this “influential” information must be based on best available science and data and supporting studies must be conducted in accordance with sound objective scientific practices and methods. EPA's Guidelines implementing the Information Quality Act expressly contemplate the correction of information disseminated by EPA that falls short of the “basic standard of quality, including objectivity, utility, and integrity,” established by either EPA's own Guidelines or those issued by OMB.

The commenter seeks correction to a number of errors and omissions in EPA's regional haze FIP with regard to CALPUFF modeling and EPA's failure to re-run the WRAP model. The commenter requests that EPA withdraw its regional haze FIP until these issues are resolved.

Response: As EPA explained in our Information Quality Guidelines, we believe “that the thorough consideration provided by the public comment process serves the purposes of the Guidelines, provides an opportunity for correction of any information that does not comply with the Guidelines, and does not duplicate or interfere with the orderly conduct of the action.” [111] Therefore, we are responding to the modeling comments and related comments regarding EPA's Guidelines and the Information Quality Act in this document.

WRAP performed regional photochemical modeling using both the Community Multi-Scale Air Quality modeling system (CMAQ) and Comprehensive Air Quality Model (CAMx) air quality models to evaluate progress toward attaining visibility goals using all projected emission changes from all source categories throughout the United States. WRAP did not perform regional photochemical modeling to evaluate the visibility impacts of individual BART sources. While WRAP did make assumptions regarding the level of emissions control that would be adopted by BART sources as part of its analyses, no state or EPA region has re-run the WRAP's regional photochemical models to assess individual BART source contributions to visibility impairment. Instead, the BART sources, the states, and EPA have used the CALPUFF model to evaluate contributions to visibility impairment from individual BART sources. This approach is consistent with the BART Guidelines that recommend that the CALPUFF model should be used to evaluate visibility impairment from individual BART sources. Additionally, while EPA supported development of WRAP CMAQ modeling in order to assist states in developing their RPGs and determining the cumulative benefit of an overall control strategy vis-à-vis the URP on the 20% worst days, our use of CALPUFF for evaluating visibility improvement from a single BART source is consistent with the BART Guidelines and also consistent with modeling performed by other states and EPA regional offices for individual BART sources.

We have responded to comments related to errors and omissions in the CALPUFF modeling in separate response to comments.

Comment: The commenter states that the revised EPA modeling, which used new information on emissions rates, did not significantly change the results identified in Wyoming's BART analyses. The commenter also states that there are small differences between EPA's and Wyoming's analyses which do not justify EPA rejecting Wyoming's BART determinations.

Response: We disagree that in all cases there are only small differences in EPA revised modeling and the State's modeling. Importantly, as described elsewhere in this document and in the docket for this action, EPA revised and corrected various inputs to the BART factors so that the analyses are consistent with the RHR and statutory requirements. While the difference at the most impacted Class I area from individual sources or units is some cases can be characterized as small, the cumulative differences from many small improvements can be significant. Whether such differences are significant will depend on the overall consideration of the BART factors. Because of the flaws in Wyoming's Start Printed Page 5116visibility and cost analyses for many of its BART sources, EPA could not be confident that Wyoming's BART determinations were reasonable without undertaking an appropriate analysis of the statutory factors.

Comment: The commenter states that Wyoming provided the required modeled visibility improvement information for SCR, and quotes from the State's analysis: “Post-Control Scenario B is directly comparable to Post-Control Scenario A as the only difference is directly attributable to the installation of SCR.” The commenter then concludes EPA did not lack the required information to evaluate visibility improvements.

Response: We have addressed this comment in a previous response, citing the requirement in the BART Guidelines that visibility must be assessed relative to the pre-control scenario (and not just incremental to other control scenarios). Moreover, there remain deficiencies, as presented elsewhere in this document and docket, with the State's BART modeling analyses that justify our rejection of the State's BART determinations for PacifiCorp. Most notably, as discussed in separate responses, the State did not assess the visibility improvement of SNCR as required by the CAA and BART Guidelines.

Comment: An older version of the CALPUFF modeling suite was used by EPA (CALPUFF model of March 2006 vintage and the CALPOST model of April 2006 vintage.) These older versions pre-date the latest Model Change Bulletin (MCB-D) of June 23, 2007. Since the analysis for the five Wyoming power plants was performed in February thru April 2012, we question why the older version was used and not the current CALPUFF 5.8 version, which was approved as the guideline version in June of 2007. We do not recommend use of the older versions of CALPUFF and CALPOST.

Response: As described in this action, our previous modeling continued to use the State's Protocol, including the older model versions, to maintain consistency with the State's modeling results. However, in this final action, we adopted a new modeling protocol that uses the current regulatory versions of the models, including the Model Change Bulletin suggested by the commenter. We determined that it was appropriate to adopt an updated modeling protocol because we made other significant changes in the State's modeling approach, and because we remodeled all emissions scenarios, there was no longer a need to use older model versions for consistency of comparison of our limited gap filling model results to the State's original modeling.

Comment: EPA should have used the most recent version of CALPUFF, or at a minimum, should have used the version that EPA requires for other regional haze SIPs. EPA has taken the position that CALPUFF Version 5.8 must be used for regional haze modeling. 77 FR 42834, 42854. However, EPA's unit-specific CALPUFF modeling in Wyoming, completed in April 2012, used CALPUFF Version 5.711a (originally released in 2004). Version 5.711a is eight years old, and several CALPUFF versions behind Version 5.8. While PacifiCorp believes the more modern and realistic CALPUFF Version 6.42 should be used, at a minimum EPA must abide by its own position and use Version 5.8 in evaluating the Wyoming regional haze SIP, which it failed to do. According to EPA's own statements, EPA's own modeling results should be discarded because EPA used an improper “alternative model” in Wyoming.

EPA should have used the most recent version of CALPUFF (Version 6.42) in Wyoming because it produces more realistic and accurate results. Version 6.42 contains needed refinements, such as a better “chemistry” module known as ISORROPIA (Version 2.1). CALPUFF Version 6.42 is more accurate because, as the FLMs have noted, Version 5.8 does not have the required settings to perform the new Method 8 visibility analysis. Additionally, CALPUFF Version 6.42 has been maintained by TRC, Inc., a private contractor, and has had many bug fixes and enhancements not included in CALPUFF Version 5.8. Most importantly, the previous chemistry modules used in Version 5.8 (and in the 5.711a Version EPA used here) also have been shown to overestimate nitrate concentrations in Wyoming by a factor of 3-4 and substantial improvements have been made to eliminate this over-prediction using the ISORROPIA module.

Despite all these advancements in modeling and modeling science, EPA conducted its modeling for its regional haze FIP in 2012 using the same (now outdated) CALPUFF version that PacifiCorp and Wyoming used five years ago, which has been shown to overestimate results by 300% to 400%. Rejecting Wyoming's modeling, and then using the same, outdated modeling approach, is arbitrary and capricious.

Response: As described in previous responses, we previously used the same modeling protocol adopted by the State for the purpose of our limited, gap filling modeling, so that we would have a consistent basis for comparison with the State's modeling. In this action we have updated the protocol to use the current regulatory versions of the models including CALPUFF version 5.8.

We did not use CALPUFF Version 6.42 because this version of the model has not been approved by EPA for regulatory use. EPA relied on version 5.8 of CALPUFF because it is EPA-approved version in accordance with the Guideline on Air Quality Models (“GAQM”, 40 CFR 51, Appendix W, section 6.2.1.e). EPA updated the specific version to be used for regulatory purposes on June 29, 2007, including minor revisions as of that date. The approved CALPUFF modeling system includes CALPUFF version 5.8, level 070623, and CALMET version 5.8 level 070623. CALPUFF version 5.8 has been thoroughly tested and evaluated, and has been shown to perform consistently with the initial 2003 version in the analytical situations for which CALPUFF has been approved. Any other version, and especially one with such fundamental differences in its handling of chemistry, would be considered an “alternative model”, subject to the provisions of GAQM section 3.2.2(b), requiring full model documentation, peer-review, and performance evaluation. No such information for the later CALPUFF versions that meet the requirements of section 3.2.2(b) has been submitted to or approved by EPA. Experience has shown that when the full evaluation procedure is not followed, errors that are not immediately apparent can be introduced along with new model features. For example, changes introduced to CALMET to improve simulation of over-water convective mixing heights caused their periodic collapse to zero, even over land, so that CALPUFF concentration estimates were no longer reliable.[112]

The change from CALPUFF version 5.8 to CALPUFF 6.4 is not a simple model update to address minor issues, but a significant change in the model science that requires its own rulemaking with public notice and comment before it can be relied on for regulatory purposes.

Furthermore, it should be noted that the U.S. Forest Service and EPA review of CALPUFF version 6.4 results for a limited set of BART applications showed that differences in its results from those of version 5.8 are driven by two input assumptions not associated Start Printed Page 5117with the chemistry changes in 6.4. Use of the so-called “full” ammonia limiting method and finer horizontal grid resolution are the primary drivers in the predicted differences in modeled visibility impacts between the model versions. These input assumptions have been previously reviewed by EPA and the FLMs and have been rejected based on lack of documentation, inadequate peer review, and lack of technical justification and validation.

Comment: The commenter states that EPA treats the results from the CALPUFF as being capable of accurately predicting visibility improvements down to the tenths or hundredths of a deciview, but that the model does not accurately predict visibility impacts at this level.

Response: As described in response to other comments, EPA recognized that there is uncertainty in the CALPUFF results, and EPA addressed this uncertainty by using the modeled 98th percentile visibility impairment rather than the maximum visibility impairment. EPA considers model changes on the order of tenths of a deciview to be useful for informing the BART decision process, consistent with BART modeling performed by other EPA regions and states.

Comment: PacifiCorp presented substantial information supporting the need to use improved and updated versions of the models and provided substantial information on the effects that the nitrogen oxides to nitrogen dioxide conversion rate and background ammonia concentrations have on modeled visibility impacts.

Response: We agree with the comment that the background ammonia concentration has a significant effect on model predicted visibility impacts. As described elsewhere in this action, we reviewed recent ambient monitoring data for ammonia and particulate ammonium, and concluded that the original background ammonia concentration of 2 ppb was inconsistent with the IWAQM Phase 2 report and monitoring data for estimating visibility impacts in Wyoming, especially in the western portion of the State. In the modeling results included in this action, we considered the default value of 0.5 ppb and also applied a seasonally varying background ammonia concentration in western Wyoming that was based on measurements from 2006 to 2011 of total ammonia and particulate ammonium at Pinedale, Wyoming.[113] We also reduced the background ammonia concentration to 1 ppb in the eastern portion of the State, and for both parts of the State we used an ammonia limiting correction for modeling multiple units from a single BART source to avoid double counting of the available ammonia. The use of more realistic ammonia background concentrations, the ammonia limiting correction, and the use of the 98th percentile modeled impact address the concern that the CALPUFF model could overestimate visibility impacts.

Comment: One commenter stated that we underestimated the background ammonia concentration in the CALPUFF modeling, and cited the IWAQM Phase 2 recommendations for default ammonia concentrations for grasslands, forest lands and arid lands, respectively, of 10, 0.5 and 1 ppb, at 20 degrees Celsius. The commenter stated that, because land use type can vary across the large domains used in the CALPUFF modeling, it would be appropriate to calculate a weighted average of the background ammonia based on the fractional land use type in the model domain.

Response: We disagree with this comment. The IWAQM Phase 2 report does not recommend calculating a weighted average of default ammonia concentration based on regional variation in land use types. The commenter provides no regulatory basis for use of a weighted average.

Comment: One commenter stated that the IWAQM recommended default background ammonia recommendations do not specifically account for strong point/area sources of ammonia, such as cattle feedlots, which are also scattered throughout the modeling domain and which generally add to the background ammonia levels. This commenter stated that some areas of the modeling domain, namely northeastern Colorado, are described as “ammonia rich”. For BART source analyses in Colorado, the recommended background ammonia value from the Colorado Department of Public Health and Environment (CDPHE) is 44 ppb, based on measurements conducted during the Northern Front Range Air Quality Study (NFRAQS), and therefore the Wyoming DEQ/EPA background ammonia concentration of 2 ppb might not carry sufficient ammonia for an accurate modeling assessment of visibility impacts within certain Colorado Class I areas.

Response: We disagree with the comment that large point/area sources are not included in estimates of background ammonia concentrations. While concentrations of ammonia of several hundred ppb can be observed near a cattle feedlot, these concentrations are not typical of regional background concentrations. Additionally, dispersion and vertical mixing occur in plumes of air transported downwind of large emissions sources, and the resulting dilution of ammonia results in lower concentrations as the plume is transported downwind of the source area. Therefore, ambient ammonia concentrations are generally greatly reduced downwind from the source. Moreover, ammonia has a short atmospheric lifetime of a few hours to a few days,[114] and removal of ammonia by deposition further reduces the concentration downwind of the source area. Therefore, it would be inappropriate to estimate background ammonia levels by measuring ammonia concentrations close to a large ammonia emissions source. Background concentrations of air pollutants are generally estimated using ambient monitoring data at background sites that are specifically selected such that there is no direct influence by large nearby point or areas sources.[115] Therefore, background monitoring data do explicitly include the contributions of emissions from large point and area sources by providing a direct measurement of the ambient concentration after transport, dilution and removal processes operate on the emissions from the source.

The commenter also cites modeling performed by the CDPHE and ammonia measurements made during the NFRAQS studies. As discussed in another response, CDPHE performed a CALPUFF model sensitivity study to evaluate the effect of background ammonia on model predicted nitrate concentrations, and found that the CALPUFF model was insensitive to variations in background concentrations greater than 10 ppb and became progressively more sensitive to background NH3 as it was reduced from 10 to 0 ppb.[116] The NFRAQS study reported measured ammonia concentrations in the Denver metropolitan area, and these measurements are not representative of background ammonia concentrations in Start Printed Page 5118rural and remote areas of central Colorado or western Wyoming.

Comment: The commenter states that Wyoming has conducted its regional haze SIP based on the modeling protocols and versions available at the time its regional haze SIP was completed. Because of this, there are limitations associated with the results obtained. However, in proposing its regional haze SIP, Wyoming has evaluated the model output with an understanding of the model's limitations. Wyoming then applied its judgment, as encouraged and required by EPA's guidelines and the CAA, which helped to mitigate the issues associated with models that over-predict the visibility improvement associated with BART controls being added. In contrast, EPA gives no consideration to the limitations of the models it uses. In the absence of using good judgment to deal with over-predictive results, it is critical that EPA use the most up-to-date and scientifically accurate models available.

We also received related comments that states have significant modeling discretion to which EPA failed to grant the proper deference. One commenter pointed out that the BART Guidelines recognize that states can make judgments regarding the use of modeling results due to the very real problems with CALPUFF, including its overestimation of visibility improvement. As EPA itself has stated, Wyoming should be free to make its own judgment about which modeling approaches are valid and appropriate. 70 FR 39123. Another commenter pointed to the statement that “we must permit States to take into account the degree of improvement in visibility that would result from imposition of BART on each individual source when deciding on particular controls.” 70 FR 39107, 39129. Another commenter stated that EPA failed to allow Wyoming to account for CALPUFF's overestimation of NOX impacts, and therefore, EPA is not affording Wyoming's BART decisions the proper deference when it comes to the modeling and applying the modeling results.

Response: As discussed elsewhere in this document in greater detail, Congress crafted the CAA to provide for states to take the lead for implementing plans, but balanced that decision by requiring EPA to approve the plans or prescribe a federal plan should the state plan be inadequate. Our action today is consistent with the statute. As also discussed elsewhere in this document, we agree that there are limitations in the original modeling performed by the state, and therefore, it was necessary to perform new modeling using more realistic background ammonia concentrations, default values, and updated model versions to provide a sound basis for evaluating BART source visibility impacts. Our revised modeling is consistent with the BART Guidelines and with visibility modeling guidance in the IWAQM Phase 2 report, and we believe that the revised modeling constitutes a sound basis for evaluating visibility impacts of BART sources and in fact is supportive of Wyoming's SIP with respect to sources where ammonia background makes a significant differences.

Comment: The commenter states that EPA should have used the most recent version of CALPUFF (Version 6.42) in Wyoming because it produces more realistic and accurate results and because Version 5.8 does not have the required settings to perform the new Method 8 visibility analysis.

Response: As described in response to another comment, we used CALPUFF version 5.8 because this is the approved regulatory version of the model, while CALPUFF version 6.42 has not been approved. CALPUFF version 5.8 does allow the option of using the Method 8 visibility analysis, and as described in our modeling protocol, we used Method 8 for our analysis. The availability of Method 8 in CALPUFF version 5.8 was one of the reasons that we determined it was important to perform new modeling using the current regulatory version of the model, rather than continuing to rely on the original protocol adopted by the State in 2006, as we had done in our previous proposal.

Comment: EPA used a different background ammonia number for modeling than it requires of the states, and ignored current science on background ammonia. Regional haze modeling, and the resulting predicted visibility improvement, is greatly influenced by the background ammonia number used in the model. EPA improperly used a constant 2 ppb background ammonia number for the Wyoming BART modeling. EPA has not provided any scientific proof showing the constant 2 ppb ammonia number is appropriate for Wyoming. The 2 ppb ammonia value overestimates visibility improvement, contrary to the approach used by Wyoming Land Use, IWAQM Guidance, WRAP protocols, and elsewhere.

Commenter suggests that the WRAP recommended the use of 1 ppb of ammonia year round for states in the region to account for seasonal variability. EPA has required states to use 1 ppb of background ammonia when conducting regional haze modeling. 76 FR 52434 (New Mexico criticized for not using 1 ppb background ammonia). At a minimum EPA should follow its own guidelines and use 1 ppb of background ammonia when conducting CALPUFF unit-specific modeling.

However, the “best” science requires the use of “variable ammonia” background numbers. IWAQM recommends ammonia background numbers of 0.5 ppb for forest, 1 ppb for dry/arid lands, and 10 ppb for agriculture/grassland. Given its geographic location and elevation levels, Wyoming undergoes seasonal swings of dry-hot summers and snow covered ground in the winter. Therefore, the use of a single ammonia concentration for the entire year in a state where the land use and land cover changes significantly between seasons results in overestimation of visibility improvements. This is particularly true in winter when agricultural activity is minimal and meteorological conditions make visibility calculations particularly sensitive to ambient ammonia concentrations. EPA has approved the use of variable gaseous ammonia concentrations before, including the “Addendum to Modeling Protocol for the Proposed Desert Rock Generating Station” and should have used them when conducting the CALPUFF modeling for Wyoming.

Sensitivity tests on ambient ammonia concentrations were performed by the CDPHE for an area in northwest Colorado. The analysis demonstrated that visibility calculations performed at Mount Zirkel Wilderness Area in northwest Colorado had limited impact when ambient ammonia concentrations were reduced from 100 to 1 ppb, but there was a significant reduction in visibility impacts when concentrations were further reduced to 0.1 ppb.

Given the evidence presented above, the use of the monthly varying ammonia would provide accurate estimates of visibility impacts from the PacifiCorp regional haze units. EPA's failure to use variable background ammonia in its modeling is arbitrary and capricious.

Response: We agree that the 2 ppb constant background ammonia concentration is inconsistent with the IWAQM Phase 2 report default values and monitored data. This value was adopted by the State in 2006 before more reliable ammonia and particulate ammonium measurements were available in Wyoming. As described in this action, we modeled using seasonally varying background ammonia concentrations in western Wyoming based on 5 years of Start Printed Page 5119monitoring data,[117] and we also modeled using the IWAQM default value of 0.5 ppb for forests. In eastern Wyoming we adopted a constant 1 ppb ammonia concentration based on the IWAQM guidance. We used an ammonia limiting correction for BART sources with multiple units throughout the State. While robust, long term monitoring data of ammonia and particulate ammonium are not available in eastern Wyoming, the BART sources in eastern Wyoming, and the South Dakota Class I areas where they contribute the greatest visibility impairment, are located closer to areas of Nebraska and the Dakotas which have large agricultural sources of ammonia emissions. Moreover, the IMPROVE monitoring at the South Dakota Class I areas show much higher winter concentrations of ammonium nitrate than do Class I areas in western Wyoming. Therefore, we believe it is appropriate to adopt higher background ammonia concentrations in eastern Wyoming than in western Wyoming, and we used a constant 1 ppb ammonia concentration in eastern Wyoming, consistent with the IWAQM guidance [118] for arid lands and also consistent with the WRAP Protocol.[119]

Comment: EPA made a modeling error in Wyoming when it used CALPOST version 5 with Method 6. FLMs recommended in 2000 the use of Method 6 to determine visibility impacts from BART eligible sources. However, for any recent PSD application and BART modeling since 2010, EPA has requested that Method 8 be used for determining impacts on visibility at nearby Class I areas.

The previously preferred Method 6 simply computes background light extinction using monthly average relative humidity adjustment factors particular to each Class I area applied to background and modeled sulfate and nitrate. Six years after the development of Method 6 in 1999, EPA released enhancements to the background light extinction equations, which use the revised IMPROVE variable extinction efficiency formulation. These enhancements take into account the fact that sulfates, nitrates and organics and other types of particles have different light extinction coefficients. Also, the background concentrations at each Class I area have been updated by EPA to reflect natural background visibility condition estimates for each Class I area for each type of particle. Additionally, relative humidity adjustment factors have been tailored separately for small particles, large particles, and to account for sea salt background concentrations.

These new enhancements to the calculation method, called Method 8, greatly improve the accuracy of the estimated visibility impact. Method 8 was added to CALPOST in 2008 and was adopted as the preferred option for determining impacts on visibility by the FLMs in their “Federal Land Managers Air Quality Related Values Work Group Guidance Document” in 2010 (FLAG 2010). The applicable background concentrations and relative humidity adjustment factors using Method 8 for each Class I area are identified in the FLAG 2010 document.

Despite the update to Method 8 in 2008 and the stated preference by the FLMs in 2010 to use Method 8, EPA conducted the Wyoming BART modeling in 2012 using the long outdated and scientifically inferior Method 6. EPA's use of Method 6, and not Method 8, is arbitrary and capricious.

Response: We agree that it is preferable to use Method 8 rather than Method 6 for evaluating visibility impacts based on the recommendation of the FLMs in FLAG 2010.[120] The older CALPUFF version 5.711 that was adopted in the State's original modeling protocol in 2006 did not have the option of using Method 8. In our previous modeling we adopted the State's original protocol to maintain a consistent basis of comparison with the State's modeling results. In this final action, we adopted an updated modeling protocol using the current regulatory version of the model, which allows the use of Method 8, and we used Method 8 for the analysis of visibility impacts.

Comment: We received numerous comments that EPA incorrectly used the maximum annual visibility impacts occurring during any given year of the 2001-2003 baseline period over which the Wyoming visibility models are run. Commenters asserted that standard practice has been, and continues to be, to average the results over the three year period as the three year average is a more robust value than the single year value used by EPA, and thus EPA should use longer term data. One commenter pointed out that consistent with the principle of using longer-term averages, baseline visibility conditions under the RHR are determined by taking the average degree of visibility impairment for the most and least impaired days for each of calendar years 2000 through 2004, and averaging the five annual values.

Response: We disagree with this comment. In our review of the CALPUFF modeling results presented by Wyoming, we cited the change in the maximum 98th percentile impact over the modeled three year meteorological period (2001-2003). As the 98th percentile value is intended to reflect the 8th high value in any year, it already eliminates seven days per year from consideration in order to account for short-term events, unusual meteorological conditions, and any over-prediction bias in the model. We also note that our approach is consistent with the method used by Wyoming in identifying subject-to-BART sources, where a source is exempt from BART only if the modeled 98th percentile change is less than 0.5 deciview at all Class I areas for each year modeled.[121] That is, whether a source is subject to BART is dependent on the maximum 98th percentile over the three year modeled meteorological period, not the average across the three year period. We find that it is reasonable to use the same approach when considering the visibility improvements associated with control options. Finally, we note that this approach is consistent with our consideration of visibility improvement in other actions, such as our FIP for Montana.

Comment: EPA's use of the maximum values in its BART NOX determinations for its regional haze FIP causes inflated visibility impacts and over-estimated improvements being used. For example, if EPA were to run its approved models, used its approved ammonia values for the western states, and used the average visibility impact over the three years rather than a maximum impact for a single year, the incremental visibility impact between installing LNB technology and SNCR at Wyodak and Dave Johnston Unit 3 drops to just 0.09 deciview. Instead, EPA has used an improper evaluation to create an inflated visibility improvement of 0.15 and 0.17 delta deciview to justify the installation of the SNCR. As a result, EPAs' BART NOX evaluations are Start Printed Page 5120invalid. The modeling results reported by Mr. Paine using the actual BART limits include values for each of the meteorological years 2001, 2002 and 2003, as well as the average values for the three years. All of the values signify a negligible visibility improvement from SNCR.

Response: We have addressed each aspect of this comment in separate responses to comments. In our previous proposal we performed CALPUFF model simulations consistent with the approach specified in the Wyoming protocol, but in this action we adopted updated model versions and used lower ammonia background concentrations that are consistent with monitoring data and IWAQM Phase 2 report. Regardless, as discussed in section IV, in consideration of our revised cost of compliance and visibility impact analyses, and of the remaining BART factors, we have changed our final NOX BART determinations for both of the units in question. We are no longer requiring SNCR for either Wyodak or Dave Johnston Unit 4.

Comment: EPA's use of the cumulative deciview improvement from several Class I areas overestimates the visibility improvement which may reasonably be anticipated because visibility impacts from a BART source may occur on different days at each Class I area. Adding the numbers in Tables 47, 54, and 56 of EPA's proposed regional haze FIP leads to the impression that a perceptible visibility improvement will occur, when in reality none of the modeled visibility improvements would be perceptible to the human eye.

Response: We disagree with this comment. In evaluating the visibility improvement associated with various control options, EPA interprets the CAA to require consideration of visibility improvement at all impacted Class I areas. Consideration of improvement at multiple Class I areas, as opposed to just benefits at the most impacted Class I area, has often been described as “cumulative visibility improvement.” Despite this terminology, however, an analysis of cumulative visibility improvement does not necessarily require that the deciview improvement at each area be summed together. While states or EPA are free to take such a quantitative approach, they are also free to use a more qualitative approach. Here, we chose to rely primarily on the visibility improvement at the most impacted Class I area, while also considering the number of additional Class I areas that would see improvement, as well as the level of improvement at each area. We did not expressly rely on a summation of visibility benefits across Class I areas, as we have done in other regional haze actions, although, as the commenter points out, this metric was included in some tables. Contrary to the commenter's assertion, however, a summation of visibility benefits is not intended to suggest that individually imperceptible levels of improvement are somehow perceptible, but rather to provide a single metric that can simultaneously capture both the number of Class I areas affected and the magnitude of improvement at those areas for comparison purposes. Moreover, we note that visibility improvement does not need to be perceptible to be deemed significant for BART purposes. We have responded in more detail to concerns regarding perceptibility elsewhere in this document.

Comment: The commenter states that EPA's cumulative visibility analyses ignore the discretion given to states in 70 FR 39107; Id. at 39123 (emphasis added); see also 77 FR 24768, 24774 (Apr. 25, 2012)

Response: As stated above, EPA primarily relied on the benefits at the area with the greatest visibility improvement from controls, but we also considered impacts and benefits at nearby areas included in the modeling analysis. The consideration of visibility benefits over multiple Class I areas is a useful metric that can further inform a BART determination.

Comment: The BART rule provides no support for EPA's “summation of cumulative impacts” approach. Rather, the BART rule makes clear that the initial focus is expected to be on the “nearest Class I area” to the facility in question. 70 FR 39104, 39162 (Separt 6, 2005). The BART rules indicate that it is appropriate to take account impacts at not only the nearest Class I area but also impacts at other nearby Class I areas, not for the purposing of summing impacts at all of those areas, but rather for the purpose of “determin(ing) whether effects at those (other) areas may be greater than at the nearest Class I area.” Id. The BART rule states: “If the highest modeled effects are observed at the nearest Class I area, you may choose not to analyze the other Class I areas any further as additional analyses might be unwarranted.” Id.

Response: See our response to comments above. In addition, the BART Guidelines provide that states, or EPA in lieu of the state, have discretion on how to assess visibility impacts.

Comment: The commenter states that the BART rule does not preclude a state from taking into account, as part of a BART assessment for a given facility, visibility impacts projected to occur in two or more Class I areas that are attributable to that facility's emissions. However, nothing in the rules requires such an analysis, and such analyses are deceptive when used in a cumulative fashion. EPA did not have the authority to disapprove Wyoming's visibility improvement analyses on the grounds that EPA prefers a different approach than the lawful and permissible approach taken by Wyoming. See Train v. Natural Res. Def. Council, Inc., 421 U.S. 60, 79 (1975).

Response: See our response to comments above and elsewhere in this document (e.g., Legal Issue section) regarding EPA's oversight authority.

Comment: EPA has improperly failed to account for the very few number of days of visibility impacts or the seasonal timing of when those few impacts occur.

Response: EPA recognizes that the BART Guidelines allow states to consider the timing of impacts in addition to other factors related to visibility impairment. However, states are not required to do so, and EPA is not required to substitute a source's desired exercise of discretion for that of the states. Furthermore, when promulgating a FIP, EPA stands in the shoes of the state. In that capacity, EPA is not required to consider the seasonality of impacts and has chosen not to do so here. Taking into account visitation contradicts the goal of the regional haze rule of improving visibility on the 20 percent best and worst days. Indeed, EPA believes that the experiences of visitors who come to Class I areas during periods other than the peak visitation season are important and should not be discounted.

Comment: A review of the unit-specific CALPUFF EPA modeling results developed for the Mount Zirkel Wilderness Area provides a vivid example of the over-estimation of visibility improvement that EPA is relying on to justify the installation of hundreds of millions of dollars in additional controls. The unit specific CALPUFF modeling would indicate that adding SCR to these units would improve visibility in Mount Zirkel by over seven deciviews.

However, the monitored data from 2001-2003 at Mount Zirkel tells a completely different story. This is the same time period used in the CALPUFF models to develop the deciview impacts for each Wyoming BART-eligible unit and to project the visibility improvements associated with the addition of control devices.

Looking at the three-year average monitored results, and assuming that Start Printed Page 5121the nitrates associated with the emissions from all sources (not just the BART-eligible EGUs) are completely eliminated, only a 0.94 deciview improvement would be expected. EPA attempts to justify over a billion dollars in controls at eight PacifiCorp units by assuming more than seven deciviews of improvement could be obtained from these eight units when the actual monitored data indicates that only a 0.94 deciview improvement would be possible if all nitrate was removed from all sources. In essence, EPA's regional haze FIP fails to recognize that, given the monitored nitrate impacts, the modeled visibility impacts are obviously grossly exaggerated.

Response: We agree with some of these concerns—that the original modeling performed by the State and EPA used a high background ammonia concentration and did not correct for ammonia limiting conditions. This particularly affected the model results in the western part of Wyoming and Class I areas such as Bridger Wilderness Area.

We have addressed this concern by adopting a new modeling protocol that makes several improvements in the model results, including the use of the current regulatory version of the model, the use of Method 8 to assess visibility impacts, the use of lower background ammonia concentrations, and ammonia limiting correction for BART sources with multiple units. We note that the model overprediction in our re-proposed modeling analyses occurred at Class I areas affected by BART sources in western Wyoming, in the region in which monitoring data showed strong seasonal variability in ammonia concentrations. In contrast, modeled nitrate impacts from BART sources in eastern Wyoming were significantly lower than observed nitrate concentrations at IMPROVE sites at Wind Cave and Badlands in western South Dakota.

There are several factors that make it challenging to directly compare CALPUFF results to measured concentrations at IMPROVE monitoring sites at Class I areas. Most significantly, the monitor operates every third day, while the model predicts concentration each day. Moreover, modeled visibility impacts from multiple BART sources cannot be summed and directly compared to measured data as all BART sources are unlikely to have their largest impacts on the same Class I area on the same day. Additionally, the model 98th percentile impact should be compared to the maximum observed monitoring data because the highest 2% of model impacts are discarded to address concerns that the model can overpredict visibility impacts.

Comment: The commenter cites a study by Terhorst and Berkman which compared CALPUFF model predicted impacts of the Mojave Power Station at the Grand Canyon to observed impacts after the facility was closed in 2005. The study concluded that there was virtually no evidence that the (Mojave) closure improved visibility in the Grand Canyon, and the commenter cites this conclusion as evidence of the unreliability of the CALPUFF model.

Response: We disagree that the Mohave Power Plant study raises questions about CALPUFF's reliability. The Nevada Division of Environmental Protection performed CALPUFF modeling to estimate the contribution of the Mohave Power Plant to visibility impairment at Grand Canyon National Park. Consistent with the BART Guidelines, the State used the CALPUFF model to evaluate the Mohave Power Plant contribution to visibility impairment relative to natural visibility conditions. Subsequently, after the Mohave Power Plant ceased operating, Terhorst and Berkman analyzed changes in monitored sulfate concentrations at the Grand Canyon and calculated the visibility impacts of those changes relative to current degraded visibility conditions. Terhorst and Berkman incorrectly concluded that the State's previous CALPUFF modeling overpredicted the Mohave Power Plant visibility impacts because Terhorst and Berkman failed to compare their results to natural visibility conditions. EPA considered and rejected comments on the proposed BART Guidelines that visibility impacts should be evaluated to relative to current degraded visibility conditions and concluded that “[u]sing existing conditions as the baseline for single source visibility impact determinations would create the following paradox: the dirtier the existing air, the less likely it would be that any control is required.” (70 FR 39124). Because Terhorst and Berkman failed to compare observed changes in sulfate concentration to natural visibility conditions, their analysis does not support the commenter's statement that CALPUFF is unreliable. This flaw in their analysis has also been recognized in a paper that responded to their analysis.[122] Finally, as presented above, the use of the CALPUFF model for regional haze is a settled manner for which the time for judicial review has passed.

Comment: The commenter states that EPA's own studies document that CALPUFF overstates results and cites a May 2012 EPA sponsored study of CALPUFF that found “the current and past CALPUFF model performance evaluations were consistent with CALPUFF tending to overestimate the plume maximum concentrations and underestimate plume horizontal dispersion.”

Response: In the BART Guidance, EPA recognized concerns that CALPUFF can overpredict visibility impacts in some cases, and therefore, as explained above, adopted the use of the 98th percentile modeled impact, rather than the maximum modeled impact, to address this concern.

Comment: EPA appears to take contrary positions in Oklahoma, where it modeled all visibility impairing pollutants together, and Wyoming, where EPA said that, based on the State's modeling, EPA “could not ascertain what the visibility improvement would be from an individual NOX or PM control option.”

Response: It appears that the commenter has confused (1) whether all pollutants were modeled together; and (2) whether all emission reductions were modeled together. All pollutants were modeled together both in modeling performed by Wyoming and by EPA for BART sources in Wyoming and Oklahoma, consistent with IWAQM Phase 2 report recommendations and with the State of Wyoming modeling protocol. The additional modeling performed by EPA was designed to evaluate visibility improvements from certain emissions reduction technologies; specifically, to compare the incremental benefits of SCR and SNCR. Each of these model simulations by EPA also included all other visibility impairing pollutants, so the approach used by EPA in Wyoming and Oklahoma is consistent.

Comment: EPA found that SCR provided only a 0.36 delta deciview incremental visibility improvement for Dave Johnston Unit 3, using EPA modeling, with an incremental cost of $7,163.00 per ton. 78 FR 34777-34778. EPA failed to justify in its proposed rule how a 0.36 delta deciview improvement, or approximately one-third that humanly detectible, justifies the tremendous cost of SCR. Likewise, EPA found that installing SNCR at Dave Johnston Unit 4 results in an incremental 0.11 delta deciview improvement over Wyoming's BART determination at an incremental cost of Start Printed Page 5122$4,655. 78 FR 34781-34782. The alleged incremental visibility benefit of installing SNCR at Wyodak is 0.12 delta deciview at an incremental cost of $3,725 per ton. 78 FR 34784-34785. EPA provides no justification for requiring such tremendous costs for such an inconsequential visibility improvement that likely falls within CALPUFF's margin of error. EPA's modeling approaches are inconsistent because EPA has determined in other states that visibility improvements greater than those used to justify SNCR at Wyodak are too small or inconsequential to justify additional pollution controls. See 77 FR 24794 (0.27 deciview improvement termed “small” and did not justify additional pollution controls in New York); 77 FR 11879, 11891 (0.043 to 0.16 delta deciview improvements considered “very small additional visibility improvements” that did not justify NOX controls in Mississippi); 77 FR 18052, 18066 (agreeing with Colorado's determination that “low visibility improvement (under 0.2 delta deciview)” did not justify SCR for Comanche units)). Tellingly, the “low visibility improvements” that Colorado found at the Comanche units not to justify post-combustion NOX controls, as agreed to by EPA, were 0.17 and 0.14 delta deciview. 77 FR 18066. In Montana, where EPA issued a regional haze FIP directly, it found that a 0.18 delta deciview improvement to be a “low visibility improvement” that “did not justify proposing additional controls” for SO2 on the source. 77 FR 23988, 24012. Here, EPA's actions requiring additional NOX controls based on little to no additional visibility improvement are arbitrary and capricious, especially when EPA did not require additional NOX controls in other states based on similar visibility improvements. This is particularly true in Montana where EPA had direct responsibility for the regional haze program.

Response: We disagree that visibility improvements at Dave Johnston Unit 3 and Wyodak are “too small or inconsequential to justify additional pollution controls.” While the visibility benefits at these units are less than what is generally considered perceptible (1.0 deciview), they are not so low as to preclude selection of the associated controls without any consideration of the remaining BART factors. The BART Guidelines are clear that states should consider visibility impacts that are less than perceptible: “Even though the visibility improvement from an individual source may not be perceptible, it should still be considered in setting BART because the contribution to haze may be significant relative to other source contributions in the Class I area. Thus, we disagree that the degree of improvement should be contingent upon perceptibility. Failing to consider less-than-perceptible contributions to visibility impairment would ignore the CAA's intent to have BART requirements apply to sources that contribute to, as well as cause, such impairment.” 70 FR 39129. When the visibility improvements are considered, we continue to find that this level of improvement, when considered along with the other statutory factors, justifies the selected BART controls.

Finally, regarding commenter's assertions that we are being inconsistent, because the commenter is only specific about visibility improvement, it is not possible for EPA to address in this response any specific concerns. As articulated in our proposed rulemakings and further explained in our responses to other comments, EPA's partial approval and partial disapproval of the Wyoming Regional Haze SIP is consistent with the CAA, the RHR, BART Rule, and EPA guidance. Our determinations considered all five factors, not just visibility improvement.

Comment: Although it is true that Wyoming did not model the visibility impact of SNCR, that fact is no justification for disapproving Wyoming's BART. Nothing in the BART Guidelines or Wyoming's BART Modeling Protocol demands modeling of SNCR, and EPA points to nothing in either document that requires modeling of SNCR.

Response: We disagree with this comment. The BART selection process requires a comparison between all technically feasible control options, not the evaluation of individual control technologies in isolation. While the BART Guidelines do not specify the order in which control options must be evaluated (e.g., beginning with the most stringent or beginning with least stringent control), they do specify that the CAA factors must be considered for all options: “In the final guidelines, we have decided that States should retain the discretion to evaluate control options in whatever order they choose, so long as the State explains its analysis of the CAA factors.” 70 FR 39130. The only exceptions are “. . . if you find that a BART source has controls already in place which are the most stringent controls available . . .”, or “. . . . if a source commits to a BART determination that consists of the most stringent controls available . . .” 70 FR 39165. In these situations, it is not necessary to complete an analysis of all five BART factors. Therefore, because neither of these criteria was met, the State was required to perform an analysis of all five BART factors for all technically feasible control options. As such, the State's failure to consider the visibility impacts of SNCR did in fact serve as appropriate grounds for EPA's disapproval of Wyoming's BART determination.

Comment: EPA modeling shows no significant visibility improvement from SNCR and without a significant improvement there can be no justification for disapproving the State's BART for Laramie River Station, and, to the contrary, EPA's modeling supports Wyoming's rejection of SNCR and choice of LNB/OFA because SNCR provides negligible visibility improvement.

Response: We have required new LNBs with OFA and SCR for the Laramie River Station, not new LNBs with OFA and SNCR, which is the control option addressed by the commenter. Accordingly, we do not find that the comment is relevant to our action. Our revised modeling shows that the visibility benefit of new LNBs with OFA and SCR for Laramie River Units 1-3 is 0.57 deciviews, 0.53 deciviews, and 0.52 deciviews, respectively. We continue to find that the visibility benefit, when taking into consideration the remaining BART factors, justifies installation of new LNBs with OFA and SCR.

Comment: Basin Electric submitted results based on more accurate modeling than EPA, which show that actual visibility improvement from SNCR would be substantially lower than assumed by EPA. There is no justification for disapproving Wyoming's BART based on a modeled visibility improvement that is such a small fraction of what is humanly perceptible.

Response: As described in response to other comments, we agree that the original modeling protocol adopted by the State was inconsistent with the BART Guidelines, IWAQM Phase 2 report and newly available ambient monitoring data, and in our revised modeling we adopted several of the changes recommended by this commenter, including the use of lower background ammonia concentration, a correction for ammonia limiting conditions for multiple units located at a single BART source, and the use of Method 8 for the evaluation of visibility impairment. However, even using these model options, we still found significant visibility impacts for SCR control at Start Printed Page 5123Laramie River. Our results are generally consistent with the modeling results submitted by the commenter which also show significant impacts. As described in another response, while the visibility benefits at each of these units individually are less than what is generally considered perceptible (1 deciview), they are not so low as to preclude selection of the associated controls without any consideration of the remaining BART factors. The BART Guidelines are clear that States should consider visibility impacts that are less than perceptible because these sources may still contribute to cumulative visibility impairment.

Comment: EPA did not assert a failure to model NOX impacts separately was a flaw in the Laramie River Station modeling, although EPA did identify this as a flaw in PacifiCorp modeling.

Response: We agree that the State evaluated NOX impacts separately for the control technologies that the State included in its modeling, however, the State did not evaluate SNCR. The other deficiencies in the State's visibility analysis, including the failure to consider the visibility impacts of SNCR, were appropriate grounds to disapprove the State's BART determination.

Comment: The Wyoming modeling did in fact isolate the impact on visibility for NOX control alternatives. Wyoming held SO2 and PM emissions constant at baseline levels while modeling varying NOX emission rates for each of the NOX control options.

Response: We agree with this comment. Nonetheless, as stated above, the other deficiencies in the State's visibility analysis, which were inconsistent with the BART Guidelines, including the failure to consider the visibility impacts of SNCR, were appropriate grounds to disapprove the State's BART determination.

Comment: EPA claims that Wyoming modeled the wrong emission rates. EPA notes that in its cost analysis it calculated a new removal efficiency for NOX control options that was different than the removal efficiency calculated by Wyoming, and claims that visibility modeling should have used the EPA efficiencies. However, EPA does not explain how modeling with the different removal efficiencies conflicts with the BART Guidelines or the CAA. As to SNCR, EPA argues that the State assumed a higher removal efficiency and thus, paradoxically, modeling with the State's removal assumption would yield greater visibility improvement than modeling with EPA's values. No such modeling was done, however. The State did no modeling for SNCR, so the State's removal efficiency was never modeled. It is an enigma how EPA can disagree with modeling with the different SNCR removal values when such modeling was never performed.

Response: We disagree with this comment. The BART Guidelines are clear on how removal efficiencies should be considered in the visibility evaluation: “Post-control emission rates are calculated as a percentage of pre-control emission rates. For example, if the 24-hr pre-control emission rate is 100 lb/hr of SO2, then the post control rate is 5 lb/hr if the control efficiency being evaluated is 95 percent.” 70 FR 39170. Therefore, because the control efficiencies assumed by the State differed from those found by the EPA, they affected the calculation of post-control emission rates for modeling purposes (and thereby the consideration of visibility impacts).

In regard to SNCR, as conceded by the commenter, the State did not provide the visibility impacts associated with the control option. As discussed elsewhere, failure to assess the visibility impacts of a technically feasible control option is in clear conflict with the requirements of the CAA and BART Guidelines. This failure alone, regardless of the control efficiency assumed for SNCR, was sufficient grounds for us to reject the State's BART determination. Moreover, the incorrect removal efficiency for SNCR assumed by the State adversely affected their analysis of cost of compliance, another statutorily required BART factor.

To put it simply, the State failed in the first instance by not considering the visibility improvement of SNCR as required by the CAA and BART Guidelines. When EPA corrected this deficiency by performing the modeling ourselves, it was necessary for us to correct the removal efficiency of SNCR (as discussed in response to other comments).

Comment: The State assumed that SCR would reduce NOX emissions from 0.21 lb/MMBtu to 0.07 lb/MMBtu—a reduction of 0.14 lb/MMBtu. EPA assumes SCR would reduce NOX emissions from 0.19 lb/MMBtu to 0.05 lb/MMBtu, a reduction of the same 0.14 lb/MMBtu. All other things being held constant, the 0.14 lb/MMBtu reduction will in both cases yield an identical reduction in the visibility impairing concentration of nitrate particulate in a Class I area.

Response: We disagree with this comment. The CALPUFF model simulations estimate the visibility impairment attributed to the emissions in each control scenario, not the relative reduction in different control scenarios. Therefore, an emissions rate of 0.07 lb/MMBtu will have 40% greater total emissions and a larger visibility impact than an emissions rate of 0.05 lb/MMBtu.

Comment: EPA argues that Wyoming should have used a baseline of the maximum 24-hour average NOX emission rate during the baseline years of 2001-2003, and instead used an annual average baseline rate. The BART Guidelines do not mandate the use of the 24-hour maximum but, rather, “recommend that the State use the highest 24-hour average actual emission rate” and that the states should have flexibility when evaluating the fifth statutory factor. The BART Guidelines by their express terms authorize states to use baseline emissions other than the 24-hour maximum rate. Use of the 24-hour maximum baseline is not mandatory, and not using that baseline is not a failure to comply with any requirement in the Guidelines. EPA itself used annual average pre-control and post-control emission rates to model visibility impacts in its Nevada FIP rulemaking.

Response: We disagree with this comment. As stated in the preamble to the BART Guidelines, “the emissions estimates used in the [visibility] models are intended to reflect steady-state operating conditions during periods of high capacity utilization.” 70 FR 39120. As such, the BART Guidelines recommend excluding emissions during periods of start-up, shutdown, or malfunction or estimating visibility impacts based on a source's allowable emissions as this could inflate the visibility impacts of a source. Rather, for sources such as power plants where States have information on a source's daily emissions, the BART Guidelines explains that an emission rate based on a source's maximum actual emissions over a 24-hour period is an appropriate gauge of a source's potential impact as it ensures that peak emission conditions are reflected but would be unlikely to lead to an overestimation of a source's potential impacts. Id. The BART Guidelines state that in developing a modeling protocol, States should “[u]se the 24-hour average actual emission rate from the highest emitting day of the meteorological period modeled (for the pre-control scenario).” Id. and 70 FR 39170.

Wyoming did not do this. Instead, in assessing the improvement in visibility associated with the use of controls in its BART determinations, Wyoming used the visibility modeling performed by PacifiCorp and Basin Electric for their facilities. Although these companies used very different approaches to estimating the baseline emission rate—Start Printed Page 5124neither of which used the 24-hour average actual emission rate—the State accepted the visibility modeling done by both and submitted the results in the Wyoming SIP. Even if the commenter were correct that the approach in the BART Guidelines is only recommended, the commenter has not provided any explanation as to why the disparate approaches used in the Wyoming SIP were appropriate for estimating the degree of visibility improvement associated with controls. Wyoming similarly provided no explanation as to why the varying approaches adopted by Basin Electric and PacifiCorp were appropriate for assessing visibility improvement. Moreover, the commenter has not established that the baseline emission rates used by Wyoming would accurately reflect visibility impacts associated with steady-state operating conditions during periods of high capacity utilization. Obviously, baseline emission rates reflecting periods of relatively lower capacity utilization would tend to underestimate peak visibility impacts. Consider for example the baseline emission rate used for Laramie River Unit 1. There, the State used a daily emission rate equating to 6,320 tons per year,[123] while, based on actual emissions data, the EPA used a daily rate equating to 8,786 tons per year.[124] Thus, the rate used by the State reflects a period of considerably lower capacity utilization that would therefore tend to underestimate peak impacts.

Regarding the emissions rates used in the Nevada regional haze SIP, the State did not use the 24-hour average of actual emissions from the highest emitting day in its BART determination for Reid Gardner Generating Station. 77 FR 50936, 50944 (Aug. 23, 2012). As part of its review of the Nevada SIP, EPA performed new visibility modeling. In that modeling exercise, EPA used Nevada's emission rates based on annual averages. Please refer to a related comment and response in the final action for that rule for a full discussion. See 77 FR 50944). Following our review of comments in that rulemaking—including comments that we should have used the Guidelines maximum 24-hour average of emissions in our visibility modeling—we scaled our estimates of the visibility impacts of controls based on the source's emissions using the Guidelines maximum 24-hour average. We took these scaled visibility impacts into account in our final action. Id. at 50945.

Comment: EPA did not use the 24-hour maximum rate for the modeling it performed in 2012. As noted in Section VIII.C, it used the same baseline emission rates used by the State. EPA-R08-OAR-2012-0026-0037. EPA did not find the State's approach was a violation of the BART Guidelines or was a reason to disapprove the State's modeling or BART determination. Having sanctioned the use of a different baseline then, EPA may not now claim it violates the BART Guidelines or a ground for disapproving the State's modeling or BART determination.

Response: We agree that we did not use the 24-hour maximum [actual] emission rates for modeling purposes in our original proposed rule published in 2012. However, we did not finalize that rule, at least in part, for the very reason that the baseline emission rates calculated by Wyoming, and subsequently used by EPA in the 2012 proposed rule, were inconsistent with the BART Guidelines. As we never finalized the original rule, we disagree with the commenter's suggestion that we somehow sanctioned Wyoming's approach. A proposed rule does not represent final agency action.

Comment: The maximum improvement modeled by EPA that would be achieved at any Class I area by adding SCR to the existing new LNB plus OFA is 0.5 delta deciview. This is below the 1.0 delta deciview level often cited as the lowest level of change that is humanly perceptible. For EPA to propose disapproval of the State's BART based on an imperceptible improvement is to propose disapproval based on a nonmaterial factor.

Response: We disagree that the visibility improvements for Laramie River or Jim Bridger are de minimis or too small to just justify the expense of requiring controls. As discussed in response to another comment, the BART Guidelines are clear that it is not necessary for the visibility improvement of a particular control option to be above the perceptible threshold. The regional haze program is premised on the fact that numerous sources are contributing to visibility impairment and numerous sources will need to reduce emissions in order to improve visibility. We continue to find that this level of improvement, when considered along with the other statutory factors, justifies the selected BART controls.

Comment: The commenter states that EPA's modeled visibility improvement overstates the improvement that would actually be achieved. The commenter submitted new modeling showing that the visibility improvement from further reductions of NOX emissions would be much smaller than that predicted by EPA. AECOM corrected four of the flaws in EPA's modeling and re-ran CALPUFF. The commenter submitted refined modeling with four adjustments: 1. The use of seasonal background ammonia concentrations; 2. Modeling of all units together with correction for ammonia limited conditions; 3. Use of a post-control emission rate of 0.07 lb/MMBtu, consistent with EPA's proposed emission limit; and 4. The use of CALPOST Method 8. AECOM's revised modeling was identical to EPA's in all other respects. The refined modeling predicted that the incremental visibility benefit of SCR at each of Laramie River Units 1, 2 and 3 would range between 0.20-0.24 delta deciview at either Badlands or Wind Cave National Park. The actual visibility improvement of SCR would be even less than predicted by the refined modeling because CALPUFF is known to substantially overstate nitrate haze.

Response: We agree that our proposed modeling was inconsistent with the BART Guidelines, IWAQM Phase 2 report and monitored data, and in our revised final modeling we adopted several of the changes recommended by this commenter, including the use of lower background ammonia concentration, a correction for ammonia limiting conditions for multiple units located at a single BART source, and the use of Method 8 for the evaluation of visibility impairment. However, even using these less conservative model options, we still found significant visibility impacts for SCR control at the Basin Electric Laramie River EGUs. We did not use the seasonal background ammonia concentration proposed by the commenter because we did not have sufficient ambient monitoring data to determine the seasonal background concentrations in eastern Wyoming.

Comment: The commenter states that nitrate haze occurs primarily in the winter when few visitors are present in Class I areas. During the peak summer visitation period, the impact of wildfires would overwhelm any marginal visibility improvement that might be achieved by SCR. The commenter cites an EPA report that stated “[A] all else being equal, impairment from anthropogenic sources is considerably more objectionable during times of the year with greatest visitor attendance (e.g., summer). Visibility objectives might, therefore, be stated in terms of acceptable frequency distributions of visibility (e.g., contrast) over the course of a year.” Source: Report to Congress under CAA Section 169A(a)(3). The Start Printed Page 5125commenter states that these factors further support Wyoming's decision to reject SCR and SNCR as BART. These technologies would not improve visibility on the worst haze days because Laramie River doesn't contribute to haze on those days, and any slight visibility improvement would occur in the winter season when few visitors enter the Class I areas. Wyoming's decision to reject SCR as BART is therefore reasonable and complies with the CAA.

Response: As discussed in response to another comment, EPA agrees that nitrate impacts are more dominant in the winter. Nonetheless, daily nitrate impacts from April through October are not trivial. EPA also agrees that the BART Guidelines allow states to consider the timing of impacts in addition to other factors related to visibility impairment. However, states are not required to do so, and EPA is not required to substitute a source's desired exercise of discretion for that of the states. Furthermore, when promulgating a FIP, EPA stands in the shoes of the state. In that capacity, EPA is not required to consider the seasonality of impacts and has chosen not to do so here. Taking into account visitation contradicts the goal of the regional haze rule of improving visibility on the 20 percent best and worst days. Indeed, EPA believes that the experiences of visitors who come to Class I areas during periods other than the peak visitation season are important and should not be discounted.

Comment: We received comments that our FIP was not warranted because the cause of visibility impairment during the times of peak visitation was wildfires and thus does not justify the control of NOX from stationary sources.

Response: See response above.

Comment: Ammonia levels at the altitude of the plume would be lower than the reported surface level ammonia concentrations, so less ammonia would be available to form visibility-impairing nitrate.

Response: We disagree that there is compelling evidence that background ammonia levels are significantly different at the altitude of the plume compared to the surface. While there are limited studies showing vertical gradients of ammonia in the troposphere,[125] these studies do not show a strong gradient within the planetary boundary layer where the plume is typically located. Moreover, as discussed in the response to another comment, it is necessary to evaluate the combined concentrations of gas ammonia and particulate ammonium to estimate the background ammonia level, so vertical gradients in measured ammonia alone are not sufficient to specify the vertical gradient in background ammonia. It is possible that decreasing temperature with altitude could affect the thermodynamic equilibrium between gas ammonia and particulate ammonium and that this could contribute to observed vertical gradients in ammonia. It is also possible that dry deposition of ammonia at the surface could create a negative vertical gradient in ammonia near the surface. We recognize that there are limited measurement studies available for total gas ammonia and particulate ammonium, and as a result there is uncertainty in the estimate of background ammonia. Given this uncertainty, we believe it is appropriate to rely on measurement studies of total gas ammonia and particulate ammonium when available and reliable as explained elsewhere in this document (along with the IWAQM Phase 2 report default values), and to rely on the IWAQM Phase 2 report where monitoring data are not available.

Comment: The commenter states that inventories show very low ammonia concentrations in the corridors between Laramie River Station and the relevant Class I areas.

Response: We disagree with the commenter's assertions. We note that there is large uncertainty in estimates of ammonia emissions inventories that are based on source activity data and emissions factors. Moreover, even when more certain estimates of ammonia emissions are available, it is not possible to estimate ambient ammonia concentrations based on emissions inventory data alone. An estimate of ambient ammonia levels would require an evaluation of modeled emissions data and the effects of transport, dispersion and removal of ambient ammonia. Direct measurements of ambient concentrations of gas ammonia and particulate ammonium provide a more reliable estimate of background ammonia than do model simulations of the emissions, transport, dispersion and removal of ammonia.

Comment: The commenter states that reliable ammonia measurements from the IMPROVE monitor located in the Wind Cave National Park were published in September 2012. Chen et al, available at AECOM Report. This monitor provides actual ground level ammonia data that is representative of the Class I areas that are relevant to Laramie River Station. AECOM Report at 4. EPA has given no explanation for its assumption of a constant 2.0 ppb background concentration in this case.

Response: We evaluated the ammonia and ammonium monitoring data submitted by the commenter in Exhibit 6, which is an extended abstract presented at the 2010 conference in Whitefish, MT.[126] The data are from a pilot study conducted from April 2011 to January 2012 designed to measure total NHX as the sum of ammonia and particulate ammonium at 9 IMPROVE sites. The pilot study includes data for IMPROVE monitoring sites at Wind Cave and Rocky Mountain National Park, which are Class I areas for which we evaluated visibility impacts in this action. We note that the pilot study data are for less than one full year and are plotted in Figure 1 of the report as monthly average concentrations. The measured values of NHX are not reported, but the plot does show seasonal variation in NHX concentrations, as expected, with higher NHX concentrations in summer and lower concentrations in winter. Annual average NHX concentrations cannot be estimated from the plot itself, but they appear to be approximately consistent with the default IWAQM ammonia background concentration of 0.5 ppb for forested areas. Given that both the Rocky Mountain and Wind Cave Class I areas have significant forest cover, the measurements in the pilot study appear to be consistent with the IWAQM Phase 2 report.

Measurements of NHx are not reported for Badlands National Park, which is a mix of bare rock and mixed-grass prairie ecosystems. Based on the IWAQM Phase 2 report, default background ammonia concentrations in the range of 1 to 10 ppb at 20 degrees Celsius would be appropriate for this region. We reviewed the ambient ammonia monitoring data on which the IWAQM Phase 2 report was based, and the data for grasslands were largely based on measurements at Pawnee National Grassland, where average ammonia levels in summer were 10 ppb. Because the Pawnee National Grassland is located close to large agricultural and livestock ammonia sources in eastern Colorado, it is uncertain if the same ammonia levels would be appropriate for the more Badlands area. Therefore, we selected a background ammonia concentration of 1 ppb for CALPUFF modeling of BART sources that impact the Wind Cave and Badlands Class I areas.Start Printed Page 5126

Comment: The commenter states that if EPA uses the maximum 24-hour NOX emissions rate when modeling baseline visibility impacts, it should also use the maximum 24-hour SO2 and PM10 emissions rates for the baseline. NOX competes with SO2 for ammonia to make either ammonium nitrate or ammonium sulfate. Setting the SO2 baseline rate at a low concentration relative to NOX skews the model to predict the formation of more ammonium nitrate and less ammonium sulfate. This magnifies the modeled benefits of reducing NOX emissions.

Response: We agree that we did not use the maximum [actual] 24-hour emission rates for SO2 and PM10 as we did for NOX. However, we have not found based on our analysis, and the commenter has not established, that doing so had any material impact on the modeled benefits associated with NOX controls. The BART sources in Wyoming that are covered in this action are subject-to-BART only for NOX and PM. In addition, we considered comments on, but did not question the validity of the State's BART analyses for PM. In fact, as explained in detail elsewhere in this document, with respect to the State's PM BART determinations, the State's SIP and existing information was adequate to find that the PM BART determinations were reasonable. Accordingly, the purpose of our modeling effort was to identify the visibility improvement associated with NOX controls, not SO2 or PM controls. And so, in evaluating the visibility of NOX controls, we held the SO2 and PM emissions constant at the rate associated with the “committed controls” identified by the State. Therefore, even if there was a discernible impact on the modeled visibility benefit of NOX controls related to our treatment of modeled emission rates for SO2 and PM, it would be common to all of the modeled NOX control scenarios and would not have favored one control option over another.

Comment: The visibility improvement from SCR will be much less than EPA claims. The modeling preformed by AECOM and Wyoming produced similar results, and both predicted much less visibility improvement than EPA.

Response: The modeling performed by Wyoming used the 2 ppb background concentration that was established in the State's protocol, and this resulted in model visibility impacts that were significantly greater than those estimated by AECOM in its modeling using lower, seasonally varying background ammonia concentrations. The ammonia concentrations in ppb used in the AECOM modeling for the months of January December were as follows: 0.3, 0.9. 0.9, 1.0, 1.0, 2.0, 2.0, 1.0, 0.8, 0.8, 0.6, and 0.3. We note that our modeling results in our original proposal also used the State's protocol, and our model results were identical to the State's modeling results for the emissions scenarios that both the State and EPA evaluated. The EPA modeling results in our revised proposal showed larger visibility impacts because we corrected the baseline emissions rates to make the emissions consistent with the BART Guidance. In the new modeling results that we performed using our revised final EPA Protocol and included in this action, we used a model configuration that is generally consistent with modeling submitted by the commenter. The revisions to the protocol include reduced background ammonia, correction for ammonia limiting conditions, updated regulatory versions of the model, and the use of Method 8. The commenter did not submit model results for all emissions scenarios in a format that can be directly compared to our tabulated model results, but our revised model results in this action appear to be generally consistent with the commenter's model results, and these results do show that SCR at Basin Electric Laramie River has appreciable visibility benefit at the Wind Cave and Badlands Class I areas.

Comment: The commenter states that the version of CALPUFF used by Wyoming and EPA (version 5.711a) relies on simplified chemistry algorithms that overstate nitrate formation and overpredict visibility impacts, and that EPA acknowledges that “the simplified chemistry in the [CALPUFF] model tends to magnify the actual visibility effects of [a] source.” 70 FR 39121. Papers by Morris et al. and Karamchandani et al. show that CALPUFF chemistry overpredicts nitrates by a factor of 2-to-4 times in winter.

Response: As described in responses to other comments and in our modeling protocol, EPA used the currently approved CALPUFF version 5.8 for modeling used in this action. EPA has acknowledged in the BART Guidelines that there is uncertainty in the CALPUFF modeled visibility impacts. EPA recognized the uncertainty in the CALPUFF modeling results when EPA made the decision, in the final BART Guidelines, to recommend that the model be used to estimate the 98th percentile visibility impairment rather than the highest daily impact value. While recognizing the limitations of the CALPUFF model in the BART Guidelines Preamble, EPA concluded that, for the specific purposes of the Regional Haze Rule's BART provisions, CALPUFF is sufficiently reliable to inform the decision making process.

Comment: The commenter states that a study by the CDPHE showed model results for nitrates will be skewed high by assuming higher than actual background levels of ammonia.

Response: The CDPHE completed a CALPUFF model sensitivity study that evaluated the effect of the background ammonia concentration on model predictions for ammonium nitrate and sulfate. The CDPHE found that CALPUFF model predicted nitrate was insensitive to variations in background concentrations greater than 10 ppb and became progressively more sensitive to background ammonia as it was reduced from 10 to 0 ppb.[127] We note that CDPHE performed a sensitivity study but did not evaluate model performance and did not identify any particular case in which model performance was skewed by the use of inappropriate background ammonia concentrations. The conclusions of the CDPHE study are fully consistent with the IWAQM Phase 2 report, which also recognized that accurate specification of background ammonia “is critical to the accurate estimation of particulate nitrate concentrations.” [128]

Comment: The commenter cites several presentations and studies that document flaws in CALPUFF's sulfate and nitrate chemistry: (1) It is out of date, overly simplistic, and inaccurate; (2) CALPUFF greatly overstated sulfate and nitrate in winter, overestimating visibility impacts by 100-1000% in many cases; and (3) that the model understated sulfate in summer; and that nitrate predictions were particularly inaccurate, overstated, and unreliable.

Response: EPA recognized the uncertainty in the CALPUFF model when EPA made the decision, in the final BART Guidelines, to recommend that the model be used to estimate the 98th percentile visibility impairment rather than the highest daily impact value. While recognizing the limitations of the CALPUFF model in the BART Guidelines, EPA concluded that, for the specific purposes of the RHR's BART provisions, CALPUFF is sufficiently reliable to inform the decision making process.

Comment: NOX emissions control has little visibility benefit during summer when visibility impairment is dominated by wildfires.

Response: EPA agrees that nitrate impacts are more dominant in the winter. The CALPUFF model results are Start Printed Page 5127consistent with these observations, with the largest modeled visibility improvements occurring from late fall to early spring. NOX emissions are precursors to ammonium nitrate, and high concentrations of ammonium are typically observed from late fall to early spring when cold temperatures and high relative humidity results in thermodynamic conditions that favor the formation of ammonium nitrate. Observed ammonium nitrate concentrations are typically low during summer because warm temperatures result in thermodynamic conditions that are not favorable to the formation of ammonium nitrate. Nonetheless, there may be higher nitrate concentrations on colder days during this period.

Comment: EPA's visibility-benefits analysis still is constrained in the re-proposed Wyoming haze plan because EPA has not identified the visibility benefits from BART controls across all of the Class I areas affected by haze-causing pollutants from Wyoming sources. Wyoming EGUs impact visibility over at least 18 Class I areas. While EPA's own visibility modeling fully supports determinations that SCR is BART for all Wyoming EGUs, the visibility benefits of SCR across all affected Class I areas are cumulatively significant and, if the RHR's fundamental purpose is to be fulfilled, they must not be ignored.

Response: EPA disagrees with this comment. The commenter's number of “at least 18 Class I areas” is derived by including Class I areas that are more than 300 km from BART sources. EPA disagrees that these Class I areas should be included in the visibility analysis. The IWAQM Phase 2 report reviewed model performance evaluations of CALPUFF as a function of distance from the source and concluded that:”Based on the tracer comparison results presented in Section 4.6, it appears that CALPUFF provides reasonable correspondence with observations for transport distances of over 100 km. Most of these comparisons involved concentration values averaged over 5 to 12 hours. The CAPTEX comparisons, which involved comparisons at receptors that were 300 km to 1000 km from the release, suggest that CALPUFF can overestimate surface concentrations by a factor of 3 to 4. Use of the puff splitting option in CALPUFF might have improved these comparisons, but there are serious conceptual concerns with the use of puff dispersion for very long-range transport (300 km and beyond). As the puffs enlarge due to dispersion, it becomes problematic to characterize the transport by a single wind vector, as significant wind direction shear may well exist over the puff dimensions. With the above thoughts in mind, IWAQM recommends use of CALPUFF for transport distances of order 200 km and less. Use of CALPUFF for characterizing transport beyond 200 to 300 km should be done cautiously with an awareness of the likely problems involved.” [129] We present additional discussion of this issue in our response to the following comment.

Comment: EPA arbitrarily failed to model visibility impacts of the various control options at all affected Class I areas, including those that are beyond 300 km from the source. EPA recently responded to a similar comment in its final action promulgating the Montana Regional Haze FIP, 77 FR 57864, for the first time supporting its truncated modeling by referencing a now-discredited 1998 report regarding CALPUFF performance. Because EPA raised this issue only after the close of the public comment period on its Wyoming regional haze action, EPA should consider the Conservation Organizations' response. See 42 U.S.C. 7607(d)(4)(B)(i).

In its response to public comments on the Montana FIP, EPA stated, “The Interagency Workgroup on Air Quality Modeling (IWAQM) Phase 2 report (EPA, 1998) reviewed model performance evaluations of CALPUFF as a function of distance from the source and concluded that: . . .[u]se of CALPUFF for characterizing transport beyond 200 to 300 km should be done cautiously with an awareness of the likely problems involved.” 77 FR 57867-68. EPA then concludes, “[t]herefore, given that the IWAQM guidance provides for the use of the CALPUFF model at receptor distances of up to 200 to 300 km, and given that EPA has already addressed uncertainty in the CALPUFF model, we believe it is reasonable to use CALPUFF to evaluate visibility impacts up to 300 km.” Id. at 57868.

We agree that CALPUFF is reliable at distances of 300 km. However, EPA's use of the IWAQM Phase 2 report to support its decision to exclude modeling at distances beyond 300 km is arbitrary. First, changes to CALPUFF since 1998 may correct problems identified in the IWAQM Phase 2 report with modeling accuracy in the 200-1,000 km range. Second, a more recent study prepared for EPA called into question the conclusions of the IWAQM Phase 2 report upon which EPA relies. See Long Range Transport Models Using Tracer Field Experiment Data (May 2012) (EPA Contract No: EP-D-07-102, Work Assignment No: 4-06). The May 2012 study concluded that “The inability of most (~90%) of the current study's CALPUFF sensitivity tests to reproduce the 1998 EPA study tracer test residence time on the 600 km receptor arc is a cause for concern.” Not only were the authors of the May 2012 study unable to reproduce the 1998 study's findings that CALPUFF overestimated pollutant concentrations at distances of 600 km, the 2012 study concluded that CALPUFF actually underestimates average pollutant concentrations at 600 km. Accordingly, reliance on CALPUFF at long distances would result in conservative estimates of visibility impacts. It is not appropriate to assume, as EPA effectively did in its Wyoming proposal, that such impacts are non-existent. EPA's failure to model and consider visibility impacts at all affected Class I areas, including those beyond 300 km, is not supported.

Because the RHR, and SIPs and FIPs promulgated to implement it, are to fulfill CAA requirements to mitigate and ultimately eliminate anthropogenic sources of haze pollution at all Class I national parks and wilderness areas, it is imperative that states and EPA use models to completely and accurately depict the visibility impact of a source to the region's Class I areas as well as projected benefits from BART. In this regard, the conclusion of the May 2012 study that CALPUFF reliably (if conservatively) identifies visibility impacts to Class I areas beyond those previously evaluated are critical, and directs EPA to supplement the incomplete analysis presented in its proposed action on the Wyoming Regional Haze plan with additional modeling, or consider the more complete modeling submitted by the conservation organizations with their August 2, 2012 comments.

Response: EPA disagrees with the commenter's assertion that changes to CALPUFF now support modeling at distances greater than 300 km. The commenter cited a May 2012 technical evaluation (Documentation of the Evaluation of CALPUFF and Other Long Range Transport Models Using Tracer Field Experiment Data [130] ) that evaluates several long range transport models based on several tracer studies. The report cited by the commenter does not refute the IWAQM Phase 2 report which states that “IWAQM recommends use of CALPUFF for transport distances of order 200 km and less. Use of CALPUFF for characterizing transport Start Printed Page 5128beyond 200 to 300 km should be done cautiously with an awareness of the likely problems involved.” [131] In fact, the May 2012 report further “emphasizes the need for a standardized set of options for regulatory CALPUFF modeling.” [132] Given these findings, EPA does not agree, as the commenter asserts, that it must consider CALPUFF modeling results from Federal Class I areas beyond 300 km. EPA therefore believes that the results of CALPUFF modeling beyond 300 km of the source should be evaluated in light of the limitations discussed in the two guidance documents cited above.

Finally, we disagree that there is any notice issue with respect to the commenter's allegations that EPA referenced the 1998 IWAQM study for the first time in our response to comments in our Montana FIP action. As quoted above, the BART guidelines specifically reference the 1998 IWAQM study with respect to CALPUFF settings.

Comment: EPA modeled visibility benefits at four Class I areas, and demonstrated visibility improvement due to SCR that approximately doubled the improvement afforded by SNCR at every Class I area modeled. 78 FR 34775-34776. EPA properly took account of the cumulative visibility improvement across all four modeled Class I areas for each unit, id. at 34776, but in fact, as the Conservation Organizations commented previously, see 8/2/2012 Conservation Organization Comments, SCR affords visibility benefits across at least six Class I areas. Thus, the cumulative visibility benefits are even greater than found by EPA, and further support a determination that SCR is BART for Laramie River Station Units 1-3.

Response: We disagree that we should have evaluated visibility impacts at all of the areas that the commenter considered in its analysis. The commenter provided CALPUFF model results at 18 areas, including areas that are not mandatory Class I areas, and at Class I areas at distances greater than 300 km from Laramie River Station. In our analysis of visibility impacts, we considered the visibility improvement at four Class I areas within 300 km of the Laramie River Station. Therefore, our modeling analyses did not ignore the visibility improvement that would be achieved at areas other than the most impacted Class I area, and we disagree with the assertion that we should have evaluated all of the areas that the commenter considered.

Comment: EPA's re-proposed Wyoming haze plan presents a unit-by-unit analysis of the visibility benefits of the installation of various BART control alternatives at Wyoming EGUs, and identifies benefits at only a subset of the affected Class I areas. However, EPA did not present evidence of the cumulative visibility benefits that would be enjoyed by Class