[Federal Register Volume 79, Number 20 (Thursday, January 30, 2014)]
[Rules and Regulations]
[Pages 5032-5222]
From the Federal Register Online via the Government Publishing Office [www.gpo.gov]
[FR Doc No: 2014-00930]



[[Page 5031]]

Vol. 79

Thursday,

No. 20

January 30, 2014

Part II





Environmental Protection Agency





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





40 CFR Part 52





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

  Federal Register / Vol. 79 , No. 20 / Thursday, January 30, 2014 / 
Rules and Regulations  

[[Page 5032]]


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

ENVIRONMENTAL PROTECTION AGENCY

40 CFR Part 52

[EPA-R08-OAR-2012-0026, FRL9905-42-R08]


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

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.

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, [email protected].

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 NH 3 mean or refer to ammonia.
    xxxiii. The initials NO X 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 PM 2.5 mean or refer to 
particulate matter with an aerodynamic diameter of less than 2.5 
micrometers.
    xxxvii. The initials PM 10 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 SO 2 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 Goals

[[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.
---------------------------------------------------------------------------

    \1\ We issued a finding of failure to submit for Wyoming only 
for the requirements of 40 CFR 51.309(g)) regarding required SIP 
provisions, including NOX BART, to address visibility at 
Class I areas other than the 16 areas covered by the Grand Canyon 
Visibility Transport Commission Report. Wyoming had submitted a SIP 
for the rest of the requirements under 40 CFR 51.309 prior to our 
January 15, 2009 finding.
---------------------------------------------------------------------------

    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).
---------------------------------------------------------------------------

    \2\ 77 FR 73926 (Dec. 12, 2012).
---------------------------------------------------------------------------

    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.
---------------------------------------------------------------------------

    \3\ WildEarth Guardians v. Jackson, 1:11-cv-CMA-MEH (D. Colo.).
---------------------------------------------------------------------------

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

    \4\ WildEarth Guardians v. Jackson, 1:11-cv-CMA-MEH (D. Colo.) 
(Dkt. Nos. 73, 74).
---------------------------------------------------------------------------

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),

[[Page 5034]]

NOX, 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).
---------------------------------------------------------------------------

    \5\ Visual range is the greatest distance, in kilometers or 
miles, at which a dark object can be viewed against the sky.
---------------------------------------------------------------------------

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.
---------------------------------------------------------------------------

    \6\ Areas designated as mandatory Class I Federal areas consist 
of national parks exceeding 6000 acres, wilderness areas and 
national memorial parks exceeding 5000 acres, and all international 
parks that were in existence on August 7, 1977. 42 U.S.C. 7472(a). 
In accordance with section 169A of the CAA, EPA, in consultation 
with the Department of Interior, promulgated a list of 156 areas 
where visibility is identified as an important value. 44 FR 69122 
(November 30, 1979). The extent of a mandatory Class I area includes 
subsequent changes in boundaries, such as park expansions. 42 U.S.C. 
7472(a). Although states and tribes may designate as Class I 
additional areas which they consider to have visibility as an 
important value, the requirements of the visibility program set 
forth in section 169A of the CAA apply only to ``mandatory Class I 
Federal areas.'' Each mandatory Class I Federal area is the 
responsibility of a ``Federal Land Manager.'' 42 U.S.C. 7602(i). 
When we use the term ``Class I area'' in this action, we mean a 
``mandatory Class I Federal area.''
---------------------------------------------------------------------------

    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\
---------------------------------------------------------------------------

    \7\ EPA's regional haze regulations require subsequent updates 
to the regional haze SIPs. 40 CFR 51.308(g)-(i).
---------------------------------------------------------------------------

    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

[[Page 5035]]

existence 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\
---------------------------------------------------------------------------

    \8\ The preamble to the RHR provides additional details about 
the deciview. 64 FR 35714, 35725 (July 1, 1999).
---------------------------------------------------------------------------

    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\
---------------------------------------------------------------------------

    \9\ Guidance for Estimating Natural Visibility Conditions Under 
the Regional Haze Rule, September 2003, EPA-454/B-03-005, available 
at http://www.epa.gov/ttncaaa1/t1/memoranda/Regional_Haze_envcurhr_gd.pdf, (hereinafter referred to as ``our 2003 Natural 
Visibility Guidance''); and Guidance for Tracking Progress Under the 
Regional Haze Rule, (September 2003, EPA-454/B-03-004, available at 
http://www.epa.gov/ttncaaa1/t1/memoranda/rh_tpurhr_gd.pdf, 
(hereinafter referred to as our ``2003 Tracking Progress 
Guidance'').
---------------------------------------------------------------------------

    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

[[Page 5036]]

stationary 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.
---------------------------------------------------------------------------

    \10\ The set of ``major stationary sources'' potentially 
subject-to-BART is listed in CAA section 169A(g)(7).
---------------------------------------------------------------------------

    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.
---------------------------------------------------------------------------

    \11\ BART-eligible sources are those sources that have the 
potential to emit 250 tons or more of a visibility-impairing air 
pollutant, were not in operation prior to August 7, 1962, but were 
in existence on August 7, 1977, and whose operations fall within one 
or more of 26 specifically listed source categories. 40 CFR 51.301.
---------------------------------------------------------------------------

    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

[[Page 5037]]

effect 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\
---------------------------------------------------------------------------

    \12\ On May 15, 2012 the EPA signed the first proposed rule on 
the Wyoming Regional Haze SIP which proposed to partially approve 
and partially disapprove the Wyoming state plan. The EPA published 
the proposed rule in the Federal Register for public comment on June 
4, 2012. This public Federal Register notice may be found at 77 FR 
33022 (June 4, 2012). EPA then obtained an extension to the Consent 
Decree deadline in order to re-propose the Wyoming regional haze 
plan based on data generated after the conclusion of the original 
comment period. In this document, all references to ``proposal'' or 
``proposal notice'' refer to the notice published on June 10, 2013 
unless otherwise stated.
    \13\ E.g., 78 FR 34777. The proposed notice also explained that 
``[t]he Agency will take the comments and testimony received, as 
well as any further SIP revisions submitted by the State, into 
consideration in our final promulgation. Supplemental information 
received may lead the Agency to adopt final SIP and/or FIP 
regulations that reflect a different BART control technology option, 
or impact other proposed regulatory provisions, which differ from 
this proposal.'' 78 FR 34777.
---------------------------------------------------------------------------

    In our 2013 proposal we proposed to disapprove the following:

[[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
--------------------------------------------------------------------------------------------------------------------------------------------------------
                                                                 Emission  limit--lb/                                                   Average cost-
              Source                       Technology *             MMBtu  (30-day       Total capital cost     Total annualized     effectiveness  ($/
                                                                  rolling  average)             ($)                 cost  ($)               ton)
--------------------------------------------------------------------------------------------------------------------------------------------------------
Dave Johnston Unit 3..............  New low-NOX burners (LNBs)  0.28 (for LNBs with    $15,976,696 (for LNBs  $1,828,137 (for LNBs  $644 (for LNBs with
                                     with overfire air (OFA)     OFA).                  with OFA).             with OFA).            OFA).
                                     and shut down in 2027; or
                                     new LNBs with OFA and
                                     selective catalytic
                                     reduction (SCR) **.

[[Page 5039]]

 
Laramie River Unit 1..............  New LNBs/OFA and SCR......  0.07.................  $180,254,572.........  $21,770,134.........  $4,461.
Laramie River Unit 2..............  New LNBs with OFA and SCR.  0.07.................  $188,826,333.........  $22,691,467.........  $4,424.
Laramie River Unit 3..............  New LNBs with OFA and SCR.  0.07.................  $188,437,953.........  $22,666,982.........  $4,375.
Wyodak Unit 1.....................  New LNBs with OFA and SCR.  0.07.................  $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
---------------------------------------------------------------------------

    \14\ Andover Technology Partners, ``Cost of NOX 
Controls on Wyoming EGUs'', October 28, 2013; Wyoming EGU BART and 
Reasonable Progress Costs--10/28/2013; Wyoming EGU BART and 
Reasonable Progress Costs for Jim Bridger--10/28/2013.
    \15\ Air Quality Modeling Protocol: Wyoming Regional Haze 
Federal Implementation Plan, U.S. EPA, January, 2014.

                                            Table 2--Summary of EPA's Laramie River Unit 1 NOX BART Analysis
--------------------------------------------------------------------------------------------------------------------------------------------------------
                                                                                                                                            Visibility
                                                                                                                                            improvement
                                                                                                                                              (Delta
                                                           Emission rate                                                    Incremental    deciview for
                                                            (lb/MMBtu;       Emission       Annualized     Average cost        cost         the maximum
                   Control technology                         annual         reduction         costs       effectiveness   effectiveness       98th
                                                             average)          (tpy)                          ($/ton)         ($/ton)       percentile
                                                                                                                                             impact at
                                                                                                                                             Badlands
                                                                                                                                          National Park)
--------------------------------------------------------------------------------------------------------------------------------------------------------
New LNBs with OFA.......................................            0.19           1,556      $2,268,806          $1,458  ..............            0.18
New LNBs with OFA and selective non-catalytic reduction             0.15           2,445       8,554,896           3,485          $6,993            0.28
 (SNCR).................................................
New LNBs with OFA and SCR...............................            0.05           4,880      21,770,134           4,461           5,449            0.57
--------------------------------------------------------------------------------------------------------------------------------------------------------


[[Page 5040]]


                                            Table 3--Summary of EPA's Laramie River Unit 2 NOX BART Analysis
--------------------------------------------------------------------------------------------------------------------------------------------------------
                                                                                                                                            Visibility
                                                                                                                                            improvement
                                                                                                                                              (Delta
                                                           Emission rate                                                    Incremental    deciview for
                                                            (lb/MMBtu;       Emission       Annualized     Average cost        cost         the maximum
                   Control technology                         annual         reduction         costs       effectiveness   effectiveness       98th
                                                             average)          (tpy)                          ($/ton)         ($/ton)       percentile
                                                                                                                                             impact at
                                                                                                                                             Badlands
                                                                                                                                          National Park)
--------------------------------------------------------------------------------------------------------------------------------------------------------
New LNBs with OFA.......................................            0.19            1823      $2,268,806          $1,244  ..............            0.18
New LNBs with OFA and SNCR..............................            0.15           2,717       8,531,631           3,140          $7,006            0.27
New LNBs with OFA and SCR...............................            0.05           5,129      22,691,467           4,424           5,871            0.53
--------------------------------------------------------------------------------------------------------------------------------------------------------


                                            Table 4--Summary of EPA's Laramie River Unit 3 NOX BART Analysis
--------------------------------------------------------------------------------------------------------------------------------------------------------
                                                                                                                                            Visibility
                                                                                                                                            improvement
                                                                                                                                              (Delta
                                                           Emission rate                                                    Incremental    deciview for
                                                            (lb/MMBtu;       Emission       Annualized     Average cost        cost         the maximum
                   Control technology                         annual         reduction         costs       effectiveness   effectiveness       98th
                                                             average)          (tpy)                          ($/ton)         ($/ton)       percentile
                                                                                                                                             impact at
                                                                                                                                             Badlands
                                                                                                                                          National Park)
--------------------------------------------------------------------------------------------------------------------------------------------------------
New LNBs with OFA.......................................            0.19            1789      $2,268,806          $1,268  ..............            0.18
New LNBs with OFA and SNCR..............................            0.15           2,706       8,643,839           3,194          $6,951            0.27
New LNBs with OFA and SCR...............................            0.05           5,181      22,666,982           4,375           5,667            0.52
--------------------------------------------------------------------------------------------------------------------------------------------------------

    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
--------------------------------------------------------------------------------------------------------------------------------------------------------
                                                                                                                                            Visibility
                                                                                                                                            improvement
                                                                                                                                              (Delta
                                                                                                                                           deciview for
                                                           Emission rate     Emission                      Average cost     Incremental     the maximum
                   Control technology                       (lb/MMBtu;       reduction      Annualized     effectiveness       cost            98th
                                                              annual           (tpy)           costs          ($/ton)      effectiveness    percentile
                                                             average)                                                         ($/ton)        impact at
                                                                                                                                              Bridger
                                                                                                                                            Wilderness
                                                                                                                                             Area) **
--------------------------------------------------------------------------------------------------------------------------------------------------------
New LNBs with SOFA......................................            0.18           4,558      $1,167,297            $256  ..............       0.17/0.23
New LNBs with SOFA and SNCR.............................            0.14           5,332       4,330,052             812          $4,088       0.20/0.27
New LNBs with SOFA and SCR..............................            0.05           7,352      19,372,105           2,635           7,447       0.27/0.37
--------------------------------------------------------------------------------------------------------------------------------------------------------
* 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
--------------------------------------------------------------------------------------------------------------------------------------------------------
                                                                                                                                            Visibility
                                                                                                                                            improvement
                                                                                                                                              (Delta
                                                                                                                                           deciview for
                                                           Emission rate     Emission                      Average cost     Incremental     the maximum
                   Control technology                       (lb/MMBtu;       reduction      Annualized     effectiveness       cost            98th
                                                              annual           (tpy)           costs          ($/ton)      effectiveness    percentile
                                                             average)                                                         ($/ton)        impact at
                                                                                                                                              Bridger
                                                                                                                                            Wilderness
                                                                                                                                              Area) *
--------------------------------------------------------------------------------------------------------------------------------------------------------
New LNBs with SOFA......................................            0.19           3,787      $1,167,297            $308  ..............       0.16/0.21
New LNBs with SOFA and SNCR.............................            0.15           4,545       4,291,184             944          $4,122       0.19/0.25
New LNBs with SOFA and SCR..............................            0.05           6,554      22,307,492           3,403           8,968       0.27/0.36
--------------------------------------------------------------------------------------------------------------------------------------------------------
* 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.


[[Page 5041]]


                                             Table 7--Summary of EPA's Jim Bridger Unit 3 NOX BART Analysis
--------------------------------------------------------------------------------------------------------------------------------------------------------
                                                                                                                                            Visibility
                                                                                                                                            improvement
                                                                                                                                              (Delta
                                                                                                                                           deciview for
                                                           Emission rate     Emission                      Average cost     Incremental     the maximum
                   Control technology                       (lb/MMBtu;       reduction      Annualized     effectiveness       cost            98th
                                                              annual           (tpy)           costs          ($/ton)      effectiveness    percentile
                                                             average)                                                         ($/ton)        impact at
                                                                                                                                              Bridger
                                                                                                                                            Wilderness
                                                                                                                                              Area) *
--------------------------------------------------------------------------------------------------------------------------------------------------------
New LNBs with SOFA......................................            0.20           3,710      $1,167,297            $315  ..............       0.14/0.19
New LNBs with SOFA and SNCR.............................            0.16           4,539       4,458,776             982          $3,972       0.17/0.23
New LNBs with SOFA and SCR..............................            0.05           6,799      22,573,920           3,320           8,015       0.26/0.35
--------------------------------------------------------------------------------------------------------------------------------------------------------
* 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
--------------------------------------------------------------------------------------------------------------------------------------------------------
                                                                                                                                            Visibility
                                                                                                                                            improvement
                                                                                                                                              (Delta
                                                                                                                                           deciview for
                                                           Emission rate     Emission                      Average cost     Incremental     the maximum
                   Control technology                       (lb/MMBtu;       reduction      Annualized     effectiveness       cost            98th
                                                              annual           (tpy)           costs          ($/ton)      effectiveness    percentile
                                                             average)                                                         ($/ton)        impact at
                                                                                                                                               Rawah
                                                                                                                                            Wilderness
                                                                                                                                              Area) *
--------------------------------------------------------------------------------------------------------------------------------------------------------
New LNBs with SOFA......................................            0.19           4,161      $1,167,297            $281  ..............       0.25/0.23
New LNBs with SOFA and SNCR.............................            0.15           4,956       4,372,457             882          $4,035       0.30/0.28
New LNBs with SOFA and SCR..............................            0.05           7,108      19,494,417           2,743           7,027       0.45/0.42
--------------------------------------------------------------------------------------------------------------------------------------------------------
* 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

[[Page 5042]]

deciviews 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]
--------------------------------------------------------------------------------------------------------------------------------------------------------
                                                                                                                                            Visibility
                                                                                                                                            improvement
                                                                                                                                              (Delta
                                                           Emission rate                                                    Incremental    deciview for
                                                            (lb/MMBtu;       Emission       Annualized     Average cost        cost         the maximum
                   Control technology                         annual         reduction         costs       effectiveness   effectiveness       98th
                                                             average)          (tpy)                          ($/ton)         ($/ton)       percentile
                                                                                                                                          impact at Wind
                                                                                                                                           Cave National
                                                                                                                                               Park)
--------------------------------------------------------------------------------------------------------------------------------------------------------
New LNBs with OFA.......................................            0.22           2,837      $1,828,137            $644  ..............            0.33
New LNBs with OFA and SNCR..............................            0.16           3,356       3,898,930           1,162          $3,988            0.39
New LNBs with OFA and SCR...............................            0.05           4,433      16,591,006           3,742          11,781            0.51
--------------------------------------------------------------------------------------------------------------------------------------------------------


                                            Table 10--Summary of EPA's Dave Johnston Unit 3 NOX BART Analysis
                                                             [20 Year remaining useful life]
--------------------------------------------------------------------------------------------------------------------------------------------------------
                                                                                                                                            Visibility
                                                                                                                                            improvement
                                                                                                                                              (Delta
                                                           Emission rate                                                    Incremental    deciview for
                                                            (lb/MMBtu;       Emission       Annualized     Average cost        cost         the maximum
                   Control technology                         annual         reduction         costs       effectiveness   effectiveness       98th
                                                             average)          (tpy)                          ($/ton)         ($/ton)       percentile
                                                                                                                                          impact at Wind
                                                                                                                                           Cave National
                                                                                                                                               Park)
--------------------------------------------------------------------------------------------------------------------------------------------------------
New LNBs with OFA.......................................            0.22           2,837      $1,699,807            $599  ..............            0.33
New LNBs with OFA and SNCR..............................            0.16           3,356       3,510,589           1,046          $3,488            0.39
New LNBs with OFA and SCR...............................            0.05           4,433      11,680,144           2,635           7,583            0.51
--------------------------------------------------------------------------------------------------------------------------------------------------------


                                            Table 11--Summary of EPA's Dave Johnston Unit 4 NOX BART Analysis
--------------------------------------------------------------------------------------------------------------------------------------------------------
                                                                                                                                            Visibility
                                                                                                                                            improvement
                                                                                                                                              (Delta
                                                           Emission rate                                                    Incremental    deciview for
                                                            (lb/MMBtu;       Emission       Annualized     Average cost        cost         the maximum
                   Control technology                         annual         reduction         costs       effectiveness   effectiveness       98th
                                                             average)          (tpy)                          ($/ton)         ($/ton)       percentile
                                                                                                                                          impact at Wind
                                                                                                                                           Cave National
                                                                                                                                               Park)
--------------------------------------------------------------------------------------------------------------------------------------------------------
New LNBs with OFA.......................................            0.14           3,114        $767,342            $246  ..............            0.41
New LNBs with OFA and SNCR..............................            0.11           3,505       2,541,600             725          $4,535            0.46
New LNBs with OFA and SCR...............................            0.05           4,377      14,158,899           3,235          13,312            0.57
--------------------------------------------------------------------------------------------------------------------------------------------------------

    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.

[[Page 5043]]



                                              Table 12--Summary of EPA's Naughton Unit 1 NOX BART Analysis
--------------------------------------------------------------------------------------------------------------------------------------------------------
                                                                                                                                            Visibility
                                                                                                                                            improvement
                                                                                                                                              (Delta
                                                                                                                                           deciview for
                                                          Emission  rate     Emission                      Average cost     Incremental     the Maximum
                   Control technology                       (lb/MMBtu;       reduction      Annualized     effectiveness       cost            98th
                                                              annual           (tpy)           costs          ($/ton)      effectiveness    percentile
                                                             average)                                                         ($/ton)        impact at
                                                                                                                                              Bridger
                                                                                                                                            Wilderness
                                                                                                                                              Area) *
--------------------------------------------------------------------------------------------------------------------------------------------------------
New LNBs with OFA.......................................            0.21           2,100        $932,466            $444  ..............       0.22/0.26
New LNBs with OFA and SNCR..............................            0.16           2,463       2,234,827             907          $3,584       0.26/0.30
New LNBs with OFA and SCR...............................            0.05           3,209       9,974,616           3,109          10,384       0.33/0.39
--------------------------------------------------------------------------------------------------------------------------------------------------------
* 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
--------------------------------------------------------------------------------------------------------------------------------------------------------
                                                                                                                                            Visibility
                                                                                                                                            improvement
                                                                                                                                              (Delta
                                                                                                                                           deciview for
                                                           Emission rate     Emission                      Average cost     Incremental     the Maximum
                   Control technology                       (lb/MMBtu;       reduction      Annualized     effectiveness       cost            98th
                                                              annual           (tpy)           costs          ($/ton)      effectiveness    percentile
                                                             average)                                                         ($/ton)        impact at
                                                                                                                                              Bridger
                                                                                                                                            Wilderness
                                                                                                                                              Area) *
--------------------------------------------------------------------------------------------------------------------------------------------------------
New LNBs with OFA.......................................            0.21           2,586        $883,900            $342  ..............       0.28/0.32
New LNBs with OFA and SNCR..............................            0.16           3,024       2,480,832             820          $3,647       0.34/0.38
New LNBs with OFA and SCR...............................            0.05           3,922      10,062,750           2,566           8,440       0.42/0.46
--------------------------------------------------------------------------------------------------------------------------------------------------------
* 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]
--------------------------------------------------------------------------------------------------------------------------------------------------------
                                                                                                                                            Visibility
                                                                                                                                            improvement
                                                                                                                                              (Delta
                                                                                                                                           deciview for
                                                           Emission rate     Emission                      Average cost     Incremental     the Maximum
                   Control technology                       (lb/MMBtu;       reduction      Annualized     effectiveness       cost            98th
                                                              annual           (tpy)           costs          ($/ton)      effectiveness    percentile
                                                             average)                                                         ($/ton)        impact at
                                                                                                                                              Bridger
                                                                                                                                            Wilderness
                                                                                                                                              Area) *
--------------------------------------------------------------------------------------------------------------------------------------------------------
Existing LNBs with OFA**................................            0.33             442        $106,393            $240  ..............       0.05/0.07
Existing LNBs with OFA and SNCR.........................            0.23           1,673       3,852,377           2,303          $3,045       0.20/0.29
Existing LNBs with OFA and SCR..........................            0.05           3,922      13,604,702           3,469           4,335       0.49/0.60
--------------------------------------------------------------------------------------------------------------------------------------------------------
* 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.

[[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
--------------------------------------------------------------------------------------------------------------------------------------------------------
                                                                                                                                            Visibility
                                                                                                                                            improvement
                                                                                                                                              (delta
                                                           Emission rate                                                    Incremental    deciview for
                                                            (lb/MMBtu;       Emission       Annualized     Average cost        cost         the maximum
                   Control technology                         annual         reduction         costs       effectiveness   effectiveness       98th
                                                             average)          (tpy)                          ($/ton)         ($/ton)       percentile
                                                                                                                                          impact at Wind
                                                                                                                                           Cave National
                                                                                                                                               Park)
--------------------------------------------------------------------------------------------------------------------------------------------------------
New LNBs with OFA.......................................            0.19           1,239      $1,272,427          $1,027  ..............            0.21
New LNBs with OFA and SNCR..............................            0.15           1,914       3,726,573           1,947           3,635            0.32
New LNBs with OFA and SCR...............................            0.05           3,735      15,073,502           4,036           6,233            0.61
--------------------------------------------------------------------------------------------------------------------------------------------------------

    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
--------------------------------------------------------------------------------------------------------------------------------------------------------
                                                                                                                                            Visibility
                                                                                                                                            improvement
                                                                                                                                              (delta
                                                           Emission rate                                                    Incremental    deciview for
                                                            (lb/MMBtu;       Emission       Annualized     Average cost        cost         the maximum
                   Control technology                         annual         reduction         costs       effectiveness   effectiveness       98th
                                                             average)          (tpy)                          ($/ton)         ($/ton)       percentile
                                                                                                                                          impact at Wind
                                                                                                                                           Cave National
                                                                                                                                               Park)
--------------------------------------------------------------------------------------------------------------------------------------------------------
LNBs with OFA *.........................................            0.20           1,226      $1,214,000            $990  ..............            0.12
LNBs with OFA and SNCR..................................            0.15           1,466       2,096,430           1,430           3,670            0.14
LNBs with OFA and SCR...................................            0.05           1,947       6,808,374           3,496           9,798            0.18
--------------------------------------------------------------------------------------------------------------------------------------------------------
* 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
--------------------------------------------------------------------------------------------------------------------------------------------------------
                                                                                                                                            Visibility
                                                                                                                                            improvement
                                                                                                                                              (delta
                                                           Emission rate                                                    Incremental    deciview for
                                                            (lb/MMBtu;       Emission       Annualized     Average cost        cost         the maximum
                   Control technology                         annual         reduction         costs       effectiveness   effectiveness       98th
                                                             average)          (tpy)                          ($/ton)         ($/ton)       percentile
                                                                                                                                          impact at Wind
                                                                                                                                           Cave National
                                                                                                                                               Park)
--------------------------------------------------------------------------------------------------------------------------------------------------------
LNBs with OFA...........................................            0.20           1,180      $1,441,146          $1,221  ..............            0.11
LNBs with OFA and SNCR..................................            0.15           1,425       2,335,022           1,638           3,645            0.14
LNBs with OFA and SCR...................................            0.05           1,916       7,037,969           3,673           9,588            0.18
--------------------------------------------------------------------------------------------------------------------------------------------------------


[[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

[[Page 5046]]

Unit 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

[[Page 5047]]

reconsider certain aspects of our decision for those sources.
---------------------------------------------------------------------------

    \16\ We are finalizing our proposed approval of the State's PM 
BART determinations. We did not receive any adverse comments that 
were sufficient to convince us that reexamination of the State's 
control costs was warranted.
---------------------------------------------------------------------------

    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 Station                LNB/SOFA           LNB/SOFA + SNCR             LNB/SOFA + SCR
----------------------------------------------------------------------------------------------------------------
Unit 1.............................  0.18 deciviews.......  0.28 deciviews.......  0.57 deciviews
Unit 2.............................  0.18 deciviews.......  0.27 deciviews.......  0.53 deciviews
Unit 3.............................  0.18 deciviews.......  0.27 deciviews.......  0.52 deciviews
                                    ----------------------------------------------------------------------------
    Total *........................  0.54 deciviews.......  0.82 deciviews.......  1.62 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,

[[Page 5048]]

with 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\
---------------------------------------------------------------------------

    \17\ Air Quality Modeling Protocol: Wyoming Regional Haze 
Federal Implementation Plan, U.S. EPA, January, 2014.

                                  Table 19--Visibility Modeling for Jim Bridger
----------------------------------------------------------------------------------------------------------------
           Jim Bridger                  LNB/SOFA         LNB/SOFA + SNCR               LNB/SOFA + SCR
----------------------------------------------------------------------------------------------------------------
Unit 1..........................  0.17/0.23 deciviews  0.20/0.27 deciviews  0.27/0.37 deciviews
Unit 2..........................  0.16/0.21 deciviews  0.19/0.25 deciviews  0.27/0.36 deciviews
Unit 3..........................  0.14/0.19 deciviews  0.17/0.23 deciviews  0.26/0.35 deciviews
Unit 4..........................  0.25/0.23 deciviews  0.30/0.28 deciviews  0.45/0.42 deciviews
                                 -------------------------------------------------------------------------------
    Total *.....................  0.72/0.86 deciviews  0.86/1.03 deciviews  1.25/1.5 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

[[Page 5049]]

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 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 Johnston                    LNB/OFA             LNB/OFA + SNCR             LNB/OFA + SCR
----------------------------------------------------------------------------------------------------------------
Unit 3.............................  0.33 deciviews.......  0.39 deciviews.......  0.51 deciviews
Unit 4.............................  0.41 deciviews.......  0.46 deciviews.......  0.57 deciviews
                                    ----------------------------------------------------------------------------
    Total *........................  0.74 deciviews.......  0.85 deciviews.......  1.08 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

[[Page 5050]]

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).

                                   Table 21--Visibility Modeling for Naughton
----------------------------------------------------------------------------------------------------------------
               Naughton                        LNB/OFA               LNB/OFA + SNCR           LNB/OFA + SCR
----------------------------------------------------------------------------------------------------------------
Unit 1...............................  0.22/0.26 deciviews....  0.26/0.30 deciviews....  0.33/0.39 deciviews.
Unit 2...............................  0.28/0.32 deciviews....  0.34/0.38 deciviews....  0.42/0.46 deciviews.
Unit 3...............................  0.05/0.07 deciviews....  0.20/0.29 deciviews....  0.49/0.60 deciviews.
                                      --------------------------------------------------------------------------
    Total *..........................  0.55/0.65 deciviews....  0.80/0.97 deciviews....  1.24/1.45 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
----------------------------------------------------------------------------------------------------------------
               Wyodak                       LNB/SOFA           LNB/SOFA + SNCR             LNB/SOFA + SCR
----------------------------------------------------------------------------------------------------------------
Unit 1.............................  0.21 deciviews.......  0.32 deciviews.......  0.61 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.

[[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

[[Page 5052]]

when 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

[[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.

[[Page 5054]]

7602(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\
---------------------------------------------------------------------------

    \18\ In response to another D.C. Circuit decision, Center for 
Energy and Economic Development v. EPA, 398 F.3d 653 (D.C. Cir. 
2005), EPA revised the RHR's provisions governing alternatives to 
source-specific BART determinations on October 13, 2006. These 
revisions did not alter the requirements for source-specific BART 
determinations that apply to Wyoming's BART determinations at issue 
here.
    \19\ 40 CFR Part 51, Appendix Y. While the Guidelines are only 
mandatory for fossil fuel-fired electric generating plants with a 
total generating capacity in excess of 750 megawatts, States are 
encouraged to follow the BART Guidelines in making BART 
determinations for other types of sources. Id. section I.H. The 
Guidelines also set specific presumptive limits for SO2 
and NOX for these large power plants, but allow states to 
apply more or less stringent limits based upon source-specific five-
factor analyses. 70 FR 39131-39132.
    \20\ Memo from Joseph W. Paise Regarding Regional Haze 
Regulations and Guidelines for BART (July 19, 2006); Additional 
Regional Haze Questions (Guidance) (Sept. 27, 2006). In addition, 
EPA issued final ``Guidance for Setting Reasonable Progress Goals 
Under the Regional Haze Program'' on June 1, 2007, but this Guidance 
is not directly relevant for individual BART determinations.
---------------------------------------------------------------------------

    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).
---------------------------------------------------------------------------

    \21\ See, e.g., 76 FR 36450 (Nevada); 77 FR 24794 (New York); 76 
FR 13944 (California); 77 FR 11798 (Rhode Island); 76 FR 27973 
(Delaware); 77 FR 12770 (Nebraska); 77 FR 18052 (Colorado); 76 FR 
16168 (Oklahoma); 77 FR 11914 (Vermont); 77 FR 11928 (Wisconsin); 76 
FR 52604 (Kansas); 76 FR 64186 (Arkansas); 77 FR 11839 (Maryland); 
76 FR 58570 (North Dakota); 77 FR 3966 (Illinois); 76 FR 76646 
(South Dakota). EPA proposed limited approval and limited 
disapproval of the Regional Haze SIPs of states covered by the Clean 
Air Interstate Rule (CAIR), due to the remand of CAIR by the D.C. 
Circuit. See, e.g. 77 FR 3691 (Jan. 25, 2012) (proposing limited 
approval and limited disapproval of Virginia's Regional Haze SIP).
    \22\ See, e.g., 76 FR 34608 (California); 76 FR 42557 
(Delaware); 76 FR 80754 (Kansas); 77 FR 19 (New Jersey); 77 FR 5191 
(District of Columbia); 77 FR 14604 (Arkansas); 77 FR 17334 
(Nevada); 77 FR 24845 (South Dakota); 77 FR 40150 (Nebraska); 77 FR 
51915 (New York).
---------------------------------------------------------------------------

    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).
---------------------------------------------------------------------------

    \23\ As presented elsewhere in this final notice and in the 
docket, the five NOX BART determinations we are 
disapproving are for the following: PacifiCorp Dave Johnston Unit 3, 
PacifiCorp Wyodak Unit 1, and Basin Electric Laramie River Units 1, 
2, and 3.
    \24\ As presented elsewhere in this final notice and in the 
docket, the four elements of the State SIP we are disapproving 
include: (1) Wyoming's RPGs; (2) The State's monitoring, 
recordkeeping, and reporting requirements in Chapter 6.4 of the SIP; 
(3) 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; and (4) 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.
---------------------------------------------------------------------------

    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

[[Page 5055]]

BART 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.
---------------------------------------------------------------------------

    \25\ 690 F.3d at 679, 682, 686.
---------------------------------------------------------------------------

    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

[[Page 5056]]

reasonableness 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

[[Page 5057]]

will 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
---------------------------------------------------------------------------

    \26\ 40 CFR part 51, appendix Y, section III.A.1.
---------------------------------------------------------------------------

    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:
---------------------------------------------------------------------------

    \27\ CAA section 169A(b)(2)(A), 42 U.S.C. 7491(b)(2)(A).

    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\
---------------------------------------------------------------------------

    \28\ 42 U.S.C. 7491(g)(2).

---------------------------------------------------------------------------
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\
---------------------------------------------------------------------------

    \29\ 40 CFR 51.308(e)(1)(ii)(A).

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.

[[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

[[Page 5059]]

should 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.
---------------------------------------------------------------------------

    \30\ As the commenter mentions, we agree that we did approve 
Wyoming's regional haze SIP submitted under Section 309 of the RHR 
(40 CFR 51.309) (77 FR 73926 (Dec. 12, 2012)), as in that action we 
determined the State met the requirements of 40 CFR 51.309 and 
related provisions.
---------------------------------------------------------------------------

    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

[[Page 5060]]

the 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

[[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.
---------------------------------------------------------------------------

    \31\ The conference agreement also revised the language ``except 
as otherwise provided pursuant to subsection (c), a requirement that 
each major stationary source (as defined in section 302(o)) which is 
in existence on the date of enactment of this section, but which has 
not been in operation for more than 15 years as of such date'' in 
H.R. 6161 to its present form. This revision does not affect any 
issue raised by the commenter.
---------------------------------------------------------------------------

    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.


[[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.
---------------------------------------------------------------------------

    \32\ In context, the statement regarding ``required progress'' 
must be understood to include BART. First, the preceding portion of 
the statement discusses States' roles in determining controls 
generally under 169A(b)(2), ``including'' the BART requirements in 
169A(b)(2)(A). The portion about EPA's supervisory role in assuring 
``required progress'' should be understood to apply to all of 
169A(b)(2), including subsection 169A(b)(2)(A).
---------------------------------------------------------------------------

    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

[[Page 5063]]

limitations 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''

[[Page 5064]]

as 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,

[[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.
---------------------------------------------------------------------------

    \33\ By statute, EPA's promulgation of a FIP must be upheld 
unless the court determines EPA's action was ``arbitrary, 
capricious, an abuse of discretion, or otherwise not in accordance 
with the law.'' CAA Section 307(d)(1)(B), (9)(A). There is no 
statutory standard of review governing EPA's disapproval of a SIP, 
however, the Supreme Court has held that where the Clean Air Act 
does not specific a standard for judicial review, ``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.'' ADEC 
at 496-97.
---------------------------------------------------------------------------

    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

[[Page 5066]]

was 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

[[Page 5067]]

Environmental 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

[[Page 5068]]

control 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

[[Page 5069]]

York 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.
---------------------------------------------------------------------------

    \34\ In contrast, elsewhere in the Act Congress has made it 
explicit that participation in state processes is required in order 
to raise objections with EPA. See CAA section 504(b)(2).
---------------------------------------------------------------------------

    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

[[Page 5070]]

Londoner 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.
---------------------------------------------------------------------------

    \35\ The cases cited by the commenter, 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, lack any reference to a 
notion of ``state substantive procedural rights'' or ``state-
sponsored procedural rights.'' The opinion in Virginia concerns the 
roles EPA and states play under the Act; the opinion does not 
discuss due process for owners of individual sources. The opinion in 
Michigan, noting that EPA's action had a rational basis, briefly 
dismisses a claim that the action violated the due process clause of 
the Fifth Amendment by discriminating against business and industry. 
Michigan, 805 F.2d at 185 n.1. Although the opinion does not make it 
explicit, the claim there thus seems to have been equal protection 
as incorporated into the Fifth Amendment, not procedural due 
process. See Bolling v. Sharpe, 347 U.S. 497 (1954). Michigan is not 
on point. Finally, the dicta in Bethlehem Steel speculates that, in 
the case of a FIP, ``EPA might have had to give interested persons 
an opportunity to submit oral as well as written comments,'' 
Bethlehem Steel, 742 F.2d at 1032, which EPA did in this case. The 
dicta, which in any case is not binding, does not say that EPA's 
experts must be available for cross-examination or that EPA is bound 
by state procedures or that the Fifth Amendment to the U.S. 
Constitution recognizes state ``substantive procedural rights.''
---------------------------------------------------------------------------

    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

[[Page 5071]]

procedures required by Congress in the CAA and EPA believes that no 
additional proceedings are warranted.
---------------------------------------------------------------------------

    \36\ See 42 U.S.C. 7607(d)(5).
    \37\ See 5 U.S.C. 553(c); see also U.S. v. Allegheny-Ludlum 
Steel Corp., 406 U.S. 742, 757 (1972).
    \38\ See 42 U.S.C. 7607(d)(5); see also Anaconda Co. v. 
Ruckelshaus, 482 F.2d 1301, 1306 (10th Cir. 1973).
    \39\ Vermont Yankee Nuclear Power Corp. v. Natural Res. Def. 
Council, Inc., 435 U.S. 519, 548 (1978).
---------------------------------------------------------------------------

    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\
---------------------------------------------------------------------------

    \40\ Kennecott Corp. v. EPA, 684 F.2d 1007, 1020 (D.C. Cir. 
1982).
    \41\ See H.R. Rep. No.95-564, 95th Cong. (1977).
---------------------------------------------------------------------------

    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.
---------------------------------------------------------------------------

    \42\ ``[I]dentification of the specific dictates of due process 
generally requires consideration of three distinct factors: First, 
the private interest that will be affected by the official action; 
second, the risk of an erroneous deprivation of such interest 
through the procedures used, and the probable value, if any, of 
additional or substitute procedural safeguards; and finally, the 
Government's interest, including the function involved and the 
fiscal and administrative burdens that the additional or substitute 
procedural requirement would entail.'' Eldridge, 424 U.S. at 335.
---------------------------------------------------------------------------

    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).
---------------------------------------------------------------------------

    \43\ LaChance v. Erickson, 522 U.S. 262, 266 (1988).
---------------------------------------------------------------------------

    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.
---------------------------------------------------------------------------

    \44\ Implementation Plans; Approvals, Disapprovals and 
Promulgations: Wyoming; Regional Haze State Implementation Plan; 
Federal Implementation Plan for Regional Haze June 10, 2013 Docket 
EPA-R08-OAR-2012-0026-0093.
    \45\ Air Quality State Implementation Plans; Approvals, 
Disapprovals and Promulgations: Wyoming; Regional Haze State 
Implementation Plan; Federal Implementation Plan for Regional Haze; 
Public Hearings Jul. 8, 2013 Docket EPA-R08-OAR-2012-0026-0098; see 
42 U.S.C. 7607(d)(5).
    \46\ Basin Electric Power Cooperative Comments, Aug. 6, 2013 
Docket EPA-R08-OAR-2012-0026-0058; Public Comment from Basin 
Electric Email Aug. 9, 2013 Docket EPA-R08-OAR-2012-0026-0148; 
Transcript from July 26, 2013 Hearings in Casper, Wyoming Aug. 8, 
2013 Docket EPA-R08-OAR-2012-0026-0108 pp. 48-83; Transcript from 
June 24, 2013 Hearings in Cheyenne, Wyoming Aug. 15, 2013 Docket 
EPA-R08-OAR-2012-0026-0100 pp. 62-67; Additional Public Comment from 
Basin Electric Laramie River Station BART CALPUFF Modeling Analysis 
Aug. 26, 2013 Docket EPA-R08-OAR-2012-0026-0227.
---------------------------------------------------------------------------

    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

[[Page 5072]]

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.''); 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.
---------------------------------------------------------------------------

    \47\ See BART Guidelines, 40 CFR Part 51, appendix Y, section 
IV.E.2.
    \48\ Central Arizona Water Conservation Dist. v. EPA, 990 F.2d 
1531, 1541 (9th Cir. 1993).
---------------------------------------------------------------------------

    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.
---------------------------------------------------------------------------

    \49\ 291 F.3d at 5-9.
    \50\ Id. at 7-8.
    \51\ EPA revised the RHR to address the court's decision in 
American Corn Growers at the same time as we promulgated the BART 
Guidelines. 70 FR 39104 (July 6, 2005). The revised RHR and the 
Guidelines were upheld by the D.C. Circuit in Utility Air Regulatory 
Group v. EPA, 471 F.3d 1333 (D.C. Cir. 2006).
---------------------------------------------------------------------------

    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

[[Page 5073]]

consider 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

[[Page 5074]]

Guidelines, 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

[[Page 5075]]

opposite 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

[[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.
---------------------------------------------------------------------------

    \52\ As explained elsewhere in this document, EPA has accepted 
some of the costs submitted in response to the proposed notice 
developed for Basin Electric, but not others.
---------------------------------------------------------------------------

    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

[[Page 5077]]

analysis 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.
---------------------------------------------------------------------------

    \53\ CAA section 169A(b) and 40 CFR 51.308(e)(1)(ii)(B).
---------------------------------------------------------------------------

    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

[[Page 5078]]

docket. 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

[[Page 5079]]

OFA.'' 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

[[Page 5080]]

result 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.

[[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\
---------------------------------------------------------------------------

    \54\ EPA-R08-OAR-2012-0026-0134, and EPA-R08-OAR-2012-0026-0068.
---------------------------------------------------------------------------

    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.

[[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.
---------------------------------------------------------------------------

    \55\ See CAA section 307(d).
    \56\ 78 FR 34738, and 78 FR 40654.
---------------------------------------------------------------------------

    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\
---------------------------------------------------------------------------

    \57\ 77 FR 72512, 72531 (Dec. 5, 2012)(BART for Apache, Cholla 
and Coronado).

    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\
---------------------------------------------------------------------------

    \58\ CCM (Tables 1.4 and 2.5 show AFUDC value as zero).
    \59\ See, e.g., 77 FR 20894, 20916-17 (Apr. 6, 2012) (explaining 
in support of the North Dakota Regional Haze FIP, ``we maintain that 
following the overnight method ensures equitable BART determinations 
. . .''); 76 FR 52388, 52399-400 (August 22, 2011) (explaining in 
the New Mexico Regional Haze FIP that the Manual does not allow 
AFUDC).

    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.

[[Page 5083]]

However, 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

[[Page 5084]]

Oklahoma 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).
---------------------------------------------------------------------------

    \60\ 40 CFR 51.302(e)(1)(ii)(B) (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.
---------------------------------------------------------------------------

    \61\ EPA is responsible for reviewing State-submitted SIPs and 
SIP revisions to ensure that they ``meet[ ] all of the applicable 
requirements of [the Act].'' CAA Section 110(k)(3); see also CAA 
Section 110(l) (EPA shall not approve SIP revision if it would 
interfere with ``any . . . applicable requirement of this 
chapter''); Oklahoma, 723 F.3d at 1204 (EPA reviews all SIPs to 
ensure plans comply with the Act). There is nothing unusual about 
regional haze SIPs in this regard--they, like any other SIPs, must 
be reviewed by EPA, and may be approved only if they meet all 
applicable requirements of the Act, including provisions related to 
visibility. See Oklahoma, 723 F.3d at 1207; North Dakota, 730 F.3d 
at 756-57.
---------------------------------------------------------------------------

    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.

[[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

[[Page 5086]]

with 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\
---------------------------------------------------------------------------

    \62\ The commenter cannot challenge EPA's duly promulgated 
regulations and Guideline. Indeed, the time for such a challenge has 
long passed, since the Guidelines were promulgated July 6, 2005, and 
could only have been challenged within 60 days. 70 FR 39,104; 42 
U.S.C. 7607(b), (d)(1)(J).
---------------------------------------------------------------------------

    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

[[Page 5087]]

determinations--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

[[Page 5088]]

CAA 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\
---------------------------------------------------------------------------

    \63\ 70 FR 39104, 39166.

    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.
---------------------------------------------------------------------------

    \64\ 70 FR 39104, 39168.
---------------------------------------------------------------------------

    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

[[Page 5089]]

usurp 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

[[Page 5090]]

capricious'' 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

[[Page 5091]]

plan 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.
---------------------------------------------------------------------------

    \65\ As we explain later in this document ``[t]hat 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.'' See 
response to comments in the modeling section for further 
information.
---------------------------------------------------------------------------

    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

[[Page 5092]]

the ``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

[[Page 5093]]

the 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.

[[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

[[Page 5095]]

Co. 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\
---------------------------------------------------------------------------

    \66\ 40 CFR 51.308(e)(1)(ii)(A).

    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.
---------------------------------------------------------------------------

    \67\ Commenters also suggest that, EPA has methodically changed 
or selectively ignored the requirements from those which were 
established in 40 CFR Part 51 and Appendix Y, which were published 
in 2005. The states' SIPs, written shortly after that period, were 
based on the rules and guidance provided at that time. Since then, 
however, EPA has arbitrarily and continually changed its 
interpretation of the regional haze regulations in order to achieve 
emission reductions and other objectives well beyond those allowed 
by the regional haze program. Here are a few examples of how EPA's 
position has changed over the past few years with respect to the 
guidance given for determining NOX BART controls.
    Appendix Y provides a presumptive BART NOX rate 
differentiated by boiler design and type of coal burned. EPA now 
requires post-combustion controls significantly more aggressive than 
the presumptive rates prescribed in Appendix Y. Appendix Y makes 
distinctions for unit size, with more aggressive controls targeted 
at the largest units. In Wyoming, EPA now proposes to require SCR on 
units as small as 160 megawatts. The preamble to the regional haze 
rules suggests that 75 percent of the electric generating units 
would have BART NOX controls cost between $100 and $1,000 
per ton. EPA is now imposing costs, based on its own calculations, 
of $3,700 to $6,000 per ton on 100 percent of PacifiCorp's Wyoming 
BART-eligible units.
    SCR controls were only expected to be cost-effective controls 
for cyclone units with high NOX emission rates. EPA is 
now proposing post-combustion NOX controls on every BART-
eligible unit in Wyoming, including the installation of eleven SCRs. 
EPA must stop changing its interpretations of the regional haze 
rules and guidelines that were formalized in 2005 and move ahead 
with approving the Wyoming BART analysis and the regional haze SIP 
which complies with those rules and guidelines.
---------------------------------------------------------------------------

    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

[[Page 5096]]

applied.'' 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

[[Page 5097]]

presumptive 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.
---------------------------------------------------------------------------

    \68\ Given the statutory mandate, a state may only avoid full 
consideration of the five statutory factors if an initial 
consideration demonstrates that further analysis is moot--for 
example, where the state demonstrates that the subject unit already 
employs the ``most stringent control available.'' 70 FR at 39165. 
Where these unique circumstances are not present, a state's failure 
to consider the five factors (for large EGUs, by complying with the 
BART Guidelines' five-step analysis) is grounds for disapproval. CAA 
Section 110(k)(3), (l); see Oklahoma, 723 F.3d at 1207-08.
---------------------------------------------------------------------------

    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

[[Page 5098]]

Wyoming. 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

[[Page 5099]]

regional 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

[[Page 5100]]

year 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

[[Page 5101]]

13211(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\
---------------------------------------------------------------------------

    \69\ 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.
    \70\ For example, EO 12866 (Sec. 10, Judicial review) explicitly 
states, ``This Executive order is intended only to improve the 
internal management of the Federal Government and does not create 
any right or benefit, substantive or procedural, enforceable at law 
or equity by a party against the United States, its agencies or 
instrumentalities, its officers or employees, or any other person.''
---------------------------------------------------------------------------

    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

[[Page 5102]]

adversely 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

[[Page 5103]]

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. Please refer elsewhere for a detailed 
analysis of the elements required by our regulations for BART 
determinations.
---------------------------------------------------------------------------

    \71\ Results of visitor surveys from 22 studies demonstrate that 
clean air and scenic vistas in national parks consistently rank as 
the top priorities of 90 percent or more of visitors. ``National 
Park Service Visitor Values & Perceptions of Clean Air, Scenic Views 
& Dark Night Skies 1988-2011,'' Natural Resource Report NPS/NRSS/
ARD/NRR-2013/632 (Feb. 2013), pp. 16-23 (including specific studies 
focused on Wyoming's national parks). Additionally, as explained in 
recent National Park Service (NPS) and U.S. Forest Service (USFS) 
reports, the national parks and U.S. Forest Service lands are 
important economic engines for local communities and businesses, 
with visitors generating significant economic activity and support 
thousands of jobs. See, ``Economic Benefits to Local Communities 
from National Park Visitation, 2011,'' National Park Service, 
Natural Resource Report NPS/NRSS/ARD/NRR-2013/632 (Feb. 2013) (NPS 
2013 Report); and ``National Visitor Use Monitoring Results, 
National Summary Report,'' USDA Forest Service (May 20, 2013 
update). In 2011, the NPS report indicates that non-local visitor 
spending to the national parks in Wyoming was $621 million, creating 
more than 9,000 jobs. NPS 2013 Report, p. 26. The Grand Teton 
National Park alone had more than 2.5 million visitors who spent 
$463 million which created more than 6,000 jobs. NPS 2013 Report, p. 
19.
---------------------------------------------------------------------------

    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

[[Page 5104]]

Guidelines. 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

[[Page 5105]]

EPA'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

[[Page 5106]]

deadline, 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.
---------------------------------------------------------------------------

    \72\ January 2, 2011 a Complaint was filed against EPA in the 
federal district court for the District of Colorado seeking 
declaratory and injunctive relief under the Clean Air Act due to 
EPA's failure to meet regulatory and statutory deadlines for the 
regional haze implementation plan. See WildEarth Guardians v. 
Jackson, Case No. 11-cv-0001-CMA-MEH (D. Colo.).
    \73\ The EPA resolved this complaint by means of a settlement 
agreement that was memorialized in a Consent Decree that was entered 
by the Court on June 6, 2011 (the ``Consent Decree'').
---------------------------------------------------------------------------

    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.
---------------------------------------------------------------------------

    \74\ EPA routinely notifies the states of these extensions. For 
example, on December 11, 2012, Region 8 Air Program Director Carl 
Daly called Wyoming DEQ's Steve Dietrich and his staff, to let them 
know the Agency had submitted a motion to the Court to re-propose 
the 309(g) rule. Mr. Daly also contacted Mr. Dietrich on March 26, 
2013 to let the State know EPA was working on further extensions, as 
the Agency needed additional time to consult with our headquarters 
offices.
---------------------------------------------------------------------------

    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

[[Page 5107]]

the 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.
---------------------------------------------------------------------------

    \75\ Additionally, the CAA defines the term ``as expeditiously 
as practicable'' to mean ``as expeditiously as practicable but in no 
event later than five years after the date of approval of a plan 
revision under this section (or the date of promulgation of such a 
plan revision in the case of action by the Administrator under 
section 110(c) [42 USCS Section 7410(c)] for purposes of this 
section).'' 42 U.S.C. 7491(g)(4).
---------------------------------------------------------------------------

    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

[[Page 5108]]

BART 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.
---------------------------------------------------------------------------

    \76\ Guidance for Estimating Natural Visibility Conditions Under 
the Regional Haze Rule, Document No. EPA-R09-OAR-2012-0345-0003-B9, 
U.S. Environmental Protection Agency, September 2003. http://www.epa.gov/ttncaaa1/t1/memoranda/rh_envcurhr_gd.pdf, page 1-1: 
``Natural visibility conditions represent the long-term degree of 
visibility that is estimated to exist in a given mandatory Federal 
Class I area in the absence of human-caused impairment. It is 
recognized that natural visibility conditions are not constant, but 
rather they vary with changing natural processes (e.g., windblown 
dust, fire, volcanic activity, biogenic emissions). Specific natural 
events can lead to high short-term concentrations of particulate 
matter and its precursors. However, for the purpose of this guidance 
and implementation of the regional haze program, natural visibility 
conditions represents a long-term average condition analogous to the 
5-year average best- and worst-day conditions that are tracked under 
the regional haze program.''
    \77\ Ibid., pages 3-1 to 3-4.
---------------------------------------------------------------------------

    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

[[Page 5109]]

modeling has demonstrated that Wyoming's BART sources are also 
significant contributors to regional haze.
---------------------------------------------------------------------------

    \78\ 78 FR at 34767 (``Table 28 also shows that Wyoming is not 
meeting the URP to meet natural visibility conditions by 2064. In 
this case, 40 CFR 51.308(d)(1)(ii) requires the State to 
demonstrate, based on the four factors in 51.308(d)(1)(i)(A), that 
the RPGs established in this SIP are reasonable for this planning 
period and that achieving the URP in this planning period is not 
reasonable. In its demonstration, the State cited many reasons why 
meeting the URP was not reasonable, including the following. First, 
emissions from natural sources greatly affect the State's ability to 
meet the 2018 URP. As discussed earlier, WEP data shows that 
emissions of OC, EC, PM2.5, and PM10 come 
mainly from natural or non-anthropogenic sources, such as natural 
wildfire and windblown dust.'')
---------------------------------------------------------------------------

    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.
---------------------------------------------------------------------------

    \79\ 40 CFR 39170.
---------------------------------------------------------------------------

    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

[[Page 5110]]

our 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\
---------------------------------------------------------------------------

    \80\ 70 FR at 39123.
    \81\ Assessment of the ``VISTAS'' Version of the CALPUFF 
Modeling System, EPA-454/R-08-007, August 2008; also see CALPUFF 
Model Change Bulletins B (MCB-B), MCB-C and MCB-D.
---------------------------------------------------------------------------

    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\
---------------------------------------------------------------------------

    \82\ 70 FR at 39121
    \83\ 70 FR at 39121-29124.
---------------------------------------------------------------------------

    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.
---------------------------------------------------------------------------

    \84\ 70 FR at 39123.
---------------------------------------------------------------------------

    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 Ni[ntilde]o'' and 
``la Ni[ntilde]a''. These events have been monitored and quantified 
since 1950 and occurrences are sub-categorized as weak, moderate and 
strong. The periodicity of strong events

[[Page 5111]]

for both ``el Ni[ntilde]o'' and ``la Ni[ntilde]a'' is every 9-11 years. 
The last strong ``el Ni[ntilde]o'' occurred in 1997. The last two 
strong ``la Ni[ntilde]a'' 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 
Ni[ntilde]o and la Ni[ntilde]a 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 Ni[ntilde]o and la Ni[ntilde]a 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 
Ni[ntilde]o and la Ni[ntilde]a 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.
---------------------------------------------------------------------------

    \85\ ``Guidance for Estimating Natural Visibility Conditions 
under the Regional Haze Rule'' Document No. EPA-R09-OAR-2012-0345-
0003-B9, U.S. Environmental Protection Agency, September, 2003.
    \86\ 70 FR at 39123.
---------------------------------------------------------------------------

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\
---------------------------------------------------------------------------

    \87\ Li, Y., Schwandner, F.M., Sewell, H.J., Zivkovich, A., 
Tigges, M., Raja, S., Holcomb, S., Molenar, J.V., Sherman, L., 
Archuleta, C., Lee, T., Collett Jr., J.L., Observations of ammonia, 
nitric acid, and fine particles in a rural gas production region, 
Atmospheric Environment (2013), doi: 10.1016/j.atmosenv.2013.10.007.
    \88\ Chen et al., A Pilot Monitoring Study of Atmospheric NHx at 
Selected IMPROVE sites AWMA Aerosol and Atmospheric Optics, 
Visibility & Air Pollution Conference, September 24-28, 2012, 
Whitefish, MT.
---------------------------------------------------------------------------

    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.
---------------------------------------------------------------------------

    \89\ Wyoming's ``BART Air Modeling Protocol'' (Protocol) is 
included in the docket in the State's Technical Support Document.
---------------------------------------------------------------------------

    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\
---------------------------------------------------------------------------

    \90\ Air Quality Modeling Protocol: Wyoming Regional Haze 
Federal Implementation Plan, U.S. EPA, January, 2014.
    \91\ Modeling Protocol: Montana Regional Haze Federal 
Implementation Plan (FIP) Support, prepared for EPA Region 8 by 
Alpine Geophysics, LLC, November 21, 2011.

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

[[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.
---------------------------------------------------------------------------

    \92\ Interagency Workgroup On Air Quality Modeling (IWAQM) Phase 
2 Summary Report And Recommendations For Modeling Long Range 
Transport Impacts (EPA-454/R-98-019), EPA OAQPS, December 1998, 
http://www.epa.gov/scram001/7thconf/calpuff/phase2.pdf.
---------------------------------------------------------------------------

    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\
---------------------------------------------------------------------------

    \93\ Langford, A.O., F.C. Fehsenfeld, J. Zachariassen, and D.S. 
Schimel (1992), Gaseous ammonia fluxes and background concentrations 
in terrestrial ecosystems of the United States, Global 
Biogeochemical Cycles, 6, 459-483.
    \94\ Interagency Workgroup On Air Quality Modeling (IWAQM) Phase 
2 Summary Report And Recommendations For Modeling Long Range 
Transport Impacts, EPA-454/R-98-019, (December 1998) pages 14-15.
---------------------------------------------------------------------------

    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.
---------------------------------------------------------------------------

    \95\ IWAQM, Ibid., page 6.
---------------------------------------------------------------------------

    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.
---------------------------------------------------------------------------

    \96\ Li, Y., Schwandner, F.M., Sewell, H.J., Zivkovich, A., 
Tigges, M., Raja, S., Holcomb, S., Molenar, J.V., Sherman, L., 
Archuleta, C., Lee, T., Collett Jr., J.L., Observations of ammonia, 
nitric acid, and fine particles in a rural gas production region, 
Atmospheric Environment (2013), doi: 10.1016/j.atmosenv.2013.10.007.
    \97\ Li et al. U.S. EPA, December, 2013.
    \98\ Air Quality Modeling Protocol: Wyoming Regional Haze 
Federal Implementation Plan.
---------------------------------------------------------------------------

    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\
---------------------------------------------------------------------------

    \99\ Wyoming Protocol, p. 12.
---------------------------------------------------------------------------

    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.
---------------------------------------------------------------------------

    \100\ Wyoming Protocol, p. 15.
---------------------------------------------------------------------------

    We describe here in more detail the ambient monitoring data from 
the site in western Wyoming and our use of those

[[Page 5113]]

data. 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.
---------------------------------------------------------------------------

    \101\ Li, Y., Schwandner, F.M., Sewell, H.J., Zivkovich, A., 
Tigges, M., Raja, S., Holcomb, S., Molenar, J.V., Sherman, L., 
Archuleta, C., Lee, T., Collett Jr., J.L., Observations of ammonia, 
nitric acid, and fine particles in a rural gas production region, 
Atmospheric Environment (2013), doi: 10.1016/j.atmosenv.2013.10.007.
    \102\ Li et al., Ibid.
    \103\ Air Quality Modeling Protocol: Wyoming Regional Haze 
Federal Implementation Plan,U.S. EPA, December, 2013, page 18, Table 
7.
---------------------------------------------------------------------------

    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.
---------------------------------------------------------------------------

    \104\ Protocol for BART-Related Visibility Impairment Modeling 
Analyses in North Dakota, November, 2005, North Dakota Department of 
Health, Division of Air Quality, 1200 Missouri Avenue, Bismarck, ND 
58506. Page 33.
    \105\ Chen et al., A Pilot Monitoring Study of Atmospheric 
NHX at Selected IMPROVE sites AWMA Aerosol and 
Atmospheric Optics, Visibility & Air Pollution Conference, September 
24-28, 2012, Whitefish, MT.
---------------------------------------------------------------------------

    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.
---------------------------------------------------------------------------

    \106\ Modeling Protocol: Montana Regional Haze Federal 
Implementation Plan (FIP) Support, prepared for EPA Region 8 by 
Alpine Geophysics, LLC, November 21, 2011.
---------------------------------------------------------------------------

    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

[[Page 5114]]

determination 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.
---------------------------------------------------------------------------

    \107\ 77 FR 39123: ``Because of the scale of the predicted 
impacts from these sources, CALPUFF is an appropriate or a 
reasonable application to determine whether such a facility can 
reasonably be anticipated to cause or contribute to any impairment 
of visibility. In other words, to find that a source with a 
predicted maximum impact greater than 2 or 3 deciviews meets the 
contribution threshold adopted by the States does not require the 
degree of certainty in the results of the model that might be 
required for other regulatory purposes. In the unlikely case that a 
State were to find that a 750 MW power plant's predicted 
contribution to visibility impairment is within a very narrow range 
between exemption from or being subject to BART, the State can work 
with EPA and the FLM to evaluate the CALPUFF results in combination 
with information derived from other appropriate techniques for 
estimating visibility impacts to inform the BART applicability 
determination. Similarly for other types of BART eligible sources, 
States can work with the EPA and FLM to determine appropriate 
methods for assessing a single source's impacts on visibility.''
---------------------------------------------------------------------------

    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.
---------------------------------------------------------------------------

    \108\ 70 FR at 39123.
---------------------------------------------------------------------------

    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

[[Page 5115]]

that ``[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.
---------------------------------------------------------------------------

    \109\ 70 FR at 39123.
---------------------------------------------------------------------------

    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.
---------------------------------------------------------------------------

    \111\ Guidelines for Ensuring and Maximizing the Quality, 
Objectivity, Utility, and Integrity of Information Disseminated by 
the Environmental Protection Agency, EPA/260R-02-008 October 2002, 
U.S. Environmental Protection Agency Office of Environmental 
Information (2810) 1200 Pennsylvania Avenue NW., Washington, DC 
20460.
---------------------------------------------------------------------------

    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-[agrave]-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

[[Page 5116]]

visibility 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\
---------------------------------------------------------------------------

    \112\ ``CALPUFF Regulatory Update'', Roger W. Brode, 
Presentation at Regional/State/Local Modelers Workshop, June 10-12, 
2008; http://www.cleanairinfo.com/regionalstatelocalmodelingworkshop/archive/2008/agenda.htm.
---------------------------------------------------------------------------

    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

[[Page 5117]]

with 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.
---------------------------------------------------------------------------

    \113\ Li et al. 2013
---------------------------------------------------------------------------

    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.
---------------------------------------------------------------------------

    \114\ Langford et al.
    \115\ For example, see EPA guidance documents that discuss 
methods for estimating background NO2 concentrations: ``Additional 
Clarification Regarding Applicability of Appendix W Modeling 
Guidance for the 1-hour NO2 NAAQS''
---------------------------------------------------------------------------

    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

[[Page 5118]]

rural and remote areas of central Colorado or western Wyoming.
---------------------------------------------------------------------------

    \116\ Supplemental BART Analysis CALPUFF Protocol for Class I 
Federal Area Visibility Improvement Modeling Analysis, DRAFT, 
revised Aug 19, 2010, Colorado Department of Public Health and 
Environment. (CDPHE) Air Pollution Control Division Technical 
Services Program, 4300 Cherry Creek Drive South, Denver, Colorado 
80246, pages 26-33.
---------------------------------------------------------------------------

    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

[[Page 5119]]

monitoring 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\
---------------------------------------------------------------------------

    \117\ Li et al., 2013.
    \118\ Interagency Workgroup On Air Quality Modeling (IWAQM) 
Phase 2 Summary Report And Recommendations For Modeling Long Range 
Transport Impacts (EPA-454/R-98-019), EPA OAQPS, December 1998.
    \119\ Tonnesen, G., Wang, Z., Morris, R., Hoats, A., Jia, Y., 
Draft Final Modeling Protocol, CALMET/CALPUFF Protocol for BART 
Exemption Screening Analysis for Class I Areas in the Western United 
States, Submitted to the Western Regional Air Partnership, August 
15, 2006.
---------------------------------------------------------------------------

    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.
---------------------------------------------------------------------------

    \120\ Federal Land Managers' Air Quality Related Values Work 
Group (FLAG) Phase I Report--Revised (2010).
    Natural Resource Report NPS/NRPC/NRR--2010/232.
---------------------------------------------------------------------------

    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.
---------------------------------------------------------------------------

    \121\ Chapter 6, Section 9(d)(i)(C) of the Wyoming Air Quality 
Standards and Regulations.
---------------------------------------------------------------------------

    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

[[Page 5120]]

invalid. 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

[[Page 5121]]

the 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.
---------------------------------------------------------------------------

    \122\ W.H. White, R.J. Farber, W.C. Malm, M. Nuttall, M.L. 
Pitchford, B.A. Schichtel, Comment on ``Effect of coal-fired power 
generation on visibility in a nearby National Park (Terhorst and 
Berkman, 2010)'', Atmospheric Environment 55 (2012) 173-178. 
doi:10.1016/j.atmosenv.2012.02.076.
---------------------------------------------------------------------------

    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

[[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

[[Page 5123]]

Laramie 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--

[[Page 5124]]

neither 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.
---------------------------------------------------------------------------

    \123\ Wyoming Department of Environmental Quality Air Quality 
Division BART Application Analysis AP-6047, Table 16, May 28, 2009.
    \124\ Air Quality Modeling Protocol: Wyoming Regional Haze 
Federal Implementation Plan, U.S. EPA, Table B.9, January, 2014.
---------------------------------------------------------------------------

    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

[[Page 5125]]

commenter 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.
---------------------------------------------------------------------------

    \125\ Levine et al., (1980) The Vertical Distribution of 
Tropospheric Ammonia, Geophys. Res. Letters, vol. 7, No. 5, 17-32.
---------------------------------------------------------------------------

    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.
---------------------------------------------------------------------------

    \126\ Chen et al., A Pilot Monitoring Study of Atmospheric 
NHX at Selected IMPROVE sites AWMA Aerosol and 
Atmospheric Optics, Visibility & Air Pollution Conference, September 
24-28, 2012, Whitefish, MT.
---------------------------------------------------------------------------

    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.

[[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\
---------------------------------------------------------------------------

    \127\ CDPHE, Ibid.
    \128\ IWAQM, page 14 and page 21
---------------------------------------------------------------------------

    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

[[Page 5127]]

consistent 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.
---------------------------------------------------------------------------

    \129\ IWAQM, p.18.
---------------------------------------------------------------------------

    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

[[Page 5128]]

beyond 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.
---------------------------------------------------------------------------

    \130\ http://www.epa.gov/scram001/reports/EPA-454-R-12-003.pdf.
    \131\ http://www.epa.gov/scram001/7thconf/calpuff/phase2.pdf, 
page 18.
    \132\ http://www.epa.gov/scram001/reports/EPA-454-R-12-003.pdf, 
page 10.
---------------------------------------------------------------------------

    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 I areas from implementation of all of the BART 
determinations in its 2013 re-proposal. To assess this shortcoming, the 
Conservation Organizations contracted with Howard Gebhart to conduct a 
cumulative visibility improvement modeling analysis that compared 
installation of the NOX BART determinations found in EPA's 
2013 re-proposal versus the State BART determinations found in the 
Wyoming Regional Haze SIP. See Gebhart Report, at 17-24. Mr. Gebhart's 
visibility modeling results show that installation of the BART 
determinations in EPA's 2013 re-proposal will result in significant 
visibility improvement at numerous Class I areas when compared to the 
Wyoming SIP. For example, installation of the BART determinations in 
EPA's 2013 re-proposal would consistently result a total deciview 
improvement of 1.0 deciview or greater over the Wyoming SIP at Badlands 
National Park, Savage Run Wilderness, and Wind Cave National Park. In 
addition, significant visibility improvements exceeding 0.5 deciviews 
were predicted at Badlands National Park, Bridger Wilderness, Mount 
Zirkel Wilderness, Rawah Wilderness, Rocky Mountain National Park, 
Savage Run Wilderness, and Wind Cave National Park. In summary, the 
Conservation Organizations' cumulative visibility improvement modeling 
analysis provides further support that significant visibility benefits 
can be achieved from the finalization of the BART determinations 
contained in EPA's 2013 re-proposal. EPA's 2013 re-proposed rule 
advances (without entirely fulfilling) the goals of the regional haze 
program to reduce visibility impairment using BART during the first 
regional haze five-year planning period. In contrast, the Wyoming 
Regional Haze SIP would fall far short of these goals.
    Response: First, we note that the modeling performed by the 
Conservation Organizations' contractor used the 2 ppb background 
ammonia concentration, and did not correct the model results for 
ammonia limiting conditions, and therefore predicts greater visibility 
impairment than did EPA's revised modeling. EPA provided information 
about the visibility improvement modeled for different BART scenarios 
at multiple Class I areas within 300 km of each BART source. EPA 
primarily relied on the benefits at the area with the greatest 
visibility improvement from controls, but we also considered the 
cumulative impacts and benefits at multiple Class I areas. EPA agrees 
that considering cumulative visibility benefits by aggregating the 
expected improvement from over multiple Class I areas is a useful 
metric that can further inform a BART determination. Such an approach 
can be useful, for example, in simplifying a complex array of 
visibility impacts, especially where a source has significant impacts 
on multiple Class I areas.
    Comment: EPA's proposed rule fails to present the cumulative 
visibility benefits of installation of SCR at Wyoming's EGUs. Instead, 
EPA only presents the visibility benefits for a single Class I area per 
source (Wind Cave National Park for all sources except the Jim Bridger 
plant (Mount Zirkel Wilderness Area)). The cumulative impact of a 
source's emissions on visibility as well as the cumulative benefit of 
emission reductions is a necessary consideration as part of the fifth-
step in the BART analysis. The statutory direction and goal of the 
regional haze program is to remedy any existing impairment of 
visibility in mandatory Class I areas. 42 U.S.C. 7491(1). The 
implementing regulations plainly anticipate the need to reduce impacts 
in multiple Class I areas, including those outside a state's borders, 
and the obligation to assess what is necessary to do so. 40 CFR 
51.308(d)(3). Further, states are required to establish reasonable 
progress goals for each Class I area, not just the one most impacted by 
a single source. Id. section 51.308(d)(1). EPA's own regional haze 
guidance document states that a cumulative visibility benefit analysis 
is generally consistent with the CAA. 70 FR 39105, 39107 (we believe 
that a State's decision to use a cumulative analysis at the eligibility 
stage is consistent with the CAA); 40 CFR Part 51, App Y. While the 
Guidelines also contemplate and even allow analysis of only the most 
impacted Class I area, such an analysis contradicts the regional 
approach towards the restoration of visibility. Moreover, given the 
number of Class I areas impacted by Wyoming sources, it is illogical 
and baseless to fictitiously limit the spectrum of source impact and 
emission control benefit. Based upon the guidance and the requirements 
of the CAA, the cumulative impact of a source's

[[Page 5129]]

emissions on visibility, as well as the cumulative benefit of emission 
reductions, should be considered as part of the fifth-step in the BART 
analysis. The FLMs, too, have urged EPA Region 8 to consider the 
cumulative visibility benefits of requiring stricter controls on BART-
eligible units in Montana. For example, at a public meeting in 
Billings, regarding the Montana Regional Haze SIP, Valerie Naylor, 
Superintendent of Theodore Roosevelt National Park stated, ``EPA placed 
too much emphasis on incremental costs and incremental benefits, while 
eliminating consideration of cumulative benefits that would be realized 
in the numerous Class I National Parks, National Wildlife Refuges, and 
Wilderness Areas impacted by Colstrip.'' The National Park Service 
(NPS) has consistently requested that cumulative visibility benefit 
analyses be conducted in other regional haze determinations. In 
addition, EPA must consider the cumulative visibility benefit of BART 
controls on multiple units of a single source. EPA's BART guidelines 
make clear that states must consider emissions from an entire source in 
determining whether a source is subject to BART, and further clarify 
that multiple units at a single utility constitute a single source. 40 
CFR Part 51, App Y, sect. II.A.
    The Conservation Organizations retained Air Resource Specialists, 
Inc. (ARS) to evaluate the cumulative visibility impact of 
NOX BART controls, and found that the cumulative benefit of 
SCR at all Wyoming BART-subject EGUs is very significant. In conducting 
its supplementary modeling, ARS used an SCR-controlled NOX 
emission rate to 0.05 lb/MMBtu to reflect the level of control 
achievable with SCR and recalculated baseline emissions to comply with 
the BART guidelines, as described in sections I.A.2 and I.C. Otherwise, 
ARS employed the same assumptions used by EPA in its analysis.
    ARS's visibility modeling addresses impacts to 18 Class I areas, 
including Savage Run Wilderness Area (which is not a mandatory Class I 
area but is managed as such by Wyoming). The ARS report addresses the 
cumulative benefit of installation of SCR at multiple units at a single 
power plant location (ex. the cumulative benefits of installation of 
SCR at all four Bridger units). The ARS Report also calculates the 
cumulative visibility benefit of installation of SCR on all BART units 
in Wyoming. Id. It should not be assumed that ARS's results document 
the highest impacts. Rather, they are presented to demonstrate 
widespread and far-reaching visibility impacts and improvements that 
can be achieved through the use of SCR.
    The cumulative visibility benefit from installation of SCR on all 
BART units in Wyoming is significant. The application of SCR control on 
Wyoming's subject-to-BART emission units is predicted to improve worst-
case visibility impairment by up to 8 deciviews at the Savage Run Class 
I area, with 4 deciviews of improvement or better at six Class I areas. 
ARS Report, Table 3-13. For the 98th percentile day, the improvement 
after SCR emissions control at all Wyoming BART-subject EGUs is as high 
as 3.5 deciviews at Wind Cave National Park. Id. At least six different 
Class I areas show improvement of 3 deciviews or more based on the 98th 
percentile day after SCR emissions control at all Wyoming BART-subject 
EGUs. Id.
    SCR controls at Wyoming's subject-to-BART units are also predicted 
to significantly reduce the number of days with visibility impacts 
above 0.5 deciview and 1.0 deciview compared to baseline emissions 
scenario. Over all 18 Class I areas modeled, the cumulative improvement 
from application of SCR on all Wyoming BART-subject EGUs is 721 fewer 
days with visibility impairment exceeding 0.5 deciview and 595 fewer 
days with visibility impairment exceeding 1.0 deciview. Id., Table 3-
14. These improvements are relatively uniformly distributed across the 
seven Class I areas most impacted by Wyoming's subject-to-BART EGUs: 
Badlands National Park, Bridger Wilderness Area, Mt. Zirkel Wilderness 
Area, Rawah Wilderness Area, Rocky Mountain National Park, Savage Run 
Wilderness Area, and Wind Cave National Park.
    Response: As described in another response, EPA did not limit its 
analysis of visibility impairment to a single Class I area. We 
evaluated visibility impairment from each BART source at multiple Class 
I areas. We presented the results for each Class I area, and we 
considered the visibility impairment at multiple Class I areas in our 
BART determination. The estimates of visibility impairment presented by 
the commenter relied on an overly conservative estimates of background 
ammonia concentrations, and therefore are likely to overestimate 
cumulative visibility benefits.
    In regard to the comment concerning the cumulative visibility 
benefit of BART controls on multiple units at a single source, see the 
response to a subsequent comment below.
    Comment: It is appropriate to consider both the degree of 
visibility improvement in a given Class I area as well as the 
cumulative effects of improving visibility across all of the Class I 
areas affected. If reducing emissions from a BART source impacts 
multiple Class I areas, then a BART determination should incorporate 
those benefits. It is not justified to evaluate impacts at one Class I 
area, while ignoring others that are similarly significantly impaired 
by the BART source. If emissions from the BART source are reduced, the 
benefits will be spread well beyond only the most-impacted Class I 
area, and these benefits are an integral part of the BART 
determination. The BART Guidelines attempt to create a workable 
approach to estimating visibility impairment. The Guidelines do not 
attempt to address the geographic extent of the impairment, but in 
effect assume that all Class I areas are created equal, i.e., 
widespread impacts in a large Class I area and isolated impacts in a 
small Class I area are given equal weight for BART determination 
purposes. To address the problem of geographic extent, we look at the 
cumulative impacts of a source on all Class I areas affected, as well 
as the cumulative benefits from reducing emissions. While there may be 
more sophisticated approaches to this problem, we believe that this is 
the most practical, given current modeling techniques and information 
available.
    Response: Contrary to the commenter's assertion, we did assess 
cumulative visibility impacts for multiple Class I areas. In our 
analysis of visibility impacts, we considered the visibility 
improvement at multiple Class I areas within the 300 kilometers of the 
modeling domain. For example, in our analysis of BART control options 
for Naughton, we considered the visibility improvement at seven Class I 
areas (Bridger Wilderness Area, Fitzpatrick Wilderness Area, Grand 
Teton National Park, North Absaroka Wilderness Area, Teton Wilderness 
Area, Washakie Wilderness Area, and Yellowstone National Park).
    Therefore, our proposed rule 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 assertions that we did 
not consider the impacts at multiple Class I areas. In the proposed 
rule, we did however focus on the visibility benefits at the most 
impacted Class I area.
    Comment: EPA has incorrectly estimated visibility improvement from 
all NOX control options at the Laramie River Station. 
Wyoming DEQ evaluated visibility improvements at the two nearest Class 
I areas and reported the

[[Page 5130]]

``The cumulative visibility improvement for SCR, as compared to LNB/
OFA, across Wind Cave National Park and Badlands National Park (based 
on the 98th percentile modeled results) was 0.52-0.54 delta deciview 
for each of the three units.'' EPA R8 evaluated the five closest Class 
I areas but reported results for only the Wind Cave National Park.
    Response: As described in a previous response, in our analysis of 
visibility impacts, we considered the visibility improvement at four 
Class I areas within 300 kilometers of Laramie River. Modeling results 
for all Class I areas considered for each BART source for the re-
proposal were available to the public during the comment period upon 
request. (See ``Summary of EPA's Additional Visibility Improvement 
Modeling''). Therefore, our proposed rule 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 assertions that we did 
not consider the impacts at multiple Class I areas. In the proposed 
rule, we did however focus on the visibility benefits at the most 
impacted Class I area.
    Comment: EPA rejected Oklahoma's visibility analyses which ``relied 
upon pollutant specific modeling to evaluate the benefits from the use 
of available SO2 emission controls.'' 76 FR 81728, 81740. 
Rather, EPA modeled in Oklahoma ``all visibility impairing pollutants 
to fully assess the visibility improvement anticipated from the use of 
controls.'' EPA argued this modeling took into account ``the complexity 
of atmospheric chemistry and chemical transformation among 
pollutants.'' In Wyoming, EPA noted that Wyoming provided ``visibility 
improvement modeling results that combine[d] the visibility improvement 
from NOX, PM and SO2 control options'' and that 
``EPA could not ascertain what the visibility improvement would be from 
an individual NOX or PM control option.'' 77 FR 33031. EPA 
appears to take contrary positions in Oklahoma and Wyoming. EPA's 
inconsistent positions are arbitrary and capricious.
    Response: As described in a response to a previous comment, it 
appears that the commenter has confused (1) whether all pollutants were 
modeled together; and (2) whether all control technologies were 
modeled. 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. The 
additional modeling performed by EPA was designed to evaluate 
visibility improvements from certain emissions reduction technologies. 
Each of these simulations also included all other visibility impairing 
pollutants, so the approach used by EPA in Wyoming and Oklahoma is 
consistent.
    Comment: We are concerned about the emissions modeled by EPA as 
presented in the ``Summary of EPA's Additional Visibility Improvement 
Modeling.'' For example, sulfuric acid mist 
(H2SO4) emissions from each PacifiCorp unit are 
assumed to double from the baseline and control scenarios that do not 
include SCR versus scenarios with SCR. The only explanation provided by 
EPA is that ``the emission rate for . . . total sulfate rates were 
increased to account for the additional production that results from 
SCR controls.'' EPA's approach in Wyoming is not consistent with its 
approach elsewhere. For example, in its modeling analysis of addition 
of SCR at Colstrip Units 1 and 2 in Montana, EPA assumed no additional 
sulfate emissions from the addition of SCR.
    Because H2SO4 must be reported as a hazardous 
air pollutant, the Electric Power Research Institute has developed a 
widely-accepted method for estimating those emissions. Our analyses 
indicate a two-orders-of-magnitude overestimation by EPA of these 
visibility-impairing emissions, which results in an underestimation of 
the visibility benefit of adding SCR.
    Response: While the method established by the Electric Power 
Research Institute may yield more accurate H2SO4 
emission rates, we have not found, and the commenter has not 
substantiated, that our treatment of H2SO4 led to 
meaningfully different modeled visibility improvement, or for that 
matter, influenced the BART determination in a material manner. In the 
modeling conducted by EPA, we set the sulfuric acid emission rates 
equal to those in the State's modeling analyses which typically doubled 
the H2SO4 emission rate between the baseline and 
SCR modeling scenarios. In comparison to the emission rates for 
SO2 and NOX, the emission rates for 
H2SO4 were trivial. For example, consider Dave 
Johnston Unit 3, where the modeled emission rates for SO2 
and NOX in the baseline scenario were 420.0 lbs/hr and 
1671.0 lbs/hr, respectively, while the modeled emission rate for 
H2SO4 was 2.6 lbs/hr. Here, in comparison to 
SO2 and NOX emissions, the emissions rate of 
H2SO4 is clearly insignificant and would have a 
limited impact on modeled visibility. The same can be said for the SCR 
scenario where the modeled emission rates for SO2 and 
NOX were 420.0 lbs/hr and 163.3 lbs/hr, respectively, while 
the modeled emission rate for H2SO4 was 5.1 lbs/
hr. In short, the H2SO4 emission rates used in 
the modeling were so low that it is apparent that they have no more 
than a negligible impact on the modeled visibility improvement.
    Comment: EPA must consider the cumulative visibility benefit of 
BART controls on multiple units of a single source. EPA's BART 
guidelines make clear that states must consider emissions from an 
entire source in determining whether a source is subject-to-BART, and 
further clarify that multiple units at a single utility constitute a 
single source. 40 CFR part 51, App Y, sect. II.A. This is not by 
accident or oversight. As EPA stated in its preamble to the BART 
Guidelines, ``[a]pplying de minimis levels on a unit by unit basis . . 
. could exempt hundreds of tons of emissions of a visibility-impairing 
pollutant from BART analysis. [I]t is possible that while emissions 
from each unit are relatively trivial, the costs of controlling 
emissions from multiple units might be cost-effective in light of the 
BART-eligible source's total emissions of the pollutant at issue.'' 70 
FR 39104, 39117. With respect to the RHR requirement that states must 
project visibility impacts of BART controls, the BART Guidelines state: 
``Once you have determined that your source or sources are subject to 
BART, you must conduct a visibility improvement determination for the 
source(s) as part of the BART determination.'' 40 CFR part 51, App Y, 
sect. IV.D.5. Thus, it is clear that both visibility impacts and 
visibility benefits are to be considered cumulatively for multiple 
units at a single source.
    This is also consistent with EPA's practice in other states. For 
example, EPA found it appropriate to consider the combined visibility 
impact of pollution controls on multiple units at a single facility in 
determining that BART is SNCR for Units 1 and 2 of the Colstrip 
facility in Montana. Failure to consider cumulative visibility impacts 
discounts the very real effect of source-specific pollution on regional 
haze and likewise the cumulative benefits of potential retrofits. EPA 
cannot demonstrate that it has properly evaluated BART controls for 
affected sources without producing and presenting such a cumulative 
analysis.
    Response: 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

[[Page 5131]]

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. The approach that we used in our BART 
decisions for Wyoming is consistent with the approach that we used for 
Montana.
    Comment: The commenter submitted results of back trajectory HYSPLIT 
modeling showing that pollutants reaching certain Class I areas on the 
high nitrate haze days did not originate from Laramie River Station. 
The commenter concludes that this analysis confirms that reducing 
NOX emissions from Laramie River would not improve 
visibility at these Class I areas.
    Response: We disagree with the comment that the HYSPLIT results 
submitted by the commenter can be used to evaluate the contribution of 
Laramie River to visibility impairment at Wind Cave National Park. The 
commenter performed HYSPLIT back-trajectory modeling for 10 days with 
high ammonium nitrate concentrations at Wind Cave National Park. The 10 
days were selected from the period from 2001 to 2010, and only two of 
these days occurred during 2001 to 2003 baseline period used for the 
BART visibility modeling. These two days were February 24, 2001 and 
February 14, 2003, when the observed ammonium nitrate at the IMPROVE 
monitoring site at Wind Cave National Park was 41 and 33 inverse 
Megameters (Mm-1), respectively. We note that there were many days 
during the 2001 to 2003 period on which observed ammonium nitrate 
levels at Wind Cave National Park were in the range from 10 to 30 Mm-
1,\133\ but the commenter did not submit HYSPLIT results for these 
days.
---------------------------------------------------------------------------

    \133\ Document with Wind Cave IMPROVE data, in the docket.
---------------------------------------------------------------------------

    HYSPLIT is a trajectory model similar to CALPUFF in that both 
models use modeled and observed wind field data to predict the 
trajectory of pollutants transported from a source area to a receptor 
location. There are differences in the formulation of the HYSPLIT and 
CALPUFF models and differences in the meteorological data used as input 
data for each model, so the predicted trajectory from each model may 
vary somewhat as a result of these differences. The most notable 
difference in the two models is that CALPUFF is designed to predict 
both the trajectory and the chemical conversion of precursor emissions 
to fine particulates and to estimate the concentrations of ammonium 
nitrate and other species at receptor sites, while HYSPLIT simply 
predicts the trajectory of the emissions but does not predict the 
chemical transformations nor the concentration of ammonium nitrate at 
receptor sites.
    We evaluated the CALPUFF results for February 24, 2001 and February 
14, 2003, and found that the HYSPLIT and CALPUFF results were 
consistent, i.e., the CALPUFF model did not attribute high levels of 
ammonium nitrate at Wind Cave National Park on these two days to 
Laramie River. The table of CALPUFF modeling results \134\ shows that 
the model predicted a contribution of nitrate from Laramie River of 
0.02 deciview on Feb 24, 2001, or 0.05% of the observed value, and on 
Feb 14, 2003, 1.697 deciview, or 5% of the observed. The small modeled 
contribution on these days is consistent with uncertainty in the 
HYSPLIT model. Because the HYSPLIT model does not estimate the 
formation of ammonium nitrate, and because HYSPLIT results were only 
submitted for two days during the 2001 to 2003 baseline modeling 
period, these HYSPLIT results are neither useful nor reliable for 
identifying emissions sources that contribute to visibility impairment 
at Wind Cave National Park. The HYSPLIT and CALPUFF results do indicate 
that sources other than Laramie River contribute to visibility 
impairment on the two days with the very highest ammonium nitrate 
levels at Wind Cave during the 2001 to 2003 baseline period. However, 
the CALPUFF results indicate that Laramie River contributes to 
visibility impairment at Wind Cave National Park.
---------------------------------------------------------------------------

    \134\ EPA CALPUFF modeling results for Laramie Rivers Station, 
in docket: CALPUFF--WY--BART--bextNO3--BE--LR--Baseline--WindCave--
12112013.
---------------------------------------------------------------------------

    Comment: EPA improperly considered ``cumulative visibility 
improvement'' when it rejected Wyoming's BART NOX analyses 
and required SCR at Naughton Unit 1 and Naughton Unit 2. (78 FR 34782). 
Other comments asserted that EPA improperly considered ``cumulative 
visibility improvement'' when it rejected Wyoming's BART NOX 
analyses and required SCR at Dave Johnston Unit 3 (78 FR 34778). 
Finally, a third set of comments asserted that EPA R8 has incorrectly 
estimated visibility improvement from all NOX control 
options at Wyodak: Wyoming DEQ evaluated cumulative visibility 
improvements at the two nearest Class I areas (Wind Cave and Badlands 
National Parks) while EPA R8 reported results for only one Class I 
area.
    Response: We disagree with these comments. 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 this 
metric was included in some tables. Finally, in our analysis of 
visibility impacts, we considered the visibility improvement at both 
Class I areas within 300 kilometers of Wyodak. The modeling results for 
the second proposal for all Class I areas considered for each BART 
source were available to the public during the comment period upon 
request. (See ``Summary of EPA's Additional Visibility Improvement 
Modeling'').

[[Page 5132]]

C. Overarching Comments on BART

1. BART-Eligible Sources
    Comment: OCI Wyoming commented that it was listed as a BART-
eligible source, but that the facility has an enforceable cumulative 
annual NOX emission limit of 175.2 tons/year. Therefore, the 
facility is not a ``major stationary source'' and is not BART-eligible.
    Response: We agree with this comment and acknowledge that OCI 
Wyoming is not a BART-eligible source.
2. Costs of Controls
    Comment: One commenter stated that it supported EPA's use of the 
CCM and the Integrated Planning Model (IPM) to calculate costs.
    Response: It is noted that EPA has revised the cost estimates found 
in the proposed rule based upon input from various commenters. The 
differences in cost for individual units may result from: (1) 
Accounting for site elevation in the SCR capital cost; (2) Change in 
SCR reagent to anhydrous ammonia from urea; (3) Change in urea SNCR 
chemical utilization for Laramie River units due to high furnace 
temperature; (4) Incorporation of some of the costs provided in 
comments; (4) Change in auxiliary electrical cost from market price to 
generating, or ``busbar,'' cost; (5) Correction of dilution water cost 
equation for SNCR; and (6) Consideration of shorter plant lifetimes in 
some instances.
    More detailed descriptions of these changes and how they were 
addressed are discussed in a report (Andover Report) and spreadsheets 
\135\ developed for EPA's responses to comments, as well as in our 
responses to the specific comments that are associated with these 
changes below.
---------------------------------------------------------------------------

    \135\ Andover Technology Partners, ``Cost of NOX 
Controls on Wyoming EGUs'', October 28, 2013; Wyoming EGU BART and 
Reasonable Progress Costs--10/28/2013; Wyoming EGU BART and 
Reasonable Progress Costs for Jim Bridger--10/28/2013.
---------------------------------------------------------------------------

    Comment: Sargent & Lundy's analysis provides realistic information 
regarding what it likely would cost to install and operate an SCR 
system at Laramie River Station. They include a cost analysis by 
Sargent & Lundy that, unlike EPA's consultant's work, follows the BART 
Guidelines and EPA's CCM, and takes into account key site-specific 
conditions at Laramie River Station. This analysis is far more accurate 
and reliable than what was done by EPA's consultant--it is a site-
specific, from-the-ground-up analysis done by an engineering firm that 
has done more NOX control projects for EGUs than any other 
firm in the U.S. Sargent & Lundy's analysis provides realistic 
information regarding what it likely would cost to install and operate 
an SCR system at Laramie River. This analysis strongly supports 
Wyoming's decision to select new LNBs and OFA as BART for Laramie River 
Station, not SNCR or SCR.
    Response: We do not agree with this comment. The BART Guidelines 
provide that: ``You should include documentation for any additional 
information 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, labor 
productivity and rates and any other element of the calculation that 
differs from the Control Cost Manual.'' 40 CFR part 51, app. Y, at 
IV.4.a.
    Thus, detailed cost documentation is necessary to the extent that 
cost assumptions differ from the CCM. In this case, several of Sargent 
& Lundy's cost assumptions for control costs at Basin Electric's 
Laramie River Station differed from the CCM, but the necessary 
supporting documentation was not provided as part of their report. 
Detailed descriptions of the deficiencies in the cost assumptions are 
described in comments specific to the units. As explained elsewhere in 
this document, EPA has accepted some of the revised costs developed for 
Basin Electric, but not others.
    Comment: We found that EPA's consultant had added 1.2% to the total 
capital investment of SCR to account for ``taxes and insurance.'' The 
CCM says: ``In many cases property taxes do not apply to capital 
improvements such as air pollution control equipment, therefore, for 
this analysis, taxes are assumed to be zero. The cost of overhead for 
an SCR system is also considered to be zero. An SCR system is not 
viewed as risk-increasing hardware (e.g., a high energy device such as 
a boiler or a turbine). Consequently, insurance on an SCR system is on 
the order of a few pennies per thousand dollars annually.'' The BART 
submittal by PacifiCorp included a 1.1% sales tax and Basin Electric 
included a 4% sales tax, both of which were applied to the purchased 
equipment costs. It is unclear if application of a sales tax is 
appropriate in Wyoming and, if so, what the correct tax rate is.
    Response: To the extent that sales or property taxes are actually 
incurred and increase the cost of the project, they should be accounted 
for in the cost. See CCM at 2.5.4.1 and 2.5.5.8. However, air pollution 
control improvements often do not result in increased property taxes. 
When these taxes are not in fact applied, they should not be included 
in the estimate. The use of 1.2% as an estimate of property taxes and 
insurance is a conservative estimate that is consistent with EPA's 
assumptions in the IPM documentation. According to the IPM 
documentation (version 4.10, Chapter 8, page 8-11): ``U.S. state 
property taxes are approximately 0.9% based on a national average 
basis. This is based on extensive primary and secondary research 
conducted by ICF using property tax rates obtained from various state 
agencies. . . . Insurance costs are approximately 0.3%. This is based 
on estimates of insurance costs on a national average basis.'' \136\
---------------------------------------------------------------------------

    \136\ http://www.epa.gov/airmarkets/progsregs/epa-ipm/BaseCasev410.html#documentation.
---------------------------------------------------------------------------

    As noted by the commenter, these costs may not in fact apply for 
environmental upgrades or may be much less than estimated. EPA did not 
have information on the applicability of property taxes at the time we 
conducted our cost estimates and conservatively assumed a reasonable 
amount. We also note that the commenter did not provide sufficient 
information to support a different property tax or insurance rate.
    With regard to sales tax, the IPM algorithm for SCR cost is based 
upon historical projects and incorporates typical levels of sales tax. 
That is, the capital costs provided by the algorithm(s) are inclusive 
of sales tax. Accordingly, for the purpose of the BART cost estimates, 
and without additional data to determine what sales taxes would 
actually apply, EPA has relied on the assumptions in the IPM algorithm.
    Comment: EPA states in its FIP Action (78 FR 34749): ``For all 
control technologies, EPA has identified instances in which Wyoming's 
source-based cost analyses did not follow the methods set forth in the 
EPA Control Cost Manual. For example, Wyoming included an allowance for 
funds used during construction and for owners costs and did not provide 
sufficient documentation such as vendor estimates or bids.''
    With respect to AFUDC, another utility (Oklahoma Gas and Electric) 
argued in a similar regional haze setting that: ``AFUDC provides a way 
of measuring the real cost of interest over the construction period. 
AFUDC accounts for the time value of money associated with the 
distribution of construction cash flows over the construction period, 
which may be approximately 18 months for an SCR project.'' Total 
capital investment, as

[[Page 5133]]

defined in the CCM, includes all costs required to purchase equipment 
needed for the control system (purchased equipment costs), the costs of 
labor and materials for installing that equipment (direct installation 
costs), costs for site preparation and building, working capital, and 
off-site facilities.
    A cost breakdown of total capital investment (as defined above) is 
presented in several examples in the CCM. For example, Table 1.4 (page 
1-32 of Section 4--NOX Controls) and Table 2.5 (page 2-44 of 
Section 4--NOX Controls) therein explicitly identify AFUDC 
as component ``E'' of the TCI, where TCI = D + E + F + G + H + I, 
where: D = Total Plant Cost; E = AFUDC; F = Royalty Allowance; G = 
Preproduction Cost; H = Inventory Capital; I = Initial Catalyst and 
Chemicals.
    References 9 and 10 on page 2-38 of the CCM explicitly include 
AFUDC as a cost component and reference two reports, by Shattuck and 
Kaplan, in support of its use. The EPA built upon this knowledge base 
and costing methodology in its publication of the CCM in 2002. Thus, 
the CCM allows the time value of money, measured by the real discount 
rate, to be incorporated into the cost estimate.
    Section 2.3.1 of the CCM (Elements of Total Capital Investment) 
describes the need for total capital investment to include all 
expenditures incurred during the construction phase of the project, 
including direct costs, indirect costs, fuel and consumables expended 
during start-up and testing, and other capitalized expenses. The only 
items explicitly mentioned to be excluded are common facilities that 
already exist at the site. AFUDC is part of the expense that will be 
incurred with the installation of a large air pollution control system, 
and the accepted practice in the utility industry and by financial 
institutions is to treat AFUDC as a capitalized expenditure. This 
approach is recognized in publications by the U.S. Department of 
Energy--Energy Information Administration, such as the Annual Energy 
Outlook, and in publications by the Electric Power Research Institute, 
such as the Technical Assessment Guide (EPRI TAG). As previously 
mentioned, the EPA clearly followed this approach in its studies of 
retrofit costs of SO2 and NOX in the years 
leading up to its publication of the CCM. Furthermore, AFUDC has been 
included in several other coal-fired boiler BART determinations, and 
AFUDC is included as a line item in EPA's Coal Quality Environmental 
Cost (CUECost) worksheets for flue gas desulfurization (FGD) control 
systems. In cases where the time value of money during the construction 
period would be significant (e.g., projects with longer construction 
periods such as the installation of SCR), the CCM clearly allows 
inclusion of AFUDC.
    PacifiCorp supports and adopts by reference Oklahoma Gas and 
Electric's argument regarding including AFUDC in project cost 
estimates. Whether or not AFUDC is included in project cost estimates 
does not materially impact the results reached under the EPA CCM 
method, its inclusion should not constitute a basis for EPA to reject 
Wyoming's entire cost assessments. PacifiCorp has provided tables that 
provide comparisons of PacifiCorp's project specific EPA CCM method 
results where AFUDC is excluded in one set of costs and is included in 
the other to demonstrate this point.
    Response: We disagree 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.\137\ Therefore, in reviewing other BART 
determinations, EPA has consistently excluded AFUDC.\138\ EPA's 
position regarding exclusion has been upheld in the United States Tenth 
Circuit Court of Appeals.\139\
---------------------------------------------------------------------------

    \137\ CCM (Tables 1.4 and 2.5 show AFUDC value as zero).
    \138\ See, e.g., 77 FR 20894, 20916-17 (Apr. 6, 2012) 
(explaining in support of the North Dakota Regional Haze FIP, ``we 
maintain that following the overnight method ensures equitable BART 
determinations * * *.''); 76 FR 52388, 52399-52400 (August 22, 2011) 
(explaining in the New Mexico Regional Haze FIP that the Manual does 
not allow AFUDC)
    \139\ Oklahoma v. U.S. EPA, 723 F.3d 1201 (10th Cir. 2013).
---------------------------------------------------------------------------

    The fact that CUECost, the EPRI TAG, and the Department of Energy 
cost estimates, and even cost estimates used as the basis for IPM 
typically include AFUDC is immaterial in this case because, for this 
purpose, overnight cost methodology is used and AFUDC is not included 
in that methodology.
    Finally, we reject the commenter's assertion that Wyoming's 
inclusion of AFUDC did not provide a material basis for EPA to 
disapprove portions of the State's SIP. Inclusion of AFUDC increases 
total project costs of SCR by several million dollars. For example, 
Attachment 4 to PacifiCorp's comment letter shows that AFUDC for Dave 
Johnston Unit 4 would add more than $9.5 million dollars to the capital 
costs of SCR. We find that amounts of this magnitude are not trivial 
when assessing the costs of compliance.
    Comment: Sargent & Lundy's cost estimate does include AFUDC, which 
accounts for the interest charges that would be incurred by Basin 
Electric during SCR construction. AFUDC is a real and a significant 
cost on capital intensive, long-term projects such as SCR installation, 
which require financing over a construction period of up to four years. 
Indeed, to exclude AFUDC would inappropriately bias the cost estimate 
in favor of high capital intensity projects. Therefore, consistent with 
industry practice, Sargent & Lundy included AFUDC, calculated based on 
a typical SCR construction project cash flow assuming a real interest 
rate of 7%.
    The inclusion of AFUDC is not, as EPA asserts, inconsistent with 
either the BART Guidelines or the CCM. See 78 FR 34749. The CCM simply 
references ``Total Capital Investment,'' which includes ``all costs 
required to purchase equipment needed for the control system,'' as well 
as ``working capital.'' CCM 2.3.1, page 2-5 (emphasis added). This 
includes costs required to purchase equipment needed for the control 
system (purchased equipment costs), the costs of labor and materials 
for installing that equipment (direct installation costs), costs for 
site preparation and building, working

[[Page 5134]]

capital, and off-site facilities. Id. Nowhere in the CCM does EPA state 
that AFUDC is not an appropriate cost, particularly with respect to 
long-term, capital intensive pollution control projects. And even if 
the CCM made such an assertion, inclusion of AFUDC in a cost estimate 
cannot be grounds for SIP disapproval because: (1) the CCM is not 
binding for purposes of making BART determinations, and (2) requiring 
states to exclude AFUDC is not consistent with Congress' general 
directive that states include ``costs of compliance'' in their BART 
determinations.
    The CAA requires states to consider in their BART determinations 
the ``costs of compliance,'' but does not further define the term. See 
42 U.S.C. 7491. EPA's regulations codify the BART factors, but neither 
the regulations nor the BART Guidelines in Appendix Y purport to 
restrict in any manner the categories of costs that states should 
consider when making a BART determination. See 70 FR 39166-39168. AFUDC 
is a ``cost of compliance.'' Basin Electric cannot fund large capital-
intensive projects like SCR without financing, and the costs related to 
such financing are real and substantial. Consideration of AFUDC is 
therefore entirely consistent with the CAA's broad reference to ``costs 
of compliance,'' and excluding AFUDC would be inconsistent.
    In this case, even if AFUDC is excluded from the total annual 
costs, the costs of installing SCR do not decrease substantially enough 
to justify SCR. Sargent & Lundy performed sensitivity analyses 
demonstrating that the cost-effectiveness of SCR at Laramie River 
remains at between $8,531 per ton of NOX removed and $9,048 
per ton of NOX removed even if AFUDC is excluded. 
Furthermore, if the maintenance cost and labor rate of 1.5% also is 
factored into the analysis, consistent with the CCM, as opposed to the 
lower 0.25% used by Sargent & Lundy, the cost-effectiveness remains 
above $8,500 per ton of NOX removed. Sargent & Lundy also 
performed a sensitivity analysis demonstrating that including property 
taxes and insurance as 1.2% of total capital investment, consistent 
with the approach taken by Andover but not with the CCM approach, more 
than offsets the exclusion of AFUDC. S&L Evaluation section 7.1.4, 
Table 10.
    Response: EPA agrees that AFUDC can be a substantial overall cost 
on large capital projects that extend over a period of several years. 
However, as noted in the previous response, the CCM clearly excludes 
AFUDC in the overnight cost method. Furthermore, as we explain in more 
detail in responses to comments that pertain to specific sources, we 
disagree with the commenter's estimates for cost effectiveness for the 
Laramie River units.
    Comment: Commenters assert that EPA's regional haze FIP is flawed 
because it failed to provide sufficient documentation, such as vendor 
estimates or bids to validate its estimates. EPA attempts to justify 
its approach by stating: ``In our revised cost analyses, we have 
followed the structure (emphasis added) of the EPA CCM, though we have 
largely used the Integrated Planning Model cost calculations to 
estimate direct capital costs and operating and maintenance costs.'' 78 
FR 34749.
    EPA did not explain what it meant by following the ``structure'' of 
the manual, versus simply following the manual. By contrast, PacifiCorp 
solicited and incorporated vendor estimates into these comments. This 
new information, which EPA must incorporate into new BART analyses to 
the extent EPA issues a final regional haze FIP, validates Wyoming's 
BART analyses cost of control estimates. In addition, it further 
quantifies the inaccuracies in EPA's development and use of purported 
new information that in no way qualifies as vendor estimates, bids, or 
any type of site-specific vendor information.
    Response: We do not agree with this comment. By following the 
``structure'' of the manual, EPA included all of the cost elements that 
the CCM indicates should be included, while excluding those that should 
not (such as AFUDC). In other words, EPA employed the overnight cost 
method as is required for BART analyses. The BART Guidelines require 
that the CCM be followed unless deviations from it are clearly 
documented and explained.
    PacifiCorp received bids from vendors and EPA has incorporated 
information from these bids into its revised cost estimates. However, 
for reasons described elsewhere in response to comments, EPA has not 
accepted all of the costs. The BART Guidelines state: ``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 CCM.'' 70 FR 39166.
    With regard to Basin Electric, vendor quotes for the Laramie River 
Station were not supplied. As Basin Electric indicated in its comments, 
``[t]he LRS cost estimates are conceptual in nature; thus, S&L did not 
procure equipment quotes specifically for the LRS control systems. 
Rather, equipment costs for the LRS projects are based on conceptual 
designs developed for the control systems, preliminary equipment sizing 
developed for the major pieces of equipment, and recent pricing for 
similar equipment.''
    In effect, like the IPM cost algorithms, the method that underlies 
Sargent & Lundy's estimate for Basin Electric is empirically based on 
past data, and not vendor quotes developed specific to Laramie River. 
We have, however, accepted some of the costs submitted by Basin 
Electric and not accepted others.
    Comment: Even if EPA had the authority to require the use of the 
CCM, which it does not, EPA's insistence on Wyoming's strict compliance 
with the costing methodology set forth in the CCM, without adjusting 
the methodology to account for important site-specific factors, leads 
to an erroneous and arbitrary and capricious result. This is not 
required by the CCM. Indeed, the manual expressly discounts the 
usefulness of the costing methodology to power plants generally and to 
SCR control systems specifically, and it acknowledges that deviation 
from the methodology may be appropriate based on a user's engineering 
judgment.
    The CCM provides general costing methodology for stationary source 
air pollution control technologies, applicable primarily to regulatory 
development where a rough order of magnitude estimate is appropriate. 
The introduction to the manual also caveats its usefulness when 
assessing control costs at power plants, which use different cost 
accounting. Cost Manual section 1.1, page 1-3. Specifically, it states 
that ``[e]lectrical utilities generally employ the EPRI Technical 
Assistance Guidance (TAG) as the basis for their cost estimation 
processes.'' Id. In a footnote, it explains that while power plants 
might still use the manual, ``comparisons between utilities and across 
the industry generally employ a process called `levelized costing' that 
is different from the methodology used here.'' Id. section 1.1, page 1-
3 n.1.
    The CCM also generates rough estimates of costs that are less 
accurate than the site-specific cost factors that are more appropriate 
for BART determinations at a large power plant. The manual is used 
heavily in regulatory development, and the costing methodology is 
geared specifically to avoid the necessity of site-specific information 
and to enable estimates to be prepared at ``relatively low cost with

[[Page 5135]]

minimum data.'' Id. section 2.2, page 2-3 (internal quotations 
omitted). However, the level of accuracy is much lower than that for 
estimates using site-specific information. The cost estimating 
procedure can provide a ``rough order of magnitude,'' estimate that is 
``nominally accurate to within  30%.'' Id. section 1.2, 
page 1-4. Indeed, ``EPA does not claim cost estimates for industry at a 
greater than study level accuracy for industrial users'' because ``the 
industrial user will necessarily have much more detailed information 
than the generic cost and sizing information.'' Id. section 2.2, page 
2-3, 2-4.
    Where the user has detailed site-specific information, the manual 
does not contemplate strict adherence to its costing methodology. Users 
may ``exercise `engineering judgment' on those occasions when the 
procedures may need to be modified or disregarded.'' Id. section 1.3, 
page 1-7. With respect to estimating factors used in cost estimates, 
``the application of an appropriate factor requires the subjective 
application of the analyst's best judgment.'' Id. section 2.5.4.1, page 
2-28. The manual is designed to provide a tool box for estimating costs 
that can be helpful to the engineer, but ``[t]he bottom line is that 
there is no clear-cut `cookbook' process through which the analyst will 
be able to make the right informed decision each time, and the 
formalized costing methodology employed by the Manual is only part of 
that process.'' Id. section 2.6, page 2-37.
    With respect to SCR cost estimations, the CCM is no more than a 
``tool to estimate study-level costs for high-dust SCR systems.'' Id. 
section 2.4, page 2-40. The ``[a]ctual selection of the most cost-
effective option should be based on a detailed engineering study and 
cost quotations from the system suppliers.'' Id. This requirement for a 
more detailed study relying on site-specific factors is necessary 
because, as EPA acknowledges, the CCM's assumptions regarding capital 
investment for SCR are inaccurate. Id. section 2.5.4.1, page 2-27. For 
systems like SCR, ``the control in question is either so large or so 
site-specific in design that suppliers design, fabricate, and construct 
each control according to the specific needs of the facility.'' Id. 
section 2.5.4.1, page 2-27.
    Thus, for these systems, ``the Manual deviates from its standard 
approach of providing study level costs and, instead, provides a 
detailed description of the factors that influence the TCI [total 
capital investment] for the analyst to consider when dealing with a 
vendor quotation.'' Id. Under these circumstances, EPA acknowledges 
that getting vendor quotes may be difficult because they cannot be done 
in an ``off-the-shelf'' fashion. Id. The engineering judgment of the 
manual's user is especially critical in estimating the costs of an SCR 
retrofit: ``Probably the most subjective part of the cost estimate 
occurs when the control system is to be installed on an existing 
facility.'' Unless the original designers had the foresight to include 
additional floor space and room between components for new equipment, 
the installation of retrofitted pollution control devices can impose an 
additional expense to ``shoe-horn'' the equipment into the right 
locations. For example, an SCR reactor can occupy tens of thousands of 
square feet and must be installed directly behind a boiler's combustion 
chamber to offer the best environment for NOX removal. For 
these boilers, there is generally little room for the reactor to fit in 
the existing space and additional ductwork, fans, and flue gas heaters 
may be needed to make the system work properly.
    To quantify the unanticipated additional costs of installation not 
directly related to the capital costs of the controls themselves, 
engineers and cost analysts typically multiply the cost of the system 
by a retrofit factor. The proper application of a retrofit factor is as 
much an art as it is a science, in that it requires a good deal of 
insight, experience, and intuition on the part of the analyst. The key 
behind a good cost estimate using a retrofit factor is to make the 
factor no larger than is necessary to cover the occurrence of 
unexpected (but reasonable) costs for demolition and installation. Such 
unexpected costs include, but are certainly not limited to, the 
unexpected magnitude of anticipated cost elements; the costs of 
unexpected delays; the cost of reengineering and re-fabrication; and 
the cost of correcting design errors. Id. section 2.5.4.2, page 2-28. 
The CCM cannot properly account for these uncertainties and thus 
provides that users can apply a ``retrofit factor'' of up to 50 percent 
to account for them. Id. EPA notes that ``[s]ince each retrofit 
installation is unique, no general factors can be developed.'' Id. 
2.5.4.2, page 2-29.
    In sum, the BART Guidelines and CCM were drafted and are to be 
applied as guidelines to assist the states in their decision making, 
not as inflexible mandates. Knowing this, states like Wyoming follow 
the BART Guidelines generally but make the necessary localized and 
individualized adjustments required to generate realistic, rather than 
formalistic, cost estimates. Accordingly, EPA should expect the states 
to deviate on occasion, not to slavishly follow the BART Guidelines and 
CCM to the point of generating artificial (and unrealistic) cost 
estimates. Yet that is now exactly what EPA contends the states must 
do. EPA's current approach to using the BART Guidelines and manual as 
grounds for disapproval without deference to the states' authority and 
local control is unreasonable, erroneous, and arbitrary and capricious.
    Response: There are only very limited situations in which a state 
or EPA can depart from the CCM cost methodology. ``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 CCM, Fifth Edition, February 
1996, EPA 453/B-96-001). In order to maintain and improve consistency, 
cost estimates should be based on the OAQPS CCM, where possible. The 
CCM 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.'' 70 FR 39166.
    The guidelines for BART determinations make it clear that the CCM 
is the intended methodology for conducting a BART cost determination. 
It also states why: To maintain and improve consistency. However, the 
CCM does state that site-specific conditions should be incorporated. 
Site-specific conditions could include space constraints, or a design 
feature that could complicate installing a control. However, the BART 
Guidelines are clear that the analyst should document any deviations 
from the CCM: ``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 CCM.'' 70 FR 39166.
    In fact, the record does not point to any unusual circumstances 
that explain why SCR costs in Wyoming should be so much higher than 
costs of SCR at other similar facilities. As will be demonstrated in 
responses to comments that are specific to the individual units, the 
commenters did not identify any unique features of their plants that 
would make of the installation of an SCR so difficult that the cost 
would be outside the range of what has been experienced elsewhere, even 
accounting for such things as elevation, which is discussed later in 
these responses to

[[Page 5136]]

comments. The BART Guidelines also provide the following explanation, 
which makes it clear that other cost methods are supplemental, not 
replacements for the CCM cost method: ``We believe that the CCM 
provides a good reference tool for cost calculations, but if there are 
elements or sources that are not addressed by the CCM or there are 
additional cost methods that could be used, we believe that these could 
serve as useful supplemental information.'' 70 FR 39127.
    Although the focus in the second quote is ensuring the remaining 
useful life is incorporated into the amortization schedule, this 
passage affirms that the CCM's annualized cost methodology should be 
followed. The following quote from the same page of the BART guidelines 
sheds light on the type of costing methodology employed by the CCM, 
``capital and other construction costs incurred before controls are put 
in place can be rolled into the first year, as suggested in EPA's OAQPS 
CCM.'' Although this passage is again focused on the remaining useful 
life, the text we reproduce is a reference to the basic CCM cost 
methodology--the overnight method. That is what is meant with the 
reference of rolling future costs into the first year. The ``all in'' 
method that OG&E used does not do that--it projects costs to a future 
date. Although the CCM does not use the term, ``overnight cost,'' it is 
widely used in industry.
    The U.S. Energy Information Administration defines ``overnight 
cost'' as ``an estimate of the cost at which a plant could be 
constructed assuming that the entire process from planning through 
completion could be accomplished in a single day. This concept is 
useful to avoid any impact of financing issues and assumptions on 
estimated costs.'' \140\ In effect, the overnight cost is the present 
value cost that would have to be paid as a lump sum up front to 
completely pay for a construction project.
---------------------------------------------------------------------------

    \140\ EIA, ``Updated Capital Cost Estimates for Electricity 
Generation Plants,'' November 2010, footnote. 2. Available at: 
http://www.eia.gov/oiaf/beck_plantcosts/?src=email.
---------------------------------------------------------------------------

    As will be described in EPA's other responses to comments regarding 
specific plants, commenters did not provide the documentation required 
under the RHR to demonstrate why their costs were so much higher than 
costs for other similar units. Such documentation would include any 
vendor quotes to include scope of supply, explanations of labor 
productivity issues with supporting documentation, and other concerns 
raised by commenters and addressed in more detail in other comments.
    Comment: Sargent & Lundy is both a design and engineering firm and 
a system supplier, and it has provided exactly the type of detailed 
scoping-level engineering study for SCR contemplated by the CCM when 
selecting the most cost-effective NOX control device. EPA 
acknowledges that with respect to SCR cost estimations, the CCM is no 
more than a ``tool to estimate study-level costs for high-dust SCR 
systems.'' CCM section 2.4, page 2-40. For systems such as SCR, ``the 
control in question is either so large or so site-specific in design 
that suppliers design, fabricate, and construct each control according 
to the specific needs of the facility.'' Id. section 2.5.4.1, page 2-
27. See also id. at section 2.3, page 2-30 (``the design is highly 
site-specific.''). Importantly, the ``[a]ctual selection of the most 
cost-effective option should be based on a detailed engineering study 
and cost quotations from the system suppliers.'' Id. at section 2.6, 
page 2-43.
    The Sargent & Lundy Evaluation provides a cost estimate not based 
on the general, broad brush assumptions set forth as examples in the 
CCM, but on a conceptual design of SCR at Laramie River based on site-
specific variables and Sargent & Lundy's extensive knowledge of, and 
experience with, SCR installations on coal-fired utility boilers. Based 
on this conceptual design, Sargent & Lundy estimated equipment costs 
using example vendor quotes for similar projects, and used appropriate 
commodity pricing references, rates for labor based on industry 
publications and locality-specific data, and, where necessary, 
allowances.
    Response: As EPA has noted in previous responses to comments, the 
CCM is a good reference tool for estimating costs. With regard to 
Sargent & Lundy's estimates at Laramie River Station, EPA has found 
deficiencies in the cost estimates or underlying assumptions that will 
be discussed in more detail in comments that are specific to units.
    Comment: One commenter asserted that the EPA's cost evaluations 
overestimated the annualized capital costs of BART options by assuming 
an unrealistically high interest rate, which is particularly extreme in 
the cost estimates for SCR because of its relatively higher capital 
costs than the other control technologies evaluated. The commenter 
calculated and submitted documentation of what the commenter considers 
to be the real cost of capital interest to PacifiCorp, which was 5.66 
percent, and to Basin Electric, which was around 2 percent. According 
to the commenter, the EPA continues to assume a much higher 7 percent 
interest rate, apparently based on the EPA's mistaken belief that this 
rate is supported by the CCM. The commenter stated that while the CCM 
states the social interest rate ``is currently set at seven percent'' 
by the Office of Management and Budget (OMB), it references the 
interest rate established by OMB in 1992, whereas the OMB updates 
interest rates yearly and the current social interest rate is 1.7 
percent for a 20-year period (citing OMB Circular A-94, App. C (revised 
Dec. 2012)).\141\ The commenter concluded that even if EPA were correct 
in applying the social interest rate, it should have used the current 
published OMB rate in accordance with the CCM's direction.
---------------------------------------------------------------------------

    \141\ http://www.whitehouse.gov/sites/default/files/omb/assets/a94/dischist-2013.pdf.
---------------------------------------------------------------------------

    The commenter went on to contend that the CCM recommends a source-
specific interest rate for BART and reasonable progress determinations, 
rather than the social interest rate applied in promulgation of 
regulations. According to the commenter, the 7-percent ``social 
interest rate'' is used to estimate the cost to society of taking an 
action. However, the CCM states that this social interest rate ``is 
probably not appropriate for industry.'' The commenter noted that the 
RHR requires the EPA to make case-by-case determinations of ``the costs 
of compliance'' for identified BART and reasonable progress options, 
which the commenter interprets as being the actual cost to the source 
of implementing the studied alternatives. The commenter indicated that 
where the EPA, the state, or industry is evaluating ``the economic 
impact that [air-pollution control] equipment would have upon the 
source,'' a source-specific interest rate is appropriate. The commenter 
concluded that the EPA erred in relying on a generic and outdated 7-
percent social interest rate that resulted in a particularly inflated 
estimate of SCR costs, and asserted that in recalculating the 
annualized capital costs of control technologies, the EPA must either 
use the current social interest rate of 1.7 percent, or more 
appropriately, source-specific rates of 5.66 percent for PacifiCorp 
Units and 2 percent for Basin Electric Units.
    Response: We have retained the use of a 7-percent interest rate in 
calculating the capital recovery factor. For cost analyses related to 
government regulations, an appropriate ``social'' interest (discount) 
rate should be used,

[[Page 5137]]

not the source's actual rate of borrowing. OMB Circular A-4, providing 
Federal agencies guidance on developing regulatory analyses, and dated 
September 17, 2003, reiterates the guidance found in the earlier 
Circular A-94: ``As a default position, OMB Circular A-94 states that a 
real discount rate of 7 percent should be used as a base-case for 
regulatory analysis. The 7 percent rate is an estimate of the average 
before-tax rate of return to private capital in the U.S. economy, based 
on historical data. It is a broad measure that reflects the returns to 
real estate and small business capital as well as corporate capital. It 
approximates the opportunity cost of capital, and it is the appropriate 
discount rate whenever the main effect of a regulation is to displace 
or alter the use of capital in the private sector.'' \142\
---------------------------------------------------------------------------

    \142\ Office of Management and Budget, Circular A-4, Regulatory 
Analysis, http://www.whitehouse.gov/omb/circulars-a004_a_4/.
---------------------------------------------------------------------------

    In addition, EPA calculated capital recovery factors using 3-
percent and 7-percent interest rates in determining cost-effectiveness 
for the Regulatory Impact Analysis for the BART Guidelines.\143\ The 3-
percent rate is mainly used when private consumption displacement is 
the main impact of a regulatory action. This cost of retrofitting power 
plants for this action displaces private capital far more than private 
consumption, so 3-percent is not an interest rate that is applicable 
here. We consider our use of an interest rate of 7-percent to calculate 
capital recovery to be a conservative approach.
---------------------------------------------------------------------------

    \143\ ``Regulatory Impact Analysis for the Final Clean Air 
Visibility Rule or the Guidelines for Best Available Retrofit 
Technology (BART) Determinations Under the Regional Haze 
Regulations,'' EPA-0452/R-05-004 (June 2005).
---------------------------------------------------------------------------

    Finally, the interest rate cited by the commenter from Appendix C 
to OMB Circular A-94, 1.7 percent, is for an altogether different 
purpose than the type of regulatory analysis supporting today's rule. 
According to the discount rate policy described in Circular A-94, 
interest rates contained in Appendix C, which reflect Treasury 
borrowing rates, are for the purpose of internal planning decisions of 
the Federal Government. This is in contrast to regulatory actions, for 
which as noted above, the circular prescribes use of a ``real discount 
rate of 7 percent.'' \144\
---------------------------------------------------------------------------

    \144\ http://www.whitehouse.gov/omb/circulars_a094#8.
---------------------------------------------------------------------------

    Comment: One commenter stated that the EPA used the IPM default 
cost for auxiliary power of $0.06/kWhr for all of the control scenarios 
evaluated, which is much higher than the auxiliary power cost commonly 
used in cost-effectiveness analyses. The commenter asserted that the 
appropriate cost of auxiliary power to use in a cost-effectiveness 
analysis is the busbar cost of power to run the plant, not the cost of 
power sold. According to the commenter, auxiliary power is the power 
required to run the plant, or power not sold, and cost-effectiveness 
analyses are based on the cost to the owner to generate electricity, or 
the busbar cost, not market retail rates. The commenter indicated that 
the site-specific data reported by PacifiCorp to the Federal Energy 
Regulatory Commission (FERC) in 2010 indicates that the busbar power 
cost for the Wyoming PacifiCorp plants is typically in the range of 
$0.02/kWhr to $0.03 lb/kWhr, and for Basin Electric's Laramie River 
Station, the company used a cost of $0.015/kWhr, which is consistent 
with the busbar power cost.
    Response: In EPA's original analysis, we used the default values 
for electricity in the IPM model, although we agree that the cost of 
power used for auxiliary loads should be the cost of the owner to 
generate the electricity rather than the market price it could be sold 
at. EPA has reviewed FERC Form 1 for PacifiCorp \145\ and has 
incorporated in our revised costs for each plant the stated cost of 
electricity per net kWh. For Laramie River Station, the costs in their 
July 2008 BART analysis \146\ are used, as these are more consistent 
with PacifiCorp's reported cost of generation (as reported to FERC) 
than the values later used by Basin Electric or by EPA in our original 
cost analysis (that our proposed action was based on), which are more 
reflective of retail power prices rather than the cost to generate.
---------------------------------------------------------------------------

    \145\ FERC Financial Report, FERC Form No. 1: ``Annual Report of 
Major Electric Utilities, Licensees and Others and Supplemental Form 
3Q: Quarterly Financial Report, PacifiCorp, Year/Period of Report 
2010/Q4,'' pgs 402.2, 403, 403.2, see line 35.
    \146\ Black & Veatch, ``Basin Electric Power Cooperative Laramie 
River Station Refined BART Visibility Modeling,'' July 24, 2008, pg 
25 of 176.
---------------------------------------------------------------------------

    Comment: One commenter argued that the IPM model is not appropriate 
for generating site-specific cost estimates to evaluate the cost-
effectiveness of BART projects because it does not account for those 
site-specific requirements that significantly impact overall project 
costs.
    Response: As described in our proposal, the IPM is a multi-regional 
linear programming model of the U.S. electric power sector. IPM relies 
upon a very large number of data inputs and provides forecasts of 
least-cost capacity expansion, electricity dispatch, and emission 
control strategies for meeting energy demand and environmental, 
transmission, dispatch, and reliability constraints. EPA has used IPM 
to evaluate the cost and emissions impacts of proposed rules, such as 
the recent Mercury and Air Toxics Standard (MATS).\147\
---------------------------------------------------------------------------

    \147\ http://www.epa.gov/airmarket/progsregs/epa-ipm/docs/SuppDoc410MATS.pdf.
---------------------------------------------------------------------------

    We wish to clarify that, for our proposed action on Wyoming's 
Regional Haze SIP, we did not actually run IPM. Rather, we used 
information from one component of IPM, specifically, the component that 
develops the costs of air pollution control technologies. Broadly 
speaking, IPM relies upon numerous components and sub-components to 
specify constraints and variable values that feed into the model 
algorithms used during an actual IPM model run. The air pollution 
control cost development component is just one of these numerous 
components. We relied upon the cost information and equations contained 
in this component by manually placing them into a spreadsheet that 
calculated the capital, operating, and maintenance costs associated 
with pollution control options. While we relied upon the results of 
these spreadsheet calculations, we did not then use those results to 
run IPM, as the type of information generated by an actual IPM model 
run (e.g., generation dispatch decisions, capacity decisions) is not 
relevant to our action.
    We documented our use of the equations from IPM's air pollution 
control technology cost component by placing the raw cost calculation 
spreadsheets in the docket for our proposal.\148\ These spreadsheets 
contain the IPM equations, corresponding variable values, selected 
notes regarding assumptions and variable ranges, as well as selected 
tables from IPM Base Case v4.10 documentation. Because we did not 
perform an actual IPM model run, the spreadsheet and contractor's 
report in the docket for our proposal sufficiently document our use of 
the cost methodologies from the IPM air pollution control cost 
component.
---------------------------------------------------------------------------

    \148\ Wyoming EGU BART and Reasonable Progress Costs--10/28/
2013; Wyoming EGU BART and Reasonable Progress Costs for Jim 
Bridger--10/28/2013.
---------------------------------------------------------------------------

    We disagree with commenters' characterization of the cost-
development methodology contained in IPM as generalized and inadequate 
for performing site-specific cost estimates. As noted in the 
documentation for IPM's cost-development methodology for SCR, the 
methodology is based upon

[[Page 5138]]

two databases of actual SCR projects.\149\ These databases include 2004 
and 2006 industry cost estimates prepared for the Midwestern Ozone 
Group, and a proprietary in-house database maintained by engineering 
firm Sargent & Lundy. The Midwestern Ozone Group information was cross-
referenced with actual 2009 projects, and escalated accordingly. 
Sargent & Lundy then used the information in these databases to develop 
the equations described in the cost component, taking into account the 
pre-control NOX emission level, degree of reduction, coal 
type, facility size, and numerous other unit-specific factors. While a 
costly engineering evaluation that included site visits in addition to 
use of satellite imagery might produce a more refined cost estimate, we 
disagree that our approach does not produce site-specific estimates. As 
noted by EPA in response to other comments, EPA's use of satellite 
imagery enabled us to evaluate each of the major site-specific issues 
raised by commenters.
---------------------------------------------------------------------------

    \149\ http://www.epa.gov/airmarkets/progsregs/epa-ipm/docs/v410/Appendix52A.pdf.
---------------------------------------------------------------------------

    Specifically, we input several site-specific factors, such as fuel 
type, baseline NOX level, reagent cost and type, level of 
NOX reduction, site-specific power and reagent costs, etc. 
into the algorithm. The algorithm also provides for adjustment of cost 
to account for retrofit difficulty. The CCM at section 2.5.4.2 (page 2-
28 of Section 1 Chapter 2) calls for a retrofit difficulty factor to 
account for the site-specific costs associated with a retrofit, such as 
demolition or moving existing equipment, etc. A retrofit factor is also 
used in the IPM algorithm, making it consistent with the approach used 
in the CCM. Per the documentation for the IPM algorithms: ``The 
formulation of the SCR cost estimating model is based upon two 
databases of actual SCR projects. The comparison between the two sets 
of data was refined by fitting each data set with a least squares curve 
to obtain an average $/kW project cost as a function of unit size. The 
data set was then collectively used to generate an average least-
squares curve fit. The least squares curve fit was based upon an 
average of the SCR retrofit projects. Retrofit difficulties associated 
with an SCR may result in capital cost increases of 30 to 50 percent 
over the base model. The least squares curve fits were based upon the 
following assumptions: Retrofit Factor =1; Gross Heat Rate = 9880; 
SO2 Rate = < 3 lb/MMBtu; Type of Coal = Bituminous; Project 
Execution = Multiple lump sum contracts.'' \150\
---------------------------------------------------------------------------

    \150\ Sargent & Lundy, ``IPM Model--Revisions to Cost and 
Performance for APC Technologies--SCR Cost Development Methodology, 
FINAL'', August 2010.
---------------------------------------------------------------------------

    Therefore, the IPM algorithm is based upon actual retrofit 
projects. As such, the average or typical retrofit found for the 
retrofit projects evaluated is assumed to use a retrofit factor of 1.0, 
and for more difficult than average retrofits, a retrofit factor 
greater than 1.0 would apply. On page 1 of the documentation of the IPM 
model for SCR, it states that ``Retrofit difficulties associated with 
an SCR may result in capital cost increases of 30 to 50% over the base 
model.'' Therefore, EPA expects that retrofit difficulty factors may 
apply up to around 1.50 at the maximum. In effect, project elements 
that are typically included in an SCR retrofit are accounted for in the 
cost estimated by the algorithm, and deviations from those typical 
costs can be addressed by a retrofit factor. In fact, the algorithm 
expressly calls for a retrofit factor that can be varied (see Table 1 
of reference, variable ``B''), which makes it consistent with the 
retrofit difficulty factor method called for in the CCM. And, because 
the IPM algorithm is based upon actual projects, it already 
incorporates contingency. Finally, the IPM algorithm can be modified 
for other effects, such as elevation, and EPA has since examined this 
and modified its estimates in this final action to correct for the 
effects of altitude.
    Comment: One commenter alleged that site elevation was not 
reasonably accounted for in EPA's cost estimates, particularly for 
PacifiCorp's Naughton Units 1 and 2 and Dave Johnston Unit 3. The 
commenter explained that algorithms in the IPM model were developed for 
a generic coal-fired power plant located at or near sea level. However, 
site elevation can have a significant impact on control system sizing 
and design. Thus, elevation of the site must be considered separately 
and factored into the unit capacity (i.e. megawatts) accordingly due to 
its effects on the flue gas volume. The commenter pointed out that 
PacifiCorp's Wyoming BART units are located at elevations ranging from 
approximately 5,000 to 7,000 feet above mean sea level. At this 
elevation, flue gas flows will be 20-30 percent higher than similarly 
sized units at mean sea level. The higher flue gas flow requires larger 
ductwork, larger reactors, and more robust support structures, and 
these items have a profound influence on the overall project cost. 
While Wyoming had this information available in its SIP, EPA failed to 
account for site elevation in its FIP.
    Response: EPA agrees with commenter that higher altitudes will 
increase the volume of flue gas, making it necessary to increase the 
cross-sectional area of associated ductwork and the SCR reactor. 
Increased flue gas volume also impacts the fan design. Consequently, 
EPA has revised its cost calculations for SCR in this final action to 
address issues associated with plant altitude. While altitude has a 
significant impact on the cost of SCR, it does not make a significant 
difference in the cost of SNCR because altitude does not affect the 
urea flow rate or the associated urea storage system, urea circulation 
system, or metering/mixing/pumping systems. There may be a slight 
increase in the number of injectors due to increased furnace cross-
section, but this is expected to be a small part of the total cost of 
an SNCR system.
    Comment: The same commenter argued that site configurations were 
not reasonably accounted for in EPA's cost estimates, particularly for 
the Naughton Units 1 and 2 and Dave Johnston Unit 3. The IPM model 
applies a retrofit factor to account for the difficulty of fitting new 
BART equipment into the existing site configuration. The Andover Report 
states that site visits were not possible. Thus, retrofit factors for 
Naughton Units 1 and 2, and Dave Johnston Unit 3 were determined based 
on a review of Google EarthTM images of the station. 
Accordingly, the Andover Report applied retrofit factors for the units 
that are highly subjective based on minimal site information. The 
commenter argued that, when preparing site-specific cost estimates, 
site visits must be conducted to evaluate the true complexity 
associated with the retrofit and to assess specific modifications to 
the plant that would be required to overcome issues associated with 
congestion, as well as difficulties associated with construction. 
Neither Andover nor EPA sought permission from PacifiCorp to visit the 
sites of the BART units, nor did Andover explain why it ``wasn't 
possible'' to do so. Both Sargent & Lundy and Babcock and Wilcox have 
extensive experience with PacifiCorp's Naughton and Dave Johnston 
facilities. Just since 2005, Sargent & Lundy has been contracted by 
PacifiCorp to perform 14 projects at Dave Johnston station and over 25 
projects at Naughton station. These projects range from site 
evaluations, studies, detailed engineering, or functioning as 
PacifiCorp's Owner's Engineer for major environmental retrofit 
engineer, procure, and construct projects. From having conducted many 
site visits at these stations, Sargent &

[[Page 5139]]

Lundy is very aware of site-specific congestion and construction 
challenges that would affect SCR installations at Naughton Units 1 and 
2 and Dave Johnston Unit 3. Similar to Sargent & Lundy's site-specific 
experience, Babcock and Wilcox has recently completed major 
environmental retrofit projects on Naughton Units 1 and 2 (wet scrubber 
additions) and Dave Johnston Unit 3 (dry scrubber and baghouse 
addition), making Babcock and Wilcox uniquely positioned to offer 
budgetary cost estimates for further retrofits to those facilities with 
significant first-hand knowledge. While Wyoming had much of this 
information available in its SIP, EPA failed to account for this site-
specific information in its FIP.
    Response: EPA disagrees with the commenter. EPA did account for 
site-specific factors when performing its cost estimates. Because SCRs 
are built on or next to the boiler structure, they are often elevated, 
and there is usually equipment in one direction (the boiler) or the 
other (other air pollution control equipment, like an electrostatic 
precipitator (ESP), scrubber, or chimney) that limits access. This 
issue is complicated further with boilers that are located adjacent to 
one another--a common configuration. Due to the height of the SCR, 
large cranes play a vital role in their construction. The location of 
cranes next to where the SCR is going to be built can be difficult.
    As noted in a paper by Babcock & Wilcox,\151\ key issues for SCR 
constructability are site access and ability to locate a crane and the 
resulting erection sequence. The erection sequence is impacted by the 
crane that is available and whether it can fit on site because the 
crane and its location will limit the size of material that can be 
lifted into place. A larger crane allows for the lifting of larger 
pieces of ductwork, resulting in fewer lifts and less fabrication in 
the air. Without adequate access for a crane and proximity to a lay-
down area for material, erection must be done with smaller pieces, 
which will require more labor and expense.
---------------------------------------------------------------------------

    \151\ Hines, J.A., Kokkinos, A., Fedock, D., ``Benefits of SCR 
Design for Constructability'', Power Gen, International 2001, 
December 11-13, 2001, Las Vegas
---------------------------------------------------------------------------

    Access around and between the boilers will determine crane location 
and location of material receiving areas. In some cases, it may be 
necessary to demolish equipment or buildings in order to gain adequate 
access. In other cases, it may be necessary or preferable to erect 
cranes on the top of the boiler structure (as was performed for the SCR 
installed at Dominion's Brayton Point Unit 3).\152\
---------------------------------------------------------------------------

    \152\ Wright, B., Erickson, C., Phillipo, M., ``Keys to Success: 
SCR Installations at Dominions Brayton Point Units 1 and 3'', 
Electric Power, May 2008.
---------------------------------------------------------------------------

    Because of its easy availability and it usefulness in providing a 
``bird's eye'' view of the site congestion (how close equipment is 
located to each other, room for a crane, etc.), site access, local 
transportation options, availability of a lay-down area to locate 
material on site, and other limitations around the site, satellite 
imagery has become a very important tool in evaluating these site-
specific factors. In fact, the major air pollution control original 
equipment manufacturers use satellite imagery to assist them in 
estimating site congestion issues, determining location of construction 
equipment and other limitations on and around the site in this way. 
Site visits are also useful, but are normally performed in addition to 
rather than in lieu of careful examination of satellite images. For 
example, in their comments to EPA in 2010, the Utility Air Regulatory 
Group used satellite photographs to demonstrate the relative difficulty 
of different SCR installations.\153\
---------------------------------------------------------------------------

    \153\ Cichanowicz, E., ``Current Capital Cost and Cost-
Effectiveness of Power Plant Emissions Control Technologies'', 
Prepared for Utility Air Regulatory Group, January 2010.
---------------------------------------------------------------------------

    While a site visit can be useful and provide additional 
information, assessing satellite imagery provides adequate information 
to determine access to the site, access around the boilers, 
availability of space for locating construction equipment and 
materials, and whether buildings or equipment must be demolished to 
make room for the equipment. Notably, the budgetary price provided to 
PacifiCorp from Babcock & Wilcox was not developed from a site visit. 
Per the cover page of the budgetary proposal, ``[g]iven the budgetary 
nature of this request, we have not made site visits to consider layout 
options: instead, we have used available drawings and made necessary 
assumptions to enable us to establish a basis to derive quantities of 
material and associated costs.''
    For SCR installations, site visits and more detailed boiler 
drawings provide additional information regarding air preheater 
location and whether it must be relocated to make room for the SCR, or 
if ductwork limitations require demolition of other large pieces of 
equipment such as ESPs. Such costs will significantly increase the cost 
of retrofitting an SCR. However, relocation of the air preheater or ESP 
was not identified as a concern by any of the commenters. Instead, most 
commenters raised retrofit issues that are commonly encountered in SCR 
retrofits, including location of SCR support steel and possible 
interferences with other equipment on site; penetration of boiler 
building by SCR ductwork; location of cranes for units that are side-
by-side; the need for increased fan capacity and associated electrical 
modifications; and stiffening of ductwork due to increased pressure 
drop from SCR. As a result, the retrofit costs in the IPM algorithms 
that were developed from actual SCR projects should capture these more 
common retrofit issues and to the extent that some situations seem more 
difficult, can be addressed with retrofit factors.
    In its cost estimates, PacifiCorp provided a long comparative table 
(over 100 rows with 25 columns of data) for Dave Johnston 3 and 
Naughton 1 and 2 showing different cost estimating methods. The table 
showed vendor budgetary pricing for Direct Capital Costs based upon a 
proposal from Babcock & Wilcox. While EPA accepts the proposal from 
Babcock & Wilcox as part of our final action, we have a few general 
comments. The proposal, while providing a detailed total scope of 
supply, provides a total cost for the project without line items. In 
addition, the items included under the Owner's scope by Black and 
Veatch are limited to: Boiler modifications; air-preheater 
modifications; medium voltage power source; asbestos, lead, and 
polychlorinated biphenyls (PCB) remediation; commercial licenses and 
permits; and spare parts. Some of these costs, such as air-preheater 
modifications, will not be required, while others will have a small to 
modest impact on the overall cost. Notably, the cost estimate includes 
items like potable water systems, fire protection, service water, other 
assorted auxiliaries, as well as roads, fences, etc. Therefore, the 
proposal is close to being ``turnkey'' and includes nearly all costs 
for the project. It also includes some items that would fall into the 
category of General Facilities. The proposal assumes that other 
modifications to the Naughton plant will result in abandonment of 
existing chimneys and ESPs on Units 1 and 2. Currently, these chimneys 
are in service. Babcock & Wilcox also determined that the existing fans 
are likely to be sufficient for the additional draft loss from the SCR. 
As was noted in the Andover Report, we think that there may be some 
substantial conservatism built into the Babcock and Wilcox estimate. 
For example, the proposal states that the offset of Naughton Units 1 
and 2 is just enough

[[Page 5140]]

to make it impractical to make a common structure for both SCR 
reactors. However, examination of the drawings in the proposal shows 
that Babcock & Wilcox has sized the ductwork to compensate for the 
offset so that the SCR reactors should be able to be supported with a 
common structure. Also, the proposal assumes that the abandoned stacks 
at both sites will be dismantled, although this does not appear to be 
necessary for Naughton Unit 2 and may not be necessary for Dave 
Johnston Unit 3 if shorter horizontal duct runs are used. Finally, 
comparison of the cost estimate provided by PacifiCorp for Naughton 
Unit 1 and 2, and Naughton 1 especially, to historical costs shows that 
the costs are well in excess of what other SCRs have cost. This is 
particularly perplexing because additional fan capacity is not needed, 
and it is not necessary to move the air preheater.
    In light of the proposal and its fairly comprehensive scope of 
supply, PacifiCorp's capital cost estimate included a number of items 
that EPA is not including in our cost estimate, as noted below:
    1. Process Contingency: Although the CCM permits a process 
contingency of 5%, in EPA's opinion this is not necessary today for SCR 
on coal-fired boilers firing the coals used in Wyoming. According to 
the Department of Energy's National Energy Technology Laboratory,\154\ 
``Process contingency is intended to compensate for uncertainty in cost 
estimates caused by performance uncertainties associated with the 
development status of a technology. Process contingencies are applied 
to each plant section based on its current technology status.'' 
According to the document, for commercially available technologies, 
process contingency could range from 0-10%.
---------------------------------------------------------------------------

    \154\ U.S. Department of Energy, National Energy Technology 
Laboratory, ``Cost Estimating Methodology for NETL Assessments of 
Power Plant Performance'', DOE/NETL-2011/1455, April 2011, pg 4.
---------------------------------------------------------------------------

    When the CCM was issued in January 2002, SCR was commercially 
available but was only emerging in application on coal-fired utility 
boilers in the U.S. According to a study by Cichanowicz,\155\ there was 
only about 13,000 MW of coal-fired capacity using SCR in the U.S. at 
the end of 2001, with nearly all SCRs installed in the prior two years, 
meaning that there was very limited long-term experience with SCR on 
coal-fired units. SCR usage on coal-fired boilers has since increased 
about ten-fold to about 130,000 MW of coal capacity (over 40% of all 
U.S. coal capacity), and is therefore a very well proven and well 
understood technology on a wide range of U.S. coals, including Powder 
River Basin coal. As a result, the process contingency for SCR on coal-
fired utility boilers should be much lower today than what it was when 
the CCM was issued in January 2002, which was 5%. EPA believes that for 
SCR applications on utility boilers burning Powder River Basin coals 
(the Wyoming utility boilers), which are very well understood SCR 
applications, there should not be any need for process contingency.
---------------------------------------------------------------------------

    \155\ From data in Cichanowicz, J., Muzio, L., Hein, M., ``The 
First 100 GW of SCR in the U.S.,--What Have We Learned?''--2006 Mega 
Symposium.
---------------------------------------------------------------------------

    2. Project Contingency: Because the cost estimates developed for 
PacifiCorp are already very conservative and based upon detailed 
estimates of the labor and materials to build the SCR, a 15% project 
contingency is excessive. According to the CCM at Section 1.1, Chapter 
1, pages 1-4: ``The accuracy of the information in the Manual works at 
two distinct levels. From a regulatory standpoint, the Manual 
estimating procedure rests on the notion of the ``study'' (or rough 
order of magnitude--ROM) estimate, nominally accurate to within  30%. This type of estimate is well suited to estimating control 
system costs intended for use in regulatory development because they do 
not require detailed site-specific information necessary for industry 
level analyses.''
    The methods and cost elements of the CCM were adapted from the 
American Association of Cost Engineers, or AACE (CCM Section 1, Chapter 
2, pages 2-5). AACE 16R-90 \156\ states that, ``Project Contingency is 
included to cover the costs that would result if a detailed-type 
costing was followed as in a definitive-type study.'' According to 
National Energy Technology Laboratory (NETL), ``AACE 16R-90 states that 
project contingency for a `budget-type' estimate (AACE Class 4 or 5) 
should be 15% to 30% of the sum of [bare erected cost], [engineering, 
procurement, and construction] fees and process contingency.'' \157\ 
AACE 18R-97 defines different classes of estimates, from Class 5 (least 
detailed) to Class 1 (most detailed).\158\ The methodology used in the 
CCM falls into a Class 4 or Class 5, while the methodology used by 
PacifiCorp's contractor, Babcock and Wilcox, is clearly a far more 
detailed estimate that does not leave out any aspect of the project. 
Therefore, the project contingency factor is not applicable. The 15% 
project contingency factor in the CCM for SCR is based upon use of the 
cost-estimating method described in the CCM to develop the Total Direct 
Capital Costs. It is not intended to apply to a detailed estimate that:
---------------------------------------------------------------------------

    \156\ AACE Recommended Practice, AACE 16R-90; www.aacei.org/non/rps/18R-90.pdf.
    \157\ U.S. Department of Energy, National Energy Technology 
Laboratory, ``Cost Estimating Methodology for NETL Assessments of 
Power Plant Performance'', DOE/NETL-2011/1455, April 2011, pg 5.
    \158\ AACE Recommended Practice, AACE 16R-87; www.aacei.org/non/rps/18R-97.pdf.
---------------------------------------------------------------------------

     Includes many cost items not explicitly included in the 
estimating method described in the CCM to develop the Total Direct 
Capital Costs and meant to be included in the 15% project contingency; 
and
     Already has substantial contingency built into it through 
conservative assumptions.
    In fact, the CCM discusses the importance of not double-counting 
contingency in multiple places, such as retrofit factor and 
contingency: ``Due to the uncertain nature of many estimates, analysts 
may want to add an additional contingency (i.e., uncertainty) factor to 
their estimate. However, the retrofit factor is a kind of contingency 
factor and the cost analyst must be careful to not impose a double 
penalty on the system for the same unforeseen conditions. Retrofit 
factors should be reserved for those items directly related to the 
demolition, fabrication, and installation of the control system. A 
contingency factor should be reserved (and applied to) only those items 
that could incur a reasonable but unanticipated increase but are not 
directly related to the demolition, fabrication, and installation of 
the system. For example, a hundred year flood may postpone delivery of 
materials, but their arrival at the job site is not a problem unique to 
a retrofit situation.'' (emphasis added). The CCM, therefore, 
explicitly anticipates that some analysts may, incorrectly, apply 
multiple contingencies for the same areas of uncertainty even when 
using the methods described in the CCM for estimating Total Direct 
Capital Costs.
    Because the cost estimates developed for PacifiCorp are already 
very conservative and based upon detailed estimates of the labor and 
materials to build the SCR, rather than study-level estimates, they 
have double-counted both the costs that are intended by the CCM to be 
included in the project contingency when using the CCM method, plus 
they have added additional contingency in the form of conservative 
assumptions to address uncertainties in their estimate. For this 
reason, a 15% project contingency is

[[Page 5141]]

excessive, and we have not revised our cost estimates to include one.
    3. General Facilities: The cost estimate from Babcock & Wilcox, 
submitted by PacifiCorp, includes several items that would fall into 
the category of General Facilities, and in order to avoid double-
counting, EPA has not included an additional line item for General 
Facilities.
    Comment: The same commenter suggested that the project-specific 
scopes were not reasonably accounted for in EPA's cost estimates, 
particularly for Naughton Units 1 and 2 and Dave Johnston Unit 3. 
Additional project-specific scope concerns (related to the addition of 
SCR on-site) include limited capacity of the existing induced-draft 
fans and auxiliary power system, as well as National Fire Protection 
Association-related equipment reinforcement requirements. Larger, more 
powerful, induced-draft fans may overload existing electrical systems, 
and the electrical systems may require significant modifications. 
Structural stiffening of the duct work, and equipment downstream of the 
boiler and upstream of the new fans may also be required by National 
Fire Protection Association regulations to operate at more negative 
pressures due to the installation of the SCR. These types of costs are 
not generally reflected in the base case IPM cost algorithms, but they 
must be taken into consideration in the development of a project-
specific cost estimate. Wyoming had this information available in the 
Wyoming SIP, but EPA failed to account for this important cost 
information in its FIP.
    Response: All SCR systems experience a pressure drop across the 
SCR, and therefore some consideration must be made to fan capacity for 
every SCR system. The algorithm used by EPA explicitly includes a 
``balance of plant'' cost line item such as an allowance for additional 
fans and auxiliary electrical work.\159\ As for the duct stiffening, 
this is frequently necessary when new fans are installed. However, as 
noted in the proposal by Babcock & Wilcox, additional fan capacity is 
not expected to be necessary at Naughton Units 1 and 2 or Dave Johnston 
Unit 3.
---------------------------------------------------------------------------

    \159\ Sargent & Lundy, ``IPM Model--Revisions to Cost and 
Performance for APC Technologies--SCR Cost Development Methodology, 
FINAL'', August 2010, Table 1, pg. 5.
---------------------------------------------------------------------------

    Comment: The same commenter alleged that Owner's costs were not 
reasonably accounted for in EPA's cost estimates, particularly for 
Naughton Units 1 and 2 and Dave Johnston Unit 3. Owner's Costs include 
a variety of non-financial costs incurred by the owner to support 
implementation of the air pollution control project. Owner's Costs are 
project-specific, but generally include costs incurred by the owner to 
manage the project, hire and retain staff to support the project, and 
costs associated with third party assistance associated with project 
development and financing. Owner's Costs are real costs that the owner 
will incur during the project and are typically included in cost 
estimates prepared for large air pollution control retrofit projects. 
In fact, EPA's Coal Quality Environmental Cost (CUECost) model includes 
Owner's Costs (or ``Home Office'' costs) in its air pollution control 
system cost estimating workbook and interrelated set of spreadsheets. 
See CUECost Workbook User's Manual Version 1.0, prepared by Raytheon 
Engineers & Contractors, Inc. and Eastern Research Group, Inc., EPA 
Contract No. 68-D7-0001, Appendix B, pages B-3 and B-6. Wyoming had 
this information available in its SIP, but EPA failed to account for 
this important cost information in its FIP.
    Response: Home office fees are Owner's costs, and these are 
accounted for in the CCM in the 10% allowance for Engineering and 
Owner's Costs. See CCM at Section 4.2, Chapter 2, page 2-44). As 
described in Table 2.5 of the CCM, engineering and home office fees 
represent 10% of purchased equipment costs. In this respect, we agree 
with the commenter's assertions that the CCM does discuss some of the 
items that roll up into these line items. For example, the CCM does 
provide for ``Engineering and Home Office Fees'' that includes the home 
office and plant support costs described in the comments. We have 
included the portion of Owner's Costs/Surcharge in the total cost, up 
to the value specified for ``Engineering and Home Office Fees'' 
indicated by the CCM, which is 10%.
    The cost factors used in the CCM include home office fees in the 
10% that is applied to engineering fees; however, the line item for 
Owner's Cost in the IPM estimate was made zero. The reason Owners Cost 
was removed is that the CCM includes owner's cost with the 10% for 
engineering and home office fees. A 10% engineering charge was already 
applied and therefore an additional allowance for home office fees 
would be greater than the cost allowed under the CCM. Even if that cost 
were added at a 5% rate, it would increase capital cost by 5%. This 
difference would not change the determination.
    Comment: The same commenter argued that regional labor concerns 
were not reasonably accounted for in EPA's cost estimates, particularly 
for Naughton Units 1 and 2 and Dave Johnston Unit 3. Regional labor 
concerns are not accounted for in the IPM model. Regional labor 
characteristics must be taken into consideration in a site-specific 
cost estimate to account for factors including labor availability, 
project complexity, local climate, and working conditions. Because the 
Naughton and Dave Johnston facilities are in relatively remote 
locations, higher labor rates must be paid to attract the kind of 
skilled workers required to construct an SCR project. In addition, the 
locations are subject to extreme cold and wind that can result in 
significant productivity and construction challenges and delays, adding 
to the overall project cost. Wyoming had this information available in 
its SIP, but EPA failed to account for this important cost information 
in its FIP.
    EPA's flawed analyses of incomplete ``new'' cost information 
directly resulted in EPA's proposed requirements for PacifiCorp to 
install SCR on Naughton Units 1 and 2 and Dave Johnston Unit 3. In 
contrast, to be responsive to EPA's request for additional information, 
PacifiCorp has solicited budgetary project-specific cost information 
from Babcock and Wilcox, an active and uniquely positioned competitive 
market participant for SCR technology, for these same units. In 
conjunction with Sargent & Lundy's expertise, PacifiCorp has 
incorporated the site-specific budgetary cost information from Babcock 
and Wilcox into updated EPA CCM side-by-side comparisons with the 
Andover Report results to further demonstrate the inaccuracies in the 
new cost information developed by EPA. (The following included tables 
to summarize the results of these comparisons.) It is important to note 
that PacifiCorp has utilized a 20-year remaining equipment life and has 
excluded AFUDC from the results in the tables for comparison purposes.
    As demonstrated by the results in the tables, EPA significantly 
understated costs per ton of pollutant removed. As such, EPA based its 
cost-effectiveness conclusions on significantly inaccurate information. 
Before taking any final action on the proposed FIP, EPA must consider 
in its final BART analyses the additional cost information being 
provided by PacifiCorp.
    Response: EPA disagrees with this comment. The commenter claims 
that remote locations require offering higher wage rates and that 
conditions at the site, including inclement weather, reduce worker 
productivity. Because the commenter claims that these are important 
factors that impact cost, the

[[Page 5142]]

commenter should have provided data to support its assertions. The 
commenter did not provide any data to demonstrate that wage rates in 
the area near its facilities are higher than in more populated areas. 
The commenter also did not provide any specific productivity factors or 
other evidence to show how the commenter arrived at its man-hour 
estimates or explain how those estimates differ from a normal 
productivity. Without such information, EPA cannot review and validate 
the commenter's claims that labor productivity is low or that labor 
cost is high.
    On the other hand, labor rates for union construction labor are 
available from the Construction Labor Research Council (CLRC),\160\ and 
these rates are consistently well below what appears to have been 
assumed in the itemized estimate provided with Basin Electric's 
comments. The difference is too large to be explained by per diem. 
Because both Basin Electric and PacifiCorp used Sargent & Lundy to 
prepare their cost estimates, it is reasonable to assume that both 
companies have made similar assumptions. Moreover, in addition to the 
high labor rates assumed, there are additional line item costs for 
overtime and per diem. As a result, the estimates provided by 
commenters appears to incorporate additional costs or provisions well 
beyond the normal costs both in the labor rates and in the line item 
for additional labor costs.
---------------------------------------------------------------------------

    \160\ Construction Labor Research Council Union Wages and 
Supplements, available from the National Construction Boilermaker 
Employers Web site, www.nacbe.com.
---------------------------------------------------------------------------

    While EPA welcomes the use of SCR vendor estimates, such as those 
used by PacifiCorp, or engineering estimates, such as those provided by 
Basin Electric, specific details supporting the estimates must be 
provided in order for them to be useful. Without details on the scope 
of supply, the estimates cannot be used as a reliable source of 
information because vendor scope could potentially be in error or could 
be duplicative of other costs included in the estimate elsewhere.
    With regard to adjustment for regional labor concerns, neither 
PacifiCorp nor Basin Electric's submittals were satisfactory. 
Deficiencies in Basin Electric's submittal with regard to regional 
labor concerns are addressed elsewhere in these responses to comments. 
The proposal from Babcock & Wilcox, while showing a total lump sum 
price, did not demonstrate how they factored in regional labor 
concerns.
    Comment: The IPM methodology relied upon in the Andover Report is 
inconsistent with the assumptions set forth in the CCM. While EPA 
states that its revised cost analyses ``followed the structure of the 
EPA CCM,'' EPA acknowledges that ``we have largely used the Integrated 
Planning Model cost calculations to estimate direct capital costs and 
operating and maintenance costs.'' 78 FR 34749. See also Andover Update 
(``In estimating the costs of controls, the following were considered: 
IPM Cost Models, US EPA Air Pollution CCM.''). The IPM model is a 
regulatory model that uses cost algorithms developed by Sargent & Lundy 
to estimate system-wide costs of air pollution technology for adoption 
of national regulations. The inputs in the IPM model do not conform to 
the methodology set forth in the CCM, and neither Andover nor EPA 
offers any explanation for the discrepancies. This failure epitomizes 
the arbitrary and capricious nature of EPA's decision making.
    A careful reading of the Andover Report evidences that Andover only 
followed the CCM on a limited basis, and in doing so, read into the CCM 
requirements that are non-existent, while ignoring wholesale many of 
the CCM's recommendations. The following language from the report 
illustrates the concern: ``The BART Guidelines recommend use of the EPA 
Air Pollution CCM, and the methodology used here for estimating costs 
is consistent with the recommendations in the manual, such as inclusion 
of taxes, insurance and administrative costs, and the use of overnight 
cost for capital cost.''
    As an initial matter, the CCM does not anywhere recommend the use 
of ``overnight cost'' for estimating capital costs. See Section XI.B.3. 
The overnight approach assumes construction of a project ``overnight,'' 
which means a party would not incur any interest charges, including 
AFUDC, or experience any cost escalations. While a ``constant dollar 
approach'' may be read to exclude escalation, the CCM does not 
recommend that users assume that the interest costs related to 
constructing a capital-intensive, multi-year project will cost the same 
as an off-the-shelf control technology that can be installed in a day. 
Furthermore, based on the acknowledgment in its report, Andover appears 
only to have followed the methodology set forth in the CCM in these 
three limited respects. In all other respects, Andover relied on the 
IPM cost algorithms for its SCR estimate, without regard to the CCM. 
For SNCR capital costs, Andover relied on an assumed $20/kW cost rather 
than using the IPM algorithms or the methodology set forth in the CCM.
    The high-level cost algorithms in the IPM model and the assumed 
$20/kW capital cost for SNCR resulted in a substantial underestimate of 
the costs of designing and constructing both the SCR and SNCR systems. 
For example, the only inputs to the IPM cost modules are unit size, 
heat input, fuel type, and NOX removal efficiency. The CCM 
uses these inputs, but also provides for using SCR design parameters, 
such as flue gas flow rate, actual stoichiometric ratio (the amount of 
reagent needed to achieve target NOX emissions), space 
velocity and area velocity (a measure of flue gas resident time in the 
SCR reactor), catalyst volume, SCR reactor dimensions, and reagent 
consumption. The CCM methodology includes similar design parameters 
specific to SNCR, which the IPM model does not include and which 
Andover did not consider in its cost estimate.
    Andover also used the IPM model to calculate indirect capital costs 
in the SCR cost estimate. The IPM model includes in its indirect 
capital cost algorithm factors for Engineering and Construction 
Management, Labor, and Contractor Profit/Fees. Andover removed from the 
IPM capital cost calculation both Owner's Costs and AFUDC. Moreover, 
the IPM Model does not include a number of other inputs that are 
included in the CCM, including preproduction costs, inventory capital, 
and initial catalyst costs, and Andover did not adjust the model to 
incorporate these additional costs. Andover's selection of certain IPM 
model inputs and exclusion of inputs in the CCM resulted in the 
substantial underestimation of the indirect capital costs necessary to 
design and install an SCR system.
    For SNCR, Andover arbitrarily assigned a capital cost of $20/kW, 
without using the IPM algorithms or performing an analysis of direct 
and indirect costs consistent with the CCM. Andover relied on the IPM 
model to calculate operating and maintenance costs except for urea 
reagent costs for SNCR, which uses factors that differ from those 
recommended in the CCM and costs utilized by Sargent & Lundy. The 
model's simplistic treatment of catalyst costs, for example, is 
inconsistent with the recommendations in the CCM and underestimates 
annual catalyst replacement costs for SCR and for SNCR. For SNCR, 
Andover adjusted the urea utilization rate from the IPM model without 
explanation, which cut in half estimated annual urea costs. Andover 
also assumed urea would be used as the reagent for SCR, which

[[Page 5143]]

increased reagent costs above those calculated by Sargent & Lundy based 
on the use of anhydrous ammonia.
    Neither Andover nor EPA explains why the cost estimate uses only 
limited portions of the CCM, or why Andover believes use of the IPM 
cost methodology is somehow more representative of the costs of control 
that would be incurred by Basin Electric for installation of SCR and 
SNCR systems. It is inherently contradictory, and therefore arbitrary 
and capricious, for EPA to base its proposed disapproval of Laramie 
River BART on the alleged failures of Wyoming to follow the CCM, while 
at the same time relying on a consultant's report that does not comply 
with those same standards.
    Response: EPA disagrees with the commenter. The methodology used by 
EPA is consistent with the CCM for the following reasons: (1) EPA used 
the overnight cost method, which excludes certain cost elements such as 
AFUDC; (2) The comparative nature of BART costs makes use of the IPM 
algorithms a reasonable approach; and (3) as demonstrated in the 
Exhibit 14 of Basin Electric's comments, use of the cost equations in 
the CCM would have actually resulted in lower costs than predicted by 
the IPM algorithms.
    Moreover, both the IPM algorithms and Sargent & Lundy's estimates 
for Basin Electric are empirically based from data collected at other 
projects. According to page 21 of Exhibit 14 to Basin Electric's 
comments: ``. . . Cost estimates prepared for LRS are based on 
equipment costs and budgetary quotes available from similar projects 
and Sargent & Lundy's experience with the design and installation of 
retrofit SNCR and SCR control systems. The LRS cost estimates are 
conceptual in nature; thus, Sargent & Lundy did not procure equipment 
quotes specifically for the LRS control systems. Rather, equipment 
costs for the LRS projects are based on conceptual designs developed 
for the control systems, preliminary equipment sizing developed for the 
major pieces of equipment, and recent pricing for similar equipment . . 
.'' As a result, the estimates provided by Basin Electric are not more 
valid than those developed by EPA.
    Commenter claims that AFUDC should be included in the cost 
analysis. As described in our responses to other comments, the CCM 
explicitly excludes AFUDC from control costs, and EPA's estimates were 
correct in excluding AFUDC. This is central to the overnight cost 
methodology.
    Commenter is critical of EPA's method for estimating SNCR capital 
cost. With regard to SNCR cost methodology, the cost of SNCR is driven 
primarily by the operating cost. Capital cost has a small effect on 
total cost of SNCR. Therefore, a simplifying assumption that yields a 
reasonable estimate of capital cost will result in an annualized cost 
accurate to within  30%. Based upon input from technology 
vendors on cost for other units \161\ and based upon the fact that 
EPA's contractor Andover has over 25 years of direct experience 
designing, specifying and optimizing SNCR systems, $20/kW was a very 
reasonable estimate. EPA's contractor also reviewed the SNCR algorithms 
developed by Sargent & Lundy for EPA and is of the opinion that in most 
cases $20/kW provides a better estimate than the IPM algorithm because 
the IPM algorithm assumes greater economies of scale than generally 
exist in SNCR applications.
---------------------------------------------------------------------------

    \161\ Email from Alex Dainoff, Fuel Tech, to James Staudt, 
Andover Technology Partners, Wednesday, June 27, 2012.
---------------------------------------------------------------------------

    On the other hand, as will be described later in this response, the 
high reported furnace temperature at Basin Electric's Laramie River 
Station means that SNCR will require a more complex injection system 
and will have a higher urea injection rate than what is typical for 
most SNCR systems. For this reason EPA has accepted the capital cost of 
SNCR estimate provided by Basin Electric. In any event, SNCR capital 
cost generally has a small effect on cost compared to operating cost.
    Commenter is critical of EPA's use of the IPM algorithm as not 
utilizing the same input design parameters. As described in other 
responses to comments, the IPM algorithm for SCR incorporates either 
directly or by inference all of the inputs the commenter has raised. 
But, commenter is incorrect about some of the claimed inputs. 
Stoichiometric ratio, space velocity, area velocity, catalyst volume, 
SCR reactor volume, and reagent consumption are not inputs to an SCR 
design, but are all direct results of design calculations using the 
inputs of initial NOX level, final NOX level (and 
by inference percent reduction), unit size, heat rate, and fuel 
characteristics--all inputs to the IPM model. Gas flow rate is also 
normally calculated based upon fuel type, unit size and heat rate--
again, IPM inputs. The IPM model therefore directly develops SCR 
reactor cost based upon all of the same inputs, but using algorithms 
based upon a database of the cost of actually constructed units. Other 
costs, such as preproduction costs and initial inventory of ammonia are 
incorporated into the capital costs. In any event, these are very small 
portions of total cost. Initial catalyst cost is incorporated into the 
direct capital costs of the project in the IPM estimate, and in any 
event, initial catalyst cost is typically a very small portion of total 
capital cost.
    Commenter is critical of EPA's treatment of the associated costs 
for replacement catalyst. Regarding treatment of catalyst cost, EPA's 
approach is reasonable, factoring in the NOX reduction, 
coal, facility size, capacity factor, and catalyst cost--all of the 
same factors that impact catalyst replacement cost. PacifiCorp, on the 
other hand, uses excessive catalyst replacement costs. PacifiCorp's use 
of $290/ft\3\ for replacement catalyst cost equates to $10,422/m\3\. 
This is nearly double the current cost of new catalyst. The difference 
cannot be accounted for in the labor to install catalyst, which is a 
very small fraction of the catalyst cost.\162\ Moreover, replacement 
catalyst is frequently regenerated catalyst which costs roughly half 
the cost of new catalyst. EPA conservatively assumed new catalyst at 
roughly $5,500/m\3\,\163\ when in fact most catalyst replenishments 
will likely be at a much lower cost due to the extensive availability 
of regenerated catalyst. As a result, EPA's catalyst costs are much 
more reasonable and are in fact, conservative in light of the 
availability of regenerated catalyst.
---------------------------------------------------------------------------

    \162\ From data in Cichanowicz, J., Muzio, L., Hein, M., ``The 
First 100 GW of SCR in the U.S.,--What Have We Learned?''--2006 Mega 
Symposium, page 4.
    \163\ Cichanowicz, J.E., ``Current Capital Cost and Cost-
Effectiveness of Power Plant Emissions Control Technologies'' 
Prepared for, Utility Air Regulatory Group, January 2010.
---------------------------------------------------------------------------

    Commenters are critical of EPA's estimates of SNCR reagent 
consumption. Because of the importance of reagent consumption on SNCR 
system capital and operating cost, chemical utilization is an important 
factor. Utilization is a measure of how efficiently the SNCR reagent 
reduces NOX. With SNCR, NOX reduction does not 
occur on a one-for-one basis as reagent is added because a portion of 
the chemical introduced does not contribute to NOX 
reduction. The utilization of reagent (normally urea) declines as 
temperature (or carbon monoxide concentration) increases because more 
of the urea becomes oxidized (forming NOX), which reduces 
the amount of reagent available for the NOX reduction 
reaction. The net difference between the amount of NOX 
reduced and the amount of NOX formed equals the overall 
reduction in NOX, and at a sufficiently high temperature, 
NOX can actually increase as urea is injected. Hence, 
furnace temperature is

[[Page 5144]]

a critical parameter in determining utilization.
    Chemical utilization is equal to the percent NOX 
reduction divided by the treatment rate, expressed in terms of 
Normalized Stoichiometric Ratio. The IPM model assumes a default 
chemical utilization rate of 15%. Commenter states that Andover 
adjusted the urea utilization rate in the IPM model without 
explanation. On the contrary, on page 3 of its memo, Andover provided 
actual utility data and explained why 25% was a more common utilization 
rate for utility boilers. Chemical utilization is a strong function of 
furnace temperature, and because chemical consumption is the single 
largest expense associated with SNCR, using an incorrect chemical 
utilization will lead to large discrepancies in cost.
    Commenter states that the furnace exit gas temperature is 
2710[emsp14][deg]F. While EPA suspects that this may be in error 
(high), it is the data provided by the company. If this temperature is 
correct, SNCR may only be marginally effective. Sargent & Lundy 
estimated a chemical utilization of 12%, which is possibly optimistic. 
Information from Fuel Tech, the leading supplier of SNCR technology, 
stated that at the unusually high furnace exit temperature of 
2500[emsp14][deg]F, well below 2710[emsp14][deg]F assumed by Sargent & 
Lundy, a 25% reduction was possible at a particular facility from a 
similar NOX baseline with a Normalized Stoichiometric Ratio 
of 1.75 (utilization of 14.3%). Hence, EPA will accept the 12% 
utilization and 20% NOX reduction assumed by Sargent & Lundy 
for Laramie River Station.\164\
---------------------------------------------------------------------------

    \164\ Email from Alexander Dainoff to Jim Staudt, June 27, 2012.
---------------------------------------------------------------------------

    The model in the IPM, which was developed by Sargent & Lundy, 
assumes a utilization rate of 15%, which Andover's data, as well as 
Fuel Tech's input, demonstrates is too low for most facilities. It is 
therefore a very conservative estimate of chemical utilization. In 
fact, Andover reviewed Sargent & Lundy's model when it was developed 
for EPA and Sargent & Lundy did not provide any supporting data 
regarding the assumed utilization rate at that time. However, in light 
of the high assumed furnace temperature at Laramie River Station, EPA 
is assuming a 12% chemical utilization at a 20% NOX removal 
rate for those units.\165\
---------------------------------------------------------------------------

    \165\ Exhibit 16 to Basin Electric comment, page 25.
---------------------------------------------------------------------------

    On the other hand, in the October 28, 2013 memo by Andover for EPA, 
test data for utility boilers was provided showing that 25% is a more 
reasonable utilization rate for most units. Moreover, Andover's 
principal has conducted numerous electric utility SNCR optimization 
programs, among them are programs described in referenced papers, which 
makes him qualified to make this determination.166 167 168 
So, in applications where more typical furnace temperatures are 
expected, a chemical utilization rate closer to 25% can be reasonably 
assumed.
---------------------------------------------------------------------------

    \166\ Staudt, J., Casill, R., Tsai, T., Ariagno, L., 
``Commercial Application of Urea SNCR for NOX RACT 
Compliance on a 112 MWe Pulverized Coal Boiler'', 1995 EPRI/EPA 
Joint Symposium on Stationary Combustion NOX Control, 
Kansas City, May 16-19, 1995.
    \167\ Staudt, J, Hoover, B., Trautner, P., McCool, S., Frey, J., 
``Optimization of Constellation Energy's SNCR System at Crane Units 
1 and 2 Using Continuous Ammonia Measurement'', The MEGA Symposium, 
Baltimore, MD, August 31-September 2, 2010.
    \168\ Staudt, J.E., Kehrer, K., Poczynec, J., Cote, R., Pierce, 
R., Afonso, R., and Sload, A., ``Optimizing Selective Non-Catalytic 
Reduction Systems for Cost-Effective Operation on Coal-Fired 
Electric Utility Boilers'', presented at ICAC Forum '98, Durham, 
March 1998.
---------------------------------------------------------------------------

    It is also likely that, in light of the higher furnace exit 
temperature at Laramie River Station, it will be necessary to utilize a 
more complex and costly injection system that requires injection of 
urea into convective regions of the furnace using multiple-nozzle 
lances in addition to the more typical wall injectors. For this reason, 
EPA is accepting the capital cost developed by Sargent & Lundy for 
Basin Electric of $16.9 million per unit.
    Regarding the reagent used for SCR, as the commenter points out, 
EPA's initial estimates assumed that urea would be used as the feed 
reagent for SCR, which results in a higher reagent cost than for 
anhydrous ammonia. Because commenters have indicated that anhydrous 
ammonia will be used as a reagent rather than urea, EPA's revised 
estimates assume ammonia as a reagent. This will result in lower 
reagent costs. EPA is also using anhydrous ammonia costs provided by 
Potash Corporation instead of the value provided by the utility.\169\ 
The cost used by EPA represents the actual delivered cost of anhydrous 
ammonia, as quoted by a major reagent supplier.
---------------------------------------------------------------------------

    \169\ Email from Potash Corporation to Andover Technology 
Partners, September 27, 2013.
---------------------------------------------------------------------------

    Basin Electric provided a site-specific estimate. EPA generally 
supports the use of vendor quotes and site-specific estimates but only 
as used within the parameters of the overnight cost methodology and the 
CCM. The BART Guidelines, are clear that ``[y]ou 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 CCM.'' \170\ When supporting 
documentation to site estimates are not provided, assumptions based 
upon these cannot be considered.
---------------------------------------------------------------------------

    \170\ 70 FR 39166, footnote 15.
---------------------------------------------------------------------------

    Much of the documentation owners cite to support additional costs 
were not provided to us. For instance, although Basin Electric provided 
a table that listed their cost line items, this spreadsheet (in pdf 
format), over 600 lines in length (and including line items such as 
4 gravel surfacing and chain link fence), was stripped of 
all cell calculations, preventing any meaningful review.
    There was also inadequate explanation for how man-hour rates were 
developed. For example, current union boilermaker rates for Lodge 101 
(Denver) range from $57.62 per hour for a Journeyman to $60.12 per hour 
for a Foreman, with apprentices at lower rates.\171\ The CLRC reports 
similar union boilermaker rates in the Mountain-Northern Plains area at 
$56.59/hr for July 2013.\172\ For non-union boilermakers, the cost is 
expected to be less. Yet, for tasks that appear to be for boilermakers 
(ductwork) rates of $90.79 per man-hour are used.\173\ This is a large 
unexplained difference in cost. It may be that part of it is per diem 
($55/day for over 70 miles and $70/day for over 120 miles per Lodge 101 
information), but per diem does not explain the full difference. Per 
diem, however, is also provided as a separate cost later in the 
estimate, making the high labor rate more difficult to explain. Some 
portion may be for overtime, but there is no explanation provided 
(overtime is also added as a separate line item later in the estimate). 
Without an explanation it is not possible to evaluate these costs, 
which clearly deviate from publicly available costs for labor. In 
addition to not providing the necessary required supporting 
documentation, Basin Electric did not follow the overnight cost 
methodology. Thus, Basin Electric has failed to meet the test that is 
required to support deviations from the CCM.
---------------------------------------------------------------------------

    \171\ Wage & Benefit Information, Western States Field 
Constructions Bargaining Agreement, Effective January 1, 2013 
through September 31, 2013.
    \172\ Construction Labor Research Council Union Wages and 
Supplements for the Period 07/01/2008-07/01/2013. Available from the 
National Association of Construction Boilermaker Employers Web site: 
www.nacbe.org.
    \173\ Appendices to Exhibit 14. Page 6 of SCR estimate (pdf page 
43).

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

[[Page 5145]]

    Comment: The IPM methodology relied upon in the Andover Report is 
inconsistent with the BART Guidelines. The Guidelines require that to 
conduct a cost analysis, a state must ``[i]dentify parameters'' for 
emission control and then ``[d]evelop cost estimates based upon those 
design parameters.'' 70 FR 39166. The cost analysis includes 
development of ``estimates of capital and annual costs,'' based on the 
CCM ``where possible,'' but ``tak[ing] into account any site-specific 
design or other conditions identified above that affect the cost of a 
particular BART technology option.'' Id. Andover failed to adhere to 
the methodology set forth in the CCM ``where possible.'' More 
importantly, however, Andover neither followed the three-step process 
in the BART Guidelines for estimating costs of compliance nor 
appropriately considered the critical site- and project-specific 
variables that affect the cost of both SCR and SNCR at Laramie River. 
EPA's failure to comply with its own Guidelines results in an 
inaccurate cost estimate that should not form the basis of a BART 
determination.
    Response: EPA disagrees with the commenter. The three step process 
was followed for all cost estimates for affected Wyoming units. The 
three-step process requires that states (or EPA): (1) Identify the 
emissions units being controlled; (2) Identify design parameters for 
emission controls; and (3) Develop cost estimates based upon those 
design parameters. The BART affected units were already identified by 
the State and confirmed by EPA, which addresses step one.
    Andover clearly identified design parameters that are included in 
the spreadsheets associated with the memo under the tab ``Plant Data 
Summary''. This includes plant data such as fuel, capacity, capacity 
factor, heat rate, baseline and controlled NOX level, 
retrofit factor and firing configuration. These parameters are directly 
used in the cost analysis that developed capital and annual costs. 
Andover performed the cost estimates as described in the spreadsheet. 
Andover considered site and project-specific parameters as described in 
other responses to comments.
    Commenters may disagree with the cost methodology used by EPA, and 
our response to comments regarding the cost methodology used is in 
other responses; however, there is no question that EPA followed the 
three-step process.
    Comment: The IPM methodology relied upon in the Andover Report is 
inconsistent with the three-step process in the BART Guidelines for 
estimating costs of compliance. The three-step process in the BART 
Guidelines requires: (1) Identification of emission units being 
controlled; (2) identification of design parameters for emission 
controls; and (3) development of cost estimates based on those design 
parameters. 70 FR 39166. The Andover Report did not adequately define 
the emission units being controlled, failed to identify appropriate 
site-specific design parameters that affect cost and performance of 
these controls; and developed cost estimates that are neither 
technically indefensible nor representative of the costs of SCR and 
SNCR systems at Laramie River.
    Andover's reliance on the overly simplistic IPM model precluded an 
appropriate BART analysis. First, because the IPM cost algorithms are 
designed to provide high-level system cost, Andover used generalized 
information regarding design and baseline operating conditions at the 
Laramie River boilers to identify the emission units to be controlled. 
Second, because the IPM model includes only four inputs, Andover's cost 
analysis could not account for unit-specific operating parameters that 
affect both design of the control system and the attendant costs of 
installing the controls.
    As a result, the Andover Report failed to comply with the 
requirement in the BART Guidelines that cost estimates account for 
site-specific variables. 70 FR 39166. The site-specific design and 
operational variables have an important effect on the costs of 
NOX control technologies at Laramie River, particularly the 
installation of SCR. Finally, because Andover used generic inputs to an 
overly simplistic model, the resulting cost estimate is not technically 
defensible. As Sargent & Lundy opined ``[b]ased on our evaluation of 
the Andover cost estimates, it is our opinion that cost estimates 
prepared by Andover are not reflective of the costs BEPC would incur to 
install air pollution control systems on LRS Units 1, 2, & 3, and that 
control technology costs included in the February 7, 2013 Andover 
Report should not be used to determine BART for the Laramie River 
generating units.''
    Response: EPA disagrees with the commenter. The three step process 
requires that states (or EPA): (1) Identify the emissions units being 
controlled; (2) Identify design parameters for emission controls; and 
(3) Develop cost estimates based upon those design parameters.
    Commenter states that ``The Andover Report did not adequately 
define the emission units being controlled, failed to identify 
appropriate site-specific design parameters that affect cost and 
performance of these controls; and developed cost estimates that are 
neither technically indefensible nor representative of the costs of SCR 
and SNCR systems at LRS.'' However, the BART affected units were 
clearly identified and defined in the analysis. The units are 
identified and described in the memo and the associated spreadsheets.
    EPA's contractor clearly identified design parameters that are 
included in the spreadsheet associated with the memo under the tab 
``Plant Data Summary''. This includes plant data such as fuel, 
capacity, capacity factor, heat rate, baseline and controlled 
NOX level, retrofit factor and firing configuration. These 
parameters are directly used in the cost analysis that developed 
capital and annual costs. This is the very same data as used by Sargent 
& Lundy in their analysis, except that EPA did not correct for 
elevation, which EPA is correcting in this response to comments and 
calculations.
    Andover then performed the cost estimates as described in the 
spreadsheet using algorithms developed by Sargent & Lundy that utilize 
the same inputs as used by Sargent & Lundy in their analysis for 
commenters. Other than a site visit and an adjustment for elevation, 
commenters have not identified any other inputs that they used that are 
different or not the direct result of the inputs utilized by EPA. With 
this response to comments EPA is correcting cost estimates for 
elevation and EPA has 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.
    We disagree with commenter's characterization of the cost 
development methodology contained in IPM as generalized. As noted in 
the documentation for IPM's cost development methodology for SCR,\174\ 
the cost estimate methodology is based upon two databases of actual SCR 
projects. These databases include 2004 and 2006 industry cost estimates 
prepared for the Midwestern Ozone Group, and a proprietary in-house 
database maintained by Sargent & Lundy. The Midwestern Ozone Group 
information was cross-referenced with actual 2009 projects, and 
escalated accordingly. Sargent & Lundy then used the information in 
these databases to develop the equations described in the cost 
component taking into account the

[[Page 5146]]

pre-control NOX emission level, degree of reduction, coal 
type, facility size, and numerous other unit-specific factors. While a 
costly engineering evaluation that included site visits in addition to 
our use of satellite imagery would potentially produce a more refined 
cost estimate, we disagree that our approach has produced cost 
estimates that do not adequately address site-specific issues. As noted 
by EPA in our responses related to satellite imagery, EPA's use of 
satellite imagery enabled us to evaluate each of the major site-
specific issues raised by commenters.
---------------------------------------------------------------------------

    \174\ http://www.epa.gov/airmarkets/progsregs/epa-ipm/docs/v410/Appendix52A.pdf.
---------------------------------------------------------------------------

    Comment: The Andover Report fails to consider site-specific 
conditions as required by the BART Guidelines. The BART Guidelines and 
the CCM emphasize the importance of taking into account site-specific 
conditions in developing a cost estimate, particularly with respect to 
construction of SCR. 70 FR 39166 (cost estimates should ``into account 
any site-specific design or other conditions . . .''). See also CCM 
section 2.4, page 2-40 (with respect to SCR, ``[a]ctual selection of 
the most cost-effective option should be based on a detailed 
engineering study and cost quotations from system suppliers.''); id. 
section 2.5.4.1, page 2-27 (``Manual deviates from its standard 
approach of providing study level costs [for SCR] and, instead, 
provides a detailed description of the factors'' influencing costs). 
Yet as Andover acknowledges in its report, its engineers did not visit 
Laramie River and had no engineering plans, process flow diagrams, or 
other site-specific information regarding Laramie River when it 
developed EPA's cost estimate. The only information Andover collected 
regarding the site was the generating capacity of the station, annual 
heat input for a baseline period, NOX emission rates for 
certain years, and the type of coal burned. As a result, Andover's cost 
estimation methodology fails to comply with the BART Guidelines or 
follow the methodology recommended by the CCM, and the final estimate 
radically underestimates the cost for SCR and SNCR at Laramie River 
Station.
    Moreover, Andover's use of the IPM model compounded its failure to 
review site-specific considerations relevant to costing SCR or SNCR at 
Laramie River Station. Indeed, the fact that the model has only four 
input parameters, and does not take into account other site-specific 
parameters that are required by the BART Guidelines and recommended by 
the CCM, renders any resulting cost estimate both technically and 
legally deficient. As noted by Sargent & Lundy, which developed the IPM 
algorithms for SCR and SNCR, ``[b]ecause of the limited number of site-
specific inputs, the IPM cost algorithms provide only order-of-
magnitude control system costs, and do not provide case-by-case 
project-specific cost estimates meeting the requirements of the BART 
Guidelines.''
    By relying on the IPM model, Andover deliberately skirted the issue 
of site-specific conditions, other than the most generic inputs of unit 
size, heat rate, coal type, and a retrofit factor. The Andover Report 
describes the retrofit factor in such a way as to emphasize its site-
specific nature, and yet makes no attempt to carefully analyze the 
site-specific variables inherent in the application of the retrofit 
factor: ``The retrofit factor is a subjective factor used to account 
for estimated difficulty of the retrofit that is unique to the 
facility. Because site visits were not possible, the retrofit factor 
was estimated from satellite images that provide some insight to the 
configuration of the units and degree of congestion around the site and 
in the vicinity of where the SCR would be installed. These factors 
impact the ability to locate large cranes on the site--that impact how 
the SCR is assembled (are large sections lifted into place or is the 
SCR ``stick built''), how much duct work is needed, if the SCR must be 
built onto a large, elevated steel structure or can be built near the 
ground, and if other equipment must be relocated to accommodate the 
space of the SCR.''
    The Andover Report never provides any analysis of these site-
specific factors in determining the appropriate retrofit factor for 
Laramie River. Andover merely notes that the difficulties of 
retrofitting an SCR unit at Units 1 and 3 is ``average,'' and applies a 
retrofit factor of 1.0 to Units 1 and 3. Andover does note that 
retrofit at Unit 2 will be more difficult and added an adjustment for 
retrofit difficulties at Unit 2, but with no explanation of what unique 
site-specific conditions contributed to the factor. Instead, the report 
notes ``a modestly higher SCR retrofit difficulty factor is assumed for 
Unit 2 because access will be poorer than for Units 1 or 3.'' In sum, 
Andover makes no adjustments for Units 1 or 3 and accords a 20% cost 
retrofit factor to Unit 2, though it had previously noted that ``when 
using the IPM Capital Cost Model, retrofit difficulties associated with 
an SCR may result in capital cost increases of 30% to 50% over the base 
model.'' Thus, even when it accords a retrofit factor, as it did for 
Unit 2, Andover inexplicably low balls that retrofit factor and the 
attendant cost increase for the SCR system. When Andover's retrofit 
factors are compared across units and across facilities, it appears 
that Andover arbitrarily assigned the retrofit factors without 
adequately accounting for site congestion and constructability issues. 
Subjective retrofit factors, especially factors randomly chosen without 
knowledge of site conditions, cannot account for site-specific 
circumstances as provided in the BART Guidelines.
    A comparison of the broad brush approach taken by Andover with the 
detailed, site specific, scoping-level study taken in the Sargent & 
Lundy evaluation illustrates the inadequacies with Andover's 
methodology. As an initial matter, Andover made a fundamental error by 
failing to account for the effect of site elevation on the project 
costs. Laramie River Station is situated at 4,750 feet above sea level, 
resulting in flue gas volumes approximately 20% greater than a 
similarly sized unit at sea level. These larger flue gas volumes will 
require larger SCR reactors, larger duct work, and increased structural 
support. None of these additional costs are accounted for in Andover's 
estimate because the IPM model assumes the plant is located at sea 
level. Indeed, IPM model guidance provides that ``elevation of the site 
must be considered separately and factored into the MW size accordingly 
due to its effects on the flue gas volume.'' Id. (citing the IPM SCR 
Cost Development Document (August 2010)).
    Andover's reliance on the IPM model also resulted in failure to 
account for a regional labor productivity factor. Regional productivity 
must be taken into consideration in a site-specific cost estimate to 
account for local workforce characteristics, labor availability, 
project location, project complexity, local climate and working 
conditions. This is a key factor for Laramie River Station, because 
experienced, knowledgeable labor is difficult to acquire and requires 
premium pay, further adding to the cost of an SCR system.
    The most substantial failing of the Andover Report, however, is its 
reliance on an aerial photograph of the plant from Google 
EarthTM to account for site-specific conditions at Laramie 
River Station. There are numerous important elements that cannot be 
discerned from an aerial photograph. Specifically, a Google 
EarthTM photograph cannot identify: (1) The site constraints 
posed by the location of the coal conveyor rooms; (2) the location of 
the existing fan buildings and the space constraints between the 
existing fan buildings and the existing electrostatic precipitators; 
(3) the necessary information to determine duct work routing and SCR 
tie-ins to the existing economizers and

[[Page 5147]]

air heaters; and (4) information regarding plant subsystems such as the 
fan capacity, equipment reinforcement, auxiliary power systems, 
electrical plant system capacity, and other plant subsystems that would 
be affected by installation and operation of the SCR systems. Further, 
the aerial photograph provides no information about where the ammonia 
handling system could be located and necessary pipe routing and other 
support systems for the ammonia handling system. Nor does it show the 
need to penetrate the 20-story boiler wall and provide related 
structural support to install duct work, to provide structural support 
columns for the SCR reactors in very tight spaces, and the need for 
special cranes to lift heavy equipment into place in a congested space.
    While Andover indicated that some of these site-specific issues are 
addressed by the retrofit factor, the fact that Andover accorded no 
retrofit factor to Laramie River Station Units 1 and 3, and low balled 
the retrofit factor for Unit 2, resulted in a failure to include site-
specific costs in its estimate for the Laramie River, in direct 
contravention of the requirements of the BART Guidelines and 
suggestions of the CCM. Site-specific conditions are illustrated in 
Section 5.1.1.1 of the Sargent & Lundy Critique. Finally, Andover 
failed to include costs for the balance of plant systems required for 
the SCR. Sargent & Lundy Critique section 5.1.1.3. These items, which 
require enlarging existing plant systems to provide for the additional 
power and airflow and other systems necessary to operate the SCRs, 
include the following: (1) Replacement of induced draft fans by larger 
fans to support the SCR units; (2) Upgrading of the existing electrical 
system to support the SCR units; (3) Structural stiffening of the duct 
work downstream of the SCRs; and (4) Expand existing control system to 
accommodate six new SCRs (two for each generating unit).
    Wyoming used actual, site-specific data regarding the BART-eligible 
sources in development of its plan. In contrast, EPA did not use site-
specific data; instead, it relied on nothing more than aerial 
photographs available in the public domain.
    Response: EPA disagrees with the commenter. First, as discussed 
elsewhere in responses, we are no longer relying on the IPM cost 
algorithms, including the application of retrofit factors, to estimate 
capital costs for the Laramie River Station units. Instead, we have 
evaluated the cost information provided by Basin Electric in comments 
and incorporated it as appropriate. Therefore, the arguments made by 
the commenter related to our use of the IPM algorithm are no longer 
relevant. Nonetheless, below we discuss how our use of the IPM 
algorithm in the proposed rule was based on reason and evidence and 
addressed the site-specific concerns raised by the commenter.
    As noted in responses to other comments, EPA adequately addressed 
site-specific issues in using the IPM SCR cost model. The SCR cost 
model for IPM, being developed from actual SCR retrofit data, 
incorporate all of the costs that would normally be associated with an 
SCR retrofit. As such, retrofit issues that are common to all SCR 
retrofits are incorporated into the cost. To the extent that there are 
additional costs, as described in other comments, these can be 
addressed with the retrofit difficulty factor.
    Basin Electric did not follow the BART guidelines in developing 
their cost analyses, and importantly, did not provide adequate 
documentation when they deviated from it. There are only very limited 
situations under which an analyst can depart from the CCM methodology 
under the BART Guidelines: ``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 CCM, Fifth Edition, February 1996, EPA 453/B-96-001). In 
order to maintain and improve consistency, cost estimates should be 
based on the OAQPS CCM, where possible. The CCM 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.'' 70 FR 39166.
    This section of the BART Guidelines makes it clear that the CCM is 
the intended methodology for conducting a BART cost determination. It 
also states why: To maintain and improve consistency. However, the CCM 
does state that site-specific conditions should be incorporated. Site-
specific conditions could include space constraints, or a design 
feature that could complicate installing a control. Importantly, a 
footnote at the bottom of the same page of the BART Guidelines makes it 
clear that the analyses should document any deviations from the CCM: 
``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 CCM.''
    The record does not point to any ``unusual circumstance'' that 
explains why Basin Electric's SCR costs are higher than costs of SCRs 
at other similar facilities, other than the use of a different cost 
methodology. In fact, there is nothing in the record to support claims 
that the cost of SCR was in fact based on detailed site-specific vendor 
bids, or is in any manner more site-specific than the costs relied upon 
by EPA in our proposed rule. As an example, the BART application 
submitted by Basin Electric, and relied upon by Wyoming, shows that 
only the cost of catalyst is based on a vendor quote.\175\ There is no 
documentation to substantiate that the remaining costs are based on 
vendor quotes or any other site-specific data. The mere fact that the 
cost analysis was submitted by Basin Electric is not a basis to 
conclude that it is somehow highly site-specific. Indeed, even the 
updated cost information recently submitted by Basin Electric during 
the comment period is conceptual in nature and not based on vendor 
quotes. As stated on page 21 of Exhibit 14 to their comments: ``The LRS 
cost estimates are conceptual in nature; thus, S&L did not procure 
equipment quotes specifically for the LRS control systems. Rather, 
equipment costs for the LRS projects are based on conceptual designs 
developed for the control systems, preliminary equipment sizing 
developed for the major pieces of equipment, and recent pricing for 
similar equipment.''
---------------------------------------------------------------------------

    \175\ Wyoming Regional Haze SIP, Attachment A, ``Basin Electric 
Power Cooperative Laramie River Station Refined BART Visibility 
Modeling'', prepared for Basin Electric Power Cooperative by Black & 
Veatch Corporation, July 24, 2008, Attachment 1, page 7 of 7.
---------------------------------------------------------------------------

    Commenter correctly notes that EPA did not account for elevation. 
EPA acknowledges that it did not account for elevation in the estimate 
when using the IPM algorithm and EPA's revised estimate does account 
for elevation. Commenter states that the regional labor productivity 
was not factored into EPA's estimate. EPA's estimate did provide an 
allowance for overtime which is a line item in the estimate labeled 
``Labor Adjustment.'' However, commenter did not provide sufficient 
information to evaluate commenter's estimate and how productivity 
factors were developed or applied in their estimate to produce their 
estimate. Labor costs comprise roughly half of the total cost of Basin 
Electric's estimate of what SCR would cost to install at Laramie River 
Station, and the significance of this cost makes the lack of 
information very important.

[[Page 5148]]

Moreover, if this is an important element of commenter's critique of 
EPA's method, they should have provided sufficient data and supporting 
justifications for EPA to evaluate commenter's estimate. Notably, this 
is an important deviation from the CCM and more detailed supporting 
data should have been provided. As noted in EPA's responses to other 
comments, the commenter has not provided any data to explain the high 
labor charge rates or claimed low productivity, as required under the 
BART Guidelines.
    Commenter claims that ``a Google Earth\TM\ photograph cannot 
identify: (1) The site constraints posed by the location of the coal 
conveyor rooms''. On the contrary, the coal conveyors are prominent 
features in the images \176\ and it is clear from the coal conveyors 
where the coal conveyor rooms are located. This location for coal 
conveyor rooms is not unusual. In general, coal conveyor rooms are 
located either to one side of the boiler or the other, depending upon 
the location of the coal pile to the boiler. Moreover, the resulting 
need to route ductwork through the boiler building wall is commonly 
performed in SCR retrofits. As such, this is not an unusual issue and 
should not significantly impact retrofit cost versus a typical 
retrofit.
---------------------------------------------------------------------------

    \176\ 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., February 7, 
2013, Figures 6a and 6b.
---------------------------------------------------------------------------

    Commenter claims that EPA did not account for ``(2) the location of 
the existing FD fan buildings and the space constraints between the 
existing FD fan buildings and the existing electrostatic precipitators 
. . .'' It is clear from the images that there is a building located 
immediately below where the SCR reactor would be located, and this is 
not an unusual situation. The location of the ESP is also very clearly 
shown on the images. The space between the ESP and the fan rooms is 
also visible from the photo in Figure 6a of the Andover memo. 
Nevertheless, having to install SCR support structure in close 
proximity to existing buildings or equipment, as may be necessary at 
Laramie River, is not unusual because SCRs are commonly erected in that 
location and buildings or other equipment are normally in the area 
below. It is also possible that SCR support structure could be built to 
largely avoid the forced draft fan buildings by extending beyond them 
with a common support structure for all three, or at least more than 
one, SCR reactor. The space between the ESP and the fan buildings is 
only significant with regard to location of a crane for erection of 
Unit 2, and this is why a higher retrofit difficulty is assumed for 
Unit 2. Middle unit crane access issues are not unusual either and have 
been addressed numerous ways, including assembling a temporary crane 
above the boiler building, as was performed at the Brayton Point plant 
for their Unit 3.
    Commenter claims that EPA did not adequately account for ``(3) the 
necessary information to determine duct work routing and SCR tie-ins to 
the existing economizers and air heaters . . .'' EPA disagrees with the 
commenter. The general routing can be estimated from the images, and 
there was nothing in the images to suggest any problems with routing 
ductwork. It was apparent that a penetration was necessary in the 
boiler building, which is routinely necessary because boilers are 
typically housed in boiler buildings and the SCRs are not, making 
penetration of a boiler building wall necessary in any SCR retrofits 
where the boiler is in a boiler building. This is the case with most 
SCR retrofits. In fact, Figures 2 and 3 of Exhibit 16 to Basin 
Electric's comments demonstrate that the ductwork tie in to the boiler 
and associated modifications are similar to what is done routinely and 
ductwork length is reasonable. For an SCR cost estimate, the most 
critical cost issues that require closer examination than possible with 
a satellite photo is if it is necessary to move major pieces of 
equipment, such as an air preheater, in order to accommodate the 
ductwork. Had it been determined that relocating the air preheater was 
necessary, this would entail some significant additional cost over what 
would be considered a ``typical'' retrofit. Basin Electric did not 
indicate any such need and therefore their costs for ductwork are 
expected to be within the normal cost for a retrofit.
    Commenter claims that EPA did not properly account for ``(4) 
information regarding plant subsystems such as the FD fan capacity, 
equipment reinforcement, auxiliary power systems, electrical plant 
system capacity, and other plant subsystems that would be affected by 
installation and operation of the SCR systems. Sargent & Lundy Critique 
section 5.1.1.2.-5.1.1.3.'' The IPM algorithm explicitly assumes that 
it will be necessary to replace the forced draft (FD) fan \177\ and 
make modifications to auxiliary electrical systems and associated 
boiler structure and has a line item cost for this. As such this was 
addressed in the cost.
---------------------------------------------------------------------------

    \177\ Sargent & Lundy, ``IPM Model--Revisions to Cost and 
Performance for APC Technologies--SCR Cost Development Methodology, 
FINAL'', August 2010.
---------------------------------------------------------------------------

    Commenter claims that ``Further, the aerial photograph provides no 
information about where the ammonia handling system could be located, 
and necessary pipe routing and other support systems for the ammonia 
handling system. Nor does it show the need to penetrate the 20-story 
boiler wall and provide related structural support to install duct 
work, to provide structural support columns for the SCR reactors in 
very tight spaces, and the need for special cranes to lift heavy 
equipment into place in a congested space.'' EPA disagrees. It is 
apparent from the image, and the open spaces on the images that have no 
structures, that there are any of a number of places at the Laramie 
River site that the ammonia storage system could be located. Selecting 
the ideal location does require closer examination of the site than 
possible with an aerial photograph. However, the impact on total 
capital cost is relatively small. With regard to modifications to the 
boiler building, this has already been discussed along with the 
installation of structural support columns for the SCR and need for 
cranes. These modifications are routinely necessary for SCR retrofits 
and would be factored into the historical SCR projects that the IPM 
algorithms are based upon.
    Expansion of the controls is another cost identified by commenters 
as not adequately addressed by EPA. EPA disagrees with commenter. Every 
SCR retrofit requires expansion of controls. So, this is incorporated 
into the IPM model. There may be specific issues that may be associated 
with tailoring the controls to the existing site that make this portion 
of the cost slightly more or slightly less expensive than normal. But, 
controls are generally a small contributor to total SCR cost and these 
differences would have a minor effect.
    Comment: When all of the site-specific and balance of plant 
conditions are included, Sargent & Lundy estimated that the capital 
cost of installing SCR at Laramie River Station is $746,906,000. This 
is twice as much as Andover's flawed cost estimate of $330,000,000. The 
discrepancy in the cost estimates is not surprising in light of 
Andover's failure to comply with the BART Guidelines and to follow the 
CCM where appropriate. For example, the site congestion and balance-of-
plant upgrades alone total approximately $290 million--costs which 
accurately reflect site-specific constraints to installing SCR, but 
which were not accounted for in the Andover Report. While the costs 
estimated in the Sargent

[[Page 5149]]

& Lundy Evaluation are significantly greater than those set forth in 
the Andover Report, they are a far more accurate and representative 
assessment of the costs of installing SCR and SNCR at Laramie River 
Station.
    Response: EPA disagrees with the commenter, having found the 
following discrepancies in the commenter's estimates for cost of SCR 
and SNCR at Laramie River Station:
1. Apparently Double-Counts General Facilities
    The itemized cost estimate in the Appendices to Exhibit 14 includes 
items that are normally incorporated into the General Facilities cost, 
while Sargent & Lundy took an additional provision for General 
Facilities (Exhibit 14 at page 31). General Facilities are costs that 
are not directly associated with the process equipment, and include 
such things as access roads, access platforms, safety equipment (such 
as eye-wash stations), etc. On the other hand, ductwork, piping, 
structural steel to support process equipment are direct capital cost 
and do not fall into the category of General Facilities. The itemized 
cost estimate by Sargent & Lundy in the Appendices to Exhibit 14, 
however, includes the cost of many items that would normally fall under 
the category of General Facilities. This includes the cost of roads and 
a parking area ($930,226 at page 2 of SCR estimate), eye wash stations 
(page 3), a pre-engineered building for the construction warehouse 
($780,000 page 8), fire protection systems (page 16), gratings, 
handrails, ladders (page 22). As a result, Sargent & Lundy double 
counts for General Facilities by having these costs accounted for in 
the itemized direct capital cost as well as in its Indirect Capital 
costs.
2. Labor Rates in Sargent & Lundy's Itemized Capital Cost in the 
Appendices to Exhibit 14 Appear To Already Have a Significant 
Contingency Built in and Additional Costs for Overtime and Per Diem--
Already Incorporating Contingency in Apparent Double-Counting of 
Contingency
    According to Basin Electric's comments, their labor rates were 
taken from the publication RS Means.\178\ However, examples of how the 
labor rates from RS Means were used to develop what was ultimately used 
in their estimate were not provided. RS Means is a subscription service 
that can cost up to $1,100, depending upon the package.
---------------------------------------------------------------------------

    \178\ Exhibit 14, page 25.
---------------------------------------------------------------------------

    However, there are publicly available labor rates for the major 
construction trades from the CLRC that are available for download at 
the National Association of Construction Boilermaker Employers' Web 
site.\179\ The Laramie River Station estimate assumes boilermaker labor 
tasks with a labor rate of $90.79/hr while according to the CLRC 
boilermaker rates in the Mountain and Northwest Plains as of July 2013 
were $56.79/hour, which is close to the values for Lodge 101 of the 
boilermakers union, ranging from $57.62 for a Journeyman to $60.12 for 
a Foreman, and lower rates for apprentices.
---------------------------------------------------------------------------

    \179\ Construction Labor Research Council Union Wages and 
Supplements, available from the National Construction Boilermaker 
Employers Web site, www.nacbe.com.
---------------------------------------------------------------------------

    The Laramie River estimate assumes pipefitter labor tasks with a 
labor rate of $81.72/hour, while according to the CLRC pipefitter rates 
in the Mountain and Northwest Plains as of July 2013 were $43.57/hour. 
Plumber rates are only slightly higher are $47.47/hour. In fact, there 
are several areas where the Laramie River itemized estimated rates in 
the Appendices far exceed Mountain-Northern Plains reported rates for 
union craft labor for July 2013. The ratio of assumed rate versus 
reported is as high as 187%. It is only for the installation of an 
architectural door that the reported rate for carpenters is even close 
to the assumed rate for Laramie River Station. Carpenters also build 
concrete forms.\180\ For carpenters doing concrete forms, the paid rate 
($65.02--see page 2 of estimate) is more than double the union rate. 
For most of these crafts, these rates cannot be explained by per diem. 
For example, if all of the boilermakers lived 120 miles from the 
location and were eligible for $70/day per diem and also drove 120 
miles each way every day of an eight-hour shift receiving $0.565/mile, 
that would only increase the hourly rate by $25.70, which does not 
explain the $34+ difference.
---------------------------------------------------------------------------

    \180\ Bureau of Labor Statistics, 47-2031 Carpenters, 
www.bls.gov/oes/current/oes472031.htm.
---------------------------------------------------------------------------

    It may be that Sargent & Lundy applied an escalation to the labor 
charges for future expected rates. If so, this is inconsistent with the 
CCM, which does not allow for this. In light of the fact that labor 
comprises the single largest expense and is nearly half of the total 
direct cost of the project--per Basin Electric's estimate in the 
Appendices to Exhibit 14--the high labor rates assumed by Sargent & 
Lundy are critical cost items that require much more complete 
explanation than was provided.
3. Additional Labor Costs
    In addition to the high labor rates incorporated into Sargent & 
Lundy's itemized estimate, roughly $47 million in additional labor 
costs are included for five ten-hour days and six ten-hour days and per 
diem.\181\ This schedule (which results in overtime) is stated to be 
necessary to attract necessary labor.\182\ No further explanation is 
provided for these costs. It is unclear why it is necessary to offer 
these in light of the fact that power plant construction labor demand 
is well off of its peak and is especially low in the Western States. As 
a key power plant construction trade, boilermaker man-hours are a good 
indication of general power plant construction activity. Boilermaker 
man-hours demonstrate that labor demand is well off of past peaks, and 
for the first two quarters of 2013, boilermaker employment in the 
Western States is 18.6% below 2012 levels for the same period. 
Boilermaker man-hours in 2012 nationally totaled 27 million, well off 
the years of 2006 through 2009 that were all above 30 million, and 
peaked in 2008 at over 40 million. 2012 levels were still below 2010's 
rate of 28 million.\183\
---------------------------------------------------------------------------

    \181\ Appendices to Exhibit 14 of Basin Electric comments, page 
4 of SCR estimate.
    \182\ Exhibit 14, page 25.
    \183\ http://nacbe.com/manhour-reports/. Also see: Staudt, J.E., 
``Engineering and Economic Factors Affecting the Installation of 
Control Technologies--An update'', Developed for U.S. EPA Clean Air 
Markets Division, December 15, 2011, figure 2-2 on page 12.
---------------------------------------------------------------------------

    It is also unclear why such high expenses are needed for overtime 
and per diem, particularly in light of the high assumed wage rates 
discussed earlier. Moreover, the need for overtime needs to be 
incorporated into a discussion of schedule, which was not provided. The 
time available for installing the SCRs may allow for spreading of 
activities over longer periods of time than in past retrofit efforts 
that may have allowed less time than the RHR, which allows for five 
years. However, there is no discussion of the need for overtime in the 
context of schedule.
4. Quantity and Cost of Materials and Impact on Labor Hours
    Throughout the spreadsheets provided as Attachments to Exhibit 14 
to Basin Electric's comments, no information was provided on how the 
quantities of materials were estimated, such as tons of steel for 
ductwork, etc. This makes it impossible to evaluate if Sargent & Lundy 
estimated the correct quantities of materials, associated material 
costs, or the associated hours

[[Page 5150]]

associated with erecting the materials. The spreadsheet was provided as 
a pdf document, and therefore any underlying equations could not be 
examined. As noted in our comments to other questions, it is necessary 
to include documentation for any additional information used for the 
cost calculations that differs from the CCM. Since the quantities of 
materials also relate to the labor needed to install the materials, 
this also impacts the labor costs estimated by Sargent & Lundy.
5. AFUDC
    Sargent and Lundy includes AFUDC in its SCR cost. This cost of 
about $22-$23 million cost per unit, for a total of $68 million, is not 
permissible under the CCM as discussed in response to other comments.
6. Process Contingency
    Although the CCM shows an allowable process contingency of 5% for 
SCR, in EPA's opinion, this is not necessary today for SCR on coal-
fired boilers firing the coals used in Wyoming. According to the 
Department of Energy's NETL: \184\ ``Process contingency is intended to 
compensate for uncertainty in cost estimates caused by performance 
uncertainties associated with the development status of a technology. 
Process contingencies are applied to each plant section based on its 
current technology status.'' According to this document, for 
commercially available technologies process contingency could range 
from 0-10%. When the CCM was issued in January 2002, SCR was 
commercially available but was only emerging in application on coal-
fired utility boilers in the U.S. According to a study by 
Cichanowicz,\185\ at the end of 2001, there was only about 13,000 MW of 
coal SCR capacity in the U.S., with nearly all installed in the prior 
two years, meaning that there was very limited long-term experience 
with SCR on U.S. coals. SCR usage on coal-fired boilers has since 
increased about ten times to about 130,000 MW of coal capacity (over 
40% of all coal capacity) and is therefore a very well proven and well 
understood technology on a wide range of U.S. coals. As a result, the 
process contingency for SCR on coal-fired utility boilers should be 
much lower today than what it was when the CCM was issued in January 
2002--5%. EPA believes that for SCR applications on utility boilers 
burning Powder River Basin coals, which are very well understood SCR 
applications, there should not be a need for process contingency.
---------------------------------------------------------------------------

    \184\ U.S. Department of Energy, National Energy Technology 
Laboratory, ``Cost Estimating Methodology for NETL Assessments of 
Power Plant Performance'', DOE/NETL-2011/1455, April 2011, pg 4.
    \185\ From data in Cichanowicz, J., Muzio, L., Hein, M., ``The 
First 100 GW of SCR in the U.S.,--What Have We Learned?''--2006 Mega 
Symposium.
---------------------------------------------------------------------------

7. Sulfur Trioxide (SO3) Mitigation Measures Are Not Needed
    Sargent & Lundy assumed that a SO3 mitigation system is 
necessary for the Laramie River SCR. We disagree as this adds 
unnecessary capital and operating cost. An SO3 mitigation 
system is unnecessary because of the relatively low sulfur content of 
the coal and the fact that the coal fly ash is high in free calcium 
oxide. The available free lime will neutralize the SO3 
making SO3 mitigation unnecessary. In fact, in the model 
they developed for the IPM, Sargent & Lundy assumed that SO3 
mitigation was unnecessary for boilers using coals with SO2 
levels below 3 lb/MMBtu, making it unnecessary for Laramie River, which 
fires a much lower sulfur coal from the Dry Fork Mine, with an 
uncontrolled SO2 rate of roughly 0.50 to 1.0 lb/
MMBtu.186 187 188 In fact, Sargent & Lundy uses the same 
rationale for arguing (correctly) that air preheater modifications are 
not necessary to address potential ammonium bisulfate formation from an 
SNCR system.\189\ Moreover, even if SO3 formation were a 
legitimate concern, low oxidation SCR catalysts are available and any 
additional cost impact would be very small. SO3 mitigation 
is not a large capital cost compared to the other costs that contribute 
to the SCR, but it is illustrative of the manner in which Sargent & 
Lundy has taken efforts to overdesign the system while adding 
unnecessary costs.
---------------------------------------------------------------------------

    \186\ Sargent & Lundy, ``IPM Model--Revisions to Cost and 
Performance for APC Technologies--SCR Cost Development Methodology, 
FINAL'', August 2010, page 5.
    \187\ http://www.westernfuels.org/member-services/mining-operations
    \188\ BNSF Mine Guide, at www.bnsf.com/customers/pdf/mineguide.pdf. SO2 calculated by multiplying sulfur 
content (expressed as a decimal) times 2 million and dividing by the 
heating value in Btu/lb.
    \189\ Exhibit 14, page 17.
---------------------------------------------------------------------------

8. Labor Productivity Factor Apparently Not Site-Specific
    Basin Electric has commented that local labor productivity is a 
major factor that impacts cost. However, it appears that the labor 
productivity factor being selected at the site may have been broadly 
applied by Sargent & Lundy at multiple sites in an inconsistent manner. 
According to Basin Electric, ``Labor productivity accounts for things 
such as labor availability, site access and working conditions, 
climate, season changes, and project size and complexity'' and it is a 
common practice on large construction projects to apply a productivity 
factor to account for local worker productivity and construction site 
conditions. A labor productivity factor of 1.15 was selected to account 
for labor productivity in the southeastern Wyoming region as compared 
to the benchmark of 1.00 for Texas, cited in Basin Electric's comments. 
Although the comments refer to the Compass International Global 
Construction Cost and Reference Yearbook, the value in that document 
for the southeastern Wyoming region is not expressly stated, and it is 
unclear if 1.15 is, in fact, the value in that document as we were not 
able to confirm the number in the document referenced.\190\
---------------------------------------------------------------------------

    \190\ Exhibit 14, page 25.
---------------------------------------------------------------------------

    In another Sargent & Lundy BART analysis, performed for Oklahoma 
Gas & Electric Sooner Units 1 & 2 and Muskogee Units 4 & 5, Sargent & 
Lundy also used a Labor Productivity factor of 1.15.\191\ Since the 
Oklahoma Gas & Electric units are in a climate far more similar to 
Texas (only about 150 miles from Texas) than Wyoming, and the Oklahoma 
plant likely draws from similar construction labor pools as Texas, it 
seems that they should have a similar productivity factor as Texas. It 
is also unusual that Sargent & Lundy would select the exact same 
productivity factor for the Oklahoma BART analysis as Laramie River 
Station's BART analysis, although these facilities are roughly six 
hundred to seven hundred miles away, with very different climates and 
draw on different labor pools. In this case, it appears that Sargent & 
Lundy has used the same productivity factor for Laramie River Station 
as for other BART analyses. In summary, there is no evidence that the 
labor productivity factors suggested by the commenter are site-
specific.
---------------------------------------------------------------------------

    \191\ Oklahoma Gas & Electric, ``Sooner Units 1 & 2, Muskogee 
Units 4 & 5 Dry FGD BART Analysis Follow-Up Report'', December 28, 
2009, see pdf pages 28 and 43.
---------------------------------------------------------------------------

9. Contractor's Fees and Profit Are Excessive
    Commenter's estimate for contractor's fees Expense and Profit total 
nearly $51 million, or 14% of the estimated total Labor, Material, 
Subcontract, and Process Equipment.\192\ By contrast, Sargent & Lundy 
estimated for the IPM algorithm total contractor fees and profits of 
10% of the estimated Labor,

[[Page 5151]]

Material, Subcontract, and Process Equipment cost. When this higher 
percentage is combined with a high direct cost, the contractor's fees 
become excessive. The high contractor fees and profits assumed in the 
Laramie River Station estimate also seem inconsistent with a weak power 
plant construction market, as demonstrated by the boilermaker man-hour 
data discussed earlier.
---------------------------------------------------------------------------

    \192\ Attachments to Exhibit 14 to Basin Electric Comments, page 
4 of SCR estimate.
---------------------------------------------------------------------------

10. Labor, Material, Subcontract, and Process Equipment Costs in Excess 
of Historical Norms, With Substantial Additional and Unnecessary Costs 
Added
    It is not unusual for owners to report excessive costs because 
owners are most interested in a cost estimate that has a very low risk 
of an overrun rather than a 30% cost estimate, which has a 
higher risk of overrun (about 50%), but is likely to be a better 
estimate of actual project cost. Commenter's estimate for total Labor, 
Material, Subcontract, and Process Equipment is $361 million. Adding 
Sargent & Lundy's estimated cost of scaffolding, freight, and 
consumables that is in the Appendices to Basin Electric's Exhibit 14 to 
their estimated Labor and Material, Subcontract, and Process Equipment 
raises the cost to $383 million (Sargent & Lundy provides no supporting 
documentation for this $22 million in additional cost). This is 27% 
above the expected cost of $301 million (with elevation accounted for) 
developed from the IPM SCR model, which is developed from actual 
project data. This suggests that Sargent & Lundy made a fairly 
conservative estimate of these costs for Basin Electric. But, in 
addition to the cost of Labor, Material, Subcontract, and Process 
Equipment, scaffolding, consumables and freight, in their estimate for 
Basin Electric, Sargent & Lundy added very high costs for overtime, per 
diem (that were not explained as required) and high costs for 
contractor's fees and profits, and then added additional project and 
process contingencies, unnecessary costs such as SO3 
mitigation and un-allowed costs such as AFUDC that increased total 
project cost to nearly $750 million--about double what they had 
estimated in 2008.\193\ The combined effect of the conservative cost 
estimates with additional contingencies or unnecessary cost adders, 
results in what appears to be an unrealistically high cost.
---------------------------------------------------------------------------

    \193\ Attachments to Exhibit 14 to Basin Electric Comments, page 
4 of SCR estimate and Exhibit 14 page 31.
---------------------------------------------------------------------------

11. Project Contingency of 15% Is Too High in Light of the Method Used 
and Very Conservative Underlying Assumptions
    Because the cost estimates developed for Basin Electric are already 
very conservative, and based upon detailed estimates of the labor and 
materials to build the SCR, a 15% project contingency is excessive. 
According to the CCM, Section 1.1 Chapter 1, page 1-4: ``The accuracy 
of the information in the Manual works at two distinct levels. From a 
regulatory standpoint, the Manual estimating procedure rests on the 
notion of the ``study'' (or rough order of magnitude--ROM) estimate, 
nominally accurate to within  30%. This type of estimate is 
well suited to estimating control system costs intended for use in 
regulatory development because they do not require detailed site-
specific information necessary for industry level analyses.''
    The methods and cost elements of the CCM were adapted from the 
American Association of Cost Engineers (AACE) (CCM Section 1, Chapter 
2, p 2-5). AACE 16R-90 \194\ states that, ``Project Contingency is 
included to cover the costs that would result if a detailed-type 
costing was followed as in a definitive-type study.'' According to 
NETL,\195\ ``AACE 16R-90 states that project contingency for a 
``budget-type'' estimate (AACE Class 4 or 5) should be 15% to 30% of 
the sum of BEC, EPC fees and process contingency.'' AACE 18R-97 \196\ 
defines different classes of estimates, from 5 (least detailed) to 1 
(most detailed). The methodology used in the CCM falls into a Class 4 
or Class 5, while the methodology used by Basin Electric in their 
comments, with hundreds of line items and thousands of input 
parameters, is clearly a far more detailed estimate that does not leave 
out any aspect of the project.
---------------------------------------------------------------------------

    \194\ AACE Recommended Practice, AACE 16R-90; www.aacei.org/non/rps/18R-90.pdf.
    \195\ U.S. Department of Energy, National Energy Technology 
Laboratory, ``Cost Estimating Methodology for NETL Assessments of 
Power Plant Performance'', DOE/NETL-2011/1455, April 2011, pg 5.
    \196\ AACE Recommended Practice, AACE 16R-87; www.aacei.org/non/rps/18R-97.pdf.
---------------------------------------------------------------------------

    The 15% project contingency factor in the CCM for SCR shown in 
Table 2.5 on page 2-44 of Section 4.2 Chapter 2 is based upon use of 
the cost estimating method described in the CCM to develop the Total 
Direct Capital Costs. It is not intended to apply to a detailed 
estimate that: (1) Includes many cost items not explicitly included in 
the estimating method described in the CCM to develop the Total Direct 
Capital Costs and meant to be included in the 15% project contingency, 
and (2) Already has substantial contingency built into it through 
conservative assumptions.
    In fact, the CCM discusses the importance of not double-counting 
contingency in multiple places such as retrofit factor and contingency 
at page 2-30 of Chapter 2--Cost Estimation: Concepts and Methodology: 
``Due to the uncertain nature of many estimates, analysts may want to 
add an additional contingency (i.e., uncertainty) factor to their 
estimate. However, the retrofit factor is a kind of contingency factor 
and the cost analyst must be careful to not impose a double penalty on 
the system for the same unforeseen conditions. Retrofit factors should 
be reserved for those items directly related to the demolition, 
fabrication, and installation of the control system. A contingency 
factor should be reserved (and applied to) only those items that could 
incur a reasonable but unanticipated increase but are not directly 
related to the demolition, fabrication, and installation of the system. 
For example, a hundred year flood may postpone delivery of materials, 
but their arrival at the job site is not a problem unique to a retrofit 
situation.'' (emphasis added). The CCM, therefore, explicitly 
anticipates that some analysts may, incorrectly, apply multiple 
contingencies for the same areas of uncertainty even when using the 
methods described in the CCM for estimating Total Direct Capital Costs.
    Because the cost estimates developed for Basin Electric are already 
very conservative and based upon detailed estimates of the labor and 
materials to build the SCR, rather than study-level estimates, they 
have double-counted both the costs that are intended by the CCM to be 
included in the project contingency when using the CCM method, plus 
they have added additional contingency in the form of conservative 
assumptions to address uncertainties in their estimate. For this reason 
a 15% project contingency is excessive for their estimate.
12. The Cost Information in Exhibit 14 Does Not Appear To Be Consistent 
With the Cost Information in the Appendices to Exhibit 14
    The table on page 4 of the SCR cost estimate for Units 1-3 shows a 
total cost of $481 million. This is inconsistent with the Total Direct 
Capital Costs shown on page 31 of Exhibit 14, which total $465 million. 
It is unclear what the cause of the $16 million difference is. In 
either case, EPA believes that the cost is not adequately explained.

[[Page 5152]]

13. Laramie River Station Does Not Require an SNCR System With Four 
Injection Zones
    The Laramie River Station is a base loaded unit, with capacity 
factors well above 80%. This means that the boiler rarely operates at 
part load. Sargent & Lundy designed the SNCR system with four injection 
zones to accommodate the ``entire load and temperature range within the 
boiler''.\197\ Because the unit will rarely operate at part load and 
the emission rate is a 30-day average, there is likely no need for four 
injection levels. Four injection levels are only required on load-
following units that spend a significant amount of time at low or 
middle loads or units that must comply with emission limits of much 
shorter averaging times, such as 24 hour averages or less. In practice, 
this system would be designed with two, or, at most, three injection 
zones. The additional injection zone adds cost in the form of 
additional injectors and furnace penetrations, and associated labor. On 
the other hand, EPA has accepted Basin Electric's estimate of the cost 
of the SNCR system. Although we believe that there would likely be 
fewer injection levels, based upon the furnace exit temperature 
reported by the company, we expect that at least one of the injection 
levels will require a convective zone injection system using multi-
nozzle lances, which will increase the cost. Therefore, these effects 
offset one another and we are accepting the cost provided by Basin 
Electric.
---------------------------------------------------------------------------

    \197\ Exhibit 14, page 17.
---------------------------------------------------------------------------

14. Similar Labor Rate Issues for SNCR as for SCR Estimate
    Examination of the labor rates for the Sargent & Lundy cost 
estimate revealed that Sargent & Lundy assumed the same high labor 
rates for crafts as they did for SCR without the rates explained 
sufficiently. They also assumed an additional $2.7 million in 
additional overtime and per diem rates that are not explained.\198\
---------------------------------------------------------------------------

    \198\ Appendices to Exhibit 14 of Basin Electric comments, page 
4 of SNCR estimate.
---------------------------------------------------------------------------

    For these reasons, the Sargent & Lundy capital cost estimates for 
SCR and SNCR are deficient, for the reasons as described above. 
However, because EPA expects that the SNCR injection system necessary 
for Laramie River Station may require more costly multiple-nozzle 
lances in at least one injection zone, this should offset the cost 
impact of the deficiencies we have identified, and we are accepting the 
capital cost of the SNCR system provided by Basin Electric equal to 
$16.9 million per unit.
    For the SCR capital cost at Laramie River, EPA is accepting some 
costs and not others, as described in more detail in supporting 
information with these comments.\199\
---------------------------------------------------------------------------

    \199\ Andover Technology Partners, ``Cost of NOX 
Controls on Wyoming EGUs'', October 28, 2013; Wyoming EGU BART and 
Reasonable Progress Costs--10/28/2013.
---------------------------------------------------------------------------

    Comment: We are very concerned to see that EPA has introduced a 
retrofit factor greater than ``1'' (the default) for 13 of the 15 EGUs 
evaluated. The IPM model used by EPA to estimate control costs in 
Wyoming already includes retrofit costs in its costing algorithms. It 
is generally accepted that retrofit projects will incur costs over and 
above those for a ``greenfield'' site, and most of those retrofit costs 
are already included in the database used to generate the IPM 
algorithms. So, unless a particular situation is so extreme as to 
warrant an additional retrofit factor, applying a retrofit factor to an 
algorithm that already includes retrofit costs is double counting those 
costs. Not only is the application of a retrofit factor not mentioned 
in the Federal Register Notice, its only supporting documentation 
appears in docket item EPA-R08-OAR-2012-0026-0086[1], ``Review of 
Estimated Compliance Costs for Wyoming Electricity Generating Units 
(EGUs)--revision of previous memo'': ``Selective Catalytic 
NOX Reduction (SCR) capital cost is estimated using the IPM 
algorithms with retrofit factors adjusted on a unit by unit basis.'' 
The retrofit factor is a subjective factor used to account for the 
estimated difficulty of the retrofit that is unique to the facility. 
Because site visits were not possible, the retrofit factor was 
estimated from satellite images that provide some insight to the 
configuration of the units and degree of congestion around the site and 
in the vicinity of where the SCR would be installed. These factors 
impact the ability to locate large cranes on the site--that impact how 
the SCR is assembled (are large sections lifted into place or is the 
SCR ``stick built''), how much ductwork is needed, if the SCR must be 
built onto a large, elevated steel structure or can be built near the 
ground, and if other equipment must be relocated to accommodate the 
space of the SCR. When using the IPM capital cost model, retrofit 
difficulties associated with an SCR may result in capital cost 
increases of 30 to 50% over the base model.\200\
---------------------------------------------------------------------------

    \200\ Sargent & Lundy, ``IPM Model--Revisions to Cost and 
Performance for APC Technologies SCR Cost Development Methodology 
FINAL'', August 2010, Project 12301-007, Perrin Quarles Associates, 
Inc. p 1.
---------------------------------------------------------------------------

    A proper estimation of retrofit factors involves more than an 
inspection of satellite images. For example, EPA Region 8 visited the 
four-unit Colstrip power plant in Montana before concluding that a 
retrofit factor of ``1'' was appropriate. Once such a site visit is 
conducted, retrofit factors should be developed for each element of the 
cost analysis--not the ``blanket'' approach used by EPA here.
    Another example is provided by Sargent & Lundy's ``Constructability 
Review'' for addition of SCR at Navajo Generating Station. Navajo 
Generating Station consists of three EGUs, with the middle unit 
constrained by a coal conveyor passing through. Even so, Sargent & 
Lundy estimated that construction effort would be only 25% greater for 
Unit 2 than for the other two units. EPA needs to clarify why they 
chose to add a retrofit factor greater than 1 (average retrofit factor 
of 1.33 for 13 of 15 units reviewed) to the costs when retrofit costs 
are already contained within data used to generate the IPM and when 
neither Wyoming, Basin Electric, or PacifiCorp included a comparable 
retrofit factor. By adding the retrofit factor, EPA has overestimated 
the costs of SCR: in the case of Dave Johnston Units 1, 2, and 4 and 
Wyodak Unit 1, this has led EPA to propose less-efficient controls than 
SCR.
    Chapter 2, ``Cost Estimation: Concepts and Methodology'' of the CCM 
provides a lengthy discussion of retrofit factors. The CCM addresses 
SCR retrofits specifically ``A correction factor for a new installation 
versus a retrofit installation is included to adjust the capital 
costs'' (Section 4, NOX Controls, Section 4.2, 
NOX Post- Combustion, Chapter 2, SCR). The CCM retrofit 
factor is $728/MMBtu/hr and, for medium-size boilers like Dave Johnston 
Unit 4 or Wyodak, this represents a 23%-24% increase in the direct 
capital cost.
    EPA inconsistently and without explanation applied ``retrofit 
factors'' that improperly increase the reported capital costs of SCR 
installation. Such retrofit factors are intended to account for the 
increased costs of unusually difficult retrofits, i.e., those that 
present more challenges than assumed for the ``typical'' SCR retrofit, 
for which costs are described by the IPM SCR cost module. The EPA 
unjustifiably determined that only two of the fifteen Wyoming EGUs 
(Laramie River Station Units 1 and 3) would be of average difficulty, 
while applying increases of between 20 and 50 percent to the remaining 
units. The EPA applied such retrofit factors even for units for which 
the source owners did not claim above-average installation challenges. 
The

[[Page 5153]]

EPA's exclusive evidence of potential retrofit challenges--satellite 
images--does not support the EPA's application of retrofit factors. 
(The commenter submitted a TSD that elaborated on some of these 
points.)
    The descriptions given of the EPA's view of the retrofit difficulty 
at each plant based on satellite images make it clear that many guesses 
were made and/or that the EPA erred on the side of high retrofit 
difficulty to be conservative. Being conservative in cost estimates may 
be acceptable if such conservatism is applied equally to all units and 
if the EPA provides a reasoned basis for its assumptions, but the EPA 
did not do so. The EPA assumed that the two units in the middle at Jim 
Bridger ``will be somewhat more difficult to achieve access for 
equipment'' and applied the highest retrofit factor of 1.5 to these 
units, while for Laramie River Unit 2, the EPA assumed more retrofit 
difficulty due to its location in the middle but only applied a 
retrofit factor of 1.2. The EPA essentially made guesses that the 
middle units may have more retrofit difficulty, and did not 
consistently apply the same retrofit factors to the middle units of 
these two plants. We found it telling that PacifiCorp's capital cost 
estimates for installation of SCR systems at all four Jim Bridger units 
(which are of equal size to each other) were identical for each unit, 
and the same is true of Basin Electric's capital cost estimates for 
installation of SCR systems at all three Laramie River units (which are 
also equal in size). Given these facts, the EPA has no basis for its 
application of a higher SCR retrofit factor for the units in the middle 
(i.e., Jim Bridger Units 2 and 3 and Laramie River Unit 2).
    To summarize, the EPA has not adequately justified the application 
of any retrofit factor to the costs of SCR at any of the EGUs in 
Wyoming, and the EPA should not apply retrofit factors to increase the 
capital costs of SCR without adequate justification for those retrofit 
factors. Further, the EPA must remember that the IPM cost module for 
SCR is based on actual cost data for SCR retrofits, and that virtually 
all SCR retrofits would have some space constraints due to most power 
plants being built without ever planning for SCR installation. The EPA 
should only apply a retrofit factor if it can justify that the cost of 
SCR installation would noticeably deviate from a typical installation.
    Response: As noted by commenter, the IPM cost model is based upon 
actual retrofits and incorporates all of the costs normally associated 
with retrofit of an SCR. This means that many of the retrofit issues 
commenters have raised are incorporated into the base cost, which can 
then be adjusted with a retrofit difficulty factor based upon the 
perceived difficulty of the retrofit relative to typical retrofits. EPA 
disagrees with the commenter in its assertion that the EPA 
inconsistently and without explanation applied ``retrofit factors'' 
that improperly increase the reported capital costs of SCR 
installation.
    EPA applied retrofit factors while carefully considering site 
conditions. Where there was uncertainty, EPA did lean toward making 
more conservative estimates, which would explain the average retrofit 
factor exceeding 1.0. Section 2.5.3.2 of the CCM discusses retrofit 
cost considerations as ``Probably the most subjective part of a cost 
estimate.'' The CCM states that, ``Since each retrofit installation is 
unique, no general factors can be developed. A general rule of thumb as 
a starting point for developing an appropriate retrofit factor is: The 
larger the system, the more complex (more auxiliary equipment needed), 
and the lower the cost level (e.g. study level, rather than detailed), 
the greater the magnitude of the retrofit factor.'' Thus, retrofit 
difficulty factor may factor in some uncertainty and be higher to 
account for that.
    In the cost estimates EPA developed, retrofit factors were 
determined from satellite images using the following considerations: 
(1) Available access to and from the site for transportation of 
equipment and available space for laying down construction materials; 
(2) Location of equipment relative to each other and whether there is a 
substantial amount of demolition necessary in order to make room for 
SCR equipment; and (3) Access for a crane. At a highly congested site, 
crane access can be difficult and may entail a more costly approach. 
Access for a crane is a particular concern for internal units when 
units are located side-by-side.
    There are no strict guidelines used for determining the actual 
value of retrofit factors. They are a matter of judgment. Per the CCM 
at 2.5.4.2 (page 2-28, Chapter 2 Cost Estimation: Concepts and 
Methodology), ``[t]he proper application of a retrofit factor is as 
much an art as it is a science, in that it requires a good deal of 
insight, experience, and intuition on the part of the analyst.'' What 
follows is the explanation for each of the retrofit factors used at 
each of the sites. As will be shown, the retrofit factors were the 
result of a thoughtful process, and were not arbitrary.
    With regard to the Dave Johnson site, this is one of the more 
congested sites in Wyoming. Per the Andover report on estimated costs 
of NOX controls: ``Based upon the close proximity of the 
boilers and associated equipment to one another, decommissioned 
chimneys that will limit access and ability to move a crane, the coal 
pile and coal conveyors that also limits access to the area of the 
units where construction would occur, the office building that is 
adjacent to Unit 1 and limits access, and the Unit 4 scrubber, retrofit 
of an SCR on Units 1-4 would likely entail a significantly higher than 
average retrofit cost. Unit 4 probably has the best access of all of 
the units because there may be some space between the boiler and the 
scrubber, but it is difficult to say for sure from the image and 
therefore a conservatively high retrofit difficulty was used for unit 4 
that is consistent with the other units.'' \201\ As noted, the site is 
fairly congested for all units, justifying a high retrofit factor of 
1.5 for all units, not just the middle units.
---------------------------------------------------------------------------

    \201\ Andover Technology Partners, ``Cost of NOX 
Controls on Wyoming EGUs'', October 28, 2013, p. 30; Wyoming EGU 
BART and Reasonable Progress Costs--10/28/2013.
---------------------------------------------------------------------------

    Jim Bridger is also limited on space, but not so much so as Dave 
Johnston: ``Based upon the satellite photo, the SCR reactors would 
likely be installed above the ESPs and ductwork routed to the boiler. 
The boilers do not appear to be unusually constrained from the 
perspective of installing SCR ductwork; however, access for 
construction equipment will be much more difficult to achieve for the 
two middle units. The scrubbers and associated piping will limit access 
somewhat. Unit 4 access will be limited somewhat by the coal conveyor 
and because it's scrubber takes up more room than the other scrubbers. 
For this reason a retrofit difficulty factor of 1.5 is assumed for the 
middle units and 1.25 for units 1 and 4.'' \202\
---------------------------------------------------------------------------

    \202\ Andover Technology Partners, ``Cost of NOX 
Controls on Wyoming EGUs'', October 28, 2013, p. 35; Wyoming EGU 
BART and Reasonable Progress Costs--10/28/2013.
---------------------------------------------------------------------------

    On the other hand, as is apparent from the satellite image, Laramie 
River is a more open site than Dave Johnson. Satellite images ``show a 
less constrained site than Dave Johnston, with good access to both 
units 1 and 3. The coal conveyor is clearly visible and will be an 
obstruction for the unit 2 SCR. As a result, retrofit difficulty of 
installing SCR is expected to be average, except possibly for unit 2 
which is located between units 1 and 3. Access of a crane will be 
somewhat more challenging for Unit 2 and an SCR retrofit difficulty of 
1.2 is assumed for estimating SCR capital cost. In all cases

[[Page 5154]]

the SCR reactor is likely to be installed above the ESP ductwork.'' 
\203\
---------------------------------------------------------------------------

    \203\ Andover Technology Partners, ``Cost of NOX 
Controls on Wyoming EGUs'', October 28, 2013, p. 39; Wyoming EGU 
BART and Reasonable Progress Costs--10/28/2013.
---------------------------------------------------------------------------

    Naughton is much more congested than Laramie River, although access 
appears slightly better than for Dave Johnson. Per the Andover report:

    Babcock & Wilcox provided cost estimates for the Naughton unit 1 
& 2 SCRs. Babcock & Wilcox's estimate assumed that it would be 
necessary to demolish stacks that will be abandoned after a planned 
scrubber installation and they also determined that additional fan 
capacity was not necessary. Babcock & Wilcox also assumed a complex 
support structure would be needed, which adds cost. Babcock and 
Wilcox also stated that units 1 and 2 are slightly offset which 
makes it impractical to build a common support structure for the SCR 
reactors; however, . . . Babcock & Wilcox has used a longer 
horizontal duct run on the unit 1 SCR which places the unit 1 and 
unit 2 SCR reactors side-by-side so that a common support structure 
is likely to be possible, offering some potential savings from what 
they have estimated. Alternatively, a shorter horizontal duct run on 
unit 1 may make it possible to avoid demolition of the unit 1 
chimney that will be abandoned. When using the IPM algorithm a 
retrofit difficulty factor of 1.3 is assumed. This was based upon 
the fact that it appeared to be a less congested site than Dave 
Johnston, but there were potential challenges, such as the chimneys, 
that could result in longer duct runs or additional demolition.
    For unit 3 it is also unclear if there is enough space to 
install the SCR reactor on the same side of the chimney as the 
boiler, which, means that demolition of that chimney may be needed. 
There appears to be access for construction equipment, such as a 
crane, in the area east of the plant (the upper part of the photo) 
and to the north of unit 3. In estimating the cost of the SCR for 
unit 3, retrofit difficulty is above average because more lengthy 
duct runs or demolition of the chimney are likely needed, and an 
assumed retrofit difficulty factor of 1.3 is assumed for unit 3.

    Hence, Naughton is assumed to be an above average retrofit 
difficulty because of the potential for some significant interference 
from some equipment and the possible need for longer than average duct 
runs.
    For Wyodak, access to the site appears good, but there are some 
possible issues that might come up if the existing (but decommissioned) 
ESP needs to be demolished, which was the reason for the above average 
retrofit factor of 1.3. Per the Andover report:

    In the event SCR were installed at the site, the SCR reactor 
would likely be located above the existing (but decommissioned) ESP 
shown between the boiler building and the chimney . . . There is 
ample room on the site for lay down of material. Location of a crane 
near the construction site appears to be possible; however, in every 
direction from the boiler there is a potential interference that 
might complicate crane location relative to the lay-down area. 
Therefore, the difficulty of this retrofit is probably average to 
perhaps above average. A conservative estimate of retrofit 
difficulty of 1.3 is assumed, although a closer examination of the 
site may show that a lower retrofit difficulty may be possible.\204\

    \204\ Andover Technology Partners, ``Cost of NOX 
Controls on Wyoming EGUs'', October 28, 2013, p. 52; Wyoming EGU 
BART and Reasonable Progress Costs--10/28/2013.
---------------------------------------------------------------------------

    As a result, the retrofit factors used were not arbitrary but the 
result of a thoughtful process of examining the site for issues that 
would affect the difficulty of the retrofit.
    Comment: EPA's application of the maximum retrofit factor (1.5) to 
Dave Johnston Units 1 and 2 is unsupported and leads to a significant 
$1.5 million/year and $800/ton overestimation of average costs. Neither 
PacifiCorp nor Wyoming proposed a retrofit factor for these units. It 
is especially surprising that EPA has applied the maximum retrofit 
factor to all four units at Dave Johnston, and that even an ``end'' 
unit like Unit 1 is considered to have the highest degree of retrofit 
difficulty. It has been our experience that end units are typically the 
easiest to retrofit, while the more difficult retrofits are associated 
with ``middle'' units. Once the SCR costs are corrected to address the 
issue discussed above, the incremental costs become $5,700-$5,800/ton 
(versus $7,050/ton at Bridger Unit 2).
    The EPA applied a retrofit factor of 1.5 to Dave Johnston Units 1-
3, citing close proximity of boilers, decommissioned chimneys and the 
coal pile. It is not clear that these issues warrant a 50 percent 
increase in SCR costs due to retrofit difficulty at Units 1-3.
    Response: EPA's estimate of retrofit factor is based upon a 
thoughtful consideration of the various factors described in the 
previous response. With regard to the Dave Johnston site, this appears 
to be one of the more difficult sites in Wyoming from the perspective 
of retrofit.
    Comment: EPA's application of the maximum retrofit factor (1.5) to 
SCR on Dave Johnston Unit 4 is unsupported and leads to a significant 
$3.8 million/year and $900/ton overestimation of average costs. Neither 
PacifiCorp nor Wyoming proposed a retrofit factor for this unit. We 
disagree with EPA's decision to apply the maximum retrofit factor to 
all four units at Dave Johnston, and that even an ``end'' unit like 
Unit 4 is considered to have the highest degree of retrofit difficulty. 
It has been our experience that end units are typically the easiest to 
retrofit, while the more difficult retrofits are associated with 
``middle'' units.
    The CCM retrofit factor is $728/MMBtu/hr and, for medium-size 
boilers like Dave Johnston Unit 4 or Wyodak. This represents a 23%-24% 
increase in the direct capital cost. For Dave Johnston Unit 4, the EPA 
applied a retrofit factor of 1.5 to the costs of SCR because ``[t]here 
may be more space available near unit 4 for a retrofit of SCR, but this 
is unclear from the photograph.'' This is a very questionable basis to 
justify increasing the costs of SCR by 50 percent.
    Response: Because of the congestion of the overall site at Dave 
Johnson plant, a large reduction of retrofit factor to well below 1.5 
was not justified even though Unit 4 is an end unit. There appears to 
be other equipment in the vicinity of Unit 4 that would obstruct access 
and maintaining a retrofit factor of 1.5 seems reasonable.
    Comment: The EPA assigned a 1.3 retrofit factor to the SCR cost 
estimate for the single unit Wyodak plant. It appears the main reason 
for applying this factor is because the SCR would likely have to be 
placed on top of the decommissioned ESP, and that space constraints 
were not an issue. Although the EPA summarized that ``. . . the 
difficulty of this retrofit is probably average to above average since 
it is common to have some relocation of equipment,'' the EPA applied a 
30 percent increase to the SCR costs for Wyodak.
    Response: As noted by the commenter, EPA determined that a retrofit 
factor of 1.3 is conservative at Wyodak. But, because of the possible 
items that are apparent (as described above) from the satellite 
photograph and that could get in the way, it was not regarded as below 
average difficulty, but perhaps something slightly above average. EPA 
therefore does not believe that the retrofit factor should be lower 
than 1.0 and it may be as high as 1.3. EPA is basing its cost analysis 
on the more conservative estimate.
    Comment: In Montana, EPA used the IPM algorithms for some sources 
and not for others, asserting only that use of IPM ``was intended to 
ensure that the direct capital costs reflect the most recent cost 
levels seen in the marketplace'' and thus did not over-estimate costs. 
77 FR 57888. EPA also used IPM for the Arizona FIP but failed to 
address how its use was consistent with either the BART Guidelines or 
the CCM. 77 FR 72512 (Dec. 5, 2012). In Colorado, EPA said the State's 
cost

[[Page 5155]]

estimates for Craig Unit 1 deviated from the CCM but accepted them 
anyway because EPA was pleased with a State law that required emission 
reductions from certain other EGUs. 77 FR 76875. EPA sometimes 
supplements the CCM with a rule that installed SCR retrofit costs must 
fall between $79/kW and $316/kW, unless the state justifies a deviation 
from this range. North Dakota SIP, 77 FR 20929; Montana FIP, 77 FR 
57889; New Mexico FIP, 76 FR 52388, 52392. EPA has proposed to apply 
this rule to the Wyoming SIP and FIP, 78 FR 34738. This cost range is 
derived from ``industry studies'' and does not appear anywhere in the 
CCM. Supplementing the CCM with this new requirement is inconsistent 
from the terms of the CCM and BART Guidelines.
    Response: We agree that we have used the IPM control cost 
algorithms in various regional haze rulemakings as noted by the 
commenter. And as noted by the commenter, our intent in using the IPM 
cost algorithms was to ensure that our capital cost estimates for SCR 
reflect those currently found in the marketplace. Elsewhere in these 
responses to comments, we have documented in some detail how our use of 
the IPM algorithms is consistent with the BART Guidelines and CCM. We 
disagree that we have supplemented the CCM with a rule that retrofit 
costs must fall between $79/kW and $316/kW--2010 dollars ($81/kW to 
$324/kW when escalated to 2013 dollars), which was the range of actual 
installed capital costs found in recent industry studies as cited in 
our final rule for North Dakota. In the North Dakota rulemaking, we 
used this information to assess whether costs supplied to EPA by states 
or sources were consistent with those observed in the industry for 
numerous other retrofits spanning a wide range of retrofit 
difficulties. As such, this in no way represents a requirement imposed 
by EPA; rather, it represents a very practical means by which EPA has 
gauged the validity of costs. We acknowledge that, given exceptionally 
difficult retrofit circumstances or other factors, it is possible for a 
particular retrofit to fall outside of this range. In fact, we note 
that our revised costs supporting this final rule are in some cases in 
excess of the cited range. Our revised SCR costs for BART EGUs, when 
represented on a dollar per kilowatt basis, range from $222/kW to $467/
kW, with a median cost of $322/kW (2013 dollars).\205\ From this, it is 
clear that we have not established a requirement that SCR capital costs 
fall within the cited range as suggested by the commenter. For our 
Colorado final action, we are currently in litigation over our approval 
of the State's BART determination for Craig.
---------------------------------------------------------------------------

    \205\ Andover Technology Partners, ``Cost of NOX 
Controls on Wyoming EGUs,'' October 28, 2013; Wyoming EGU BART and 
Reasonable Progress Costs--10/28/2013; Wyoming EGU BART and 
Reasonable Progress Costs for Jim Bridger--10/28/2013.
---------------------------------------------------------------------------

    Comment: EPA inappropriately claimed that ``Wyoming's SCR capital 
costs on a $/kW basis often exceeded real-world industry costs'' (78 FR 
34748) and then refers to industry studies conducted between 2002 and 
2007 that report installed unit capital costs actually incurred by 
owners broadly ranging ``from $79/kW to $316/kW (2010 dollars).'' Id. 
EPA also noted ``instances'' in its proposed FIP ``in which Wyoming's 
source-based cost analyses did not follow the methods set forth in the 
EPA CCM.'' EPA is simply incorrect in stating that Wyoming's analyses 
were flawed and did not reflect real-world industry costs for the units 
being analyzed. The commenter states that they are presenting 
information on the ``real-world'' costs for the upcoming Jim Bridger 
Units 3 and 4 SCR projects, which recently were competitively bid for 
engineering, procurement, and construction contracts to be installed in 
accordance with the requirements in the Wyoming SIP. These real-world 
costs, in turn, can easily be compared to the costs assessed by Wyoming 
and by EPA in their BART determinations.
    Even when including AFUDC, the Wyoming SIP cost basis aligns 
closely with the EPA's cost basis, with each agency again understating 
real world costs for these projects. By extension, this real-world cost 
information for Jim Bridger Units 3 and 4 validates the methodology 
used by Wyoming to determine cost information for each of PacifiCorp's 
BART Units. This information clearly disputes EPA's claims in its FIP 
that Wyoming ``did not properly or reasonably take into consideration 
the costs of compliance'' and that its SCR cost analyses exceeded real 
world industry costs and were flawed. Id.
    Response: We disagree that it was incorrect for EPA to state 
Wyoming's cost analyses for SCR were flawed. As discussed in our 
proposed rule, EPA found several deficiencies with Wyoming's cost 
analyses, including: Inclusion of AFUDC, inclusion of some 
inappropriate owner's costs, insufficient documentation to support 
vendor estimates or bids, and use of incorrect baseline emission rates. 
78 FR 34749. These deficiencies represented a departure from the 
procedures outlined in the CCM and BART Guidelines and, particularly 
when taken collectively, had a material impact on the cost estimates. 
We have addressed why each of these items are inconsistent with the CCM 
and BART Guidelines in other response to comments here.
    Moreover, since the time of the State's analyses, EPA has been made 
aware of additional pertinent information by commenters, much of which 
has been incorporated into the revised costs presented in this final 
rulemaking. Examples include: Inclusion of certain costs submitted by 
the facility owner's where appropriately documented, shorter useful 
life for one facility, correction for elevation, use of ammonia instead 
of urea as SCR reagent, revised SNCR reagent consumption for some 
facilities, and use of busbar costs for auxiliary power in place of 
market prices.
    Regarding whether Wyoming's costs exceeded real world industry 
costs, see EPA's response to comment immediately above.
    Comment: Wyoming did not overestimate the costs of SCR. The EPA 
claimed to have identified a number of flaws in Wyoming's cost analyses 
for SCR (78 FR 34748), but only identified one flaw--that ``Wyoming's 
SCR capital costs on a $/kW basis often exceeded real-world industry 
costs.'' The EPA's use of the word ``often'' indicates that Wyoming's 
costs did not always exceed real-world costs, but the EPA did not 
explain which costs exceeded real-world costs and which did not.
    The EPA specifically alleged only that the cost estimates for Dave 
Johnston Units 3 and 4, Naughton Units 1, 2, and 3, and Wyodak ``are in 
excess of the range of capital costs documented by various studies for 
actual installations,'' and that the EPA based this conclusion of five 
industry studies conducted between 2002 and 2007. The EPA did not 
explain why the State was wrong to rely on vendor submitted, 
engineered, site-specific cost estimates instead of reports of 
installations at other facilities as long as a decade ago. The State's 
costs of compliance are based on site-specific capital costs, operating 
costs, and maintenance costs provided by the companies in their 
applications for a state BART permit, and over 50 percent of the costs 
of compliance is driven by the capital cost to engineer and physically 
install a SCR system. Such costs must be evaluated on a case-by-case 
basis in accordance with Appendix Y. Variable costs, including reagent 
usage (ammonia), account only for 2 to 7 percent of SCR costs.
    The BART Guidelines not only allow, but encourage states to take 
into account

[[Page 5156]]

site-specific conditions that impact the cost of installing emission 
controls. Until the EPA explains why it was unreasonable for Wyoming to 
prefer site-specific, real-world costs over speculative extrapolation 
of costs incurred at other facilities many years past, the EPA cannot 
lawfully displace the State's judgment simply because EPA prefers one 
approach over the other.
    Response: We disagree that EPA only identified one material flaw in 
Wyoming's costs estimates for SCR. See EPA's response to comment 
immediately above where we identify several flaws. Because Wyoming's 
approach to estimating SCR costs was not consistent with the BART 
Guidelines and CCM, it was appropriate for EPA to revise these costs in 
our proposed rule.
    We agree that the BART Guidelines encourage states to take into 
account site-specific conditions that impact the cost of installing 
emission controls. However, we disagree with the commenter's 
characterization of the State's costs of compliance for SCR as site-
specific in nature on the mere basis that they were submitted by the 
sources. There is nothing in the record to support claims that these 
costs were in fact based on detailed site-specific vendor bids, or are 
in any manner more site-specific than those costs relied upon by EPA in 
our proposed rule. As an example, the BART application submitted by 
PacifiCorp for Dave Johnston Unit 3, and relied upon by Wyoming, states 
that: ``Costs and schedules for the LNBs and OFA, SNCR, and SCR were 
furnished to CH2M HILL by PacifiCorp, developed using Sargent and 
Lundy's internal proprietary database, and supplemented (as needed) by 
vendor-obtained price quotes. The relative accuracy of these cost 
estimates is stated by S&L to be in the range of plus or minus 20 
percent.'' \206\
---------------------------------------------------------------------------

    \206\ Wyoming Regional Haze SIP, Attachment A, ``BART Analysis 
for Dave Johnston Unit 3,'' prepared for PacifiCorp by CH2MHILL, 
December 2007, page 3-7.
---------------------------------------------------------------------------

    From this, it is clear that PacifiCorp, and thereby also the State, 
based SCR costs on ``S&Ls internal database'' and not a unique 
quotation specific to Dave Johnston Unit 3 supplied by an SCR vendor. 
Moreover, while the BART application refers to ``vendor-obtained 
quotes,'' it does not make clear for which items these quotes were 
obtained, if any, nor are any quotes for SCR included in the BART 
application. Instead, the total installed capital cost of SCR (with 
combustion controls) is shown in PacifiCorp's economic analysis as a 
single line item with a value of $83,301,164, but without any detail or 
supporting documentation.\207\ In an update to its initial BART 
application, PacifiCorp subsequently increased the capital cost to 
$129,700,000, but again without any detail or supporting 
documentation.\208\
---------------------------------------------------------------------------

    \207\ Wyoming Regional Haze SIP, Attachment A, ``BART Analysis 
for Dave Johnston Unit 3,'' prepared for PacifiCorp by CH2MHILL, 
December 2008, Attachment 1.
    \208\ Wyoming Regional Haze SIP, Attachment A, ``Addendum to 
Dave Johnston Unit 3 BART Report,'' prepared for PacifiCorp by 
CH2MHILL, March 26, 2008, Attachment 1.
---------------------------------------------------------------------------

    We note that the capital cost estimates for SCR presented by EPA in 
our proposed rule were also based on the Sargent & Lundy databases as 
these in turn underlie the IPM cost algorithms. As such, the commenter 
is mistaken when characterizing Wyoming's capital costs as superior to 
those from EPA. However, the costs presented by EPA went on to correct 
the deficiencies that we have identified elsewhere (e.g., improper 
calculation of baseline emissions). Therefore, the capital costs 
provided from each agency were ultimately generated in a similar 
manner, but only the overall costs generated by EPA were in keeping 
with the BART Guidelines and CCM.
    It is notable that, in order to address our concerns regarding lack 
of site-specific costs and associated documentation, the Wyoming 
sources have submitted additional cost information during the comment 
period for the proposed rule. The PacifiCorp comments include capital 
costs based on a vendor budgetary quote from Babcock and Wilcox, as 
opposed to capital costs based on the Sargent & Lundy databases. 
Similarly, Basin Electric has submitted a consultant's report that, 
while conceptual in nature and without vendor-based equipment costs, 
provides a more detailed analysis of SCR costs for Laramie River than 
before. These submittals contain more recent and more detailed cost 
information than relied upon by Wyoming in their cost analyses. The 
submission of these updated costs from the sources, intended to supply 
more site-specific costs to EPA, belies claims by the commenter that 
the costs originally used by the State were ``vendor submitted, 
engineered, site-specific cost estimates.''
    Comment: We would like to point out that while the EPA makes a 
point of saying in their proposed rule that they have followed their 
own guidelines in the CCM, that manual has not been updated since 2002. 
Representative Lummis of Wyoming has authored language to require the 
EPA to update its cost manual for the first time in over a decade. The 
old data in the old handbook no longer reflects the true costs of 
designing, engineering and installing controls. Before rejecting state 
data on the cost of compliance, the EPA must engage states and 
regulating entities to acquire real-world cost data and use that data 
to update its manual.
    Response: We consider the use of the broader costing methodology 
used by the CCM, the overnight method, as crucial to our ability to 
assess the reasonableness of the costs of compliance. Evaluation of the 
cost of compliance factor requires an evaluation of the cost-
effectiveness associated with the various control options considered 
for the facility. A proper evaluation of cost-effectiveness allows for 
a reasoned comparison not only of different control options for a given 
facility, but also of the relative costs of controls for similar 
facilities. If the cost-effectiveness of a control technology for a 
particular facility is outside the range for other similar facilities, 
the control technology may be rejected as not cost-effective. In order 
for this type of comparison to be meaningful, the cost estimates for 
these facilities must be performed in a consistent manner. Without an 
``apples-to-apples'' comparison of costs, it is impossible to draw 
rational conclusions about the reasonableness of the costs of 
compliance for particular control options. Use of the CCM methodology 
is intended to allow a fair comparison of pollution control costs 
between similar applications for regulatory purposes.
    Just as importantly, while we have followed the broad methodology 
of the CCM as required by the BART Guidelines, we have also accounted 
for the cost of controls currently observed in the marketplace. In 
particular, our use of the cost calculations taken from the IPM, 
released in 2010, is designed to reflect modern day costs. Moreover, 
operation and maintenance costs for items such as labor, reagent, and 
catalyst, reflect current market values. In short, we have adhered to 
the broad overnight cost methodology specified in the CCM, while 
updating both capital and operation and maintenance costs to reflect 
current market conditions. Therefore, the commenter is mistaken in 
asserting that our costs are based on outdated information.
    Comment: EPA's regional haze FIP also is improper because it 
assumes BART NOX controls over $5,000 per ton are ``cost 
effective.'' (See e.g., 77 FR 33053.) Appendix Y, on the other hand, 
states that BART NOX control costs per ton above $1,500 are 
not ``cost

[[Page 5157]]

effective.'' In the preamble to the BART Guidelines, EPA suggests that 
75% of the EGUs would have BART NOX removal costs between 
$100 and $1,000 per ton, and almost all of the remaining EGUs could 
install sufficient BART NOX control technology for less than 
$1,500 per ton. EPA also recognized in the preamble that SCR was 
generally not cost effective for EGUs, except for EGUs with cyclone 
boilers (where the cost per ton was less than $1,500 per ton, with an 
average of $900 per ton). Based upon EPA's Preamble, BART 
NOX control technology that costs more than $1,500 per ton 
should not be considered ``cost effective.'' Here, EPA found BART 
NOX controls with a ``cost effectiveness'' number much more 
than $1,500 per ton to be ``cost effective.'' Therefore, EPA should 
withdraw its regional haze FIP.
    Response: We disagree with this comment. 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: ``States 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. Because the preamble generally 
discusses costs, this does not obviate the need for states (or EPA in 
the case of a FIP) to identify the best system of continuous emission 
control technology on a case-by-case basis considering the five 
factors. While EPA described various dollar-per-ton costs as ``cost-
effective'' in various preambles (e.g., 70 FR 39135-39136), EPA did not 
establish an upper cost effectiveness threshold for BART 
determinations.
    Comment: Far from stating that the CCM must be the exclusive source 
of cost information, the BART Guidelines state that ``[t]he basis for 
equipment cost estimates also should be documented, either with data 
supplied by all equipment vendor (i.e., budget estimates or bids) or by 
a referenced source (such as the EPA CCM]).'' Although the BART 
Guidelines then say that cost estimates should he based on the CCM, it 
also says that the CCM should only he used ``where possible.'' The 
Guidelines go on to say that the CCM ``addresses most control 
technologies in sufficient detail for a BART analysis.''
    The CCM does not say that it addresses `all' control technologies, 
just ``most'', implying that the CCM does not supply all of the 
necessary information. Further, the Guidelines state that the cost 
analysis should ``take into account any site-specific design or other 
conditions identified above that affect the cost of a particular BART 
technology option.'' Again, the CCM acknowledges that there are 
conditions, design scenarios, etc. that are not addressed in the CCM 
but that exist in the real world that must be addressed.
    Response: We acknowledge that our BART guidelines state, ``In order 
to maintain and improve consistency, cost estimates should be based on 
the [CCM], where possible'' and that ``[w]e believe that the [CCM] 
provides a good-reference tool for cost calculations, but if there are 
elements or sources that are not addressed by the Control CCM or there 
are additional cost methods that could be used, we believe that these 
could serve as useful supplemental information.'' The CCM contains two 
types of information: (1) Study level cost estimates of capital and 
operation and maintenance costs for certain specific types of pollution 
control equipment, such as SCR, and (2) a broader costing methodology, 
known as the overnight method. We agree that the language of the BART 
Guidelines does not require strict adherence to the study level 
equations and cost methods used to estimate capital and operating and 
maintenance costs.
    We consider the use of the broader costing methodology used by the 
CCM, the overnight method, as crucial to our ability to assess the 
reasonableness of the costs of compliance. Evaluation of the cost of 
compliance factor requires an evaluation of the cost-effectiveness 
associated with the various control options considered for the 
facility. A proper evaluation of cost-effectiveness allows for a 
reasoned comparison not only of different control options for a given 
facility, but also of the relative costs of controls for similar 
facilities. If the cost-effectiveness of a control technology for a 
particular facility is outside the range for other similar facilities, 
the control technology may be rejected as not cost-effective. In order 
for this type of comparison to be meaningful, the cost estimates for 
these facilities must be performed in a consistent manner. Without an 
``apples-to-apples'' comparison of costs, it is impossible to draw 
rational conclusions about the reasonableness of the costs of 
compliance for particular control options. Use of the CCM methodology 
is intended to allow a fair comparison of pollution control costs 
between similar applications for regulatory purposes. This is why the 
BART guidelines specify the use of the CCM where possible and why it is 
reasonable for us to insist that the CCM methodology be observed in the 
cost estimate process. The overnight method has been used for decades 
for regulatory control technology cost analyses, and its use ensures 
equitable BART determinations across states and across sources.
    Comment: Although EPA contends that States must conform in all 
respects to the Agency's CCM, its own consultant ignores the Manual 
when calculating capital costs and operating and maintenance costs, and 
instead uses an entirely different methodology called the IPM. 78 FR 
34749. EPA tries to finesse this problem by asserting that the 
consultant followed ``the structure of'' the CCM and BART Guidelines, 
id., but that simply is not true. The IPM is a fundamentally different 
tool and uses a fundamentally different methodology than the CCM--it 
does not follow the CCM. Therefore, to rely on the consultant's cost 
report to disapprove Wyoming's cost analysis and BART analysis would be 
arbitrary and capricious, and not in accordance with law.
    Response: We disagree. As noted elsewhere in these responses to 
comments, in our revised cost estimates, we have followed the broad 
methodology of the CCM, referred to as the overnight method, while 
updating capital and operating and maintenance costs to reflect current 
real-world costs. In doing so, we directed our consultant (Andover) to 
reconcile anything in the IPM cost algorithms that would be 
inconsistent with the CCM's overnight method. For example, the IPM cost 
algorithms include AFUDC, which as we have established elsewhere in 
these response to comments, is not part of the overnight costs. 
Accordingly, our consultant eliminated this cost when utilizing costs 
derived IPM cost algorithms. In effect, we have ``squared'' the IPM-
based costs with the methodology required by the CCM.
    Comment: EPA's average cost effectiveness for combustion controls 
and SCR for the Laramie River units is higher than Wyoming's average 
cost effectiveness. Compare, e.g., 78 FR 34773, Table 36 (Wyoming's SCR 
average cost effectiveness of $3,372/ton for Unit 1) with 78 FR 34775, 
Table 39 (EPA's SCR average cost effectiveness of $3.718/ton for Unit 
1). The higher the cost effectiveness of a given technology, the 
stronger the case for rejecting it. If the State was justified in 
rejecting SCR based on its lower predicted cost of SCR, it would be 
even more justified in

[[Page 5158]]

rejecting SCR if it had used EPA's higher cost. The outcome would not 
have changed, and so any error alleged by EPA is not material.
    EPA's incremental cost effectiveness for combustion controls plus 
SCR, compared with the cost effectiveness of combustion controls plus 
SNCR, is lower than Wyoming's incremental cost effectiveness. However, 
in considering cost effectiveness for purposes of both its SIP 
disapproval and its FIP proposal, EPA cites and relies primarily on the 
average cost effectiveness for SCR, not the incremental cost. 78 FR 
34776 (``[T]he cost-effectiveness for new LNBs with OFA and SCR ranges 
from approximately $3600/ton to $3900/ton with significant visibility 
improvement at the most impacted Class I area. . . . When considering 
the cost effectiveness and visibility improvement of new LNBs plus OFA 
and SCR, it is within the range of what EPA has found reasonable for 
BART in other SIP and FIP actions.'') EPA refers to incremental cost 
only incidentally--not as an affirmative reason for disapproving 
Wyoming's BART. Id. (``We also propose to find that the incremental 
cost-effectiveness does not preclude the selection of new LNBs with OFA 
and SCR.'').
    Response: We disagree. The commenter fails to note that the 
visibility improvement presented by EPA in our proposed rule is higher 
than that found by the State. The tables cited by the commenter show a 
visibility improvement from SCR of 0.44 deciviews resulting from the 
State's analysis, while EPA's analysis showed a visibility improvement 
of 0.79 deciviews. We found that, when balancing all of the BART 
factors, this level of visibility improvement was significant enough to 
justify the costs associated with SCR. In our revised visibility 
modeling analysis for this final rule, we have presented a lower 
visibility improvement for Laramie River Unit 1 of 0.57 deciviews. We 
continue to find that this level of visibility improvement, and 
consideration of the other BART factors, warrants installation of SCR. 
The same can be said for the other two Laramie River units.
    Additionally, the SCR costs and visibility improvement for the 
Laramie River units provided by the State and EPA are not directly 
comparable. In the BART application submitted by Basin Electric, and 
relied upon by the State, and unlike in the case of SNCR, no additional 
combustion controls are assumed in the SCR control scenario. Since the 
time that Basin Electric submitted the BART application to the State, 
additional combustion controls have been installed on the Laramie River 
units. We have taken account of these additional controls in our 
analyses. When assessing the emission reductions from SCR (or SNCR), 
and the associated costs and visibility improvement, we incorporated 
the actual emission rates currently being achieved with the additional 
combustion controls. We have presented the costs of compliance and 
visibility for the additional combustion controls plus SCR, much in the 
same way that the State presented the same factors for the PacifiCorp 
units.
    Comment: The costs of SCR plus combustion controls are cost 
effective at all of the Wyoming EGUs regardless of whether the costs 
are based on EPA's cost analyses or the commenter's 2012 cost analyses 
conducted for the original Wyoming SIP. SCR costs for each EGU in 
Wyoming, show that SCR plus combustion controls is very cost effective 
for all BART-subject EGUs and also Dave Johnston Units 1 and 2. 
Further, even EPA's June 2013 cost estimates for SCR plus combustion 
controls show that these controls are cost effective at all Wyoming 
EGUs, despite what the commenter believes are deficiencies in EPA's 
cost effectiveness analyses that overestimate the costs of SCR plus 
combustion controls. These costs are within the range that has been 
required or proposed of other similar sources to meet BART as follows:
     Final NOX BART determination for San Juan Units 
1-4 requires installation of SCR at all four units to meet a 
NOX emission limit of 0.05 lb/MMBtu, found that the costs 
ranged from $1,987/ton to $2,651/ton of NOX removed, in 2010 
dollars.
     EPA Region 9 has proposed SCR as BART for Four Corners 
Units 1-5 to meet a NOX limit of 0.11 lb/MMBtu at a cost 
effectiveness of $2,515/ton to $3,163/ton in 2008 dollars. That 
converts to $2,407/ton to $3,028/ton in 2010 dollars.
     In its FIP for Montana, EPA found that the cost 
effectiveness of SCR controls for Colstrip Units 1 and 2 of 
approximately $3,200/ton per unit (in 2010 dollars) was reasonable.
     In its FIP for Arizona regional haze, EPA is requiring SCR 
along with combustion controls to meet BART at the BART-subject coal-
fired units at Apache, Cholla, and Coronado power plants at cost 
effectiveness values ranging from $2,275/ton to $3,472/ton.
    Response: We agree that the costs for SCR plus combustion controls 
presented in our proposed rule, taken without consideration of the 
remaining BART factors, may generally be considered cost effective. 
However, the CAA and RHR require a consideration of all five BART 
factors. For example, a control that is considered cost effective may 
not be warranted if the visibility improvement is minor. Also, there 
may be occasions that, while the average cost effectiveness of a 
control is reasonable, the incremental cost effectiveness may not be. 
In short, EPA must weigh more than just the cost effectiveness when 
considering BART.
    Also note that, as described elsewhere in these responses to 
comments, we have revised the SCR cost estimates that we presented in 
our proposed rule. In today's final rule, we have again balanced the 
costs along with the remaining BART factors when considering the 
selection of BART controls.
    Comment: EPA's proposed FIP is arbitrary, capricious, and contrary 
to law for a number of reasons, including that the EPA's BART analyses 
ignored relevant data. Wyoming based its BART analyses on site-
specific, engineered, vendor submitted bids for installing emission 
controls (citing the SIP Attachment A materials related to Laramie 
River Station), and Basin Electric has submitted to EPA comments 
extensively explaining the bases for these cost estimates, including 
the substantial technical difficulty of installing SNCR and SCR at 
Laramie River Station due to the design of the three units. The EPA has 
disregarded the site-specific cost estimates submitted for Laramie 
River Station and the other BART sources in Wyoming, and the EPA has 
alleged without any specificity that Wyoming did not properly or 
reasonably take into consideration the costs of compliance. The EPA 
relied on the IPM Model with retrofit factors adjusted on a source-by-
source basis, instead of relying on the site-specific costs.
    EPA's October 23, 2012 revised cost memo states that ``[t]he 
retrofit factor is a subjective factor used to account for the 
estimated difficulty of the retrofit that is unique to the facility'' 
and noted that these retrofit factors were determined without site 
visits, but based on satellite images of the facilities. At EPA's 
public hearing in Casper, Wyoming, on July 26, 2013, Basin Electric's 
consultant, Kenneth Snell, explained to EPA in detail how the satellite 
images fail to reveal multiple conditions specific to Laramie River 
Station that make installing SCR far more expensive than EPA's 
consultant assumed. EPA's failure to rebut those positions is arbitrary 
and, moreover, that the EPA's methodology--

[[Page 5159]]

relying on a subjective interpretation of satellite images--is itself 
arbitrary and capricious because it strains credulity to claim that one 
can assess retrofit costs by simply looking at hazy satellite pictures 
of a power plant.
    Response: EPA disagrees. First, Wyoming's BART analysis \209\ was 
based upon a 2008 cost estimate by Basin Electric, which, when adjusted 
for the fact that they were performed prior to addition of combustion 
controls, resulted in similar capital costs for SCR and similar cost 
effectiveness. Moreover, and as noted in previous responses, the costs 
submitted by Wyoming should not be considered site-specific estimates, 
and therefore superior to EPA's costs, on the mere basis that they were 
submitted by a source. In any case, with their comments on EPA's 
reproposal, Basin Electric has roughly doubled their claimed cost of 
SCR, but these were not part of the Wyoming BART analysis. These new 
costs submitted by Basin Electric are presumably intended to be more 
site-specific in nature than those originally submitted to the State.
---------------------------------------------------------------------------

    \209\ Wyoming Department of Environmental Quality Air Quality 
Division BART Application Analysis AP-6047 May 28, 2009.
---------------------------------------------------------------------------

    Second, the new costs offered by Basin Electric were found to be 
deficient in a number of respects that are discussed more specifically 
in other responses to comments. The new costs estimates included 
numerous costs that were inadequately explained or without any 
supporting documentation. The new cost estimates also did not include 
vendor quotes. Per Basin Electric's Exhibit 14, page 21: ``The LRS 
[Laramie River Station] cost estimates are conceptual in nature; thus, 
S&L did not procure equipment quotes specifically for the LRS control 
systems.''
    EPA also disagrees with the commenter regarding the use of 
satellite images for assessing retrofit difficulty. As noted in 
responses to other comments, because they provide a unique ``bird's 
eye'' view, satellite images are routinely used to evaluate conditions 
at a site: Available space for a crane, access to and from the site, 
interferences that may exist at the site boundary, interferences 
between major pieces of equipment, available space for laying down 
material. With regard to SCR installation, satellite images cannot 
reveal whether or not the air preheater must be relocated to 
accommodate SCR ductwork; however, none of the commenters indicated 
that any affected Wyoming BART sources found it necessary to relocate 
their air preheater. Satellite images cannot reveal the ``ideal 
location'' for reagent storage equipment, although this is not a large 
impact on cost. When possible and resources allow, site visits may also 
provide useful data in addition to satellite images, but these are 
generally performed in addition to rather than in lieu of analysis of 
satellite images.
    Mr. Snell's comments are largely addressed in other comments and 
broadly fall into three areas: (1) Criticism of EPA's use of the IPM 
algorithm for estimating SCR cost; (2) Assertions that EPA failed to 
take into account site-specific factors affecting cost; (3) Assertions 
that EPA failed to take into account balance of plant systems that 
would need to be upgraded. Each of these items raised by Mr. Snell as 
well as the specific issues within each item has been addressed 
elsewhere in other responses to comments.
    Comment: It has been our experience that the effectiveness of SNCR 
is highly dependent upon the characteristics of each boiler. EPA states 
that SNCR typically reduces NOX an additional 20 to 30% 
above combustion controls without excessive NH3 slip. 
NOX reduction with SNCR is known to be greater at higher 
NOX emission rates than lower rates. Accordingly, EPA has 
estimated that the NOX reduction from SNCR as 30% for 
initial NOX greater than 0.25 lb/MMBtu, 25% for 
NOX from 0.20 to 0.25 lb/MMBtu and 20% for NOX 
less than 0.20 lb/MMBtu.
    To support this statement, EPA cites a memo from Jim Staudt, 
Andover Technology Partners (``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., February 7, 2013, p 7), but this memo provides no evidence or 
documentation to support the assumptions that these control levels can 
be achieved. Such assumptions, whether or not supported, can 
significantly affect the outcome of a BART determination, as EPA 
explained regarding Laramie River: ``Therefore, EPA predicts that the 
reduction that can be achieved with SNCR at the Laramie River units is 
20%, which is much lower than the 48% assumed by Wyoming. This 
significantly reduces the tons reduced by SNCR which is in turn used in 
the calculation of cost effectiveness. It also affects the incremental 
cost effectiveness between SNCR and SCR (both in combination with 
additional combustion controls).'' The use of incremental costs in this 
manner is extremely sensitive to bias due to the interjection of 
control strategies based upon invalid assumptions of control 
efficiency.
    Another commenter stated that the EPA is wrong to claim that 
Wyoming overestimated the ability of SNCR to reduce NOX. The 
commenter made the following points in support of this claim:
     The CCM claims that ``[r]eductions of up to 65% have been 
reported for some field applications of SNCR in tandem with combustion 
control equipment such as low NOX burners (LNB).''
     Wyoming's estimates are entirely consistent with 
demonstrated SNCR effectiveness. One study clearly concluded that 
``SNCR has the capability of NOX reductions in the range of 
30-60%, depending on the specific retrofit application.'' See EPRI, 
Cardinal 1 Selective Non-Catalytic Reduction (SNCR) Demonstration Test 
Program, at 1-2 (2000). That study showed, for example, that a 600 MW 
unit equipped with LNB could reduce NOX by an amount greater 
than EPA's ``typical'' results.
     The EPA's AP 42, Fifth Edition, Volume I, Chapter 1: 
External Combustion Sources recognizes that ``[t]he effectiveness of 
SNCR depends on the temperature where reagents are injected; mixing of 
the reagent in the flue gas; residence time of the reagent within the 
required temperature window; ratio of reagent to NOX and the 
sulfur content of the fuel that may create sulfur compounds that 
deposit in downstream equipment.''
    The commenter concluded that EPA's own literature, as well as other 
studies, recognize that SNCR effectiveness is highly contextual and 
that it can achieve reductions far in excess of Wyoming's estimates.
    The commenter asserted that the EPA, without explanation, 
disregarded its own position on the contextual nature of SNCR 
effectiveness, and in turn disregarded Wyoming's well-reasoned analysis 
by relying instead on ``typical'' NOX reductions. The 
commenter believes that the EPA has practiced arbitrary decision making 
because the EPA did not explain in its proposal why it now prefers a 
generic approach to SNCR effectiveness in reducing NOX over 
its previously expressed recognition that effectiveness depends on a 
host of facility-specific factors.
    Response: EPA agrees with the commenter that the effectiveness of 
SNCR is highly dependent upon the characteristics of each boiler, and 
those characteristics include furnace temperature, furnace CO 
concentration, NOX level and other factors, but furnace 
temperature, CO concentration, and

[[Page 5160]]

NOX level are most important. The tendency of NOX 
reduction to decrease as the NOX concentration is reduced is 
a well-established phenomenon. Utility boiler upper furnace 
temperatures are typically in the range of 2000-2300 [deg]F, but can 
sometimes be lower or higher. As described in Section 1.2.3, SNCR 
Performance Parameters in the Control CCM, and also by Sun, Hofmann and 
Pachaly in 1990, and by Muzio, Montgomery, Quartucy and Texeira in 
1993,210 211 the percentage reduction in NOX is 
strongly impacted by the residence time, furnace temperature and the 
starting, or baseline, NOX. Because most utility boiler 
furnace temperatures and residence times fall into an expected range, 
the possible NOX reduction is generally related to baseline 
NOX and Figure 1.5 of the CCM demonstrates the effect of 
baseline NOX and temperature on NOX reduction. Of 
course, there are some units that may fall outside the typical range of 
furnace temperatures or CO levels, and can achieve higher or lower 
levels of NOX reduction. As noted in our response to other 
comments, the furnace temperatures at Laramie River Station as reported 
by Basin Electric in their recently submitted comments are much higher 
than typical, and this will limit the possible NOX 
reduction.
---------------------------------------------------------------------------

    \210\ Sun, W., Hofmann, J., and Pachaly, R., ``Post-Combustion 
NOX Reduction With Urea--Theory and Practice'', Seventh 
Annual International Pittsburgh Coal Conference, September 10-14, 
1990.
    \211\ Muzio, L., Montgomery, T., Quartucy, G., Texeira, D., 
``The Effect of Residence Time On SNCR Processes'', EPRI/EPA 1992 
Joint Symposium on Stationary Combustion NOX Control, Bal 
Harbor, FL, May 24-27, 1993.
---------------------------------------------------------------------------

    On the other hand, EPA disagrees that EPA ``disregarded its own 
position on the contextual nature of SNCR effectiveness, and in turn 
disregarded Wyoming's well-reasoned analysis''. On the contrary, EPA 
carefully considered the contextual situation at Laramie River Station 
and the State's analysis in reaching its opinion. Experience has shown 
that for utility boilers NOX reductions of 48% using SNCR 
alone have only been possible from much higher NOX baselines 
than exist at Laramie River Station. In practice, facility owners have 
generally found that, when using SNCR, the lowest cost approach is to 
first reduce NOX as far as possible with combustion controls 
and then use SNCR for additional reductions beyond what combustion 
controls can provide. As a result, SNCR is rarely used alone to provide 
48% NOX reduction on electric utility boilers because the 
baseline levels in practice are typically too low to achieve such high 
NOX reduction through SNCR.
    The Cardinal Station citation raised by the commenter is from a 
test on a 600 MW unit that had a NOX baseline of around 450-
500 ppm of NOX \212\--in the range of about 0.6 to 0.7 lb/
MMBtu, well above the emission rate of the Laramie River Station units, 
which, after additional combustion controls is about 0.19 lb/MMBtu 
(annual). This unit achieved 25% NOX reduction at full load 
and 30% NOX reduction at 350 MW in long term tests. As a 
result, this project does not support the possibility of 48% 
NOX reduction with SNCR at Laramie River Station, which has 
a much lower baseline NOX level than at the Cardinal 
Station.
---------------------------------------------------------------------------

    \212\ Stallings, J., ``Cardinal 1 Selective Non-Catalytic 
Reduction (SNCR) Demonstration Test Program. EPRI Report 1000154, 
July 2000, pages 4-7 and 8-1.
---------------------------------------------------------------------------

    Finally, in Exhibit 14 to Basin Electric's comments,\213\ Sargent & 
Lundy states that a ``33% reduction is not likely to be achievable'' 
and conditionally indicates that a 20% reduction should be achievable 
from a baseline emission rate of 0.19 lb/MMBtu. EPA agrees that based 
upon the information that is available, 20% is a more reasonable level 
of reduction to expect from SNCR at Laramie River Station.
---------------------------------------------------------------------------

    \213\ Exhibit 14, pages 15, 16.
---------------------------------------------------------------------------

    Comment: Contrary to EPA's assertion, Wyoming's estimate of the 
reduction achievable with SNCR does not depart from the BART 
Guidelines. The Guidelines do not specify the effectiveness of SNCR, so 
there is no contradiction. EPA observes that Wyoming assumed that after 
installation of combustion controls (new LNBs and OFA), SNCR would 
reduce NOX emissions from 0.23 lb/MMBtu to 0.12 lb/MMBtu, a 
48% reduction. EPA, however, insists that its consultant contends that 
SNCR typically reduces NOX by 20% to 30%, depending on the 
level of NOX going to the SNCR unit. According to EPA's 
consultant, when the input level of NOX is 0.19 lb/MMBtu, 
which EPA says was the annual average at Laramie River Station Unit 1 
in 2012, then after installation of new LNBs and OFA the reduction 
achievable with SNCR is only 20%. 78 FR 34748. The consultant says that 
would reduce the NOX emission rate only to 0.15 lb/MMBtu. 
Andover Report at 7.
    The only authority cited by EPA's consultant for the assumed 20% 
reduction is an October 15, 2012 email from Fuel Tech. Id. at 13. No 
information is provided by EPA or its consultant about the expertise of 
Fuel Tech, who at Fuel Tech sent the email, why an email from Fuel Tech 
should be deemed reliable, persuasive or authoritative, or why it 
should take precedence over Wyoming's analysis.
    EPA's statements in the Montana FIP demonstrate that EPA's critique 
of Wyoming's estimate is misplaced. In that case, EPA determined that 
with an inlet concentration of 0.20 lb/MMBtu, SNCR can reduce 
NOX emissions by 25%, as compared to the 20% EPA endorses 
for a nearly identical inlet concentration at Laramie River Station. 77 
FR 23988, 24023, 24032, 24039; 77 FR 57864, 57885-57886. EPA relied on 
information from Fuel Tech to support the feasibility of a 25% 
NOX reduction at this inlet concentration. 77 FR 57885. EPA 
explained that: ``[H]igher NOX reductions can be achieved at 
mid to low load heat inputs, possibly up to 40%. Given that the 
Colstrip Unit 1 and 2 frequently operate at below full load, it is 
likely that on an annual basis SNCR can achieve better than the 25% 
emission reduction assumed by EPA.
    EPA further explained that its review of Clean Air Markets Division 
(CAMD) emissions data showed that ``there are many EGUs equipped with 
SNCR (with combustion controls) that are achieving an emissions rate of 
0.15 lb/MMBtu or lower on a monthly basis.'' Id. at 57886 (emphasis 
added). For example, Boswell Unit 4 had a NOX rate of 0.35 
lb/MMBtu with LNB and close-coupled over fire air (CCOFA). Id. With 
SNCR and SOFA, the unit achieved a monthly NOX rate between 
0.11 and 0.14 lb/MMBtu over a full 12 month period--a reduction of 60% 
to 69%. Id. In response to comments that EPA had overstated the 
benefits of SNCR, EPA stated that it would not adopt a higher post-SNCR 
emission rate ``without a showing that there are circumstances unique 
to Colstrip Unit 1 and 2 that would prevent SNCR from achieving the 
same reductions as at Boswell Unit 4.'' Id.
    Response: As noted in other comments, EPA carefully considered the 
contextual situation at Laramie River Station in reaching its opinion. 
The Wyoming analysis \214\ indicated that NOX was reduced by 
SNCR from 0.23 lb/MMBtu to 0.12 lb/MMBtu. This seemed to be a higher 
level of NOX reduction than expected and inconsistent with 
other experience. The BART analysis suggests no additional 
NOX reduction from OFA versus LNB. Experience has shown that 
for utility boilers NOX reductions of 48% using SNCR alone 
have only been possible

[[Page 5161]]

from much higher NOX baselines than exist at Laramie River 
Station, and a NOX reduction from 0.19 lb/MMBtu (the 
NOX emissions rate after installation of combustion 
controls) to 0.12 lb/MMBtu--roughly 37% reduction--is not likely to be 
feasible either. In practice, facility owners have generally found 
that, when using SNCR, the lowest cost approach is to first reduce 
NOX as far as possible with combustion controls and then use 
SNCR for additional reductions beyond what combustion controls can 
provide. Those coal-fired utility units that the commenter states are 
achieving below 0.15 lb/MMBtu emission rates and are equipped with SNCR 
are also using combustion controls--most often LNBs and SOFA--that 
lower the NOX sufficiently that less than 20% NOX 
reduction is necessary to achieve under 0.15 lb/MMBtu.
---------------------------------------------------------------------------

    \214\ Department of Environmental Quality, Air Quality Division, 
BART Application Analysis, AP-6047, Laramie River Station, May 28, 
2009.
---------------------------------------------------------------------------

    As a result, SNCR is rarely used alone to provide 48% 
NOX reduction on electric utility boilers because the 
baseline levels in practice are typically too low to achieve such high 
NOX reduction through SNCR. In fact, Exhibit 14 to Basin 
Electric's comments, Sargent & Lundy states that ``33% reduction is not 
likely to be achievable'' and conditionally indicates that 20% 
reduction should be achievable from a baseline emission rate of 0.19 
lb/MMBtu. This is more consistent with what EPA has determined.
    EPA also cited input from SNCR technology supplier, Fuel Tech, 
which supports EPA's opinion that an expected NOX reduction 
would be in the range of 20%. Fuel Tech is the largest supplier of SNCR 
technology to the electric utility industry and is therefore a very 
knowledgeable source of information on SNCR.
    Comment: Wyoming did not underestimate the usage and cost of urea, 
and its estimate regarding urea does not conflict with the BART 
Guidelines. EPA contends that producer prices for urea have increased 
over the past three years and that Wyoming's analysis is defective 
because it does not take those price increases into account. EPA, 
however, cannot use information not available at the time of Wyoming's 
BART determination to second-guess that determination. EPA's own 
Guidelines counsel that in making a BART determination, a state should 
consider technologies ``available before the close of the State's 
public comment period,'' but explicitly provide that ``in order to 
provide certainty in the process,'' a state ``need not consider 
technologies that become available after this date.'' 40 CFR Part 51, 
App. Y., section IV(D)(2)(3) (emphasis added). This makes sense. Absent 
some time cutoff, a state's SIP would be in a constant state of flux, 
subject to constant challenge based on ever changing information and 
technology not available to the State at the time it made its BART 
determination. This is particularly true given the amount of time it 
takes EPA to review a state's SIP. It is also consistent with 40 CFR 
51.308(f), which requires states to reevaluate and revise their 
regional haze SIPs every ten years. That regulation clearly 
contemplates that states have a duty to take into account new 
information only in connection with the required periodic SIP 
revisions--not on an ongoing basis.
    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 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 made a number of the same points, stating that 
changes in urea prices are not a valid basis for disapproving the 
state's cost analyses, and even if they were, EPA's facts are mistaken. 
According to the commenter, the EPA asserted that the BART sources 
underestimated the cost of SNCR and EPA supported this conclusion by 
stating that Wyoming underestimated ``SNCR reagent (urea) usage and 
cost.'' The commenter indicated that the EPA did not explain how 
Wyoming underestimated urea usage, but the EPA asserted that ``prices 
for urea have increased in the last three years'' since Wyoming 
submitted its plan to EPA.
    This commenter finds it remarkable that EPA would claim that a 
change in urea prices in the time since Wyoming submitted its SIP 
somehow invalidates the SIP, indicating that the time that has elapsed 
since Wyoming submitted its plan to EPA is due in large part to EPA's 
failure to take timely action on Wyoming's plan. According to the 
commenter, the EPA did not claim that Wyoming's analyses were invalid 
when Wyoming submitted its plan in January 2011, and the EPA did not 
explain how the change in urea market prices led Wyoming to 
unreasonable conclusions. The commenter stated that the EPA appears to 
believe that Wyoming and other states must constantly update their BART 
analyses to account for changing urea market prices up until the date 
that EPA takes final action on the plan. The commenter asserted that 
under this theory, the EPA can hold SIPs hostage, waiting for commodity 
prices to change, and then disapprove SIPs on that basis alone. The 
commenter indicated that the EPA cited no legal basis for this theory.
    The commenter noted that the BART Guidelines expressly acknowledge 
that ``[i]n order to provide certainty in the process,'' states ``need 
not consider technologies that become available after [the close of the 
comment period on the state plan] (citing 40 CFR part 51, App. Y, 
section IV(D)(2)(3)). The commenter believes that in order ``to provide 
certainty in the process,'' the EPA cannot claim that state plans are 
perpetually subject to invalidation as a result of changing commodity 
prices. The commenter stated that the State's price for urea does not 
conflict with the BART Guidelines, and EPA offers no evidence that its 
price is more reliable than the State's price. Commenter provided the 
following additional statements: Even if urea prices have increased, 
assumptions regarding such prices do not constitute a failure to follow 
the BART Guidelines because the Guidelines do not specify what the 
price is or how it should be determined. EPA relies on its consultant's 
report to claim that prices have increased, but that report also says 
that there has been significant variability in cost. Andover Report at 
7-8. There is no analysis by the consultant as to whether, given the 
cited price variability, the current price is likely to go up or down 
in the future or what the actual cost of urea is likely to be going 
forward. The consultant relies on a single source at a single point in 
time to pick a urea price to apply for the life of an SNCR 
installation, with no consideration of the price variability. The 
reliability of the resulting price is no greater than a roll of the 
dice at Las Vegas, and EPA offers no explanation why its consultant's 
price is superior to the State's price. It is merely different,

[[Page 5162]]

and this is yet another technical issue upon which EPA is required to 
defer to the State's assessment.
    EPA's consultant modified its initial report issued in October 2012 
in the revised report issued in February 2013, purportedly to account 
for an alleged change in urea prices. However, the February 2013 report 
cites the very same source for current urea prices as the October 2012 
report. Compare Andover Report (Oct. 23, 2012) at 7. n.23, EPA docket 
cite EPA-R08-OAR-2012-0026-0081, with Andover Report (Feb. 7, 2013) at 
7 n.22, EPA docket cite EPA-R08-OAR-2012-0026-0086. Both reports cite 
the same 10/12/2012 email from Doug Kirk of Fuel Tech. The additional 
10/15/2012 email from Jennifer Zagorsky of Potash Corp that is cited in 
the October report is deleted from the February report. There is no 
explanation for the change.
    Moreover, 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.
    If the State rejected SNCR based on an allegedly too-low cost of 
urea, perhaps EPA could argue that the State was wrong in rejecting 
SNCR. But EPA makes no such argument. It asserts only that its 
consultant's urea price is different from the State's price. As 
explained above respecting SNCR, to succeed in arguing that the State's 
rejection of SNCR justifies disapproval of the State's BART, EPA would, 
at a minimum, have to show that the State was arbitrary and capricious. 
The choice of BART is the prerogative of the State, and the State is 
charged with evaluating and balancing all five BART factors and 
deciding how much weight to give to each factor. EPA may not disapprove 
the State's judgment merely because it disagrees with the State on what 
is a reasonable cost, or how the State balanced costs with other BART 
factors. In fact, nothing in EPA's proposal takes issue with how 
Wyoming weighed or balanced the BART factors, or with the State's 
judgment regarding the terms of the settlement agreement on which the 
State's BART determination for Laramie River was based. EPA's complaint 
is not with the State's judgment in applying the BART factors. Rather, 
it is that the State used information with which EPA disagrees. But 
that cannot justify disapproval of the State's cost analysis or BART 
determination.
    Another commenter argued that the information EPA relied on to 
conclude that urea market prices have increased is itself outdated, 
noting that the report EPA cited as support for its urea price claim 
was completed October 23, 2012, and relied on vendor emails from Fuel 
Tech and PotashCorp dated October 12, 2012 and October 15, 2012, 
respectively, to conclude that urea cost approximately $650 per ton. 
The commenter pointed out that the same report recognizes that ``there 
has been significant variability in [urea] cost,'' and added that since 
the date of that report, urea prices have continued to vary 
significantly, falling by roughly 50 percent.\215\ The commenter noted 
that in its February 2013 revised cost analyses, EPA acknowledged the 
beginning of the price decrease, pegging urea costs at $450 per ton. 
The commenter added that urea prices are today far closer to Wyoming's 
price assumptions than EPA's, which commenter stated were among the 
highest prices for urea in the last four years. Therefore, the 
commenter asserted, even if changes in commodity prices following SIP 
submission were a valid basis for disapproving SIP analyses that relied 
on prices at the time of SIP development, the EPA is factually mistaken 
to claim that Wyoming unreasonably underestimated urea prices; rather, 
the EPA has unreasonably overestimated urea prices by supporting its 
analysis with an abnormally high price that is not reflective of the 
current market.
---------------------------------------------------------------------------

    \215\ Citing PotashCorp., Market Data, August 14, 2013, which 
can be found at http://www.potashcorp.com/customers/markets/market_data/prices.
---------------------------------------------------------------------------

    Response: 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.
    Also, regardless of the cost of urea, EPA found material errors 
with the State's cost analyses for SNCR that required that we revise 
the analysis. In particular, as described in some detail in response to 
other comments, the State significantly overestimated the ability of 
SNCR to reduce NOX at Laramie River Station. There, the 
State assumed that SNCR would reduce NOX by 48%. In response 
to comments above, EPA has definitively established, using information 
from a number sources, including Basin Electric's own consultant, as 
well as a major SNCR supplier, that SNCR cannot approach this level of 
control in the case of Laramie River Station. As such, it was 
appropriate, if not obligatory, for EPA to revisit the SNCR analysis 
for Laramie River Station.
    Finally, today we are providing updated SNCR cost analyses in order 
to address information provided by various commenters. Because we have 
taken into consideration input from a number of commenters when 
revising costs, we believe that they represent the most informed and 
robust costs for SNCR presented yet. In particular, we have revised the 
costs for Laramie River to reflect high furnace temperatures and low 
reagent utilization (a factor not considered in Wyoming's analysis). 
And we have also updated the SNCR costs to reflect the most recently 
available cost of reagent as delivered to Wyoming. Our analyses are 
consistent with our response on a similar comment in the Legal Section 
of this final action.
    Comment: EPA erroneously calculated urea costs. EPA made two 
fundamental and significant errors that have the effect of overstating 
the costs of SNCR, which in turn justified the EPA's conclusion that 
SCR is cost effective. The errors are as follows:
     EPA mistakenly converted pounds to tons in its calculation 
of operation and maintenance costs for urea. See EPA's Revised Cost 
Analyses for Jim Bridger Units 1-4--Detailed Spreadsheet Supporting 
Analyses (NOX-SNCR tab, rows 62-64) (Bridger Costs); EPA's 
Revised Cost Analyses for

[[Page 5163]]

Wyoming Sources--Detailed Spreadsheet Supporting Analyses 
(NOX-SNCR--0 1--03 tab, rows 62-64) (EPA Costs). The cost 
formula multiplies the urea rate (pounds/hour) times the cost (dollars/
ton) and divides that product by the source's megawatt rating to yield 
a dollar per megawatt hour cost for urea. In converting pounds to tons, 
EPA mistakenly divided by 1,000, when it should have divided by 2,000 
(the number of pounds in a ton).
     EPA incorrectly calculated the water dilution variable for 
operation and maintenance costs in urea. See Bridger Costs 
(NOX-SNCR tab, rows 62-64); EPA Costs (NOX-SNCR--
01--03 tab, rows 62-64). EPA's cost calculation incorporates the wrong 
spreadsheet cell (auxiliary power cost). It should have instead 
incorporated spreadsheet cell for the hourly water rate in thousands of 
gallons per hour.
    Response: The reagent cost calculation is correct. The urea rate 
(assuming 100% urea) is multiplied by the cost for 50% by weight urea 
and is multiplied by 2 (to account for the fact that the cost is for 
50% by weight urea) and then divided by 2000 (for the tons to pounds 
conversion). The effect is to divide by 1000, which is the equation 
shown. Commenter is correct that there was an error in the dilution 
water cost calculation. The error has been corrected in EPA's revised 
cost estimates. The error has negligible impact on the estimated cost 
of SNCR.
    Comment: EPA asserts that it was an error for Wyoming to evaluate 
SNCR using a controlled emission rate of 0.12 lb/MMBtu, which is about 
a 48% reduction from 0.23 lb/MMBtu (the rate without new LNB and OFA). 
78 FR 34748. EPA claims that after combustion controls reduce emissions 
to 0.19 lb/MMBtu, SNCR can achieve only a 20% further reduction, to 
0.15lb/MMBtu. Id., citing a Fuel Tech vendor report. EPA declined to 
accept Wyoming's conclusion that SNCR would cut emissions by 20% to 
30%. Id.
    This conflicts with EPA's findings in the North Dakota FIP. 77 FR 
20898. EPA found that SNCR plus LNB and SOFA at Coal Creek Station 
Units 1 and 2 could reduce NOX from a baseline of 0.22 lb/
MMBtu to 0.115 lb/MMBtu, which is a 48% reduction. EPA explained that 
after combustion controls reduced emissions to a degree, SNCR by itself 
would cut emissions another 25%, despite the facility's claim that SNCR 
would achieve only a 20% reduction. Id., citing a Fuel Tech report.
    Response: EPA does not dispute that SNCR in combination with 
combustion controls can reduce NOX by 48% in some cases. As 
described in our response to other comments, EPA does not agree that 
48% reduction of NOX is possible at Laramie River Station 
using SNCR alone.
    Comment: We agree with EPA that on an annual basis SCR can achieve 
emission rates of 0.05 lb/MMBtu or lower. We recommend that EPA 
consider that some coal-fired EGUs are achieving lower emissions; e.g., 
our search of the CAMD database found seven conventional coal-fired 
EGUs averaging 0.04 lb/MMBtu or lower on an annual basis in 2012. 
Unlike SNCR, for SCR the ability to achieve low NOX 
emissions is less a function of boiler characteristics and more a 
function of SCR design; it is generally accepted that SCR can reduce 
NOX emissions by 80-90+%. However, the average control 
efficiency assumed by EPA for all Wyoming EGUs was 75% (74% median 
value).
    The efficiency of NOX removal is determined primarily by 
the amount of catalyst used, as pointed out by Hitachi in an email from 
Hitachi to EPA Region 9 regarding SCR at the Navajo Generating Station. 
In response to a question from the EPA on SCR NOX 
performance guarantee, Hitachi replied that a 3 plus 1 SCR design could 
be designed to guarantee NOX emissions of 0.05 lb/MMBtu on a 
30-day rolling average. However, Hitachi also stated that the utility 
and their engineer need to determine what margin needs to be applied to 
insure the unit is capable of achieving less than the permit level on a 
30-day rolling average. The EPA stated that in an engineering study 
performed by Sargent & Lundy that with a NOX permit limit 
between 0.07 and 0.08 lb/MMBtu, the SCR would be designed for 0.05 lb/
MMBtu. The difference between 0.05 and 0.07 is the margin necessary for 
compliance. By underestimating the efficiency of SCR and potentially 
overestimating the efficiency of SNCR, EPA has overestimated the 
incremental costs for SCR.
    Response: We agree with the information provided by the commenters 
that SCR technology has, in some cases, the potential to achieve 
emissions of less 0.05 lb/MMBtu (annual). However, emission limits 
associated with BART do not need to meet the lowest emission rate 
achieved with that technology at any coal-fired power plant. 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.'' 40 CFR 51.308(e)(1)(ii)(A).
    In determining the controlled emission level, EPA must consider 
emission rates that are practically achievable in light of routine 
variations in operation and understanding that the SCR must be designed 
to maintain emissions below the required limit. SCRs in the U.S. are 
typically either 2 plus 1, or 3 plus 1 systems, with two or three 
initially full catalyst layers plus a spare layer for future catalyst 
additions. EPA is not aware of, nor has commenter provided information 
for, any 4 plus 1 SCR systems operating on coal-fired utility boilers. 
Therefore, EPA would favor more commonly used 2 plus 1 or 3 plus 1 SCR 
designs rather than the 4 plus 1 system described in commenter's 
citation from Hitachi.
    Additionally, the BART Guidelines state that: ``[i]n assessing the 
capability of the control alternative, latitude exists to consider 
special circumstances pertinent to the specific source under review, or 
regarding the prior application of the control alternative'' (40 CFR 
Part 51, Appendix Y, section IV.D.3) and that ``[t]o complete the BART 
process, you must establish enforceable emission limits that reflect 
the BART requirements''. (40 CFR Part 51, Appendix Y, section V). The 
five-factor BART analysis described in the Guidelines is a case-by-case 
analysis that considers site-specific factors in assessing the best 
technology for continuous emission controls. After a technology is 
determined as BART, the BART Guidelines require establishment of an 
emission limit that reflects the BART requirements, but does not 
specify that the emission limit must represent the maximum level of 
control achieved by the technology selected as BART.
    While the BART Guidelines and the RHR do not preclude selection of 
the maximum level of control achieved by a given technology as BART, 
the emission limit set to reflect BART must be determined based on a 
consideration and weighting of the five statutory BART factors. 
Therefore, limits set as BACT during PSD review, or emission rates 
achieved from the operation of individual facilities under an emissions 
trading program (e.g., CAA Interstate Rule) may provide important 
information, but should not be construed to automatically represent the 
most appropriate BART limit for a given technology.
    As noted in our response to other comments, EPA does not believe 
that we have overestimated the performance of SNCR, nor does EPA 
believe that the performance of SCR has been underestimated.

[[Page 5164]]

    Comment: EPA's errors in calculating SCR costs resulted in 
significantly skewed cost-effectiveness determinations for every unit 
analyzed. EPA overstated the costs per ton of SCR by between 33 and 99 
percent. Although even EPA's cost estimates for SCR are within the 
range that the EPA previously has found reasonable, the revised cost 
analyses correcting what the commenter believes are EPA errors make it 
clear that SCR is cost effective for every Wyoming EGU.
    Response: Commenter's assertion that EPA's costs are too high is 
largely based upon disagreement over the retrofit factors used, EPA's 
inclusion of a provision for taxes and insurance, and disagreement with 
use of 7% interest in determining the capital recovery factor. EPA has 
responded to each of these issues in other comment responses and has 
developed revised cost estimates that will incorporate changes where 
EPA believes the changes are warranted.
    Comment: Wyoming has overestimated the cost of SCR. Wyoming has not 
provided justification or documentation for their cost estimates. We 
(the commenter) were not provided with any vendor estimates or bids, 
and Wyoming did not use the CCM, as recommended by the BART Guidelines. 
For example, the cost estimates used by Wyoming and EPA contained 
AFUDC, which is not allowed by the CCM and has been rejected by EPA 
Region 8 in other analyses. As a result, total capital costs estimated 
by Wyoming for SCR exceeded $300/kW at ten of the fifteen EGUs 
evaluated. EPA has compiled a graphic presentation of SCR capital costs 
adjusted to 2009 dollars. The EPA data confirm that SCR capital costs 
typically range from $73-$243/kW. Wyoming has not demonstrated unique 
features for the Wyoming EGUs that would justify cost estimates so much 
higher than the range for the industry.
    Response: We agree that in some cases Wyoming has overestimated the 
cost of SCR. In order to address the cost analysis deficiencies noted 
by the commenter, EPA has performed revised cost analyses for EGUs 
where the cost of SCR is pertinent. In our revised cost analyses, we 
have followed the structure of the CCM, though we have used the IPM 
cost models to estimate direct capital costs and operating and 
maintenance (O&M) costs.
3. Consideration of the Five Factors
    Comment: We received numerous comments that the State followed the 
requirements of the RHR and CAA, and simply did not come to the same 
conclusions as EPA. Commenters stated that Wyoming's BART 
determinations were based on a consideration of all five BART factors 
and that the State weighed each factor appropriately.
    Response: We disagree with this comment. As discussed in detail in 
section VII.C.3.a of our proposed rulemaking, the EPA identified 
numerous issues and errors with the State's cost analyses, including 
the fact that the State underestimated the cost of SNCR and 
overestimated the cost of SCR; the State overestimated the emission 
reductions from SNCR; the State underestimated the control efficiency 
of SCR; the State overestimated the capital costs for SCR; and the 
State allowed for some costs not allowed by the CCM and thus their cost 
analyses did not meet the requirements of the RHR. 78 FR 34748.
    Likewise, for the visibility improvement modeling, EPA discussed in 
detail in section VII.C.3.b of our proposed rulemaking why the State's 
visibility modeling did not meet the requirements of the RHR (78 FR 
34749). As stated in our proposed rulemaking, Wyoming did not consider 
the visibility improvement associated with SNCR, which is clearly in 
conflict with the requirements set forth in section 169A(g)(2) of the 
CAA, as well as in the implementing regulations, 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.'' In addition, it was not possible for EPA, or any 
other party, to ascertain the visibility improvement that would result 
from the installation of the various NOX control options 
because Wyoming modeled the emission reductions for multiple pollutants 
together in its SIP. Finally, Wyoming did not establish baseline 
emission rates used for modeling in a manner consistent with BART 
Guidelines. That is, Wyoming did not use ``the 24-hour average actual 
emission rate from the highest emitting day of the meteorological 
period modeled (for the pre-control scenario).'' 70 FR 39170. Instead, 
Wyoming modeled baseline emission rates reflective of permitted 
emission limits, leading to both an underestimation, and in some cases, 
overestimation of visibility impacts.
    Therefore, contrary to the commenters claim, today's action is the 
result of Wyoming's failure to meet certain statutory and regulatory 
requirements, and not a simple matter of the State and EPA arriving at 
different conclusions.
    Comment: Wyoming's BART NOX determinations for the 
Naughton power plant further demonstrate Wyoming's consideration and 
balancing of all five factors, including visibility improvement, and 
its individualized consideration for each unit. For Naughton Units 1 
and 2, Wyoming found that costs of compliance (total capital costs and 
cost effectiveness), power losses (energy impacts) caused by post-
combustion NOX controls, environmental considerations 
related to chemical reagents used with post-combustion NOX 
controls (non-air quality environmental impacts), and visibility 
improvement information indicated that LNBs and OFA are BART 
NOX. However, for Naughton Unit 3, based upon its much 
greater ``visibility improvement'', Wyoming determined that SCR is BART 
NOX. Wyoming's BART NOX analyses across the 
Naughton Plant's three units demonstrate Wyoming's consideration and 
weighing of all five BART factors, including the decision to require 
different levels of BART NOX controls across various units 
at the same plant when Wyoming determined that the visibility 
improvements and other factors at one unit justified more stringent 
control. This example is yet one more indication, contrary to EPA's 
assertions, that Wyoming did adequately consider ``visibility 
improvement'' information in each of its BART determinations, including 
Wyoming deciding in its discretion the ``weight and significance'' 
appropriate for each BART factor at each BART unit.
    Response: We disagree with the commenter's assertion that the 
State's determination for the Naughton units shows how the State 
considered all five factors when the information that the State was 
relying on was not accurate as pointed out in our response above.
    Comment: PacifiCorp submitted its BART studies to Wyoming in 2007, 
and the State completed its BART analyses during 2008. At that time the 
remaining useful life of all PacifiCorp BART units was considered to be 
at least 20 years. Primarily due to EPA's delays in dealing with the 
Wyoming Regional Haze SIP, this assumed twenty-year life span is no 
longer a valid basis for certain units. EPA now must take into account 
the current useful life of the units, rather than the useful life 
assumed under Wyoming's BART analyses completed at a different point in 
time. Dave Johnston Unit 3's current depreciable life ends in 2027 and 
the life for Naughton Units 1 and 2 ends in 2029.
    As a practical matter, the SCRs required under the regional haze 
FIP at Dave Johnston Unit 3 and Naughton Units 1 and 2 could not be 
installed until shortly before the end of 2018, due to the regulatory 
processes that apply to PacifiCorp's major investment

[[Page 5165]]

decisions, as well as the associated permitting and competitive 
procurement timelines. At that time, the useful life for Dave Johnston 
Unit 3 will be nine years, and for Naughton Unit 1 and 2 eleven years. 
EPA must use these shorter useful lives in its BART analyses. Taking 
into consideration the remaining useful lives of these particular BART 
units clearly demonstrates that EPA's current assessed cost 
effectiveness conclusions (whether using the Andover Report costs or 
PacifiCorp's updated information) do not support the installation of 
SCR on these units because they are not cost effective. To the extent 
EPA needs to include firm retirement dates commensurate with the 
depreciable lives for purposes of finalizing the regional haze FIP, 
then PacifiCorp requests that EPA do so.
    Response: We agree in part. However, because of our revised cost 
and visibility analyses, and our conclusions regarding BART that stem 
from those analyses, the comment is no longer pertinent to all of the 
units in question.
    Using a remaining useful life of 20 years, our revised analysis for 
Naughton Unit 1 shows that the cost effectiveness of new LNBs with OFA 
and SCR is $3,109/ton, while the incremental cost effectiveness is 
$10,384/ton. The visibility improvement associated with new LNBs with 
OFA and SCR is 0.33 deciviews. Similarly, using a remaining useful life 
of 20 years, our revised analysis for Naughton Unit 2 shows that the 
cost effectiveness of new LNBs with OFA and SCR is $2,566/ton, while 
the incremental cost effectiveness is $8,440/ton. The visibility 
improvement associated with new LNBs with OFA and SCR is 0.42 
deciviews. Given these costs and visibility improvements, taken along 
with the other BART factors, we no longer find that SCR is warranted 
for Naughton Units 1 or 2, even assuming a longer remaining useful 
life. Therefore, because the commenter suggested alternative control 
options in lieu of the proposed SCR, which we would otherwise not 
require, the comment is no longer pertinent to these two units. 
However, as described below, it remains relevant to Dave Johnston Unit 
3.
    Using a remaining useful life of 20 years, our revised analysis for 
Dave Johnston Unit 3 shows that the cost effectiveness of LNBs with OFA 
and SCR is $2,635/ton, while the incremental cost effectiveness is 
$7,583/ton. The visibility improvement associated with new LNBs with 
OFA and SCR is 0.51 deciviews. Given these costs and visibility 
improvement, taken along with the other BART factors, we continue to 
find that SCR would be warranted for Dave Johnston Unit 3. However, 
using a remaining useful life of 9 years, as identified by PacifiCorp, 
our analysis shows that the cost effectiveness of LNBs with OFA and SCR 
is $3,742/ton, while the incremental cost effectiveness is $11,781/ton. 
Given the costs that result from the shorter remaining useful life, 
along with other BART factors, we find that SCR is not warranted. As a 
result, we find that combustion controls (LNBs) and an earlier 
retirement date are BART for Dave Johnston Unit 3.
    We note that depreciable life is the result of financial accounting 
rules, such as for tax purposes, and is determined by capital 
investments in the plant and associated accounting rules for the timing 
of depreciation of those capital investments. As a result, the 
depreciable life is often shorter than the economic life of the 
facility. Economic life, which is the actual expected viable life of 
the facility, is the key consideration in regard to the remaining 
useful life (one of the five BART factors). As a result, depreciable 
life is not relevant to a BART analysis unless the depreciable life 
that results from a capital investment for BART is longer than the 
economic life of the facility, in which case asset impairment charges 
could result at the end of the economic life. Nonetheless, we 
understand PacifiCorp's comment as meaning that, for financial reasons, 
they would prefer to shutdown the units on an accelerated schedule in 
lieu of installing SCR.
    Finally, while PacifiCorp has presented revised cost information 
along with their comments, we have not accepted these costs without 
examination. As described in other responses, while allowing some of 
the costs suggested by PacifiCorp, we have not allowed others. More 
information regarding our cost analyses for the units in question can 
be found in the cost report located in the docket.\216\
---------------------------------------------------------------------------

    \216\ Andover Technology Partners, ``Cost of NOX 
Controls on Wyoming EGUs'', October 28, 2013; Wyoming EGU BART and 
Reasonable Progress Costs--10/28/2013; Wyoming EGU BART and 
Reasonable Progress Costs for Jim Bridger--10/28/2013.
---------------------------------------------------------------------------

    Comment: In its proposed rule, the EPA found that the limits and 
technologies mandated in the rule are cost effective based on 
amortizing those costs over a 20 year period. Here, the Agency's cost 
modeling is seriously flawed as many of the units subject to the new 
rule have remaining lives significantly less than 20 years. For 
example, Dave Johnston has a remaining life of only 14 years and 
Naughton 16 years. Amortizing the larger investment required by the FIP 
over these shorter lives would cause rates to go up even more, casting 
doubt on the veracity of the EPA's conclusion that the FIP is cost 
effective.
    Response: See response above. We note, however, that we are using 
the remaining useful life periods as presented by PacifiCorp in the 
comment above for Dave Johnston Unit 3, which differ from this 
commenter's numbers for remaining useful life.
    Comment: We received comments that the State considered the energy 
and non-air quality environmental impacts of compliance when developing 
the State's plan. The commenters went on to say that it is not apparent 
that EPA addressed the energy and non-air quality impacts in their 
analyses.
    Response: We disagree with this comment. Throughout our proposed 
rulemaking, we consistently acknowledged that we are proposing to 
accept the State's energy and non-air quality impacts analysis (e.g. 78 
FR 34759). In the State analyses for all BART sources, it states that 
the energy and non-air quality impacts do not preclude the selection of 
any of the control technologies the evaluated for the BART sources. In 
weighing all of the BART factors ourselves, we agree with this 
conclusion and adopt it as our assessment of the energy and non-air 
quality impacts.
    Comment: There are three types of energy impacts that should be 
considered. These include the energy associated with operating the 
controls, the energy that must be provided when the unit is removed 
from service in order to install the controls, and most importantly to 
Wyoming and its citizens, the energy that must be replaced when the 
emissions controls prescribed for a given unit are not economically 
justifiable and result in accelerated unit retirements and 
replacements.
    The latter scenario is of particular concern because the EPA has 
now proposed SCR controls for PacifiCorp's Naughton Unit 1, Naughton 
Unit 2 and Dave Johnston Unit 3. Unlike the Wyoming SIP, the EPA's FIP 
requires controls that are not expected to be justifiable and would 
result in accelerated unit retirements and replacements, potential 
natural gas conversions, and the associated costs and socio-economic 
impacts of removing major coal-fueled generation resources from service 
in areas of Wyoming that rely heavily on these facilities.
    Response: The commenter raises concerns about energy impacts,

[[Page 5166]]

specifically: The energy associated with operating the controls, the 
energy that must be replaced when the unit is taken out of service, and 
the energy that must be supplied if the unit is retired in lieu of 
addition of controls. The energy associated with operating the controls 
are accounted for in the variable operating cost of SNCR and SCR. Most 
of the construction occurs with the unit operating, but the unit must 
be shut down when ductwork tie-ins are made to the SCR. Regarding 
replacement energy when the unit is taken out of service, the 
generation units have periodic outages of several weeks for major 
maintenance items, such as turbine overhaul where there is adequate 
time to make the tie-ins for the equipment. It is reasonable to assume 
that facility owners would schedule outages for the SNCR or SCR 
retrofits during periods when other maintenance is being performed that 
requires the unit to be out of service, and this is what is commonly 
done in practice. EPA has allowed five years after the final rule to 
meet the emission limits, which should provide companies ample 
opportunity to schedule retrofit activities during a normally scheduled 
outage. As a result, retrofit of NOX controls would not have 
a significant impact on the energy production of the generating unit.
    As for the energy that must be replaced if a unit is retired, the 
CAA and BART Guidelines do not explicitly require that this impact be 
taken into consideration as part of the non-air quality and energy 
impacts.
    Comment: EPA's proposed approach is a myopic effort to focus on 
only one portion of what is supposed to be a multi-faceted decision. 
Appendix Y became law after notice-and-comment rulemaking, and states 
are justified in relying on it when crafting their regional haze SIPs. 
Indeed, EPA made clear that the Appendix Y guidelines ``are designed to 
help States and others . . . determine the level of control technology 
that represents BART for each source.''
    BART determinations are composite decisions, with many facts and 
data from each of the five BART factors playing a role in the ultimate 
BART determination as decided by Wyoming. EPA's proposal to pluck out a 
single BART factor (visibility improvement) as the sole justification 
for rejecting Wyoming's entire NOX BART determination for 
some units is arbitrary and capricious because it makes a single factor 
more important than any of the others and also more important than the 
composite BART determination as a whole. EPA's approach also disregards 
each of the five BART factors as Wyoming evaluated them and ignores the 
``weight and significance'' of each factor alone, and in combination 
with the others, as Wyoming determined in its BART decisions.
    Response: We disagree with this comment. The commenter is not 
correct in asserting that EPA rejected the State's BART determinations 
for certain sources based only on a single BART factor--visibility 
improvement. EPA's rejection of the State's BART determination was 
based on EPA's consideration of all five BART factors. Nowhere in our 
notice do we indicate that we are rejecting the State's BART 
determination based solely on the consideration of visibility 
improvement. Moreover, as noted elsewhere in these responses to 
comments, we found several instances in which the State's analyses were 
inconsistent with the RHR and BART Guidelines, requiring that EPA 
revise the State's analyses.
    Comment: Use of the BART guidelines is only required for sources 
located at electric generating facilities with a total capacity greater 
than 750 megawatts. See 40 CFR 51.308(e)(ii)(B). Only three power 
plants in Wyoming met these criteria: Basin Electric's Laramie River 
Station, PacifiCorp's Jim Bridger, and PacifiCorp's Dave Johnston 
plants. For consistency, and as a matter of State discretion, Wyoming 
went above and beyond the requirements by following the five-step 
process for all BART sources, not solely the three aforementioned large 
electric generating facilities. EPA should commend Wyoming for taking 
this approach, not use it as an excuse for invalidating the SIP.
    Response: We agree that the BART guidelines are only mandatory 
under the regional haze regulations for ``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 guidelines for other BART sources does not mean that the state 
has unfettered discretion to act unreasonably or inconsistently with 
the CAA and our regulations. Where the BART guidelines are not 
mandatory, a state must still meet the requirements of the CAA and our 
regulations. In other words, the State must still adopt and apply the 
best available retrofit technology, considering the statutory factors.
    Comment: Based on the erroneous claims that the SIP incorrectly 
analyzed costs, calculated baseline NOX emissions, and 
modeled visibility improvement, EPA proposes a FIP for eight BART 
sources in Wyoming. For each of these sources, EPA proposes to approve 
all of the State's BART NOX analyses, except for the cost of 
compliance, baseline emissions, and visibility factors. In other words, 
EPA approves the State's analyses of some BART factors, but not the 
others.
    EPA, however, does not explain how it weighed the five BART factors 
after substituting its cost of compliance, baseline emissions, and 
visibility modeling for the State's. For example, for the Laramie River 
Station units, EPA reiterates its disagreement with the State's 
analyses and shows how its analyses change those factors. 78 FR 34776. 
But EPA does not explain how it analyzed those new factor conclusions 
in relation to the remaining Wyoming BART factors that EPA proposes to 
approve. For each of the eight BART units, EPA takes the same approach, 
failing to explain how it balanced the multiple BART factors.
    Response: We disagree with this comment. As stated above, EPA came 
to its conclusions on the State's BART determinations based on a 
consideration of the five factors on an individual source basis. We 
considered the visibility benefits and costs of control together by 
weighing the costs in light of the predicted visibility improvement and 
the other BART factors.
    Comment: There are no threshold minimum acceptable cost 
effectiveness levels, nor any requirements regarding how much weight a 
state must give to cost factors versus other factors such as 
visibility. EPA provides no explanation regarding what it views as a 
reasonable cost factor, or why or how such a factor should be balanced 
with visibility factors.
    Response: See response above.
    Comment: EPA must consider the energy that must be replaced when 
the emissions controls prescribed for a given unit are not economically 
justifiable and result in accelerated unit retirements and 
replacements. This scenario is of particular concern because the EPA 
has now proposed SCR controls for PacifiCorp's Naughton Unit 1, 
Naughton Unit 2 and Dave Johnston Unit 3. Unlike the Wyoming SIP, the 
EPA's FIP requires controls that are not expected to be justifiable and 
would result in accelerated unit retirements and replacements, 
potential natural gas conversions, and the associated costs and socio-
economic impacts of removing major coal-fueled generation resources 
from service in areas of Wyoming that rely heavily on these facilities.
    Response: As noted above, the CAA and BART Guidelines do not 
explicitly require that these impacts be taken into

[[Page 5167]]

consideration as part of the non-air quality and energy impacts.
    Comment: EPA's assertion that Wyoming underestimated the ability of 
SCR to reduce NOX was arbitrary. The EPA cited no legal or 
factual support for its assertion (at 78 FR 34748) that SCR can achieve 
emission rates of 0.05 lb/MMBtu or lower on an annual basis. EPA 
approved Colorado's use of a 0.07 lb/MMBtu annual emission rate for SCR 
at coal-fired power plants because the EPA explained (at 77 FR 76871, 
76873) that rate ``is within the range of actual emission rates 
demonstrated at similar facilities in EPA's Clean Air Markets Division 
(CAMD) emission database.'' EPA also said in that Colorado action that 
an emission rate as low as 0.05 lb/MMBtu can be achieved only ``in some 
cases[.]'' In its proposed disapproval of Wyoming's SIP, the EPA has 
not explained why Wyoming's analyses are distinct from Colorado's.
    Response: We disagree. In fact, the cost analyses submitted by both 
PacifiCorp and Basin Electric in comments support EPA's assumption that 
0.05 lb/MMBtu is achievable on an annual basis. PacifiCorp's comments 
include a budgetary price estimate for three units from Babcock & 
Wilcox indicating an outlet NOX rate of 0.04 lb/MMBtu.\217\ 
Though Babcock & Wilcox does not specify the averaging time basis of 
this rate, because emission rates are lower over longer averaging 
times, the emission rate would only be lower if not already expressed 
on an annual basis. Similarly, the report prepared for Basin Electric 
by Sargent & Lundy indicates an annual emission rate of 0.05 lb/MMBtu 
for the Laramie River units.\218\ Therefore, it does not appear that 
either Sargent & Lundy or Babcock & Wilcox dispute whether SCR is 
capable of achieving an annual emission rate of 0.05 lb/MMBtu. In 
addition, the commenter has not provided any information to 
substantiate that SCR cannot achieve an actual annual emission rate of 
0.05 lb/MMBtu.
---------------------------------------------------------------------------

    \217\ Letter from Babcock & Wilcox Power Generation Group, Inc., 
to PacifiCorp Energy, page 3, August 19, 2013.
    \218\ Laramie River Station SNCR and SCR Cost Estimates, 
prepared for Basin Electric Power Cooperative by Sargent & Lundy, 
Table 1, page 14, August 26, 2013.
---------------------------------------------------------------------------

    Further, as noted by other commenters, information in the CAMD 
database reveals that there a number of coal-fired EGUs retrofitted 
with SCR which are achieving actual emissions of 0.05 lb/MMbtu or less 
on an annual basis. It is important to note that the commenter is 
questioning the annual emission rate achievable with SCR (0.05 lb/
MMBtu) that EPA assumed for the purpose of calculating cost 
effectiveness. By contrast, when establishing a 30-day emission limit 
for SCR, the annual rate must be adjusted upward to account for: (1) A 
margin for compliance, (2) a shorter averaging period, and (3) start-up 
and shutdown emissions. Therefore, EPA agrees that a 30-day rolling 
average emission limit of 0.07 lb/MMBtu is appropriate for SCR. In 
fact, we have approved this emission limit for Wyoming sources where 
the State has required the installation of SCR. However, we continue to 
find that it was appropriate for EPA to use the anticipated actual 
annual emission rate, as opposed to the allowable 30-day limit, in 
calculating cost effectiveness. The approach taken by EPA is consistent 
with the BART Guidelines: In general, for the existing sources subject-
to-BART, you will estimate the anticipated annual emissions based upon 
actual emissions from a baseline period. 70 FR 39167. That is, cost 
effectiveness is more appropriately based on the reduction in annual 
emissions, not the change in allowable emissions.
    Finally, we disagree that we have treated Wyoming in a manner 
distinct from Colorado with regard to the control effectiveness of SCR. 
As noted by the commenter, in Colorado we held that SCR can achieve an 
annual emission rate of 0.05 lb/MMBtu. However, in Colorado we also 
held that it was unlikely that an analysis performed around this rate 
would have altered the state's conclusions regarding BART. For units 
where Colorado did require the installation of SCR (Craig Unit 2, 
Hayden Units 1 and 2, and Pawnee), Colorado established a 30-day 
rolling average emission limit of 0.07 to 0.08 lb/MMbtu. These emission 
limits are commensurate with those established in Wyoming by both EPA 
and the State for SCR.
4. Visibility Improvement
    Comment: The implementation by EPA of its NOX FIP is an 
overreach of its authority given the record in this case. In 
particular, as it relates to the Laramie River Station, EPA arbitrarily 
requires in its FIP the installation of SCRs to address regional haze. 
However, the facts reveal that the installation of SCRs is not 
justified because--even based on EPA's own calculations--the visibility 
improvement that could be achieved is imperceptible. Nowhere in EPA's 
proposed rule does it evaluate its FIP as achieving an improvement in 
visibility at an individual Class I area that meets the standard 
deciview definition, i.e., a full deciview being equal to the amount of 
visibility improvement that is detectable by the human eye. In 
addition, EPA failed in its analysis to consider the predicted change 
in visibility between control options and whether the incremental 
improvement for a given control is even perceptible. Conceding these 
facts, the imposition of a FIP to achieve an imperceptible improvement 
in visibility is arbitrary and capricious.
    Response: We disagree that the visibility improvements for Laramie 
River Station or other BART sources are de minimis or too small to just 
justify the expense of requiring controls. 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. 70 FR 
39129.
    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. 
The same facts apply to the commenter's assertions on incremental 
visibility improvement.
    In addition, we received numerous general comments that controls on 
sources were not warranted because the visibility improvement was less 
than the perceptible amount of 1.0 deciview, to which we respond in the 
same way.
    Comment: EPA's reliance on cumulative analysis of visibility 
improvement is contrary to the CAA. The aggregate approach EPA is 
employing in its proposed rule has been rejected by the D.C. Circuit. 
The Court held that an EPA requirement for a group consideration of 
visibility impacts was not allowed by the CAA. Instead, EPA must 
consider all five BART factors for each source. As the Court explained, 
with the cumulative approach, ``it is therefore entirely possible that 
a source may be forced to spend millions of dollars for new technology 
that will have no appreciable visibility improvement.''
    Response: We disagree that our consideration of visibility 
improvement was contrary to the CAA. Here the commenter has conflated 
two separate issues related to cumulative visibility analyses. In the 
D.C. Circuit ruling, American Corn Growers Ass'n v. EPA, 291 F.3d 1 
(D.C. Cir. 2002), the issue was related to the cumulative visibility

[[Page 5168]]

impacts from multiple sources. There, the court held that a source may 
be unduly required to install controls because of the emissions from 
other sources. By contrast, in the instance related to our proposed 
rule, the issue is related to the cumulative visibility impact to 
multiple Class I areas from a single source. Therefore, there is no 
relationship between the approach rejected by the D.C. Circuit Court 
and that used in our assessment of visibility improvement.
    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. 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. The alleged incremental visibility benefit 
of installing SNCR at Wyodak is 0.12 delta deciview at an incremental 
cost of $3,725. 78 FR 34784-85. 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: As stated elsewhere in our response to comments, we must 
consider the five factors for each facility when making a BART 
determination. Even though one factor (such as visibility improvement) 
may be similar between two units, it must be weighed in the context of 
the other BART factors. In addition, as we discuss in other response to 
comments, in accordance with the BART Guidelines, controls may be 
warranted even in instances where the visibility benefit is less than 
perceptible.
    We note that, in light of comments submitted during the public 
comment period, we have revised our BART determinations for the 
Naughton Units 1 and 2, the Dave Johnston Units 3 and 4, and Wyodak. 
See sections III.B and V.D for a discussion on our consideration of the 
BART factors and our BART determination for these units.
5. PM BART Determinations
    Comment: We received comments that EPA's BART determinations with 
respect to PM emissions from Wyoming EGUs are flawed. One commenter 
pointed out that contrary to the BART guidelines, EPA failed to propose 
BART limits on condensable PM and total PM (PM2.5 + 
PM10), focusing instead solely on filterable PM. Commenters 
went on to state that EPA underestimated the control effectiveness of 
baghouses, which should be able to achieve a limit of 0.010 lb/MMBtu or 
even lower, and thus EPA overestimated the costs effectiveness for 
baghouses.
    Response: We disagree with both points made in this comment. On the 
first point, the BART Guidelines do not explicitly require that states 
establish separate emission limits for condensable PM.\219\ However, we 
do recognize, by merit of the compliance test methods specified for PM 
(e.g. EPA Method 5B), that the BART emission limits in the Wyoming SIP 
only pertain to filterable PM.
---------------------------------------------------------------------------

    \219\ The BART Guidelines do not specify that States must 
establish a BART limit for both PM10 and 
PM2.5. The BART Guidelines provide the following: ``You 
must look at SO2, NOX, and direct particulate 
matter (PM) emissions in determining whether sources cause or 
contribute to visibility impairment, including both PM10 
and PM2.5.'' [Appendix Y to Part 51, section III.A.2.] 
This language in the BART Guidelines was intended to clarify to 
States that when determining whether a source is subject to BART, 
the modeling evaluation to determine the source's impact on 
visibility has to account for both PM10 and 
PM2.5 emissions. There are several instances in which we 
state in both the preamble to the RHR, and in the BART Guidelines 
that PM10 may be used as indicator for PM2.5 
in determining whether a source is subject to BART. Neither the RHR 
nor the BART Guidelines specify that states must make separate BART 
determinations for PM10 and PM2.5. Therefore, 
we disagree that we must evaluate separate limits or disapprove the 
PM BART determination for the Wyoming SIP on the basis that a BART 
determination for PM2.5 was not made.
---------------------------------------------------------------------------

    On the second point, the commenter has not provided any data or 
information to substantiate that using a lower limit (i.e., 0.010 lb/
MMBtu) for baghouses would have changed the PM BART determinations. 
Given that the cost effectiveness for baghouses was generally 
excessively high, we do not expect that using a lower limit would have 
changed the BART determination. Using Jim Bridger Unit 1 as an example, 
an emission limit of 0.015 lb/MMBtu results in an emission reduction of 
709 tpy, while using an emission limit of 0.010 lb/MMBtu results in an 
emission reduction of 829 tpy (calculated in the same manner as in 
Wyoming's BART determination: 6,000 MMBtu/hr heat input and 7,884 hours 
of operation). In this example, the cost effectiveness of the new 
polishing fabric filter was $8,980/ton, and the incremental cost 
effectiveness was $16,396/ton. Given these costs, we have no reason to 
conclude that such a modest difference in the reduction (120 tpy) would 
lead to a meaningful improvement in visibility. This is particularly 
true since, on a per unit mass basis, PM emissions have a lower 
visibility impact than SO2 or NOX. A similar 
conclusion can be drawn for other EGUs where baghouses were considered.
6. Incremental Costs and Visibility
    Comment: Wyoming and EPA have placed undue weight on incremental 
costs and incremental benefits. Wyoming and EPA have essentially based 
their BART and reasonable progress determinations on incremental costs 
and incremental benefits. (In almost every case, Wyoming stated that 
the average cost-effectiveness of the proposed BART technologies for 
NOX are all reasonable.) However, in discussing average and 
incremental costs, EPA BART Guidelines explain: ``The average cost 
(total annual cost/total annual emission reductions) for each may be 
deemed to be reasonable. However, the incremental cost of the 
additional emission reductions to be achieved may be very great. In 
such an instance, it may be inappropriate to choose control B, based on 
its high

[[Page 5169]]

incremental costs, even though its average cost may be considered 
reasonable.'' Although EPA does not explain in its BART Guidelines what 
it considers ``very great'' and ``high'' incremental costs, it goes on 
to provide an example of how incremental cost is calculated, and 
explains: ``The incremental cost of Option 1, then, is $20,000 per ton, 
11 times the average cost of $1,900 per ton.''
    The clear implication of EPA's advice in the BART Guidelines is 
that incremental costs become a deciding factor only if they greatly 
exceed average costs. Instead, EPA has determined that incremental 
costs only twice the ``reasonable'' average costs are excessive. In 
doing so, EPA ignores the established fact that pollution control costs 
increase exponentially with control efficiency, which means that 
incremental costs will always exceed average costs.
    Response: We disagree with most aspects of this comment, but do 
agree with the commenter that EPA has not defined what the terms ``very 
great'' or ``high'' mean when pertaining to incremental costs. We do 
not agree with the commenter that the one of the examples EPA provided 
in 40 CFR part 51, Appendix Y, should be interpreted to mean that 
incremental costs only become a deciding factor if they greatly exceed 
average costs by some magnitude over twice the average costs. In 
addition, incremental costs are to be considered within the context of 
the five factors, including average cost effectiveness and visibility 
improvement. Our BART determinations reflect the statement in the BART 
Guidelines the commenter referenced in that while average cost 
effectiveness may be reasonable, the EPA determined that the high 
incremental costs in some instances made the selection of more 
stringent controls not to be reasonable, when considered with 
visibility improvement. We discuss in each instance our evaluation of 
incremental and average costs and explain our conclusions.
    Comment: Incremental visibility improvement is not mentioned in the 
reasonable progress provisions or BART Guidelines, and EPA cannot 
create a new criterion for the sole purpose of eliminating a control 
option that is reasonably cost-effective and would yield a significant 
visibility improvement. If EPA is going to compare costs and visibility 
benefits, it must do so in a transparent and objective manner, and 
state its criteria for acceptance or rejection of a control strategy. 
Relatively subjective statements about costs being ``high'' or 
visibility improvements ``small'' are not sufficient to justify the 
decisions.
    Response: We disagree with this comment. The RHR states ``When 
making this determination [BART Step 5 on visibility impacts], you have 
flexibility in setting absolute thresholds, target levels of 
improvement, or de minimis levels since the deciview improvement must 
be weighed among the five factors, and you are free to determine the 
weight and significance to be assigned to each factor.'' 70 FR 39170. 
EPA concludes that in exercising its discretion, a state or EPA may 
consider the incremental degree of visibility improvement, which is a 
part of visibility improvement. EPA's consideration of incremental 
visibility improvement in our proposed action in Wyoming is also 
consistent with EPA actions in other states (e.g., Kansas (76 FR 
80754), Nebraska (77 FR 40150), and Oregon (76 FR 38997)). In comparing 
control options and selecting one, it is natural to compare the 
visibility improvement (that is, to compute the incremental visibility 
improvement) for each option.
    Comment: EPA in some cases rejected the best systems of continuous 
emission reduction as BART based on a subjective judgment that the 
incremental costs of concededly superior controls are not warranted by 
the visibility benefits they yield. However, EPA has failed to offer 
any rationale for these cost-benefit determinations, let alone the 
increment threshold applied. As a result, EPA's conclusions are at odds 
with the EPA's own analysis demonstrating that installing the most 
effective controls will yield needed visibility improvements.
    EPA's approach is inconsistent with the purpose of the RHR and the 
five-factor BART analysis. The CAA identifies the elimination of human-
caused visibility impairment in Class I areas as the purpose and 
required outcome of the haze program. 42 U.S.C. 7491(a)(1). Congress 
directed states and the EPA to impose the best system of continuous 
emission reduction on BART-subject sources, and identifies BART as the 
feasible, cost-effective technology that produces the most visibility 
benefits. For NOX emissions at Wyoming EGUs, EPA's source-
specific BART analyses uniformly point to SCR plus combustion controls 
as the appropriate technology.
    To avoid this result, however, EPA puts the technologies that it 
has already determined are feasible and cost-effective through an 
incremental benefit filter in which it assesses not just which control 
technology makes the most visibility improvement, but how much more 
progress it makes over the second best technology relative to their 
costs. EPA applies this additional filter without disclosing what the 
threshold of improvement over the next best technology or the ratio of 
incremental improvement to incremental cost has to be, instead simply 
declaring that ``the cost effectiveness value [of SCR] is significantly 
higher than [inferior technology] and there is a comparatively small 
incremental visibility improvement over the [inferior technology].''
    If haze plans only compel installation of controls with lower 
incremental costs and large incremental benefits (whatever those might 
be), then it may be impossible to reach the goal of attaining natural 
conditions in the Class I areas. This is especially true for Wyoming, 
where there are many large pollutant sources affecting many Class I 
areas. The level of visibility improvement that can be achieved through 
reduction of emissions from any one source might always be deemed too 
small to justify the cost of controls, in which case we will never be 
able to eliminate that last increment of haze pollution because it is 
too small to justify. While EPA or states may argue that additional 
emissions reductions can be achieved in the future, the opportunity to 
reduce haze-causing emissions in initial SIPs/FIPs by requiring BART, 
as recognized and directed by Congress in the CAA, is the best chance 
to make significant progress on this pervasive pollution problem. EPA's 
use of the incremental benefit analysis to eliminate the best 
pollution-reduction systems does not comply with the law.
    Response: We do not agree with this comment. As stated above, EPA 
based its decisions on the BART determinations based on a careful 
weighing of the five factors, including average and incremental cost 
effectiveness. Much like average cost effectiveness, EPA has not 
established a threshold for incremental cost effectiveness as each BART 
determination is an individual decision based on the five BART factors. 
In accordance with the BART Guidelines, for each BART-eligible 
facility, we considered incremental cost effectiveness, and when 
weighed with the other BART factors, we reasonably concluded that more 
or less stringent controls were not warranted.
7. Other Comments on BART
    Comment: The majority of BART sources were constructed between 1962 
and 1977. They have a typical life

[[Page 5170]]

expectancy of 50-60 years. They likely will be retired before 2064 and 
replaced with state-of-the-art power generation technology and 
pollution control equipment. This will be a major factor in achieving 
the 2064 natural background goal for nitrate when these units are 
replaced. Thus, there is no need for controls on these sources now.
    Response: While the goal of the regional haze program is to achieve 
natural visibility conditions in all mandatory Class I Federal areas by 
2064, the statute explicitly calls for a program of reductions over 
time, and incremental reasonable progress towards the long-term goal. 
The requirement for states to implement BART applies during the first 
planning period ending in 2018 and is the first increment of progress. 
Furthermore, the remaining useful life of a facility is one of the five 
factors considered for BART. Thus, for example, if a facility has made 
a federally-enforceable commitment to either shut down or change fuels 
by a date certain, the shortened useful life of the facility is 
incorporated into the cost analysis as part of the amortization of 
total capital costs.
    Comment: Wyoming's SIP is silent with respect to BART emissions 
limits during malfunctions and emergencies. However, EPA proposes a FIP 
requirement that: ``These [BART] emission limitations shall apply at 
all times, including startups, shutdowns, emergencies, and 
malfunctions.'' 77 FR 33061. As EPA has previously noted, EPA's 
proposed FIP requirement for Wyoming is not required by the RHR: 
``Kansas' inclusion of the startup, shutdown, and malfunction 
provisions as exemptions from the BART emission rates are not required 
elements of the regional haze SIPs to be developed and submitted by 
States pursuant to section 169 of the CAA.'' See 76 FR 52604, 52618. 
EPA has also stated that ``EPA's disapproval of the startup, shutdown, 
and malfunction provisions . . . does not trigger an obligation on the 
part of EPA to issue a FIP pursuant to section 110(c) of the CAA, 42 
U.S.C. 7410(c).'' Id.
    Yet, EPA proposes to impose a FIP for startup, shutdown, emergency, 
and malfunction emissions for Wyoming sources despite EPA's prior 
statements that such exemptions do not trigger an EPA obligation to 
issue a FIP. EPA's proposed action for Wyoming is arbitrary, 
unauthorized and unlawful.
    Wyoming does not agree with EPA's proposal to include emergencies 
and malfunctions in 40 CFR 52.2636(c)(2). Permitted emission limits 
should reflect the potential-to-emit (PTE) of a stationary source. The 
PTE refers to a stationary source's maximum capacity to emit under its 
physical and operational design. In estimating a source's PTE, Wyoming 
has consistently only utilized emissions that are anticipated to occur 
on a continuous or regular basis under the source's physical and 
operational design. See United States v. Louisiana-Pacific Corp., 682 
F. Supp. 1141, 1158 (D. Colo. 1988). Emissions that occur outside of a 
source's physical and operational design or, are unplanned, are not 
included in PTE estimates, and are addressed instead in accordance with 
Wyoming's enforcement discretion. The Wyoming Supreme Court recently 
upheld Wyoming's approach. See Sierra Club v. Wyoming Depart of Envtl. 
Quality, 251 P.3d 310, 2011 WY 42 (Wyo. 2011). Therefore, Wyoming 
requests that EPA withdraw its proposed FIP provision addressing 
emergencies and malfunctions.
    Response: We disagree with this comment. The RHR states that 
``Section 302(k) of the CAA requires emissions limits such as BART to 
be met on a continuous basis. Although this provision does not 
necessarily require the use of continuous emissions monitoring, it is 
important that sources employ techniques that ensure compliance on a 
continuous basis.'' 70 FR 39172. The rule goes on to state that 
``[m]onitoring requirements generally applicable to sources, including 
those that are subject to BART, are governed by other regulations.'' 
See, e.g., 40 CFR part 64 (compliance assurance monitoring); 40 CFR 
70.6(a)(3) (periodic monitoring); 40 CFR 70.6(c)(1) (sufficiency 
monitoring) (70 FR 39172). Therefore, it is clear that the rule 
intended for BART emission limits to be met on a continuous basis and 
did not provide either explicitly or implicitly exceptions for startup, 
shutdown, or malfunction. Furthermore, it has been EPA's longstanding 
position that SIP provisions generally cannot contain automatic 
exemptions for startup, shutdown, and malfunction.\220\ With respect to 
PTE, the comment does not identify how the arguments about PTE are 
relevant to a BART emissions limit. Finally, regarding claims of 
inconsistency with the final action for Kansas, commenter quotes from 
the proposed not the final agency action. As explained in the final 
agency action, EPA did not take final action on those portions of the 
Kansas submittal, the state withdrew them.\221\
---------------------------------------------------------------------------

    \220\ See for example, EPA's September 20, 1999, guidance, 
``State Implementation Plans: Policy Regarding Excess Emissions 
during Malfunctions, Startup and Shutdown,'' cited in the next 
footnote.
    \221\ As EPA Region 7 explained in their final action 76 FR 
80754, 80755-6 (Dec. 27, 2011): ``As EPA explained in the proposed 
notice, the Consent Agreements exempted periods of startup and 
shutdown for both Kansas City Power and Light and Westar Energy from 
compliance with applicable emission limits, which were not narrowly 
defined, and exempted periods of malfunction for Westar Energy. EPA 
proposed to disapprove the exemptions because they are inconsistent 
with the Clean Air Act and EPA's September 20, 1999, guidance, 
``State Implementation Plans: Policy Regarding Excess Emissions 
during Malfunctions, Startup and Shutdown.'' Steven Herman, 
Assistant Administrator for Enforcement and Compliance Assurance, 
and Robert Perciasepe, Assistant Administrator for Air and 
Radiation, ``State Implementation Plans (SIPs): Policy Regarding 
Excess Emissions During Malfunctions, Startup, and Shutdown,'' 
September 20, 1999; and 52 FR (45109 November 24, 1987).
    EPA subsequently received a letter from the State dated December 
1, 2011, withdrawing the SSM provisions in the Consent Agreements in 
their entirety from the regional haze SIP. Specifically, the 
following four provisions were withdrawn from EPA's consideration 
for approval in the regional haze SIP:
    1. All references to, ``excluding periods of startup and 
shutdown'' in Paragraph 23 of the Kansas City Power and Light 
Company regional haze agreement;
    2. The reference to, ``excluding periods of startup, shutdown 
and malfunction'' in footnote 1 of Appendix A to the Westar Energy, 
Inc. regional haze agreement;
    3. All references to, ``excluding periods of startup and 
shutdown'' in Chapter 9.3.1 of the Kansas regional haze SIP;
    4. And the sentence, ``The Agreements between KDHE and the 
affected BART sources currently exclude emissions associated with 
startup, shutdowns, and malfunctions (SSM) in the agreed upon 
emission limits'' in Chapter 9.5 of the Kansas regional haze SIP.
    Since the SSM provisions were withdrawn by the State, and are 
therefore no longer before EPA, neither EPA's proposed disapproval 
of these exemptions, nor the comments on that proposed disapproval, 
are relevant to this final action.
---------------------------------------------------------------------------

    Comment: The EPA's proposed FIP states only that subject-to-BART 
sources must comply within five years of adoption of the FIP. This 
blanket schedule of compliance for FIP sources is contrary to the CAA. 
For one thing, by its very language, the EPA's proposed FIP fails to 
ensure that subject-to-BART sources ``procure, install, and operate, as 
expeditiously as practicable'' any additional controls that may 
represent BART as required by the CAA. See 42 U.S.C. 7491(b)(2)(A) and 
(g)(4). The EPA only requires that sources comply within five years, 
but does not actually require sources to comply with BART limits 
established in the FIP ``as expeditiously as practicable.'' Thus, EPA's 
proposed FIP fails to implement the statute. Furthermore, simply 
stating verbatim in the FIP that ``sources shall comply with the 
emission limitations and other requirements of this section within five 
years of the effective date of this rule'' fails to give force and 
effect to the statutory requirements that

[[Page 5171]]

compliance occur as ``expeditiously as practicable.'' Here, the CAA is 
clear that in mandating ``expeditious'' compliance, FIPs must ensure 
that subject-to-BART sources comply as soon as possible. In this case, 
the EPA's proposed FIP simply fails to ensure compliance with BART as 
soon as possible. It lacks any concrete dates by which subject-to-BART 
sources must comply, other than to state that sources must comply 
within the statutory maximum compliance date of five years.
    However, the CAA is clear that if a source can comply with BART 
before five years, it must comply by that earlier date. See 42 U.S.C. 
7491(g)(4). Simply deferring to the five-year deadline undermines the 
Congressional intent behind the ``as expeditiously as practicable'' 
provision. It is notable that the EPA actually required ``expeditious 
compliance'' for Jim Bridger Units 3 and 4.
    Response: We have reviewed the compliance dates for meeting BART 
limits that are contained in the SIP. Given the magnitude of the 
retrofits being undertaken, we believe that five years from the 
effective date of this final rule is as expeditiously as practicable. 
We note that our compliance dates for Jim Bridger Units 3 and 4 are 
based on the fact that those are the dates in the State's SIP which we 
are approving for these two units.
    Comment: Compliance with the perceived dictates of the CAA need not 
be as inflexible as contemplated in the EPA's proposal. By exploring 
and employing creative solutions, it is possible to reduce emissions to 
satisfy the CAA while ensuring reasonable value and more cost-effective 
expenditures for PacifiCorp's ratepayers. Two recent examples of 
successful creative alternatives that will save ratepayers many 
millions of dollars include the recent proposals of (1) Public Service 
of New Mexico for its San Juan Generating Station, and (2) PacifiCorp 
regarding its Naughton Unit 3 in Wyoming.
    Notably, the EPA's revised 2013 proposal for Wyoming implicates ten 
of PacifiCorp's coal-fueled units. Given the number of affected 
PacifiCorp generation plants, Wyoming appears to be a particularly 
fertile ground for encouraging the type of alternative solution that 
satisfied the CAA with regard to San Juan and Naughton Unit 3. We urge 
the EPA, in response to these comments, to signal its willingness to 
consider all feasible compliance options that PacifiCorp may offer 
(including those that the EPA has no authority to order) to provide the 
lowest-cost solution for ratepayers in achieving emissions reductions.
    Response: We agree with the commenter's points that there is some 
flexibility under the CAA to meet the requirements of the RHR. As with 
past actions, EPA is willing to consider alternatives compliance 
proposals that are put forth.
    Comment: Wyoming's regional haze program has been underway for 
several years. Under EPA's RHR, BART controls were expected to be 
installed by the end of 2013. Wyoming appropriately and effectively 
developed and implemented a regional haze program that met the 2013 
timeline. As required by the Wyoming SIP, and with the one exception of 
Naughton Unit 3 which has a deadline of 2014, PacifiCorp has fully 
implemented Wyoming's BART requirements for its Wyoming BART units. As 
a result, in 2013 alone, there will be 76,000 fewer tons of visibility 
impairing pollutants emitted by PacifiCorp BART units than was emitted 
in 2004.
    Had Wyoming waited for EPA's final FIP, none of these reductions 
would have occurred to date. In other words, the Wyoming SIP required 
regional haze reductions to begin earlier and extend over a longer 
period of time than EPA's FIP. It is striking to note that from 2005-
2021 the State's regional haze program will have removed 243,000 tons 
more NOX from PacifiCorp's Wyoming facilities than EPA's 
proposed FIP.
    In 2022, the EPA's FIP begins providing an annual benefit of 5,100 
tons per year. Ironically this benefit only lasts for six years, when 
the units at which EPA's proposed FIP requires more stringent controls 
are retired. By 2027, the Wyoming Regional Haze SIP will have removed 
over 210,000 more tons of NOX from PacifiCorp's units than 
the EPA's proposed FIP, with a significantly lower cost (more than $300 
million less in capital) and will require significantly lower 
expenditures in operation and maintenance between 2022 and 2027.
    Response: We acknowledge that the emission reductions already 
achieved by PacifiCorp's Wyoming facilities are substantial. However, 
the emission reductions already achieved at the PacifiCorp facilities 
do not release EPA from its obligation under the CAA to review 
Wyoming's SIP, or to promulgate a FIP where we find that the SIP fails 
to comply with the CAA or RHR.
    We disagree that the SIP will result in greater emission reductions 
than the FIP. As discussed in section III.B, in response to comments 
received during the public comment period, we have made several changes 
to our proposed BART determinations for the PacifiCorp units. Even so, 
our final rule today continues to achieve greater emission reductions 
than the Wyoming SIP for the PacifiCorp units. For Wyodak, our BART 
determination (new LNBs with OFA and SCR) results in an additional 
2,496 tons per year when compared to the SIP. For Dave Johnston Unit 3, 
though PacifiCorp has the option to shut down the unit in 2027, our 
BART determination (new LNBs with OFA and SCR) results in an additional 
1,597 tons per year when compared to SIP. Clearly, even though we are 
no longer requiring some of the BART controls which we proposed, the 
FIP achieves greater emission reductions than the SIP at any point in 
time. Regardless, the BART determination for any BART source is founded 
on a consideration of the statutory BART factors, and not a comparison 
of overall reductions achieved between a federal and state plan.

D. BART Sources

1. Basin Electric Laramie River Station Units 1-3
a. General Comments
    Comment: We received numerous comments expressing concern over the 
economic impact our proposed FIP would have for customers of Basin 
Electric, and the commenters urged us to approve the State's regional 
haze SIP. The commenters went on to point out that Basin Electric is a 
non-profit electric cooperative that must pass costs on directly to 
consumers. One commenter noted that the projected cost to install SCR 
for each of the three units at Laramie River will be $200 million for 
Western Minnesota and Missouri Energy Services members. If that cost is 
spread over a 10-year period, the cost would be $110 million a year, 
which relates to an increase in electric rates of 8 percent.
    Response: In considering the costs of compliance, the BART 
Guidelines instruct states and EPA to evaluate several metrics, 
focusing specifically on average cost-effectiveness and incremental 
cost-effectiveness, not total capital cost or total annual cost. EPA 
has found that the average and incremental cost-effectiveness of SCR is 
reasonable for all three units at Laramie River. While the BART 
Guidelines suggest that total capital cost and total annual cost, as 
well as incidental increases in prices to consumers, can be considered 
as part of an affordability demonstration, Basin Electric did not 
provide the necessary detailed information to suggest that installing 
SCR at Laramie River would be

[[Page 5172]]

unaffordable, either for the cooperative or its rate payers. 
Consequently, we believe that our analysis of the costs of control, 
which focused on cost-effectiveness, was appropriate.
    Comment: We received numerous comments that EPA's FIP would result 
in additional costs of $600-$700 million for the owners of Laramie 
River Station with no perceptible visibility improvement.
    Response: See the response above. We have addressed the issue of 
perceptible visibility improvement in section V.C.4. As explained in 
the introductory section and elsewhere, the visibility improvements 
from controls at Laramie River are significant, even when considered on 
a unit by unit basis.
b. NOX BART Determination
    Comment: One commenter provided a spreadsheet with cost 
calculations for each of the affected Laramie River units.
    Response: EPA has reviewed the spreadsheet.\222\ The major 
difference in calculations relates to selection of retrofit factor and 
cost of property taxes and insurance (excluded by commenter). Commenter 
indicates that all facilities have a retrofit difficulty of 1.0. EPA 
disagrees and has provided our reasons for retrofit factors in other 
comments. In addition, for certain units where we have incorporated new 
cost information submitted by the facility owner's during the comment 
period, we are no longer applying a retrofit factor. Finally, property 
taxes and insurance costs should be included, but only to the extent 
that they are actually realized. See details in the cost report 
included in the docket.\223\
---------------------------------------------------------------------------

    \222\ Wyoming EGU BART and Reasonable Progress Costs--10/2013.
    \223\ Andover Technology Partners, ``Cost of NOX 
Controls on Wyoming EGUs'', October 28, 2013.
---------------------------------------------------------------------------

    Comment: EPA's decision to change its initial NOX BART 
proposal for Laramie River Station Units 1-3 from SNCR to instead 
propose requiring SCR is well-supported by EPA's analysis. When site-
specific information is appropriately considered, the costs of SCR at 
Laramie River Station are even lower than EPA estimated. EPA used a 
``social interest rate'' of 7 percent in its analysis when the plant's 
owner used an interest rate of only 6 percent; the EPA accepted the 
owner's claimed costs of new LNBs with OFA (after subtracting 
disallowed costs), even though other data submitted by the company 
demonstrated lower costs for these combustion controls; and the EPA's 
cost estimates assumed unreasonably high auxiliary power costs of 
$0.06/kilowatt hour (``kWhr''), when even the owner assumed an 
auxiliary power cost of $0.015/kWhr in its cost-effectiveness analyses. 
Making the suggested cost changes to the analysis would result in cost 
effectiveness values ranging from $3,244/ton to $3,532/ton, as opposed 
to the EPA's values ranging from $3,589/ton to $3,903/ton. The 
substantial visibility benefits afforded by SCR on Laramie River Units 
1-3 also justify a finding that SCR is BART on these units. The costs 
and visibility improvements are consistent with what other states in 
their SIP or EPA in a FIP have found reasonable for BART controls.
    Response: EPA has addressed each of the issues raised by the 
commenter in other responses. EPA has provided revised cost estimates 
based upon input and consideration of all commenters.
    Comment: The Sargent & Lundy Evaluation demonstrates that the costs 
of installing SCR at Laramie River Station are excessive and supports 
Wyoming's determination that OFA plus LNB constitutes BART. In response 
to EPA's proposed SIP disapproval and FIP, Basin Electric requested 
that Sargent & Lundy prepare detailed and site-specific cost estimates 
for installation and operation of SNCR and SCR at the Laramie River 
Station. Sargent & Lundy, ``SNCR and SCR Cost Estimates, Laramie River 
Station'' (August 26, 2013) (S&L Evaluation). Sargent & Lundy is a 
leading engineering, design, and consulting firm and a system supplier 
that has extensive experience with the specification, evaluation, 
selection, and implementation of emission control technologies and 
coal-fired power plants, including more than 98 projects for the 
control of NOX emissions. S&L Evaluation section 2. Indeed, 
Sargent & Lundy has participated in the installation of more than 72 
SCR systems and 26 SNCR systems. Id. The Sargent & Lundy Evaluation 
follows the BART Guidelines and uses the methodology in the CCM where 
possible, while addressing site-specific variables that are critical to 
reaching an accurate cost estimate for these NOX control 
technologies at Laramie River Station.
    The Sargent & Lundy Evaluation estimates that the total capital 
costs of SCR would exceed $746 million, while annual costs of an SCR 
system for the Laramie River units would total more than $86 million. 
S&L Evaluation, Tables 3, 7. Total capital costs for installing SNCR on 
all three units, on the other hand, would be approximately $50.6 
million with annual costs of approximately $20 million. Id. Tables 2, 
6. Moreover, the cost effectiveness of SNCR based on the Sargent & 
Lundy Evaluation would be between $6,967 to $7,013 per ton of 
NOX removed. Cost effectiveness values for SCR range from 
$8,531 per ton of NOX removed to $9,048 per ton of 
NOX removed, with an incremental cost effectiveness compared 
to SNCR of between $9,157 per ton to $9,862 per ton. Id. Table 7. The 
Sargent & Lundy Evaluation demonstrates that the costs of installing 
SCR at Laramie River Station are excessive and supports Wyoming's 
determination that OFA plus LNB constitutes BART.
    Response: As noted on page 21 of Exhibit 14 of Basin Electric's 
comments: ``. . . Cost estimates prepared for Laramie River Station are 
based on equipment costs and budgetary quotes available from similar 
projects and Sargent & Lundy's experience with the design and 
installation of retrofit SNCR and SCR control systems. The Laramie 
River Station cost estimates are conceptual in nature; thus, Sargent 
and Lundy did not procure equipment quotes specifically for the Laramie 
River Station control systems. Rather, equipment costs for the Laramie 
River Station projects are based on conceptual designs developed for 
the control systems, preliminary equipment sizing developed for the 
major pieces of equipment, and recent pricing for similar equipment . . 
.''
    As noted in EPA's response to other comments, EPA has found a 
number of deficiencies in Sargent & Lundy's estimates and disagrees 
with the costs they have arrived at for SCR. Also as described in 
response to other comments, in light of recently submitted information, 
EPA has accepted Basin Electric's estimated capital cost of SNCR.
    Comment: The Sargent & Lundy Evaluation and resulting cost estimate 
is far more accurate than the study level estimate contemplated by the 
CCM and the IPM algorithms relied upon by EPA in its SIP disapproval 
and FIP. The Sargent & Lundy Evaluation takes into consideration site-
specific design and operating parameters and provides a conceptual, or 
scoping-level, estimate for SNCR and SCR at Laramie River. S&L 
Evaluation section 4.5.
    The BART Guidelines state that ``cost estimates should be based on 
the OAQPS Control Cost Manual, where possible'' and in those cases 
where the CCM addresses the control technology in ``sufficient detail 
for a BART analysis.'' 70 FR 39166. In all cases, however, ``[t]he cost 
analysis should also take into account any site-specific design or 
other conditions . . . that affect the cost of a particular BART 
technology option.'' Id. The CCM

[[Page 5173]]

describes various technologies and provides general costing 
methodology, but EPA acknowledges that the methodology is intended to 
provide a ``rough order of magnitude'' estimate of costs that is 
accurate to within 30%. CCM section 1.2, page 1-4. This 
rough estimate is appropriate for regulatory development because it can 
be prepared at a ``relatively low cost with minimum data.'' Id. section 
2.2, page 2-3 (internal quotations omitted). See also S&L Evaluation at 
section 4.5. But ``EPA does not claim cost estimates for industry at 
greater than study level accuracy for industrial users'' because ``the 
industrial user will necessarily have much more detailed information 
than the generic cost and sizing information.'' CCM section 2.2, page 
2-4, 2-5.
    The BART Guidelines may reference the CCM because it provides a 
simple and less costly methodology for estimating costs, but neither 
the Guidelines nor the CCM require use of a less accurate methodology 
where more accurate methodologies are appropriate. Indeed, the BART 
Guidelines require consideration of site-specific variables that in 
some cases, such as with SCR, are not factored into the examples 
provided by the CCM. 70 FR 39166. Under these circumstances, the Manual 
``offers the user an opportunity for greater accuracy than that used by 
regulators'' and gives users the discretion to ``exercise `engineering 
judgment' on those occasions when the procedures need to be modified or 
disregarded.'' CCM section 1.2, page 2-4, section 1.3, page 1-7.
    The cost estimates prepared by Sargent & Lundy are scoping-level 
estimates, which required the use of numerous site-specific design 
parameters that are not included in the general CCM equations and 
reflect all costs to install the control systems, taking into account 
site-specific variables and physical constraints. S&L Evaluation 
section 4.5, Attachments A1, A2. These estimates are far more accurate 
than EPA's estimates, reinforce Wyoming's BART determination, and 
demonstrate there is no basis for EPA to disapprove the State's action.
    Response: We agree that source-specific costs can be useful in the 
BART analysis and agree with a number of the cost estimates in the 
Sargent & Lundy analysis. However, as noted in our response to other 
comments, EPA found Sargent & Lundy'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 described in our responses to other comments, in light of 
recently submitted information, EPA has accepted Basin Electric's 
estimated capital cost of SNCR.
    Comment: The Sargent & Lundy's estimate for SCR considers critical 
site-specific variables that are not captured by the CCM. The Sargent & 
Lundy estimate for SCR is based on an in-depth, detailed study of site-
specific costs conducted by a team of engineers with extensive 
experience in SCR installations. In order to establish SCR control 
system design parameters and to prepare inputs for the capital cost 
estimate, Sargent & Lundy engineers performed a site walkdown to 
identify site constraints for the SCRs and associated plant 
modifications and reviewed operating conditions at Units 1 through 3 
affecting flue gas conditions at the SCR inlet. Sargent & Lundy then 
developed general arrangement drawings for the SCRs and new ductwork, 
on which estimations for material quantities were made. Finally, 
Sargent & Lundy evaluated the existing forced draft (FD) fan buildings 
to determine whether the existing buildings could support the SCR 
structures based on the conceptual design.
    Sargent & Lundy's evaluation identified important design 
considerations affecting the SCR cost estimates, many of them directly 
related to the considerations that EPA acknowledges in the CCM make 
broad brush cost estimates for SCR nearly impossible. CCM section 
2.5.4.2, page 2-28 (installation of SCR can ``impose an addition 
expense to `shoe-horn' the equipment into the right locations'' where 
``there is generally little room for the reactor to fit in the existing 
space and additional ductwork, fans, and flue gas heaters may be needed 
to make the system work properly.''). The considerations include, inter 
alia, congested existing plant configuration, and limited auxiliary 
power available at the station. S&L Evaluation section 4.3.1.
    The site congestion at Laramie River Station substantially 
complicates installation of SCR systems because the location of the FD 
fan buildings limits the open area available to drop support columns 
for the SCR. Id. Based on the site walkdown and review of drawings 
provided by Basin Electric, the conceptual design placed the SCRs 
directly above the existing FD fan buildings, which will require that 
the SCR support columns penetrate the FD fan buildings. This, in turn, 
would require the construction of deep foundations for the SCR support 
columns in a congested area resulting in challenging and time-consuming 
efforts to ensure adequate support. Id. Another related complicating 
factor relates to constructability issues. All three units are 
constructed side-by-side in a row with little space between them, which 
limits crane placement and would require selection of larger, more 
expensive cranes during installation. Id. Another major design 
consideration identified by Sargent & Lundy is the need for entirely 
new auxiliary power equipment for the SCR and replacement induced draft 
(ID) fans because the existing ID fans currently are running at full 
capacity. Id.
    In addition, Sargent & Lundy's Evaluation took into consideration, 
among other factors, the following site-specific conditions that affect 
the cost of SCR at Laramie River Station: (1) Boiler Building 
Reinforcement. SCR duct work will penetrate the existing boiler 
building structural columns, thereby requiring that the boiler building 
structural supports be redesigned and rebuilt and engineered to ensure 
continued support of the 20-story boiler building. (2) SCR Reactors and 
Catalyst. The conceptual design calls for two reactors per unit using 
anhydrous ammonia as the reagent. To achieve required NOX 
emission reductions on a consistent basis, three layers of catalyst 
would be required and the SCRs would need to be designed to hold four 
layers of catalyst. (3) Ammonia System. The conceptual design located 
the anhydrous ammonia system in a remote location from the units and, 
therefore, the cost estimate assumed that all three SCR units would 
share a single ammonia storage facility. (4) Structural Stiffening. 
Structural stiffening of the ductwork and equipment downstream of the 
boiler and upstream of the new ID fans would be required by federal 
regulation to operate at more negative pressures due to installation of 
the SCR.
    Response: EPA has reviewed the submitted comments, and believes 
that for each of the items cited, insufficient information was provided 
to justify why the cost of SCR at Laramie River Station would be so 
much higher than for other SCRs. Commenter cites the location of the 
SCR reactor as an issue. This is an

[[Page 5174]]

issue that is common to every SCR retrofit. Based upon information used 
by EPA and information submitted by Basin Electric, there is no 
indication that location of the SCR reactor will be any more difficult 
than at any other site. In fact, the location is a rather common 
location. There was no indication that major equipment would need to be 
relocated. Therefore, in this respect Laramie River does not appear to 
be any more difficult than a typical SCR retrofit.
    Pertaining to site congestion, Laramie River is no more congested 
than a typical facility that retrofits SCR, and in some respects is 
less congested. There is greater difficulty in retrofitting unit 2 (the 
middle unit), and EPA accounted for that with a higher retrofit 
difficulty factor. Commenter indicates that the SCR support steel will 
interfere with equipment at the ground level, specifically, the FD fan 
buildings, requiring installation of steel and deep foundations in a 
congested area. SCRs are rarely installed at ground level and are 
normally installed above other equipment. SCR support steel is 
therefore commonly installed in this area below the SCR and there 
typically is other equipment that interferes with this. This is not an 
unusual situation and is not a reason for SCR cost at Laramie River 
Station to be higher than a typical retrofit. Side-by-side installation 
is common, and EPA has accounted for that with a higher retrofit 
difficulty factor for unit 2.
    Most boilers are inside a boiler building and SCRs are always built 
outside the boiler building, making it always necessary to route 
ductwork through the boiler building wall or through the roof. Making 
penetrations for SCR ductwork through the boiler building wall is very 
common in SCR retrofits, and this is not a reason to justify a higher 
cost for an SCR retrofit at Laramie River station. SCR reactor 3 plus 1 
and in two sections is a common SCR arrangement for a boiler of this 
size and does not justify a higher than average cost for Laramie River. 
All SCR systems have ammonia storage facilities and typically try to 
combine storage for all units at a site together. Ammonia storage is 
not a major cost item and where the system is located on the site will 
not make a large difference in overall cost. Pertaining to the need for 
an additional fan, the cost estimate used by EPA had a specific line 
item cost for the fan and associated costs for electrical and other 
modifications. Structural stiffening of ductwork is typically required 
when an ID fan is added. The cost estimate by EPA included provision 
for this. Notwithstanding these points, EPA has accepted parts of Basin 
Electric's cost estimate where those costs are supported. See EPA's 
response to other comments for more information.
    Comment: Sargent & Lundy estimated capital costs based on the 
conceptual design of SCR installation at Laramie River Station and in-
depth itemized studies, not the type of generic cost factors set forth 
in the CCM. As a result, the Sargent & Lundy Evaluation provides cost 
estimates that reflect more accurately the actual costs Basin Electric 
would incur for installation of SCR at Laramie River.
    The Sargent & Lundy Evaluation uses an SCR design with itemized 
budgetary cost estimates for major equipment items and site-specific 
costs. S&L Evaluation section 4.5, Attachment A2. For example, the 
estimate includes line-item costs for upgrades, replacements, or 
installations of the following plant subsystems to support SCR 
operation at the Laramie River Station: (1) Foundation work to support 
SCR systems; (2) Economizer ductwork modifications; (3) Larger ID fans 
will be required on all three units, requiring replacement of the 
existing ID fans; (4) Existing electrical systems are not capable of 
handling the new fan loads and SCR control systems and will require 
significant upgrades; (5) Structural stiffening of the duct work 
downstream of the air heater and upstream of the new ID fans; (6) The 
existing Distributed Control System needs to be expanded; (7) Dry 
sorbent injection control systems will be required on Units 1 and 2 for 
SO3 mitigation, resulting from the wet scrubbers installed 
on those units; and (8) Ammonia unloading area construction, including 
two storage tanks and tank equipment, as well as ammonia delivery and 
vaporization equipment.
    Because of the site-specific nature of these items, the assumptions 
in the CCM would not be adequate to account for them, yet these types 
of system upgrades add substantial cost to the SCR installation.
    Sargent & Lundy prepared direct capital cost estimates for each of 
these systems, including all costs associated with equipment, labor, 
and freight. S&L Evaluation section 4.5. As EPA acknowledges, due to 
the site-specific nature of SCR, detailed vendor quotes are difficult 
to obtain because they cannot be done in an ``off-the-shelf'' fashion. 
CCM section 2.5.4.1, page 2-27. Sargent & Lundy, however, has used 
example vendor quotes for major pieces of equipment, including ammonia 
handling system, unity auxiliary transformers, catalyst modules, and 
sootblowers and sonic horns, and adjusted the quotes as necessary to 
account for the site-specific factors such as Laramie River Stations' 
boiler size, flue gas rates, flue gas temperatures, and inlet and 
outlet NOX concentrations. S&L Evaluation section 4.4.1; 
Attachment A2. Sargent & Lundy also provided process equipment cost 
estimates for manufactured equipment in Attachment E of its Evaluation.
    Sargent & Lundy estimated material and commodity costs by 
multiplying the quantity of the material needed to install the system 
based on the conceptual design by the unit cost for the commodity, 
which was estimated using Sargent & Lundy in-house data, vendor 
catalogs, and industry publications. Id. section 4.4.2. The basis for 
the estimates of materials is set forth in detail in the Sargent & 
Lundy Evaluation. Id. section 4.4.2.1. Labor costs were estimated based 
on man-hour estimates from industry publications, union craft rates for 
southeastern Wyoming, and a local labor productivity factor. Id. See 
also id. at Attachment F (Example Industry Publications--Commodity 
Costs and Man-Hour Estimates). Where the conceptual design provided 
insufficient detail on which to generate an estimated cost, Sargent & 
Lundy used allowances based on the typical scope of similar projects. 
Id. section 4.4.3.
    Sargent & Lundy considered both fixed and variable operating and 
maintenance costs. S&L Evaluation section 4.5. Variable costs for SCR 
include costs of anhydrous ammonia and catalyst replacement costs, 
while fixed costs include property taxes and insurance. Id. section 4.5 
and Table 5. Sargent & Lundy's analysis did not use the maintenance 
materials and labor cost in the CCM of 1.5% of Total Capital Investment 
because it results in significantly higher than expected maintenance 
costs than reported by industry. Rather, Sargent & Lundy used a lower 
maintenance materials and labor cost of 0.25% of Total Capital 
Investment, which results in a conservative estimate of operating and 
maintenance costs. Id. section 6.1.2.
    Indirect capital costs were estimated based on total direct capital 
costs using the factors set forth in EPA's CCM. S&L Evaluation section 
4.4.5. For large projects like SCR, with project durations of between 
1.5 years and four years, Sargent & Lundy typically would account for 
escalation, reflecting the increases in equipment, material, and labor 
costs that occur during the duration of the project. Id. section 
4.4.6.1. Sargent & Lundy has taken a

[[Page 5175]]

conservative approach in its estimate, however, and has calculated 
estimates in constant 2013 dollars without including escalation, which 
is consistent with the constant dollar approach discussed in the CCM. 
Id.; CCM section 4.2, page 2-43.
    Response: As noted on page 21 of Exhibit 14 of Basin Electric's 
comments: ``. . . Cost estimates prepared for Laramie River Station are 
based on equipment costs and budgetary quotes available from similar 
projects and Sargent & Lundy's experience with the design and 
installation of retrofit SNCR and SCR control systems. The Laramie 
River Station cost estimates are conceptual in nature; thus, Sargent 
and Lundy did not procure equipment quotes specifically for the Laramie 
River Station control systems. Rather, equipment costs for the Laramie 
River Station projects are based on conceptual designs developed for 
the control systems, preliminary equipment sizing developed for the 
major pieces of equipment, and recent pricing for similar equipment . . 
.''
    The approach used by Sargent & Lundy is essentially how the cost 
algorithms for IPM were developed, upon which EPA relied. Both are 
empirically-based estimates that, as demonstrated in our response to 
other comments, use many of the same inputs. However, Sargent & Lundy 
developed a very detailed cost estimate that includes many line items 
that would otherwise be included in the Project Contingency or other 
areas, such as General Facilities, and thereby double-counted these 
costs.
    EPA has addressed comments relative to capital cost estimates in 
our other responses. EPA has reviewed the assumed variable operating 
costs and has commented on them in other comments. Regarding fixed 
operating costs, the IPM algorithm represents information from actual 
facilities, and is therefore used in EPA's analysis. As far as indirect 
capital costs, EPA agrees that escalation should not be included 
because the CCM requires use of the overnight method.
    Comment: When site-specific conditions are taken into 
consideration, the costs of installing SCR at Laramie River Station 
would total nearly $747 million, with annual costs of $86,074,000. The 
per unit breakdown of capital costs are set forth in Table 5 and Table 
3 of the Sargent & Lundy Evaluation. Direct costs include: Equipment, 
material, labor, spare parts, special tools, consumables, and freight. 
Total project costs include equipment costs for the SCR, ammonia 
handling system, and balance-of-plant systems including the ID fan, 
auxiliary power system, electrical system, and dry sorbent injection 
control systems on Laramie River Units 1 and 2. See S&L Evaluation 
section 4. Indirect costs include: General facilities, engineering and 
home office fees, contingencies, preproduction costs, and initial 
catalyst fills. See Cost Manual section 2.5.2, pages 2-41 through 2-47; 
S&L Evaluation section 4.4.5.
    In all cases, Sargent & Lundy used methodology that results in a 
conservative estimate of total costs taking into account the unique, 
site-specific factors discussed above. The Sargent & Lundy Evaluation 
explains the effect of these factors on the cost estimate, consistent 
with the BART Guidelines and the CCM. S&L Evaluation section 5.2.1; CCM 
section 4.2, page 2-21. In addition, as acknowledged by EPA's own 
consultant, SCR capital costs have risen significantly over the past 
decade. S&L Evaluation section 5.2.1, citing Cichanowicz, Edward J., 
``Current Capital Cost and Cost-Effectiveness of Power Plant Emissions 
Control Technologies,'' (January 2010).
    Response: EPA disagrees with commenter and has identified numerous 
deficiencies in the cost estimate developed by Sargent & Lundy for the 
Laramie River Station that are discussed in previous responses to 
comments. Furthermore, the IPM cost algorithm used had already been 
adjusted to address escalation to 2009 dollars and, per the memo by 
EPA's contractor for this action, the Chemical Engineering Plant Cost 
Index (CEPCI) \224\ was used to escalate costs after that period. Per 
the memo for the IPM algorithm: ``The data sets were escalated to 
update the MOG information to 2009 and all of the data was cross 
referenced with current 2009 projects. The MOG and S&L cost data were 
updated to reflect the changes in equipment and labor rates. The CEPCI 
index for power plants was used to escalate the costs. The Handy-
Whitman index was also used to escalate the project costs to account 
for regional effects; the results were compared with the CEPCI index 
and were within 2% for total project costs.'' \225\
---------------------------------------------------------------------------

    \224\ The CEPCI is an industry index that allows for adjustment 
of plant construction costs from one period to another.
    \225\ Sargent & Lundy, ``IPM Model--Revisions to Cost and 
Performance for APC Technologies SCR Cost Development Methodology 
FINAL'', August 2010, Project 12301-007, Perrin Quarles Associates, 
Inc. p 1.
---------------------------------------------------------------------------

    Comment: The Sargent & Lundy Evaluation considers site-specific 
data, including operating parameters for the Laramie River Station 
units and design parameters for SNCR that were developed based on input 
from Basin Electric and on Sargent & Lundy's extensive experience. S&L 
Evaluation section 4.1, Table 1. See also id. at Attachment A1. Prior 
to undertaking the cost estimate, Sargent & Lundy developed a 
conceptual design taking into consideration site-specific design and 
operating parameters. S&L Evaluation section 4.2.1. The Evaluation 
highlights each of these project-specific considerations, which 
include, among other items: (1) Considerations Related to Use of Urea 
as the Reagent. The SNCR estimate is based on use of urea as a reagent, 
which would be delivered by truck and unloaded into fiberglass 
reinforced plastic storage tanks. The tanks would be cross tied and 
solution would be transferred using stainless steel piping. Centrifugal 
pumps would be needed to pump solution to metering modules and variable 
frequency drives would be used to maintain constant pressure. Finally, 
distribution modules would be needed to provide diluted urea solution 
and atomizing air to individual injectors. The design, quantity, type 
and placement of injectors are critical to SNCR performance. (2) 
Furnace Modifications. Penetrations in the boiler water wall would be 
required at injector locations; and to support injector penetrations, 
water wall tubes would need to be removed and replaced with tubes 
curved around the location. Also, reinforcement may be necessary to 
support the injectors. (3) Process and Freeze Protection Tracing 
System. A freeze protection system would be necessary for outdoor 
piping and instruments. The system would be designed to accommodate 
both normal plant operations and extended shutdowns during cold 
weather. S&L Evaluation section 4.2.2.
    The equipment costs were estimated based on SNCR original equipment 
manufacturers for control systems on similar coal-fired boilers in 
light of the conceptual design of the control technology. Equipment 
costs were developed for SNCR metering skids and injectors, 
compressors, reagent storage tanks, and related ancillary equipment. 
Id. section 4.4.1. Consistent with the SCR cost estimate, material and 
commodity quantities for structural, mechanical and electrical items 
were developed for each subsystem included as part of the SNCR system 
and provided as inputs to the cost calculation. Id. section 4.4.2.1. 
Material and commodity pricing was based on in-house data, vendor 
catalogs, and industry publications. Id. Where the

[[Page 5176]]

conceptual design provided inadequate detail on which to base costs, 
allowances were used. Id. section 4.4.3. Similar to the SCR cost 
estimate, labor costs were based on local labor rates, with an estimate 
of man-hours required for installation of each line item in the SNCR 
estimate. Id. section 4.4.4.
    Indirect capital costs were based on the CCM, using default factors 
set forth in Section 4.2, Chapter 1, Table 1.4. Id. section 4.4.5. The 
estimate includes variable and fixed operating and maintenance costs, 
including costs for urea. Id. section 4.5. Sargent & Lundy did not 
include either escalation or AFUDC in its cost estimate because the 
installation of SNCR systems are less capital and time intensive and 
can be done in a shorter period of time than SCR systems. S&L 
Evaluation section 4.4.6.3.
    Based on methodology consistent with the CCM, total site-specific 
costs for installation of SNCR at Laramie River Station are 
approximately $50.5 million, with annual costs of $19.75 million. The 
breakdown of total costs per unit can be found at Table 2 of the 
Sargent & Lundy Evaluation.
    Response: Based upon information provided by Basin Electric in 
their comments, EPA has accepted Basin Electric's estimated capital 
cost for SNCR and the estimated chemical utilization.
    Comment: The Sargent & Lundy Evaluation supports Wyoming's BART 
determination. The average effectiveness of both SNCR and SCR are high, 
and the incremental cost of SCR compared to SNCR is even higher. 
Sargent & Lundy calculated the cost effectiveness of SNCR and SCR 
technologies based on emission rates resulting from Laramie Rivers' 
installation of LNB and OFA, consistent with the BART Guideline's 
directive to use baseline emission rates that ``represent a realistic 
depiction of anticipated annual emissions for the source.'' 70 FR 
39167. For comparison purposes, Sargent & Lundy also performed a 
sensitivity analysis showing the values generated by using the 2001 
through 2003 emission rates, which EPA's relied upon in making the BART 
determination in its proposed FIP. S&L Evaluation section 6.1.3, Tables 
11-12. The cost effectiveness of SCR still remains between $5,955 and 
$6,298 costs per ton of NOX removed and incremental cost 
effectiveness from SNCR is above $9,000 per ton of NOX 
removed. Id. Table 12. These cost effectiveness values remain 
prohibitive, and reinforce Wyoming's determination that OFA plus LNB 
constitutes BART at Laramie River and EPA's 2012 rejection of SCR as 
not cost effective.
    The Sargent & Lundy Evaluation further supports Wyoming's BART 
determination for Laramie River Station by providing a more precise 
cost estimate for both SCR and SNCR. At significant expense, Basin 
Electric arranged for an evaluation of the costs of compliance with SCR 
and SNCR at a level of detail that far exceeds what the CCM requires, 
but represents the gold standard for estimating the costs of compliance 
for a control technology. This evaluation demonstrates that the costs 
to install SCR at Laramie River would reach $750 million, far above 
what EPA estimates in the proposed SIP disapproval and FIP. In sum, the 
Sargent & Lundy Evaluation supports Wyoming's BART determination, and 
its decision not to require SCR, with a detailed consideration of the 
costs of compliance for Laramie River. Moreover, the Evaluation 
highlights the fundamental inadequacies in EPA's own cost estimates, on 
which EPA bases both its decision to disapprove the SIP and the BART 
determinations in its FIP.
    Response: EPA disagrees with commenter and has identified numerous 
deficiencies in the cost estimate developed by Sargent & Lundy for the 
Laramie River Station that are discussed in previous responses to 
comments. EPA calculated emission reductions from emission rates 
indicative of pre-BART levels in 2001-2003, prior to addition of 
combustion controls for BART. The estimated reduction of NOX 
by SNCR and SCR used baseline levels that were based upon actual 
emission rates achieved after the addition of those combustion 
controls.
    Comment: The EPA's expert, Andover, used aerial photographs to 
assess the structural and mechanical changes necessary for installation 
of the SCR. As witness Ken Snell demonstrated at the July 26, 2013, 
public hearing, however, an aerial photograph is wholly inadequate to 
assess site-specific conditions that affect SCR costs. Those relevant 
site-specific conditions include, among others, the following: (1) Site 
elevation--Laramie River is situated at 4,750 feet above mean sea level 
(MSL), a fact which affects the flue gas volume which require a larger 
SCR reactor, duct work and structural support; (2) Regional labor 
productivity factor--necessary to account for local workforce 
characteristics, labor availability, project location, project 
complexity, local climate and working conditions; (3) Location of 
conveyor rooms--aerial photo cannot reveal conveyor rooms located in 
boiler buildings; (4) Location of FD Fan buildings--aerial photo cannot 
reveal the location of the existing FD fan buildings; (5) Space 
constraints--aerial photo cannot determine the space constraints 
between the FD fan buildings and existing ESPs; (6) Ammonia handling--
aerial photo cannot provide information about where the ammonia 
handling system required for an SCR could be located, or where pipe 
routing could be placed; (7) Ductwork routing--aerial photo does not 
provide information regarding ductwork routing and SCR tie-ins to the 
existing economizers and air heaters; and (8) Subsystems ignored--
aerial photo does not provide any information regarding plant 
subsystems such as ID fan capacity, equipment reinforcement, auxiliary 
power systems, electrical system capacity, or other plant subsystems.
    Failure to take into consideration the site-specific, plant-
specific characteristics for installation of SCR systems necessarily 
leads to a significant underestimation of the costs to install such 
control technology. EPA's expert Andover, using generalized data and an 
aerial photograph, estimated the total capital investment for 
installation of an SCR at $330,000,000. However, when actual site 
conditions are considered--i.e. site elevation, regional productivity 
factors, site congestion, balance-of-plant subsystem upgrades and other 
indirect costs--expert Ken Snell estimated the total capital investment 
at $746,906,000. Failure of EPA's expert to take into consideration the 
Laramie River Station's specific characteristics and plant 
configuration omits approximately $460,000,000 in very real costs. To 
turn a blind eye to site-specific characteristics that have a major 
impact on costs of installation skews the EPA's cost analysis by more 
than 100 percent. Basing the Laramie River Station BART determination 
on EPA's cost estimates would be arbitrary and capricious.
    EPA proposes to reject Wyoming's SIP despite a finding that 
``Wyoming considered all five steps above in its BART determinations'' 
because of alleged ``flaws and deficiencies'' in the cost assumptions 
and methodology, including Wyoming's alleged failure to ``follow the 
methods set forth in the EPA Control Cost Manual.'' 78 FR 34748-34749. 
To address these alleged deficiencies, EPA hired its own consultant, 
Andover Technology Partners (``Andover''), to perform an independent 
cost analysis of installing SNCR and SCR at the Laramie River Station. 
Andover, Review of Estimated Compliance Costs for Wyoming Electricity 
Generating Units (EGUs)

[[Page 5177]]

(Oct. 23, 2012), EPA-R08-OAR-2012-0081 (``Andover Report''); Andover, 
Review of Estimated Compliance Costs for Wyoming Electricity Generating 
Units (EGUs)--revision of previous memo (Feb. 7, 2013), EPA-R08-OAR-
2012-0086 (``Andover Update''). Ironically, it is the cost methodology 
relied upon in the Andover Report that deviates from the BART 
Guidelines and the CCM, thereby making Andover's analysis inconsistent 
with EPA's claim that cost estimates should not deviate from the CCM. 
As a result, EPA's cost estimates are less accurate than the Wyoming 
cost estimates and the Sargent & Lundy Evaluation and do not form a 
legally supportable basis on which to base either a SIP disapproval or 
the promulgation of a FIP.
    Sargent & Lundy, at Basin Electric's request, has provided a 
critique of the Andover Report that highlights the numerous technical 
irregularities in the cost estimate relied upon by EPA for the proposed 
disapproval of the Wyoming SIP and the FIP. Sargent & Lundy Laramie 
River Station Andover Report Comparison (August 26, 2013), Exhibit 16 
(``S&L Critique'') to commenter 0148. The Sargent & Lundy Critique 
demonstrates that EPA's reliance on the Andover Report is not in 
accordance with section 169A of the CAA and the BART Guidelines for 
three reasons: (1) Andover relied primarily on the IPM for cost 
methodology, which is not consistent with the BART Guidelines or the 
CCM methodology and never was intended to be used to develop a site-
specific cost estimate. (See section 2.3, EPA Use of the IPM Cost 
Models.); (2) Andover failed to take into account site-specific 
conditions and resulting balance of plant systems required for SCR and 
SNCR and therefore did not comply with the directive in the BART 
Guidelines that cost estimates ``take into account any site-specific 
design or other conditions . . .'' 70 FR 39166.; (3) Andover ignored 
NOX reductions achieved to date by existing control 
equipment, thereby artificially increasing the cost effectiveness of 
SCR.
    If EPA is basing its SIP disapproval on failure to adhere to the 
methodology set forth in the CCM, EPA's reliance on the cost estimation 
in the Andover Report is wholly inappropriate and imposes an arbitrary 
double standard. Not only does the Andover Report rely on methodology 
that deviates from the CCM, but EPA's approach is inconsistent with the 
requirements of the BART Guidelines because it does not adhere to the 
three-step approach for cost estimation set forth in the Guidelines and 
fails to appropriately account for ``site-specific design or other 
conditions'' that ``affect the cost of a particular BART technology 
option.'' 70 FR 39166. Reliance on the Andover Report for disapproval 
of the Wyoming SIP or imposition of a FIP would constitute arbitrary 
and capricious decision making and would run contrary to the very 
provisions of law on which EPA proposes to base its decision.
    Response: EPA disagrees with commenter and has identified numerous 
deficiencies in the cost estimate developed by Sargent & Lundy for the 
Laramie River Station that are discussed in previous responses to 
comments. EPA has also noted in other responses to comments that, 
except for elevation, each of the site-specific issues raised by 
commenter has been addressed, or commenter has not provided adequate 
information to support their assertion that there are unique costs that 
are not accounted for in EPA's cost estimate. We disagree with the 
characterization of the cost development methodology contained in IPM 
as inconsistent with BART guidelines. As noted in the documentation for 
IPM's cost development methodology for SCR, the cost estimate 
methodology is based upon two databases of actual SCR projects.\226\ 
These databases include 2004 and 2006 industry cost estimates prepared 
for the Midwestern Ozone Group, and a proprietary in-house database 
maintained by engineering firm Sargent & Lundy. The Midwestern Ozone 
Group information was cross-referenced with actual 2009 projects, and 
as explained elsewhere in this document, escalated accordingly. Sargent 
& Lundy then used the information in these databases to develop the 
equations described in the cost component taking into account the pre-
control NOX emission level, degree of reduction, coal type, 
facility size, and numerous other unit-specific factors. While a costly 
engineering evaluation that included site visits in addition to our use 
of satellite imagery might be useful, we disagree that our approach is 
not sufficiently site specific to satisfy BART guidelines. As noted by 
EPA in previous responses, EPA's use of satellite imagery enabled us to 
evaluate each of the major site-specific issues.
---------------------------------------------------------------------------

    \226\ http://www.epa.gov/airmarkets/progsregs/epa-ipm/docs/v410/Appendix52A.pdf.
---------------------------------------------------------------------------

    Comment: EPA argues that ``[w]hen considering the cost 
effectiveness and visibility improvement of new LNBs plus OFA and SCR, 
it is within the range of what EPA has found reasonable for BART in 
other SIP and FIP actions.'' 78 FR 34776. EPA's cost effectiveness and 
visibility improvement numbers for Laramie River Station Units 1-3 are 
within the range of what EPA has found not to be reasonable for BART. 
The case for rejecting SCR becomes even more compelling when EPA's 
numbers are corrected to comply with the BART Guidelines, CCM, and EPA 
guidance, and to reflect site specific conditions. With these comments, 
Basin Electric is submitting updated and more accurate reports with 
cost estimates and visibility modeling results based on inputs that are 
more correct and consistent with EPA's BART Guidelines. Average and 
incremental cost effectiveness values for SCR at Laramie River Station 
in these reports are far higher than assumed by EPA, and visibility 
improvement associated with SCR is far lower than EPA assumed. For EPA 
to disapprove the State's BART determination for Laramie River Station 
and proceed with its FIP in light of this new information would be 
egregiously inconsistent with BART actions it has taken for other 
sources.
    The following discussion explains that even with EPA's cost and 
visibility values for Laramie River Station, its proposed action at 
Laramie River Station is inconsistent with actions elsewhere and EPA 
should withdraw its proposed disapproval.
    The comparison to Gerald Gentleman Station (GGS) Units 1-2 is 
striking. SCR was rejected at GGS despite substantially lower costs and 
very similar visibility improvement. GGS is a valid point of comparison 
despite the fact that Nebraska adopted the Transport Rule as a BART 
alternative. 77 FR 40159. EPA did not make a final determination as to 
whether to select SCR as BART for GGS because Nebraska became subject 
to the Transport Rule and relied on that as a BART alternative. Id. 
However, EPA's proposed rule discusses the costs and benefits of SCR. 
77 FR 12770, 12779 (March 2, 2012). In its proposed rule, EPA agrees 
with Nebraska's decision to reject SCR at an average cost effectiveness 
of $2,297/ton and an incremental cost effectiveness of $5,445/ton (both 
as calculated by Nebraska). The projections of visibility improvement 
were the same in the proposal and the final rule, i.e. 0.62 delta 
deciview for each of GGS Units 1 and 2. The proposed rule states that 
``EPA agrees that the State's NOX BART determination for GGS 
is reasonable.'' 77 FR 12779. EPA never retracted that conclusion.
    The difference between EPA's pending proposals for Wyodak Unit 1 
and Laramie River Station Units 1-3 are equally striking. The average 
and incremental cost effectiveness is virtually the same for these 
units. So are

[[Page 5178]]

the baseline and incremental visibility improvement figures. Yet EPA 
proposes to require LNB, OFA and SNCR at Wyodak Unit 1, as opposed to 
LNB, OFA and SCR at each Laramie River Station unit. 78 FR 34785. EPA 
bases these differing outcomes on the fact that SCR at Wyodak Unit 1 
would achieve a cumulative visibility improvement of 1.16 deciviews, 
whereas SCR at Laramie River Station Units 1-3 would reportedly achieve 
cumulative visibility improvements of 2.12, 1.97, and 2.29. Id. 
Cumulative visibility improvement at multiple Class I areas is not a 
valid criterion for use in BART determinations. Disregarding the 
invalid cumulative criterion, EPA inconsistently eliminated SCR as BART 
at Wyodak based on cost and visibility values very similar to EPA's 
cost and visibility values for Laramie River Station.
    Also noteworthy is EPA's decision to reject SCR as BART at Healy 
Unit 1 because of its $5,300/ton cost effectiveness, 0.786 deciview 
visibility improvement from the LNB/OFA baseline, and 0.17 deciview 
incremental improvement compared to SNCR. EPA recalculated the costs 
after publishing its proposed rule to account for various potential 
useful life scenarios. The $5,300/ton figure shown here is for a 30 
year life. For a 20 year useful life, SCR would cost $5,900/ton. EPA 
concluded that these costs are ``not justified'' given the visibility 
improvement (which was not recalculated after proposal).
    A comparison of the costs and visibility impacts of installing and 
operating SCR at Laramie River Station Units 1-3 to the costs and 
visibility impacts of SCR at the facilities listed above quickly shows 
that SCR must be rejected as BART. EPA has made no attempt to explain 
why it proposes to disapprove the State's BART for Laramie River 
Station and proposes SCR instead, when it has eliminated SCR at other 
facilities based on similar information. To the extent the information 
at the other facilities is a little different than at Laramie River, 
EPA has not and cannot show that the difference is significant or a 
reason to treat the facilities differently.
    Response: We disagree with the commenter's assertion that our 
proposed action for Laramie River Station, as it relates to the 
consideration of SCR as BART, was inconsistent with our proposed action 
for other BART sources in Wyoming or with EPA actions in other States.
    Regarding NOX BART for Gerald Gentlemen Station in 
Nebraska, we note that our proposed approval of the State's 
NOX BART determination, as described by the commenter, does 
not reflect final agency action. In our final rulemaking for Nebraska, 
where the State is subject to the Transport Rule and FIP for 
NOX, we provided the following:

    Given the emission reductions provided by the NOX 
emission limits associated with Nebraska's NOX BART 
determination of LNB and OFA for GGS Units 1 and 2, which strengthen 
the Nebraska SIP, in conjunction with the existing Transport Rule 
FIP which already applies to Nebraska and has been determined to 
provide greater reasonable progress than BART, in today's action, 
EPA is finalizing its proposed approval of Nebraska's SIP as 
satisfying the requirements of the Regional Haze Rule with respect 
to BART for NOX, and therefore do not inquire further 
here as to whether the cost effectiveness of SCR is low enough and 
the associated deciview improvement significant enough to reasonably 
determine that SCR is BART for GGS Units 1 and 2.\227\
---------------------------------------------------------------------------

    \227\ 77 FR 40159 (emphasis added).

Therefore, because the Transport Rule removed the need for EPA to 
consider SCR for Gerald Gentlemen Station any further, there is no 
factual basis to determine whether our consideration of SCR in Wyoming 
differs from that in Nebraska. In simpler terms, the commenter has 
erred by drawing a comparison with a BART determination that was never 
finalized.
    We also disagree that our proposed NOX BART 
determinations for Laramie River Station were inconsistent with that 
for Healy Unit 1 in Alaska. There, the cost effectiveness of SCR, using 
a 20 year lifetime comparable to that used for Laramie River Station, 
was found to be $5,900/ton.\228\ This cost effectiveness is greater 
than that for any of the Wyoming units for which EPA proposed SCR as 
BART, or for which EPA is finalizing SCR as BART today. The cost 
effectiveness of SCR (with combustion controls) for Laramie River 
Station units estimated by EPA in our proposed rule ranged from $3,589/
ton to $3,903/ton \229\--at least 34% less than at Healy Unit 1. While 
we have revised these costs effectiveness estimates for today's rule, 
they remain well below the $5,900/ton cost effectiveness that EPA 
calculated for Healy Unit 1.
---------------------------------------------------------------------------

    \228\ 78 FR 10548.
    \229\ 78 FR 34775-34776.
---------------------------------------------------------------------------

    Finally, while the costs and visibility improvement (at one Class I 
area) for the Laramie River units and Wyodak described in our proposed 
rule may have been similar, we disagree that the cumulative visibility 
benefit was not a valid criterion for use in BART determinations. Refer 
to the modeling section above where we address our consideration of 
cumulative visibility benefits.
    We have addressed the updated cost estimates and visibility 
modeling submitted by Basin Electric during the public comment period 
in other response to comments.
    Comment: SCR should be rejected for Laramie River Station Units 1-3 
because of its high cost per deciview of visibility improvement. EPA 
did not report the $/deciview for Laramie River Station, but dividing 
the annualized costs by the visibility improvement (from the pre-LNB/
OFA baseline) reveals that SCR would cost between $23.0 million and 
$27.8 million per deciview of improvement in the Class I area with the 
greatest visibility improvement. These figures are very similar to the 
$/deciview numbers seen at Martin Drake, Colstrip, Corette, and GGS, 
and substantially higher than the $10.8 million and $20 million figures 
reported for Lakeland Electric and JEA Northside. The Laramie River 
figures are also well above the $14 million to $18 million per deciview 
range that has frequently been seen as cost effective. EPA has not 
considered or justified SCR at Laramie River in light of its high cost 
per deciview.
    Not only has EPA proposed to mandate SCR for Laramie River at $/
deciview levels where SCR has previously been rejected, EPA has 
frequently refused to apply the $/deciview metric in accordance with 
the BART Guidelines. The Guidelines expressly allow states to evaluate 
control technologies based on ``cost-effectiveness measures (such as $/
deciview),'' 70 FR 39170, but EPA has generally opposed reliance on 
such a standard by the states. While acknowledging that the dollar per 
deciview is ``an additional cost effectiveness metric that can be 
employed along with $/ton for use in a BART evaluation,'' EPA has not 
used $/deciview because it is ``unnecessary,'' it ``complicates the 
BART analysis,'' and it is ``difficult to judge.'' 77 FR 57871. In 
other rulemakings, EPA has stated that ``[w]e do not generally 
recommend the use of this metric as it can be complicated to use and 
the results can be difficult to assess.'' 77 FR 76871, 76873 (Dec. 31, 
2012). EPA also has objected to the $/deciview metric because it claims 
that metric is based on the impacts of a single day. 77 FR 57871.
    EPA's reasons for refusing to consider the $/deciview metric are 
frivolous. Even if the $/deciview metric is complicated or difficult to 
use, that does not distinguish it from any other aspect of the BART 
determination process. EPA's concern that the $/deciview metric is 
based on a single day holds no

[[Page 5179]]

water. EPA sometimes bases the visibility improvement (delta deciview) 
of potential control technologies on the ``maximum 98th percentile 
impact,'' meaning the 98th percentile day with the highest deciview 
improvement during the relevant period, and proposes to do so in its 
analysis of Laramie River. 78 FR 34775. Indeed, it is that day's 
deciview improvement which is used to calculate $/deciview. The $/
deciview metric has the virtue of directly comparing cost to visibility 
improvement. It would constrain EPA's current use of a nebulous sliding 
scale where the agency gives itself the latitude to point to any one of 
several variables as justification for overriding the State's choice of 
BART or for making its own.
    Response: We disagree that our reasons for not considering the $/
deciview metric are frivolous. We maintain that, for the reasons 
discussed in other regional haze actions, as cited by the commenter, 
the $/deciview metric is problematic and does not offer any better 
basis for making BART determinations than those used by EPA here--cost 
effectiveness, incremental cost effectiveness, and visibility 
improvement. Moreover, the BART Guidelines do not require EPA or the 
states to conduct a $/deciview analysis when evaluating the visibility 
improvement factor. Instead, the BART Guidelines allow flexibility in 
this area, stating that: ``You have flexibility to assess visibility 
improvements due to BART controls by one or more methods. You may 
consider the frequency, magnitude, and duration components of 
impairment.'' 70 FR 39170. While the BART Guidelines suggest cost per 
deciview as a possible parameter for consideration, its use is entirely 
discretionary. There are numerous examples of BART analyses conducted 
by states and EPA that have not calculated this metric.
    Comment: The Laramie River Station began commercial operation in 
July, 1980, with a permitted limit for NOX emissions of 0.71 
lb/MMBtu. However, Laramie River was able to significantly outperform 
its permitted limits, achieving an average emission rate that was much 
lower, approximately 0.45 lb/MMBtu. In 1996 and 1997, Laramie River 
replaced burner nozzles on all three units and again reduced its 
NOX emission rates, to an average of about 0.27 lb/MMBtu. 
Now, pursuant to Wyoming's BART permit, Laramie River is required to 
further reduce its NOX emissions to a limit of 0.21 lb/MMBtu 
and 14,474 tons/year in 2014, and reduce emissions even further by the 
end of 2017, to 12,773 tons/year (equivalent to 0.158 lb/MMBtu). By 
2017, the Wyoming regional haze SIP will have required the station to 
reduce its NOX emission rate by 65% from the NOX 
rate emitted when the units were originally started up. This 
demonstrates that the State has achieved very substantial 
NOX emission reductions without undue and wasteful expense.
    To put the reductions already achieved in perspective, the 
combination of past reductions and future required reductions results 
in total NOX reductions at Laramie River of 0.29 lb/MMBtu 
(from 0.45 lb/MMBtu to 0.158 lb/MMBtu). This has been done and will be 
done at a significant but reasonable cost. In contrast, EPA proposes to 
require the expenditure of nearly $750 million dollars to reduce 
NOX emissions further, from 0.158 1b/MMBtu to 0.05 lb/MMBtu, 
a reduction of only 0.11 lb/MMBtu, less than half of what has already 
been accomplished.
    Response: We disagree with this comment. The RHR and BART 
Guidelines instruct states to calculate the cost-effectiveness and 
visibility improvement associated with the various control options 
against a realistic emissions baseline. For the purposes of BART, most 
states, including Wyoming, used a baseline period of 2000-2004, which 
corresponds to the five-year period that followed the promulgation of 
the RHR. Setting a baseline that predates the promulgation of the RHR, 
as the commenter suggests, would be inappropriate because it would 
allow emission reductions that were achieved as the result of 
compliance with other CAA programs to be attributed instead to BART. 
Thus, any reductions achieved at Laramie River between 1980 and 1997 
cannot be credited to the source owner, but must be incorporated into 
the baseline, as both the State and EPA properly did in this case. In 
regards to reductions achieved at Laramie River that have been or will 
be achieved due to compliance with the State's BART determination, 
these reductions will also occur under EPA's FIP. Consequently, a more 
accurate way of comparing the State's regional haze SIP to EPA's FIP is 
to subtract the ultimate emission rate achieved by each plan from the 
baseline. Using the commenter's emission rates, the State's regional 
haze SIP would reduce emissions at Laramie River by 0.112 lb/MMbtu from 
the baseline, while EPA's FIP will reduce emissions by 0.22 lb/MMbtu. 
For a more detailed discussion of baseline emissions and the flaws in 
the commenters' logic, see our response to similar comments on the 
consideration of existing controls in use at a source.
    Comment: The costs of installation of the SCRs at Laramie River 
Station impose an economic impact that is unjustified by the facts of 
this case. The regulation of regional haze is focused on improving 
visibility, not public health. Yet, the improvement in visibility that 
EPA suggests will be created by installation of SCRs, as opposed to the 
Wyoming SIP's LNBs/OFA, does not carry with it a significant 
improvement in visibility. By EPA's own calculations, installation of 
SCRs will result in only a 0.79 deciview visibility improvement at the 
most impacted Class I area, Badlands National Park, and those 
calculations substantially overstate the visibility improvement that 
would actually be achieved. By its very definition, this small 
incremental improvement in visibility is not even perceptible by the 
human eye.
    Response: We disagree with this comment. As stated in section 
IV.C.5 above, 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.\230\
---------------------------------------------------------------------------

    \230\ The preamble to the BART Guidelines state, ``[e]ven 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 39104, 39129, July 6, 2005).
---------------------------------------------------------------------------

    Comment: EPA asserts that Wyoming's alleged ``deviations'' from the 
BART Guidelines and CCM form adequate grounds for rejection of its 
SIP--yet the BART determinations included in EPA's proposed FIP eschew 
the very standards to which it holds the State. EPA's estimate of the 
``cost of compliance'' for installation of SCR and SNCR at Laramie 
River Station is grounded in an outside consultant's report that 
expressly dismisses the recommendations of EPA's own CCM in favor of a 
methodology that is inconsistent with the directives of the BART 
Guidelines and fails to account

[[Page 5180]]

for critical site-specific factors that affect the cost of these 
technologies at Laramie River. The resulting cost estimates are not 
representative of the costs that Basin Electric would incur for the 
installation of SCR and SNCR and do not form a legally supportable 
basis on which to promulgate a FIP. To finalize a FIP based on the cost 
effectiveness estimates in the Andover Report would constitute 
arbitrary and capricious decision making and would run contrary to the 
same provisions of law on which EPA bases its disapproval of Wyoming's 
SIP.
    The Andover Report uses a high-level model that was never intended 
to be used to estimate site-specific costs. But neither Andover nor EPA 
offers any explanation of why reliance on the IPM model is more 
appropriate than either the CCM recommendations or a site-specific 
scoping level study such as the Sargent & Lundy Evaluation. Indeed, 
EPA's preamble notes that the cost estimate relied primarily on the IPM 
model, but then simply parrots the conclusions of the Andover Report 
without further analysis or discussion. Although EPA has the discretion 
to rely on a model of its choice, EPA's reliance on the IPM model to 
estimate costs requires both an explanation of the assumptions made and 
a defense of this particular methodology--particularly because EPA has 
proposed disapproval of Wyoming's SIP for failure to adhere strictly to 
the CCM methodology. See Appalachian Power, 249 F.3d at 1053. EPA has 
failed to ``make plain its course of inquiry, its analysis and its 
reasoning'' and therefore promulgation of a FIP based on the Andover 
Report would be arbitrary and capricious. Olenhouse, 42 F.3d at 1575.
    EPA's BART determinations relied upon methodology that does not 
comply with the BART Guidelines. Specifically, Andover did not adhere 
to the three-step process in the BART Guidelines for estimating the 
``costs of compliance'' because the analysis set forth in the report 
did not adequately define the emission units being controlled, failed 
to identify site-specific design parameters that affect cost and 
performance of the controls, and used the IPM model to develop cost 
estimates that are neither technically defensible nor representative of 
the costs of SCR and SNCR systems at Laramie River.
    First, Andover's use of the IPM model--which requires only four 
inputs--is so general that it failed to adequately define the operating 
parameters of Laramie River Station Units 1 through 3. Second, both the 
IPM model and Andover's manipulation of the cost algorithms ignored 
numerous site-specific variables that would have a substantial effect 
on the costs incurred by Basin Electric. These variables include, among 
other items, failure to account for the elevation of Laramie River and 
the complexities of SCR at the facility. Finally, Andover used out-of-
date and inaccurate emissions from Laramie River Units 1 through 3 on 
which to base its cost effectiveness analysis. These emissions 
estimates did not take into consideration the reductions that result 
from the installation of OFA and LNB at these units, which have reduced 
the baseline NOX emissions to 0.19 lb/MMBtu. As a result of 
these inadequacies, the Andover Report makes inaccurate cost estimates 
that are not representative of the costs that Basin Electric would 
incur for installation of either SNCR or SCR. By relying on the Andover 
Report, EPA has ``complete[ly] failed to consider the criteria that 
should inform'' its BART determination, and a court would accord EPA's 
BART determination no deference and would determine that it is 
arbitrary, capricious, an abuse of discretion and not in accordance 
with the law. Nat. Resources Defense Council, 725 F.2d at 771; see also 
Appalachian Power, 249 F.3d at 1052; Sierra Club v. Costle, 657 F.3d at 
333.
    Response: Each of the commenter's claims have been addressed above 
in other responses and elsewhere in this document. In these responses 
we have substantiated that the cost methodology employed by EPA, 
including use of the IPM-based cost algorithms, is consistent with the 
BART Guidelines and CCM. Moreover, we modified our cost estimates in 
response to site-specific information provided by Basin Electric during 
the comment period. Therefore, we reject the commenter's assertions 
that (1) we have dismissed the recommendations of the CCM in favor of a 
methodology that is inconsistent with the directives of the BART 
Guidelines, and (2) failed to account for critical site-specific 
factors.
    We have addressed the commenter's concern regarding whether our 
cost-effectiveness analysis reflects the relatively lower emissions 
achieved with recent combustion control updates (OFA and LNB) in 
section V.A.12 above.
    Comment: Wyoming concluded that SCR would lower the NOX 
emission rate of Laramie River Units 1-3 to 0.07 lbs/MBtu on a 30-day 
rolling average basis and used the 0.07 lbs/MMBtu controlled 
NOX rate to estimate costs. 78 FR 34748; WDEQ Revised NOX 
BART Impact Analysis AP-6047A (January 3, 2011) (``2011 Revised BART 
Analysis'') at 3, Table 2-2, docket cite EPAR08-OAR-2012-0026-0003.
    The State's administrative record supports its selection of 0.07 
lb/MMbtu on a 30-day rolling average as an appropriate post-SCR 
NOX emissions rate. Wyoming first presented this emissions 
rate in its BART Application Analysis AP-6047. Wyoming BART Analysis at 
8, Table 2. Wyoming explained its rationale in its December 31, 2009 
response to comments on BART Permit AP-6047. Available at EPA-R08-OAR-
2012-0026-0058, Exh. 3.
    Wyoming then provided a table comparing SCR control efficiencies at 
seven similar coal fired EGUs. The NOX emission rate 
selected by Wyoming is squarely within the range of control 
efficiencies identified by the State's search. Of note are the 
NOX emission rates for the Iatan Station (0.08 lb/MMbtu, 30-
day average), Big Cajun II Power Plant (0.07 lb/MMBtu annual average), 
and OPPD--Nebraska City Station (0.07 lb/MMbtu, 30-day average). Id. 
Wyoming's explanation and supporting data negate any contention that 
the State violated the CAA or acted unreasonably when it chose to 
evaluate SCR using a NOX control efficiency of 0.07 lb/MMbtu 
on a 30-day rolling average basis.
    EPA contends that whereas Wyoming assumed that adding SCR controls 
at Laramie River would achieve a control effectiveness of 0.07 lb/
MMBtu, ``EPA has determined that on an annual basis SCR can achieve 
emission rates of 0.05 lb/MMBtu or lower.'' 78 FR 34748. EPA provides 
no explanation and cites nothing to support how it ``determined'' this 
to be the case. It doesn't associate the 0.05 lb/MMBtu with any 
specific facility or unit--it just makes this blanket assertion.
    Nor can EPA assert that Wyoming's 0.07 lb/MMBtu assumed control 
level conflicts with the CAA or the BART Guidelines. Nothing in the 
Guidelines dictates what SCR can achieve, and EPA cites no provision of 
the Guidelines to support its claim. In fact, EPA's blanket claim that 
0.05 lb/MMBtu must always be used itself conflicts with the Guidelines, 
which make clear that BART is a site-specific determination, not a 
blanket finding. The notion that EPA can apply an across-the-board 
value and thereby deprive the State of its ability to exercise 
discretion on an individual case basis is contrary to the holding in 
American Corn Growers, 291 F.3d at 7-10.
    The lack of support for EPA's claim that the 0.07 lb/MMBtu is a 
``flaw'' is reinforced by EPA's own acceptance of this value in BART 
analyses by other States. For example, in Colorado, commenters on EPA's 
proposed

[[Page 5181]]

approval of the State's regional haze SIP, including BART 
determinations, argued that the State was wrong in assuming that at 
Tri-State's Craig Station Units 1 and 2, SCR would achieve only a 0.07 
lb/MMBtu NOX emission rate on an annual basis; however, EPA 
did not disapprove Colorado's BART determinations for this reason or 
find that this was an error. EPA's response to the comments stated that 
it agreed that ``SCR in some cases can achieve annual NOX 
emission rates as low as 0.05 lb/MMBtu'' but that the ``annual emission 
rate assumed by Colorado, 0.07 lb/MMBtu, is within the range of actual 
emission rates demonstrated at similar facilities in EPA's Clean Air 
Markets Division (CAMD) emission database.'' 77 FR 76871, 76873. 
Similarly, although commenters argued that SCR at Alaska's Healy Unit 1 
could achieve a NOX emission rate of 0.035 lb/MMBtu, EPA 
evaluated using a rate of 0.07 lb/MMBtu to evaluate SCR. 78 FR 10546, 
10548. EPA evaluated SCR using an emission rate of 0.06 lb/MMBtu for 
Nevada's Reid Gardner Generating Station. 77 FR 21896, 21903 (calling 
this a ``mid-range option'').
    Also, EPA accepted an even higher post-SCR NOX emission 
rate of 0.10 lb/MMBtu for Jeffrey Energy Center Units 1 and 2. Kansas 
SIP Approval, 76 FR 80754, 80756. This emission rate was ``within the 
range of effectiveness that the State believed to be reasonable as a 
retrofit control on older tangential-fired units.'' Id. EPA deferred to 
Kansas, noting that ``EPA believes the State's decision to choose a 
control efficiency within the middle of the range for the purpose of 
estimating cost is a reasonable approach and is acceptable according to 
the BART Guidelines.'' Id. If it was not error and not unreasonable or 
arbitrary for Colorado and Kansas to use 0.07 or 0.10 lb/MMBtu, it 
cannot be error, or unreasonable, or arbitrary, for Wyoming to use the 
same or lower value.
    Response: We have addressed the control effectiveness of SCR above 
in section V.C.3 above. Again, we agree that it was appropriate for 
Wyoming to set the 30-day rolling average emission limit for SCR 
installations at 0.07 lb/MMBtu. And again, EPA's use of an actual 
annual emission rate of 0.05 lb/MMBtu for cost calculation purposes is 
supported by information supplied by control equipment consultants or 
vendors and submitted along with comments from PacifiCorp and Basin 
Electric.
    We also note that the commenter has compared the 30-day allowable 
rates established at certain facilities to the annual emission rate 
used by EPA to calculate cost effectiveness. These values are not 
directly comparable. This is because: (1) The former is on a 30-day 
basis, while the latter is on an annual basis, and (2) the former is an 
allowable emission limit, while the latter is an actual emission rate.
    The remaining comments have been addressed elsewhere in this 
document.
    Comment: For the same reasons that SNCR is not a relevant basis for 
disapproving Wyoming's BART for Laramie River Station, the price of 
urea is likewise not relevant. The price of urea relates only to SNCR 
technology, not to SCR. As noted above, SNCR is not a relevant factor 
to support EPA's rejection of the State's NOX BART for 
Laramie River Station because neither the State's SIP nor EPA's 
proposed FIP chooses SNCR as BART. SNCR has been taken off the table by 
EPA, so the attempt to base its BART disapproval on SNCR issues is 
specious. EPA cannot reasonably base its disapproval on the State's 
alleged failure to properly consider the cost of a technology that EPA 
itself rejects.
    Response: We disagree. 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. And if, as EPA has established in other 
responses, the analysis of one of those options, such as SNCR, was 
flawed, then the State could not sensibly identify the best available 
option among all of the control options considered. Therefore, 
regardless of whether the State or Wyoming rejected SNCR as BART, it is 
mistaken for the commenter to suggest that the analysis of SNCR was 
somehow immaterial in the selection of BART.
    It is particularly important that the costs of SNCR be properly 
estimated in relation to the calculation of incremental cost 
effectiveness. (The incremental cost of effectiveness should be 
calculated in addition to the average cost effectiveness. 70 FR 39167). 
The cost of SNCR affects the incremental cost effectiveness between SCR 
and SNCR, as well as incremental cost effectiveness between SNCR and 
combustion controls. If the cost of SNCR is incorrect, the incremental 
cost effectiveness between control options will also be incorrect. This 
underscores the point that, if the underlying assumptions were flawed, 
the State could not have reasonably chosen between competing control 
options.
    We have addressed the price of urea in a separate response.
    Comment: EPA is proposing that the FIP NOX BART emission 
limit for Basin Electric Laramie River Unit 1, Unit 2, and Unit 3 is 
0.07 lb/MMBtu (30-day rolling average). While we are generally pleased 
with EPA's proposal, we note that EPA's analysis is based on only 74% 
NOX control by the SCRs, and still results in each EGU 
contributing 0.5 deciview to visibility impairment at Badlands National 
Park.
    Based on an evaluation of the rolling 30-boiler operating day 
average NOX emission rates from Laramie River Units 1-3 with 
emissions data available in EPA's CAMD database, a NOX limit 
of 0.07 lb/MMBtu on a rolling 30-boiler operating day basis would only 
require Laramie River Units 1-3 to achieve 61-70 percent NOX 
removal across the SCR systems. A 0.05 lb/MMBtu NOX emission 
limit applicable on a rolling 30-boiler operating day average basis 
would only require 73-79 percent NOX removal across the SCR, 
which is readily achievable.
    Response: We have addressed the control effectiveness of SCR above 
in section V.C.3 above.
    Comment: Moreover, the Sargent & Lundy evaluation demonstrates, 
based on a detailed scoping-level cost analysis, that SCR will cost 
approximately $9,000 per ton of NOX removed, further 
demonstrating the arbitrariness of EPA's proposed disapproval of the 
State's NOX BART determination for Laramie River Station. 
Sargent & Lundy Evaluation, Table 7.
    Response: We disagree that SCR will cost $9,000/ton as indicated by 
the commenter. We have incorporated certain costs claimed by Sargent & 
Lundy in their evaluation, but not others. We have addressed the 
Sargent

[[Page 5182]]

& Lundy cost analysis, including our rationale for not accepting 
certain costs, in detail in other responses. Our revised cost analysis 
of SCR plus combustion controls, indicates that the cost effectiveness 
for the three units is between $4,375/ton and $4,461/ton.
    Comment: EPA should have used 30-day average emission limits in the 
cost effectiveness analysis, rather than expected/actual emission 
rates, to be consistent with how EPA and states have done other BART 
cost effectiveness calculations.
    Response: We disagree. As we have stated in other responses, our 
use of the anticipated actual annual emission rate is consistent with 
the BART Guidelines. As we previously noted, cost effectiveness is more 
appropriately based on the reduction in annual emissions, not the 
change in allowable emissions.
    Comment: We received a comment that EPA's consultant did not take 
into account site-specific data for Laramie River Station.
    Response: We have addressed this issue in a separate response. As 
noted there, we have incorporated many of the costs suggested by Basin 
Electric's consultant, Sargent and Lundy, in our revised costs 
supporting this final action.
    Comment: Wyoming has underestimated the cost of SNCR. Wyoming 
estimated LNB+OFA+SNCR would cost $2,056-$2,109/ton. EPA calculated the 
incremental costs of SCR versus LNB+OFA+SNCR, its preferred control 
option, and estimated incremental costs of $7,054-$7,242/ton. We are 
concerned that Wyoming underestimated the cost of SNCR, which biases 
its emphasis on incremental costs against SCR. We calculated the costs 
of SNCR using the CCM (with the reagent correction used by EPA for 
Montana), and heat inputs and emission estimates from CAMD data for 
2001-2003. Based upon application of the CCM, we estimate SNCR cost-
effectiveness at $2,358-$2,536/ton, which is $300-$400/ton higher than 
Wyoming's estimates.
    Response: We agree that Wyoming has underestimated the cost of SNCR 
for the Laramie River Station units. In order to address deficiencies 
in Wyoming's SNCR cost estimates for the Laramie River Station units 
identified by commenters, such as the control effectiveness of SNCR, we 
have conducted a revised cost analysis.
    Comment: Wyoming has underestimated the ability of SCR to reduce 
emissions. In estimating the annual cost-effectiveness of the 
LNB+OFA+SCR option, Wyoming assumed 0.07 lb/MMBtu, which represents 74% 
control efficiency on an annual average basis, as opposed to the 
generally-accepted 90%. Wyoming has not provided any documentation or 
justification to support the higher emission rates used in its 
analyses. In other recent BART actions, EPA has determined that SCR can 
achieve 0.05 lb/MMBtu on an annual basis. Such an underestimate at 
Laramie River Station biases the cost-benefit analysis against SCR and 
is inconsistent with other EPA analyses.
    Response: The commenter has incorrectly assumed that a 90% control 
efficiency can be achieved in all SCR applications regardless of the 
input NOX emission rate or other parameters. In addition, we 
note that the emission rate analyzed by Wyoming, 0.07 lb/MMBtu, was on 
a 30-day rolling average basis, not an annual basis. Regardless, we 
agree that SCR can in most cases achieve a performance rate of 0.05 lb/
MMBtu on an annual basis. (See the section IV.C.4 of this document for 
more information regarding the control effectiveness of SCR). We have 
revised the SCR costs for the Laramie River Station units accordingly.
    Comment: The final state BART determination sets NOX 
emission limits of 0.21 lb/MMBtu, 30-day average, and related lb/hour 
and tons/year limits. However, EPA does not analyze these limits and 
find they are unreasonable. It analyzes instead a NOX 
emission limit of 0.23 lb/MMBtu, which is not the actual final BART 
limit but rather an initial limit in the BART permit that was appealed 
and was changed in the settlement of that appeal and incorporated in 
the final SIP. Therefore, EPA's disapproval pertains to a BART limit 
that is different than the actual BART limit. Using the wrong BART 
limit is arbitrary and unreasonable.
    Response: In our revised cost and visibility analyses for the 
Laramie River Station BART units, we have addressed the issue described 
by the commenter. However, we have not analyzed the 0.21 lb/MMBtu limit 
directly as it is assessed on a 30-day rolling average basis. Instead, 
as described above, we have used the actual annual emission rate of 
0.19 lb/MMBtu demonstrated since the installation of new LNBs and OFA. 
Our approach is consistent with the BART Guidelines which state: ``. . 
. you will estimate the anticipated annual emissions based upon actual 
emissions from a baseline period.'' 70 FR 39167.
    Comment: The purpose of the regional haze program is to improve 
visibility in Class I areas. The amount of emission reductions by 
itself, without any connection to visibility improvement, is irrelevant 
because without some connection to visibility improvement we cannot 
judge the significance of such reductions in light of the ``overarching 
purpose of the regional haze program'' to protect visibility in Class I 
areas. Since EPA acknowledges that SNCR would not improve visibility by 
a perceptible amount, the amount of NOX emission reductions 
standing alone does not further the purpose of the program.
    Response: We disagree that the visibility improvements for Laramie 
River Station are de minimis or too small to justify the expense of 
requiring controls. 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: ``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.
    EPA followed the BART Guidelines in determining what BART was for 
each unit, taking into account the five factors, including visibility 
improvement and the cost effectiveness of controls (which includes an 
assessment of the dollars per ton removed).
2. Jim Bridger Units 1-4
a. NOX BART Determination
    Comment: In estimating the annual cost-effectiveness of the LNB/
SOFA+SCR option, Wyoming assumed 0.07 lb/MMBtu on an annual average 
basis. Based on the 0.026 lb/MMBtu NOX emission rate 
predicted for the LNB/SOFA option, and the 0.20 lb/MMBtu annual 
emission rates demonstrated by all four Bridger units, outlet emissions 
at 0.07 lb/MMBtu represent only a 65%-73% SCR control efficiency as 
opposed to the generally-accepted 90%. Wyoming has not provided any 
documentation or justification to support the higher emission rates 
used in its analyses. In other recent BART actions, EPA has determined 
that SCR can achieve 0.05 lb/MMBtu on an annual basis. Such an 
underestimate at Bridger biases the cost-benefit analysis against SCR 
and is inconsistent with other EPA analyses.
    Response: The commenter has incorrectly assumed that a 90% control

[[Page 5183]]

efficiency can be achieved in all SCR applications regardless of the 
input NOX emission rate or other parameters. In addition, we 
note that the emission rate analyzed by Wyoming, 0.07 lb/MMBtu, was on 
a 30-day rolling average basis, not an annual basis. Nonetheless, we 
agree that SCR can, in most cases, achieve a performance rate of 0.05 
lb/MMBtu on an annual basis.
    Comment: Based on an evaluation of the rolling 30-boiler operating 
day average NOX emission rates from Jim Bridger Units 1-4 
over the period of January 1, 2011 through March 31, 2013 with 
emissions data available in EPA's CAMD, a NOX limit of 0.07 
lb/MMBtu on a rolling 30-boiler operating day basis would only require 
Jim Bridger Units 1-4 to achieve 66-68 percent NOX removal 
across the SCR systems. The commenter asserted that a 0.05 lb/MMBtu 
NOX emission limit applicable on a rolling 30-boiler 
operating day average basis would only require 75-77 percent 
NOX removal across the SCR, which the commenter believes is 
readily achievable. (The commenter's arguments regarding the achievable 
level of NOX control with SCR are summarized elsewhere in 
this document.)
    Response: We have addressed the control effectiveness of SCR above 
in section V.C.3 above.
    Comment: EPA's reliance on selected ``affordability'' language in 
its BART Guidelines does not support EPA's decision to exempt all Jim 
Bridger from SCR BART requirements. First, Congress established five 
factors--no more--that EPA must consider when making source-by-source 
BART determinations. 42 U.S.C. 7491(g). Applying those five factors 
alone, EPA determined that SCR was BART for all four Bridger Units. 78 
FR 34756. By considering the ``affordability'' of BART controls across 
PacifiCorp's entire fleet, EPA has arbitrarily relied on factors which 
Congress has not intended it to consider. Motor Vehicle Mfrs. Ass'n v. 
State Farm Mut. Auto. Ins. Co., 463 U.S. 29, 43 (1983f); Pac. Coast 
Fed'n of Fishermen's Ass'ns, Inc. v. Nat'l Marine Fisheries Serv., 265 
F.3d 1028, 1034 (9th Cir. 2001); see also North Carolina v. Envtl. 
Prot. Agency, 531F.3d 896, 906 (D.C. Cir. 2008) (standard of review is 
the same under the APA and the CAA, 42 U.S.C. 7607).
    Response: We disagree with the commenter that when considering the 
five factors alone, SCR is BART on all the Jim Bridger units. As 
discussed in section III.B.6 above, when considering the five factors, 
we find it unreasonable to require SCR as BART on these two units and 
instead we are approving the State's LTS for (all four or Units 1 and 
2) the Jim Bridger units. We are not relying on the affordability 
analysis in making this final determination.
    Comment: 40 CFR part 51, Appendix Y, section IV.E.3 makes clear 
that the affordability analysis should be limited to the economic 
impact of the BART unit at issue--not to a utility's fleet-wide BART 
obligations as a whole. EPA's application of this language to 
PacifiCorp's BART obligations at other power plants is improper. In 
addition, the BART Guidelines establish a very narrow test for applying 
the affordability language, which is whether requiring installation of 
the control technology would ``have a severe impact on plant 
operations.'' 40 CFR part 51, app. Y, section IV.E.3. In such 
circumstances, the BART Guidelines suggest that EPA prepare ``an 
economic analysis that demonstrates, in sufficient detail for public 
review, the specific economic effects, parameters, and reasoning.'' EPA 
has not prepared an economic analysis demonstrating the specific 
economic, parameters, and reasoning for its decision to exempt the Jim 
Bridger facility from further BART controls. Instead, EPA simply 
concludes that ``it would be unreasonable to require any further 
retrofits at this source within five years of our final action'' based 
on PacifiCorp's BART obligations at other facilities. 78 FR 34756. 
Further, PacifiCorp has not presented evidence that installation of SCR 
at each Bridger unit within the first regional haze planning period 
would cause any noticeable economic impact, let alone ``severe'' 
impact, such a shutdown of one or all of the units.
    Response: As discussed in more detail in a response to another 
comment below, we agree that PacifiCorp has not provided sufficient 
evidence to show that the installation of SCR at the Jim Bridger units 
within five years after our final action would cause a severe economic 
impact. We are basing our decision to not require SCR for BART based on 
our weighing of the five factors.
    Comment: The BART Guidelines also suggest that if the agency grants 
an affordability exemption from the best level of control, it must then 
select a ``slightly lesser degree of control.'' 40 CFR part 51, app. Y, 
section IV.E.3. EPA's proposal does not require any additional level of 
control under BART. Instead, EPA's 2013 re-proposal selects the pre-
existing LNB/OFA as BART for each unit. EPA's choice of LNB/OFA as BART 
does not even represent the ``second best'' control technology for 
eliminating NOX related visibility impairment--which would 
be SNCR. 78 FR 34756 (Table 13).
    Response: As stated above, we are not basing our BART determination 
for Jim Bridger Units 1 and 2 on an affordability argument.
    Comment: The BART Guidelines ``affordability'' language recommends 
that states (or EPA) consider ``whether other competing plants in the 
same industry have been required to install BART controls if this 
information is available.'' 40 CFR part 51, app. Y, section IV.E.3. 
This provision suggests that SCR requirements should be applied 
consistently among competing utility companies. EPA's exemption of 
PacifiCorp from SCR obligations at the Jim Bridger units is 
inconsistent with its actions at other competing utility companies with 
large coal fleets, to which EPA has not offered ``affordability'' 
exemptions from SCR requirements. For example, EPA has required Salt 
River Project, a competing utility, to install SCR on numerous coal-
plant units in its fleet, including Coronado and Navajo (Arizona), 
Craig and Hayden (Colorado) and Four Corners (New Mexico). Each of 
EPA's BART determinations for these plants requires installation of SCR 
as BART within the mandated five-year implementation deadline.
    EPA's 2013 re-proposal for Jim Bridger exempts the plant from both 
the SCR requirement and the five-year implementation deadline, 
potentially giving PacifiCorp an advantage over the competing 
utilities. Thus, EPA's 2013 re-proposal rule is not only inconsistent 
with its own BART Guidelines, it is competitively unfair.
    Response: We agree that other utility companies have had to install 
SCR within the five year BART window and that evidence provided by 
PacifiCorp does not support delaying controls on Jim Bridger Units 1 
and 2 until 2022 and 2021, respectively. Nonetheless, as stated 
earlier, we are no longer basing our BART determination for Jim Bridger 
Units 1 and 2 on an affordability argument.
    Comment: We are very familiar with the ``affordability'' provisions 
of the BART Guidelines and have dealt with this issue in Arizona 
(Apache power plant) and Washington (Alcoa's Intalco primary aluminum 
smelter). In both of those cases, the company requesting the 
affordability exemption from BART provided extensive documentation 
(much of it confidential) to EPA and the FLMs to support its request. 
It was only after a thorough review by EPA that the affordability 
exemptions were approved. In this case, it appears that the only 
information presented by PacifiCorp to

[[Page 5184]]

support its request is its ``assertions'' dated July 12, 2012. We 
believe that a more rigorous analysis is necessary in order for EPA, 
FLMs, and the public to be assured that the additional time being 
proposed by EPA is necessary and appropriate. For example, an important 
part of such an analysis would be the ``installation schedule'' that 
PacifiCorp has designed in order to minimize the number of units that 
are out of service system-wide for installation of emissions controls 
at any one time. Currently, the only schedule available in the docket 
is the July 2012 letter from PacifiCorp to EPA in which PacifiCorp 
simply reiterates the dates proposed for its ``Installation 
Requirements.''
    Response: As stated in other responses to comments, we agree that 
the information provided by PacifiCorp was not sufficient to support 
the delay of SCR controls on Jim Bridger Units 1 and 2, but we are not 
relying on that information in our BART determination.
    Comment: EPA's proposal for Jim Bridger Units 1 and 2 is a ``do 
nothing'' BART determination. Although the Agency claims that it is 
proposing to approve the State's proposal to require the use of LNBs 
for Units 1 and 2 and for both units to meet an emission rate of 0.28 
lb/MMBtu over a 30-day rolling period, this emission rate is actually 
higher than what Units 1 and 2 are currently emitting and worse, does 
not reflect the presumptive BART limits set forth in 40 CFR part 51, 
Appendix Y. Most significantly, it defies the statutory intent of 
Congress in establishing the regional haze program under the CAA.
    According to EPA's CAMD Web site, both Units 1 and 2 already 
consistently achieve 30-day rolling average NOX emissions 
lower than 0.20 lb/MMBtu. The data illustrates that Jim Bridger Units 1 
and 2 consistently achieve NOX emission rates below 0.20 lb/
MMBtu on a monthly basis and have done so since 2010. To this end, the 
definition of BART explicitly states that it must represent a 
``reduction'' in each pollutant that causes or contributes to 
visibility impairment. See 40 CFR 51.301 (setting forth definition of 
BART).
    Furthermore, although a state must take into account the five 
factors set forth under 42 U.S.C. 7491(g)(2), nothing in the CAA or the 
EPA's regulations implementing the regional haze program suggest or 
remotely imply that a state could allow emission increases as BART. 
Accordingly, EPA must, at a minimum, disapprove of Wyoming's 
NOX BART determinations for Jim Bridger Units 1 and 2 and 
adopt a FIP that establishes BART limits that are consistent with the 
CAA and that represent actual emission reductions.
    Response: We disagree with the commenter that BART is an emission 
limit of 0.07 lb/MMBtu at Jim Bridger Units 1 and 2 within five years 
of our final action. As discussed previously, based on our weighing of 
the five factors, we do not find it reasonable to require SCR for BART 
on Jim Bridger Units 1 and 2 and instead we are approving the State's 
LTS for these units.
    Comment: The need to promulgate a FIP is underscored by the EPA's 
own BART guidelines. According to those guidelines, tangentially fired 
boilers burning subbituminous coal, such as Jim Bridger Units 1 and 2, 
are presumed to be able to cost-effectively meet a NOX 
emission rate of 0.15 lb/MMBtu on a 30-day rolling average basis. See 
40 CFR part 51, Appendix Y, Table 1. EPA's claim that it would not be 
cost-effective to meet an emission rate below 0.28 lb/MMBtu as BART for 
Units 1 and 2 is therefore undercut by the Agency's own extensive 
analysis and conclusion that a 0.15 lb/MMBtu rate is presumed to be 
appropriate. Tellingly, the EPA nowhere in its proposed rule analyzes 
or addresses why a 0.28 lb/MMBtu rate is appropriate in light of the 
Agency's own presumptive BART limits for NOX emissions from 
tangentially-fired boilers burning subbituminous coal.
    Response: We disagree with the commenter's assertion that our 
approval of non-presumptive BART emission limits for Jim Bridger Units 
1 and 2 is flawed. In the BART Guidelines EPA explained that:

    For coal-fired EGUs greater than 200 MW located at greater than 
750 MW power plants and operating without post-combustion controls 
(i.e. SCR or SNCR), we have provided presumptive NOX 
limits, differentiated by boiler design and type of coal burned. You 
may determine that an alternative control level is appropriate based 
on a careful consideration of the statutory factors. For coal-fired 
EGUs greater than 200 MW located at power plants 750 MW or less in 
size and operating without post-combustion controls, you should 
likewise presume that these same levels are cost-effective. You 
should require such utility boilers to meet the following 
NOX emission limits, unless you determine that an 
alternative control level is justified based on consideration of the 
statutory factors.\231\
---------------------------------------------------------------------------

    \231\ 40 CFR Part 51, Appendix Y, section IV.E.5.

Therefore, the presumptive emission limits in the BART Guidelines are 
rebuttable, and the five statutory factors enumerated in the BART 
Guidelines provide the mechanism for establishing different 
requirements. Specifically, as explained in the preamble to the BART 
---------------------------------------------------------------------------
Guidelines:

    If, upon examination of an individual EGU, a State determines 
that a different emission limit is appropriate based upon its 
analysis of the five factors, then the State may apply a more or 
less stringent limit.\232\
---------------------------------------------------------------------------

    \232\ 70 FR 39132.

    Thus, the establishment of presumptive BART emission limits does 
not preclude states or EPA from setting limits that differ from those 
presumptions, even where the control technology is the same as that 
associated with the presumptive limits (in this case, combustion 
controls). The five-factor analysis performed by Wyoming demonstrates 
that, because of the nature of the coal fired at these units, the 
presumptive limit of 0.15 lb/MMBtu is not attainable. Wyoming supported 
this conclusion with information from an established vendor of 
combustion controls.\233\ We concur with those conclusions and find 
that Wyoming's BART emission limits for Jim Bridger Units 1 and 2 were 
established in a manner consistent with the BART Guidelines.
---------------------------------------------------------------------------

    \233\ Wyoming Department of Environmental Quality Air Quality 
Division, BART Application Analysis AP-6040, May 28, 2009, pages 7-
9.
---------------------------------------------------------------------------

    Comment: EPA's proposal is fundamentally flawed because it makes a 
mockery of the CAA. Despite acknowledging that BART should be the 
installation of SCR and compliance with a 0.07 lb/MMBtu emission rate 
for Units 1 and 2, the EPA determined that, when considering the cost 
of such controls, they would not be reasonable. Certainly, the CAA 
allows the EPA to consider the ``cost of compliance'' in setting BART 
(42 U.S.C. 7491(g)(2)), but the CAA does not allow the EPA to 
completely avoid requiring BART based on cost considerations. Here, 
EPA's proposal to approve Wyoming's SIP with regards to BART for Jim 
Bridger Units 1 and 2 amounts to a proposal to require nothing (if not 
a proposal to allow an increase in emissions). In essence, EPA's 
proposal amounts to a determination that BART is not required for Jim 
Bridger Units 1 and 2, even though Congress clearly stated that these 
coal-fired EGUs are subject-to-BART. Although Congress allowed the EPA 
to consider costs in establishing BART, the EPA cannot use costs as a 
reason to completely forego requiring BART. Put another way, the EPA 
cannot defeat Congress' intent to require BART by cobbling together an 
interpretation of the CAA that effectively nullifies the regional haze 
BART requirements under the Act.

[[Page 5185]]

    Response: We disagree with the commenter. As discussed elsewhere, 
we are basing on approval of the State's SIP on a weighing of the five 
BART factors, including costs and visibility improvement.
    Comment: EPA's proposal for Jim Bridger Units 1 and 2 ``does not 
meet a test for being `better than BART'. . . [because] [t]he 
accelerated installation of BART at Bridger Units 3 and 4 does not 
offset the increased emissions from delaying SCR installation at 
Bridger Units 1 and 2 beyond the normal five-year BART window.'' To 
date, EPA has failed to make any demonstration that its 2013 Proposal 
would ``achieve greater reasonable progress than would be achieved 
through the installation and operation of BART.'' 40 CFR 51.308(2)(i). 
Moreover, EPA's proposal also fails to comply with the ``better than 
BART'' regulatory requirement mandating that ``all necessary emission 
reductions take place during the period of the first LTS for regional 
haze'' which concludes at the end of 2017. Id. section 51.308(b), 
(e)(2)(iii). Accordingly, EPA's proposal has not satisfied the 
regulatory requirements for a ``better than BART'' alternative.
    Response: We agree that EPA's proposal does not meet a test for 
being better than BART but have not suggested such a concept in our 
proposed or final rulemaking actions.
    Comment: EPA's proposed action on the Wyoming 308 regional haze SIP 
requested comments on whether to require installation of BART controls 
on Jim Bridger Units 1 and 2 by 2021-2022 rather than within the 
legally required five-year timeframe mandated by the regional haze 
regulations. 77 FR 33054. EPA is taking comment on the alternative 
timeline for SCR installation in response to PacifiCorp's claim that 
``the schedule for installation of emission control devices envisioned 
in [EPA's BART proposal] would be excessively costly and would pose 
service interruption risks for electrical energy customers over a large 
part of the region.'' Recent admissions by a PacifiCorp official in a 
separate Wyoming Public Service Commission proceeding undermines 
PacifiCorp`s arguments. PacifiCorp argued to the Wyoming Public Service 
Commission that procurement and installation of multiple SCRs creates 
both a cost and time savings, not an increase at Jim Bridger Units 3 
and 4. This fact is also true for installation of SCRs at Jim Bridger 
Units 1 and 2. This PacifiCorp admission is further proof that EPA 
should not permit PacifiCorp to delay installation of SCRs at Jim 
Bridger Units 1 and 2 and instead must require compliance within five 
years as is required by the BART regulations.
    Response: We have responded to this comment in other responses.
    Comment: EPA is taking comment on the alternative timeline for SCR 
installation in response to PacifiCorp's claim that the schedule for 
installation of emission control devices envisioned in EPA's BART 
proposal would be excessively costly and would pose service 
interruption risks for electrical energy customers over a large part of 
the region. EPA acknowledges that BART for all the units at Jim Bridger 
is SCR when the units are considered individually based on the five 
factors. However, EPA suggests that a different BART determination 
under the alternative approach is lawful if the five factors are 
considered across all the units in the PacifiCorp system. Not so. BART 
is a source-by-source determination.
    Response: We have responded to this comment in other responses.
    Comment: Considerations of PacifiCorp's fleet size and cumulative 
costs are outside the five-factor analysis for BART. Furthermore, 
PacifiCorp's concern about the feasibility of installing BART controls 
over its large fleet is unfounded. With proper planning, there is no 
reason to expect excessive costs or service interruption due to BART 
requirements. Indeed, PacifiCorp's large number of EGUs would appear to 
give PacifiCorp the unique ability to avoid service disruptions by 
maintaining adequate capacity from operating units while other units 
are offline. Further, other utilities have installed SCR systems on 
multiple units within very short periods of time.
    PacifiCorp's ability to install SCR on multiple units is also not 
constrained by the availability of SCR systems. In response to 
questioning of whether PacifiCorp has had any difficulties procuring or 
installing SCR systems, particularly an SCR for its Naughton Unit 3 
facility, PacifiCorp stated it had received four proposals from SCR 
system suppliers and as such did not experience notable difficulties 
procuring and obtaining the SCR system. EPA should not modify its BART 
proposal for Jim Bridger Units 1 and 2 based on PacifiCorp's 
unsupported claims of hardship.
    Response: We have responded to this comment in other responses.
    Comment: The EPA's re-proposed Wyoming haze plan reiterates EPA's 
prior finding that BART is SCR for each Jim Bridger unit considered 
individually (78 FR 34756). Based on the EPA's five-factor 
NOX BART analyses for Jim Bridger Units 1-4, the EPA must 
find that SCR is BART to meet a NOX emission rate of 0.05 
lb/MMBtu on all four units.
    The EPA's cost-effectiveness analyses identified costs for SCR on 
all four Bridger Units that are within the range that EPA has 
identified as reasonable for other units, including in this same 
proposal. 78 FR 34754-57 (SCR cost-effectiveness of $2,393/ton on Jim 
Bridger Unit 1, $3,015/ton on Unit 2, $2,961/ton on Unit 3, and $2,492/
ton on Unit 4) as compared with, e.g., 78 FR 34776 (finding cost-
effectiveness of $3,600/ton to $3,900/ton for SCR on Laramie River 
Units 1-3 to be ``within the range of what EPA has found reasonable for 
BART in other SIP and FIP actions''). However, the EPA's estimate of 
costs is significantly inflated, and the true costs are even lower than 
EPA found because the EPA used unjustified ``retrofit factors,'' 
interest rate, and auxiliary power costs. The recalculation of costs 
using what the commenter believes are corrected inputs resulted in 
significantly lower SCR costs of $1,801 to $1,959/ton at all four 
Bridger units. On this basis, SCR is extremely cost effective on these 
units. SCR on these units would afford substantial visibility benefits. 
The EPA has no basis in the BART factors, including the important 
factors of compliance costs and visibility improvement, to reject SCR 
as BART on Bridger Units 1-4.
    Response: As discussed elsewhere, we have responded to the 
commenter's points about how costs are calculated for the BART units. 
Regardless, we determined that SCR was not reasonable for BART based on 
our weighing of the five factors.
    Comment: The EPA's 1st Proposal from its June 2012 Proposal/2013 
Proposal in the Alternative, which would approve Wyoming's 
NOX emission limits and SCR compliance timeframes for Jim 
Bridger Units 3 and 4 and would require Jim Bridger Units 1 and 2 to 
install SCR within five years of EPA's final action on the Wyoming 
regional haze plan, would result in lower NOX emissions on 
an annual basis than any of EPA's other NOX proposals at the 
Jim Bridger Power Plant. This schedule also likely reflects the most 
economical installation of SCR at all four of the Jim Bridger units 
because the engineering for SCR installation, including the design of 
the construction phase, can all be done during the same time frame, the 
construction equipment can remain on-site for the duration of the 
installations, and much of the installation work can be done

[[Page 5186]]

simultaneously, which will save on labor and construction equipment.
    There are numerous examples of installations of multiple SCRs at 
numerous units at the same power plant site over short timeframes, 
including: (1) W.A. Parish Units 5-8 (SCRs installed over 2003-2004); 
(2) All four units of the Big Bend power plant (SCRs installed during 
2007-2010); (3) Units 1-5 of the Clifty Creek power plant (SCRs 
installed 2002-2003); (4) Winyah Units 1-4 (SCRs installed in 2005); 
(5) Over the period of 2001-2006, TVA has installed SCRs at 18 units at 
four power plants. On this basis, the number and timing of SCR 
installations required at PacifiCorp plants as a result of 
NOX BART determinations can be accomplished, as it has been 
done before.
    Response: As stated above, we find that PacifiCorp has not 
presented sufficient evidence that the economic effects of installation 
of SCR on Jim Bridger Units 1 and 2 within five years would affect the 
viability of continued plant operations, but based on weighing of the 
five factors, we find that it is not reasonable to require SCR for 
BART.
    Comment: We agree with EPA that Wyoming's proposal to require 
installation of SCR in 2021 and 2022 as part of Wyoming's LTS does not 
satisfy the CAA or its implementing regulations. Having determined that 
SCR plus LNBs/SOFA is the best system of continuous emission control, 
is cost effective, and will result in significant visibility 
improvement, EPA is required to find that the controls are BART. 40 CFR 
51.308(e)(i)(2)(A). Under the RHR, BART must be installed ``as 
expeditiously as practicable, but in no event later than five years 
after approval of the implementation plan revision.'' 40 CFR 
51.308(e)(iv). Thus, EPA lacks discretion to approve Wyoming's proposal 
to require PacifiCorp to install BART technology beyond the five-year 
time frame.
    Response: See responses above.
    Comment: PacifiCorp submitted comments in support of delaying 
controls on Jim Bridger Units 1 and 2 until 2022 and 2021 respectively. 
(EPA issued a Notice of Data Availability pertaining to this 
information on July 24, 2012. 77 FR 43205). The main points raised in 
their comments are as follows: \234\
---------------------------------------------------------------------------

    \234\ See July 12, 2012 comments from PacifiCorp in the docket.
---------------------------------------------------------------------------

     Because of the size and multi-state nature of its 
generation fleet, PacifiCorp and its customers are unreasonably 
impacted by the RHR. PacifiCorp provides regulated electric service to 
more than 1.7 million customers in California, Idaho, Oregon, Utah, 
Washington and Wyoming with a net system capacity of 10,597 megawatts. 
PacifiCorp operates 75 generating units across the western U.S. 
PacifiCorp owns and operates 19 coal-fueled generating units in Utah 
and Wyoming, and owns 100% of Cholla Unit 4, a coal-fueled generating 
unit in Arizona. In addition, PacifiCorp has an ownership interest in 
Craig Units 1 and 2 and Hayden Units 1 and 2 in Colorado.
     As evidenced by the emission reduction projects which 
PacifiCorp has already installed in accordance with the Utah and 
Wyoming regional haze SIPs, PacifiCorp is not opposed to making 
emission reductions that are cost effective for its customers and that 
achieve environmental benefits, as required by law. PacifiCorp has 
undertaken projects to comply with the Utah and Wyoming SIPs at a cost 
of approximately $1.3 billion (PacifiCorp's share of $1.4 billion of 
total project costs) between 2005 and 2011. Those projects, in 
conjunction with projects completed through 2012, have reduced 
emissions of SO2 by approximately 58% and emissions of 
NOX by approximately 46%.
     Just as modeled visibility improvements associated with 
PacifiCorp's emission reduction projects do not stop artificially at a 
state border, EPA's analysis of the impacts of its proposed FIP for a 
large, multi-state system like PacifiCorp's should not be limited to 
only those facilities and customers located within Wyoming's borders. 
EPA's actions impacting large, multi-state systems in one state must 
also consider the cumulative impacts of all of its actions in all other 
states that affect the same system.
     Given the number of facilities operated by PacifiCorp and 
the facilities in which the company has an ownership interest in and is 
required to pay costs for the installation of regional haze-related 
controls, accelerated and additional controls under the proposed FIP 
result in approximately $500 million of additional capital expenditures 
plus an incremental annual cost of $16-24 million to operate those 
controls in the next five years. In addition, an EPA proposal for 
stringent control requirements in Utah (i.e., SCR) within five years 
would add approximately $750 million in capital expenditures, plus 
approximately $7 million to $9 million annually in operating costs and 
approximately $4 million annually for catalyst replacement projects. 
All of these costs will be put on the backs of PacifiCorp and its 
customers in an extremely short-time frame.
     In addition to the regional haze requirements, 
PacifiCorp's coal-fueled generating fleet, including the BART-eligible 
units, must accommodate controls for compliance with the Mercury and 
Air Toxics Standards (MATS) during the same timeframe. While the 
scrubbers and baghouses already installed at many of the PacifiCorp 
facilities pursuant to the Utah and Wyoming regional haze SIPs position 
the company well to comply with the acid gas and non-mercury metals 
limits under the MATS requirements, additional work will be necessary 
to comply with the mercury emission limits by April 2015.
     PacifiCorp's customers cannot absorb increasing 
environmental costs. To accommodate, among other cost increases, the 
costs of the environmental controls already installed on PacifiCorp's 
coal-fueled generating facilities, PacifiCorp has filed with its 
utility regulatory authorities annual cases to increase customer rates. 
PacifiCorp's customers have consistently participated in these cases to 
express concerns regarding increases in electric rates. While EPA may 
view its proposal to accelerate the installation of controls and 
require additional controls at PacifiCorp's facilities as just another 
utility complaining to avoid the consequences of large investments in 
controls, EPA's proposal has a very real impact on customers.
     As a regulated utility, PacifiCorp has a legal obligation 
to supply reliable electric service at reasonable rates as set by state 
utility commissions; it also has a legal requirement to supply its 
customers as much electricity as they want, when they want it. While 
the installation of emissions controls on multiple units in a short 
period of time creates substantial challenges from a project management 
perspective, these challenges are exacerbated by increased risk factors 
that jeopardize PacifiCorp's ability to meet its underlying utility 
obligations and challenge the reliability of the system.
     When considered independently from other environmental 
requirements, the retrofits required under either regional haze 
compliance scenario are not anticipated to impose undue stress on the 
national supply chain for specialized labor, materials, and equipment. 
However, analyses of compliance with the MATS have raised concerns that 
requiring much of the U.S. coal fleet to retrofit or retire in a three 
to five year-time frame (partially overlapping the compliance time 
period under the regional haze program) will

[[Page 5187]]

challenge the equipment construction industry.
     Wyoming and EPA are legally required to consider the 
economic and system impacts on PacifiCorp and its customers. As EPA's 
BART Guidelines explain: ``There may be unusual circumstances that 
justify taking into consideration the . . . economic effects of 
requiring the use of a given control technology. These effects would 
include effects on product prices . . . Where these effects are judged 
to have a severe impact on plant operations you may consider them in 
the selection process, but you may wish to provide an economic analysis 
that demonstrates, in sufficient detail, for public review, the 
specific economic effects, parameters, and reasoning.'' 70 FR 39171. 
Given the large number of BART impacted units owned by PacifiCorp in 
different states, these unusual circumstances justify Wyoming's BART 
actions on PacifiCorp's facilities and PacifiCorp's customers.
    Response: PacifiCorp argues that springtime scheduling of the unit 
outages and outage extensions needed for ``tie-in'' of retrofitted 
controls could challenge system reliability in certain years--in 
PacifiCorp's forecast, power demand plus reserves would temporarily 
exceed available supply. EPA believes that this forecast is unrealistic 
because PacifiCorp constrains itself almost entirely to use of its own 
generation supplies (ignoring other available generators in its region) 
and limits its assessment to springtime-only outages in its system-wide 
outage planning examples. PacifiCorp indicates that spring outages are 
economically preferred due to the historical availability of cheap 
hydro replacement power and the typically higher alternative costs of 
purchased replacement power at other times. However, PacifiCorp 
provides no information on the availability or net cost of replacement 
power to meet demand, nor does PacifiCorp identify any alternative 
retrofit outage schedules. This is a significant omission because 
alternative retrofit outage schedules that avoid reliability issues 
through non-coincident temporary uses of purchased power, even if such 
temporary power purchases may cost more than power typically provided 
by the facilities experiencing an outage, might have a very small 
levelized net retail cost impact when applied to customers system-wide. 
In short, PacifiCorp ties its own hands in its provided analysis, 
ignoring proven and cost-effective strategies for maintaining electric 
reliability to allow facility upgrades in a timely fashion.
    EPA notes that PacifiCorp overstates the purported regulatory 
burden on its generating resources by claiming that the company ``has 
not yet identified a viable control suite that will allow it to comply 
with the [Mercury and Air Toxics Standards, or MATS] provisions at the 
Carbon plant in Utah. As a result, while not finally determined, it is 
anticipated that Carbon Units 1 and 2 will be required to be shut down 
in the 2015 timeframe, resulting in the loss of 172 megawatts of 
generation from PacifiCorp's system.'' Such an assumption is unfounded 
and ignores the EPA's clear explanation in the final MATS that under 
the CAA, state permitting authorities can also grant sources an 
additional year as needed for technology installation. EPA expects this 
option to be broadly available. EPA is also providing a pathway for 
reliability critical units to obtain a schedule with up to an 
additional year to achieve compliance. This pathway is described in a 
separate enforcement policy document.\235\ As a result, the comment 
does not identify any specific conflict between MATS compliance 
planning at the Carbon facility and regional haze compliance planning 
at the Jim Bridger units at issue in this rule.
---------------------------------------------------------------------------

    \235\ Memorandum from Cynthia Giles, Assistant Administrator of 
the Office of Enforcement and Compliance Assurance, ``The 
Environmental Protection Agency's Enforcement Response Policy for 
Use of Clean Air Act Section 113(a) Administrative Orders in 
Relation to Electric Reliability and the Mercury and Air Toxics 
Standard'' (Dec. 16, 2011).
---------------------------------------------------------------------------

    In developing their argument, PacifiCorp borrows a ``WetFGDeq'' 
concept that EPA used in its nation-wide analysis of the feasibility of 
retrofitting all controls that might be needed for timely compliance 
with MATS. PacifiCorp uses EPA's nationally applicable WetFGDeq concept 
to compare annual WetFGDeq MW amounts of the limited site-specific 
retrofit activity that PacifiCorp has actually conducted during the 
past two decades, and might conduct in the future under the SIP, to the 
annual amount that they might have to conduct in the future under the 
proposed FIP. Based on this comparison, PacifiCorp states the following 
(see page 20 of 23; also see Figure 8 of PacifiCorp's July 12, 2012 
comments): ``The differences between the SIP Scenario and the 
Aggressive BART Scenario are fairly substantial on an equivalent Wet 
FGD basis. In the SIP Scenario, only one year exceeds the 2010-2011 
levels of retrofit investment (of about 225 MW/year), while retrofits 
placed in service in 2017 (675 MW) substantially exceed the previous 
historic maximum of 475 MW by 200 MW and two years are above the 2010-
2011 level. The control installation requirements under the EPA 
Aggressive BART Scenario would result in more work, less time, and 
increased costs.''
    EPA does not disagree that the proposed FIP may entail more 
PacifiCorp project management and construction effort (in one year, 
2017) than the SIP would require, or than PacifiCorp has actually 
experienced as an individual company in the past. However, EPA does not 
consider the relatively small absolute amounts of the differences (200-
300 MW) to be a serious obstacle for any large utility, given a bevy of 
retrofit experience of this magnitude by like companies in the past, on 
similar schedules.
    Comment: In making any BART determinations on a large, multi-
jurisdictional system such as PacifiCorp's, the regulating agency must 
consider the broad scope of the impacts of its decisions on customers 
and generating system reliability as a whole. Wyoming considered these 
factors in developing its regional haze SIP: ``The Division believes 
that the size of PacifiCorp's fleet of coal-fired units presents unique 
challenges when reviewing costs, timing of installations, customer 
needs, and state regulatory commission requirements. Information has 
been supplied by PacifiCorp elaborating on additional factors to be 
considered in PacifiCorp's BART determination (see PacifiCorp's 
Emissions Reductions Plan in Chapter 6 of the Wyoming technical support 
document).''
    Wyoming's consideration of these factors was appropriate. While 
PacifiCorp agrees with EPA's proposed conclusions regarding the 
reasonableness and timing of installation of controls at Jim Bridger 
Units 1 and 2, EPA's focus on affordability impermissibly fails to 
consider the unusual circumstances and broader impacts of its action on 
PacifiCorp's other BART Units. EPA's selection of SCR controls at 
Naughton Units 1 and 2 and at Dave Johnson Unit 3 will affect the 
viability of continued unit operations. Installation of SCR controls at 
these three units, particularly given the cost of controls and their 
remaining useful life, create such ``unusual circumstances'' that 
justify taking into consideration the conditions of the plant and the 
economic effects of requiring the use of a given control technology.
    The timing and reasonableness of the eight SCR and two SNCR and 
LNBs required in EPA's proposed action must

[[Page 5188]]

be considered in the context of the additional controls required at 
PacifiCorp's units in Arizona (Cholla Unit 4 with SCR required by 2017) 
and its share of units in Colorado (Hayden Unit 1 with SCR in 2015, 
Hayden Unit 2 with SCR in 2016, Craig Unit 1 with SNCR in 2017 and 
Craig Unit 2 with SCR required in 2016) and the potential for 
additional controls required at four of PacifiCorp's BART-eligible 
units in Utah within five years after final action. EPA's failure to 
consider the ``unusual circumstances'' contemplated under its Appendix 
Y Guidance means the agency acted in a manner that is arbitrary and 
capricious in its overall assessment (or lack thereof) of the effects 
of its actions on PacifiCorp's generation fleet.
    Response: See our response to the comment above.
    Comment: Pursuant to 40 CFR 51.308(e), the State included 
provisions in its 309(g) regional haze SIP to address BART. When 
evaluating each permit application, the State determined BART for each 
source by evaluating visibility control options presented in the 
applications using the methodology prescribed in 40 CFR part 51, 
Appendix Y.
    The use of the BART guidelines contained in Appendix Y is only 
required for sources located at EGUs with a total capacity greater than 
750 MW, which for Wyoming were Basin Electric's Laramie River Station 
and PacifiCorp's Jim Bridger and Dave Johnston plants. However, for 
consistency, the State chose to follow the guidelines for all BART 
sources, including those located at the trona facilities. By using the 
guidelines of Appendix Y for all sources, the State established a 
consistent framework for performing BART evaluations. Finally, when 
selecting the ``best alternative,'' the State considered additional 
impacts to both the plant and the State. Appendix Y affords the 
determining authority discretion to consider additional impacts. See 70 
FR at 39171.
    The State's BART analysis not only considered all statutory 
factors, but also considered the significant impact on energy costs to 
PacifiCorp's Wyoming rate payers if the controls, including SCR, were 
required to be installed in the BART timeframe of five years after SIP 
approval. While the State did not have the resources to perform a 
highly technical and complex analysis to quantify the potential cost 
impact of requiring installation of SCR controls on all of PacifiCorp's 
Wyoming fleet, additional information was provided by PacifiCorp for 
public review.
    In addition to identifying costs in terms of capital expenditures, 
the State also considered the logistical challenges unique to 
PacifiCorp. The State is not aware of any other company faced with as 
many potential add-on control installations as PacifiCorp. 
Additionally, the State noted potential reliability issues related to 
the extended downtimes needed for the installation of SCR systems on 
multiple units within the BART timeframe. The impact of taking down 
large units, like Jim Bridger Units 1-4, each rated at a nominal 530 
MW, for extended outages increases the possibility of power shortages, 
not to mention increased power cost if PacifiCorp must purchase 
additional power at spot market prices to meet demand.
    Response: The commenter raises many of the same points that 
PacifiCorp raised in its July 12, 2012 comments on our third proposal 
in the alternative, and we have responded to the commenters points in 
our response above. EPA does not find the arguments for delaying 
controls put forth by PacifiCorp or the commenter to be compelling.
    Comment: In making any BART determinations on a large, multi-
jurisdictional system such as PacifiCorp's, the regulating agency must 
consider the broad scope of the impacts of its decisions on customers 
and generating system reliability as a whole. Wyoming considered these 
factors in developing its regional haze SIP: ``The State believes that 
the size of PacifiCorp's fleet of coal-fired units presents unique 
challenges when reviewing costs, timing of installations, customer 
needs, and state regulatory commission requirements.'' Information has 
been supplied by PacifiCorp elaborating on additional factors to be 
considered in PacifiCorp's BART determination (see PacifiCorp's 
Emissions Reductions Plan in Chapter 6 of the Wyoming Technical Support 
Document). Wyoming's consideration of these factors was appropriate. 
EPA's rejection of these factors was not appropriate.
    Given the large number of BART impacted units owned by PacifiCorp 
in different states, these ``unusual circumstances'' justify Wyoming 
and EPA considering the impact of EPA's BART decision-making in the 
western U.S. on PacifiCorp and its customers.
    Response: We have responded to the commenter's points in our 
responses above. As stated, EPA does not find the arguments for 
delaying controls put forth by the State or PacifiCorp to be 
compelling.
    Comment: Congress has defined ``best available retrofit 
technology'' as ``an emission limitation based on the maximum degree of 
reduction of each pollutant subject to regulation under this chapter. . 
. .'' 42 U.S.C. 7479(3). Congress also narrowly defined which sources 
would be exempt from BART. Section 169A(c) of the CAA exempt fossil 
fuel power plants exceeding 750 megawatts only if the ``owner or 
operator of any such plant demonstrates to the satisfaction of the 
Administrator that such power plant is located at such a distance from 
all areas . . . that such power plant does not or will not, by itself 
or in combination with other sources, emit any air pollutant which may 
reasonably be anticipated to cause or contribute to significant 
impairment of visibility in any such area.'' Id. at section 7491(c)(2). 
Any such exemption must be agreed to by the FLMs. Id. at section 
7491(c)(3). PacifiCorp has not submitted evidence demonstrating that 
the Jim Bridger coal plant--with a net generating capacity of 2,120 
megawatts, 78 FR 34753--will not cause or contribute to significant 
visibility impairment in any Class I area. To the contrary, EPA's own 
visibility modeling shows that Bridger has significant visibility 
impacts at numerous Class 1 areas. 78 FR 34754-34758. As such, EPA may 
not exempt the Jim Bridger plant from BART.
    Response: We agree with the commenter that PacifiCorp has not 
submitted any evidence that the Jim Bridger plant is located at such a 
distance from all Class I areas that the plant will not, by itself or 
in combination with other sources, emit any air pollutant which may 
reasonably be anticipated to cause or contribute to significant 
impairment of visibility in any such area.
    Comment: EPA should require installation of SCR at each Jim Bridger 
unit within the five-year regulatory deadline because this approach 
offers the greatest visibility improvement.
    Response: See responses above.
    Comment: EPA proposes that Wyoming's determination of 
NOX BART for Jim Bridger units 1 and 2 as new LNB plus OFA 
is reasonable and that it would be unreasonable of the EPA to require 
any further retrofits at these units within five years of EPA's final 
action. 78 FR 34756. The State supports EPA's proposed approval of 
NOX BART as LNB plus OFA for Jim Bridger Units 1 and 2. EPA 
also proposes to approve the State's LTS of NOX control for 
Jim Bridger Units 1 and 2 as the SCR-based emission rate of 0.07 lb/
MMBtu with compliance dates of December 31, 2021, for Unit 2 and 
December 31, 2022, for Unit 1.
    Based on facts PacifiCorp raised concerning the additional 
requirements

[[Page 5189]]

in the proposed FIP for Wyoming, the finalized FIP for Arizona, and the 
possibility of additional requirements in a future FIP or SIP for Utah, 
the additional time allowed PacifiCorp to install controls under the 
State's LTS on Jim Bridger Units 1 and 2 is warranted under the 
affordability provisions in the BART Guidelines. 40 CFR part 50, App. 
Y, section IV(E)(3); see also 78 FR 34756. Wyoming therefore supports 
EPA's proposed approval.
    Response: We disagree with the points raised by the commenter in 
the second paragraph and have addressed their points in previous 
responses to comments. Nonetheless, we are approving the State's SIP 
for Jim Bridger Units 1 and 2.
    Comment: Wyoming strongly urges EPA to stand by its proposed 
approval of Wyoming's SIP requiring Jim Bridger Unit 1 to meet the 0.07 
lb/MMBtu emission rate prior to December 31, 2021 and Unit 2 to meet 
the 0.07 lb/MMBtu emission rate prior to December 31, 2022. However, 
Wyoming encourages EPA to approve Wyoming's LTS for Jim Bridger Units 1 
and 2 as submitted, rather than approve only the SCR portion, in order 
to preserve future flexibility for ensuring adequate emission controls.
    Response: We agree with this comment to the extent that the 
regulatory requirements we are adopting for monitoring, recordkeeping, 
and reporting only require that Jim Bridger Units 1 and 2 meet an 
emission limit of 0.07 lb/MMBtu on a 30-day rolling average. Our 
regulatory language does not require PacifiCorp to install SCR to meet 
these limits. EPA is approving Wyoming's LTS for Jim Bridger Units 1 
and 2 as submitted.
    Comment: EPA acted arbitrarily by not evaluating SNCR for the Jim 
Bridger units. EPA's proposed regional haze FIP is defective because 
EPA did not follow the BART Guidelines when conducting a five-factor 
analysis for potential BART NOX controls. As the BART 
Guidelines explain, states (and EPA when it substitutes itself for the 
state) must evaluate ``the control effectiveness of all the technically 
feasible control alternatives. . . .'' Here, EPA failed to do so by not 
analyzing SNCR for the Jim Bridger plant. In fact, EPA admits it did 
not conduct a full BART analysis for SNCR for the Jim Bridger units: 
``Because of our examination of the factors lead us to propose SCR as 
reasonable for BART, we have eliminated SNCR for further 
consideration.''
    Response: We disagree with this comment. Our proposed rulemaking 
notice clearly shows that we considered SNCR in our analysis (see Table 
9 and Table 11 of the proposed rulemaking action).
    Comment: One commenter stated that our proposed rule creates 
unnecessary regulatory uncertainty by saying we propose to approve the 
State's compliance deadlines for Jim Bridger Units 1 and 2, but then go 
on to say we are seeking comment on a 2017 compliance deadline. They go 
on to say that EPA must state unequivocally that they approve of the 
State's existing compliance deadlines for Jim Bridger Units 1 and 2.
    Response: We have responded to this comment in previous responses.
    Comment: We received numerous general comments in favor of our 
proposed approval of the State's SIP for the Jim Bridger Units 1-4.
    Response: See responses to other comments above.
    Comment: PacifiCorp supports EPA's proposed action to afford 
``considerable deference'' to the Wyoming SIP with respect to what 
controls are reasonable and when they should be implemented at Jim 
Bridger Units 1 and 2, and that it would be unreasonable to require any 
further retrofits at this source within five years of EPA's final 
action. This is especially true given the extremely limited visibility 
improvement that would be achieved if SCRs were installed within the 
BART time period at Jim Bridger Units 1 and 2.
    Further, PacifiCorp does not believe EPA, having reached the 
conclusion that it would be unreasonable to require further retrofits 
at Jim Bridger within five years, can reverse its decision simply by 
inviting comment on an alternative proposal without further 
consideration of the broader impacts of forcing more aggressive 
controls within a five-year period.
    Response: We have responded to this comment in previous responses.
    Comment: We received numerous comments that an earlier compliance 
deadline for the installation of SCR at Jim Bridger Units 1 and 2 would 
be a significant burden and would be costly to PacifiCorp consumers.
    Response: We have responded to this comment in other responses to 
comments.
    Comment: We agree with EPA that SCR represents BART for all four 
Bridger units, but recommend a lower 30-day rolling average emission 
limit (e.g. 0.06 lb/MMBtu) to reflect the true capabilities of SCR.
    Response: We have addressed the control effectiveness of SCR above 
in section V.C.3.
    Comment: We received comments that BART for NOX 
emissions at Jim Bridger Units 1-4 must be based on SCR and LNBs/SOFA, 
which represents the best system of continuous emissions reduction and 
that commenters agree with EPA's proposal to require this technology as 
BART. Commenters went on to state that EPA must revise its BART-based 
NOX emission limit for Units 1-4 from 0.07 lb/MMBtu to no 
higher than 0.05 lb/MMBtu, which the selected technology can easily 
achieve.
    Response: We have addressed the control effectiveness of SCR above 
in section V.C.3.
b. PM BART Determination
    Comment: The fabric filter option discussed by Wyoming represents 
PacifiCorp's estimate that application of a Compact Hybrid Particulate 
Collector unit in addition to using flue gas conditioning with the 
existing electrostatic precipitators can reduce emissions an additional 
50% resulting in a PM10 emission rate of 0.015 lb/MMBtu. 
Considering that EPA Region 9 proposed that the Desert Rock power plant 
meet 0.010 lb/MMBtu, we believe that the Compact Hybrid Particulate 
Collector option could achieve the same limit.
    Response: See our response to a similar comment in section IV.C.6 
of this rulemaking.
    Comment: Neither Wyoming nor EPA completed the five-step BART 
process for PM10 emissions. EPA asserted that: ``The State 
did not provide visibility improvement modeling for fabric filters, but 
EPA is proposing to conclude this is reasonable based on the high cost 
effectiveness of fabric filters at each of the units. In addition, we 
anticipate that the visibility improvement that would result from 
lowering the limit from 0.03 lb/MMBtu to 0.015 lb/MMBtu would be 
insignificant based on the State's analysis.''
    We have several concerns with these conclusions: (1) EPA cannot 
simply abort the five-step process once it has determined a technology 
to be technically feasible; (2) EPA has overlooked the environmental 
impact of SO3 emissions that may be released as a result of 
PacifiCorp's FGC BART proposal; (3) Wyoming has underestimated the 
effectiveness of the fabric filter option; and (4) Wyoming's fabric 
filter costs are overestimated. For example, the cost estimates used by 
Wyoming contained escalation, extra contingencies, and AFUDC, which are 
not allowed by the CCM and have been rejected by EPA in other analyses. 
The total for these improper costs exceeds $7 million per fabric 
filter.
    Even taken at face value, the cost/ton deemed ``high'' by EPA for 
Units 2 and

[[Page 5190]]

3 are similar to or lower than cost/ton values accepted as reasonable 
(for NOX) by states and by EPA in other analyses. EPA should 
complete a proper five-step PM10 BART analysis by re-
evaluating the Compact Hybrid option on the basis of its ability to 
achieve a lower limit (e.g., 0.010 lb/MMBtu), evaluating costs in 
accordance with the BART Guidelines, comparing its cost-effectiveness 
to other baghouse installations to properly assess the 
``reasonableness'' of its cost, and determining the degree of 
visibility improvement that would result from a lower PM10 
limit.
    Response: See our response to a similar comment in section IV.C.6 
of this rulemaking.
3. Dave Johnston Units 3 and 4
a. NOX BART Determination
    Comment: Wyoming has underestimated the ability of SCR to reduce 
emissions. In estimating the annual cost-effectiveness of the 
LNB+OFA+SCR option, Wyoming assumed 0.07 lb/MMBtu on an annual average 
basis. Based on the 0.28 lb/MMBtu NOX emission rate 
predicted for the LNB+OFA option, and the 0.23 lb/MMBtu annual emission 
rates demonstrated by Johnston Unit 3 in 2011, outlet emissions at 0.07 
lb/MMBtu represent only a 70%-75% SCR control efficiency as opposed to 
the generally-accepted 90%. Based on the 0.15 lb/MMBtu NOX 
emission rate predicted for the LNB+OFA option, outlet emissions at 
0.07 lb/MMBtu represent only a 53% SCR control efficiency on Unit 4. 
Wyoming has not provided any documentation or justification to support 
the higher emission rates used in its analyses. In other recent BART 
actions, EPA has determined that SCR can achieve 0.05 lb/MMBtu on an 
annual basis. Such an underestimate at Johnston biases the cost-benefit 
analysis against SCR and is inconsistent with other EPA analyses.
    Response: The commenter has incorrectly assumed that a 90% control 
efficiency can be achieved in all SCR applications regardless of the 
input NOX emission rate or other parameters. In addition, we 
note that the emission rate analyzed by Wyoming, 0.07 lb/MMBtu, was on 
a 30-day rolling average basis, not an annual basis. Regardless, we 
agree that SCR can in most cases achieve a performance rate of 0.05 lb/
MMBtu on an annual basis. (See section IV.C.4 of this rulemaking for 
more information on the control effectiveness of SCR). We have revised 
the SCR costs for Dave Johnston Units 3 and 4 accordingly.
    Comment: Wyoming has overestimated the cost of SCR. A survey of 
industry SCR cost data (conducted for the Utility Air Regulatory Group) 
and EPA IPM estimates show that typical SCR costs for units the size of 
the Johnston units would be $180-$300/kW. Wyoming's cost estimates for 
SCR on Units 3 and 4 are $488 and $436/kW, respectively, which exceed 
real-world industry costs ($50-$300/kW) and industry estimates, leading 
us to believe that capital and annual costs are overestimated.
    Response: See our response regarding the cost of SCR in section 
IV.C.5 of this rulemaking.
    Comment: Neither PacifiCorp nor Wyoming provided justification or 
documentation for their cost estimates. We were not provided with any 
vendor estimates or bids, and PacifiCorp and Wyoming did not use the 
CCM. For example, the cost estimates used by Wyoming contained AFUDC, 
which is not allowed by the CCM and has been rejected by EPA in other 
analyses. The total for these improper costs exceeds $13 million.
    Response: In order to address the cost analysis deficiencies noted 
by the commenter, EPA has performed revised cost analyses for Dave 
Johnston Units 3 and 4. In our revised cost analyses, we have followed 
the structure of the CCM, though we have used the IPM cost models to 
estimate direct capital costs and O&M costs.
    Comment: Dave Johnston Unit 4 could very likely achieve a 
NOX rate as low as 0.03 lb/MMBtu, which reflects 80% 
NOX control across the SCR. A lower NOX emission 
limit would increase the cost of the total system, but the cost 
effectiveness of the system is actually improved because of the greater 
NOX removal. The cost effectiveness of SCR plus LNBs/OFA at 
Dave Johnston Unit 4 to meet a 0.03 lb/MMBtu NOX rate would 
be $1,803/ton of NOX removed. EPA should require Dave 
Johnston Unit 4 to install SCR plus LNBs/OFA to meet a NOX 
rate of 0.03 lb/MMBtu or, at worst, no higher than 0.05 lb/MMBtu.
    Response: We have addressed the control effectiveness of SCR above 
in section V.C.3. Again, we have not selected LNBs with OFA and SCR for 
Dave Johnston Unit 4 due to the high incremental cost effectiveness, 
when considered within the context of the five factors.
    Comment: EPA's use of undefined incremental cost effectiveness 
versus incremental visibility benefit threshold is arbitrary in concept 
and in its application. It is arbitrary in concept because EPA has not 
provided any reasoned basis for its approach let alone disclosed the 
threshold it applies. It is arbitrary in application, because in the 
case of Dave Johnston Unit 4, the visibility benefits of SCR do justify 
its cost, as EPA has found for other units.
    Response: We disagree. We have made our determination based on a 
weighing of the five factors. One of the factors to be considered is 
cost-effectiveness, both average and incremental. For Dave Johnston 
Unit 4, we have determined that the incremental costs, when considered 
with the other BART factors, does not make the selection of SCR 
reasonable.
    Comment: The EPA failed to support its conclusion that SNCR, rather 
than SCR, is BART for Dave Johnston Unit 4. The EPA's sole basis for 
rejecting SCR as BART for Dave Johnston Unit 4 was the incremental cost 
effectiveness of the control, which EPA estimated to be $11,951, but 
the EPA has not supported this line-drawing with reference to the 
statutory BART factors or purpose of the regional haze program. Without 
providing objective standards or rationale to support its 
determination, the EPA's judgment that the incremental cost 
effectiveness of SCR on Dave Johnston Unit 4 is too high appears 
arbitrary.
    The EPA's analysis of incremental cost effectiveness for Dave 
Johnston Unit 4 also cannot drive the Agency's BART determination 
because EPA overestimated the cost of SCR for Unit 4. The EPA found the 
average cost effectiveness of SCR to be reasonable for the unit, but 
the costs are even lower than the EPA assumed because the EPA used 
unjustified ``retrofit factors,'' interest rate, and auxiliary power 
costs.
    Response: As stated in our response above, we have made our 
determination based on a weighing of the five factors. One of the 
factors to be considered is cost-effectiveness, both average and 
incremental. For Dave Johnston Unit 4, we have determined that the 
incremental costs, when considered with the other BART factors, does 
not make the selection of SCR reasonable.
    In addition, we have revised the costs of SCR for Dave Johnston 
Unit 3 in support of our final rulemaking. The revised costs are no 
longer based on a retrofit factor, but instead are based on a budgetary 
price from an equipment vendor submitted by PacifiCorp during the 
comment period. We feel that use of the vendor data for SCR provides a 
more accurate capital cost than when using a retrofit factor. For 
reasons described in separate responses, we continue to find that use 
of the social discount rate of 7% is appropriate for regulatory 
applications such as BART determinations. We have corrected the

[[Page 5191]]

auxiliary power costs to reflect busbar, and not market price of power; 
however this has a small affect on overall costs. Our cost methodology 
and assumptions are described in detail in the cost report that can be 
found in the docket.
    Comment: SCR on Dave Johnston Unit 4 would likely result in even 
greater NOX emission reductions than EPA assumed, further 
undermining the Agency's exclusive reliance on incremental cost 
effectiveness to reject SCR as BART. An emission rate lower than 0.05 
lb/MMBtu on a 30-day average is achievable for Dave Johnston Unit 4 
because this unit operates combustion controls that independently 
control NOX emissions to a 30-day average of 0.13 lb/MMBtu 
(compared with the 0.22 lb/MMBtu NOX-emission rate achieved 
at Unit 3). At this emission level, SCR would only need to remove 66.4 
percent of NOX emissions to achieve an emission rate of 0.05 
lb/MMBtu, while SCR is capable of achieving NOX reductions 
of 90 percent. The EPA should have evaluated the cost-effectiveness of 
SCR on Dave Johnston Unit 4 based on even greater NOX 
emissions reductions that are readily achievable.
    Response: We disagree. First, the commenter has incorrectly assumed 
that a 90 percent reduction in NOX is achievable with SCR 
regardless of inlet rate or other parameters. In most cases, SCRs are 
designed for a performance emission rate, such as in lb/MMBtu, and not 
the anticipated percent reduction. In the case of Dave Johnston Unit 4, 
the low emissions currently being achieved with combustion controls are 
not an indication that the SCR would achieve greater reductions than 
estimated by EPA. In fact the exact opposite is true: the lower the 
inlet rate to the SCR, the less NOX that will be removed as 
there are simply fewer tons to remove. Finally, we note that we have 
revised our cost calculations to support the determinations in today's 
final rule. In our revised analysis, we calculate that the incremental 
cost effectiveness of SCR is $13,312/ton, as opposed to $11,951/ton. 
This reinforces our conclusion that SCR is not appropriate for Dave 
Johnston Unit 4. Our cost methodology and assumptions are described in 
detail in the cost report that can be found in the docket.
b. Alternative Control Technology Proposal
    Comment: Dave Johnston Unit 3 was retrofitted with LNB and 
separated OFA in the spring of 2010, and Unit 4 was retrofitted with 
the same technology in early 2009. EPA recognizes that Unit 3 has a 
current annual NOX emission rate of about 0.22 lb/MMBtu, and 
Unit 4 has a rate of about 0.14 lb/MMBtu. The potential additional 
NOX controls that may be added to these units include SNCR 
and SCR. Should an alternate control technology be considered by EPA 
for Dave Johnston Unit 3, SNCR is preferable to SCR for Dave Johnston 
Unit 3 when considering all currently available information and the 
current emissions performance of the unit.
    Even though the cost of SNCR is unacceptably high for Unit 3 
(approximately $5,500 per ton NOX removed), it is still far 
less than the tremendously expensive cost of SCR ($15,769 per ton 
NOX removed for Unit 3), particularly when taking into 
account the incrementally small modeled visibility improvement between 
the technologies.
    Response: As described in section III.B of this document, we have 
re-evaluated the cost of compliance for Dave Johnston unit 3 to reflect 
a shorter remaining useful life (9 years as opposed to 20 years) 
because PacifiCorp has volunteered to install SNCR and retire the unit 
in 2027 in lieu of installing SCR under our proposed rule. As we 
explain there, our revised BART analysis shows that neither SNCR nor 
SCR is reasonable over this shorter remaining useful life. However, our 
analysis continues to support a conclusion that SCR is warranted if the 
costs of compliance are calculated over a 20-year remaining useful 
life. Therefore, we have also included an to give PacifiCorp the option 
to meet a 0.07 lb/MMBtu emission limit (assumes installation of SCR) 
within five years of today's action instead of shutting down the unit.
    Comment: With respect to Dave Johnston Unit 4, EPA has concluded 
that SNCR is BART for that unit. As such, PacifiCorp has only provided 
updated SNCR information for Unit 4, considering all currently 
available information and the current emissions performance of the 
unit. The cost of SNCR for Unit 4 is unacceptably high and not cost 
effective (approximately $12,000 per ton NOX removed). The 
alternate control technology for Dave Johnston Unit 4 would be LNB/OFA, 
as is currently installed today.
    Response: We disagree with the cost effectiveness estimates 
provided by the commenter. Nonetheless, as described above, in 
consideration of comments received during the public comment period, we 
have re-evaluated our cost of compliance estimates for Dave Johnston 
Unit 4. Our revised costs, when taken along with the remaining BART 
factors, no longer show that SNCR is warranted for Dave Johnston Unit 
4. Therefore, we agree that BART for Dave Johnston Unit 4 is the 
currently installed combustion controls (LNB/OFA).
4. Naughton Units 1-3
a. NOX BART Determination
    Comment: The EPA proposes to accept Wyoming's SIP proposal to 
identify SCR as BART on Naughton Unit 3 and to require SCR to reduce 
NOX emissions from Naughton Unit 1 and 2. The EPA properly 
recognized that the costs of SCR on Naughton Units 1-3 are reasonable. 
However, the EPA's estimate of costs is significantly inflated, and the 
true costs are even lower than EPA found because the EPA used 
unjustified ``retrofit factors,'' interest rate, and auxiliary power 
costs. The recalculation of costs using what the commenter believes are 
corrected inputs resulted in significantly lower SCR costs of $1,501 to 
$1,788/ton at all three Naughton units. On this basis, SCR is very cost 
effective on these units and at the low end of the cost threshold when 
scanning NOX reduction costs elsewhere. SCR also is 
justified by the visibility benefits it would afford, which 
additionally supports EPA's findings that SCR reflects BART for 
Naughton Units 1-3.
    Response: EPA believes that the retrofit factors used in cost 
estimates were reasonable and has described in detail the reasoning for 
the retrofit factors in other responses. EPA also discussed the 
reasoning for the assumed interest rate in responses to other comments. 
EPA has revised its cost estimates and has made changes where EPA 
believed that input from commenters justified changes.
    Comment: EPA should evaluate the feasibility and cost-effectiveness 
of further NOX reductions that could be achieved by a more-
efficient SCR. EPA is proposing that the FIP NOX BART 
emission limit for Naughton Unit 1, Unit 2, and Unit 3 is 0.07 lb/MMBtu 
(30-day rolling average). While we are generally pleased with EPA's 
proposal, we note that EPA's analysis is based on only 76% 
NOX control by the SCRs on Units 1 and 2, and 85% control by 
the SCR on Unit 3. This still results in Unit 2 contributing 0.5 
deciviews and Unit 3 contributing 0.9 deciviews to visibility 
impairment at Badlands National Park.
    Response: We have addressed the control effectiveness of SCR above 
in section V.C.3 above.
    Comment: A NOX limit of 0.07 lb/MMBtu on a rolling 30-
boiler operating day basis would only require Naughton Units 1 and 2 to 
achieve 71 percent NOX removal and Naughton Unit 3 to 
achieve 80 percent NOX removal across the SCR system based 
on an evaluation of

[[Page 5192]]

available CAMD emissions data. A 0.05 lb/MMBtu NOX emission 
limit applicable on a rolling 30-boiler operating day average basis 
would require 79 percent NOX removal across the SCR at 
Naughton Units 1 and 2 and 85.7 percent NOX removal at 
Naughton Unit 3, which is achievable.
    Response: We have addressed the control effectiveness of SCR above 
in section V.C.3 above.
    Comment: We received comments that Wyoming has underestimated the 
ability of SCR to reduce emissions. Commenters stated that in 
estimating the annual cost-effectiveness of the LNB+OFA+SCR option, 
Wyoming assumed 0.07 lb/MMBtu on an annual average basis. Based on the 
0.026-0.37 lb/MMBtu NOX emission rate predicted for the 
combustion control option, outlet emissions at 0.07 lb/MMBtu represent 
only 73%-81% SCR control efficiency as opposed to the generally 
accepted 90%. Commenters went on to point out that in other recent BART 
actions, EPA has determined that SCR can achieve 0.05 lb/MMBtu on an 
annual basis.
    Response: The commenters have incorrectly assumed that a 90% 
control efficiency can be achieved in all SCR applications regardless 
of the input NOX emission rate or other parameters. In 
addition, we note that the emission rate analyzed by Wyoming, 0.07 lb/
MMBtu, was on a 30-day rolling average basis, not an annual basis. 
Regardless, we agree that SCR can in most cases achieve a performance 
rate of 0.05 lb/MMBtu on an annual basis. (See section IV.C.4 of this 
rulemaking for more information regarding the control effectiveness of 
SCR). We have revised the SCR costs for the Naughton units accordingly.
    Comment: A survey of industry SCR cost data (conducted for the 
Utility Air Regulatory Group) and IPM estimates show that typical SCR 
costs for units the size of the Naughton units would be $280-$330/kW. 
Wyoming's cost estimates for SCR are $412-$531/kW, which exceed real 
world industry costs ($50-$300/kW) and industry estimates, leading us 
to believe that capital and annual costs are overestimated.
    Response: See our response regarding the cost of SCR in the section 
V.B.2 of this rulemaking.
    Comment: Neither PacifiCorp nor Wyoming provided justification or 
documentation for their cost estimates. We were not provided with any 
vendor estimates or bids, and PacifiCorp and Wyoming did not use the 
CCM. For example, the cost estimates used by Wyoming contained AFUDC 
which is not allowed by the CCM and has been rejected by EPA in other 
analyses. The total for these improper costs exceeds $17 million.
    Response: In our revised SCR cost analysis for the Naughton units, 
we followed the framework of the CCM (although we derived direct 
capital costs and O&M costs using the more recent approach found in the 
IPM cost models). For example, we did not allow for owner's costs and 
AFUDC. Therefore, we have addressed the concerns raised by the 
commenter.
b. Alternative Control Technology Proposal
    Comment: EPA requested additional information on the conversion of 
Naughton Unit 3 from a coal fired unit to a natural gas fired unit. 78 
FR 34760. EPA must evaluate PacifiCorp's fuel conversion in accordance 
with Appendix Y as a ``better-than-BART'' alternative and not as a BART 
control technology option because EPA had made clear in its BART 
Guidance that ``it is not [EPA's] intent to direct States to switch 
fuel forms, e.g. from coal to gas,'' as part of the BART analysis. 70 
FR 39104, 39164. PacifiCorp voluntarily submitted its permit 
application to convert Naughton Unit 3 to natural gas, the State issued 
a federally enforceable permit requiring such conversion. Compliance 
with the permit is therefore not voluntary.
    The permitted NOX performance level of Naughton Unit 3 
after conversion to natural gas is 0.08 lb/MMBtu based on a 30-day 
rolling average and not 0.10 lb/MMBtu based on a 30-day rolling average 
as stated in PacifiCorp's permit application. Additionally, the 
permitted NOX mass emission rate is 250 lb/hr based on a 30-
day rolling average, which is protective of visibility and lower than 
the BART-determined NOX rate of 259 lb/hr based on the same 
averaging period. Finally, annual NOX emissions will be 
reduced from the BART level of 1,134 tons to 519 tons.
    Response: We tentatively agree that the conversion of Naughton Unit 
3 to natural gas is better-than-BART for that individual unit, however, 
the State has not provided a SIP for EPA's action on this option. EPA 
does not have the authority to approve the conversion without a SIP 
submittal, and is, therefore, approving the State's BART determination 
for SCR at Naughton Unit 3 without making a final determination on 
whether the conversion is better-than-BART for that unit. In lieu of 
our approval of the State's BART determination for Naughton Unit 3, EPA 
is committed to take expedited action on a future SIP revision for 
Naughton Unit 3 reflecting the conversion if the State submits such a 
revision. That action would constitute our final determination on the 
conversion.
    Comment: Rather than install the control equipment required by the 
Wyoming SIP, PacifiCorp will convert the unit to fire natural gas by 
the end of 2017. A construction permit allowing the conversion has been 
issued by Wyoming, and PacifiCorp is moving ahead with a request for 
Wyoming to modify the Wyoming SIP to accommodate this change. The 
construction permit issued by Wyoming requires Naughton Unit 3 to cease 
burning coal by December 31, 2017 and to be retrofitted to natural gas 
as its fuel source by June 30, 2018. PacifiCorp requests that EPA's 
final FIP include this compliance alternative for Naughton Unit 3.
    Response: See our response above.
    Comment: The Conservation Organizations support the Naughton Unit 3 
conversion to natural gas within the first five-year regional haze 
planning period as a better-than-BART alternative to installation of 
SCR on Unit 3. We recognize that a gas conversion will virtually 
eliminate SO2 emissions as well as greatly reduce 
NOX and PM emissions resulting in significant visibility 
benefits.
    However, to the extent that EPA is considering whether the Naughton 
Unit 3 is better than BART as proposed for all three Naughton Units 
(i.e., whether the conversion may be approved ``instead of . . . BART 
as proposed'' for Naughton Units 1 and 2, 78 FR 34783), the 
Conservation Organizations object. The Conservation Organizations 
conducted visibility modeling to determine whether PacifiCorp's 
proposed natural gas conversion at Unit 3 (with LNB and OFA at Units 1 
and 2) would result in greater visibility improvement than would EPA's 
re-proposed BART alternative of SCR at all three Naughton Units. The 
analysis shows that EPA's re-proposed SCR BART determination 
consistently results in greater visibility improvement over the gas 
conversion scenario. Thus, the conversion of Naughton Unit 3 to gas 
with LNB/OFA on Units 1 and 2 does not satisfy the ``better-than-BART'' 
standards of the regional haze regulations. Whether or not PacifiCorp 
converts Naughton Unit 3 to natural gas, EPA must require the 
installation of SCR to meet an emission limit of 0.05 lb/MMBtu to 
satisfy BART for Naughton Units 1 and 2.
    Response: We acknowledge the support for a natural gas conversion 
as a better-than-BART alternative for Naughton Unit 3. If the State 
submits a SIP revision reflecting the conversion,

[[Page 5193]]

we will take expedited action on it. As discussed elsewhere, we are 
approving the State's SIP submittal for all Naughton Units based on our 
consideration of the five BART factors. The remainder of the comment is 
therefore not relevant.
5. Wyodak
    Comment: Wyoming has underestimated the ability of SCR to reduce 
emissions. In estimating the annual cost-effectiveness of the 
LNB+OFA+SCR option, Wyoming estimated 0.07 lb/MMBtu on an annual 
average basis. Based on the 0.18 lb/MMBtu NOX emission rate 
predicted for the LNB+OFA option, outlet emissions at 0.07 lb/MMBtu 
represent only a 61% SCR control efficiency as opposed to the 
generally-accepted 90%. Wyoming has not provided any documentation or 
justification to support the higher emission rates used in its 
analyses. In other recent BART actions, EPA has determined that SCR can 
achieve 0.05 lb/MMBtu on an annual basis. Such an underestimate at 
Wyodak biases the cost-benefit analysis against SCR and is inconsistent 
with other EPA analyses.
    Response: The commenter has incorrectly assumed that a 90% control 
efficiency can be achieved in all SCR applications regardless of the 
input NOX emission rate or other parameters. In addition, we 
note that the emission rate analyzed by Wyoming, 0.07 lb/MMBtu, was on 
a 30-day rolling average basis, not an annual basis. Regardless, we 
agree that SCR can in most cases achieve a performance rate of 0.05 lb/
MMBtu on an annual basis. (See section IV.C.4 of this rulemaking for 
more information regarding the control effectiveness of SCR). We have 
revised the SCR costs for Wyodak accordingly.
    Comment: A survey of industry SCR cost data and EPA IPM estimates 
show that typical SCR costs for units the size of Wyodak would be $180-
$280/kW. Wyoming's cost estimates for SCR are $474/kW, which exceed 
real-world industry costs ($50-$300/kW) and industry estimates, leading 
us to believe that capital and annual costs are overestimated.
    Response: See our response regarding the cost of SCR in section 
IV.C.5 of this rulemaking.
    Comment: Neither PacifiCorp nor Wyoming provided justification or 
documentation for their cost estimates. We were not provided with any 
vendor estimates or bids, and PacifiCorp and Wyoming did not use the 
CCM. For example, the cost estimates used by Wyoming contained AFUDC, 
which is not allowed by the CCM and has been rejected by EPA in other 
analyses. The total for these improper costs exceeds $8 million.
    Response: In our revised SCR cost analysis for Wyodak, we followed 
the framework of the CCM (although we derived direct capital costs and 
O&M costs using the more recent approach found in the IPM cost models). 
For example, we did not allow for owner's costs and AFUDC. Therefore, 
we have addressed the concerns raised by the commenter.
    Comment: The addition of SCR at Wyodak should be required because 
it is consistent with the other BART determinations EPA has made. EPA 
is proposing that the FIP NOX BART is new LNBs with OFA plus 
SNCR at an emission limit of 0.17 lb/MMBtu. EPA proposes to eliminate 
new LNBs with advanced OFA plus SCR. The cumulative cost effectiveness 
of adding SCR to Wyodak is equivalent to EPA's accepted values at 
Laramie River Unit 2. Based upon cost and visibility improvement, we 
believe that SCR is BART for Wyodak. As EPA stated in its notice, 
``cost-effectiveness and visibility improvement are within the range of 
other EPA FIP actions.'' Even though cumulative visibility improvement 
is relatively low, so are SCR costs.
    A NOX limit of 0.07 lb/MMBtu on a rolling 30-boiler 
operating day basis would only require Wyodak to achieve 67 percent 
NOX removal across the SCR system based on an evaluation of 
available CAMD emissions data. A 0.05 lb/MMBtu NOX emission 
limit applicable on a rolling 30-boiler operating day average basis 
would require 76 percent NOX removal across the SCR, which 
the commenter believes is achievable.
    Response: As a result of other comments we have received, we are 
finalizing a NOX BART determination of new LNBs with OFA 
plus SCR for Wyodak. We agree with the portion of this comment that 
states this is consistent with other EPA BART determinations. We 
disagree with the remainder of the comment. As we have discussed in 
other responses, we are not required, nor have we chosen to, use the $/
deciview metric, let alone the same on a cumulative basis, when 
assessing BART.
    We have addressed the control effectiveness of SCR above in section 
V.C.3.
    Comment: For Wyodak, EPA is proposing that the FIP NOX 
BART is new LNBs with OFA plus SNCR at an emission limit of 0.17 lb/
MMBtu. EPA proposes to eliminate new LNBs with advanced OFA plus SCR 
because: ``Although the cost-effectiveness and visibility improvement 
are within the range of other EPA FIP actions, we find that the 
cumulative visibility improvement of 1.16 deciviews for new LNBs with 
OFA plus SCR is low compared to the cumulative visibility benefits that 
will be achieved by requiring SCR at Dave Johnston Unit 3 (2.92 
deciview), Laramie River Unit 1 (2.12 deciview), Laramie River Unit 2 
(1.97 deciview), Laramie River Unit 3 (2.29 deciview), Naughton Unit 1 
(3.54 deciview), and Naughton Unit 2 (4.18 deciview).''
    Because the cumulative visibility improvement from EPA's proposed 
control strategy is barely half of the visibility improvement that EPA 
rejected as ``low,'' then visibility improvement cannot be the only 
factor relied upon by EPA in making its BART determination. We can only 
conclude that EPA is somehow relating visibility improvement to another 
factor. For example, after correcting for the unsupported 1.3 retrofit 
factor at this relatively simple, single-EGU facility, the cost-
effectiveness of adding SCR is $16 million/deciview at Wind Cave 
National Park, and $10 million/cumulative deciview. By comparison, 
based upon EPA estimates, addition of SCR to Laramie River Unit 3 
results in $28 million/deciview at the most-impacted Class I area, and 
addition of SCR to Laramie River Unit 2 yields $10 million/cumulative 
deciview. The cumulative cost effectiveness of adding SCR to Wyodak is 
equivalent to EPA's accepted values at Laramie River Unit 2.
    Response: We disagree with the suggested use of the $/deciview 
metric. As we have discussed in other responses, we are not required, 
nor have we chosen to, use the $/deciview metric, let alone the same on 
a cumulative basis, when assessing BART. Even if we had, the 
commenter's position is predicated on their assertion that EPA 
inappropriately applied a retrofit factor for SCR at Wyodak. As we have 
discussed in other responses, we disagree that it was inappropriate to 
apply a retrofit factor of 1.3. However, as explained below, we agree 
that we should not have relied on the basis stated in our proposal to 
reject SCR. For Wyodak, we find that the visibility improvements at two 
Class I areas, when weighed with the other BART factors, makes SCR 
reasonable as BART.
    Comment: Based upon cost and visibility improvement, we believe 
that SCR is BART for Wyodak. Under the EPA proposal, Wyodak would still 
contribute over 0.7 deciview impairment at Wind Cave National Park (and 
exceed 0.5 deciviews at Badlands National Park). With the addition of 
SCR, impairment would drop to less

[[Page 5194]]

than 0.5 deciviews at all Class I areas. As EPA stated in its proposal, 
``cost-effectiveness and visibility improvement are within the range of 
other EPA FIP actions.'' Even though cumulative visibility improvement 
is relatively low, so are SCR costs. Addition of SCR at Wyodak should 
be required because it is consistent with the other BART determinations 
EPA has made here.
    Response: After further consideration, we agree that it was 
inappropriate for EPA to reject SCR as BART for Wyodak based on the 
rationale that Wyodak's emissions affect fewer Class I areas than 
Wyoming's other BART-eligible sources. Where consideration of the five 
factors demonstrates that a control is reasonable in light of the 
visibility improvement that will occur at the most impacted Class I 
area, as was the case here for Wyodak and Wind Cave, the fact that 
additional Class I areas will also experience visibility improvement 
can only bolster the case for that control's selection, not undermine 
it. In other words, the fact that Wyodak's emissions affect two Class I 
areas instead of six or seven is irrelevant if the improvement at just 
one Class I area is sufficient to warrant a control's selection as 
BART. Consequently, we have reassessed the five factors for Wyodak and 
now conclude, even after taking into account our revised cost estimates 
and visibility modeling, that LNB/OFA + SCR is NOX BART for 
Wyodak Unit 1.
    Comment: SCR with an emission limit of 0.05 lb/MMBtu should be 
required as NOX BART for Wyodak, rather than an SNCR-based 
limit of 0.17 lb/MMBtu as EPA proposes. EPA properly recognized that 
the cost-effectiveness and incremental cost-effectiveness of SCR to 
reduce Wyodak's NOX emissions are reasonable, but 
nonetheless proposed to reject SCR on the basis of purportedly 
insufficient cumulative visibility benefits. EPA's proposed 
determination is improper, because EPA has failed to justify why 
incremental visibility benefits over the large number of Class I Areas 
impacted by Wyodak's NOX emissions should not be required to 
achieve reasonable progress toward the national visibility goal, 
particularly in light of the fact that none of the Wyoming Class I 
areas affected by Wyodak's NOX emissions are projected to 
achieve the Uniform Rate of Progress (URP) in 2018. Moreover, while EPA 
evaluated the impacts of Wyodak's NOX emissions only at Wind 
Cave and Badlands National Parks, our supplemental modeling shows that 
SCR to control Wyodak's NOX emissions will nearly eliminate 
the plant's perceptible visibility impacts at 18 Class I areas. EPA's 
visibility justification for rejecting SCR as BART was improper because 
Congress has directed EPA to require BART ``for the purpose of 
eliminating or reducing any [visibility] impairment'' caused by the 
source. 42 U.S.C. 7491(b)(2)(A). Installing SCR at Wyodak would resolve 
this impairment.
    Response: See our response to the previous comment. While we do not 
agree with the commenter's assertion that Wyodak's emissions have 
perceptible visibility impacts at 18 Class I areas or that the URP is 
relevant for purposes of a BART determination, we do agree that our 
decision to eliminate SCR based on cumulative visibility improvement 
was improper.
    Comment: EPA properly recognized that the cost-effectiveness and 
incremental cost-effectiveness of SCR to reduce Wyodak's NOX 
emissions are reasonable, but nonetheless proposed to reject SCR on the 
basis of purportedly insufficient cumulative visibility benefits. EPA's 
estimate of costs is significantly inflated, and the true costs are 
lower than EPA found because the EPA used unjustified ``retrofit 
factors,'' interest rate, and auxiliary power costs. On these bases, 
the EPA should require SCR at Wyodak as BART.
    Response: See our previous two responses.
6. Trona Mines
a. FMC Westvaco and General Chemical Green River
    Comment: EPA should reconsider whether SCR plus combustion controls 
is BART for the FMC Westvaco Units NS-1A and NS-1B. At $3,493/ton, as 
presented by EPA, SCR may be a cost-effective option. Furthermore, EPA 
should evaluate whether the cost of SCR for FMC Westvaco Units NS-1A 
and NS-1B were calculated correctly.
    Response: Although EPA has not re-evaluated the cost of SCR at the 
FMC Westvaco Units, we note the relatively low visibility improvement 
from SCR for each unit (0.24 deciviews). Because of the low visibility 
improvement from SCR, we do not find that a reconsideration of costs 
would necessarily have led EPA or the State to a different conclusion 
regarding the selection of SCR.
    Comment: At a minimum, EPA must require SNCR and LNB + SOFA as BART 
for NOX at the Westvaco plant. EPA determined that this 
enhanced technology could achieve a 0.21 lb/MMBtu NOX 
emissions rate. This would result in a 70% reduction in NOX 
emissions from current levels, rather than just a 50% reduction that 
would result from the 0.35 lb/MMBtu emission rate currently proposed. 
Requiring SNCR would lead to NOX emissions reductions of 
1,903 tpy. SNCR in addition to LNB+SOFA is highly cost effective at 
$673/ton. This is well within the range of BART costs that EPA has 
found reasonable for SNCR at other facilities, including facilities in 
Wyoming. For example, EPA proposes to reject Wyoming's NOX 
BART proposal for Wyodak Unit 1, and instead to require LNB+OFA+SNCR as 
BART, finding the technology cost effective at $958/ton, a higher cost 
than the same technology at the Westvaco boilers.
    Response: We disagree with this comment. First, SNCR at each of the 
units would achieve a 0.19 deciview improvement, with an incremental 
visibility improvement of 0.06 deciviews. The cost effectiveness for 
LNBs compared to LNBs with SNCR is more than double ($304/ton compared 
to $673/ton). Based on this information, we find it reasonable for the 
State not to determine SNCR is BART for these units based on a 
consideration of the five factors, including the visibility 
improvement.
    Comment: Requiring SNCR at the FMC Westvaco plant would improve 
visibility at affected Class I areas. EPA states that Wyoming's 
visibility modeling for this facility demonstrated a 0.19 deciview 
improvement at the Bridger Wilderness Area from the installation of 
SNCR on each boiler. In fact, Wyoming's modeling demonstrated a 0.198 
deciview visibility improvement for the maximum 98th percentile impact 
at Bridger Wilderness Area. The combined visibility improvement due to 
SNCR at both Westvaco boilers is nearly 0.4 deciviews at the Bridger 
Wilderness Area alone. EPA found it appropriate to consider the 
combined visibility impact of pollution controls on multiple units at a 
single facility in determining that BART is SNCR for Units 1 and 2 of 
the Colstrip facility in Montana and should likewise consider the 
combined visibility impact of SNCR on the two Westvaco boilers.
    Response: We have addressed a similar comment above in section V.B 
of this rulemaking. We recognize that there may be some efficiencies in 
installing SNCR on two units (e.g., a common reagent supply system), 
but expect that this would provide only a modest reduction in annual 
costs. We do not find that the combined benefit for the two FMC 
Westvaco boilers, 0.4 deciviews, is a basis for requiring SNCR.
    Comment: Wyoming's modeling also showed that SNCR could virtually 
eliminate the visibility impairment at the Bridger Wilderness Area 
caused by

[[Page 5195]]

the FMC Westvaco Units NS-1A and NS-1B, reducing the number of days of 
noticeable visibility impairment caused by each boiler from eleven to 
just one. Visibility in the Bridger Wilderness Area is presently 
diminished by 4.6 deciviews from natural conditions and, under EPA's 
proposed action, it will not achieve natural conditions until 2165. A 
0.4 deciview visibility improvement at the Bridger Wilderness Area is 
particularly significant in light of new sources of haze-causing 
pollution from the oil and gas industry that will affect this area. 
NOX emissions from Wyoming oil and gas development are 
expected to more than double in the current regional haze planning 
period, from 14,725 tpy in 2002 to 34,142 tpy in 2018, yet EPA does not 
propose any NOX emissions reductions from this sector. 
Accordingly, it is imperative for Wyoming and EPA to reduce 
NOX emissions from every other source to the greatest extent 
possible, including by requiring SNCR to be installed at the FMC 
Westvaco Units NS-1A and NS-1B.
    Response: We disagree with this comment. BART is a source-by-source 
analysis taking into consideration the five factors. The BART 
Guidelines and RHR do not require states or EPA to take into 
consideration the state being able to achieve the URP for a Class I 
area in its determination of BART for individual BART units.\236\
---------------------------------------------------------------------------

    \236\ In determining the measures necessary to make reasonable 
progress and in selecting RPGs for mandatory Class I areas within 
Wyoming, the State took into account the following four factors into 
consideration: Costs of compliance; time necessary for compliance; 
energy and non-air quality environmental impacts of compliance; and 
remaining useful life of any potentially affected sources. CAA 
section 169A(g)(1) and 40 CFR 51.308(d)(1)(i)(A). 40 CFR 
51.308(d)(1)(ii) allows for a slower rate of improvement in 
visibility than the URP, as long as it is demonstrated that based on 
these four factors, it is not reasonable to achieve the URP and that 
the selected RPG is reasonable. CAA section 169A(g)(1) and 40 CFR 
51.308(d)(1)(i)(A).
---------------------------------------------------------------------------

    Comment: Wyoming's modeling, upon which EPA relied, excluded all 
Class I areas beyond 300 km from the Westvaco facility. However, there 
is no demonstration that Class I areas further afield are not impacted 
by the Westvaco facility. As a comparison, EPA recently approved the 
South Dakota regional haze SIP which includes BART limits for the Big 
Stone facility, for which the nearest Class I area is over 400 km away.
    Response: We explained in response to another comment the reasons 
why we did not evaluate visibility impairment at Class I areas at 
distances greater than 300 km. Regarding the South Dakota regional haze 
SIP, there are no Class I areas within 300 km of the Big Stone 
Facility. Therefore, it was reasonable for the state to evaluate 
visibility impacts at the nearest Class I area even though the distance 
was greater than 300 km. We note that the BART rule provides some 
flexibility to the states in the approach used to evaluate visibility 
impairment. The fact that South Dakota chose to evaluate visibility 
impacts at a distance greater than 300 km does not impose a similar 
requirement on other states.
    Comment: Considering just the two Class I areas modeled, the 
installation and operation of SNCR would result in a cumulative maximum 
98th percentile visibility improvement of 0.304 deciviews from each 
unit, or 0.608 from both units combined. This cumulative visibility 
improvement at two Class I areas is significant and amply justifies 
SNCR, at a minimum, as BART.
    Response: In the proposed rule, we did focus on the visibility 
benefits at the most impacted Class I area. We considered the 
visibility benefits at the other Class I areas, but did not consider 
the benefits sufficient to warrant a change in our determination as to 
the appropriate level of control.
    Comment: Although the cost effectiveness and visibility 
improvements due to SNCR and LNB+SOFA standing alone justify a 
determination that this combination of technologies is BART, EPA 
apparently agreed with Wyoming that the incremental costs of requiring 
SNCR were not justified by the resulting visibility improvement. EPA's 
consideration of incremental cost effectiveness and visibility benefit 
is arbitrary given the lack of any objective criteria and in any event, 
must not be viewed in a vacuum.
    Response: We disagree with this comment. As stated above, we find 
it reasonable, based on a consideration of the five factors, including 
the low visibility improvement, for the State to find that SNCR and 
LNBs was not reasonable for BART.
    Comment: Although the State and EPA determined that addition of 
combustion controls is BART for the three BART boilers at these two 
facilities, it is unclear how they arrived at these conclusions. The 
visibility improvement from EPA's proposed controls for the trona 
plants are less than the visibility improvement that EPA rejected as 
``low'' in the EGU BART analyses, so it appears that EPA is using 
different criteria for these facilities or relating visibility 
improvement to another factor, which we assume to be some combination 
of cost and visibility improvement. (Otherwise, one would always choose 
the control strategy with the greatest visibility improvement.)
    However, it appears that EPA did not evaluate the cost analyses 
presented by the companies and the State, so we are concerned that the 
cost analyses for these two trona plants may suffer for the same 
problems that we pointed out to EPA before regarding the EGUs. For 
example, although Boiler D at Green River is the same size as the FMC 
boilers: (1) FMC evaluated addition of new combustion controls in 
combination with SNCR or SCR, Green River did not. (2) The capital cost 
of adding SNCR at Green River Boiler D is more than four times FMC. (3) 
EPA presented cost-effectiveness of SNCR as $3,176/ton at Green River 
Boiler D. The actual cost-effectiveness, based on EPA's annual cost and 
emission reduction, is $1,637/ton. (4) FMC assumed that SCR could 
reduce NOX by 31% to 0.10 lb/MMBtu, Green River assumed 80% 
NOX reduction to 0.14 lb/MMBtu. (EPA typically assumes that 
SCR can achieve 0.05 lb//MMBtu on an annual basis.) (5) SCR capital 
cost is $43 million at FMC, $19 million for Green River Boiler D. (6) 
EPA presented cost-effectiveness of SCR as $3,510/ton at Green River 
Boiler D. The actual cost-effectiveness, based on EPA's annual cost and 
emission reduction, is $2,339/ton.
    It is apparent that EPA must have been considering the costs of 
controls, but, in view of the substantial discrepancies noted above, 
those costs are questionable. In view of these discrepancies, we 
question how EPA rejected the more-effective control technologies (SNCR 
and SCR) that produce greater visibility improvements for the proposed 
controls.
    Response: We disagree with this comment. Even if the cost of SNCR 
and SCR were reduced, we find that the visibility improvement (Boiler 
C--0.08 deciviews for SNCR and 0.14 deciviews for SCR; Boiler D--0.12 
deciviews for SNCR and 0.17 deciviews for SCR) would not warrant the 
selection of post-combustion controls for BART.
b. FMC Granger Trona Mine
    Comment: EPA proposes to approve Wyoming's determination that the 
FMC Granger trona mine, while BART-eligible, is not subject-to-BART. 
The basis for EPA's proposed approval is that the visibility impact of 
this facility at the Bridger Wilderness Class I area would be 0.39 
deciviews, and EPA proposes to ``agree with Wyoming that 0.5 deciviews 
is a reasonable threshold for determining whether its BART-eligible 
sources are subject-to-BART.'' EPA should reconsider its determination 
that the Granger facility is not subject-to-BART. In making the

[[Page 5196]]

subject-to-BART determination at least three considerations must be 
incorporated pursuant to EPA's BART guidelines: Whether the source 
causes or contributes to visibility impairment in a Class I area, the 
number of emissions sources affecting a Class I area, and the magnitude 
of the individual source impacts. Wyoming determined that the Granger 
plant was not subject-to-BART because its visibility impairment level 
at the Bridger Wilderness was predicted to be 0.39 deciviews--below the 
0.5 deciview threshold. Wyoming's determination was flawed however 
because it apparently did not consider the other factors essential to a 
subject-to-BART determination, i.e., the number of emissions sources 
affecting the Class I area and the magnitude of the individual sources' 
impacts.
    There are a large number of pollution sources affecting visibility 
in the Bridger Wilderness Area, including significant impacts from 
thousands of operating oil and gas wells that are not BART-eligible. 
This fact highlights the need for maximum feasible emissions from each 
source contributing to impairment at the Bridger Wilderness Area, 
particularly sources like the FMC Granger trona mine, which is eligible 
for BART controls.
    Response: We disagree with this comment. 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 deciviews, 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. Given the relatively limited 
impact on visibility from these seven sources, we continue to agree 
with Wyoming that 0.5 deciviews is a reasonable threshold for 
determining whether its BART-eligible sources are subject-to-BART. In 
addition, the commenter points to the impacts from oil and gas at Class 
I areas. The BART Guidelines do not require states to consider the 
impacts from sources other than BART-eligible sources when defining the 
threshold for determining what sources are subject-to-BART. While the 
Guidelines first say that, in setting a contribution threshold, states 
should consider the number of ``emissions sources'' affecting the Class 
I area at issue, the Guidelines then go on to clarify that states may 
use a lower contribution threshold based on the location of a large 
number of ``BART-eligible'' sources within the State that are proximate 
to the Class I area at issue.

E. Reasonable Progress

1. RPGs
    Comment: 40 CFR 51.308(d)(1) of the RHR requires states to 
establish goals (in deciviews) that provide for reasonable progress 
towards achieving natural visibility conditions for each Class I area 
of the state. These are goals, not standards. Goals are typically 
understood as levels aimed for but not necessarily met. Early on in the 
process, EPA considered setting ``presumptive targets'' but eliminated 
them before the final rule. EPA also says that the RPGs established by 
the state are not directly enforceable. In spite of this, EPA has 
proposed to FIP the Wyoming RPGs.
    EPA does not specifically define the word ``goal,'' but the RHR 
does describe what must be considered when the goals are set. Wyoming 
has set six reasonable progress goals and every one of them met that 
criteria. EPA does not even argue with this basic fact. When setting 
the goals, the state must do a reasonable progress analysis. The State 
of Wyoming complied with this requirement as well.
    Response: EPA disagrees with this comment. Wyoming's selected RPGs 
do not meet the requirements of the RHR. In establishing RPGs, Wyoming 
must make two demonstrations. First, the State must demonstrate how the 
four statutory reasonable factors, as applied to potentially affected 
sources, were taken into consideration in selecting the goals. 40 CFR 
51.308(d)(1)(i). In addition, if Wyoming establishes an RPG that 
provides for a slower rate of improvement than the URP, the State must 
demonstrate, based on the four statutory reasonable progress factors, 
that achieving the URP is not reasonable and that the selected RPG is 
reasonable. 40 CFR 51.308(d)(1)(ii). In determining whether the 
selected RPGs in fact provide for reasonable progress towards natural 
visibility conditions, EPA must evaluate these two demonstrations. 40 
CFR 51.308(d)(iii).
    EPA's interpretation of the statute and the RHR is that BART 
sources should also be identified as anthropogenic sources of 
visibility impairment for purposes of developing the long-term 
strategy. 40 CFR 51.308(d)(3)(iv). Correspondingly, BART sources should 
be considered ``potentially affected sources'' and evaluated for 
controls using the reasonable progress factors. See 40 CFR 
51.308(d)(1)(i)(A). However, due to the similarity of the reasonable 
progress and BART factors, it is reasonable for states to rely on their 
BART determinations to fulfill the requirements of 51.308(d)(1)(i)(A) 
and 51.308(d)(1)(ii) (if applicable), in other words to demonstrate 
that the reasonable progress factors were reasonably considered for 
those sources for the first planning period. This interpretation is 
consistent with guidance EPA has issued for states regarding meeting 
reasonable progress requirements.\237\ However, the Wyoming submittal 
states that the reasonable progress ``four factor analysis . . . is a 
method for evaluating potential control strategies for facilities that 
are not eligible for Best Available Retrofit Technology (BART) or 
better-than-BART programs.'' Wyo. 309(g) Submittal at 115. Thus, the 
Wyoming submittal on its face fails to meet the requirements of the 
statute and the RHR. To the extent that Wyoming can be said to have 
relied on its BART determinations to establish that the State 
reasonably considered the reasonable progress factors for those 
sources, that reliance fails for those sources for which we are 
disapproving the BART determinations. In addition, as the State's RPGs 
fall short of the URP, the State failed to adequately demonstrate, 
based on the four statutory reasonable progress factors, that achieving 
the URP was not reasonable and that, the selected RPG is reasonable. 
Given our evaluation of these demonstrations, we have determined that 
the selected RPGs do not provide for reasonable progress towards 
natural visibility conditions.
---------------------------------------------------------------------------

    \237\ US EPA, Office of Air Quality Planning and Standards, 
Guidance for Setting Reasonable Progress Goals Under the Regional 
Haze Program (June 1, 2007).
---------------------------------------------------------------------------

    In addition, although we are not disapproving the State's ultimate 
determination to not impose controls on non-BART sources, we note that 
(as explained in more detail below) the State unreasonably relied on 
impermissible factors to reach those determinations. Thus, the State 
failed to demonstrate that it was reasonable, based on consideration of 
the statutory reasonable progress factors, to not meet the URP. In 
other words, although we are approving the State's decision as part of 
its long-term strategy to not

[[Page 5197]]

impose controls on the non-BART sources the State listed, we are still 
disapproving the State's RPGs.
    Because the State failed to meet the requirements of 
51.308(d)(1)(i) and (ii) when the State selected its RPGs as part of 
the State's Regional Haze SIP, EPA is obligated to promulgate a 
regional haze FIP to meet those requirements. That the RPGs are named 
``goals,'' not standards, and are not directly enforceable is 
irrelevant to this obligation.
    Comment: Wyoming explained to EPA that Wyoming could not compel 
these reasonable progress sources to put on controls without a State 
rule, and that rule would have to include a visibility impact analysis. 
Wyoming was willing to commit to developing such a rule in the next 
planning period, but it did not have the time or resources left to 
complete that task and get the SIP submitted to the EPA for the first 
planning period. Wyoming's administrative rulemaking process requires 
about nine months to a year to develop and finalize rules. Wyoming 
believes that it has taken an important first step in the process, and 
it appears to be more than many other states were making.
    Wyoming also believes that it made more sense to develop a 
comprehensive State reasonable progress rule that could be used for the 
next SIPs to address regional haze. That rule would take extra time 
that EPA was not willing to give the State. EPA told the State 
repeatedly that ``The Regional Haze Rule does not allow for commitments 
to potentially implement strategies at some later date that are 
identified under reasonable progress.'' The State is still dumbfounded 
by this kind of response for a rule that goes out to 2064, especially 
where EPA itself has recognized the one-step-at-a-time doctrine.
    Response: EPA disagrees with this comment. While we understand the 
State's position on its limits on its authority, time, and resources, 
EPA first promulgated the reasonable progress requirements under the 
RHR on July 1, 1999, and we issued our guidance on setting RPGs in 
September 2007. Wyoming submitted its Regional Haze SIP on January 12, 
2011. Wyoming does not explain why the State did not have an adequate 
amount of time to develop a regional haze SIP that meets the 
requirements for reasonable progress.
    In any case, the State's limits on its authority, time, and 
resources are not permissible factors for EPA to take into account when 
assessing the State's Regional Haze SIP. Instead, we must assess 
whether it meets the requirements of the RHR, and in particular the 
requirements for reasonable progress towards natural visibility 
conditions. We note that we are approving certain portions of Wyoming's 
Regional Haze SIP with respect to reasonable progress requirements.
    To the extent that the notion of ``one-step-at-a-time'' is relevant 
in this context, as explained elsewhere, Wyoming's Regional Haze SIP 
fails to adequately make the very first step towards natural visibility 
conditions: Achieving reasonable progress during the first planning 
period. A commitment to completing that first step in a future planning 
period cannot substitute for actually completing the first step within 
schedule. Wyoming cites no authority to the contrary; conditional 
approval under CAA section 110(k)(4) does not apply as Wyoming has made 
no commitment to adopt specific enforceable measures within one year to 
remedy the deficiencies. Again, whatever the constraints imposed on the 
State by time, resources, and authority, those constraints cannot be 
taken into account in assessing whether the State has met the 
requirements for the first planning period. In this case, Wyoming has 
not met those requirements with respect to reasonable progress.
    Comment: States are required, when setting RPGs, to determine the 
rate of progress needed to attain natural visibility conditions by 
2064. The State did that and included it in the SIP. EPA's RHR also 
requires that if the rate is slower than the rate established by 
drawing a straight line between baseline visibility and natural 
conditions, that it must be explained why. The data clearly show that 
the primary reason that the State will not reach natural conditions by 
2064 is that smoke from wildfires controls the slope of the line.
    Response: EPA disagrees with this comment, which understates the 
requirements of the RHR for setting RPGs. We agree that Wyoming did 
appropriately determine the URP needed to attain natural visibility 
conditions by 2064 and we are approving that determination. However, 
when a state selects an RPG that provides for a slower rate of 
improvement in visibility than the URP, it is not the case that all the 
state must do is ``explain why.'' Instead, the state must demonstrate, 
based on the statutory reasonable progress factors as applied to 
potentially affected sources, that the URP is not reasonable and that 
the selected RPG is reasonable. 40 CFR 51.308(d)(1)(ii). Under the RHR, 
40 CFR 51.308(d)(1)(iii), and under section 110 of the Act (as 
discussed elsewhere) we are required to evaluate the state's 
demonstration.
    As discussed elsewhere, the State did not reasonably consider the 
statutory reasonable progress factors for potentially affected sources. 
As a result, the State also failed to adequately demonstrate, based on 
the four statutory reasonable progress factors as applied to 
potentially affected sources, that achieving the URP was not reasonable 
and that the selected RPG is reasonable. We therefore are disapproving 
Wyoming's selected RPGs.
    With respect to the comment's reference to wildfires, we provide a 
detailed response to comments relating to wildfires and natural 
conditions in the modeling section of this response to comments.
    Comment: EPA cannot remove the reasonable progress goals for the 
State. Wyoming followed the process outlined in the RHR. EPA is not 
following the RHR by proposing a control requirement for a specific 
source to replace six RPGs for an entire state. The RHR does not allow 
for the substitution of RPGs with control strategies. EPA's proposed 
disapproval is contrary to law.
    Response: EPA disagrees with this comment. As discussed above, we 
are disapproving the State's selected RPGs because they do not meet the 
requirements of the RHR. In addition, the commenter is mistaken in 
stating that EPA is ``replacing'' RPGs with control requirements for a 
specific source. This statement conflates two separate but related 
requirements of the RHR. First, states must set RPGs in accordance with 
40 CFR 51.308(d)(1). Second, states must submit a LTS, including 
enforceable emissions limitations, compliance schedules, and other 
measures as necessary to achieve the RPGs. 40 CFR 51.308(d)(3).
    We are disapproving Wyoming's RPGs because they do not meet the 
requirements of 51.308(d)(1), as detailed above. Separately, we 
proposed to disapprove Wyoming's determination to not impose 
enforceable emissions limitations at Dave Johnston Units 1 and 2. Thus, 
we did not propose to ``replace'' the RPGs with control requirements; 
instead, we proposed to provide both. However, as explained elsewhere, 
on the basis of the cost and visibility information that EPA developed, 
we are now approving Wyoming's determination (although not the State's 
rationale) to not impose enforceable emissions limitations at Dave 
Johnston Units 1 and 2. We nonetheless continue to disagree with the 
comment.
    Comment: There is no way Wyoming can control the impacts from 
wildfire

[[Page 5198]]

smoke on visibility. Therefore, it will be a very long time, 126 to 161 
years, before controlling manmade sources can ever overcome the smoke 
impacts, and that is assuming that smoke impacts never increase. The 
length of time for other western states is even longer, and EPA has 
approved those SIPs. Wyoming included this explanation along with 
identifying other sources that impact visibility, but EPA disagreed 
with the Wyoming assessment, saying not all reasonable controls were 
implemented during the first planning period. Specifically, EPA 
disagreed with Wyoming's determination to not impose controls on Dave 
Johnston Units 1 and 2. Because Wyoming did not impose controls on Dave 
Johnston Units 1 and 2, EPA has proposed to disapprove Wyoming's RPGs.
    Wyoming believes that EPA's reasoning for disapproving the State's 
RPGs is flawed and arbitrary. First, the State set goals based on 
regional modeling projections done for the entire western U.S. To the 
best of our knowledge, that is the same process that every other state 
in the western U.S. used and many of them now have approved RPGs in 
spite of the fact that it will take hundreds of years in all of the 
western Class I areas to reach ``natural conditions.'' In North Dakota, 
for example, it will take between 156 and 232 years to reach natural 
conditions at affected Class I areas. It would be impossible to set 
deciview goals without regional modeling, unless the State wanted to 
wildly guess at it.
    Response: EPA disagrees with this comment. We are not disapproving 
Wyoming's RPGs solely on the basis that they fall short of achieving 
the URP. Instead, as explained above, we are disapproving them on the 
basis that the State has failed to demonstrate that the four statutory 
reasonable progress factors were appropriately considered. The State 
has also failed to demonstrate, again based on the four statutory 
reasonable progress factors, that achieving the URP is unreasonable and 
that the State's selected RPGs are reasonable. The comment's reference 
to wildfires is beside the point, as the existence of wildfires does 
not relieve the State of all responsibility to reasonably consider the 
statutory reasonable progress factors for potentially affected sources. 
We elsewhere provide a detailed response to comments relating to 
wildfires and natural conditions in the modeling section of this 
response to comments.
    Comment: While EPA ``anticipates'' that controls at Dave Johnston 
Units 1 and 2 would result in measurable visibility improvement in 
regional modeling demonstrations, and that ``anticipation,'' not 
modeling, therefore justifies dispensing with Wyoming's RPGs, Wyoming 
does not. When the WRAP modeled all of the emission reductions from the 
entire western U.S. (including Wyoming emission reductions for all of 
the pollutants) for this first planning period, Wyoming saw an 
improvement of 0.6 deciviews at the Yellowstone site, and a 0.5 
deciview improvement at the North Absaroka and Bridger sites on the 
worst days. The numbers are even smaller or zero for the best days. 
These improvements from much larger emission reductions for multiple 
pollutants are almost imperceptible. Therefore, it is highly unlikely 
that emission reductions for NOX from two units would make 
enough difference to show up as an ``improvement'' in regional scale 
modeling, and thereby justify setting different RPGs.
    Response: We disagree with the approach suggested in this comment. 
Below, we discuss the use of CALPUFF (instead of the regional scale 
modeling the comment suggests) to determine visibility improvement from 
controls on Dave Johnston Units 1 and 2. We also respond to comments 
regarding regional scale modeling in section V.B.
    Comment: The EPA proposes to impose reasonable progress controls on 
Dave Johnston Units 1 and 2, more stringent NOX BART 
controls on Dave Johnston Unit 3, Jim Bridger Units 1 and 2, Wyodak 
Unit 1, and Laramie Units 1, 2, and 3. These EPA proposed controls are 
more stringent than what was assumed by the WRAP in modeling Wyoming's 
RPGs. Wyoming established its RPGs based on the regional modeling 
projections completed in the WRAP process. In proposing these 
reasonable progress controls, EPA is also proposing RPGs that are 
consistent with the controls, thereby rejecting Wyoming's proposed 
RPGs.
    In rejecting Wyoming's RPGs and imposing its own, EPA did not re-
run the WRAP model; instead, the agency essentially guessed ``that the 
additional controls would result in an increase in visibility 
improvement during the 20% worst days,'' thereby warranting the more 
stringent controls at these units.
    EPA's proposal to reject Wyoming's RPGs is not warranted. First, 
the mere assumption that additional controls will result in greater 
visibility improvement cannot reasonably be supported without modeling 
data. EPA admits that it ran no modeling that would support its best 
guess that visibility would improve with the installation of more 
stringent controls. Second, EPA's proposal to place controls on the 
Dave Johnston Units 1 and 2 is flawed. The Dave Johnston Units 1 and 2 
are not BART-eligible units. When Wyoming considered the WRAP model 
data, it concluded that putting controls on the Dave Johnston Units 1 
and 2 would not result in an improvement in visibility. Without any 
improvement in visibility coming from placing controls on these non-
BART units, Wyoming reasonably concluded that there was no reason to 
change its RPGs.
    Response: We disagree with this comment. WRAP performed regional 
photochemical modeling using both the CMAQ and 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, 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. As discussed 
earlier in this rulemaking and the docket for this final action, EPA 
modeled visibility impairment from individual sources in making its 
determination of BART and reasonable progress controls. Thus, the 
comment is inaccurate in stating that EPA ran no modeling to assess 
whether controls on Dave Johnston Units 1 and 2 would improve 
visibility. With respect to the assertion that Wyoming considered the 
WRAP model data and decided that the data showed controls on Dave 
Johnston Units 1 and 2 would not improve visibility, the Wyoming SIP 
submittal does not reflect that. In evaluating the reasonable progress 
factors for Dave Johnston Units 1 and 2 (which was selected by the 
State as a potentially affected source) the SIP submittal states: ``LNB 
or LNB w/OFA seem to be the most reasonable choice[s] for the Dave 
Johnston Electric Generating Station boilers BW41 and BW42 based on the 
four factor analysis. The implementation of new control technologies on 
the two boilers are discussed in further detail in Chapter 8 (Section 
8.3.4), Long-Term Strategy.'' Nonetheless, in section 8.3.4, the SIP 
stated: ``The Air Quality Administrator cannot, per Wyoming Statute 35-
11-202, establish emission control

[[Page 5199]]

requirements except through State rule or regulation. Furthermore, the 
Wyoming statute requires the Administrator to consider the character 
and degree of injury of the emissions involved. In this case, 
visibility modeling would be required to assess the degree of injury 
caused by the emissions. Modeling is not available at this time to 
determine impacts from emission reduction.'' As we explain elsewhere, 
these are not permissible reasons to ignore the four statutory 
reasonable progress factors. Nonetheless, our revised visibility 
modeling leads us to the conclusion that it was not unreasonable for 
the State to not impose controls on Dave Johnston Units 1 and 2, even 
though the State's basis for doing so was inadequate.
    Comment: Wyoming's adoption of an alternative SO2 
program, even if it were valid, does not relieve Wyoming of its 
obligation to develop and implement a LTS that includes measures 
necessary to reduce visibility-impairing emissions of SO2, 
PM, and NOX to achieve RPGs for non-Colorado Plateau Class I 
areas. Accordingly, EPA must determine whether Wyoming's RPGs for its 
non-Colorado Plateau Class I areas are adequate.
    Response: We agree that Wyoming must develop a LTS to address 
reasonable progress for non-Colorado Plateau Class I areas. As our 
proposed notice indicates, we proposed to disapprove the State's RPGs. 
We also proposed to implement additional controls under reasonable 
progress. We are completing the action to disapprove the State's RPG's 
today, and as explained elsewhere in this section, we are not 
finalizing requirements for additional controls under reasonable 
progress.
    Comment: With the exception of the controls required on Naughton 
Unit 3, PacifiCorp has installed all of the BART controls required by 
the Wyoming BART permits and the regional haze SIP. These controls were 
installed from 2005 through 2012. The actual monitored visibility 
impairment demonstrates that Wyoming has made significant progress in 
reducing nitrate concentrations and further demonstrates that the RPGs 
are on track through the 2008-2017 planning period. EPA's FIP is not 
``necessary'' to meet RPGs for nitrates in these Class I areas. As a 
result, EPA should withdraw its FIP.
    Response: EPA disagrees with this comment. As explained above, the 
State was required to assess the four statutory reasonable progress 
factors for potentially affected sources and reasonably determine 
potential controls, and we are required to evaluate the State's 
determination. The State did not demonstrate reasonable progress for 
those determinations that we are disapproving. As a result, we must 
disapprove the State's RPGs and promulgate a FIP for them. We also note 
that the comment does not explain the relationship between the State's 
RPGs and changes in monitored visibility impairment as the result of 
installed controls, as the State's RPGs were not remodeled to reflect 
the controls selected by the State.
    Comment: Any discussion of the appropriate NOX control 
levels required under the RHR should include an assessment of the 
existing visibility levels to understand what pollutants are driving 
visibility impairment in Wyoming. Measured visibility impairment at 
Wyoming's IMPROVE monitoring stations shows that the contribution from 
nitrates, which are visibility impairing pollutants that result from 
NOX emissions, play a lesser role in visibility impairment 
in Wyoming than particulate organic mass or sulfates. The latest 
available IMPROVE data (2000-2009) from the WRAP Technical Support 
System reveals the following about the two Class I areas that were most 
closely examined for impacts from Wyoming BART sources: (1) Currently, 
the air in those Class I areas is very clear, with overall visibility 
among the best in the entire country; (2) When visibility is not good, 
i.e., when you can't see across the vista, it is likely because of 
smoke from wildfires; (3) The contribution to visibility impairment 
from nitrate particles, which are as a result of emissions of 
NOX, is small.
    The State believes it has made a good case that fire contributes 
more to visibility impairment than nitrates at Class I areas most 
affected by Wyoming sources. The State has made great progress in 
reducing the manmade contribution to visibility impairment from power 
plants, even when the manmade contribution has much less impact to 
visibility impairment than other components. EPA's proposed disapproval 
and FIP are not supported by a record that demonstrates small 
visibility improvements predicted by a CALPUFF model replete with 
uncertainty when the actual, measured levels of nitrates at Class I 
areas affected by those sources is so small. Wyoming's SIP is 
adequately supported because Wyoming considered these and other factors 
in arriving at the selected levels of NOX controls for 
Wyoming sources and the schedule for the installation of those 
controls.
    Response: We disagree with this comment. Regardless of the 
considerations presented in the comment, the State was required, at a 
minimum, to evaluate the four statutory reasonable progress factors for 
potentially affected sources and to reasonably determine controls, and 
we are required to evaluate the State's determination. In evaluating 
the factors for Dave Johnston Units 1 and 2, which was selected in the 
State's SIP as a potentially affected source, the SIP submittal states: 
``LNB or LNB w/OFA seem to be the most reasonable choice[s] for the 
Dave Johnston Electric Generating Station boilers BW41 and BW42 based 
on the four factor analysis. The implementation of new control 
technologies on the two boilers are discussed in further detail in 
Chapter 8 (Section 8.3.4), Long-Term Strategy.'' Nonetheless, in 
section 8.3.4, the SIP stated: ``The Air Quality Administrator cannot, 
per Wyoming Statute 35-11-202, establish emission control requirements 
except through State rule or regulation. Furthermore, the Wyoming 
statute requires the Administrator to consider the character and degree 
of injury of the emissions involved. In this case, visibility modeling 
would be required to assess the degree of injury caused by the 
emissions. Modeling is not available at this time to determine impacts 
from emission reduction.''
    As explained above, it is unreasonable and impermissible for the 
State to disregard its four factor analysis on the basis that the State 
lacked the necessary modeling and that reasonable progress requirements 
could be postponed until the next planning period. The considerations 
presented by the comment do not change this.
    In addition, section 110(a)(2)(E)(i) of the Act requires that SIPs 
provide necessary assurances that, among other things, the State has 
adequate authority and resources to carry out the plan. The SIP 
language we quote above instead denies that the State has the proper 
authority and resources to meet the requirements of the RHR, in 
particular the requirement that the long-term strategy ``include 
enforceable emissions limitations . . . and other measures as necessary 
to achieve the reasonable progress goals.'' 40 CFR 51.308(d)(3). As a 
result, Wyoming's Regional Haze submittal fails to meet the 
requirements of section 110(a)(2)(E)(i), which is applicable to 
``[each] implementation plan submitted by a State under [the CAA],'' 
including the Regional Haze submittal.
    Comment: An area of the RHR that is unusual is in the timing of the 
implementation of the rule. It is the most forward looking of all the 
rules

[[Page 5200]]

with requirements to be carried out by the grandchildren of the people 
who are currently working on the rule, with an end date of 2064. While 
EPA has established long-term targets through the acid rain program and 
ozone attainment requirements in a 10-20 year time frame, they have 
never set goals that were 60 years down the road. This is significant 
because EPA recognized that the problem was complicated and that it 
would take at least this much time to solve it.
    EPA's strategy included breaking up the long-range goal of 
achieving natural conditions by 2064 into many smaller pieces. EPA 
included a requirement for states to submit comprehensive SIP revisions 
in 2018 and every ten years thereafter. In addition to the 
comprehensive SIP revisions, states will also be required under 40 CFR 
51.308(g) to submit progress reports in the form of a SIP revision 
every five years, with the first revision due in 2013. Between both the 
comprehensive SIP revisions and the progress report SIP revisions, 
states will be working on 16 more SIP revisions, at a minimum, to 
address regional haze. The State views these upcoming SIP revisions on 
regional haze as opportunities to build on the first SIP, and that the 
current rush by the EPA to get so many reductions procured in the first 
time period as unnecessary. It is unnecessary because the State has 
submitted a plan to reduce NOX from BART sources by 45,153 
tons, and an additional 19,677 tons through the LTS in the first 
planning period. There are few states in the country that can 
demonstrate this magnitude of emission reductions Wyoming has secured.
    EPA recognized in the RHR preamble that many factors will change 
over time and that it may be possible to procure emission reductions in 
the future that cannot be accomplished during an earlier period. EPA 
expected reductions to occur over time and did not expect states to 
front end load this program with emission reductions.
    The RHR provides states with the time necessary to intelligently 
address the very complicated problem of regional haze. Wyoming asks EPA 
to recognize their own intentions to roll out this program step by step 
and approve the State's decision to require SCR on PacifiCorp Units 1 
and 2 of the Jim Bridger Power Plant in 2021 and 2022 as part of the 
LTS. The State also asks that EPA give the State the time it needs to 
create a rule to address reasonable progress, which would include 
reductions at the PacifiCorp Dave Johnston Plant, Units 1 and 2. 
Wyoming plans to create a general reasonable progress rule in the next 
planning period to address future reductions.
    Response: While we recognize the emission reductions achieved by 
the State for the first planning period and that the regional haze 
program is a long-term program, the State must still meet the 
requirements of 40 CFR 51.308 for the first planning period. As we 
stated in our proposal notice, the State's plan does not fully meet the 
requirements for BART and reasonable progress. Because we have found 
that the State's SIP submission did not adequately satisfy the RHR 
requirements in full, we have not only the authority, but a duty to 
promulgate a FIP that meets those requirements. The EPA disagrees that 
the additional emission reductions required by our proposed FIP are 
unnecessary, as we have demonstrated that the State's SIP does not meet 
the requirements of 40 CFR 51.308. Our FIP action is only intended to 
ensure that CAA requirements are satisfied in accordance with our 
authority under the CAA.
    Comment: We received numerous comments that monitoring data shows 
that the worst visibility days are due to wildfires and that EPA should 
be focusing on these emissions and not on nitrate emissions from 
stationary sources, which have little impact on poor visibility days. 
One commenter pointed out data from Class I areas in Wyoming that show 
organic carbon and elemental carbon, which are indicators of wildfire, 
are major contributors on poor visibility days compared to nitrates. 
Another commenter stated that the only EPA policy to address fires is 
the Interim Air Quality Policy on Wildland and Prescribed Fires which 
has not been updated since 1998 and that the EPA is not taking action 
on this core issue.
    Response: While we agree that industrial facilities are not the 
only causes of haze, we disagree with the thrust of this comment. We 
provide a detailed response to comments relating to wildfires and 
natural conditions in the modeling section of this response to 
comments. Regardless of the contribution from wildfire emissions, 40 
CFR 51.308(d)(3)(iv) states, ``The State must identify all 
anthropogenic sources of visibility impairment considered by the State 
in developing its long-term strategy. The State should consider major 
and minor stationary sources, mobile sources, and area sources.'' As 
discussed elsewhere, in its submittal the State identified a number of 
stationary sources as potential contributors to visibility impairment 
(i.e. potentially affected sources). The State was required, at a 
minimum, to evaluate the five statutory BART factors and four statutory 
reasonable progress factors for potentially affected sources and to 
reasonably determine controls, and we are required to evaluate the 
State's determination. 40 CFR 51.308(e) and 40 CFR 51.308(d)(3)(iv), 
respectively. The requirements of 40 CFR 51.308(d)(3)(iv) and 40 CFR 
51.308(e) are not dependent on the showing of a certain amount of 
impairment from point sources.
    Comment: The CAA and the RHR require SIPs to set forth goals, 
expressed in deciviews, that assure ``reasonable progress toward 
meeting the national goal'' of ``natural visibility conditions [in 
Class I areas] by the year 2064.'' 42 U.S.C. 7491(a)(4), (b); 40 CFR 
51.308(d)(1)(i)(A). The goals ``must provide for an improvement in 
visibility for the most impaired days over the period of the 
implementation plan and ensure no degradation in visibility for the 
least impaired days over the same period.'' 40 CFR 51.308(d)(1). To 
establish these goals, a state must also ``[a]nalyze and determine the 
rate of progress needed to attain natural visibility conditions by the 
year 2064,'' by ``compar[ing] baseline visibility conditions to natural 
visibility conditions [in Class I areas] and determin[ing] the uniform 
rate of visibility improvement'' necessary to achieve natural 
conditions by 2064. 40 CFR 51.308(d)(1)(i)(B).
    Wyoming's SIP meets these requirements. See SIP, at 114-31. The SIP 
calculates and compares baseline and natural visibility conditions, Id. 
at 114-15, analyzes the rate of progress needed to attain natural 
visibility conditions by 2064, Id., and establishes a uniform rate of 
progress, Id. Wyoming also ensured improvement in visibility on the 
most impaired days and no degradation on the least impaired days. See 
Id. at 115 (Table 7.2.1). And, most importantly, the SIP establishes 
reasonable progress goals. Id. at 127-131. The CAA and the RHR also 
require states to make reasonable progress determinations for 
particular sources by ``[c]onsider[ing] the costs of compliance, the 
time necessary for compliance, the energy and non-air quality 
environmental impacts of compliance, and the remaining useful life of 
any potentially affected sources, and includ[ing] a demonstration 
showing how these factors were taken into consideration in selecting 
the goal.'' 40 CFR 51.308(d)(l)(i)(A).
    Wyoming also met this requirement. The SIP clearly explains how 
Wyoming considered these factors and identified sources impacting 
visibility in Class I areas. See SIP, at 116-17. Wyoming then

[[Page 5201]]

explained in its SIP how it applied the factors to each individual 
source. See Id. at 117-27. The SIP therefore meets the requirements of 
the Act and the RHR.
    Response: As discussed elsewhere, we have evaluated Wyoming's BART 
and reasonable progress determinations and we are disapproving them for 
Dave Johnston Unit 3, Wyodak Unit 1, and Laramie River Station Units 1-
3. Because the State did not reasonably consider the statutory BART 
factors for these sources, the State also failed to adequately 
demonstrate (to the extent that the State relied on its BART 
determinations to demonstrate the required consideration of the 
reasonable progress factors) that the reasonable progress factors were 
appropriately considered in establishing the RPGs. The State also 
failed to adequately demonstrate, based on the statutory BART and 
reasonable progress factors, that achieving the URP was not reasonable 
and that the selected RPG is reasonable. Given our evaluation of these 
two demonstrations and the comments received, we have determined that 
the selected RPGs do not provide for reasonable progress towards 
natural visibility conditions.
    In making this determination, we are not limited to merely noting 
whether the State has submitted an analysis that purports to consider 
the BART and reasonable progress statutory reasonable progress factors. 
Instead, we evaluate whether the State reasonably assessed the 
statutory BART and reasonable progress factors as applied to 
potentially affected sources and, based on those factors, reasonably 
determined whether controls were required for this planning period. In 
this case, the State did not do so.
    As discussed earlier, because the State failed to meet the 
requirements of Sec.  51.308(d)(1)(i) and (ii) when the State selected 
its RPGs as part of the State's Regional Haze SIP, EPA is obligated to 
promulgate a regional haze FIP to meet those requirements.
    We do agree that the State did correctly calculate and compare 
baseline and natural visibility conditions, analyzed the rate of 
progress needed to attain natural visibility conditions by 2064, and 
established a URP. We agree that Wyoming's SIP ensured improvement in 
visibility on the most impaired days and no degradation on the least 
impaired days, as does our FIP.
    Comment: EPA acknowledges that Wyoming evaluated the requisite four 
factors in its reasonable progress determinations. 78 FR 34785. But, 
EPA asserts that Wyoming incorrectly calculated costs in those 
determinations. Id. EPA, however, does not explain how Wyoming 
incorrectly calculated costs. EPA asserts first that ``EPA's rationale 
for disapproving the State's reasonable progress determination[s] . . . 
can be found in Section VIII.B of [the proposal].'' ld. at 34763. 
Section VIII.B--the location of EPA's supposed ``rationale''--only 
reiterates EPA's general allegation of deficiencies in the control cost 
estimates. Id. at 34785. EPA therefore has not described with any 
meaningful degree of specificity the supposed errors that justify 
rejecting the State's reasonable progress determinations. EPA's failure 
to provide an intelligible justification for its action is unlawful and 
arbitrary, and precludes Wyoming from offering a more meaningful 
response.
    Response: We disagree. First, the commenter fails to fully disclose 
EPA's proposed rationale for disapproving the State's reasonable 
progress determination for Dave Johnston Units 1 and 2. The commenter 
cites language related to our finding of deficiencies with the State's 
cost analysis (at 78 FR 34785), but fails to cite our fuller 
explanation for disapproving the State's determination a few pages 
later (at 78 FR 34787): ``We disagree with the State's reasoning for 
not adopting reasonable progress controls for Dave Johnston Unit 1 and 
Unit 2. If the State determined that it needed to adopt a rule or 
perform modeling to adequately assess and, if warranted, require 
reasonable progress controls, the State should have completed these 
steps before it submitted its regional haze SIP. The RHR does not allow 
for commitments to potentially implement strategies at some later date 
that are identified under reasonable progress or for the State to take 
credit for such commitments.''
    We offered this rationale in response to the State's argument that 
no controls were reasonable because: (1) the State's four factor 
analysis was limited, in that no guidance was provided by EPA for 
identifying significant sources and EPA did not establish contribution 
to visibility impairment thresholds (a potential fifth factor for 
reasonable progress determinations), (2) the State cannot, per Wyoming 
Statute 35-11-202, establish emission control requirements except 
through State rule or regulation, (3) the Wyoming statute requires the 
State to consider the character and degree of injury of the emissions 
involved--information that State claimed not to have, and (4) the State 
believes it has taken a strong and reasonable first step in identifying 
potential contributors to visibility impairment, and that the next step 
of creating an appropriate rule or regulation will be accomplished in 
the next SIP revision. 78 FR 34786. Therefore, our proposed rationale 
for disapproving the State's reasonable progress determination for Dave 
Johnston extended beyond our concerns with the cost analysis.
    Even so, contrary to the commenter's assertions, and though perhaps 
not to the level of detail desired by the commenter, we did 
sufficiently explain our concerns with deficiencies in Wyoming's cost 
analyses, including those for Dave Johnston Units 1 and 2. Most 
notably, as described in Section VII.C of the proposed rule, we 
recognized that Wyoming had understated ``the ability of SCR to reduce 
NOX.'' This was most pronounced at Dave Johnston Units 1 and 
2 where the State assumed that SCR would only reduce NOX to 
an emission rate of about 0.09 lb/MMBtu (equivalent to an 80% reduction 
from 2001-2003 baseline). As we have established elsewhere in response 
to comments, in this instance SCR has the ability to reduce 
NOX to an emission rate of 0.05 lb/MMBtu or less. Therefore, 
it is clear that the State underestimated the emission reductions that 
can be achieved with SCR, and thereby miscalculated the cost 
effectiveness. And while EPA did not find that SCR was warranted for 
Dave Johnston Units 1 and 2, it was nonetheless necessary to correctly 
calculate the cost effectiveness of all of the technically feasible 
controls in order to rationally evaluate the State's decision to not 
impose any controls and to (had we been compelled to impose a FIP) 
select from among competing control options.
    Comment: The RHR clearly states that every implementation plan must 
include reasonable progress goals. 40 CFR 51.308(d)(l). Those goals 
must be expressed in deciviews and must provide for visibility 
improvement on the most impaired days and no degradation on the least 
impaired days during the planning period. Id. In EPA's own words, RPGs 
are ``[t]he vehicle for ensuring continuing progress towards achieving 
the natural visibility goal,'' 78 FR 34743, which is the focal point of 
the regional haze program, see 42 U.S.C. 749l(a)(l).
    EPA proposes to disapprove the State's reasonable progress goals. 
78 FR 34767. In the same sentence, EPA claims to be proposing a FIP to 
replace those goals, which EPA asserts can be found in Section VIII.C 
of the notice. Section VIII.C reveals, however, that EPA has in fact 
failed to establish replacement RPGs. See Id. at 34788.

[[Page 5202]]

EPA does not set forth RPGs in deciviews, nor does it provide for 
visibility improvement on the most impaired days with no degradation on 
the least impaired days. See Id. Instead, EPA merely ``anticipates'' 
that its FIP would lead to improved visibility. Id. EPA's anticipation 
falls far short of the plain requirements of the RHR--concrete, 
deciview-based reasonable progress goals that provide for improved 
visibility on the worst days and no degradation on the best days. EPA's 
failure to establish RPGs to replace the SIP goals EPA proposes to 
disapprove is therefore unlawful.
    EPA justifies its failure to establish the requisite RPGs by 
explaining that it ``could not re-run the modeling due to time and 
resource constraints [.]'' Id. This excuse stands in stark contrast to 
EPA's response to similar claims the State raised in the context of 
reasonable progress. For example, the State explained to EPA that the 
State could not complete its evaluation of the impacts to visibility 
from oil and gas sources until the WRAP completes its emission 
inventory study. Id. at 34764-34765. EPA responded that ``If the State 
determined that additional information was need . . . the State should 
have developed the information.'' Id. at 34765. Similarly, the State 
explained to EPA that it needed to conduct additional modeling before 
it could justify controls for the Mountain Cement kiln. Id. at 34765-
34766. Again setting forth its dual standard, EPA responded that ``If 
the State determined that it needed to adopt a rule or perform modeling 
. . . the State should have completed these steps before it submitted 
its regional haze SIP.'' Id. at 34766.
    Response: EPA disagrees with this comment to the extent it argues 
that we should approve the State's RPGs. We note that the State did 
not, in setting its RPGs, re-run its modeling to reflect the State's 
selected controls. Instead, the State relied on WRAP modeling that 
reflected certain generic assumptions about the level of controls. See 
Wyo. 309(g) SIP, pages 53 and 127. As we have explained elsewhere, 
regardless of how the State quantified its RPGs, they cannot be 
approved, as the State failed to appropriately consider the four 
statutory reasonable progress factors for the sources the State 
selected as potentially affected sources. See 40 CFR 
51.308(d)(1)(i)(A). In addition, the State cannot rely on the BART 
determinations that we are disapproving to show reasonable progress for 
those sources. Because we must disapprove the State's RPGs, and RPGs 
are a required component of a regional haze SIP, we must promulgate our 
own. We note that the RPGs are not directly enforceable. 40 CFR 
51.308(d)(1)(v). The elements that directly impact sources and 
visibility are the emissions limitations in the long-term strategy, 
including those for BART and those for the reasonable progress sources.
    Comment: When determining the responsibility for regional haze, 
Sweetwater County strongly believes that the DEQ and EPA need to 
investigate the contribution to Wyoming's haze problem by sources 
located outside of the United States, especially from countries like 
China that do not appreciate the necessity for strong environmental 
regulations. If we do not consider the effects of air pollution 
contributing to our nation's and our State's air quality issues, we 
open the door for unfair competition. To assign the entire cost of 
Wyoming's haze and air pollution to Wyoming industries without 
considering the effects of offshore sources is unfair to our 
industries, and it would cause unnecessary impacts to the economy of 
Wyoming and the United States.
    Response: While sources outside Wyoming do contribute to haze in 
the Class I areas within Wyoming, that does not preclude the State's or 
our obligation to evaluate sources within the State according to the 
five BART factors and the four reasonable progress factors and to 
require additional controls where necessary. In addition, we note that 
the State did evaluate the sources of contribution to Class I areas in 
the State (see e.g., Chapter 5 of the SIP).
    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 IMPROVE 
network to show the contributions to light extinction from organic 
carbon, elemental carbon, and nitrate.
    Response: EPA does recognize this issue and has taken it into 
consideration in this action on the Wyoming SIP and in our final FIP. 
We agreed that Wyoming did appropriately determine the URP needed to 
attain natural visibility conditions by 2064 and we are approving that 
determination. We are not disapproving Wyoming's RPGs solely on the 
basis that they fall short of achieving the URP. Instead, as explained 
above, we are disapproving them on the basis that the State has failed 
to demonstrate that the four statutory reasonable progress factors were 
appropriately considered. As stated previously, regardless of the 
contribution from wildfire emissions, 40 CFR 51.308(d)(3)(iv) states, 
``The State must identify all anthropogenic sources of visibility 
impairment considered by the State in developing its long-term 
strategy. The State should consider major and minor stationary sources, 
mobile sources, and area sources.'' As discussed elsewhere, in its 
submittal the State identified a number of stationary sources as 
potential contributors to visibility impairment (i.e. potentially 
affected sources) and was required, at a minimum, to evaluate the five 
statutory BART factors and four statutory reasonable progress factors 
for potentially affected sources and to reasonably determine controls, 
and we are required to evaluate the State's determination. 40 CFR 
51.308(e) and 40 CFR 51.308(d)(3)(iv), respectively.
2. Reasonable Progress Sources
a. Oil and Gas Sources
    Comment: We received comments that volatile organic compound (VOC) 
emissions from the oil and gas industry must be controlled under 
reasonable progress. Commenters asserted that EPA acknowledged that oil 
and gas sources emit haze-causing VOCs but inexplicably failed to 
analyze whether reducing such VOC emissions is reasonable. One 
commenter pointed out that EPA has just designated Sublette County (and 
portions of Sweetwater and Lincoln Counties) in nonattainment with the 
8-hour ozone national ambient air quality standard, so there is no 
doubt the ozone levels in Sublette County are of great concern. 
Commenters pointed out that ozone severely impairs visibility; the 
failure to consider strategies to limit oil and gas industry VOC 
emissions was a significant oversight on both the part of the State and 
EPA. Thus, commenters concluded that EPA must correct this problem by 
analyzing and imposing reasonable progress controls on oil and gas 
industry VOC emissions.
    Commenters pointed out that there are numerous opportunities to 
reduce VOC emissions from the oil and gas industry. These include 
requiring all oil and gas fields in the State to control VOC emissions 
to the same extent currently required in the Pinedale Anticline and 
Jonah fields pursuant to the State's BACT guidelines, implementing 
recommendations from the Upper Green River Basin Air Quality

[[Page 5203]]

Citizens Advisory Task Force, and adoption of a statewide offset 
program.
    Response: We disagree with this comment. The commenters did not 
provide any evidence of the impact of VOC emissions on visibility in 
Class I areas.
    Comment: Wyoming claims that regulation of drilling rigs is 
problematic because drilling rigs are mobile sources over which states 
have limited CAA authority. EPA is not similarly constrained and may 
require emissions reductions from drilling rigs in a FIP. Replacement 
of Tier 2 engines with Tier 4 engines on drilling rigs has a cost 
effectiveness value as low as $900/ton, which is very reasonable.
    Response: EPA disagrees with this comment. The costs noted by the 
commenter for controls for drill rig engines are the lower end of the 
costs presented by the State. For replacement of Tier 2 engines with 
Tier 4 engines, the State presented costs of $900 to $2400 per ton of 
NOX removed, but the commenter cited only the $900 per ton 
figure. To the extent that drill rig engines could be regulated under 
the RHR, this range of costs is not so low that we are prepared to 
disapprove the State's determination in the reasonable progress 
context.
    Comment: EPA states it disagrees with the State's reasoning for not 
adopting reasonable progress controls for the for oil and gas sources. 
It is our view that, having made this finding, it is inappropriate for 
EPA to then propose approval of the State's control plan, a plan which 
would involve no new controls on the oil and gas sector. Having found 
that the State's RPGs were not justified, the EPA must put in place a 
FIP establishing RPGs for the oil and gas sector or ask the State to 
revise its plan.
    There are numerous available means for controlling NOX 
from the oil and gas sector, which is the primary focus that EPA has. 
For example, the State has begun regulating NOX emissions 
from drill rigs on the Pinedale Anticline and Jonah Field. The State 
has put in place a number of regulations on those drill rigs. We 
believe there is no reason this could not be extended to other fields 
in other portions of the state.
    Response: We disagree with this comment. First, we did not propose 
approval of the State's control plan in its entirety. Instead, we 
proposed to disapprove the State's reasonable progress determination 
for Dave Johnston Units 1 and 2; we also proposed to disapprove the 
State's RPGs. We then proposed a FIP for the RPGs. While we are 
approving the State's reasonable progress determination for Dave 
Johnston Units 1 and 2, we are still finalizing a FIP for the RPGs, as 
we have disapproved some of the State's BART determinations. Second, as 
we stated in our proposal, although we disagree with the State's 
reasoning with respect to the oil and gas sector, after considering the 
costs presented by the State, we find that they are not so low that we 
are prepared to disapprove the State's determination in the reasonable 
progress context. With respect to NOX emissions generally 
from the oil and gas sector, as discussed elsewhere, Wyoming applies 
minor source BACT to these sources. For drill rig engines in 
particular, see our response above. Finally, with respect to visibility 
impacts of NOX emissions from oil and gas sources on Class I 
areas, this comment provided no particular data. We respond below to 
other comments on visibility impacts of oil and gas sources.
    Comment: The State provided sound reasoning for not adopting 
reasonable progress controls for oil and gas sources. Wyoming is an oil 
and gas production state, along with Colorado, North Dakota, New 
Mexico, Montana, and Utah. One of the biggest challenges faced by these 
WRAP states has been to inventory the emissions from this industry. At 
the beginning of the regional haze process, a comprehensive emission 
inventory of oil and gas production operations in the western region 
that covered both point and area sources had not been developed. No 
methodology had been developed to produce an inventory of this scope. 
The WRAP oil and gas states collaborated to develop and implement a 
uniform procedure for estimating area source emissions from oil and gas 
operations. WRAP initiated a study to focus on estimating emissions of 
pollutants with the potential to impair visibility near Class I areas 
in the West, particularly NOX emissions.
    Developing this inventory has been one of the most important tasks 
that needed to be completed before any of the western states could 
begin to look at imposing controls for improving visibility. In 
addition to developing these critical inventories, the State has also 
been very active in identifying and controlling emissions from the oil 
and gas industry. Wyoming has been ahead of the curve when it comes to 
controlling emissions from this industry to protect health standards. 
The EPA's recently finalized national oil and gas regulations to reduce 
air pollutants from the oil and gas production industry were patterned 
in large part after what Wyoming has been doing since the early 1990's. 
Since 2005, the State has been spending more time and resources to 
study and control emissions from natural gas production than any other 
sector.
    When it came time to address visibility impacts associated with the 
oil and gas industry for the RHR, Wyoming completed the required 
reasonable progress analysis. Wyoming also laid out reasons for why the 
time was not right for requiring additional controls on the industry to 
reduce visibility impairment, including lacking the very critical 
information to be supplied by the WRAP inventory study. In spite of 
Wyoming's diligent efforts, EPA disagrees with the State's reasoning 
for not adopting reasonable progress controls for the industry during 
the first planning period.
    EPA has completely misunderstood the purpose of the collaborative 
study to develop and implement a uniform procedure for estimating area 
source emissions from oil and gas operations. Wyoming could not have 
developed such a procedure on their own, and it continues to make no 
sense for each state in the West to develop independent emission 
inventories that cannot be compared to neighboring state inventories 
for a regional effort. EPA should understand this better than any 
individual state, since it relies on consistency in comprehensive 
national inventories to develop sound national rules. While Wyoming 
waits for the WRAP inventory study to be completed, it has not been 
idle with respect to developing information on the oil and gas 
industry. The State has invested huge resources in understanding 
emissions from this industry and EPA's suggestion that the State ``just 
develop the information'' shows a total lack of understanding of the 
problem.
    EPA's whole issue is about substituting its view regarding timing 
in place of Wyoming's reasoned judgment. Wyoming is hopeful that as it 
addresses ozone nonattainment it can also demonstrate the co-benefits 
to improving visibility just as EPA has done in the East by developing 
an ozone control strategy that also demonstrates adequate visibility 
improvement. Wyoming's effort goes beyond the first planning period, 
and is in accordance with the RHR. Wyoming respectfully requests that 
EPA acknowledge that Wyoming participation in the regional inventory 
development process satisfies reasonable progress for this first 
planning period.
    Response: We do commend the State for the work it is doing on 
developing more comprehensive information on oil and gas emissions 
although we disagree with this comment. As we stated in our

[[Page 5204]]

proposed notice, we disagree with the State's reasoning for not 
adopting reasonable progress controls for oil and gas sources. If the 
State determined that additional information on emission data from oil 
and gas sources was needed to potentially control oil and gas sources, 
the State should have developed the information in time for 
incorporation into their SIP.
    Comment: Wyoming's booming oil and gas industry has a significant 
and growing impact on visibility in the State's national parks and 
wilderness areas. Given the close proximity of some of Wyoming's 
largest planned oil and gas fields to the Bridger and Fitzpatrick 
wilderness areas--between just 10 and 200 miles--these magnificent 
lands in western Wyoming suffer the greatest visibility impairment due 
to oil and gas activities. The 4,399 additional approved wells in the 
Pinedale Anticline Oil and Gas Exploration and Development Project 
alone are projected to degrade visibility in the Bridger Wilderness by 
up to 6.1 deciviews, and to cause impacts greater than 1.0 deciview on 
45 days of each year. This impact is in addition to the impairment 
caused by the existing 1,819 wells in the Pinedale Anticline area, and 
the impacts from the numerous other existing and planned oil and gas 
fields in the region.
    Wyoming and EPA are obligated to reduce haze-causing emissions from 
the State's oil and gas industry to achieve ``reasonable progress'' 
toward the national goal of eliminating human-caused visibility 
impairment in Class I areas, 42 U.S.C. 7491(b)(2)(B), and doing so by a 
target year of 2064, 40 CFR 51.308(d)(1)(i)(B), (ii). See also 42 
U.S.C. 7491(b) (requiring ``measures as may be necessary to make 
reasonable progress toward meeting the national [visibility] goal''). 
Under Wyoming's Regional Haze SIP, natural visibility conditions would 
not be reached in Wyoming's Bridger and Fitzpatrick Wilderness Areas 
until 2165--more than 100 years past the 2064 goal set by EPA. Wyo. 
309(g) SIP at 115. Although EPA's proposed FIP includes additional 
measures that would hasten visibility improvement, EPA projects that 
Wyoming Class I areas still will not achieve the URP necessary to 
restore natural visibility by 2064. 78 FR 34788. Thus, EPA must 
demonstrate the reasonableness of its decision not to require emissions 
reductions from oil and gas activities that could make greater progress 
toward restoring natural visibility. 40 CFR 51.308(d)(1)(ii).
    Both Wyoming and EPA have failed to demonstrate that regulating 
emissions from Wyoming oil and gas development activities is not 
reasonable, in light of the facts that pollution-control technologies 
are technologically feasible, cost effective, and would improve 
significantly visibility across several affected Class I areas. EPA 
properly ``disagree[s] with the State's reasoning for not adopting 
reasonable progress controls for oil and gas sources.'' 78 FR 34765. 
Specifically, EPA rejects the State's view that it needs more time to 
collect information before it regulates the industry, stating ``[i]f 
the State determined that additional information was needed to 
potentially control oil and gas sources, the State should have 
developed the information.'' Id. EPA also rejects Wyoming's claim that 
it needs up to two years to develop necessary regulations, because 
``[i]f regulations are needed to implement reasonable progress 
controls, the State must develop them as part of the regional haze 
SIP.'' See also id. at 34764 n.43. The Conservation Organizations agree 
that Wyoming is not excused from regulatory requirements to commit 
reasonable emissions reductions from the oil and gas industry in the 
current planning period simply because Wyoming thinks more information 
about oil and gas activity emissions would be desirable. See 78 FR 
34765. As we pointed out in previous comments, ample information about 
oil and gas industry emissions and their visibility impacts has already 
been developed and published in numerous state and federal 
environmental impact statements.
    Response: EPA disagrees with the portions of this comment that take 
issue with our proposed action. We acknowledge the comment's support 
for our statement that Wyoming could not rely on the lack of data for 
the State's determination for oil and gas sources. With respect to the 
projected visibility impacts of future oil and gas production, we note 
that the analysis cited by the commenter relied on a background ammonia 
level of 1 ppb to determine visibility impacts on the Bridger 
Wilderness. Elsewhere, we explain why we reconsidered use of a 
background ammonia level of 2 ppb for modeling visibility impacts to 
the Bridger Wilderness; as a result we remodeled using both a monitored 
monthly varying concentration and an IWAQM default of 0.5 ppb for 
background ammonia. Thus, the analysis cited by the commenter may 
overstate visibility impacts. Furthermore, modeling of the visibility 
impacts alone does not quantify the potential visibility benefits of 
the controls the commenter supports.
    The comment cites 40 CFR 51.308(d)(ii), which requires states (or 
EPA in this instance) to demonstrate, when the RPGs fall short of the 
URP, that the RPGs are reasonable and achieving the URP is 
unreasonable. As we stated in our proposal, we found this to be the 
case due to the results of the four-factor analyses along with 
emissions from sources outside the WRAP domain. The commenter does not 
take issue with the latter, and we explain elsewhere that we continue 
to think that the controls considered by Wyoming are not so cost-
effective that it was necessarily unreasonable for Wyoming to require 
them.
    Comment: Although Wyoming's January 2011 SIP identified in 
particular a need for the WRAP to complete its ``Phase III'' inventory 
of Wyoming oil and gas emissions before requiring additional 
regulations of the industry, that inventory was completed in November 
2012. WRAP prepared technical memorandums specific to three areas in 
Wyoming--the Powder River Basin, the Wind River Basin, and the Greater 
Green River Basin--identifying both baseline emissions in 2006 and 
projected emissions in 2015. Indeed, WRAP even has completed ``Phase 
IV'' of its emissions inventory project, updating oil and gas industry 
baseline emissions as of 2009 for specific regions, including all three 
regions of Wyoming that were evaluated in Phase III. Thus, Wyoming has 
no justification based on incomplete data for refusing to identify oil 
and gas emissions control technology to satisfy reasonable progress 
requirements. And there should be no reason for EPA to accept Wyoming's 
invalid and outdated claim that more emissions information is needed 
when that information was available for more than six months prior to 
EPA's most recent Wyoming regional haze proposal.
    Response: We disagree with this comment to the extent that it 
argues we should not approve Wyoming's decision to not impose controls 
on oil and gas sources. We did state in our proposal that the lack of 
emissions data was not an appropriate justification for Wyoming's 
decision to not impose controls on oil and gas sources. Instead, we 
proposed to approve Wyoming's decision based on the cost of controls 
and on the application of minor source BACT. The comment does not 
identify anything in the November 2012 data that affects that rationale 
and does not explain how emissions data would change the cost of 
controls or the application of the SIP-approved minor source BACT 
provisions. Thus the comment does not give a reason for us to change 
our decision.
    Comment: While EPA rejects Wyoming's rationale for refusing to 
limit haze causing pollutants from this

[[Page 5205]]

booming industry, EPA provides insufficient rationale of its own to 
justify the omission. EPA provides two reasons for proposing to accept 
the State's plan not to require NOX emissions reductions 
from Wyoming oil and gas sources. First, ``the most reasonable controls 
are for compressor engines, which the State already controls through 
its minor source BACT requirements.'' 78 FR 34765 & n.25 (citing 
Wyoming Air Quality Standards and Regulations, Chapter 6, Section 2). 
Second, ``while the costs of some controls are within the range of cost 
effectiveness values Wyoming, other states, and we have considered as 
reasonable in the BART context, they are not so low that we are 
prepared to disapprove the State's conclusion in the reasonable 
progress context.'' Id. at 34765. Neither contention is supportable.
    EPA is wrong that compressor engine NOX emissions are 
regulated through Wyoming's minor source BACT requirements. Wyoming's 
minor source BACT guidelines for the oil and gas industry only regulate 
VOC and hazardous air pollutants, not NOX. The guidelines 
make no provisions for NOX controls at all. See State of 
Wyoming, Oil and Gas Production Facilities Chapter 6, Section 2 
Permitting Guidance (presenting controls that apply to VOC and 
hazardous air pollutants, but not NOX).
    Moreover, EPA's singular focus on compressor engines overlooks the 
numerous other opportunities to significantly reduce haze-causing 
emissions from oil and gas operations. As even Wyoming's analysis 
demonstrates, cost-effective options are available to achieve high 
control efficiency of NOX emissions from drill rig engines, 
turbines, and process heaters. See 78 FR 34764 (Table 26).
    Response: We disagree with this comment. Chapter 6, Sections 2 and 
4, which are approved into the State's SIP, both require BACT for new 
source compressor engines for regulated pollutants, which includes 
NOX and VOC. These regulatory requirements should not be 
confused with the State's oil and gas permitting guidance, which is not 
part of the SIP. The State guidance document provides additional 
compliance information for select sources of oil and gas VOC emissions, 
such as dehydration units, pumps and tanks. There are many sources 
which are regulated by the State's SIP and required to apply controls 
that are not included in the oil and gas permitting guidance.
    We also explained the reason we discussed compressor engines in 
particular: the cost of controls for those sources was the most 
reasonable. For other oil and gas sources, the costs were generally 
higher. As we stated in our proposal, those costs were not so low that 
EPA could find it necessarily unreasonable for the State to not have 
adopted them. The comment gives us no reason to think otherwise.
    Comment: EPA's justification that the costs of available controls 
are reasonable, but not so low that EPA is willing to require them, is 
both arbitrary and factually flawed. See 78 FR 34765 (``the costs of 
some controls are within the range of cost effectiveness values 
Wyoming, other states, and we have considered as reasonable in the BART 
context''). EPA's justification is arbitrary because it has not 
identified any objective threshold or rationale for reaching the 
determination that costs, although low, are still too high to justify 
modifying Wyoming's SIP determination.
    Indeed, EPA rejected Wyoming's determination not to require 
reasonable progress controls for Dave Johnston Units 1 and 2, where the 
controls would cost approximately $1,000/ton of NOX removed. 
See id. at 34788 (``Given predicted visibility improvement of 
approximately 0.30 deciviews per unit at the most impacted Class I area 
and the fact that Wyoming's RPGs will not meet the URP, we find that it 
was unreasonable for the State to reject these very inexpensive 
controls.''). EPA's statement that control technologies with similar--
and even lower--costs were not justified for the oil and gas industry 
cannot be squared with this determination for Dave Johnston Units 1 and 
2. See id. at 34765. As shown in Table 26 (of the proposed FIP), 
emissions controls for compressor engines are available in the $16 to 
$1,200/ton range. Id. at 34764. Enhanced NOX-control 
technologies for drill rigs have cost-effectiveness values of $900 to 
$1,000/ton. Id.
    Controls for NOX emissions from turbines are very cost 
effective at around $560/ton. Id. All of these costs are at or below 
the costs that were deemed ``very reasonable'' at the Dave Johnston 
power plant and which led to EPA rejecting the State's reasonable 
progress control proposal. If finalized, EPA's contrary proposal for 
the Wyoming oil and gas industry would be arbitrary.
    Response: EPA disagrees with this comment. The comparison with the 
costs of controls at Dave Johnston Units 1 and 2 is not apropos. First, 
as explained elsewhere, certain oil and gas sources are subject to the 
State's SIP-approved construction permit program, including the 
requirement for minor source BACT. On the other hand, as explained 
below in response to PacifiCorp's comments, PacifiCorp did not identify 
(nor is EPA aware of) any NOX control measures for Dave 
Johnston Units 1 and 2. Second, we did not propose to reject the 
State's determination for these units solely on the basis of the cost-
effectiveness of controls. In addition, the State relied on 
impermissible factors to disregard the results of its own four-factor 
analysis. Third, to assist in determining whether the state's 
determination for Dave Johnston Units 1 and 2 was reasonable or not, we 
have quantified the visibility benefits of controls and decided that 
the State's determination was not so unreasonable that we were prepared 
to disapprove it. Neither the commenter nor EPA has equivalent data for 
the oil and gas sources that the commenter mentions. The visibility 
benefits of the commenter's suggested controls would of course vary 
considerably depending on the location of the source and other factors, 
and the data the commenter cites elsewhere regarding the bulk 
visibility impacts of oil and gas development do not address visibility 
benefits. Thus, the comparison with Dave Johnston gives no reason to 
change our decision. Finally, the RHR does not require EPA to establish 
a hard-and-fast dollar per ton threshold or other numeric criteria for 
determining when a State's decision to not impose controls on 
reasonable progress sources is unreasonable; rather all four factors 
are to be considered under the reasonable progress provisions of the 
RHR.
    Comment: Control technologies to reduce oil and gas industry 
NOX emissions are inexpensive and justified. Wyoming did not 
identify the cost of available controls as an impediment to their 
implementation, and Wyoming's own analysis demonstrated that cost-
effective controls to reduce oil and gas industry emissions are 
available. See Wyo. 309(g) SIP at 123-26. In addition to Wyoming's 
generic analysis, the Conservation Organizations have identified 
available control technology. For example, the Bureau of Land 
Management (BLM) recommended basic pollution-reduction strategies such 
as replacing internal combustion engines for compressors with gas 
turbines, installing SCR on drilling rig engines, using electric or 
natural gas-powered drilling rigs, and centralizing production 
facilities to reduce truck traffic. The cost-effectiveness of such 
technologies is a reason for requiring them as reasonable progress 
measures; costs are not a basis for allowing Wyoming to avoid 
requirements to

[[Page 5206]]

reduce the large impact of the State's oil and gas industry on Class I-
area visibility.
    Response: EPA disagrees with this comment. First, the commenter 
incorrectly suggests that a ``generic'' analysis, rather than a source-
specific analysis, of the cost of controls for oil and gas sources 
violates the RHR. In the reasonable progress context, the cost-of-
compliance factor can be interpreted to encompass either the cost of 
compliance for individual sources or the cost of compliance for source 
categories. The language of 40 CFR 51.308(d)(1)(i)(A), cited by the 
commenter, does not explicitly require a source-specific analysis of 
the costs of compliance, contrary to the commenter's assertion. With 
respect to the control measures identified by BLM and referred to by 
the commenter, neither the comment, nor the letter cited in the 
comment, nor the records of decisions by the BLM cited by the letter, 
provide any data on the cost-effectiveness of these measures. The 
comment has no basis to describe the control measures identified by BLM 
as cost-effective.
    Comment: When a SIP fails to establish an emissions reduction 
strategy that would achieve natural visibility conditions by 2064, as 
is the case in Wyoming, the state must demonstrate that the underlying 
``uniform rate of progress'' is ``not reasonable[,] and that the 
progress goal adopted by the State is reasonable.'' 40 CFR 
51.308(d)(1)(ii); see also EPA, Guidance for Setting Reasonable 
Progress Goals Under the Regional Haze Program, at 2-3 (June 1, 2007) 
(demonstration should ``identify and analyze the measures aimed at 
achieving the uniform rate of progress and . . . determine whether 
these measures are reasonable''). EPA proposes RPGs that leave 
visibility impaired in Wyoming's affected Class I areas well beyond the 
2064 goal set by EPA. See 78 FR 34788. In light of EPA's rejection of 
every one of Wyoming's justifications for its conclusion that 
reasonable progress controls on the oil and gas industry are not 
reasonable, and EPA's lack of any valid rationale of its own to 
conclude that such measures are not reasonable, the failure to adopt 
any measures to reduce haze-causing emissions from the oil and gas 
industry cannot be supported and must be changed in the final rule.
    Response: EPA disagrees with this comment. In our proposal we 
specifically stated our rationale for agreeing with Wyoming's 
determination to not impose controls on oil and gas sources during this 
planning period. We respond elsewhere to the commenter's disagreement 
with that rationale. Because we are disapproving the State's RPGs, as 
part of our FIP we are imposing RPGs that are consistent with the 
controls in our FIP and the controls that we are approving in the 
State's SIP. We stated in our proposal that it was reasonable for the 
RPGs to fall short of the URP based not only on our consideration of 
the four statutory reasonable progress factors, but also based on 
emissions from sources outside the WRAP domain, which the commenter 
does not take issue with.
    Comment: EPA proposes to approve Wyoming's reasonable progress 
determinations for oil and gas sources. 78 FR 34765. However, EPA 
states that it ``disagree[s] with the State's reasoning for not 
adopting reasonable progress controls for oil and gas sources.'' Id. 
Wyoming explained in its SIP that it required additional information 
before it can determine whether and to what extent additional controls 
are necessary for oil and gas sources. Id. EPA thinks Wyoming should 
have obtained the additional information before submitting its SIP, 
though, EPA does not hold itself to this same standard. Nonetheless, 
EPA has previously recognized Wyoming's expertise and leadership in 
regulating the air quality impacts of oil and gas development. 76 FR 
52738, 52757 (Aug. 23, 2011). In light of Wyoming's leadership in 
regulating air pollution from oil and gas development, EPA should 
approve Wyoming's reasonable progress determination for oil and gas 
sources.
    Response: EPA disagrees with this comment. We are approving 
Wyoming's reasonable progress determination for oil and gas sources, 
although not on the State's basis. The notice cited in the comment 
relates to EPA's development of new source performance standards for 
oil and gas sources, which is not relevant to this action. As we 
explained in our proposal and elsewhere in these responses, the RHR 
sets certain requirements for reasonable progress for the first 
planning period and does not provide for deferring those requirements 
to later planning periods; thus, the State's basis for its reasonable 
progress determination for oil and gas sources is invalid. Finally, 
while we did not re-run the WRAP modeling to quantify our RPGs, the 
State did not modify its RPGs or re-run the WRAP modeling to reflect 
the controls the State selected.
    Comment: We received numerous comments urging EPA to require 
pollution controls on the oil and gas industry.
    Response: There are a number of provisions in the CAA that 
potentially apply to oil and gas sources. With respect to the 
requirements of the RHR for those sources, we have evaluated Wyoming's 
submittal and we are approving it.
b. Dave Johnston Units 1 and 2
    Comment: Wyoming did not evaluate the effectiveness of the 
LNB+OFA+SCR option. Instead, Wyoming assumed addition of SCR to these 
currently uncontrolled EGUs would only reduce NOX emissions 
by 79% down to 0.12 lb/MMBtu on an annual average basis, although it is 
generally assumed that SCR can reduce NOX emissions by 90% 
or down to 0.05 lb/MMBtu (or lower). Wyoming has not provided any 
documentation or justification to support the higher emission rates 
used in its analyses. Such an approach at Johnston adversely biases the 
cost-benefit analysis and is inconsistent with other EPA analyses.
    Response: The commenter has incorrectly assumed that a 90% control 
efficiency can be achieved in all SCR applications regardless of the 
input NOX emission rate or other parameters. In addition, we 
note that the emission rate analyzed by Wyoming, 0.07 lb/MMBtu, was on 
a 30-day rolling average basis, not an annual basis. Regardless, we 
agree that SCR can in most cases achieve a performance rate of 0.05 lb/
MMBtu on an annual basis. (See section IV.C.4 of this rulemaking for 
more information on the control effectiveness of SCR.) We have revised 
the SCR costs for Dave Johnston Units 1 and 2 accordingly. However, as 
explained elsewhere, the revised costs for SCR have not led us to 
change our determination that the State was reasonable in not selecting 
SCR for Units 1 and 2.
    Comment: Wyoming has assumed that Dave Johnston Unit 1 and Unit 2 
emitted at 0.57 lb/MMBtu on an annual basis and used this as the 
baseline condition from which to calculate the control efficiency it 
used for each control option. However, our review of CAMD data back to 
2000 shows that the highest annual NOX emission rate for 
Unit 1 was 0.474 lb/MMBtu (2002) and 0.460 lb/MMBtu for Unit 2 (2006). 
For the 2001-2003 baseline period, annual NOX emissions were 
0.46 and 0.44 lb/MMBtu for Johnston Unit 1 and Unit 2, respectively. 
Thus, Wyoming's proposal to reduce NOX to 0.20 lb/MMBtu with 
LNB+OFA represents a 56% reduction instead of 65% assumed by Wyoming.
    Response: In our revised cost analysis, we used baseline emissions 
for Dave Johnston Units 1 and 2 that reflect annual average emissions 
between 2001 and 2003, as found in the CAMD emissions system. These 
baseline rates

[[Page 5207]]

are 0.45 lb/MMbtu and 0.41 lb/MMBtu, respectively. This corresponds to 
a 56.0% and 54.6% reduction, respectively.\238\ Therefore, our revised 
cost analysis has addressed the concern raised by the commenter. As 
explained elsewhere, our revised costs have been taken into account, 
along with our revised visibility modeling, in our decision to approve 
the State's determination to not impose controls at Units 1 and 2.
---------------------------------------------------------------------------

    \238\ Staudt memo, Tables 2 and 3.
---------------------------------------------------------------------------

    Comment: The Conservation Organizations agree with EPA that 
reasonable progress controls for NOX emissions are needed 
for Dave Johnston Units 1 and 2. EPA correctly found that it was 
unreasonable for Wyoming to reject cost effective NOX 
controls that would improve visibility. EPA proposes to require only 
LNBs/OFA to achieve a NOX emission limit of 0.20 lb/MMBtu 
(30-day rolling average). While we commend EPA for proposing a FIP to 
reduce NOX emissions from Dave Johnston Units 1 and 2, we 
urge EPA to require SCR plus LNBs/OFA to meet a NOX emission 
limit of 0.05 lb/MMBtu to achieve reasonable progress. Although EPA 
concluded that the cost of SCR is not justified by the projected 
visibility improvement, EPA's analysis unreasonably assumed that SCR 
would only achieve a NOX emission rate of 0.12 lb/MMBtu, 
even though an emission rate of 0.05 lb/MMBtu is readily achievable. 
Correcting for this error, it appears that SCR at Dave Johnston Units 1 
and 2 is very cost effective at $2,001 and $1,987/ton of NOX 
removed, respectively. Accordingly, EPA should reconsider requiring SCR 
at Dave Johnston Units 1 and 2 to meet reasonable progress 
requirements.
    Response: As discussed in our proposed rulemaking, we have revised 
the SCR cost analysis for Dave Johnston Units 1 and 2 to reflect the 
installation of LNB and OFA. However, our revised cost effectiveness 
values of $3,496/ton and $3,672/ton, respectively, are much higher than 
those suggested by the commenter. We also note that the incremental 
costs for this option are high, at $9,798 and $9,588 per ton, 
respectively. In light of this, and our revised modeling results, we do 
not find it unreasonable for the State to not have imposed SCR on these 
units.
    Comment: EPA's conclusion that the addition of SCR is not justified 
due to the ``small incremental visibility improvement'' is based upon a 
flawed visibility analysis that over-values the addition of LNB + OFA 
and under-values the addition of SCR. Furthermore, the degree of 
visibility improvement is not one of the four statutory factors to be 
considered under the reasonable progress provisions of the RHR. 
Incremental visibility improvement is not mentioned anywhere in the 
reasonable progress provisions or BART Guidelines and EPA cannot create 
a new criterion for the sole purpose of eliminating a control option 
that is reasonably cost-effective and would yield a significant 
visibility improvement.
    Response: As discussed elsewhere in this rulemaking, we have 
corrected the modeling analysis for Dave Johnston Units 1 and 2, and 
the commenter's concerns regarding our methodology have been addressed. 
Our revised modeling analysis shows that the visibility improvement 
associated with SCR with LNB and OFA at Units 1 and 2 is 0.18 deciviews 
and 0.18 deciviews, respectively. The visibility improvement associated 
with LNB and OFA is 0.12 deciviews and 0.11 deciviews, respectively. We 
continue to find that the additional visibility improvement is not 
significant enough to warrant selection of SCR with LNB and OFA for 
these reasonable progress sources. As discussed earlier, we also find 
that the visibility improvement from LNBs and OFA does not justify us 
requiring reasonable progress controls on these two units. While it is 
true that incremental visibility improvement is not among the four 
statutory reasonable progress factors, the RHR does not prohibit EPA 
from assessing visibility improvement, in addition to the four 
statutory reasonable progress factors, when considering controls at 
potentially affected sources. We did not create a new criterion for the 
sole purpose of eliminating SCR at Dave Johnston Units 1 and 2; 
instead, we think it appropriate to consider visibility improvement 
when assessing control options for reasonable progress, especially when 
taking into account the purposes of the RHR. In comparing control 
options and selecting one, it is appropriate to compare the visibility 
improvement (that is, to compute the incremental visibility 
improvement) for each option.
    Comment: EPA is proposing that the FIP NOX BART for Dave 
Johnston Units 1 and 2 is LNBs with OFA at an emission limit of 0.22 
lb/MMBtu (30-day rolling average). EPA provided no reason for rejecting 
addition of SCR even though: (1) Cost/ton was $3,300-$3,400, which is 
less than the $3,900/ton accepted at Laramie River Unit 3; (2) 
Visibility at the most-impacted Class I area would improve by more than 
0.4 deciview (which is greater than the 0.3 deciview improvement for 
EPA's proposal; (3) Cumulative visibility improvement would exceed 0.6 
deciviews (versus EPA's proposed 0.43 deciview improvement for Dave 
Johnston Unit 2 at Wind Cave and Badlands); (4) Cost-effectiveness is 
$15 million/deciview at Wind Cave (versus $27,798,246/deciview at 
Badlands due to application of SCR to Laramie River Unit 3); (5) 
Cumulative cost-effectiveness is less than $10 million/deciview (versus 
$10,140,825/cumulative deciview due to application of SCR to Laramie 
River Unit 2.)
    We believe that SCR is Reasonable Progress for Dave Johnston Units 
1 and 2. Under the EPA proposal, Dave Johnston Units 1 and 2 would each 
contribute over 0.9 deciview impairment at Wind Cave National Park (and 
0.7 deciviews at Badlands National Park). With the addition of SCR, 
impairment would drop to less than 0.5 deciviews for each unit.
    Response: We disagree with this comment. We have responded in 
detail to the use of a $/deciview metric in section V.D.1.b of this 
final rulemaking action. In addition, as stated above, our revised 
modeling analysis shows that the visibility improvement associated with 
SCR with LNB and OFA is 0.18 deciviews and 0.18 deciviews, 
respectively. By contrast, the visibility improvement associated with 
LNB and OFA, is 0.12 deciviews and 0.11 deciviews, respectively. We 
continue to find that the additional visibility improvement is not 
significant enough to warrant selection of SCR with LNB and OFA for 
these reasonable progress sources, and as discussed earlier in our 
response to comments, we do not find the visibility improvement, when 
considered with the other reasonable progress factors, from LNBs and 
OFA warrants the implementation of reasonable progress controls. In 
addition, as we discuss above, the revised incremental costs for SCR 
that we present above are sufficiently high for us to conclude that it 
is reasonable to not impose SCR on Dave Johnston Units 1 and 2.
    Comment: EPA acknowledged that, for a reasonable progress analysis, 
only four factors must be analyzed. Indeed, the CAA clearly requires 
only four factors be analyzed. 42 U.S.C. 7491(g)(1). EPA employed the 
four-factor reasonable progress analysis for the other two Wyoming 
reasonable progress sources: oil and gas sources and the Mountain 
Cement Company plant. EPA has approved other regional haze SIPs where 
the state employed this same four-factor analysis, including Nevada. 
For both the oil and gas sources and the Mountain Cement Company

[[Page 5208]]

plant, EPA disagreed with Wyoming's reasonable progress analysis and 
found ``cost effective'' NOX controls could be employed, but 
EPA did not require those NOX controls because the costs 
were ``not so low that we are prepared to disapprove the State's 
conclusion in the reasonable progress context.'' If EPA found the 
NOX controls ``cost effective'', then PacifiCorp is unclear 
what additional cost analysis was performed, or what the statutory or 
regulatory basis for EPA's additional cost analysis may be. EPA does 
not differentiate PacifiCorp's Dave Johnston Units 1 and 2 from the oil 
and gas sources or the Mountain Cement Company plant.
    Also, EPA has approved other reasonable progress SIPs where the 
state is not meeting the URP, but has determined that no reasonable 
progress controls are required for the initial planning period. (See 77 
FR 30248, 30256-30257; SIP Approval for Idaho). Here, EPA admitted that 
Wyoming ``provided a four factor analyses that adequately evaluated the 
required factors'' for Dave Johnston Units 1 and 2, but then 
arbitrarily concluded ``it is also appropriate for this facility to 
consider a fifth factor for evaluating potential reasonable progress 
control options--the degree of visibility improvement that may 
reasonably be anticipated from the use of reasonable progress 
controls.''
    EPA justified its decision by citing to EPA guidance on states 
setting reasonable progress goals. However, the referenced guidance 
does not support EPA's position for several reasons:
     The guidance concedes it is ``merely guidance and that 
States or the . . . [EPA] may elect to follow or deviate from this 
guidance, as appropriate.'' EPA cannot find Wyoming acted 
``unreasonably'' when it chose not to apply discretionary guidance.
     The guidance identifies several factors that EPA did not 
include in its proposed regional haze FIP, such as the ``control 
measures and associated emission reductions that are expected to result 
from compliance with existing rules.'' EPA cannot criticize Wyoming for 
not following the guidance when EPA itself chose not to apply part of 
the same guidance in the EPA regional haze FIP.
     The guidance suggests that air quality models be used to 
estimate ``the improvement in visibility that would result from the 
implementation of the control measures you have found to be reasonable 
and compare this to the uniform rate of progress.'' Here, EPA has no 
``modeling results'' demonstrating the alleged improvement in 
visibility from the suggested NOX controls and the impact on 
the URP.
     The States, not EPA, are to determine the 
``reasonableness'' of RPGs and are given flexibility to do so.
     The guidance clearly indicates that a state must support 
its RPGs ``based on the statutory factors,'' which EPA admits Wyoming 
did.
     Finally, the guidance explains that no additional 
reasonable progress controls may be needed for the first planning 
period.
    Response: EPA disagrees with this comment. With respect to the 
reasonable progress determination for the Mountain Cement facility, the 
cost effectiveness of potential controls is generally higher than the 
controls we proposed for Dave Johnston Units 1 and 2. As we stated in 
our proposal, although the costs for potential controls for the 
Mountain Cement facility might be considered reasonable in the BART 
context, in the reasonable progress context those costs were not so low 
that we were prepared to disapprove the State's determination to not 
impose controls. That was not the case for Dave Johnston Units 1 and 2, 
for which combustion controls were significantly more cost-effective. 
Similar reasoning applies to the cost-effectiveness of controls for the 
oil and gas sources; in addition, as we noted in the proposal, Wyoming 
generally applies minor source BACT to these sources.
    As a result, EPA determined that we should perform visibility 
modeling to assess the visibility benefits of controls on Dave Johnston 
Units 1 and 2. As explained elsewhere, we considered it appropriate to 
assess, in addition to the four statutory reasonable progress factors, 
the visibility improvement from potential controls at these units, 
particularly in light of the purposes of the RHR. In this instance, the 
revised visibility modeling has confirmed that the State's decision to 
not impose controls on Dave Johnston Units 1 and 2 (although not the 
State's rationale) was not so unreasonable that EPA is compelled to 
disapprove it.
    EPA also disagrees that EPA's proposal was inconsistent with the 
cited notice proposing action on the Idaho Regional Haze SIP. In that 
notice, EPA stated, ``EPA agrees with Idaho's conclusion that 
additional controls of non-BART point sources for reasonable progress 
purposes are not reasonable at this time, because even though there are 
cost effective controls identified, visibility improvement is 
anticipated to be relatively small.'' (77 FR 30248, 30256, May 22, 
2012) (emphasis added). To derive that conclusion, EPA examined the 
modeled visibility impacts for the BART eligible sources and noted 
that, in Idaho's case, the sources with a Q/d of less than 26 had 
visibility impacts of less than 0.5 deciviews. EPA conservatively 
inferred from this that other potentially affected sources in Idaho 
with a Q/d of less than 20 would likely also have visibility impacts of 
less than 0.5 deciviews. In contrast, our original proposal showed 
modeled benefits at each unit of 0.3 deciviews from combustion 
controls. As a result, we reject the comparison with the notice 
proposing action on the Idaho Regional Haze SIP. Nonetheless, based on 
our revised modeling we have reconsidered our proposed determination to 
require LNBs and OFA on these two units, and now do not find the 
State's decision to not impose controls to be unreasonable.
    Furthermore, we are not disapproving the State's determination of 
which sources should be considered potentially affected sources. 
Wyoming reasonably used a Q/d threshold of 10 for determining the set 
of potentially affected sources, and the State selected (among others) 
Dave Johnston Units 1 and 2. That Idaho used a different threshold does 
not show that Wyoming's choice was unreasonable. However, even if EPA 
in the first instance was selecting potentially affected sources, we 
might also consider it reasonable to select Dave Johnston 1 and 2 based 
on a Q/d threshold.
    We disagree that we cited our reasonable progress guidance as part 
or whole of our basis for proposing to disapprove the State's 
reasonable progress determination for Dave Johnston Units 1 and 2. 
Instead, we briefly cited the guidance for the unremarkable proposition 
that the State must at a minimum consider the four statutory reasonable 
progress factors but could also take into account other relevant 
factors. Our proposed disapproval was not based on that proposition, 
but was based on the State's inadequately supported determination to 
not impose controls, which relied on impermissible factors. Although 
the commenter argues that it is the State, and not EPA, that should 
determine reasonable progress, as explained above we are required to 
evaluate the State's reasonable progress determinations. 40 CFR 
51.308(d)(1)(iii).
    We disagree with the commenter's other statements regarding the 
guidance. While the State did assess the four statutory reasonable 
progress factors for Dave Johnston Units 1 and 2, the guidance does not 
suggest that a state may ignore the results of that assessment for 
impermissible reasons such as a claimed lack of authority. The

[[Page 5209]]

guidance also does not suggest that a state may per se choose to impose 
no reasonable progress controls regardless of the state's assessment of 
the four statutory factors.
    Guidance aside, both the Act and the RHR explicitly require the 
state to consider the four factors for potentially affected sources. 
CAA section 169A(g)(1) (``shall be taken into consideration''); 40 CFR 
51.308(d)(1)(i)(A). Assessing the four factors but ignoring the results 
of that assessment for invalid reasons such as lack of authority to 
impose controls does not amount to considering the factors and violates 
the explicit requirements of the Act and the RHR.
    With respect to control measures expected to result from compliance 
with existing rules, the commenter did not identify any such 
NOX control measures for Dave Johnston Units 1 and 2, nor 
are we aware of any. Finally, we did independently run CALPUFF to model 
the visibility improvement from potential controls at Dave Johnston 
Units 1 and 2 and it is part of the basis for our final decision.

F. General Comments

1. Replacement of FIP Elements With SIP
    Comment: EPA has proposed to disapprove the monitoring, 
recordkeeping, reporting, and RAVI portions of Wyoming's SIP. 78 FR 
34788. Wyoming acknowledges these deficiencies in its SIP and commits 
to making the necessary revisions. However, Wyoming will revise its SIP 
in a manner that comports with statutory and regulatory processes. 
Unlike EPA, Wyoming will not shortcut legal processes designed to 
ensure FLM consultation and public participation to meet an arbitrary 
deadline EPA has established with special interest groups in litigation 
to which Wyoming was not a party. Such arbitrary deadlines defeat the 
cooperative federalism Congress intended to guide CAA implementation by 
needlessly expediting the process, tying EPA's hands, and precluding 
the State from an opportunity to revise its SIP. In this context, EPA's 
promise--to ``propose approval of a SIP revision as expeditiously as 
practicable if the State submits such a revision and the revision 
matches the terms of our proposed FIP,'' id. 34738--rings hollow.
    Response: We appreciated the State's willingness to make the SIP 
revisions necessary to correct the deficiencies with the monitoring, 
recordkeeping, reporting, and RAVI requirements. Once EPA receives the 
SIP revisions from the State, EPA will work as expeditiously as 
practicable to review such revisions and approve the State's revisions 
if they meet the terms of our FIP. We have responded to other comments 
elsewhere in this document.
2. Public Comment
    Comment: DEQ and Governor Mead requested that EPA defer its hearing 
until sixty days after the date EPA first released its proposal, with 
an additional thirty days of comment after the hearing. See, e.g., 
Letter from Todd Parfitt, Director, DEQ, to Shaun McGrath, Region 8 
Administrator, EPA (June 14, 2013). Although EPA agreed to hold two 
additional public hearings and provide an additional thirty days for 
public comment, EPA did not provide the time for public participation 
that Wyoming requested, evidently because EPA wanted to meet the 
deadline for final action it established with the special interest 
groups. Thus, while EPA did not hesitate to extend that deadline on 
multiple occasions when it benefitted EPA and the special interest 
groups, EPA refused to provide the additional time Wyoming requested 
for the benefit of the State.
    We also received comments from other parties that we should extend 
the public comment period so that there is more time to review and 
comment on our action. Some commenters specifically requested a 60-day 
extension of the comment period. One commenter noted that if the 
driving force for the short timeframe in this instance is a consent 
decree to which the EPA is a party, that no agreement between an agency 
and any interested party, whether as part of litigation or not, should 
or can modify requirements of law for a meaningful opportunity for 
public comment.
    Response: EPA took several steps to provide the opportunity for 
meaningful public comment. In addition to the initial 60-day public 
comment period, we extended the public comment period from August 9, 
2013, until August 26, 2013. In doing so, we took into consideration 
how an extension might affect our ability to consider comments received 
on the proposed action and still comply with the terms of the consent 
decree deadline,\239\ which at the time required our final action 
signed by the Administrator on or before November 21, 2013. 
Additionally, we could not extend the comment period any further and 
still have time to respond to the immense amount of public comments we 
anticipated receiving. As the commenter notes, EPA also added two 
public hearings and we received substantial comments at these 
additional hearings. We find that the comment period provided for the 
Wyoming Regional Haze SIP and FIP exceeds CAA requirements and is 
reasonable and consistent with what the Agency has provided on other 
FIP and SIP actions. For example, EPA provided a 60-day comment period 
for both the Montana and North Dakota regional haze actions (see 77 FR 
23988 and 76 FR 58570, respectively.) In addition, in our first 
proposal on Wyoming regional haze (77 FR 33022), we provided a 60-day 
comment period with no objections from interested parties.
---------------------------------------------------------------------------

    \239\ WildEarth Guardians v. McCarthy, Case No. 1:11-cv-0001-
CJA-MEH.
---------------------------------------------------------------------------

    Finally, the State and impacted sources have had many years to 
prepare and submit an approvable SIP to EPA. As detailed in the Docket 
for this action, the State received numerous detailed comment letters 
from EPA on many issues and also participated in meetings with EPA. 
Indeed, the fact that the State was able to prepare an extensive 33-
page document and provide extensive comments at the various public 
hearings, all within the allotted time period, supports EPA's 
contention that the 77-day time period for this proposed rulemaking was 
reasonable.\240\
---------------------------------------------------------------------------

    \240\ In fact, the State received the proposed notice on May 28, 
2013, two business days after the proposal was signed, and the 
proposed notice was posted on the Region's Internet site on May 28, 
2013, well in advance of the Federal Register publication on June 
10, 2013.
---------------------------------------------------------------------------

3. Economic Concerns
    Comment: We received numerous comments that the FIP will cost 
anywhere from an additional $300 million to $1 billion more than the 
State's SIP, but provide no perceptible improvement in visibility when 
compared with the SIP. We received numerous comments that EPA's FIP 
would lead to higher electricity costs to consumers and job losses at a 
time when the economy and people cannot afford an additional burden.
    Response: We disagree with these comments. We have addressed the 
issue of perceptible visibility improvement in section V.C.5 of this 
final rulemaking. In addition, it is not EPA's intention to endanger 
the economic viability of or to place an undue burden on PacifiCorp or 
Basin Electric's customers. EPA has considered the comments on these 
issues very carefully. Regarding the legal basis for our decision, 
neither the CAA nor the RHR requires states or EPA to consider the 
affordability of controls or ratepayer impacts as part of a BART 
analysis. Rather, the CAA and RHR require consideration of the 
following factors, which as detailed elsewhere in

[[Page 5210]]

our notice, we have fully considered: ``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.''
    Comment: One commenter noted that visitors to Wyoming's parks 
notice when the air is dirty, which can have a direct impact on 
tourism, the second largest business in the State. According to 
Wyoming's Office of Tourism, the travel and tourism industry in Wyoming 
creates 30,000 jobs and generates $730 million in employment earnings 
and $2.8 billion in travel expenditures annually. Over 3.5 million 
people visit Yellowstone National Park each year. The Commenter also 
indicated that the proposed plan will also reduce health care costs in 
the State. Combined, coal plants in Wyoming emit over 60,000 tons of 
NOX pollution and almost 65,000 tons of SO2 
pollution annually. The Clean Air Task Force estimates that coal plant 
pollution in the State results in over $850 million in preventable 
health care costs.
    Response: We acknowledge the commenter's points.
    Comment: We received a comment that EPA failed to calculate the 
costs of the proposed rule that will be passed on to residential and 
business customers.
    Response: Explained elsewhere in this document, we have taken these 
costs into consideration.
    Comment: We received numerous comments that EPA's action could have 
the potential to shift the energy balance in favor of gas, rather than 
coal, and that this shift may force utilities to convert their power 
plants from coal to natural gas. Commenters expressed particular 
concern over the potential conversion of Naughton Unit 3, and possibly 
Naughton Units 1 and 2. Commenters expressed concern over the potential 
impacts natural gas conversion could have on local economies.
    Response: As stated above, the CAA and RHR require consideration of 
the five statutory factors. Based on our consideration of these 
factors, EPA determined the appropriate emission limit for BART for 
each unit. Sources have the choice of how to meet that limit, including 
conversion to natural gas. EPA's action does not require any source to 
convert to natural gas, as all of the requirements in our FIP can be 
met with combustion and post-combustion control technology.
    Comment: One commenter stated that coal-fired plants in Wyoming 
have installed over one billion dollars in additional air quality 
controls and that, according to EPA standards, Wyoming has better 
visibility than virtually any other state in the country. Therefore, it 
seems unreasonable, illogical and, frankly, irrational that the EPA 
would demand Wyoming businesses and homeowners foot the bill for 
another one billion dollars in emission controls that have little 
probability of improving the quality of lives or the livelihoods of our 
citizens and, in fact, has a great potential to harm our people and our 
state.
    Response: EPA disagrees with this comment. EPA carefully considered 
the five statutory factors and determined that there are additional, 
cost-effective controls that will result in significant visibility 
improvement in Wyoming's Class I areas, and that these controls 
represent BART.
    Comment: One commenter was concerned that the investments mandated 
under the FIP will have significant adverse impacts on the quality and 
reliability of service provided to Wyoming ratepayers. The SIP is a 
well vetted plan by the State and its stakeholders that, in association 
with other regulatory requirements such as the construction authority 
process, assures that Wyoming utilities will be able to comply with its 
requirements with the least amount of impact to customers. The FIP, on 
the other hand, with its more stringent control requirements and 
accelerated compliance deadlines, will assure not only that compliance 
is needlessly expensive, but that it is also rushed, that scheduled 
outages cannot be timed to minimize the cost of replacement power, and 
that third party vendors will have free reign in determining how much a 
particular project costs. To the extent that schedules cannot be met, 
non-compliant plants will be forced out of service until the work is 
done. Such outages will necessitate the purchase of replacement power 
in the market and will result in diminishing system reserves, all of 
which will jeopardize system reliability and increase costs for 
ratepayers.
    Response: We appreciate the commenter's concerns, but note that the 
commenter has provided no data to support these assertions.
    Comment: The companies working with Wyoming have scheduled shutdown 
and installation on a schedule that will allow them to maintain service 
to their customers. The new timeline demanded in the re-proposal would 
threaten both service interruptions and an increased risk of having to 
spot purchase energy which would be an additional increase of costs to 
residential, business, manufacturing, and agricultural customers.
    Response: We appreciate the commenter's concerns, but note that the 
commenter has provided no data to support these assertions. 
Additionally, CAA section 169A(b)(2)(A) requires subject-to-BART 
sources to install BART and comply with any applicable emission limits 
``as expeditiously as practicable.'' The Act defines this term to mean 
``as expeditiously as practicable but in no event later than five years 
after . . . the date of promulgation.'' CAA section 169A(g)(4). 
Consequently, the final rule appropriately provides that the BART units 
must comply with the emission limits as expeditiously as practicable 
but in no event later than five years after the date of promulgation of 
the final rulemaking.
4. National Ambient Air Quality Standards (NAAQS)
    Comment: The EPA is duty-bound to ensure that the proposed SIP does 
not interfere with attainment and maintenance of the NAAQS, in 
accordance with section 110(l) of the CAA. Thus, the EPA must ensure 
that the proposed SIP and the proposed FIP adequately limit air 
pollution in order to safeguard public health.
    In this case, we are concerned that in proposing to approve 
portions of Wyoming's regional haze plan, the EPA has not demonstrated 
that the proposal adequately safeguards the 2008 8-hour ozone NAAQS 
(see 40 CFR 50.15), the newly promulgated 1-hour nitrogen dioxide 
(``NO2'') NAAQS (see 40 CFR 50.11(b)), the newly promulgated 
1-hour SO2 NAAQS (see 40 CFR 50.17), the 2006 24-hour 
PM2.5 NAAQS (see 40 CFR 50.13), and the 2012 annual 
PM2.5 NAAQS (see 78 FR 3086 (Jan. 15, 2013)).
    We are particularly concerned that the EPA overlooked its 110(l) 
obligations under the CAA given that, although the Proposed Rule may 
lead to emission reductions, no analysis or assessment has been 
prepared to demonstrate that even after these emission reductions, the 
recently promulgated NAAQS will be met. In this case, we are 
particularly concerned that the recently promulgated 1-hour 
NO2 and SO2 NAAQS could be jeopardized, as well 
as the recently promulgated 2012 PM2.5 annual NAAQS. Indeed, 
many, if not most, of the proposed emission rates are based on 30-day 
rolling averages. There is no indication that meeting emission rates on 
a 30-day rolling average will ensure that 1-hour NAAQS will be 
sufficiently protected. Indeed, a source could comply with a 30-day 
rolling average limit, yet still emit enough pollution on

[[Page 5211]]

an hourly basis to cause or contribute to violations of the NAAQS, 
thereby interfering with attainment or maintenance. Further, there has 
been no analysis at all as to whether the recently promulgated 
revisions to the annual PM2.5 NAAQS will be protected.
    In this case, the EPA must either disapprove the Wyoming SIP over 
the State's failure to perform a 110(l) analysis or prepare its own 
110(l) analysis to demonstrate that the SIP will not interfere with 
attainment or maintenance of the NAAQS. Furthermore, the EPA must 
demonstrate that its FIP will not interfere with attainment or 
maintenance of the NAAQS. The EPA has not done so, rendering its 
proposed rule substantively flawed.
    Response: CAA section 110(l) provides that EPA ``shall not approve 
a revision of a plan if the revision would interfere with any 
applicable requirement concerning attainment and reasonable further 
progress . . ., or any other applicable requirement of'' the CAA. The 
commenter has not provided any evidence that the Wyoming Regional Haze 
SIP will interfere with any applicable requirement concerning 
attainment and reasonable progress or any other applicable requirement 
of the CAA, or that further analysis under 110(l) is necessary. To the 
contrary, the commenter acknowledges that the Regional Haze SIP 
revision will lead to emission reductions.
    The commenter asserts that it is not enough that the SIP will lead 
to emission reductions and that EPA must determine that the SIP will 
ensure the NAAQS are met. We disagree. The CAA and EPA's regulations 
require regional haze SIPs to address visibility impairment in 
mandatory Class I areas; attainment of the NAAQS is provided for 
through a separate SIP process. It is EPA's consistent interpretation 
of section 110(l) that a SIP does not interfere with attainment and 
maintenance of the NAAQS if the SIP at least preserves the status quo 
air quality by not relaxing or removing any existing emissions 
limitation or other SIP requirements. EPA does not interpret section 
110(l) to require a full attainment or maintenance demonstration for 
each NAAQS for every SIP revision. See, e.g., Kentucky Resources 
Council, Inc., v. EPA, 467 F.3d 986 (6th Cir. 2006); see also, 61 FR 
16050, 16051 (April 11, 1996) (actions on which the Kentucky Resources 
Council case were based).
    Thus, in this action, we need not determine whether a 30-day limit 
is adequate to protect a shorter-term NAAQS because the regional haze 
SIP is not required to ensure attainment of the NAAQS. The fact that 
the regional haze SIP specifies 30-day limits will not preclude Wyoming 
from adopting limits with a shorter averaging time, if at some future 
date such limits are found to be necessary and required by the CAA to 
protect the NAAQS.
5. Other
    Comment: We received over 250 comments in a general mass mailer 
campaign in support of our action. We received over 220 mass mailer 
comments on behalf of National Parks Conservation Association in 
support of our action. We also received numerous general comments from 
individuals and organizations in support of our action.
    Response: We acknowledge the commenters' support of our proposed 
action.
    Comment: We received over 850 mass mailer comments opposed to our 
action. We also received numerous general comments from individuals and 
organizations in opposition to our action.
    Response: We note the commenters' opposition to our proposed 
action.
    Comment: We received numerous general comments in opposition to our 
action that stated that the State's plan was good enough, that it would 
achieve appropriate emission reductions, and that it represented a 
balanced approach.
    Response: We note the commenters' opposition to our proposed 
action.
    Comment: We received numerous general comments in opposition to our 
FIP that stated that the visibility in Wyoming is not hazy and that 
Wyoming has some of the best air quality in the country.
    Response: We note the commenters' general qualitative observations, 
but note that the commenters did not provide any quantitative 
information to substantiate their comment.
    Comment: We received numerous comments that we should approve 
Wyoming's SIP because it represents collaboration between the State, 
industry, local governments, and the public.
    Response: We note the commenters' points, but as stated earlier, 
EPA can only approve a state's SIP if it meets the requirements of the 
CAA and EPA's implementing regulations.
    Comment: We received numerous comments that EPA's FIP will only 
reduce NOX by 2,900 tpy more than the Wyoming's SIP, which 
reduces NOX by 63,000 tpy. Other commenters went on to say 
that EPA's FIP will basically achieve the same emission reductions the 
State's SIP would by 2022.
    Response: We disagree with this comment. EPA's calculations show 
that our proposed FIP will result in approximately 17,000 tpy more 
NOX reductions than the State's SIP, through 2022 and 
beyond.
    Comment: We received numerous comments that the proposed FIP, along 
with other EPA regulations, are in support of EPA's hidden agenda to 
kill the coal industry and shut down coal-fired power plants.
    Response: As stated earlier, EPA's proposed action was based on its 
careful consideration of the five statutory factors in the CAA and 
related statutory and regulatory requirements.
    Comment: We received numerous comments that the State's SIP was 
created through coordination with PacifiCorp and other Wyoming 
industries and that it is based on sound science that complies with the 
CAA and provides a balance between achieving compliance with the RHR 
while ensuring reliable, affordable electricity.
    Response: We note the commenters' points, but as stated earlier, 
EPA can only approve a state's SIP if it meets the requirements of the 
CAA and EPA's implementing regulations.
    Comment: We received a comment that it is generally more hazy now 
than it was ten years ago and that the commenter was supportive of 
reducing haze.
    Response: We note the commenter's support.
    Comment: We received a comment that provided data that showed 
voters in Wyoming were supportive of continued implementation of the 
CAA and environmental protections for our environment. The commenter 
went on to say that the majority of voters thought environmental 
protection and a good economy were compatible and encouraged EPA to 
finalize its proposed action. The commenter urged EPA to ignore 
negative media attention its action has drawn, stating that the 
negative publicity was being driven by economic interests.
    Response: We note the commenter's support of our proposed rule.
    Comment: EPA has applied selective comment response to the 
development of its re-proposal and the public comment process, which is 
inappropriate. EPA's process has lacked transparency, particularly to 
the State. EPA has not acknowledged the Governor's comments submitted 
last year. The EPA has not acknowledged the DEQ's comments submitted 
last year. DEQ was not consulted in the re-proposal process. It would 
appear that EPA only considered select comments that support its 
predetermined agenda.

[[Page 5212]]

    Response: Consistent with our statutory obligations, we have 
evaluated all written and oral comments on the proposal rulemaking 
(placing all the comments received in the docket for this action at 
www.regulations.gov); determined whether any revisions to the proposed 
rule are warranted; and prepared the final rulemaking and supporting 
information. The final rulemaking decisions are accompanied by the 
bases for the decisions, explanations of major changes from the 
proposals, and a response to each of the significant comments submitted 
in written or oral presentations during the comment period, which 
includes responses to such comments submitted by the Governor and DEQ.
    Comment: For years, Wyoming has pursued developing a collaborative 
and professional relationship with the EPA, but with the regional haze 
SIP process, the EPA has not reciprocated the same cooperative effort. 
This lack of effort on the part of the EPA does not represent the 
intent of what performance partnership agreements are put in place to 
accomplish.
    Wyoming is a leader in collaboration. Whether it is hydraulic 
fracturing, Sage Grouse Core Area Development, or carbon sequestration, 
Wyoming has demonstrated a willingness, and really eagerness, to work 
with federal agencies, local government, and industry to create 
solutions that not only minimize detrimental impact, but may actually 
do the opposite: Encouraging sustainable economic growth in Wyoming.
    EPA's imposition of the FIP would pour a bucket of cold water on 
the solutions resulting from this type of collaboration. In its place, 
EPA risks disenfranchising ratepayers when industry has little choice 
but to transfer the costs associated with retrofitting the EGUs. 
Moreover, because EPA failed to consider the primary cause of regional 
haze in Wyoming and the Interior West--smoke from wildfires--it risks 
alienating local government and state cooperating agencies who will 
perceive EPA as being out of touch with the regulated community.
    Response: We disagree with this comment. EPA values its 
relationship with the State of Wyoming and prior to our proposed action 
had numerous meetings with State and industry representatives to 
explore ways in which the State could have addressed our long standing 
concerns with the approvability of the State's Regional Haze SIP. 
Regrettably, we were unable to find a path forward during those 
discussions that could have resulted in the submission of a fully 
approvable regional haze SIP. Nevertheless, we remain committed to 
working collaboratively with the State on future regional haze actions 
and encourage the State to submit a SIP revision that could potentially 
replace all or a portion of our FIP. We do note that in a previous 
action we finalized full approval of the State's 309 (SO2) 
portion of the Regional Haze SIP. In this action we will also be 
finalizing approval of many aspects of the State's 309g (NOX 
and PM) portion of the regional haze plan.
    Comment: We received numerous comments that the regulation of 
regional haze is focused on improving visibility, not public health.
    Response: We agree with the commenter that the CAA's visibility 
program and the RHR are focused on improving visibility and not public 
health.
    Comment: The Conservation Organizations submitted comments on July 
23, 2012 urging EPA not to finalize its proposal to approve Wyoming's 
participation in a Western Backstop Trading Program in lieu of 
satisfying BART requirements for SO2. Under 40 CFR 51.309, 
states within the Grand Canyon Visibility Transport region, including 
Wyoming, may adopt a BART-alternative for the state's SO2 
emissions provided that, among other things, the program is shown to 
provide for greater reasonable progress than would be achieved by 
application of BART pursuant to 40 CFR 51.308(e)(2). Wyoming`s 
alternative program does not satisfy this requirement. Accordingly, 
Wyoming must comply with BART requirements for all haze-causing 
pollutants, including SO2.
    Response: We finalized approval of the State's 309 SIP that 
includes the requirements for the Western Backstop Trading Program on 
December 12, 2012 (77 FR 73926). Because this comment pertains to that 
final rulemaking, it is not germane to this final rulemaking action.
    Comment: Unlike other programs, the regional haze program requires 
regular updates and reviews to ensure that reasonable progress is being 
made towards the ultimate goal ending in 2064. In fact, the State will 
be required to submit a progress report to EPA in 2013 and a new 
regional haze SIP in 2018. EPA should approve the Wyoming Regional Haze 
SIP, and reserve most of its arguments and concerns expressed in its 
regional haze FIP for consideration in Wyoming's 2018 regional haze SIP 
submittals. In the meantime, EPA can be assured that the significant 
emission reductions required under the Wyoming Regional Haze SIP, 
nearly all of which already have been installed, will continue to 
contribute to visibility improvement.
    Response: Because we have found that the State's Regional Haze SIP 
did not adequately satisfy the RHR requirements in full, we have a duty 
to promulgate a FIP during this planning period that meets those 
requirements.
    Comment: EPA pays undue attention to the ``health'' issues in its 
FIP. For reasons it does not explain, EPA's FIP discusses the asserted 
health impacts of fine particulates, when health impacts are not part 
of the BART analysis. The regional haze program is not a health-based 
program; rather, it is focused on aesthetics.
    Response: We disagree with this comment. In our proposed notice, we 
stated that ``PM2.5 can also cause serious health effects 
and mortality in humans and contributes to environmental effects such 
as acid deposition and eutrophication.'' 78 FR 34741. The commenter 
suggests that this brief informational statement somehow means that we 
based our BART determinations in part on the health impacts of 
PM2.5. This is not the case, as we clearly based our BART 
determinations on the five statutory factors, as required by section 
169(a) of the CAA and the RHR.
    Comment: In light of EPA's apparent coordination with the special 
interest groups and the particular influence those groups seemed to be 
exerting over EPA's regional haze program, Wyoming and eleven other 
states submitted to EPA a Freedom of Information Act (FOIA) request 
seeking communications between EPA and the special interest groups 
related to EPA action on regional haze SIPs. See Letter from P. Clayton 
Eubanks, Deputy Solicitor General, Office of Oklahoma Attorney General, 
to FOIA Officer, EPA (Feb. 6, 2013) (FOIA Request). EPA denied the 
states' public records request on the ground that the states' fee 
waiver request was invalid because the states ``have not expressed a 
specific intent to disseminate the information to the public.'' Letter 
from Larry F. Gottesman, National FOIA Office, EPA, to Clayton Eubanks, 
Deputy Solicitor General, Office of Oklahoma Attorney General (Feb. 22, 
2013). But see FOIA Request, at 5-9 (Feb. 6, 2013) (describing in 
detail the states' intent to disseminate the information to the 
public).
    The states appealed that plainly erroneous decision. See Letter 
from P. Clayton Eubanks, Deputy Solicitor General, Office of Oklahoma 
Attorney General, to National FOIA Officer, EPA (March 15, 2013) 
(Exhibit 4). On May 2, 2013, EPA's Office of General Counsel informed 
the states that it needed ``a

[[Page 5213]]

brief extension of time''--until May 15, 2013--to respond to the 
states' appeal. Electronic mail from Lynn Kelly, Attorney-Advisor, EPA 
Office of General Counsel, to P. Clayton Eubanks, Deputy Solicitor 
General, Office of Oklahoma Attorney General. Two weeks later, EPA 
again informed the states that it needed more time to review the 
appeal, promising a decision by May 31, 2013. Electronic mail from Lynn 
Kelly, Attorney-Advisor, EPA Office of General Counsel, to P. Clayton 
Eubanks, Deputy Solicitor General, Office of Oklahoma Attorney General 
(May 15, 2013).
    On that date, EPA denied the states' FOIA request, claiming the 
states' request ``fails to adequately describe the records sought[.]'' 
Letter from Kevin M. Miller, Assistant General Counsel, EPA Office of 
General Counsel, to P. Clayton Eubanks, Deputy Solicitor General, 
Office of Oklahoma Attorney General, at 1 (May 31, 20 13). But see FOIA 
Request, at 1-3 (describing in detail the records sought). In the face 
of EPA's blatant attempts to frustrate the states' right to access 
public records directly related to matters of great importance to the 
states and the public, the states sued EPA in federal court. Compl., 
Oklahoma v. EPA, No. 5:13-cv-00726-M (W.D. Okla. July 16, 2013).
    In related litigation seeking the documents that the states 
requested, as well as others, a federal judge has questioned EPA's 
truthfulness and concluded ``that leaders in EPA may have purposefully 
attempted to skirt disclosure under the FOIA.'' Mem. Op., at 13, 
Landmark Legal Found. v. EPA, No. 12-1726 (D.D.C. Aug. 14, 2013). One 
cannot help but to similarly question EPA's honesty and wonder what EPA 
is trying to hide.
    Response: EPA disagrees with this comment. EPA has not coordinated 
with environmental organizations regarding the outcome of this action. 
As we explain elsewhere, nothing in the consent decree requires any 
particular substantive outcome concerning Wyoming's Regional Haze SIP. 
With respect to the allegations made relating to FOIA litigation, EPA 
has fully responded to those claims in federal court. In any case, the 
issues in the FOIA litigation mentioned by the commenter, such as 
whether the FOIA requester reasonably described the records sought, are 
unrelated to the commenter's unsupported allegations of coordination 
with environmental organizations.
    With respect to the commenter's insinuations of bias, EPA firmly 
rejects them. We have given careful consideration to all comments and 
views submitted, regardless of their origin. In response to some 
comments--both from industry and from environmental organizations--we 
have acknowledged the merits of the comments and accordingly adjusted 
not only our technical analyses, but also our final determinations. We 
have also, at our discretion, considered comments from both industry 
and from environmental organizations that were submitted after the 
close of the comment period. It is hard to imagine what better evidence 
there could be that EPA is willing and able to rationally consider 
arguments and does not have an unalterably closed mind on the issues in 
this action. See Air Transp. Ass'n of Am., Inc. v. Nat'l Mediation Bd., 
663 F.3d 476, 487 (D.C. Cir. 2010) (standard for prejudgment of 
rulemaking issues) (citing Ass'n of Nat'l Advertisers, Inc. v. FTC, 627 
F.2d 1151 (D.C. Cir. 1979), cert. denied, 447 U.S. 921 (1980)).

VI. Non-Relevant Comments From EPA's Original Proposal

    The following is a summary of the significant comments, criticisms, 
and new data we received on our initial June 4, 2012 proposed 
rulemaking, which we are not responding to because they are no longer 
relevant to the action we proposed on June 10, 2013, or the specific 
regional haze related action we are taking in this final rulemaking.

A. General Comments

    Comment: EPA is proposing to calculate compliance with tons per 
year (tpy) BART emissions limits on a rolling 12-month basis. Based on 
EPA's proposal, the owner/operator is to calculate and record a new 12-
month rolling average emission rate from the arithmetic average of all 
valid hourly emission rates from the continuous emissions monitoring 
systems for the current month and the previous 11 months, and to report 
the result in tons. The calculation and compliance determination shall 
be performed at the end of each calendar month.
    Wyoming established BART emissions limits based on a 30-day rolling 
average in accordance with 40 CFR part 51 Appendix Y. Wyoming also 
established annual emissions limits for units with BART limits. For all 
units with BART limits, except Units 1 through 3 at Basin Electric's 
Laramie River Station, Wyoming based the annual emissions limits on the 
30-day averaged lb/hr emissions limit and full-time operation for 8,760 
hours per year.
    Wyoming deliberately established these limits on a calendar year 
basis to reduce recordkeeping and reporting burdens, without being any 
less stringent than what was prescribed under Appendix Y. Every year, 
when the Title V emissions inventory is submitted for each of these 
facilities, the reported annual emissions can be compared to the annual 
BART limits established in the State permits to determine compliance. 
Requiring compliance with a 12-month rolling average will result in 
unnecessary monitoring, recordkeeping, and reporting, as the 12-month 
rolling emissions limit would be based on full-time operation of the 
unit and the more stringent 30-day averaged lb/hr value.
    Comment: We received a comment from a coalition of physicians that 
expressed concerns over the health impacts from air pollution, 
particularly particulate matter and ozone.
    Comment: We received a comment from a member of the public that 
points out the amount of coal production in Wyoming and its 
contribution to carbon dioxide and climate change.

B. Basin Electric Laramie River

    Comment: Wyoming has overestimated the ability of SNCR to reduce 
emissions. EPA is basing its BART determination on the assumption that 
LNB+OFA+SNCR can achieve 0.12 lb/MMBtu on a 30-day rolling average. 
This means that addition of SNCR must reduce NOX emissions 
from the LNB+OFA strategy by another 48%. Given the sensitivity of SNCR 
to boiler operation, size, and configuration, we are concerned that 
SNCR may not be able to achieve the proposed level of performance on a 
consistent basis. For example, our query of CAMD data for 2011 found no 
EGUs with SNCR (out of 3,621 coal-fired EGUs) that met 0.12 lb/MMBtu 
each month.
    Comment: EPA should ensure that SNCR plus LNBs/OFA are capable of 
meeting the proposed NOX limit of 0.12 lb/MMBtu, which would 
reflect a 43% NOX removal efficiency for SNCR. This level of 
removal is approximately twice that considered for other Wyoming 
facilities, as well as facilities in other state BART analyses. 
Further, in commenting on EPA's BART proposal for Montana's Colstrip 
Units 1 and 2, the NPS researched 3,621 coal-fired EGUs with SNCR and 
found only two units that could meet 0.15 lb/MMBtu consistently on a 
monthly basis. Accordingly, we question whether SNCR plus combustion 
controls can achieve a 0.12 lb/MMBtu NOX limit at Laramie 
River Station Units 1-3. Should EPA find that this level of control 
with SNCR is achievable, and fail to impose adequate BART limits 
reflective of SCR capabilities; we request the EPA ensure the proposed 
NOX limits are made

[[Page 5214]]

enforceable as a backstop and that if greater removal efficiencies are 
achievable, rates be revised downwards within a 12-month period from 
the date the technology becomes operable. We also request that level of 
ammonia slip not exceed 5ppm, and such limit likewise be made 
enforceable.
    Comment: EPA's BART analysis for Laramie River Station Units 1-3 
improperly relied on cost estimates from Basin Electric that 
overestimated capital costs and significantly underestimated operation 
and maintenance costs for SNCR. Particularly, Basin Electric 
underestimated the cost of reagent. We corrected these errors in our 
revised cost calculation using the Sargent & Lundy SNCR cost module 
from the IPM, and evaluated the cost of SNCR to reduce NOX 
from the 0.21 lb/MMBtu Wyoming BART limit for LNBs/OFA down to EPA's 
proposed FIP limit of 0.12 lb/MMBtu which reflects a NOX 
reduction across the SNCR of 43%. We also evaluated the cost 
effectiveness for an SNCR system designed to achieve 20% NOX 
removal, which would equate to a NOX rate of 0.17 lb/MMBtu. 
Our revised estimates show the cost effectiveness of SNCR plus 
combustion controls is between $2,435/ton and $2,623/ton to meet a 0.12 
lb/MMBtu NOX rate (or between $2,062/ton and $2,368/ton to 
meet a 0.17 lb/MMBtu NOX rate). These corrected cost 
effectiveness values for meeting a 0.12 lb/MMBtu NOX rate 
with SNCR are higher than the cost effectiveness values, between 
$2,056/ton and $2,109/ton, assumed by EPA.
    Comment: EPA relied on SCR cost estimates provided by Basin 
Electric that failed to include LNBs/OFA and therefore assumed 
unreasonably high construction and operation costs for the SCR.
    Comment: A comparison of SCR to EPA's preferred LNB+OFA+SNCR option 
shows incremental costs less than $4,000/ton which are well below 
values EPA typically accepts. For example, in its proposal to 
disapprove part of the North Dakota plan, EPA cited the ``. . . 
relatively low incremental cost effectiveness between the two control 
options ($4,855 per ton) . . .'' For Laramie River Station, the 
National Park Service (NPS) estimates of incremental costs of SCR are 
only slightly greater than SCR's average costs, which are reasonable 
when compared to costs accepted by other states and EPA.
    Comment: Although Basin Electric opposes EPA's SIP disapproval and 
FIP, it supports the agency's decision not to require SCR at Laramie 
River Station in its proposed FIP. EPA eliminated the option of LNBs/
OFA plus SCR from consideration as BART for the 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 OFA.'' Basin Electric agrees with and 
supports the EPA on this issue.
    Comment: EPA's analysis for Basin Electric's Laramie River Station 
acknowledged that additional emissions reductions from LNB and OFA were 
proposed in the SIP, but did not assess the benefit of these lower 
emissions rates. The short term mass emissions rate was reduced to 
1,220 lb/hr for Laramie River Station on Units 1 and 2, and reduced to 
1,254 lb/hr for Laramie River Station Unit 3. EPA explained that since 
the State did not provide additional cost information for the lower 
limits, revised modeling based on 0.21 lb/MMBtu was not performed, nor 
did EPA perform a revised cost analysis.
    The reduced mass emissions are based on a performance level of 0.19 
lb/MMBtu, which will be achieved by the installation of LNB with OFA at 
no additional cost. Accounting for additional reduction in emissions, 
without changing the control costs, yields a significantly higher 
incremental cost, approximately $3,300, between LNB with OFA and SNCR. 
If EPA had accounted for additional reduction from LNB and OFA and 
resulting smaller incremental visibility improvement (less than 0.23 
delta deciviews), the State anticipates that SNCR would not have been 
BART. This conclusion is based on EPA's determination in the proposed 
FIP for Montana that ``. . . the cost of SOFA+SCR ($3,195/ton) [lower 
than the incremental cost of SNCR] is not justified by the visibility 
improvement of 0.404 deciviews at Theodore Roosevelt National Park and 
0.378 deciviews at UL Bend.'' 77 FR 24027.
    When evaluating Wyoming's 309(g) Regional Haze SIP, EPA should have 
considered the additional NOX emissions reductions achieved 
in the SIP before making a final determination on the approvability of 
the SIP.
    Comment: EPA's comparison of annual emissions from the installation 
of SNCR to the annual emissions cap established in a settlement is not 
consistent. Operation of SNCR on Laramie River Station Units 1 through 
3 is anticipated to result in annual emissions of 8,468 tons per year. 
The State established an annual facility emissions cap of 12,773 tons 
per year in a permit. The 8,468 tons per year value is based on an 
average of 2001-2003 actual heat input for each boiler, while the 
12,773 tons of NOX is a facility cap on potential emissions. 
Typically, actual emissions are notably less than potential emissions 
in order to maintain a margin of compliance. It is the State's 
expectation that the emissions difference between EPA's SNCR 
determination and the State's reduced facility emissions cap will be 
less than the calculated 4,305 tons per year. A more direct comparison 
would be to base the difference on the potential to emit for each of 
the three Laramie River Station units (0.12 lb/MMBtu times the maximum 
heat input rate times 8,760 hours of operation) which yields an annual 
facility NOX emission rate of 10,218 tons per year, and a 
difference from the State's emission cap of 2,555 tons.
    Comment: EPA states ``We find it was unreasonable for the State not 
to determine that LNBs with OFA plus SNCR was NOX BART for 
LRS Units 1-3.'' This statement provides no insight into the agency's 
reasons for disapproving the State's NOX BART for Laramie 
River Station.

C. Jim Bridger Units 1-4

    Comment: In its regional haze FIP, EPA identifies the relatively 
high incremental cost effectiveness of SCR ($5,721 per ton) for the 
Bridger units. Nevertheless, EPA does not account for this number in 
its own BART decision making. EPA also fails to accord any deference to 
Wyoming's consideration of these same costs. Such action is arbitrary 
and contrary to EPA's actions in other states. Here, EPA erred by not 
considering any incremental costs for Bridger, and by not honoring 
Wyoming's consideration of costs for the Bridger units.
    Comment: In response to EPA's request for comment on alternative 
approaches for the Jim Bridger BART determination, EPA received general 
comments in favor of our third proposal in the alternative. Although 
these commenters supported our third proposal in the alternative, the 
commenters did not provide quantitative information to support their 
position.
    Comment: EPA is seeking comment on an alternative that would allow 
PacifiCorp to install SCR at Jim Bridger Units 3 and 4 within 5 years 
from the date of EPA's final action. EPA's reasoning is that this 
alternative would allow PacifiCorp the flexibility to determine the 
implementation schedule for BART controls on all four Jim Bridger 
units. Because EPA's initial proposal to require BART installation by 
2016 best complies with the statutory requirement that BART be 
installed and operated as expeditiously as practicable, 42 U.S.C. 
7491(b)(2)(A), we support EPA's proposal over the alternative.

[[Page 5215]]

    Comment: EPA likely underestimated the visibility benefits 
attributable to SCR to control NOX emissions in other ways. 
First, EPA failed to follow its BART guidelines in estimating pre- and 
post-control emissions. EPA's baseline for modeling included the PM and 
SO2 limits that will be established by the regional haze 
plan, rather than using a pre-SIP baseline (typically from the time 
period of 2001-2004) as required by EPA's BART guidelines. This 
approach resulted in an understatement of visibility improvement from 
NOX BART controls. Further, EPA deviated from its BART 
guidelines by modeling baseline emissions that were based on allowable 
emission rates rather than actual maximum 24-hour rates. As a result, 
EPA's baseline is based on long-term average emissions that mask actual 
maximum visibility impairment. Further, EPA greatly overestimates the 
ammonia slip (SO4 emissions) associated with SCR. For 
example, actual increases in SO4 emissions due to operation 
of SCR at each of the Jim Bridger units are 7.89 lb/hr per unit, far 
lower than EPA's assumed increase in SO4 emissions of 54.0 
lb/hr per unit. This error, too, likely caused EPA to understate the 
visibility benefits of SCR.
    The Conservation Organizations conducted modeling analyses to 
examine how widespread the impacts from each BART-subject source were 
and to analyze the widespread visibility improvements that would result 
if all units were required to install SCR along with combustion 
controls at 0.05 lb/MMBtu limits routinely achieved to meet 
NOX BART. The Conservation Organizations did not attempt to 
address all of EPA`s errors in their supplemental modeling of 
visibility impacts. Had the Conservation Organizations done so (i.e., 
changed baseline to reflect pre-SIP emissions of SO2, PM as 
well as NOX, and reduced the projected increase in sulfates 
to more reasonable levels), we assume that even greater visibility 
benefits would have been demonstrated with SCR required as BART at all 
BART-subject units.

D. Dave Johnston Units 3 and 4

    Comment: EPA relied on Wyoming's cost-effectiveness analysis of SCR 
plus LNBs/OFA at Dave Johnston Unit 3, which significantly 
overestimates the cost of this technology. While EPA presented a cost 
effectiveness of $3,243/ton, our revised cost-effectiveness calculation 
using the Sargent & Lundy SCR cost module shows that SCR plus LNBs/OFA 
at Dave Johnston Unit 3 to meet a NOX limit of 0.05 lb/MMBtu 
is much lower: $1,632/ton.
    Comment: EPA relied on Wyoming's dramatic underestimation of the 
cost for SNCR at Dave Johnson Unit 3. Wyoming's BART analysis used 
costs provided by PacifiCorp, which greatly underestimated reagent 
costs for SNCR. Further, although PacifiCorp's SNCR cost analysis was 
based on achieving a NOX rate of 0.22 lb/MMBtu, Wyoming 
stated that SNCR was assumed to meet a NOX rate of 0.19 lb/
MMBtu. This skewed the cost-effectiveness results by making SNCR appear 
less costly and more effective. Our revised analysis using the Sargent 
& Lundy SNCR cost module from the IPM to achieve a 0.22 lb/MMBtu 
NOX rate (20% removal across the SNCR) demonstrated that 
SNCR has a higher cost-effectiveness value than is presented in the 
proposed rule. While EPA presented and relied on a cost effectiveness 
for SNCR of $721/ton, correcting the flaws in that analysis 
demonstrates a cost effectiveness for SNCR of $1,443/ton. Based on 
these corrected cost calculations, the cost of SCR plus LNBs/OFA is 
$1,632/ton, not appreciably higher than the cost of SNCR at $1,443/ton.
    Comment: EPA makes the same errors in its BART analysis for Dave 
Johnston Unit 4 as it did for Unit 3. EPA has proposed to approve 
Wyoming's NOX BART determination for Dave Johnston Unit 4, 
requiring LNBs with advanced OFA to achieve a NOX emission 
limit of 0.15 lb/MMBtu. Although the cost effectiveness of SCR plus 
LNBs with OFA at Dave Johnston Unit 4 presented by Wyoming and relied 
upon by EPA was abundantly reasonable at $2,210 per ton of 
NOX removed, EPA apparently agreed with Wyoming that the 
incremental cost effectiveness of these controls compared to LNBs/OFA 
was too high to justify SCR as BART.
    Wyoming calculated the cost effectiveness of SCR plus LNBs/OFA to 
be $2,210/ton with a projected maximum visibility improvement of 0.97 
deciviews. In comparison, Wyoming and EPA found that these same 
controls constitute BART at Naughton Unit 3, even though the cost was 
greater, $2,830/ton, and the projected maximum visibility improvement 
was roughly the same, 1.0 deciviews. Wyoming and EPA also found that 
SCR plus LNBs and SOFA met BART for Jim Bridger Units 3 and 4 at a cost 
effectiveness of $2,258/ton and a projected visibility improvement of 
0.80 and 0.82 deciviews, respectively. Further, EPA found that SCR plus 
LNBs and SOFA met BART for Jim Bridger Units 1 and 2 at a cost 
effectiveness of $2,258/ton and a visibility improvement of 0.76 
deciviews and 0.82 deciviews, respectively. EPA's contrary conclusion 
for Dave Johnston Unit 4 is arbitrary and inconsistent with EPA's other 
decisions.
    Comment: Although EPA's data in the proposed rule demonstrated that 
SCR plus LNBs/OFA is cost effective at $2,210/ton of NOX 
removed, EPA overestimated the cost of SCR for Dave Johnston Unit 4 in 
the same way it did for Dave Johnston Unit 3. We recalculated the cost 
effectiveness of SCR plus combustion controls at Dave Johnston Unit 4 
using the Sargent & Lundy SCR IPM Cost Module to meet a NOX 
rate of 0.05 lb/MMBtu. The revised cost analysis shows a cost 
effectiveness of $1,837/ton for these controls.
    Comment: EPA appears to have placed undue weight on incremental 
costs. In its proposal to disapprove part of the North Dakota plan, EPA 
cited the ``. . . relatively low incremental cost effectiveness between 
the two control options ($4,855 per ton) . . .'' For Johnston units 3 
and 4, the NPS estimates of incremental costs of SCR are two--three 
times greater than LNB+OFA+SCR's average costs, which are reasonable 
when compared to costs accepted by other states and EPA.

E. Naughton Units 1-3

    Comment: We received comments that the cost analysis of SCR at 
Naughton is over inflated. One commenter estimated that, using a 
capital cost of $266/kW, LNB+OFA+SCR for Unit 1 would remove 3,249 tpy 
and cost $2,098/ton. The commenter went on to say that they estimated 
the cost for the addition of SCR to Unit 2 and Unit 3 would be $2,037 
and $2,844/ton. A commenter estimated the costs for the addition of SCR 
to Unit 3 would be $1,788/ton. Another commenter estimated the cost of 
SCR of $1,550/ton for Naughton Unit 1 and $1,501/ton for Naughton Unit 
2.
    Comment: Even taken at face value, the $2,750 and $2,848 costs per 
ton estimated by Wyoming for LNB+OFA+ SCR on Naughton Units 1 and 2, 
respectively, are similar to or lower than the cost/ton values accepted 
as reasonable in other BART analyses, including Wyoming's and EPA's 
conclusion that addition of OFA+SCR at $2,830/ton is reasonable for 
Naughton Unit 3.
    Comment: Despite our concerns with the visibility modeling 
conducted by EPA, taken at face value, the annual costs and visibility 
improvements (presented by EPA) associated with the addition of SCR 
result in cost-effectiveness of $9.6 million/deciview for Naughton Unit 
1, $11.5 million/deciview for Unit 2, and $15.7 million/deciview for 
Unit 3 (which EPA deemed reasonable) at the nearest Class I area.

[[Page 5216]]

All three of these estimates are below or within the range of average 
cost/deciview accepted as ``reasonable''.
    Comment: EPA relied solely on an incremental cost-benefit rationale 
to reject SCR as BART for Naughton Units 1 and 2. This conclusion is 
improper where SCR plus combustion controls is cost effective, even 
using EPA's inflated numbers, and its visibility benefit would be 
significant. Indeed, the costs and visibility benefits of SCR at Units 
1 and 2 are nearly identical to the costs and visibility benefits of 
SCR at Unit 3, at which EPA found SCR to be BART. Specifically, SCR at 
Naughton Units 1 and 2 has a cost effectiveness of $2,750/$2,848 per 
ton of NOX removed and results in maximum visibility 
improvements of 1.07/1.10 deciviews. Given these very similar numbers, 
EPA's determination that SCR is BART at Unit 3 but SCR is not BART at 
Units 1 and 2 is arbitrary.
    Comment: EPA has placed undue emphasis on incremental cost 
effectiveness is even more improper considering its inaccuracy. EPA 
stated incorrectly the incremental cost effectiveness of LNBs/OFA plus 
SCR compared to LNBs/OFA as $8,000/ton. However, this calculation 
actually refers to the incremental cost effectiveness of the SCR option 
compared to the SNCR BART option. The incremental cost effectiveness 
between the SCR option and the LNB/OFA option based on Wyoming's cost 
and emission estimates provided in its BART Application Analyses are 
$6,665/ton for Unit 1 and $6,518/ton for Unit 2.

F. Wyodak

    Comment: We received comments that the costs for SCR were 
overinflated. One commenter estimated that LNB+OFA+SCR would remove 
3,773 tpy and cost $3,475/ton. Another commenter estimated that based 
on the Sargent & Lundy SCR cost module, the revised cost estimate for 
these controls to meet a NOX limit of 0.05 lb/MMBtu is 
$2,602/ton of NOX removed.
    Comment: As it did for Dave Johnston Unit 3, EPA also relied on 
Wyoming's gross underestimate of the costs of SNCR. Rather than the 
$958/ton of NOX removed assumed by EPA, the more accurate 
cost effectiveness of SNCR based on the Sargent & Lundy SNCR IPM cost 
module is $3,139/ton. Thus, SCR is more cost effective than SNCR to 
control NOX at Wyodak.
    Comment: EPA has placed undue weight on incremental costs and 
incremental benefits. Our analysis of the LNB+OFA+SCR option shows an 
incremental cost of $3,726/ton for adding SCR to LNB+OFA. Our estimates 
of incremental costs of SCR are only slightly greater than 
LNB+OFA+SCR's average costs, which are reasonable when compared to 
costs accepted by other states and EPA.

G. Dave Johnston Units 1 and 2

    Comment: The Conservation Organizations agree with EPA that 
reasonable progress controls for NOX emissions are needed 
for Dave Johnston Units 1 and 2. EPA correctly found that it was 
unreasonable for Wyoming to reject cost effective NOX 
controls that would improve visibility. EPA proposes to require only 
LNBs/OFA to achieve a NOX emission limit of 0.20 lb/MMBtu 
(30-day rolling average). While we commend EPA for proposing a FIP to 
reduce NOX emissions from Dave Johnston Units 1 and 2, we 
urge EPA to require SCR plus LNBs/OFA to meet a NOX emission 
limit of 0.05 lb/MMBtu to achieve reasonable progress. Although EPA 
concluded that the cost of SCR is not justified by the projected 
visibility improvement, EPA's analysis unreasonably assumed that SCR 
would only achieve a NOX emission rate of 0.12 lb/MMBtu, 
even though an emission rate of 0.05 lb/MMBtu is readily achievable. 
Correcting for this error, it appears that SCR at Dave Johnston Units 1 
and 2 is very cost effective at $2,001 and $1,987/ton of NOX 
removed, respectively. Accordingly, EPA should reconsider requiring SCR 
at Dave Johnston Units 1 and 2 to meet reasonable progress 
requirements.
    Comment: EPA deviated from the BART Guidelines in the way it 
estimated the emission rates it used in its modeling analyses. For Dave 
Johnston Unit 1 and Unit 2, EPA assumed that NOX emissions 
would drop from 1,012.5 lb/hr (base case) to 354.375 lb/hr with the 
addition of LNB+OFA and to 202.5 lb/hr with addition of SCR. However, 
our review of 2001-2003 daily CAMD data found that daily NOX 
emissions from Johnston Unit 1 and Unit 2 during 2001-2003 never 
exceeded 680 lb/hr. EPA modeling analysis cannot be relied upon to 
estimate ``a comparatively small incremental visibility improvement'' 
because the emissions modeled are incorrect.
    Comment: The EPA conclusion that the addition of SCR is not 
justified due to the ``small incremental visibility improvement'' is 
based upon a flawed visibility analysis that over-values the addition 
of LNB+OFA and under-values addition of SCR. Furthermore, the degree of 
visibility improvement is not one of the four statutory factors to be 
considered under the reasonable progress provisions of the RHR. 
Incremental visibility improvement is not mentioned anywhere in the 
reasonable progress provisions or BART Guidelines and EPA cannot create 
a new criterion for the sole purpose of eliminating a control option 
that is reasonably cost-effective and would yield a significant 
visibility improvement.

H. Modeling

    Comment: EPA conducted visibility impact modeling from the Wyoming 
sources for its BART and reasonable progress analyses. Unfortunately, 
EPA failed to present and/or fully explain the results of its modeling 
to the public, thus preventing a complete analysis of the benefits of 
installation of SCR on the Wyoming sources. Accordingly, we request 
that EPA provide clarification on the following issues: (1) Please 
clarify whether the spreadsheet provided in response to our FOIA 
request represents EPA's complete and final modeling results. If not, 
we request that EPA post all final visibility modeling results to the 
docket for this rulemaking, including any post-processing of modeling 
results, and allow the public to submit comment on the modeling 
results; (2) Please describe and clarify any discrepancies between 
EPA's proposed rule and its final modeling results; (3) Please explain 
how EPA calculated the visibility results presented in its proposed 
rule and provide all data upon which these calculations were made. In 
addition, we request that EPA re-run its visibility impact modeling for 
years 2001-2003 and incorporate all proposed changes to its modeling 
provided in this comment letter, in the TSD, and in the Expert Report 
of Howard Gebhart, including, but not limited to utilization of the 
correct version of the model, correct emission rates, and compressive 
inclusion of impacted Class I areas. We also request that EPA post its 
new corrected modeling results to the docket for this rulemaking and 
provide the public with an adequate opportunity to respond and comment 
on the new visibility impact modeling.
    Comment: Wyoming DEQ evaluated visibility improvements at the three 
nearest Class I areas--Bridger, Fitzpatrick, and Mount Zirkel 
Wilderness Areas--and reported the ``cumulative 3-year averaged 
visibility improvement from Post-Control Scenario A across the three 
Class I areas . . .'' We requested to DEQ that the other eight Class I 
areas within 300 km of Bridger (Grand Teton National Park, Yellowstone 
National Park, Rocky Mountain National Park, Washakie

[[Page 5217]]

Wilderness Area, Teton Wilderness Area, Flat Tops Wilderness Area, 
Rawah Wilderness Area, and Eagles Nest Wilderness Area) be included in 
the modeling analysis. However, instead of expanding the modeling 
analysis, EPA R8 reported results for only the Mount Zirkel Wilderness 
Area.
    Comment: EPA R8 has incorrectly estimated visibility improvement 
from all NOX control options: WY DEQ evaluated visibility 
improvements at the four nearest Class I areas and reported the ``The 
cumulative 3-year averaged 98th percentile visibility improvement from 
Post-Control Scenario A summed across all four Class I areas achieved 
with Post-Control Scenario B was 0.754 delta deciviews from Unit 3 and 
0.405 delta from Unit 4.'' EPA R8 reported results for only one Class I 
area. PacifiCorp apparently considered cost a useful metric when it 
made the following statements for its Unit 3 BART proposal: 
``the incremental cost effectiveness for Scenario 1 compared to the 
Baseline is reasonable at $0.4 million per day and $14.4 million per 
deciview to improve visibility at Badlands NP'' and for its Unit 
4 BART proposal, ``the incremental cost effectiveness for 
Scenario 1 compared to the Baseline is reasonable at about $800,000 per 
day and $31.7 million per deciview.'' PacifiCorp's conclusions are 
consistent with those reached across the country that the average cost 
per deciview proposed by either a state or a BART source is $14-$18 
million, with a maximum of almost $50 million per deciview proposed by 
Colorado at the Martin Drake power plant. Combining the modeling 
results provided by EPA R8 (which we believe have underestimated SCR 
benefits) and Wyoming DEQ's cost analyses (which we believe have 
overestimated SCR costs), addition of SCR at Dave Johnston Unit 3 would 
improve visibility by 1.16 deciview at a cost of $14 million per 
deciview at the most-impacted Class I area. Likewise, addition of SCR 
at Dave Johnston Unit 4 would improve visibility by 0.97 deciview at a 
cost of $17 million per deciview. Not only is addition of SCR cost-
effective (even by PacifiCorp's criteria), it would be even more cost-
effective if the issues we have noted above are addressed. By 
overestimating costs of SCR and underestimating control efficiency and 
visibility benefits, EPA R8 concluded that combustion controls plus 
SNCR is BART for Unit 3 and combustion controls are BART for Unit 4, 
rather than SCR.
    Comment: EPA states that a change of 1.0 deciview is perceptible 
and causes visibility impairment and a change of 0.5 deciviews, 
although not perceptible, is considered to contribute to visibility 
impairment. 40 CFR part 51, App. Y, section III.A.1, 70 FR 39120. 
Sources that do not have an impact of 0.5 deciviews or more may be 
exempted from BART altogether. 40 CFR part 51, App. Y, section III.A.1. 
In Wyoming, EPA approved the State's selection of a 0.5 deciview 
threshold for exempting sources from BART, based on the ``relatively 
limited impact on visibility'' from sources under the threshold.
    In Colorado, the state established criteria that SNCR would be 
required as BART only if the cost effectiveness for SNCR was less than 
$5,000/ton and the visibility improvement was greater than 0.2 
deciviews. Although EPA stated it did ``not necessarily agree'' that 
these criteria would always be appropriate for determining BART, it 
proposed to approve all BART determinations the state made using these 
criteria.
    The modeled visibility improvement using the final BART permit 
levels that would be achieved with SNCR at Laramie River is one-tenth 
of what EPA contends is humanly perceptible, one-fifth of the level 
used to exempt Wyoming sources from BART due to relatively limited 
visibility impact, and one-half the SNCR threshold used by Colorado to 
establish limits that EPA proposed to approve. This de minimis 
improvement rebuts EPA's disapproval of the State's NOX BART 
for Laramie River, and supports the State's final BART determination. 
Even if EPA were entitled to disapprove a state's BART determination 
based on a standard of ``unreasonableness,'' it cannot be unreasonable 
for the state to fail to require additional SNCR controls that would 
offer tiny and imperceptible visibility improvements at enormous cost. 
However one characterizes the facts, millions of dollars would be spent 
every year to install and operate SNCR.
    Moreover, the modeled visibility improvements for the Jim Bridger 
units resulting from the requirement to install SCR (as BART under the 
EPA regional haze FIP and as part of the LTS under the Wyoming Regional 
Haze SIP) are too small to justify the overall expense of requiring 
these controls. Spending hundreds of millions of dollars for 
imperceptible visibility changes does not meet the intent, or purpose, 
of the regional haze program.
    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. EPA is proposing to accelerate the timeline 
for installing the Jim Bridger Unit 1 SCR from 2022 to 2017 and the Jim 
Bridger Unit 2 SCR from 2021 to 2017. Even when relying on the CALPUFF 
models that significantly overestimate the visibility impacts, EPA's 
proposal will only result in imperceptible visibility improvements for 
only eleven days a year until the SCRs would have been installed as 
required by the State's plan.
    In a similar manner, the days of impacts need to be considered when 
evaluating the additional controls that EPA proposes to install on 
Wyodak and Dave Johnston Unit 3. EPA's modeling of Wyodak indicates 
that installing and incurring the additional costs for the SNCR will 
not only result in an imperceptible 0.15 deciview of visibility 
improvement, but the days per year the unit is modeled to impact the 
park by greater than 0.5 deciviews will be reduced from sixteen to 
twelve days; a benefit of only four days per year.
    EPA's modeling of Dave Johnston Unit 3 indicates that installing 
and incurring the additional costs for SNCR will result not only in an 
imperceptible 0.17 deciview improvement, but the days per year the unit 
is modeled to impact the park by greater than 0.5 deciviews will only 
be reduced from nine to six days; a benefit of only three days per 
year. None of these imperceptible modeled visibility improvements 
occurring during only a few days a year justify the tremendous cost of 
controls required under EPA's regional haze FIP.

VII. Statutory and Executive Order Reviews

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

    This action is not a ``significant regulatory action'' under the 
terms of Executive Order 12866 (58 FR 51735, October 4, 1993) and is 
therefore not subject to review under Executive Orders 12866 and 13563 
(76 FR 3821, January 21, 2011). As discussed in section C below, the 
final FIP applies to only three facilities and five BART units.\241\ It 
is therefore not a rule of general applicability.
---------------------------------------------------------------------------

    \241\ Dave Johnston Unit 3, Laramie River units 1, 2, and 3, and 
the unit at Wyodak.
---------------------------------------------------------------------------

B. Paperwork Reduction Act

    This action does not impose an information collection burden under 
the provisions of the Paperwork Reduction Act, 44 U.S.C. 3501 et seq. 
Burden is defined at 5 CFR 1320.3(b). Because the final FIP applies to 
just three facilities, the Paperwork Reduction Act does not apply. See 
5 CFR 1320(c).

[[Page 5218]]

C. Regulatory Flexibility Act

    The Regulatory Flexibility Act (RFA) generally requires an agency 
to prepare a regulatory flexibility analysis of any rule subject to 
notice and comment rulemaking requirements under the Administrative 
Procedure Act or any other statute unless the agency certifies that the 
rule will not have a significant economic impact on a substantial 
number of small entities. Small entities include small businesses, 
small organizations, and small governmental jurisdictions.
    For purposes of assessing the impacts of today's final rule on 
small entities, small entity is defined as: (1) A small business as 
defined by the Small Business Administration's (SBA) regulations at 13 
CFR 121.201; (2) a small governmental jurisdiction that is a government 
of a city, county, town, school district or special district with a 
population of less than 50,000; and (3) a small organization that is 
any not-for-profit enterprise which is independently owned and operated 
and is not dominant in its field.
    After considering the economic impacts of today's final rule on 
small entities, I certify that this action will not have a significant 
economic impact on a substantial number of small entities. The Regional 
Haze FIP that EPA is finalizing for purposes of the regional haze 
program consists of imposing federal controls to meet the BART 
requirement for NOX emissions on five specific BART units at 
three facilities in Wyoming. The net result of this FIP action is that 
EPA is finalizing direct emission controls on selected units at only 
three sources. The sources in question are each large electric 
generating plants that are not owned by small entities, and therefore 
are not small entities. The final partial approval of the SIP merely 
approves state law as meeting Federal requirements and does not impose 
additional requirements. See, e.g., Mid-Tex Electric Cooperative, Inc. 
v. FERC, 773 F.2d 327 (D.C. Cir. 1985)(hereinafter Mid-Tex).

D. Unfunded Mandates Reform Act (UMRA)

    Title II of the Unfunded Mandates Reform Act of 1995 (UMRA), Public 
Law 104-4, establishes requirements for federal agencies to assess the 
effects of their regulatory actions on State, local, and Tribal 
governments and the private sector. Under section 202 of UMRA, EPA 
generally must prepare a written statement, including a cost-benefit 
analysis, for final rules with ``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. Before promulgating an EPA rule for which a 
written statement is needed, section 205 of UMRA generally requires EPA 
to identify and consider a reasonable number of regulatory alternatives 
and adopt the least costly, most cost-effective, or least burdensome 
alternative that achieves the objectives of the rule. The provisions of 
section 205 of UMRA do not apply when they are inconsistent with 
applicable law. Moreover, section 205 of UMRA allows EPA to adopt an 
alternative other than the least costly, most cost-effective, or least 
burdensome alternative if the Administrator publishes with the final 
rule an explanation why that alternative was not adopted. Before EPA 
establishes any regulatory requirements that may significantly or 
uniquely affect small governments, including Tribal governments, it 
must have developed under section 203 of UMRA a small government agency 
plan. The plan must provide for notifying potentially affected small 
governments, enabling officials of affected small governments to have 
meaningful and timely input in the development of EPA regulatory 
actions with significant federal intergovernmental mandates, and 
informing, educating, and advising small governments on compliance with 
the regulatory requirements.
    Under Title II of UMRA, EPA has determined that this final rule 
does not contain a federal mandate that may result in expenditures that 
exceed the inflation-adjusted UMRA threshold of $100 million ($150 in 
2013 when adjusted for inflation) by State, local, or Tribal 
governments or the private sector in any one year. The private sector 
expenditures that will result from the FIP, including BART controls for 
Basin Electric Laramie River Station Units 1-3 ($67,128,584 per year) 
\242\, and PacifiCorp Dave Johnston Unit 3 ($11,680,144 per year) \243\ 
and Wyodak ($15,073,502) \244\, are $93,882,230 per year. This 
calculation assumes that PacifiCorp would choose to install SCR on Dave 
Johnston Unit 3, and not to otherwise voluntarily retire the unit, an 
option which the FIP allows. Additionally, we do not foresee 
significant costs (if any) for state and local governments. Thus, 
because the annual expenditures associated with the FIP are less than 
the inflation-adjusted threshold of $150 million in any one year, this 
rule is not subject to the requirements of sections 202 or 205 of UMRA. 
This rule is also not subject to the requirements of section 203 of 
UMRA because it contains no regulatory requirements that might 
significantly or uniquely affect small governments.
---------------------------------------------------------------------------

    \242\ Andover Technology Partners, ``Cost of NOX 
Controls on Wyoming EGUs'', October 28, 2013; Wyoming EGU BART and 
Reasonable Progress Costs--10/2013.
    \243\ Andover Technology Partners, ``Cost of NOX 
Controls on Wyoming EGUs'', October 28, 2013; Wyoming EGU BART and 
Reasonable Progress Costs--10/2013.
    \244\ Andover Technology Partners, ``Cost of NOX 
Controls on Wyoming EGUs'', October 28, 2013; Wyoming EGU BART and 
Reasonable Progress Costs--10/2013.
---------------------------------------------------------------------------

E. Executive Order 13132: Federalism

    Executive Order 13132, Federalism, (64 FR 43255, August 10, 1999) 
revokes and replaces Executive Orders 12612 (Federalism) and 12875 
(Enhancing the Intergovernmental Partnership). Executive Order 13132 
requires EPA to develop an accountable process to ensure ``meaningful 
and timely input by State and local officials in the development of 
regulatory policies that have federalism implications.'' ``Policies 
that have federalism implications'' is defined in the Executive Order 
to include regulations that have ``substantial direct effects on the 
States, on the relationship between the national government and the 
States, or on the distribution of power and responsibilities among the 
various levels of government.'' Under 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 final regulation. EPA also may not issue 
a regulation that has federalism implications and that preempts state 
law unless the Agency consults with state and local officials early in 
the process of developing the final regulation.
    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 under the 
CAA to include in its SIP provisions to meet the visibility 
requirements of Part C of Title I of the CAA and to prohibit emissions 
from interfering with other states measures to protect visibility. 
Thus, Executive Order 13132 does not apply to this action.

[[Page 5219]]

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

    Executive Order 13175, entitled Consultation and Coordination with 
Indian Tribal Governments (65 FR 67249, November 9, 2000), requires EPA 
to develop an accountable process to ensure ``meaningful and timely 
input by tribal officials in the development of regulatory policies 
that have tribal implications.'' This final rule does not have tribal 
implications, as specified in Executive Order 13175. It will not have 
substantial direct effects on tribal governments. Thus, Executive Order 
13175 does not apply to this rule.

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

    EPA interprets EO 13045 (62 FR 19885, April 23, 1997) as applying 
only to those regulatory actions that concern health or safety risks, 
such that the analysis required under section 5-501 of the EO has the 
potential to influence the regulation. This action is not subject to EO 
13045 because it implements specific standards established by Congress 
in statutes. However, to the extent this final rule will limit 
emissions of NOX and PM, the rule will have a beneficial 
effect on children's health by reducing air pollution.

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

    This action is not subject to Executive Order 13211 (66 FR 28355 
(May 22, 2001)), because it is not a significant regulatory action 
under Executive Order 12866.

I. National Technology Transfer and Advancement Act

    Section 12(d) of the National Technology Transfer and Advancement 
Act of 1995 (``NTTAA''), Public Law 104-113, 12(d) (15 U.S.C. 272 note) 
directs EPA to use voluntary consensus standards in its regulatory 
activities unless to do so would be inconsistent with applicable law or 
otherwise impractical. Voluntary consensus standards are technical 
standards (e.g., materials specifications, test methods, sampling 
procedures, and business practices) that are developed or adopted by 
voluntary consensus standards bodies. NTTAA directs EPA to provide 
Congress, through OMB, explanations when the Agency decides not to use 
available and applicable voluntary consensus standards.
    This final rulemaking does not involve technical standards. 
Therefore, EPA is not considering the use of any voluntary consensus 
standards.

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

    Executive Order 12898 (59 FR 7629, February 16, 1994), establishes 
federal executive policy on environmental justice. Its main provision 
directs federal agencies, to the greatest extent practicable and 
permitted by law, to make environmental justice part of their mission 
by identifying and addressing, as appropriate, disproportionately high 
and adverse human health or environmental effects of their programs, 
policies, and activities on minority populations and low-income 
populations in the United States.
    We have determined that this final action will not have 
disproportionately high and adverse human health or environmental 
effects on minority or low-income populations because it increases the 
level of environmental protection for all affected populations without 
having any disproportionately high and adverse human health or 
environmental effects on any population, including any minority or low-
income population. This final rule limits emissions of NOX 
from three facilities and five BART units in Wyoming. The partial 
approval of the SIP merely approves state law as meeting Federal 
requirements and imposes no additional requirements beyond those 
imposed by state law.

K. Congressional Review Act

    The Congressional Review Act, 5 U.S.C 801 et seq., as added by the 
Small Business Regulatory Enforcement Fairness Act of 1996, generally 
provides that before a rule may take effect, the agency promulgating 
the rule must submit a rule report, which includes a copy of the rule, 
to each House of the Congress and to the Comptroller General of the 
United States. Section 804 exempts from section 801 the following types 
of rules (1) rules of particular applicability; (2) rules relating to 
agency management or personnel; and (3) rules of agency organization, 
procedure, or practice that do not substantially affect the rights or 
obligations of non-agency parties. 5 U.S. 804(3). EPA is not required 
to submit a rule report regarding today's action under section 801 
because this is a rule of particular applicability. This rule finalizes 
a FIP for three sources.

L. Judicial Review

    Under section 307(b)(1) of the CAA, petitions for judicial review 
of this action must be filed in the United States Court of Appeals for 
the appropriate circuit by March 31, 2014. Pursuant to CAA section 
307(d)(1)(B), this action is subject to the requirements of CAA section 
307(d) as it promulgates a FIP under CAA section 110(c). Filing a 
petition for reconsideration by the Administrator of this final rule 
does not affect the finality of this action for the purposes of 
judicial review nor does it extend the time within which a petition for 
judicial review may be filed, and shall not postpone the effectiveness 
of such rule or action. This action may not be challenged later in 
proceedings to enforce its requirements. See CAA section 307(b)(2).

List of Subjects in 40 CFR Part 52

    Environmental protection, Air pollution control, Incorporation by 
reference, Intergovernmental relations, Nitrogen dioxide, Particulate 
matter, Reporting and recordkeeping requirements, Sulfur oxides, 
Volatile organic compounds.

    Dated: January 10, 2014.
Gina McCarthy,
Administrator, U.S. EPA.
    40 CFR part 52 is amended to read as follows:

PART 52--APPROVAL AND PROMULGATION OF IMPLEMENTATION PLANS

0
1. The authority citation for part 52 continues to read as follows:

    Authority: 42 U.S.C. 7401 et seq.

Subpart ZZ--Wyoming

0
2. Section 52.2620, in the table in paragraph (e) is amended by adding 
an entry for ``XXIII. Wyoming State Implementation Plan for Regional 
Haze for 309(g)'' at the end of the table.


Sec.  52.2620  Identification of plan.

* * * * *
    (e) * * *

[[Page 5220]]



--------------------------------------------------------------------------------------------------------------------------------------------------------
     Name of nonregulatory SIP       Applicable geographic or  State submittal date/    EPA approval date
             provision                  nonattainment area          adopted date        and citation \3\                    Explanations
--------------------------------------------------------------------------------------------------------------------------------------------------------
 
                                                                      * * * * * * *
XXIII. Wyoming State                Statewide................  Submitted:             1/30/14, [Insert      Excluding portions of the following: Chapter
 Implementation Plan for Regional                               1/12/2011...........   Federal Register      6.4, Chapter 6.5.5, Chapter 6.5.7, Chapter
 Haze for 309(g).                                                                      page number where     6.5.8, and Chapter 7.5. We are excluding
                                                                                       the document          portions of these chapters because EPA
                                                                                       begins.].             disapproved: (1) The NOX BART
                                                                                                             determinations for: (1) Laramie River Units
                                                                                                             1-3, Dave Johnston Unit 3, and Wyodak Unit
                                                                                                             1; (2) the State's monitoring,
                                                                                                             recordkeeping, and reporting requirements
                                                                                                             for BART units; (3) the State's reasonable
                                                                                                             progress goals.
--------------------------------------------------------------------------------------------------------------------------------------------------------

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

    \3\ In order to determine the EPA effective date for a specific 
provision listed in this table, consult the Federal Register notice 
cited in this column for the particular provision.

0
3. Add Sec.  52.2636 to subpart ZZ to read as follows:


Sec.  52.2636  Implementation plan for regional haze.

    (a) Applicability. (1) This section applies to each owner and 
operator of the following emissions units in the State of Wyoming for 
which EPA approved the State's BART determination:
    (i) FMC Westvaco Trona Plant Units NS-1A and NS-1B (PM and 
NOX);
    (ii) TATA Chemicals Partners (previously General Chemical) Boilers 
C and D (PM and NOX);
    (iii) Basin Electric Power Cooperative Laramie River Station Units 
1, 2, and 3 (PM);
    (iv) PacifiCorp Dave Johnston Power Plant Unit 3 (PM);
    (v) PacifiCorp Dave Johnston Power Plant Unit 4 (PM and 
NOX);
    (vi) PacifiCorp Jim Bridger Power Plant Units 1, 2, 3, and 4 (PM 
and NOX);
    (vii) PacifiCorp Naughton Power Plant Units 1, 2, and 3 (PM and 
NOX); and
    (viii) PacifiCorp Wyodak Power Plant Unit 1 (PM).
    (2) This section also applies to each owner and operator of the 
following emissions units in the State of Wyoming for which EPA 
disapproved the State's BART determination and issued a NOX 
BART Federal Implementation Plan:
    (i) Basin Electric Power Cooperative Laramie River Station Units 1, 
2, and 3;
    (ii) PacifiCorp Dave Johnston Unit 3; and
    (iii) PacifiCorp Wyodak Power Plant Unit 1.
    (b) Definitions. Terms not defined below shall have the meaning 
given them in the Clean Air Act or EPA's regulations implementing the 
Clean Air Act. For purposes of this section:
    (1) BART means Best Available Retrofit Technology.
    (2) BART unit means any unit subject to a Regional Haze emission 
limit in Table 1 and Table 2 of this section.
    (3) CAM means Compliance Assurance Monitoring as required by 40 CFR 
part 64.
    (4) Continuous emission monitoring system or CEMS means the 
equipment required by this section to sample, analyze, measure, and 
provide, by means of readings recorded at least once every 15 minutes 
(using an automated data acquisition and handling system (DAHS)), a 
permanent record of NOX emissions, diluent, or stack gas 
volumetric flow rate.
    (5) FIP means Federal Implementation Plan.
    (6) The term lb/hr means pounds per hour.
    (7) The term lb/MMBtu means pounds per million British thermal 
units of heat input to the fuel-burning unit.
    (8) NOX means nitrogen oxides.
    (9) Operating day means a 24-hour period between 12 midnight and 
the following midnight during which any fuel is combusted at any time 
in the BART unit. It is not necessary for fuel to be combusted for the 
entire 24-hour period.
    (10) The owner/operator means any person who owns or who operates, 
controls, or supervises a unit identified in paragraph (a) of this 
section.
    (11) PM means filterable total particulate matter.
    (12) Unit means any of the units identified in paragraph (a) of 
this section.
    (c) Emissions limitations. (1) The owners/operators of emissions 
units subject to this section shall not emit, or cause to be emitted, 
PM or NOX in excess of the following limitations:

                                            Table 1 to Sec.   52.2636
 [Emission limits for BART units for which EPA approved the State's BART and Reasonable Progress determinations]
----------------------------------------------------------------------------------------------------------------
                                                                                                  NOX emission
                                                                                 PM emission      limits-- lb/
                            Source name/BART unit                             limits--lb/MMBtu    MMBtu (30-day
                                                                                                rolling average)
----------------------------------------------------------------------------------------------------------------
FMC Westvaco Trona Plant/Unit NS-1A.........................................              0.05              0.35
FMC Westvaco Trona Plant/Unit NS-1B.........................................              0.05              0.35
TATA Chemicals Partners (General Chemical) Green River Trona Plant/Boiler C.              0.09              0.28
TATA Chemicals Partners (General Chemical) Green River Trona Plant/Boiler D.              0.09              0.28
Basin Electric Power Cooperative Laramie River Station/Unit 1...............              0.03               N/A
Basin Electric Power Cooperative Laramie River Station/Unit 2...............              0.03               N/A
Basin Electric Power Cooperative Laramie River Station/Unit 3...............              0.03               N/A
PacifiCorp Dave Johnston Power Plant/Unit 3.................................             0.015               N/A

[[Page 5221]]

 
PacifiCorp Dave Johnston Power Plant/Unit 4.................................             0.015              0.15
PacifiCorp Jim Bridger Power Plant/Unit 1 \1\...............................              0.03         0.26/0.07
PacifiCorp Jim Bridger Power Plant/Unit 2 \1\...............................              0.03         0.26/0.07
PacifiCorp Jim Bridger Power Plant/Unit 3 \1\...............................              0.03         0.26/0.07
PacifiCorp Jim Bridger Power Plant/Unit 4 \1\...............................              0.03         0.26/0.07
PacifiCorp Naughton Power Plant/Unit 1......................................              0.04              0.26
PacifiCorp Naughton Power Plant/Unit 2......................................              0.04              0.26
PacifiCorp Naughton Power Plant/Unit 3......................................             0.015              0.07
PacifiCorp Wyodak Power Plant/Unit 1........................................             0.015               N/A
----------------------------------------------------------------------------------------------------------------
\1\ The owners and operators of PacifiCorp Jim Bridger Units 1, 2, 3, and 4 shall comply with the NOX emission
  limit for BART of 0.26 lb/MMBtu and PM emission limit for BART of 0.03 lb/MMBtu and other requirements of this
  section by March 4, 2019. The owners and operators of PacifiCorp Jim Bridger Units 1, 2, 3 and 4 shall comply
  with the NOX emission limit for reasonable progress of 0.07 lb/MMBtu by: December 31, 2022 for Unit 1,
  December 31, 2021 for Unit 2, December 31, 2015, for Unit 3, and December 31, 2016, for Unit 4.


                        Table 2 to Sec.   52.2636
  [Emission limits for BART units for which EPA disapproved the State's
                BART determination and implemented a FIP]
------------------------------------------------------------------------
                                                                 NOX
                                                               emission
                                                              limit--lb/
                   Source name/BART unit                      MMBtu (30-
                                                                 day
                                                               rolling
                                                               average)
------------------------------------------------------------------------
Basin Electric Power Cooperative Laramie River Station/Unit         0.07
 1.........................................................
Basin Electric Power Cooperative Laramie River Station/Unit         0.07
 2.........................................................
Basin Electric Power Cooperative Laramie River Station/Unit         0.07
 3.........................................................
PacifiCorp Dave Johnston Unit 3............................        *0.07
PacifiCorp Wyodak Power Plant/Unit 1.......................         0.07
------------------------------------------------------------------------
* (or 0.28 and shut-down-by December 31, 2027).

    (2) These emission limitations shall apply at all times, including 
startups, shutdowns, emergencies, and malfunctions.
    (d) Compliance date. (1) The owners and operators of PacifiCorp Jim 
Bridger Units 1, 2, 3, and 4 shall comply with the NOX 
emission limit of 0.26 lb/MMBtu and PM emission limit of 0.03 lb/MMBtu 
and other requirements of this section by March 4, 2019. The owners and 
operators of PacifiCorp Jim Bridger Units 1, 2, 3 and 4 shall comply 
with the NOX emission limit of 0.07 lb/MMBtu by: December 
31, 2022 for Unit 1, December 31, 2021 for Unit 2, December 31, 2015, 
for Unit 3, and December 31, 2016, for Unit 4.
    (2) The owners and operators of the other BART sources subject to 
this section shall comply with the emissions limitations and other 
requirements of this section by March 4, 2019.
    (3) Compliance alternatives for PacifiCorp Dave Johnston Unit 3. 
(i) The owners and operators of PacifiCorp Dave Johnston Unit 3 will 
meet a NOX emission limit of 0.07 lb/MMBtu (30-day rolling 
average) by March 4, 2019; or
    (ii) Alternatively, the owners and operators of PacifiCorp Dave 
Johnston Unit 3 will permanently cease operation of this unit on or 
before December 31, 2027.
    (e) Compliance determinations for NOX. (1) For all BART units other 
than Trona Plant units:
    (i) CEMS. At all times after the earliest compliance date specified 
in paragraph (d) of this section, the owner/operator of each unit shall 
maintain, calibrate, and operate a CEMS, in full compliance with the 
requirements found at 40 CFR part 75, to accurately measure 
NOX, diluent, and stack gas volumetric flow rate from each 
unit. The CEMS shall be used to determine compliance with the emission 
limitations in paragraph (c) of this section for each unit.
    (ii) Method. (A) For any hour in which fuel is combusted in a unit, 
the owner/operator of each unit shall calculate the hourly average 
NOX emission rate in lb/MMBtu at the CEMS in accordance with 
the requirements of 40 CFR part 75. At the end of each operating day, 
the owner/operator shall calculate and record a new 30-day rolling 
average emission rate in lb/MMBtu from the arithmetic average of all 
valid hourly emission rates from the CEMS for the current operating day 
and the previous 29 successive operating days.
    (B) An hourly average NOX emission rate in lb/MMBtu is 
valid only if the minimum number of data points, as specified in 40 CFR 
part 75, is acquired by both the pollutant concentration monitor 
(NOX) and the diluent monitor (O2 or 
CO2).
    (C) Data reported to meet the requirements of this section shall 
not include data substituted using the missing data substitution 
procedures of subpart D of 40 CFR part 75, nor shall the data have been 
bias adjusted according to the procedures of 40 CFR part 75.
    (2) For all Trona Plant BART units:
    (i) CEMS. At all times after the compliance date specified in 
paragraph (d) of this section, the owner/operator of each unit shall 
maintain, calibrate, and operate a CEMS, in full compliance with

[[Page 5222]]

the requirements found at 40 CFR part 60, to accurately measure 
NOX, diluent, and stack gas volumetric flow rate from each 
unit, including the CEMS quality assurance requirements in appendix F 
of 40 CFR part 60. The CEMS shall be used to determine compliance with 
the emission limitations in paragraph (c) of this section for each 
unit.
    (ii) Method. (A) For any hour in which fuel is combusted in a unit, 
the owner/operator of each unit shall calculate the hourly average 
NOX emission rate in lb/MMBtu at the CEMS in accordance with 
the requirements of 40 CFR part 60. At the end of each operating day, 
the owner/operator shall calculate and record a new 30-day rolling 
average emission rate in lb/MMBtu from the arithmetic average of all 
valid hourly emission rates from the CEMS for the current operating day 
and the previous 29 successive operating days.
    (B) An hourly average NOX emission rate in lb/MMBtu is 
valid only if the minimum number of data points, as specified in 40 CFR 
part 60, is acquired by both the pollutant concentration monitor 
(NOX) and the diluent monitor (O2 or 
CO2).
    (f) Compliance determinations for particulate matter. Compliance 
with the particulate matter emission limit for each BART unit shall be 
determined from annual performance stack tests. Within 60 days of the 
compliance deadline specified in paragraph (d) of this section, and on 
at least an annual basis thereafter, the owner/operator of each unit 
shall conduct a stack test on each unit to measure particulate 
emissions using EPA Method 5, 5B, 5D, or 17, as appropriate, in 40 CFR 
part 60, Appendix A. A test shall consist of three runs, with each run 
at least 120 minutes in duration and each run collecting a minimum 
sample of 60 dry standard cubic feet. Results shall be reported in lb/
MMBtu. In addition to annual stack tests, the owner/operator shall 
monitor particulate emissions for compliance with the BART emission 
limits in accordance with the applicable Compliance Assurance 
Monitoring (CAM) plan developed and approved by the State in accordance 
with 40 CFR part 64.
    (g) Recordkeeping. The owner/operator shall maintain the following 
records for at least five years:
    (1) All CEMS data, including the date, place, and time of sampling 
or measurement; parameters sampled or measured; and results.
    (2) Records of quality assurance and quality control activities for 
emissions measuring systems including, but not limited to, any records 
required by 40 CFR part 75. Or, for Trona Plant units, records of 
quality assurance and quality control activities for emissions 
measuring systems including, but not limited to appendix F of 40 CFR 
part 60.
    (3) Records of all major maintenance activities conducted on 
emission units, air pollution control equipment, and CEMS.
    (4) Any other CEMS records required by 40 CFR part 75. Or, for 
Trona Plant units, any other CEMs records required by 40 CFR part 60.
    (5) Records of all particulate stack test results.
    (6) All data collected pursuant to the CAM plan.
    (h) Reporting. All reports under this section shall be submitted to 
the Director, Office of Enforcement, Compliance and Environmental 
Justice, U.S. Environmental Protection Agency, Region 8, Mail Code 
8ENF-AT, 1595 Wynkoop Street, Denver, Colorado 80202-1129.
    (1) The owner/operator of each unit shall submit quarterly excess 
emissions reports for NOX BART units no later than the 30th 
day following the end of each calendar quarter. Excess emissions means 
emissions that exceed the emissions limits specified in paragraph (c) 
of this section. The reports shall include the magnitude, date(s), and 
duration of each period of excess emissions, specific identification of 
each period of excess emissions that occurs during startups, shutdowns, 
and malfunctions of the unit, the nature and cause of any malfunction 
(if known), and the corrective action taken or preventative measures 
adopted.
    (2) The owner/operator of each unit shall submit quarterly CEMS 
performance reports, to include dates and duration of each period 
during which the CEMS was inoperative (except for zero and span 
adjustments and calibration checks), reason(s) why the CEMS was 
inoperative and steps taken to prevent recurrence, and any CEMS repairs 
or adjustments. The owner/operator of each unit shall also submit 
results of any CEMS performance tests required by 40 CFR part 75. Or, 
for Trona Plant units, the owner/operator of each unit shall also 
submit results of any CEMs performance test required appendix F of 40 
CFR part 60 (Relative Accuracy Test Audits, Relative Accuracy Audits, 
and Cylinder Gas Audits).
    (3) When no excess emissions have occurred or the CEMS has not been 
inoperative, repaired, or adjusted during the reporting period, such 
information shall be stated in the quarterly reports required by 
paragraphs (h)(1) and (2) of this section.
    (4) The owner/operator of each unit shall submit results of any 
particulate matter stack tests conducted for demonstrating compliance 
with the particulate matter BART limits in paragraphs (c) of this 
section, within 60 calendar days after completion of the test.
    (5) The owner/operator of each unit shall submit semi-annual 
reports of any excursions under the approved CAM plan in accordance 
with the schedule specified in the source's title V permit.
    (i) Notifications. (1) The owner/operator shall promptly submit 
notification of commencement of construction of any equipment which is 
being constructed to comply with the NOX emission limits in 
paragraph (c) of this section.
    (2) The owner/operator shall promptly submit semi-annual progress 
reports on construction of any such equipment.
    (3) The owner/operator shall promptly submit notification of 
initial startup of any such equipment.
    (j) Equipment operation. At all times, the owner/operator shall 
maintain each unit, including associated air pollution control 
equipment, in a manner consistent with good air pollution control 
practices for minimizing emissions.
    (k) Credible evidence. Nothing in this section shall preclude the 
use, including the exclusive use, of any credible evidence or 
information, relevant to whether a source would have been in compliance 
with requirements of this section if the appropriate performance or 
compliance test procedures or method had been performed.

0
4. Add Sec.  52.2637 to subpart ZZ to read as follows:


Sec.  52.2637  Federal implementation plan for reasonable attributable 
visibility impairment long-term strategy.

    As required by 40 CFR 41.306(c), EPA will ensure that the review of 
the State's reasonably attributable visibility impairment long-term 
strategy is coordinated with the regional haze long-term strategy under 
40 CFR 51.308(g). EPA's review will be in accordance with the 
requirements of 40 CFR 51.306(c).

[FR Doc. 2014-00930 Filed 1-29-14; 8:45 am]
BILLING CODE 6560-50-P