WildEarth Guardians v. EPA , 770 F.3d 919 ( 2014 )

FILED
United States Court of Appeals
PUBLISH Tenth Circuit
October 21, 2014
UNITED STATES COURT OF APPEALS
Elisabeth A. Shumaker
TENTH CIRCUIT Clerk of Court
WILDEARTH GUARDIANS;
HEAL UTAH; NATIONAL PARKS
CONSERVATION ASSOCIATION;
POWDER RIVER BASIN
RESOURCE COUNCIL; SIERRA
CLUB;
Petitioners,
v. Nos. 12-9596, 13-9502, 13-9506,
13-9507, 13-9508, 13-9509,
UNITED STATES 13-9510
ENVIRONMENTAL PROTECTION
AGENCY; GINA McCARTHY,
Administrator, United States
Environmental Protection Agency,
Respondents.
--------------------
PUBLIC SERVICE COMPANY OF
NEW MEXICO; PACIFICORP;
NEW MEXICO ENVIRONMENT
DEPARTMENT; BASIN
ELECTRIC POWER
COOPERATIVE; STATE OF
WYOMING; UTAH ASSOCIATED
MUNICIPAL POWER SYSTEM;
UTAH DIVISION OF AIR
QUALITY; CITY OF
ALBUQUERQUE,
Intervenors,
and
AMERICAN COALITION FOR
CLEAN COAL ELECTRICITY,
Amicus Curiae.
PETITIONS FOR REVIEW OF FINAL DECISIONS ISSUED BY
THE UNITED STATES ENVIRONMENTAL PROTECTION AGENCY
Nos. EPA-R06-OAR-2009-0050, EPA-R08-OAR-2011-0400,
EPA-R08-OAR-2011-0114, EPA-RO6-OAR-2008-0702
Jenny K. Harbine, Earthjustice, Bozeman, Montana (John Barth, Hygiene,
Colorado, and Ashley D. Wilmes, WildEarth Guardians, Boulder,
Colorado, with her on the briefs), for Petitioners.
Chloe H. Kolman, United States Department of Justice, Environment &
Natural Resources Division, Washington, D.C. (Stephanie J. Talbert,
United States Department of Justice, Environment & Natural Resources
Division, Washington, D.C., Robert G. Dreher, Acting Assistant Attorney
General, United States Department of Justice, Environment & Natural
Resources Division, Washington, D.C.; M. Lea Anderson, Of Counsel,
United States Environmental Protection Agency, Washington, D.C.;
Matthew C. Marks, Of Counsel, United States Environmental Protection
Agency, Washington, D.C.; Brian Tomasovic, Of Counsel, United States
Environmental Protection Agency, Dallas, Texas; Sara L. Laumann, Of
Counsel, United States Environmental Protection Agency, Denver,
Colorado, with her on the brief), for Respondent.
E. Blain Rawson, Ray Quinney & Nebeker, P.C., Salt Lake City, Utah
(Emily Smith Loeffler, Quinney & Nebeker, P.C., Salt Lake City, Utah,
Michael G. Jenkins, Assistant General Counsel, PacifiCorp Energy with
him on the brief), for Intervenor PacifiCorp Energy.
Matthias L. Sayer, Assistant Attorney General, Wyoming Office of
Attorney General, Cheyenne, Wyoming (Jay A. Jerde, Deputy Attorney
General, Wyoming Office of Attorney General, Cheyenne, Wyoming, with
him on the brief), for Intervenor State of Wyoming.
2
Christopher L. Colclasure, Holland & Hart LLP, Denver, Colorado, for
Intervenor Basin Electric Power Cooperative;
Richard L. Alvidrez and Robert H. Clark, Miller Stratvert P.A.,
Albuquerque, New Mexico; Kallie H. Kuehl, Corporate Counsel,
Albuquerque, New Mexico, on the brief for Intervenor Public Service
Company of New Mexico.
Jeffrey M. Kendall, General Counsel and William G. Grantham, Assistant
General Counsel, for New Mexico Environment Department, on the brief
for Intervenor New Mexico Environment Department.
Carol Parker, Assistant City Attorney and Adelia W. Kearny, Deputy City
Attorney, Albuquerque, New Mexico, on the brief for Intervenor City of
Albuquerque.
H. Michael Keller and Mary Jane E. Galvin-Wagg, Van Cott, Salt Lake
City, Utah; Mason Baker, General Counsel, Salt Lake City, Utah, on the
brief for Intervenor Utah Associated Municipal Power Systems.
John E. Swallow, Utah Attorney General and Christian C. Stephens,
Assistant Attorney General, Salt Lake City, Utah; Craig W. Anderson,
Division Chief and Assistant Attorney General, Environment Division,
Utah Attorney General’s Office, Salt Lake City, Utah, on the brief for
Intervenor Utah Division of Air Quality.
Paul M. Seby and Marian C. Larsen, Seby Larsen LLP, Denver, Colorado,
on the brief for Amicus Curiae American Coalition for Clean Coal
Electricity.
Before BACHARACH, SEYMOUR, and MURPHY, Circuit Judges.
BACHARACH, Circuit Judge.
This appeal grows out of the Clean Air Act. In an effort to comply with
the statute, three states (New Mexico, Utah, and Wyoming), one city (City of
Albuquerque), and one county (Bernalillo County) adopted a regional cap-and-
3
trade program regulating sulfur-dioxide emissions over the Colorado Plateau. 1
Under this program, each participant obtained a ceiling on sulfur-dioxide
emissions. If the ceiling was met, polluters would get allocations of sulfur
dioxide that could be emitted. With these allocations, polluters had a choice.
They could use the allocations or cut emissions and trade the unused portions of
the allocations.
The program required approval of the Environmental Protection
Agency. In determining whether to approve the program, the EPA had to
apply its regulations. Under these regulations, states could satisfy the
Clean Air Act by ensuring installation of the best available retrofit
technology in all eligible major sources that contributed to visibility
impairment. This mode of compliance is referred to as “BART.” States
affecting visibility over the Colorado Plateau were allowed to use an
alternative program in lieu of BART. But this alternative program had to
be better than BART in improving air visibility.
1
Final Rule, Approval and Promulgation of State Implementation Plans;
Wyoming, 77 Fed. Reg. 73,926, 73,926 (Dec. 12, 2012); Final Rule, Approval,
Disapproval and Promulgation of State Implementation Plans; Utah, 77 Fed. Reg.
74,355, 74,355 (Dec. 14, 2012); Final Rule, Approval and Promulgation of State
Implementation Plans; New Mexico, 77 Fed. Reg. 70,693, 70,693 (Nov. 27,
2012); Final Rule, Approval and Promulgation of State Implementation Plans;
City of Albuquerque-Bernalillo County, 77 Fed. Reg. 71,119, 71,119 (Nov. 29,
2012).
4
New Mexico, Utah, Wyoming, the City of Albuquerque, and Bernalillo
County persuaded the EPA that the trading program would yield better results
than BART because:
● the program covered polluters that would not have been subject
to BART,
● the program encompassed emissions from new sources, which
would not have been subject to BART, and
● the program encouraged polluters to expedite equipment
upgrades and to operate below full capacity.
Five environmental groups filed petitions for review, 2 arguing that
the EPA should not have approved the trading program. To decide these
petitions, we must determine whether the EPA acted arbitrarily and
capriciously in finding that the trading program was better than BART.
We conclude that the EPA’s decision was neither arbitrary nor capricious.
Thus, we deny the petitions for review.
I. The Clean Air Act and the EPA’s Regulatory Framework
The petitions require an understanding of the statutory and regulatory
requirements for alleviation of air pollution.
A. Statutory Requirement for EPA Guidelines
The Clean Air Act requires the EPA to establish regulations to ensure
“reasonable progress” toward the improvement in visibility and
2
The Petitioners are WildEarth Guardians, Heal Utah, National Parks
Conservation Association, Powder River Basin Resource Council, and
Sierra Club.
5
“compliance with the requirements of [42 U.S.C. § 7491].” 3 42 U.S.C.
§ 7491(a)(4). In light of this requirement, the EPA had to establish
regulations requiring states to develop implementation plans to improve
visibility and adopt, maintain, and enforce air quality standards. Id. §§
7410(a)(1), 7491.
Under the statutory scheme, the EPA would then review the state
implementation plans to ensure compliance with the Clean Air Act and
implementing regulations. Id. §§ 7410(a)(3)(B), 7492(e)(2); see Oklahoma
v. EPA, 723 F.3d 1201, 1204 (10th Cir. 2013). Once approved, state
implementation plans would be enforceable as federal law. 42 U.S.C.
§§ 7413, 7604.
States implementing the BART requirement do so in two steps: (1)
identify the sources subject to BART, and (2) determine the particular
technologies required for individual sources. 40 C.F.R. § 51.308(e)(1); see
Util. Air Regulatory Grp. v. EPA, 471 F.3d 1333, 1335-36 (D.C. Cir.
2006). In considering the required technologies, states must consider five
factors for each BART-eligible source:
(1) the costs of compliance;
3
“Reasonable progress” is measured by comparing “the costs of
compliance, the time necessary for compliance, . . . the energy and nonair
quality environmental impacts of compliance, and the remaining useful life
of any existing [regulated] source” (known as the “four factors”). See 42
U.S.C. § 7491(g)(1); 40 C.F.R. § 51.308(d)(1)(i)(A).
6
(2) the energy and nonair quality environmental impacts of
compliance;
(3) the existing pollution control technologies already in place;
(4) the remaining useful life of the source; and
(5) the improvement in visibility anticipated from the use of given
technologies.
42 U.S.C. § 7491(g)(2).
B. Regulations Governing the Colorado Plateau
Congress also enacted legislation requiring the EPA to establish a
visibility transport commission to study regional haze in the Grand Canyon
and to recommend curative action. 42 U.S.C. § 7492(f).
To comply, the EPA established the Grand Canyon Visibility
Transport Commission, which would “assess scientific, technical, and other
information related to adverse visual air quality impacts from potential or
projected emissions growth from sources located in the Transport Region.”
Joint App. at 71. Upon completion of this assessment, the Transport
Commission would report to the EPA on appropriate measures to improve
visual air quality on the Colorado Plateau. Id. 4
4
The EPA expanded the scope of the Grand Canyon Visibility
Commission’s review to include sixteen Class I areas in the vicinity of the
Grand Canyon. With this expansion, the Commission addressed visual air
quality in the “Golden Circle” of parks and wilderness areas in the
Colorado Plateau. See Notice of Meeting, Grand Canyon Visibility
Transport Commission, 56 Fed. Reg. 57,522, 57,523 (Nov. 12, 1991).
7
1. The Grand Canyon Visibility Transport Commission
The Transport Commission analyzed the effects of regional haze in
sixteen Class I areas 5 affected by pollution in nine states (Arizona,
California, Colorado, Idaho, Nevada, New Mexico, Oregon, Utah, and
Wyoming). 6 42 U.S.C. § 7492(c)(1); Final Rule, Regional Haze
Regulations, 64 Fed. Reg. 35,714, 35,770 (July 1, 1999).
Based on this analysis, the Transport Commission recommended a
regional cap-and-trade program for sulfur dioxide that would go into effect
when participating states exceed an emissions target. Joint App. at 56.
Details of the program would be worked out later.
The functions of the Transport Commission were passed on to the
Western Regional Air Partnership, which continued the study and
recommended a plan. Id. at 190. The plan included:
(1) milestones to measure reductions in regional emissions of
sulfur dioxide, and
5
Class I federal areas include all regions as of August 7, 1977,
consisting of national wilderness areas and national memorial parks
exceeding 5,000 acres, national parks exceeding 6,000 acres, and
international parks. See 42 U.S.C. § 7472(a).
6
The sixteen Class I areas are the Grand Canyon National Park,
Sycamore Canyon Wilderness, Petrified Forest National Park, Mount Baldy
Wilderness, San Pedro Parks Wilderness, Mesa Verde National Park,
Weminuche Wilderness, Black Canyon of the Gunnison Wilderness, West
Elk Wilderness, Maroon Bells Wilderness, Flat Tops Wilderness, Arches
National Park, Canyonlands National Park, Capital Reef National Park,
Bryce Canyon National Park, and Zion National Park. 40 C.F.R.
§ 51.309(b)(1).
8
(2) a trading program for the nine states.
The trading program acted as a “backstop,” which would be triggered only
if the milestones were reached.
2. The Regional Haze Rule
In 1999, the EPA adopted the Transport Commission’s
recommendations in its Regional Haze Rule, 40 C.F.R. §§ 51.308, 51.309.
This rule requires states to develop programs that assure reasonable
progress toward meeting the national goal of addressing visibility
impairment in Class I areas. 40 C.F.R. § 51.300(a). Sections 51.308 and
51.309 create two methods of compliance.
Under the first method, states can submit an implementation plan
containing emission limitations applying BART for each BART-eligible
source impairing visibility in a Class I area. 40 C.F.R. § 51.308(e).
The second method is authorized in 40 C.F.R. § 51.309. Through
this method, states could use the Transport Commission’s cap-and-trade
program if participants would expect better results than they would have
had under BART regulations. The cap-and-trade program is known as the
“309 program.” 7
The 309 program establishes voluntary measures to reduce sulfur-
dioxide emissions through milestones providing “steady and continuing
7
States opting for a 309 program still had to comply with § 51.308
with respect to any other Federal Class I area not encompassed in the 309
program. 40 C.F.R. § 51.309(a).
9
emissions reductions through 2018.” 40 C.F.R. § 51.309(d)(4)(i). After
2018, the milestone remains constant until the states submit revised
implementation plans. Id. § 51.309(d)(4)(vi)(A). These milestones must
provide a “50 to 70 percent reduction in [sulfur dioxide] emissions from
1990 actual emission levels by 2040.” Id. § 51.309(d)(4)(i).
If sulfur-dioxide emissions surpass the milestone, a backstop
regional emission trading program would be triggered. Under the program,
sources are given a set volume of emissions. Any source exceeding its
allowance must pay a penalty and suffer a loss in its allotted emissions.
Joint App. at 226-27. To encourage early reductions in emissions, the
trading program provided additional allocations to sources that reduce
emissions ahead of schedule.
Upon approval of an implementation plan, the EPA would regard the
state to be in compliance through 2018 with the reasonable-progress
requirement for the sixteen Class I areas encompassed in the 309 program.
40 C.F.R. § 51.309(a). For additional Class I areas not covered in the 309
program, the state had to show long-term strategies under § 308. Id.
§ 51.309(g).
3. The D.C. Circuit Court’s Rulings
After the Western Regional Air Partnership submitted its report,
Arizona, New Mexico, Oregon, Utah, Wyoming, the City of Albuquerque,
and Bernalillo County chose to participate in the 309 program.
10
Before the EPA acted on these participants’ submissions, the D.C. Circuit
Court of Appeals invalidated part of the § 51.308(e) methodology
(requiring evaluation of progress by considering emission reductions in the
aggregate). Am. Corn Growers Ass’n v. EPA, 291 F.3d 1, 8-9 (D.C. Cir.
2002).
The EPA continued to apply the invalidated methodology in the
context of determining whether the 309 program was better than BART, but
the D.C. Circuit Court of Appeals again struck down the EPA’s action in
Center for Energy & Economic Development v. EPA, 398 F.3d 653, 660
(D.C. Cir. 2005). There the court upheld the EPA’s view that an
alternative program could satisfy the reasonable progress goals. But, the
court held that the EPA should not have used the invalidated methodology.
Ctr. for Energy & Econ. Dev., 398 F.3d at 654.
4. Regional Haze Rule Revisions
In 2006, the EPA responded to these decisions by revising the
Regional Haze Rule, making the evaluation of the final BART factor a
source-by-source determination rather than one based on an evaluation of
emission reductions in the aggregate. Final Rule, Regional Haze
Regulations; Revisions to Provisions Governing Alternative to Source-
Specific BART Determinations, 71 Fed. Reg. 60,612, 60,612-13 (Oct. 13,
2006). Thus, the participating states had to resubmit implementation
plans.
11
5. Subsequent Implementation Plans Adopting 309 Program
Arizona and Oregon decided not to participate in the 309 program.
But New Mexico, Utah, Wyoming, the City of Albuquerque, and Bernalillo
County resubmitted plans for a 309 program. In the new plans, the
participants adjusted the emission milestones to account for withdrawal of
Arizona and Oregon and reductions already achieved under the 2003
milestones. Joint App. at 426, 430-38.
The new implementation plans set the following regional milestones:
● 269,083 tons of sulfur dioxide in 2008,
● 234,903 tons of sulfur dioxide in 2009,
● 200,722 tons of sulfur dioxide in 2010-2012,
● 185,795 tons of sulfur dioxide in 2013,
● 170,868 tons of sulfur dioxide in 2014,
● 155,940 tons of sulfur dioxide in 2015-17, and
● 141,849 tons of sulfur dioxide in 2018 and beyond.
Id. at 461.
When determining whether the 309 program would outperform
BART, the participants considered BART-eligible sources and other
sources. Because presumptive rates were not established for the other
sources, the states analyzed individual sources to determine the emission-
rate benchmark for sources that were ineligible under BART. Id.
12
The 309 program set the 2018 milestone to the BART benchmark
based on the presumptive BART in Appendix Y. But New Mexico, Utah,
Wyoming, the City of Albuquerque, and Bernalillo County determined that
the 309 program would outperform BART by:
 encouraging early cuts in emissions,
 including non-BART stationary sources, covering 63 more
sources that produce emissions,
 capping growth in new sources,
 addressing not only stationary sources but also mobile sources,
fire, and clean air corridors (which are not covered by BART),
and
 establishing a “mass-based cap,” which created an absolute
limit on allowable emissions (unaffected by demand
fluctuations or operational malfunctions that could increase
emissions).
In 2011, New Mexico, Utah, Wyoming, the City of Albuquerque, and
Bernalillo County revised their implementation plans adopting the 309
program. In late 2012, the EPA approved the plans, finding that the 309
program would achieve greater reasonable progress than BART. Id. at 1-
53. The Petitioners challenge the EPA’s approval of the 309 program.
II. Standard of Review
The Clean Air Act authorizes judicial review of the EPA’s approval
of state implementation plans, but does not designate the applicable
standard of review. 42 U.S.C. § 7607(b)(1). In conducting this review, we
13
are bound by the Administrative Procedure Act. See Oklahoma v. EPA,
723 F.3d 1201, 1211 (10th Cir. 2013) (“We follow the standards of the
Administrative Procedure Act . . . in reviewing the EPA’s actions under the
[Clean Air Act].”).
Under the Administrative Procedure Act, we can reverse agency
action only if it is “arbitrary, capricious, an abuse of discretion, or
otherwise not in accordance with law.” 5 U.S.C. § 706(2)(A). This
standard requires us to determine whether the agency considered the
relevant data and rationally explained its decision. See In re FCC, 753
F.3d 1015, 1041 (10th Cir. 2014). Under this standard, we will not disturb
an agency action unless the agency
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 is so implausible
that it could not be ascribed to a difference in view or the
product of agency expertise.
Motor Vehicle Mfrs. Ass’n of the U.S., Inc. v. State Farm Mut. Auto. Ins.
Co., 463 U.S. 29, 43 (1983).
If the agency’s “‘path may reasonably be discerned’” from its
explanation, we will not disturb the action even when the explanation is
not entirely clear. Alaska Dep’t of Envtl. Conservation v. EPA, 540 U.S.
461, 497 (2004) (quoting Bowman Transp., Inc. v. Ark.-Best Freight Sys.,
Inc., 419 U.S. 281, 286 (1974)). When an agency acts under an “‘unwieldy
14
and science-driven statutory scheme[] like the Clean Air Act,’” we afford
the agency “‘particular deference.’” Nat’l Ass’n of Clean Air Agencies v.
EPA, 489 F.3d 1221, 1229 (D.C. Cir. 2007) (quoting Bluewater Network v.
EPA, 372 F.3d 404, 410 (D.C. Cir. 2004)).
III. The EPA’s Approval of the Implementation Plans
The environmental groups argue that:
● the 309 program does not achieve greater reasonable progress
than implementation of BART,
● the 309 program will not achieve reasonable progress toward
eliminating visibility impairment because only three of the nine
eligible states participated, and
● New Mexico’s program is deficient based on the failure to
analyze emissions from the Escalante coal plant.
We reject each argument.
A. The EPA’s Determination that the 309 Program Was Better
than BART
The Regional Haze Rule establishes three steps for states to follow
when determining whether an alternative program is better than BART:
(1) establish a BART benchmark and predict emission reductions if
BART were implemented;
(2) predict the emission reductions from an alternative program;
and
(3) compare the two.
40 C.F.R. § 51.308(e)(2).
15
The environmental groups contend that the EPA acted arbitrarily and
capriciously because:
● the BART benchmark improperly adopted the presumptive
emission rate established in Appendix Y of the BART
guidelines,
● the participants misapplied the “clear weight of the evidence”
standard, and
● the participants overstated the effectiveness of the 309 program
by inappropriately considering qualitative factors.
The EPA’s approval of the better-than-BART determination was not
arbitrary or capricious because:
● the presumptive BART benchmark was appropriate under the
Clean Air Act and Regional Haze Rule,
● the Regional Haze Rule allows comparison of BART to the
alternative measure through a “clear weight of the evidence”
standard, and
● the participants properly considered qualitative factors in
determining the effectiveness of the 309 program.
1. BART Benchmark
When approving the implementation plans, the EPA concluded: “The
BART benchmark calculation . . . was not intended to assess actual
emissions at BART subject sources nor was it intended to assess the
control capabilities of later installed controls. Instead, the presumptive
[sulfur dioxide] emissions rate served as a necessary simplifying
assumption.” Wyoming Rule, 77 Fed. Reg. at 73,929-30.
16
The environmental groups challenge this conclusion, arguing that
source-specific data showed that actual emissions fell below the BART
benchmark. For this argument, the environmental groups rely on a report
prepared by their expert witness, Ms. Vicki Stamper. Ms. Stamper
concluded that source-by-source BART analyses and actual emissions were
lower than the presumptive BART benchmark used to approve the 309
program.
Relying on Ms. Stamper’s conclusions, the environmental groups
contend that:
● the participants should have conducted a source-by-source
BART analysis instead of relying on the presumptive BART
benchmark, and
● the BART benchmark was inappropriate.
The first argument is untimely. Under federal law, a petitioner has
only 60 days to sue after the agency acts. 42 U.S.C. § 7607(b)(1); see
Utah v. EPA, 750 F.3d 1182, 1184 (10th Cir. 2014). The environmental
groups failed to comply with this deadline.
The environmental groups’ second challenge, involving the validity
of the BART benchmark, fails on the merits.
a. The 309 Program’s Presumptive BART Benchmark
The first challenge requires an understanding of:
● the regulatory requirements for states to establish the BART
benchmark, and
17
● how the participants used the presumptive BART benchmark in
their better-than-BART analyses.
As noted above, participants in the 309 program had to determine the
“BART benchmark,” which represents the expected emissions under a
BART regime. See 40 C.F.R. § 51.308(e)(2).
The BART benchmark is typically set by determining how much
sulfur dioxide would be emitted by each BART-regulated source. See id.
§ 51.308(e)(2)(i)(C) (“This analysis must be conducted by making a
determination of BART for each source subject to BART and covered by
the alternative program as provided for in [the subsection outlining the
BART determination].”).
But an exception exists when the alternative program is designed to
achieve a requirement other than BART, such as the reasonable progress
goals. In this situation, a source-by-source BART determination is not
necessary to determine the BART benchmark. Instead, the state could
determine the BART benchmark “based on both source-specific and
category-wide information, as appropriate.” Id.
The 309 program was designed to implement something other than
BART: the recommendations of the Transport Commission toward
eradication of regional haze over the Colorado Plateau. Accordingly, the
participants did not need to conduct source-by-source BART
18
determinations to establish the BART benchmark. And the participants did
not do so.
Instead, the participants relied on the Western Regional Air
Partnership’s better-than-BART analysis. As the BART benchmark, the
Air Partnership determined that “[a]ll utilities that were determined to be
subject to BART were assumed to be operating at the presumptive emission
rate established in 40 CFR Part 51, Appendix Y [0.15 pound per million
British thermal units].” Joint App. at 435.
The “Appendix Y” presumptive emission rate refers to the EPA’s
2005 amendment to the Regional Haze Rule, which added guidelines to
instruct states analyzing individual sources under BART. See 70 Fed. Reg.
39,131-32; 40 C.F.R. pt. 51, App. Y. In Appendix Y, the EPA established
a presumptive BART emission rate of 0.15 pound per million British
thermal units for BART-eligible sources.
The Western Regional Air Partnership relied on Appendix Y’s
presumptive emission rate for all but two BART-eligible sources. For
these two sources (the Hunter and Huntington power plant units), a lower
BART emission rate of 0.12 pound per million British thermal units was
used based on limits already in place. Joint App. at 448.
b. Timeliness
The environmental groups argue that the participants should have set
the BART benchmark by predicting emissions for each BART-regulated
19
source (rather than relying on the presumptive rate in Appendix Y). This
argument is not timely.
When amending the Regional Haze Rule in 2006, the EPA recognized
that “the [Appendix Y] presumptions represent[ed] a reasonable estimate of
a stringent case BART.” 71 Fed. Reg. at 60,619. In light of the
reasonableness of this estimate, the EPA decided in 2006 that participants
could rely on the presumptive rate when attempting to meet a requirement
other than BART. By using the presumptive rate, participants could avoid
the need to predict emissions for each source under a BART system of
regulation. Id. at 60,618-19.
The environmental groups argue that the participants should have
conducted their own source-specific BART analyses. In the absence of
these analyses, the environmental groups contend that the EPA arbitrarily
approved use of the presumptive BART benchmark. In effect, this
contention challenges the EPA’s 2006 amendment establishing use of the
presumptive BART benchmark. We can adopt this view only if we
conclude that the EPA erred in adopting the Appendix Y BART as a
presumptive benchmark. 8
8
The environmental groups deny that they are asserting a need for a
source-by-source BART determination. But in her expert report, Ms.
Stamper stated that § 51.308(e)(2)(i)(C) required a source-by-source BART
determination. Joint App. at 684. And the environmental groups rely
heavily on this report in criticizing the presumptive BART benchmark.
20
It is too late for the Petitioners to make this argument. If the
environmental groups wished to challenge adoption of Appendix Y as the
presumptive BART emission rate, they had to file a petition for review
within 60 days of the EPA’s publication of the 2006 amendment to the
Regional Haze Rule. See 42 U.S.C. § 7607(b)(1); Utah v. EPA, 750 F.3d
1182, 1184 (10th Cir. 2014). That amendment was published in the
Federal Register on October 13, 2006, and the petitions for review were
not filed until more than six years later (December 2012 and January
2013). See 71 Fed. Reg. at 60,612; 40 C.F.R. Pt. 51, App. Y.
Accordingly, we lack jurisdiction over a challenge to the EPA regulation
authorizing use of Appendix Y in lieu of a source-by-source determination.
See Utah, 750 F.3d at 1184; Utah v. EPA, ___ F.3d ___, Nos. 13-9535, 13-
9536, 2014 WL 4345770, at *5 (10th Cir. Sept. 3, 2014).
Pet’rs’ Opening Br. at 34, 41 (arguing that the presumptive BART was
“much less stringent than source-by-source BART determinations”).
In their reply brief, the environmental groups retreat from this
argument: “Contrary to EPA’s assertions in its answering brief, Petitioners
do not contend that states participating in the 309 Program must conduct
source-by-source BART determinations based on the methodology in 40
C.F.R. § 51.308(e)(1) and the BART Guidelines in order to develop a valid
BART benchmark.” Pet’rs’ Reply Br. at 6.
Even if the environmental groups had not retreated from this
argument, it would have been untimely. Thus, we need not address the
parties’ disagreement over Ms. Stamper’s opinions on emissions from
BART-regulated sources. See, e.g., Industry Intervenors’ Response Br. at
34.
21
c. Actual Emissions Lower than the Presumptive BART
Emission Rate
The environmental groups also invoke § 51.308(e)(2)(i)(C), arguing
that the EPA should have considered whether use of category-wide
information was “appropriate.” Pet’rs’ Reply Br. at 26; see 40 C.F.R.
§ 51.308(e)(2)(i)(C). 9 This argument stems from unreasonable assumptions
about the information available to the states when they submitted their
plans.
According to the environmental groups, most BART-eligible sources
emitted less sulfur dioxide than the presumptive benchmark would allow
and the EPA elsewhere projected even lower presumptive emission rates.
In the face of this data, the environmental groups argue that
§ 51.308(e)(2)(i)(C) would prohibit states from relying on Appendix Y’s
presumptive BART rate.
9
The environmental groups argue:
Petitioners challenge EPA’s application of th[e] authorization
[to use a simplifying presumption] and interpretation to
establish the BART benchmark despite readily available
category-wide and source-specific evidence that the
presumptive rate is not an “appropriate” simplifying
assumption because it grossly underestimates the emission
reductions achievable by installing BART at the affected
sources.
Pet’rs’ Reply Br. at 26.
22
This argument is based largely on the report of Ms. Stamper, who
said that 17 of the BART-eligible sources had emission rates that dipped
below the rates allowed in Appendix Y. Reliance on Ms. Stamper’s report
is misguided. Ms. Stamper relied on contemporaneous measures of
emissions post-dating the participants’ implementation plans, and the EPA
regulations expressly allowed reliance on the presumptive rate.
The Western Regional Air Partnership submitted its better-than-
BART determination in October 2010, and the participants relied on this
determination in their 2011 implementation plans. Joint App. at 435; see
Final Rule, Approval and Promulgation of State Implementation Plans;
Wyoming, 77 Fed. Reg. 73,926, 73,926 (Dec. 12, 2012); Final Rule,
Approval, Disapproval and Promulgation of State Implementation Plans;
Utah, 77 Fed. Reg. 74,355, 74,355 (Dec. 14, 2012); Final Rule, Approval
and Promulgation of State Implementation Plans; New Mexico, 77 Fed.
Reg. 70,693, 70,693 (Nov. 27, 2012); Final Rule, Approval and
Promulgation of State Implementation Plans; City of Albuquerque-
Bernalillo County, 77 Fed. Reg. 71,119, 71,119 (Nov. 29, 2012).
The environmental groups contend that the participants should have
accounted for actual emissions. The EPA could reasonably conclude that
inclusion of Ms. Stamper’s data would have been infeasible, for the better-
than-BART determination resulted from coordinated efforts by the
23
participants over several years 10 and much of the omitted data did not even
exist until this process had almost come to an end. Thus, the EPA rejected
the environmental groups’ insistence that the participants should have
incorporated the new data. Joint App. at 30-31. This conclusion was not
arbitrary or capricious. See San Luis & Delta-Mendota Water Auth. v.
Jewell, 747 F.3d 581, 620-21 (9th Cir. 2014) (holding that the Fish and
Wildlife Service’s choice of a baseline, though imperfect, was not arbitrary
or capricious because removal of the imperfections would not have been
feasible).
Reliance on Ms. Stamper’s data was not only infeasible, but also
invalid under the EPA regulations. These regulations expressly allowed
participants to use the presumptive benchmark to predict emissions instead
of assessing how much pollution would be emitted from each source under
a BART regime. 71 Fed. Reg. at 60,618-19. Ms. Stamper’s analysis
suggests that the presumptive benchmark is overly generous for some
sources. But imprecision is inherent in the nature of a simplifying
assumption.
In arguing that the EPA disregarded site-specific information, the
environmental groups refer to two units (the Hunter Unit 1 and the Dave
Johnson Unit 4) and point out that the EPA used actual emissions in the
Cross-State Air Pollution Rule. See Final Rule, Regional Haze; Revision
10
Joint App. at 175, 426.
24
to Provisions Governing Alternatives to Source-Specific BART
Determinations, 77 Fed. Reg. 33,642, 33,649 (June 7, 2012); Proposed
Rule, Regional Haze; Revisions to Provisions Governing Alternatives to
Source-Specific BART Determinations, 76 Fed. Reg. 82,219, 82,225-26
(Dec. 30, 2011).
The EPA’s use of actual emissions in one rule does not require the
EPA to use actual emissions in every rule. And, the regulations expressly
allow participants to use the benchmark in lieu of actual emissions. Thus,
the EPA interpreted its Regional Haze Rule and concluded:
● “[T]here is no need to develop a precise estimate of the
emissions reductions that could be achieved by BART in order
simply to compare two programs,” and
● “the [Appendix Y] presumptions represent a reasonable
estimate of a stringent case BART.”
71 Fed. Reg. at 60,618-19.
This interpretation was reasonable. Section 51.308 mandates the use
of source-specific and category-wide information “as appropriate.” 40
C.F.R. § 51.308(e)(2)(i)(C). Information may be appropriate in one
context, but not another. Section 51.308 provides flexibility in what may
be considered, and the EPA reasonably interpreted that provision.
d. The EPA’s Statements Regarding Appendix Y’s Presumptive
BART
The environmental groups also argue that the presumptive rate (0.15
pound per million British thermal units) is rebuttable and serves only as
25
the starting point of the BART analysis. Pet’rs’ Opening Br. at 42. This
argument is rejected.
For this argument, the environmental groups refer to other rules in
which the EPA has clarified the BART analysis for states and the role of
Appendix Y. Id. at 43. For example, the environmental groups point to
the rejection of Arkansas’ implementation plan, where the EPA said that
states must “‘consider the level of control that [was] currently achievable
at the time the BART analysis [was] being conducted.’” Id. (quoting Final
Rule; Approval and Promulgation of Implementation Plans; Arkansas, 77
Fed. Reg. 14,604, 14,613-14 (Mar. 12, 2012)).
This argument overlooks a critical distinction. In the cited instances,
the states were conducting a BART analysis. Here, they weren’t. Instead,
the participants in our case were conducting a better-than-BART
determination. This analysis required a comparison of the 309 program to
the BART benchmark, which adopted Appendix Y’s presumptive BART as
a simplifying assumption. The environmental groups have not identified
any authority requiring a source-by-source analysis for states conducting a
309 program.
2. Comparison of the 309 Program to BART
At the second and third stages of the better-than-BART analysis, the
participant must:
26
(1) predict the emission reductions achieved by implementing the
alternative program (the second stage), and
(2) compare the effectiveness of the alternative measure to the
effectiveness of implementing BART (the third stage).
40 C.F.R. § 51.308(e)(2)(i)(D), (E). The environmental groups challenge
the EPA’s approval at both stages.
At the third stage, the groups claim that the participants used an
improper method of comparison. And at the second stage, the groups
allege improper reliance on qualitative factors to bolster the effectiveness
of the 309 program. Because the applicability of qualitative factors at the
second stage depends on the method used at the third stage, we first
address whether the participants used the proper method to compare the
309 program to BART.
a. Comparison of the Milestones (in the 309 Program) to BART
The EPA compared the 309 program as a whole to BART. The
environmental groups suggest in their reply brief that the EPA should have
compared the 309 program milestones (rather than the 309 program as a
whole) to the effectiveness of BART. See 40 C.F.R. § 51.309(d)(4)(i).
But we cannot entertain this suggestion because it was unexhausted and
omitted in the environmental groups’ opening brief.
Under the Clean Air Act, “[o]nly an objection to a rule or procedure
which was raised with reasonable specificity during the period for public
comment (including any public hearing) may be raised during judicial
27
review.” 42 U.S.C. § 7607(d)(7)(B); see EPA v. EME Homer City
Generation, L.P., __ U.S.__, 134 S. Ct. 1584, 1602 (2014) (holding that
satisfaction of § 7607(d)(7)(B) was mandatory, but not jurisdictional). But
the environmental groups failed to raise this issue in the EPA proceedings.
That failure renders the claim unexhausted. See Oklahoma v. EPA, 723
F.3d 1201, 1214-15 (10th Cir. 2013).
Even in the present action, the environmental groups did not raise the
issue until they filed their reply brief. By then it was too late to raise a
new issue. See M.D. Mark, Inc. v. Kerr-McGee Corp., 565 F.3d 753, 768
n.7 (10th Cir. 2009) (the “general rule in this circuit is that a party waives
issues and arguments raised for the first time in a reply brief”).
We decline to entertain this issue, for it was not exhausted or raised
in the environmental groups’ opening brief.
b. Consideration of Qualitative Factors
The EPA determined that the 309 program as a whole was more
effective than a fully implemented BART regime because the 309 program:
(1) included non-BART sources of sulfur-dioxide emissions,
(2) included new sources of emissions,
(3) created a “mass-based” cap covering emissions in the
aggregate, and
(4) encouraged early reductions in emissions.
28
The environmental groups characterize this rationale as qualitative and
argue that the EPA should instead have focused solely on quantitative
considerations. We reject this argument.
i. Failure to Use § 51.308(e)(3)’s Method
According to the environmental groups, a quantitative approach was
required in 40 C.F.R. § 51.308(e)(3). Instead of using one of the
quantitative methods, the groups continue, the EPA improperly applied a
qualitative standard (“clear weight of the evidence”).
Section 51.308(e)(2)(i) sets out the process to compare an alternative
program to BART. This section did not initially articulate a method for
the comparison. The EPA considered a method that would compare the
“expected visibility improvement under the alternative program and under
BART according to the criteria established in § 51.308(e)(3).” Proposed
Rule, Regional Haze Regulations; Revisions to Provisions Governing
Alternative to Source-Specific BART Determinations, 70 Fed. Reg. 44,154,
44,158 (Aug. 1, 2005). The EPA also sought comment on: (1) whether
§ 51.308(e)(3) provided the sole way to demonstrate greater reasonable
progress, or (2) whether qualitative factors could be considered. Id.
In 2006, the EPA determined that § 51.308(e)(3) should not serve as
the only means to show “greater reasonable progress.” Thus, the EPA
amended § 51.308(e)(2)(i) to add “E,” which authorized use of the “clear
weight of evidence” standard as a way of showing that the alternative
29
program was better than BART. Final Rule, Regional Haze Regulations;
Revisions to Provisions Governing Alternative to Source-Specific BART
Determinations, 71 Fed. Reg. 60,612, 60,622 (Oct. 13, 2006). Under
§ 51.308(e)(2)(i)(E), the better-than-BART analysis may be made “under
paragraph (e)(3) of this section or otherwise based on the clear weight of
evidence that the trading program or other alternative measure achieves
greater reasonable progress than would be achieved through the installation
and operation of BART at the covered sources.” 40 C.F.R.
§ 51.308(e)(2)(i)(E).
Accordingly, the regulation establishes two ways that a state can
compare a 309 program to BART. The state can use the two quantitative
methods stated in § 51.308(e)(3) or apply a qualitative standard (the clear
weight of evidence). See 71 Fed. Reg. at 60,622 (“With respect to the use
of a ‘weight of evidence’ approach as an alternative to the methodology of
section 51.308(e)(3), we support the use of such a test as an alternative to
the methodology set forth in section 51.308(e)(3).”).
The participants chose the qualitative standard, which was
permissible under the EPA’s interpretation of its regulations. See Fed.
Express Corp. v. Holowecki, 552 U.S. 389, 397 (2008) (“Just as we defer to
an agency’s reasonable interpretations of [its authorizing] statute when it
issues regulations in the first instance, . . . the agency is entitled to further
30
deference when it adopts a reasonable interpretation of regulations it has
put in force.” (citation omitted)).
ii. Qualitative v. Quantitative Factors
The environmental groups argue that even if § 51.308(e)(3) did not
furnish the exclusive methodology, the participants should not have relied
on qualitative factors because:
(1) the EPA sought comment on, but did not adopt, a “qualitative”
means of evaluating whether an alternative program was better
than BART, and
(2) when adding a “clear weight of the evidence” standard, the
EPA identified only quantitative emissions and visibility data
as appropriate for a better-than-BART determination.
These arguments do not suggest arbitrariness or capriciousness by the
EPA. Both arguments depend on the environmental groups’ interpretation
of the EPA regulations. The EPA expressly concluded that a participant
could use the “clear weight of the evidence” standard. When using this
standard, however, the EPA sanctioned consideration of “all available
information.” 11 There was no prohibition against the consideration of
qualitative evidence.
11
In its final rule adding the “clear weight of the evidence” standard as
one means of determining that an alternative program was better than
BART, the EPA clarified:
“Weight of evidence” demonstrations attempt to make use of all
available information and data which can inform a decision
while recognizing the relative strengths and weaknesses of that
information in arriving at the soundest decision possible.
31
It is true that the EPA provided examples that are quantitative. See
Final Rule, Regional Haze Regulations; Revisions to Provisions Governing
Alternative to Source-Specific BART Determinations, 71 Fed. Reg. 60,612,
60,622 (Oct. 13, 2006). But the EPA pointed out that these examples were
not exhaustive and that the determination should be based on “all available
information and data which can inform a decision while recognizing the
relative strengths and weaknesses of that information in arriving at the
soundest decision possible.” Id.
Because this language supports the EPA’s interpretation of its
regulation, we do not regard the use of qualitative factors as arbitrary or
Factors which can be used in a weight of evidence
determination in this context may include, but not be limited to,
future projected emissions levels under the program as
compared to under BART, future projected visibility conditions
under the two scenarios, the geographic distribution of sources
likely to reduce or increase emissions under the program as
compared to BART sources, monitoring data and emissions
inventories, and sensitivity analyses of any models used. This
array of information and other relevant data may be of
sufficient quality to inform the comparison of visibility impacts
between BART and the alternative program. In showing that an
alternative program is better than BART and when there is
confidence that the difference in visibility impacts between
BART and the alternative scenarios are expected to be large
enough, a weight of evidence comparison may be warranted in
making the comparison. The EPA will carefully consider the
evidence before us in evaluating any [state implementation
plans] submitted by States employing such an approach.
Final Rule, Regional Haze Regulations; Revisions to Provisions Governing
Alternative to Source-Specific BART Determinations, 71 Fed. Reg. 60,612,
60,622 (Oct. 13, 2006) (emphases added).
32
capricious. See Fed. Express Corp. v. Holowecki, 552 U.S. 389, 397
(2008).
c. The Qualitative Factors
The EPA relied on four factors that could be considered
“qualitative”:
(1) The trading program included sources not subject to BART
regulation;
(2) the trading program discouraged emissions from new sources
more effectively than a BART regime would have done;
(3) the trading program included an aggregate cap on emissions,
which would have decreased emissions more effectively than
BART; and
(4) the trading program encouraged earlier reductions than under a
BART regime.
Joint App. at 31-32. These considerations provided a reasonable basis for
the EPA’s approval of the 309 program.
i. Emission Reductions from Non-BART Sources
In concluding that the 309 program would outperform BART, the
EPA relied in part on inclusion of “all sources with emissions greater than
100 tons/year of [sulfur dioxide].” See id. at 516. The threshold for
regulation under BART would have been much higher. See 40 C.F.R.
§ 51.301 (stating that sources are eligible for BART if they can emit 250
tons of sulfur dioxide per year, were built between 1962 and 1977, and fall
within one of the specified source categories).
33
The environmental groups criticize the EPA for considering potential
reductions in emissions involving non-BART sources. The groups argue
that rather than consider non-BART sources, the EPA should have confined
its analysis to BART-eligible sources.
We disagree. The environmental groups are relying on regulatory
language applicable to the first step of the better-than-BART analysis (the
determination of the BART benchmark), not the comparison of BART to
the alternative program.
Under 40 C.F.R. § 51.308(e)(2)(i), subsections “A” through “C”
outline the requirements to determine the BART benchmark, the first step
of the better-than-BART determination. 40 C.F.R. § 51.308(e)(2)(i)(A)-
(C). Subsection “D” addresses the second step: “[a]n analysis of the
projected emissions reductions achievable through the trading program or
other alternative measure.” Id. § 51.308(e)(2)(i)(D). And, as previously
discussed, subsection “E” states how one compares the relative successes
of the alternative program and BART, the third and final step in the better-
than-BART analysis. Id. § 51.308(e)(2)(i)(E) (requiring a comparison of
“the trading program or other alternative measure” with BART).
Accordingly, the environmental groups are mistaken. The EPA could
reasonably read “D” and “E” to allow comparison of BART to the entirety
of the alternative program (including the non-BART-eligible sources).
34
Thus, the EPA acted reasonably when it considered non-BART sources at
the second and third steps of the better-than-BART determination.
ii. Emission Reductions from New Sources
The environmental groups also argue that the EPA incorrectly
considered emission reductions from new sources. New sources would not
be subject to BART because they would have been built after 1977. See id.
§ 51.301. The groups contend that new sources are already subject to
regulations that are more effective than a cap on emissions in the 309
program. We reject this contention.
As the environmental groups state, new sources of emissions are
independently regulated by the Clean Air Act. 42 U.S.C. §§ 7411, 7475,
7503; see United States v. DTE Energy Co., 711 F.3d 643, 644-45 (6th Cir.
2013) (“New Source Review [under the Clean Air Act] forbids the
construction of new sources of air pollution without a permit.”). Thus,
even without the trading program, new sources would need to show that
they meet emission standards based on the “best available control
technology.” 42 U.S.C. § 7475(a)(4). But, the EPA could reasonably
conclude that the 309 program would go beyond the existing regulatory
process in reducing emissions from new sources.
The environmental groups contend that a cap would prove
meaningless by allowing new sources to obtain allocations equaling the
maximum emissions already allowed. We disagree.
35
Under the new-source regulatory scheme, new-source emissions were
limited but not capped. The EPA set out to establish a cap through
approval of 309 programs. With caps, the EPA expected polluters to adopt
voluntary measures to reduce emissions. Joint App. at 431-32, 438. The
EPA coupled this strategy with regulation for new sources. Id. This two-
fold strategy for new sources had a reasonable foundation: The EPA hoped
to reduce emissions for new sources by regulating them and encouraging
voluntary reductions in emissions. Id. at 31-32.
According to the environmental groups, the cap is ineffective
because it accommodated construction of all projected new electric
generating units proposed, which renders the “cap” on future sources no
better than the new-source regulatory scheme. Id. at 212, 432. Under the
309 program, however, the 2018 milestone continues as an emission cap
for sulfur dioxide until the participants obtain approval of revised
implementation plans. Accordingly, any post-2018 growth will be limited
unless a revised implementation mandates otherwise. In view of this
strategy by the EPA, its consideration of new sources was not arbitrary or
capricious.
iii. “Mass-Based Cap” on Sulfur-Dioxide Emissions
The environmental groups also contend that:
(1) the EPA improperly relied on the purported benefits of a
“mass-based cap” on sulfur-dioxide emissions,
36
(2) the mass-based cap cannot outperform BART because the cap
assumes that sources were operating at 85% capacity when
many of those sources were actually operating at lower
capacity,
(3) setting the assumption of capacity so high allows sources to
actually increase emissions, and
(4) BART would reduce emission rates across all operations even
when they are operating at less than full capacity.
The EPA disagreed and had a reasonable foundation for its disagreement.
Id. at 31-32.
The Western Regional Air Partnership designed the mass-based cap
to allow for an increase in operating capacity at existing sources in light of
a projected increase in electrical needs. By setting the assumed capacity at
85%, designers of the program established room for sources to adapt to
future needs. The EPA approved the mass-based cap only after concluding
that a cumulative limit on emissions would be more effective than BART.
It is true that a source’s presumptive capacity may be higher than the
actual capacity at any given time. But this possibility does not render the
EPA action arbitrary or capricious. The participants followed the concept
stated by the Western Regional Air Partnership, setting a cap based on
projected increases in electrical needs and accommodation of future
growth. Id. at 32. Based in part on the Air Partnership’s analysis, the
EPA determined that the 309 program would be better than a BART system
of regulation. Id. This determination was not arbitrary or capricious.
37
iv. Early Emission Reductions
In 1996, the Transport Commission recommended that the market
trading program “contain specific provisions to encourage and reward early
emission reductions, including reductions achieved before 2000.” Id. at
437 (internal quotation marks omitted). Following this recommendation,
the participants provided additional allocations to sources that reduce
emissions ahead of schedule.
In its 2010 report, the Western Regional Air Partnership concluded
that participants in the trading program had decreased sulfur-dioxide
emissions:
● 25% between 1990 and 2000 in the nine states eligible to
participate in the 309 program, and
● an additional 31% between 2000 and 2008 in the participating
states.
Id. at 438. The Western Regional Air Partnership attributed these
reductions to the 2003 implementation plans.
The environmental groups question the connection between the early
reductions and the 309 program. But the EPA never attributed the early
reductions to the 309 program. Instead, the EPA simply said that it could
not discount the possibility of a causal relationship. For example, when
approving the 309 program, the EPA stated that it could not “discount that
the 2003 309 [state implementation plan] submittal may have already
influenced sources to upgrade their plants before any case-by-case BART
38
determination under Section 308 may have required it.” Final Rule,
Approval and Promulgation of State Implementation Plans; Wyoming, 77
Fed. Reg. 73,926, 73,930 (Dec. 12, 2012).
In oral argument, the EPA acknowledged that it was aware of the
early reductions, but did not explicitly attribute them to the 309 program.
Instead, the EPA argued that proof of a causal relationship was
unnecessary. Oral Arg. 31:45-35:29. We agree: The EPA was not
required to prove a causal relationship between the already-achieved
emission reductions and the decade-long progression of the 309 program.
Rather, in its better-than-BART determination of the 309 program, the EPA
had to predict whether the alternative program would yield greater
reductions than a fully-implemented BART regime. See 40 C.F.R.
§ 51.308(e)(2)(i).
The existing reductions tended to support the soundness of a strategy
encouraging early reductions through the 309 program. The EPA had no
need to go further by proving actual causation between the strategy and the
early reductions. Thus, the EPA did not act arbitrarily or capriciously in
considering the early-reduction incentives.
d. Summary
The approval of the participants’ better-than-BART determination
was not arbitrary or capricious, and we reject the criticism of the EPA’s
39
reliance on qualitative factors and application of the “clear weight of the
evidence” standard.
B. “Critical Mass” of Participating States in 309 Program
We must also address the soundness of the 309 program based on the
number of states and tribes refusing to participate. Six out of the nine
eligible states refused to participate, as did every one of the 211 eligible
tribes. Joint App. at 426, 652. The environmental groups argue that
without greater participation, the 309 program was doomed to fail. The
EPA acted reasonably in rejecting this argument.
1. Timeliness
The EPA contends that this challenge was not raised in a timely
manner. For this contention, the EPA characterizes the challenge as an
attack on the Regional Haze Rule.
We disagree with this characterization. The environmental groups
are not questioning the absence of a critical mass requirement in the
Regional Haze Rule. Instead, the groups are contending that the
participating states are too few to satisfy the statutory goal of reasonable
progress. The groups’ contention addresses the EPA’s approval, rather
than the validity of the Regional Haze Rule. This contention is timely.
40
2. The Absence of a Statutory or Regulatory Requirement of
Minimum Participation
Though the argument is timely, it is invalid because neither the Clean
Air Act nor the EPA regulations require participation by a certain number
of states or tribes. See 40 C.F.R. § 51.309(a), (e).
Without a statutory or regulatory requirement, the environmental
groups rely on the EPA’s proposed 2002 rulemaking. There the EPA
stated:
The requirements in 40 CFR 51.309, if revised, will be
the product of a substantial effort by many States, Tribes,
Federal agencies, and other interested parties, extending over a
number of years from the work of the [Grand Canyon Visibility
Transport Commission] to that of the [Western Regional Air
Partnership]. The EPA recognizes, however, that the States and
Tribes do have the option of implementing the regional haze
rule under 40 CFR 51.308 rather than 40 CFR 51.309. Because
the objective of 40 CFR 51.309 is to provide a regional
approach to protecting air quality at the 16 Class I areas on the
Colorado Plateau, EPA believes that there must be a “critical
mass” of States participating for 40 CFR 51.309 [state
implementation plans] to be approvable.
Proposed Rule, Proposed Revisions to Regional Haze Rule to Incorporate
Sulfur Dioxide Milestones and Backstop Emissions Trading Program, 67
Fed. Reg. 30,418, 30,420 (May 6, 2002); see also Final Rule, Revisions to
Regional Haze Rule to Incorporate Sulfur Dioxide Milestones and
Backstop Emissions Trading Program, 68 Fed. Reg. 33,764, 33,770 (June
5, 2003) (“The EPA continues to believe, as discussed in the proposal, that
41
judgments on the issue of ‘critical mass’ are best left to the [Western
Regional Air Partnership].”).
Later in this proposed revision, however, the EPA indicated that it
would “defer to the [Western Regional Air Partnership’s] judgment on the
issue of ‘critical mass,’ and . . . request[ed] comment on this proposal.”
67 Fed. Reg. at 30,427. And the Western Regional Air Partnership did not
require participation by a minimum number of states or tribes. See Joint
App. at 174-267. 12 Accordingly, the EPA did not impose such a
requirement. See 77 Fed. Reg. at 24,769-70 (“Section 51.309 does not
require the participation of a certain number of States to validate its
effectiveness.”).
12
Instead, the Air Partnership proposed further study on the number
and diversity of sources needed to make the program viable:
The Annex has been developed based on the Grand
Canyon Visibility Transport Commission recommendations,
which assumed that all of the states and tribes in the transport
region would participate in the program. The regional haze
rule establishes two paths for states: implement the
Commission recommendations, including the backstop trading
program under §309; or develop an independent plan under
§308. An important issue still to be addressed is the effect on
the trading program if one or more states and tribes do not
choose to participate. Will there be enough sources or enough
diversity in the program to create a viable market? Will the
administrative costs of the program be justifiable with a
smaller group of states and tribes? To address these questions,
the [Western Regional Air Partnership] needs to evaluate the
economics of the program, and determine the critical mass that
is needed to create a viable program.
Joint App. at 234.
42
3. The Environmental Groups’ Arguments on the Soundness of
the EPA’s Conclusion
The environmental groups argue that not enough states are
participating to allow reasonable progress because:
● the three states participating in the 309 program contribute
only a small percentage of the sulfur dioxide in Utah’s Class I
areas and the Colorado Plateau, and
● sources in the three participating states could shift emissions to
unregulated sources.
The arguments do not render the EPA’s determination arbitrary or
capricious because states remain regulated under BART when they decline
to participate in the 309 program. With continued regulation under BART,
the EPA reasonably concluded that the 309 program could work effectively
even without participation from heavy polluters. Joint App. at 29-30.
The environmental groups challenge the factual basis for this
conclusion because:
● the 3 states generating the greatest emissions (Nevada,
California, and Arizona) chose not to participate,
● the 309 program excludes dozens of coal-fired power plants,
● the 309 program encompasses only 15 coal-fired power plants,
and
● the participating states contribute only 36% of the sulfur-
dioxide emissions over the Colorado Plateau.
These factual arguments do not undermine the reasonableness of the
EPA’s prediction. Notwithstanding exclusion of many heavy polluters, the
EPA legitimately predicted that the 309 program would make “reasonable
43
progress” toward improvement of visibility over the Colorado Plateau. Id.
The excluded sources would still be regulated, though not under the 309
program.
The environmental groups counter that:
● the exclusions prevent the 309 program from qualifying as a
“regional” program, and
● the existence of a regional program is necessary for the EPA to
satisfy the statutory and regulatory purposes.
Pet’rs’ Reply Br. at 46-47 (citing 42 U.S.C. § 7492(c)). We disagree.
Section 7492 states that “[w]henever . . . the current or projected
interstate transport of air pollutants from one or more States contributes
significantly to visibility impairment in class I areas located in the affected
States, the Administrator may establish a transport region for such
pollutants that includes such States.” 42 U.S.C. § 7492(c)(1) (emphasis
added). Accordingly, the statutory authorization of regional programs does
not require a minimum number of participating states. See id.
The environmental groups also assert that exclusion of major coal-
fired power plants from the nonparticipating states would lead sources in
the three participating states to shift emissions to unregulated sources.
Pet’rs’ Opening Br. at 51. To illustrate this assertion, the environmental
groups refer to power plants owned by PacifiCorp. PacifiCorp could shift
emissions from power plants in Wyoming and Utah (states participating in
the 309 program) to power plants in Arizona and Colorado
44
(nonparticipating states). This shift would allow PacifiCorp to comply
with the milestones established in the 309 program while increasing
emissions in the nonparticipating states. According to the environmental
groups, this shift would impede the overall reduction of emissions in the
region and could even worsen visibility.
As support, the environmental groups cite the EPA’s statements in
the Clean Air Interstate Rule:
Inclusion of all units substantially in the electricity sales
business minimizes the potential for shifting utilization, and
emissions, from regulated to unregulated units in that business
and thereby freeing up allowances, with the result that total
emissions from generation of electricity for sale exceed the
[Clean Air Interstate Rule] emissions caps. The fact that units
in the electricity sales business are generally interconnected
through their access to the grid significantly increases the
potential for utilization shifting.
Final Rule, Rule to Reduce Interstate Transport of Fine Particulate Matter
and Ozone (Clean Air Interstate Rule), 70 Fed. Reg. 25,162, 25,277 (May
12, 2005) (emphasis added). 13
13
The D.C. Circuit Court of Appeals struck down the Clean Air
Interstate Rule, reasoning that it violated the statutory prohibition against
contribution of pollution in downwind states from sources within the
upwind states. North Carolina v. EPA, 531 F.3d 896, 908 (D.C. Cir.
2008); see North Carolina v. EPA, 550 F.3d 1176, 1178 (D.C. Cir. 2008)
(en banc) (deciding to leave the Clean Air Interstate Rule in place until the
EPA could promulgate additional regulations). The D.C. Circuit Court of
Appeals reached this conclusion because the cap-and-trade program in the
Clean Air Interstate Rule would not “assure that upwind states will abate
their unlawful emissions as required by section 110(a)(2)(D)(i)(I).” North
Carolina, 531 F.3d at 906. Essentially, the D.C. Circuit Court of Appeals
agreed that upwind states participating in the regional trading program
45
The environmental groups overlook a vital distinction: Even when a
state does not participate in the 309 program, it must comply with
§ 51.308. Accordingly, approval was not arbitrary or capricious based on
the refusal of 6 states and 211 tribes to participate.
4. Inconsistency and a Lack of Explanation
The environmental groups argue that the EPA changed its position
regarding the “critical mass” of participating states without sufficient
explanation. See Pet’rs’ Opening Br. at 53. We reject this argument.
An unexplained deviation from past practice can render an agency’s
decision arbitrary and capricious, but inconsistency with past practice “is
not a basis for declining to analyze the agency’s interpretation[s].” Nat’l
Cable & Telecomms. Ass’n v. Brand X Internet Servs., 545 U.S. 967, 981
(2005). “[I]f the agency adequately explains the reasons for a reversal of
policy, ‘change is not invalidating.’” Id. (quoting Smiley v. Citibank
(S.D.), N.A., 517 U.S. 735, 742 (1996)).
The EPA never stated, one way or the other, whether a critical mass
of participating states was necessary for the success of a 309 program.
Instead, the EPA explicitly deferred to the judgment of the Western
Regional Air Partnership, which did not make a recommendation on
could trade emissions with other states to avoid the statutory duty to
reduce emissions. Id. Here, however, the environmental groups do not
assert that shifting of emissions between sources would allow the
participating states to avoid their statutory duties.
46
whether to require a minimum number of states. See 67 Fed. Reg. at
30,427; Joint App. at 174-267. Because the EPA did not render an opinion
on the critical-mass requirement, its approval of the 309 program was not
arbitrary and capricious based on an alleged inconsistency with prior
policy.
C. Emissions from the Escalante Coal Plant in New Mexico’s
Implementation Plan
The environmental groups also challenge the EPA’s approval of New
Mexico’s implementation plan in areas beyond the Class I areas subject to
the 309 program. In this challenge, the groups argue that the EPA did not
account for emissions from the state’s second-largest non-BART coal
plant, the Escalante coal plant. 42 U.S.C. § 7491(b)(2)(B); 40 C.F.R.
§ 51.308(d)(1), (3). We reject this argument.
1. Background
The environmental groups’ argument requires examination of the
regulatory and factual setting for New Mexico’s implementation plan.
a. Reasonable Progress Goals in § 51.308(d)(1)
The regulations require states to establish reasonable progress goals
through deciviews that would: (1) improve visibility during the most
impaired days, and (2) ensure no degradation in visibility on the least
impaired days. 40 C.F.R. § 51.308(d)(1). The related analysis involves
two steps.
47
In the first step, states consider four factors:
(1) the cost 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.
Id. § 51.308(d)(1)(i)(A).
In the second step, states determine the rate of required progress by
comparing the baseline visibility conditions to natural visibility conditions
that are expected by 2064. Id. § 51.308(d)(1)(i)(B). In this step, the state
considers:
● what progress is needed to obtain natural visibility conditions
by 2064, and
● what would be needed for the duration of the implementation
plan.
Id.
If the state determines that it cannot reach the uniform rate of
progress, it must demonstrate that a slower rate of progress is reasonable
and that the greater rate of progress is unreasonable. Id.
§ 51.308(d)(1)(ii).
b. New Mexico’s Reasonable Progress Goals
In its plan, New Mexico applied the four-factor analysis and
determined that the uniform rate of progress would not be reasonably
48
achievable. Final Rule, Approval and Promulgation of State
Implementation Plans; New Mexico, 77 Fed. Reg. 70,693, 70,701-02 (Nov.
27, 2012). This determination required the state to demonstrate that its
slower rate of progress would be reasonable under the four-factor analysis
articulated in § 51.308(d)(1)(i)(A). New Mexico complied with this
requirement in part based on the Western Regional Air Partnership’s
analysis.
At New Mexico’s request, the Western Regional Air Partnership
conducted an additional source-specific analysis of three petroleum
refineries in New Mexico. Joint App. at 411. New Mexico used this
source-specific analysis to argue that it could not achieve natural visibility
conditions by 2064. Id. at 564. To defend its less ambitious goal, New
Mexico pointed to natural causes of pollution (such as local wildfires) and
predicted improvement in visibility during the most impaired days and
preservation of existing visibility on the best days. Id. at 563-64.
This reasoning prompted criticism. In response, New Mexico said it
would “examine and consider implementing additional emission reductions
in the [state implementation plan] analysis for 2013.” Id. at 508. As
promised, New Mexico analyzed emissions from additional power plants.
Id.
49
c. The Escalante Coal Plant
Though New Mexico expanded its analysis, it did not examine
emissions at the Escalante Coal Plant. That omission gives rise to the
present challenge.
The Escalante Coal Plant is a 250-megawatt coal-fired power plant
outside of Albuquerque, New Mexico. The environmental groups allege
that the omission proves fatal because this plant “emits thousands of tons
per year of haze-causing nitrogen oxides and is located within 200 miles of
at least 5 of New Mexico’s Class I areas located outside of the Colorado
Plateau.” Pet’rs’ Opening Br. at 57. The EPA counters that the Escalante
plant’s emissions are far lower than the emissions from the only BART
source in New Mexico (the San Juan Generating Station). Resp.’s Br. at 54
n.13.
2. Waiver
The EPA contends that the environmental groups did not exhaust this
allegation because they did not cite § 51.308(d)(1), (3) or urge the need for
analysis of the Escalante plant. Resp.’s Br. at 53-54. We disagree.
In comments to the EPA, the environmental groups asserted:
EPA’s proposal relies on the [Western Regional Air
Partnership’s] general, non-source specific analysis of potential
reasonable progress source categories. See, Docket EPA-R06-
2009-0050-0014, Appendix E. The [Western Regional Air
Partnership’s] general source category analysis fails to identify
any specific New Mexico sources that may be subject to
reasonable progress controls. Id. The [Western Regional Air
50
Partnership’s] general source analysis is also factually
incorrect. Table 6-1 of the [Western Regional Air
Partnership’s] analysis indicates that there is no [particulate
matter, sulfur dioxide, or nitrogen oxide] emissions from coal
fired boilers in New Mexico. Id. at p. 340. To the contrary,
coal fired boilers at SJGS, Escalante coal plant, Raton coal
plant, and Four Corners all emit significant quantities of these
criteria pollutants. Thus, reliance on the [Western Regional
Air Partnership] general source report for approval of the New
Mexico [state implementation plan] is arbitrary and capricious
due to its factual inaccuracy.
Joint App. at 753.
This comment put the EPA on notice of the current argument
regarding the Escalante plant. As the EPA points out, the environmental
groups did not argue that New Mexico was required to analyze the
Escalante plant. But the comment alerted the EPA to the issue. See S.
Coast Air Quality Mgmt. Dist. v. EPA, 472 F.3d 882, 891 (D.C. Cir. 2006)
(“[C]ommenters must be given some leeway in developing their argument
before this court, so long as the comment to the agency was adequate
notification of the general substance of the complaint.”). The commenter
was alleging a need to address emissions from all industrial sources in New
Mexico, including the Escalante plant.
The groups’ failure to cite § 51.308(d)(1), (3) is not fatal. The
environmental groups discussed the reasonable progress goal, and this
discussion provided “‘adequate notification of the general substance of the
complaint.’” Natural Res. Def. Council v. EPA, 571 F.3d 1245, 1259 (D.C.
Cir. 2009) (quoting S. Coast Air Quality Mgmt. Dist., 472 F.3d at 891).
51
3. Consideration of the Escalante Plant
Citing 40 C.F.R. § 51.308(d)(3)(iv), the environmental groups
contend that New Mexico’s reasonable-progress analysis should have
included an analysis of the Escalante plant 14 instead of relying solely on:
(1) the Western Regional Air Partnership’s general, non-source specific
analysis of potential controls for eight source categories, which did not
include the Escalante plant, and (2) source-specific analyses for three New
Mexico petroleum refineries.
The environmental groups contend that New Mexico had “to
undertake a ‘source-specific’ analysis to determine whether to require
measures, such as installation of new air pollution control technology, to
achieve the reasonable progress goals.” Pet’rs’ Opening Br. at 55. In
support, the groups cite § 51.308(d)(1)(i)(A), the subsection outlining the
four factors to be considered in the reasonable-progress analysis. See 40
14
In their opening brief, the environmental groups make a source-
specific argument regarding the Escalante plant. See, e.g., Pet’rs’ Opening
Br. at 54 (“EPA’s approval of the New Mexico [state implementation plan]
also was arbitrary because New Mexico failed to evaluate whether
emissions reductions from the Escalante coal plant were necessary to
achieve reasonable progress.”). In their reply brief, however, the groups
attempt to expand their argument, stating that New Mexico failed to
properly consider all electric generating units (as a category) for non-309
program Class I areas. See, e.g., Pet’rs’ Reply Br. at 56. This source-
category argument was not raised in the environmental groups’ opening
brief; thus, we will not consider this argument. See Silverton Snowmobile
Club v. U.S. Forest Serv., 433 F.3d 772, 783 (10th Cir. 2006) (“[W]e have
held that ‘[t]he failure to raise an issue in an opening brief waives that
issue.’” (quoting Anderson v. U.S. Dep’t of Labor, 422 F.3d 1155, 1174
(10th Cir. 2005))).
52
C.F.R. § 51.308(d)(1)(i)(A). As the EPA points out, however, this
subsection does not require a source-specific analysis.
Rather, the “source-specific” language originates in
§ 51.308(e)(2)(i)(C), which discusses the better-than-BART analysis. See
40 C.F.R. § 51.308(e)(2)(i)(C). This subsection is distinct from the four
factors in § 51.308(d)(1) that govern the determination of reasonable
progress. Accordingly, we reject the environmental groups’ argument that
the EPA had to engage in a source-specific analysis for a reasonable-
progress determination. Nothing in the Regional Haze Rule or the Clean
Air Act required New Mexico to conduct a four-factor analysis of the
Escalante plant.
Two parts of the Regional Haze Rule allowed New Mexico to rely on
the Western Regional Air Partnership’s four-factor analysis. First,
§ 51.308(d)(3)(iii) permits a state conducting a reasonable-progress
determination to “rely[] on technical analyses developed by the regional
planning organization.” 40 C.F.R. § 51.308(d)(3)(iii). Second, § 51.309
allows states to base determinations of reasonable progress “on
assessments conducted by the States and/or a regional planning body.” Id.
§ 51.309(g)(1). Under both subsections, New Mexico could base their
determination of reasonable progress on the Western Regional Air
Partnership’s assessments.
53
Neither the Clean Air Act nor the Regional Haze Rule requires
source-specific analysis in the determination of reasonable progress. Thus,
the EPA’s approval of New Mexico’s plan was not rendered arbitrary or
capricious based on the alleged failure to conduct a four-factor analysis of
the Escalante coal plant.
IV. Conclusion
The EPA did not act arbitrarily or capriciously when it approved the
participants’ implementation plans. Thus, we deny the petitions for
review.
54