Heal Utah, et al. v. EPA, et al., No. 21-9509 (10th Cir. 2023)
Annotate this Case
Justia Opinion Summary
Want to stay in the know about new opinions from the Tenth Circuit US Court of Appeals?
Sign up for free summaries delivered directly to your inbox. Learn More ›
You already receive new opinion summaries from Tenth Circuit US Court of Appeals. Did you know we offer summary newsletters for even more practice areas and jurisdictions? Explore them here.
The issue this case presented for appellate review centered on the air pollution controls on certain coal-fired power plants in Utah that contributed to regional haze. This haze impaired visibility in national parks and wilderness areas across the United States (known as Class I areas). Following Congress’s direction in the Clean Air Act (the CAA or Act) to regulate regional haze, EPA promulgated the Regional Haze Rule to restore natural background visibility conditions in Class I areas by the year 2064. To comply with the CAA’s regional haze requirements, states with Class I areas, or states releasing emissions that may affect visibility in those areas, had to implement the best available retrofit technology (BART) on certain existing sources of air pollution or, alternatively, adopt measures that achieved greater reasonable progress towards improving visibility than BART. The Act required each state to develop a state implementation plan (SIP) for mitigating emissions that contribute to regional haze. The EPA then reviewed the SIP to determine if it satisfied the Act. EPA twice disapproved Utah’s SIPs addressing visibility-impairing emissions at power plants operated by Respondent-Intervenor PacifiCorp. Eventually, EPA approved Utah’s July 2019 revised SIP. In the Final Rule, EPA endorsed Utah’s decision to adopt an alternative measure instead of BART to control for visibility-impairing emissions at the power plants. Petitioners Heal Utah, National Parks Conservation Association, Sierra Club, and Utah Physicians sought review of the Final Rule. According to Petitioners, EPA abused its discretion by approving Utah’s revised SIP because Utah’s alternative measure did not satisfy the CAA’s national visibility goals. They also argued EPA failed to respond to certain comments Petitioners submitted during the rulemaking process. Finding no abuse of discretion, the Tenth Circuit denied the petition for review.
Download PDF
Appellate Case: 21-9509 Document: 010110902437 Date Filed: 08/14/2023 FILED United States Court of Appeals Tenth Circuit PUBLISH August 14, 2023 UNITED STATES COURT OF APPEALS FOR THE TENTH CIRCUIT _________________________________ Christopher M. Wolpert Clerk of Court HEAL UTAH; PARKS CONSERVATION ASSOCIATION; SIERRA CLUB; UTAH PHYSICIANS FOR A HEALTHY ENVIRONMENT, Petitioners, v. Page: 1 No. 21-9509 UNITED STATES ENVIRONMENTAL PROTECTION AGENCY; MICHAEL S. REGAN, Administrator, United States Environmental Protection Agency, Respondents. --------------------------------------DESERET GENERATION & TRANSMISSION CO-OPERATIVE; PACIFICORP; STATE OF UTAH; UTAH ASSOCIATED MUNICIPAL POWER SYSTEMS; UTAH MUNICIPAL POWER AGENCY Intervenors - Respondents _________________________________ PETITION FOR REVIEW OF AN ORDER FROM THE ENVIRONMENTAL PROTECTION AGENCY (EPA No. EPA-R08-2015-0463; FRL-10003-90-Region 8) _________________________________ Appellate Case: 21-9509 Document: 010110902437 Date Filed: 08/14/2023 Page: 2 Jenny K. Harbine of Earthjustice, Bozeman, Montana (John Barth, Hygiene, Colorado with her on the briefs), for Petitioners. Miranda M. Jensen, Attorney, United States Department of Justice, Environment & Natural Resources Division, Washington, D.C. (Todd Kim, Assistant Attorney General, United States Department of Justice, Environment & Natural Resources Division, Washington, D.C.; Daniel P. Schramm and Stephanie J. Talbert, Of Counsel, United States Environmental Protection Agency, with her on the brief), for Respondent. E. Blaine Rawson of Ray Quinney & Nebeker, P.C., Salt Lake City, Utah (Steven G. Jones, Ray Quinney & Nebeker, P.C., Salt Lake City, Utah; Marie Bradshaw Durrant, Assistant General Counsel, PacifiCorp with him on the brief), for Intervenor-Respondent PacifiCorp. Steven J. Christiansen, David C. Reymann, and Erin H. St. John of Parr Brown Gee & Loveless, Salt Lake City, Utah; David Crabtree, General Counsel, Salt Lake City, Utah, on the brief for Intervenor-Respondent Deseret Power. Christian Stephens and Marina Thomas, Assistant Attorneys General, and Stanford E. Purser, Deputy Utah Solicitor General, Office of the Utah Attorney General, Salt Lake City, Utah, on the brief for IntervenorRespondent State of Utah. H. Michael Keller and Artemis D. Vamianakis of Fabian Vancott, Salt Lake City, Utah; Mason Baker, General Counsel, Salt Lake City, Utah, on the brief for Intervenor-Respondent Utah Associated Municipal Power Systems. Alan I. Robbins and Deborah D. Roby of Jennings, Strouss & Salmon, P.L.C., Washington, D.C., for Intervenor-Respondent Utah Municipal Power Agency. _________________________________ Before PHILLIPS, MURPHY, and ROSSMAN, Circuit Judges. _________________________________ ROSSMAN, Circuit Judge. _________________________________ 2 Appellate Case: 21-9509 Document: 010110902437 Date Filed: 08/14/2023 Page: 3 This case concerns air pollution controls on certain coal-fired power plants in Utah that contribute to regional haze. Regional haze is caused when sunlight encounters very small particles in the air. Some light is absorbed by the particles; other light is scattered. This haze impairs visibility in national parks and wilderness areas across the United States (known as Class I areas). Following Congress’s direction in the Clean Air Act (the CAA or Act) to regulate regional haze, EPA promulgated the Regional Haze Rule to restore natural background visibility conditions in Class I areas by the year 2064. The Act empowers the Environmental Protection Agency to promulgate regulations ensuring reasonable progress toward this national goal. To comply with the CAA’s regional haze requirements, states with Class I areas, or states releasing emissions that may affect visibility in those areas, must implement the best available retrofit technology (BART) on certain existing sources of air pollution or, alternatively, adopt measures that achieve greater reasonable progress towards improving visibility than BART. The Act requires each state to develop a state implementation plan (SIP) for mitigating emissions that contribute to regional haze. The EPA then reviews the SIP to determine if it satisfies the Act. EPA twice disapproved Utah’s SIPs addressing visibility-impairing emissions at power plants operated by Respondent-Intervenor PacifiCorp. 3 Appellate Case: 21-9509 Document: 010110902437 Date Filed: 08/14/2023 Page: 4 In 2016, after finding Utah’s SIPs did not satisfy the CAA’s requirements, EPA issued a federal implementation plan (FIP). Utah submitted a revised SIP in July 2019, which was intended to replace the federal plan. Eventually, EPA approved Utah’s July 2019 revised SIP. See 85 Fed. Reg. 75860 (Nov. 27, 2020) (codified at 40 C.F.R. pt. 52) (Final Rule). 1 In the Final Rule, EPA endorsed Utah’s decision to adopt an alternative measure instead of BART to control for visibility-impairing emissions at the power plants. Petitioners Heal Utah, National Parks Conservation Association, Sierra Club, and Utah Physicians now seek review of the Final Rule. According to Petitioners, EPA abused its discretion by approving Utah’s revised SIP because Utah’s alternative measure does not satisfy the CAA’s national visibility goals. They also argue EPA failed to respond to certain comments Petitioners submitted during the rulemaking process. On these grounds, Petitioners ask this court to vacate the Final Rule. Exercising Petitioners refer to EPA’s November 27, 2020, rule as the “Rollback Rule.” See, e.g., Final Opening Br. at 1 (defining the rule at issue in this appeal as the “2020 Final Rule” or “Rollback Rule”). It is undisputed EPA withdrew its 2016 FIP in the Final Rule at issue in this petition. See, e.g., 85 Fed. Reg. 75860. In this opinion, we refer to EPA’s challenged action as the Final Rule instead of adopting Petitioners’ terminology. The Final Rule is included in the record at Deferred Jt. App. at 8-22. 1 4 Appellate Case: 21-9509 Document: 010110902437 Date Filed: 08/14/2023 Page: 5 jurisdiction under 42 U.S.C. § 7607(b)(1), we determine EPA did not abuse its discretion and deny the petition for review. I To situate the issues before us, we first detail the technical statutory and regulatory backdrop of this case, focusing particularly on the CAA’s regional haze program and EPA’s corresponding implementing regulations. We next turn to the procedural history of the matter before us and then discuss the challenged rulemaking. A Under the CAA’s “cooperative-federalism approach,” US Magnesium, LLC v. EPA, 690 F.3d 1157, 1159 (10th Cir. 2012), the federal government develops baseline air quality standards and oversees states’ progress toward attaining those standards. See Oklahoma v. EPA, 723 F.3d 1201, 1204-05 (10th Cir. 2013) (describing the respective roles of the federal government and the states in regulating air quality). The CAA requires states to adopt SIPs, which explain how the state will meet federal air quality requirements. See 42 U.S.C. § 7410(a)(1). EPA reviews SIPs “to ensure that the plans comply with the [CAA].” Oklahoma v. EPA, 723 F.3d at 1204. EPA must reject a SIP that does not satisfy the CAA. See 42 U.S.C. § 7410(c)(1). When that happens, EPA promulgates a FIP—a federal implementation plan designed to ensure a state satisfies the CAA. See id. 5 Appellate Case: 21-9509 Document: 010110902437 Date Filed: 08/14/2023 Page: 6 SIPs can be understood as living documents—a state may, and at times, must—revise an existing SIP to address changes needed to attain or maintain federal air quality standards. See, e.g., 42 U.S.C. § 7410(a)(2)(H). In 1977, Congress amended the CAA, adding Section 169A, to protect visibility in “mandatory class I Federal areas.” 2 Generally, the Act requires states with Class I areas to submit a SIP with “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 Act specifically provides for regulating “major stationary source[s]” of air pollutants. 3 Id. § 7491(b)(2)(A). To comply with the CAA, states must require major stationary sources to “procure, install, and operate, as expeditiously as practicable . . . the best available retrofit technology [or See 42 U.S.C. § 7491(a)(1). The CAA defines “class I” areas to include all “international parks,” “national wilderness areas which exceed 5,000 acres in size,” “national memorial parks which exceed 5,000 acres in size,” and “national parks which exceed [6,000] acres in size,” which were in existence on August 7, 1977. 42 U.S.C. § 7472(a). “[M]andatory class I Federal areas” is defined as “Federal areas which may not be designated as other than class I.” Id. § 7491(g)(5). There are 156 mandatory class I Federal areas in the country. See Regional Haze Regulations, 64 Fed. Reg. 35714, 35714 (July 1, 1999) (codified at 40 C.F.R. §§ 51.300 to 309). We refer to these areas as “Class I” areas. 2 “Major stationary source” refers to fixed (as opposed to mobile) sources of air pollution under 42 U.S.C. § 7491(g)(7), having the potential to emit 250 tons or more of any pollutant. See § 7491(g)(7) (defining “major stationary source” and listing qualifying source categories). 3 6 Appellate Case: 21-9509 Document: 010110902437 Date Filed: 08/14/2023 Page: 7 BART]” to control visibility impairing emissions if the source “emits any air pollutant which may reasonably be anticipated to cause or contribute to any impairment of visibility in any [Class I area].” Id. We call this the CAA’s BART requirement. 4 EPA’s implementing regulations define “BART” 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 C.F.R. § 51.301. The CAA requires states to identify the sources that must comply with the CAA’s BART requirement, based on the type and amount of the source’s emissions, and determine what technology represents BART for each source. See 40 C.F.R. pt. 51, App. Y, § I.A. In 1990, Congress amended the CAA again, adding Section 169B, to address regional haze. See Clean Air Act Amendments, Pub. L. No. 101-549, § 816, 104 Stat. 2399 (Nov. 15, 1990) (codified at 42 U.S.C. § 7492). Regional A major stationary source is considered “BART-eligible” if it: (1) was in existence on August 7, 1977, but had not been in operation for more than 15 years; (2) falls within one of the source categories specified in 42 U.S.C. § 7491(g)(7); and (3) has the potential to emit 250 tons or more of any pollutant. 42 U.S.C. § 7491(b)(2)(A); 40 C.F.R. § 51.301; see also 40 C.F.R. pt. 51, App. Y, § I.C (establishing guidelines for BART determinations under the Regional Haze Rule, or RHR). A BART-eligible source is subject to the CAA’s BART requirement if it “emits any air pollutant which may reasonably be anticipated to cause or contribute to any impairment of visibility in any [Class I area].” 42 U.S.C. § 7491(b)(2)(A). 4 7 Appellate Case: 21-9509 Document: 010110902437 Date Filed: 08/14/2023 Page: 8 haze is “visibility impairment that is caused by the emission of air pollutants from numerous anthropogenic sources located over a wide geographic area.” 40 C.F.R. § 51.301. EPA promulgated the Regional Haze Rule (RHR) in 1999. See Regional Haze Regulations, 64 Fed. Reg. at 35714. The RHR mandates states to “develop programs to assure reasonable progress toward” the CAA’s national visibility goal and declares its own goal of restoring visibility in Class I areas to natural conditions by 2064. 40 C.F.R. §§ 51.300(a), 308(d). According to the RHR, states were required, by 2007, to submit SIPs addressing regional haze visibility impairment. 5 These regional haze SIPs needed to identify the sources of pollution within the state that were subject to the BART requirement and to detail the state’s plan to satisfy the CAA. See 40 C.F.R. § 51.308(d). The RHR provides at least two paths for compliance with the CAA’s BART requirement. As relevant here, a state could choose to implement the best available retrofit technology on its BART-eligible sources, as provided in 42 U.S.C. § 7491(b)(2)(A). See 40 C.F.R. § 51.308(e) (directing a state to These SIPs covered the first implementation period of the regional haze program. See 40 C.F.R. § 51.308(b). The RHR also requires states to submit periodic comprehensive revisions to their regional haze SIPs. See id. § 51.308(f). States had to submit revised regional haze SIPs by July 31, 2021, to comply with the RHR’s second implementation period. See id. This case only involves the requirements of the first implementation period. 5 8 Appellate Case: 21-9509 Document: 010110902437 Date Filed: 08/14/2023 Page: 9 “submit an implementation plan . . . representing BART . . . for each BART-eligible source”). Or, a state could select an “alternative measure” to BART—but only if the proposed alternative is shown to achieve “greater reasonable progress [toward natural visibility conditions] than would be achieved through the installation and operation of BART.” Id. § 51.308(e)(2). In other words, to improve visibility under the RHR, a state’s SIP must require the installation and operation of BART controls on sources subject to the CAA’s BART requirement or propose an alternative that produces better-than-BART results in the relevant Class I areas. See id. § 51.308(e). Whether an alternative measure achieves greater reasonable progress can be demonstrated by the “clear weight of evidence,” § 51.308(e)(2)(i)(E), or—as here—the results of air quality dispersion modeling, § 51.308(e)(3). 6 Dispersion modeling is a technique used to estimate the concentration of certain pollutants surrounding a source of emissions. It uses mathematical formulations to simulate how air pollutants scatter in the ambient atmosphere and to predict the resulting impact on visibility. See The RHR’s “clear weight of evidence” option “attempt[s] 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.” See Regional Haze Regulations Revisions, 71 Fed. Reg. 60612-01, 60622 (Oct. 13, 2006). EPA has provided a non-exhaustive list of factors relevant to that analysis. See id. 6 9 Appellate Case: 21-9509 Document: 010110902437 Date Filed: 08/14/2023 Page: 10 id.; see also 40 C.F.R. pt. 51, App. Y, § IV.D.5 (providing guidelines for performing dispersion modeling). To assess compliance with the RHR, dispersion modeling focuses on the differences in visibility improvement achieved by the proposed alternative measure compared to BART for each impacted Class I area. See 40 C.F.R. § 51.308(e)(3). A state’s proposed alternative measure demonstrates greater reasonable progress if dispersion modeling shows (1) Visibility does not decline in any Class I area compared to the baseline, and (2) There is an overall improvement in visibility across all affected Class I areas compared to BART. See id. § 51.308(e)(3)(i)-(ii). Visibility improvements are measured based on the results of the worst and best 20% of days in each modeled scenario. Id. § 51.308(e)(3). Here, it is undisputed Utah has correctly identified its subject-to-BART sources—four electric generating units at two coal-fired power plants operated by PacifiCorp: Hunter Units 1 and 2 and Huntington Units 1 and 2 (the Plants). 7 The Plants cause or contribute to visibility The Hunter power plant is located in Castle Dale, Utah and consists of three electric utility steam generating units; only Units 1 and 2 are subject to BART. See Proposed Action on Utah’s June 2015 Submission, 81 Fed. Reg. 2004, 2013 (Jan. 14, 2016) (discussing the Plants). The Huntington power plant is located in Huntington City, Utah, and consists of two electric utility steam generating units; both are subject to BART. Id. 7 10 Appellate Case: 21-9509 Document: 010110902437 Date Filed: 08/14/2023 Page: 11 impairment in Class I areas—see Approval, Disapproval and Promulgation of Air Quality Plan, 81 Fed. Reg. 43894 (July 5, 2016)—and therefore fall within the regulatory purview of the RHR, see 40 C.F.R. § 51.308(d). This case also involves additional sources of air pollution operating in Utah at the time relevant to this petition: Carbon Units 1 and 2 (Carbon or Carbon plant) and Hunter Unit 3. As we will discuss, Utah’s proposed alternative measure relied in part on emissions reductions at these other sources, including the decrease in visibility-impairing emissions that resulted when PacifiCorp permanently closed the Carbon plant in August 2015. B Over the past two decades, Utah submitted a number of SIPs in its effort to satisfy the RHR’s requirements. The petition before us challenges EPA’s Final Rule approving Utah’s revised SIP, submitted on July 3, 2019, and supplemented on December 3, 2019 (the July 2019 Revised SIP). We begin by briefly discussing one of Utah’s previous regional haze SIPs, and EPA’s treatment of that submission, because it involves issues relevant to the petition now before us. We then discuss Utah’s July 2019 Revised SIP, EPA’s approval of that SIP, and Petitioners’ challenges to EPA’s approval. The Plants each use tangentially fired pulverized coal boilers which burn bituminous coal from the Deer Creek Mine in Utah. Id. 11 Appellate Case: 21-9509 Document: 010110902437 Date Filed: 08/14/2023 Page: 12 On June 4, 2015, Utah submitted a revised regional haze SIP to EPA. See Approval, Disapproval and Promulgation of Air Quality Plan, 81 Fed. Reg. at 43895-96. Utah’s June 2015 submission asked EPA to approve a measure alternative to installing and operating BART controls for nitrogen oxide (NOx) emissions at the Plants. See id. at 43894. Utah’s proposed alternative measure pointed to a combination of NOx emissions reductions at the Plants and Hunter Unit 3, which Utah maintained could be achieved through combustion control upgrades already installed at those units between 2006 and 2014. See Proposed Rule, 85 Fed. Reg. 3558, 3559, 3563. Utah also proposed to take credit for reductions in emissions resulting from the Carbon power plant’s permanent closure in August 2015. See id. at 3559. Utah maintained its proposed alternative measure satisfied the RHR’s “greater reasonable progress” standard based on the “clear weight of evidence” under 40 C.F.R. § 51.308(e)(2)(i)(E). See id. at 3563. On January 14, 2016, EPA published a proposed rule in response to Utah’s June 2015 revised SIP. See Proposed Action on Utah’s June 2015 Submission, 81 Fed. Reg. 2004 (Jan. 14, 2016). There, EPA proposed and solicited comments on two different courses of action—deemed “co-proposals.” Id. at 2006. One proposal reflected EPA’s approval of Utah’s June 2015 submission in its entirety. Id. This meant EPA would promulgate 12 Appellate Case: 21-9509 Document: 010110902437 Date Filed: 08/14/2023 Page: 13 a final rule finding Utah’s proposed alternative measure achieved greater reasonable progress based on the “clear weight of evidence.” See id. The other proposal reflected EPA’s partial approval of Utah’s submission; for the disapproved elements, EPA would promulgate a FIP mandating the installation of BART controls for NOx emissions at the Plants. Id. at 2007. After holding a public hearing and considering comments, EPA issued a final rule adopting the second co-proposal—partially approving and partially disapproving Utah’s June 2015 submission and implementing a FIP for NOx emissions at the Plants. 8 According to EPA, it was a “close call” whether Utah’s proposed alternative measure achieved “greater reasonable progress” than BART toward achieving natural visibility conditions. 9 See Approval, Disapproval and Promulgation of Air Quality Plan, 81 Fed. Reg. at 43894-95. The FIP included (1) EPA’s determination of what qualified as BART for NOx emissions at the Plants; (2) corresponding NOx emissions limits for each of the Plants’ units—a 30-day rolling average of 0.07 lb/MMbtu—which “reflect[ed] installation and operation of [selective-catalytic reduction] plus the existing upgraded combustion controls”; and (3) monitoring, recordkeeping, and reporting requirements to the extent needed to implement the FIP’s requirements. Proposed Rule, 85 Fed. Reg. at 3563 (describing EPA’s 2016 FIP). 8 EPA explained that while “some of the metrics [Utah] included in its weight-of-evidence analysis . . . appear to support a conclusion that the BART Alternative achieves greater reasonable progress than BART[,] . . . several other metrics in [Utah’s] analyses did not appear to support a conclusion that the BART Alternative achieves greater reasonable progress.” Approval, Disapproval and Promulgation of Air Quality Plan, 81 Fed. Reg. at 43895. 9 13 Appellate Case: 21-9509 Document: 010110902437 Date Filed: 08/14/2023 Page: 14 Approval, Disapproval and Promulgation of Air Quality Plan, 81 Fed. Reg. at 43895-96. Ultimately, EPA rejected Utah’s proposed alternative measure, finding Utah had not demonstrated compliance with federal requirements. Id. at 43896. According to EPA, however, Utah “retain[ed] its authority to submit a revised state plan consistent with CAA and Regional Haze Rule (RHR) requirements. An approvable SIP submission will result in the modification or withdrawal of the FIP.” Id. at 43894. 10 On July 3, 2019, Utah submitted a revised SIP proposing an alternative to EPA’s 2016 FIP. 11 The alternative measure proposed was identical in substance to what Utah had advanced in its June 2015 revised SIP—namely, reductions in visibility-impairing NOx emissions through previously-upgraded combustion controls at the Plants and Hunter Unit 3 and the permanent closure of the Carbon plant. The difference was how Utah sought to demonstrate that the alternative measure satisfied the Utah, PacifiCorp, and others petitioned this court for review of EPA’s July 5, 2016, final rule. See Utah v. EPA, Case Nos. 16-9541, 16-9542, 16-9543 & 16-1945 (10th Cir. Mar. 17, 2017). That case was later dismissed after EPA promulgated the Final Rule at issue here. See Utah v. EPA, Case No. 16-9541 (10th Cir. Jan. 11, 2021). 10 On December 3, 2019, Utah supplemented its July 3, 2019, SIP, which included amendments to monitoring, record keeping, and reporting. See, e.g., Proposed Rule, 85 Fed. Reg. at 3564. The matters addressed in Utah’s December 2019 supplement are not before us. 11 14 Appellate Case: 21-9509 Document: 010110902437 Date Filed: 08/14/2023 Page: 15 RHR’s “greater reasonable progress” standard. See Proposed Rule, 85 Fed. Reg. 3558, 3563-64 (Jan. 22, 2020) (Proposed Rule). Before, Utah had relied on the “clear weight of evidence” under § 51.308(e)(2)(i)(E). But this time, Utah pointed to the results of dispersion modeling under § 51.308(e)(3). The petition before us challenges the dispersion modeling Utah used in its July 2019 Revised SIP, so we describe aspects of that modeling in some detail. The dispersion modeling at issue compared projected visibility improvements from implementing the BART controls required by EPA’s 2016 FIP with visibility improvements predicted under Utah’s proposed alternative measure. See Proposed Rule, 85 Fed. Reg. at 3564-66. A contractor for PacifiCorp performed the dispersion modeling with EPA’s guidance. Id. at 3566-68. The dispersion modeling used the Comprehensive Air Quality Model with Extensions (CAMx), which “simulates air quality over many geographic scales and treats a wide variety of inert and chemically active pollutants.” 12 Id. at 3566. The CAMx model used data developed by the Western Air Quality Study. Id. at 3567. Following EPA’s guidance, PacifiCorp made some adjustments to the starting-point data and While Petitioners challenge the CAMx dispersion modeling EPA relied on in this case (as we discuss in detail later), Petitioners do not challenge EPA’s use of CAMx modeling for analyzing visibility improvements generally. 12 15 Appellate Case: 21-9509 Document: 010110902437 Date Filed: 08/14/2023 Page: 16 “performed a new base case model simulation and performance evaluation.” Id. at 3568. EPA observed PacifiCorp’s changes “resulted in substantial improvements in model performance” that were “expected to provide more accurate predictions of the visibility benefits of changes in NOx emissions.” See id. PacifiCorp then used the CAMx model to forecast three scenarios key to this petition: (1) a 2025 BART Baseline—what would happen if Utah did not implement its proposed alternative measure or the BART controls required by EPA’s 2016 FIP; (2) a 2025 BART Benchmark—what would happen if Utah implemented BART controls at the Plants, as required by the 2016 FIP; and (3) a 2025 BART Alternative—what would happen if Utah did not implement the BART controls required by the 2016 FIP and instead used its proposed alternative measure. We briefly explain these three scenarios. The BART Baseline represented projected emissions from the Plants, Carbon Units 1 and 2, and Hunter Unit 3, and the resulting impact on visibility that would occur in 2025 without the FIP-required controls or any alternative measures. See id.; Final Rule, 85 Fed. Reg. at 75870. This scenario’s projections used historic data for each source’s emissions from 2001-2003. 16 Appellate Case: 21-9509 Document: 010110902437 Date Filed: 08/14/2023 Page: 17 The BART Benchmark scenario represented projected emissions and the resulting impact on visibility that would occur in 2025 based on applying BART controls at the Plants, as required in the 2016 FIP. See Proposed Rule, 85 Fed. Reg. at 3568; Final Rule, 85 Fed. Reg. at 75870. This scenario started with the BART Baseline data, then simulated the impact of the BART controls on visibility benefits. See Proposed Rule, 85 Fed. Reg. at 3568; Final Rule, 85 Fed. Reg. at 75870. The BART Alternative scenario represented projected emissions and the resulting future impact on visibility in 2025 based on Utah’s proposed alternative measure. See Proposed Rule, 85 Fed. Reg. at 3568. All other inputs from the BART Baseline scenario remained unchanged to control for the impact Utah’s proposed alternative would have on 2025 projections. See id. For each scenario, the CAMx model projected visibility in 2025 on the 20% best and 20% worst visibility days, as required under the RHR. See id. at 3569; see also § 51.308(e)(3). According to Utah, the dispersion modeling demonstrated its BART Alternative achieved an overall improvement in visibility on these days compared to the BART Benchmark. See Proposed Rule, 85 Fed. Reg. at 3569. On January 22, 2020, EPA proposed approving Utah’s July 2019 Revised SIP and withdrawing its 2016 FIP. See id. at 3558. EPA explained 17 Appellate Case: 21-9509 Document: 010110902437 Date Filed: 08/14/2023 Page: 18 the dispersion modeling showed the BART Alternative, when compared to the BART Benchmark, achieved an overall improvement in visibility in the relevant Class I areas. The improvement was measured in deciviews, which is “the unit of measurement . . . for quantifying in a standard manner human perceptions of visibility.” 40 C.F.R. § 51.301. The dispersion modeling predicted the BART Alternative would achieve an overall improvement in visibility compared to the BART Benchmark of 0.00494 deciviews on the 20% best days and 0.00058 deciviews on the 20% worst days. See Proposed Rule, 85 Fed. Reg. at 3569. A public comment period and a public hearing followed EPA’s publication of the Proposed Rule. 13 Id. at 3558. Among other concerns, Petitioners challenged EPA’s reliance on the dispersion modeling used in the July 2019 Revised SIP. Petitioners argued Utah’s dispersion modeling was “technically and legally flawed” and could not “be relied upon to justify the proposed BART Alternative as providing for greater reasonable In the Final Rule, EPA explained PacifiCorp, HEAL Utah, Sierra Club, National Parks Conservation Association, Utah Physicians for a Healthy Environment, Natural Resources Defense Council, Edison Electric Institute, Ute Mountain Ute Tribe, and Salt Lake City’s Capitol Hill Action Group submitted “detailed written comments” on the Proposed Rule. Final Rule, 85 Fed. Reg. at 75861. EPA responded to these comments in the Final Rule and also prepared an accompanying Response to Comments (RTC) document that “provides detailed responses to all significant comments received.” Id. 13 18 Appellate Case: 21-9509 Document: 010110902437 Date Filed: 08/14/2023 Page: 19 progress” than the BART Benchmark. Deferred Jt. App. at 858. In support, Petitioners pointed to a report by Howard Gebhart, an air quality modeling expert. See id. at 848-49, 858. Mr. Gebhart’s report was included as an exhibit to Petitioners’ comments on the Proposed Rule. See id. at 849 n.21 (including report as Exhibit 1), 870-93 (Exhibit 1). The adequacy of EPA’s response to these comments is contested, as we will discuss. On November 27, 2020, EPA issued the Final Rule. See Final Rule, 85 Fed. Reg. at 75860. Relying on the results demonstrated by dispersion modeling, EPA found Utah’s BART Alternative achieved greater reasonable progress toward improving overall visibility than the BART Benchmark. Id. Upon its approval of the July 2019 Revised SIP, EPA withdrew the 2016 FIP. Id. C On January 19, 2021, Petitioners timely filed the instant petition seeking review of the Final Rule. 14 On February 4, 2021, this court granted EPA’s request to hold the matter in abeyance for 120 days, “to provide an opportunity for new Agency leadership to review the underlying rule in conformance with the Respondents do not challenge Petitioners’ standing or the timeliness of their petition. 14 19 Appellate Case: 21-9509 Document: 010110902437 Date Filed: 08/14/2023 Page: 20 President’s recent Executive Order on ‘Protecting Public Health and the Environment and Restoring Science to Tackle the Climate Crisis,’” published at 86 Fed. Reg. 7037 (Jan. 20, 2021). EPA then filed several motions to extend the initial abeyance period. On November 29, 2021, the parties jointly asked to lift the abeyance. This court granted the parties’ request and permitted several entities to intervene. 15 II On appeal, Petitioners advance several arguments challenging the Final Rule. Petitioners’ two primary arguments take issue with the dispersion modeling underlying Utah’s proposed alternative measure and contest the EPA’s reliance on it. Petitioners also argue EPA abused its discretion by failing to respond to comments identifying problems with the dispersion modeling’s results. We begin by discussing the applicable legal standards that guide our analysis. Then, we turn to the parties’ arguments. We conclude EPA did not abuse its discretion in promulgating the Final Rule and deny the petition for review. The State of Utah, PacifiCorp, Deseret Generation & Transmission Co-Operative, Utah Associated Municipal Power Systems, and Utah Municipal Power Agency sought to intervene in support of Respondent EPA under Rule 15(d) of the Federal Rules of Appellate Procedure. Rule 15(d) permits parties to intervene in proceedings seeking review or enforcement of an agency order. In this opinion, we refer to Respondent EPA and Respondent-Intervenors collectively as Respondents. 15 20 Appellate Case: 21-9509 Document: 010110902437 Date Filed: 08/14/2023 Page: 21 A The CAA authorizes judicial review of EPA’s approval of state implementation plans, but it does not designate the applicable standard of review. 42 U.S.C. § 7607(b)(1). However, “we are bound by the Administrative Procedure Act” when reviewing EPA’s approval of state implementation plans. WildEarth Guardians v. EPA, 770 F.3d 919, 927 (10th Cir. 2014) (citing Oklahoma v. EPA, 723 F.3d at 1211). The Administrative Procedure Act requires a reviewing court to “hold unlawful and set aside agency action” found to be “arbitrary, capricious, an abuse of discretion, or otherwise not in accordance with law.” 5 U.S.C. § 706(2)(A). We will not disturb 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 v. State Farm Mut. Auto. Ins. Co., 463 U.S. 29, 43 (1983); accord WildEarth Guardians, 770 F.3d at 927. We afford an agency particular deference where it acts under an “unwieldy and science-driven statutory scheme[] like the Clean Air Act.” Nat’l Ass’n of Clean Air Agencies v. EPA, 489 F.3d 1221, 1220 (D.C. Cir. 2007) (quoting Bluewater Network v. EPA, 372 F.3d 404, 410 (D.C. Cir. 21 Appellate Case: 21-9509 Document: 010110902437 Date Filed: 08/14/2023 Page: 22 2004)). And our deference to the agency is “especially strong” when “the challenged decisions involve technical or scientific matters within the agency’s area of expertise.” Morris v. U.S. Nuclear Regul. Comm’n, 598 F.3d 677, 691 (10th Cir. 2010) (citation omitted). Of course, deference to agency action is not unlimited. See Defs of Wildlife v. Babbitt, 958 F. Supp. 670, 679 (D.D.C. 1997) (explaining deference to an agency’s scientific or technical expertise “is not unlimited” and “the presumption . . . may be rebutted if its decisions, even though based on scientific expertise, are not reasoned”). B Petitioners seek to vacate the Final Rule because, in their view, the EPA abused its discretion in approving a SIP that did not satisfy the requirements of the CAA and RHR. The BART Benchmark used in the dispersion modeling was flawed, Petitioners contend, because it arbitrarily assumed the Carbon plant would continue producing sulfur dioxide (SO2) emissions at historical levels. According to Petitioners, the BART Benchmark scenario failed to reflect the reduction in Carbon’s SO2 emissions from the plant’s presumptive compliance with a separate EPA requirement—the Mercury and Air Toxics Standards (MATS) rule. We discern no abuse of discretion. Recall, the BART Benchmark and BART Alternative scenarios each started with the BART Baseline’s emissions and visibility projections for 22 Appellate Case: 21-9509 Document: 010110902437 Date Filed: 08/14/2023 Page: 23 2025. Each scenario then applied changes to the BART Baseline based on the policy choice the respective scenario sought to model. The BART Benchmark modeled the impact of EPA’s 2016 FIP-imposed BART controls on the Plants. The BART Benchmark “assumed that . . . the Carbon plant . . . would emit pollutants consistent with the 2001-2003 baseline.” Final Rule, 85 Fed. Reg. at 75861. The BART Alternative modeled the impact of Utah’s proposed alternative measure, which included taking credit for emissions reductions from Carbon’s permanent closure in 2015. On appeal, Petitioners argue the BART Benchmark scenario impermissibly assumed more emissions than actually would have been permitted by law. According to Petitioners, the BART Benchmark incorrectly relied on the assumption that Carbon would continue emitting SO2 at historic levels. Including Carbon’s past SO2 emissions in the BART Benchmark was unreasonable, Petitioners argue, because it distorted the regulatory reality—that is, if Carbon had not been shut down in 2015, the plant would have needed to reduce emissions to comply with EPA’s MATS rule. The BART Benchmark should have reflected Carbon’s presumptive compliance with MATS. Without that, Petitioners contend, “EPA failed to set up a comparison that would allow a legitimate determination that the BART Alternative would actually achieve ‘an overall improvement in visibility’ than BART, as required by the Act.” Final Opening Br. at 38; see 23 Appellate Case: 21-9509 Document: 010110902437 Date Filed: 08/14/2023 Page: 24 also Oral Arg. at 10:21-37 (contending the BART Benchmark set up an “impossible future emissions scenario for the Carbon coal plant that [didn’t] allow for a meaningful comparison” with the BART Alternative). 16 Petitioners’ argument seems reasonable at first blush. Why shouldn’t the visibility improvements predicted by the better-than-BART alternative be considered against a world where compliance with MATS is already mandated? But applicable law and the standards that guide our review require us to reject the argument. We conclude the EPA did not abuse its discretion for at least three reasons. First, the plain language of the RHR contradicts Petitioners’ argument that the BART Benchmark scenario should have accounted for Carbon’s presumptive compliance with MATS. The RHR details how a state must show if a proposed alternative measure achieves greater reasonable progress than source-specific BART controls. See 40 C.F.R. § 51.308(e)(2). Especially relevant here is § 51.308(e)(2)(i)(C), which says a state considering the BART benchmark side of the comparison must analyze “the best system of continuous emission control technology available and According to Respondents, Petitioners seek to challenge EPA’s BART alternative regulations themselves—namely, §§ 51.308(e)(2)(iv) and (3)—a claim that would be time-barred under the CAA. See EPA Response Br. at 42-45 (citing 42 U.S.C. § 7607(b)(1)). We are not persuaded. We understand Petitioners’ argument as challenging EPA’s application of the RHR, not the validity of the RHR or any of its individual provisions. 16 24 Appellate Case: 21-9509 Document: 010110902437 Date Filed: 08/14/2023 Page: 25 associated emission reductions achievable for each source . . . subject to BART and covered by the alternative program.” (Emphasis added.) The BART benchmark scenario may only include emissions reductions associated with BART controls on sources subject to BART. See id. It is undisputed Carbon is not a source subject to BART. See Final Reply Br. at 6 (identifying Carbon as “non-BART source[]”); EPA Response Br. at 12 (same). In the Final Rule, EPA directly responded to comments on this issue. See Final Rule, 85 Fed. Reg. at 75861-62. EPA explained that “the Carbon Units are not BART sources. Accordingly, reductions from [the Carbon plant] should not be included in determining emissions reductions from the BART Benchmark under 40 C.F.R. § 51.308(e)(2)(i)(C).” Id. at 75862 (footnote omitted). As EPA correctly recognized, analyzing the BART Benchmark scenario in the way Petitioners advocate would run afoul of the RHR. See id. at 75861. Second, EPA offered an additional, reasonable explanation for using Carbon’s historical SO2 emissions in the BART Benchmark. In the Final Rule, EPA explained that, because Carbon’s emissions reductions were fully creditable in the BART Alternative scenario, it needed to assume Carbon’s continued emissions in the BART Benchmark to properly compare the relative improvements between those two modeled scenarios. See Final 25 Appellate Case: 21-9509 Document: 010110902437 Date Filed: 08/14/2023 Page: 26 Rule, 85 Fed. Reg. at 75861. EPA stated, “[a]ssuming continued emissions from sources that would not be subject to BART controls in the BART Benchmark scenario, when such emissions would be eliminated under the BART Alternative, is simply a necessary analytical step for making a proper comparison” under 40 C.F.R. § 51.308(e). Final Rule, 85 Fed. Reg. at 75861. According to EPA, it was necessary to “compare emissions reductions under each scenario from the same baseline year” to appropriately “determine if the Alternative achieves greater reasonable progress.” Id. at 75862. EPA’s methodological approach to the RHR’s requirements “involve[s] technical or scientific matters within the agency’s area of expertise” to which we afford “especially strong” deference. Morris, 598 F.3d at 691 (citation omitted). And under the circumstances, we see no basis to conclude that, in deciding how Carbon would be treated in the BART Benchmark scenario, the EPA failed to “consider[] the relevant data and rationally explain[] its decision.” WildEarth Guardians, 770 F.3d at 927; see State Farm, 463 U.S. at 43 (explaining an agency has not acted arbitrarily or capriciously when it gives a “satisfactory explanation for its action including a ‘rational connection between the facts found and the choice made’”) (citation omitted). Third, we agree with Respondents that persuasive authority from the D.C. Circuit supports EPA’s decision not to reduce Carbon’s emissions in 26 Appellate Case: 21-9509 Document: 010110902437 Date Filed: 08/14/2023 Page: 27 the BART Benchmark scenario. In Utility Air Regulatory Group v. EPA (UARG II), the D.C. Circuit considered and rejected an argument similar to the one Petitioners advance here. 885 F.3d 714 (D.C. Cir. 2018). Petitioners in UARG II challenged EPA’s determination that states could use emissions reductions arising from compliance with the 2011 Cross-State Air Pollution Rule (CSAPR) as a BART alternative measure under the RHR. Id. at 717. In the challenged rule, EPA determined “CSAPR’s requirements were stringent and effective enough for it to serve as a better-than-BART alternative for states participating in CSAPR, thus excusing states from compliance with BART itself.” Id. Challenging EPA’s methodology, petitioners argued the agency should not have compared BART on its own (without CSAPR in place) to the BART alternative on its own (CSAPR without BART in place). See id. at 720. Petitioners reasoned EPA should have conducted its analysis differently because CSAPR was implemented under a provision of the CAA separate from the BART requirement and states would have had to comply with CSAPR anyway. See id. In rejecting petitioners’ argument, the D.C. Circuit pointed to its earlier decision where it held “an emissions control program in place to satisfy an unrelated statutory provision is not disqualified from serving as a better-than-BART alternative.” Id. (referencing Util. Air Regul. Grp. v. EPA (UARG I), 471 F.3d 1333, 1341 (D.C. Cir. 2006)). The court explained 27 Appellate Case: 21-9509 Document: 010110902437 Date Filed: 08/14/2023 Page: 28 that in UARG I, it “affirmed EPA’s comparison between BART-without-[the Clean Air Interstate Rule (CAIR) 17] and CAIR-without-BART to determine the adequacy of CAIR as a BART alternative,” based on § 51.308(e)(3) of the RHR. See id. at 720-21. The court therefore rejected petitioners’ argument that “the status quo for a better-than-BART alternative to improve must be a world that already includes CSAPR in operation.” Id. at 720. Petitioners in our case advance substantially the same argument the D.C. Circuit rejected in UARG II. That is, the BART Alternative needed to be compared against a scenario that included the Carbon plant’s presumptive compliance with the MATS rule—just like CSAPR in UARG II. The UARG cases provide persuasive authority supporting our determination EPA acted reasonably by comparing the BART Benchmark (with BART measures) with the BART Alternative (without BART measures) under the RHR. Stated differently, EPA reasonably compared the BART Benchmark—without considering Carbon’s compliance with MATS—to the BART Alternative, which appropriately included emissions reductions associated with Carbon, a non-BART source. Petitioners have advanced no compelling contrary argument. CAIR was a predecessor to CSAPR, which became defunct after CSAPR’s promulgation. See UARG II, 885 F.3d at 717. 17 28 Appellate Case: 21-9509 Document: 010110902437 Date Filed: 08/14/2023 Page: 29 C We now turn to Petitioners’ argument that Utah’s BART Alternative did not actually achieve an overall improvement in visibility under 40 C.F.R. § 51.308(e)(3)(ii). As relevant here, EPA determined Utah satisfied the RHR’s greater reasonable progress standard because the dispersion modeling showed the BART Alternative achieved “an overall average improvement over the BART benchmark of 0.00494 deciviews across all Class I areas on the 20 percent best days and 0.00058 deciviews on the 20 percent worst days.” Final Rule, 85 Fed. Reg. at 75871. Petitioners argue the Final Rule should be vacated because EPA “irrationally relied on differences between the BART Alternative and BART Benchmark that are so minute, they are effectively zero.” Final Opening Br. at 27. In support, Petitioners point to a report by air-quality-modeling expert Howard Gebhart. 18 See Final Opening Br. at 27. According to Petitioners, the “administrative record does not contain any evidence that the 0.00058 deciview[s] difference between the BART Alternative and BART Benchmark will result in ‘an overall improvement in visibility’ in the real world.” Id. at 28. Petitioners also contend “uncertainties, both inherent in Mr. Gebhart’s report, dated May 15, 2019, states that it “contains technical comments concerning the State of Utah Draft State Implementation Plan (SIP) Amendments proposed for Regional Haze.” Deferred Jt. App. at 871. 18 29 Appellate Case: 21-9509 Document: 010110902437 Date Filed: 08/14/2023 Page: 30 the modeling exercise itself and in the veracity of the underlying emissions input assumptions render EPA’s reliance on such inconsequential differences in the modeling results arbitrary.” Id. at 28. We acknowledge the important concern identified by Petitioners. The Final Rule relied on very small differences in modeled visibility improvements between the BART Benchmark and BART Alternative scenarios—differences of less than one thousandth of a deciview. However, we must conclude EPA did not abuse its discretion by determining, for purposes of compliance with the RHR, Utah’s BART Alternative achieved greater reasonable progress than the BART Benchmark. A BART alternative measure satisfies the RHR’s “greater reasonable progress” standard if dispersion modeling shows, as relevant here, the alternative achieves an “overall improvement in visibility, determined by comparing the average differences between BART and the alternative over all affected Class I areas.” 40 C.F.R. § 51.308(e)(3)(ii). According to the RHR, this showing must be based on comparing visibility impacts resulting from implementing BART controls on BART-eligible sources with those resulting from the proposed alternative for each impacted Class I area for the worst and best 20 percent of days. See id. § 51.308(e)(3). According to Respondents, the RHR does not establish a minimum threshold of improvement necessary to satisfy § 51.308(e)(3)’s two-prong 30 Appellate Case: 21-9509 Document: 010110902437 Date Filed: 08/14/2023 Page: 31 “greater reasonable progress” standard and, thus, EPA’s approval of Utah’s July 2019 Revised SIP aligns with the RHR’s plain language. We agree. We see nothing in the RHR that sets a minimum threshold for modeled visibility improvement to constitute an “overall improvement in visibility” under § 51.308(e)(3)(ii). And Petitioners point to nothing in the CAA or EPA’s regulatory framework that suggests otherwise. 19 Based on the RHR’s plain language and the standards that guide our review, we conclude Utah’s dispersion modeling demonstrated the BART Alternative satisfied the RHR’s requirements—it achieved an overall improvement in visibility compared to the BART Benchmark, based on the average visibility differences across all affected Class I areas, as required by § 51.308(e)(3). Accordingly, we perceive no basis to conclude EPA abused its discretion by relying on a small but measurable overall improvement in visibility when considering whether Utah’s BART Alternative satisfied the RHR. We note Petitioners appear to concede EPA may consider small relative visibility benefits when evaluating competing emissions controls under the RHR. See Final Opening Br. at 28 (stating “EPA may consider relative visibility benefits that are less than the level of perceptibility when evaluating air pollution controls under the regional haze program”). But Petitioners maintain the relative differences in visibility shown by the dispersion modeling in this case were not just small or below the level of perceptibility, they were effectively zero. 19 31 Appellate Case: 21-9509 Document: 010110902437 Date Filed: 08/14/2023 Page: 32 D Finally, we address Petitioners’ argument that EPA failed to meaningfully respond to its comments on the Proposed Rule challenging Utah’s dispersion modeling. Specifically, Petitioners point to Mr. Gebhart’s comments about uncertainties in the dispersion modeling itself, and in the veracity of the modeling’s emissions assumptions, which render EPA’s reliance on small differences in visibility improvement arbitrary. 20 According to Petitioners, “EPA did not respond to the heart of Mr. Gebhart’s concern,” which was that “the asserted relative benefit of the BART Alternative—just 0.00058 deciview[s]—is ‘essentially zero.’” Final Opening Br. at 30. By “nonetheless determining that the BART Alternative would achieve greater ‘overall improvement in visibility’ than BART,” Petitioners argue, “EPA overlooked contrary record evidence and failed to consider an important aspect of the problem.” Id. (citing N.M. Env’t Improvement Div. v. Thomas, 789 F.2d 825, 830 (10th Cir. 1986)). Respondents contend EPA comprehensively responded to comments challenging the dispersion modeling Utah submitted in support of its 2019 This argument relies on several pages of Mr. Gebhart’s May 15, 2019, report, which—as described—Petitioners attached as an exhibit to their comments on the Proposed Rule. See Deferred Jt. App. at 871-93. 20 32 Appellate Case: 21-9509 Document: 010110902437 Date Filed: 08/14/2023 Page: 33 Revised SIP and EPA adequately explained why it relied on the dispersion modeling’s results. We agree with Respondents. “An agency must consider and respond to significant comments received during the period for public comment.” Perez v. Mortg. Bankers Ass’n, 575 U.S. 92, 96 (2015). “An agency’s failure to respond to relevant and significant public comments generally ‘demonstrates that the agency’s decision was not based on a consideration of the relevant factors.’” Lilliputian Sys., Inc. v. Pipeline & Hazardous Materials Safety Admin., 741 F.3d 1309, 1312 (D.C. Cir. 2014) (quoting Thompson v. Clark, 741 F.2d 401, 409 (D.C. Cir. 1984)). Ultimately, agencies must “examine the relevant data and articulate a satisfactory explanation for its action including a ‘rational connection between the facts found and the choice made.’” State Farm, 463 U.S. at 43 (quoting Burlington Truck Lines v. United States, 371 U.S. 156, 168 (1962)). Agency action is arbitrary or capricious where the agency “entirely failed to consider an important aspect of the problem” or “offered an explanation for its decision that runs counter to the evidence before [it].” Id. Contrary to Petitioners’ assertion, the record confirms EPA responded to the concerns identified by Mr. Gebhart. See Deferred Jt. App. at 9-19, 937-65. For example, EPA’s Response to Comment (RTC) document 33 Appellate Case: 21-9509 Document: 010110902437 Date Filed: 08/14/2023 Page: 34 reproduced several of Mr. Gebhart’s comments in full and then responded to them in turn. See id. at 948-50, 951-55. As we will describe, EPA’s robust responses satisfy applicable law. In the RTC document, EPA acknowledged, just as Mr. Gebhart had observed, CAMx modeling has inherent uncertainties. Id. at 954 (“We also agree with the comment that CAMx is still an approximation of physical processes in the atmosphere and has inherent uncertainties.”). And EPA stated, “We have elsewhere in this RTC document explained how these inherent uncertainties or biases were addressed.” Id. On this point, EPA said the dispersion modeling Utah submitted incorporated EPA’s guidance for “model performance evaluations and criteria for determining acceptable model performance.” Id. at 949; see also id. at 954-55. EPA explained it developed this guidance to address the inherent uncertainties identified by Mr. Gebhart. See id. at 954 (explaining how “[c]onsistent with EPA guidance, the modeling analysis submitted by Utah includes measures to address uncertainty in the model simulation, including the use of the SMAT-CE analysis to correct for model bias”). EPA also responded to Mr. Gebhart’s concerns that some uncertainties and biases may have played out differently in the BART Alternative and BART Benchmark modeled scenarios. At oral argument, Petitioners’ counsel emphasized EPA’s purported failure to address this 34 Appellate Case: 21-9509 Document: 010110902437 Date Filed: 08/14/2023 Page: 35 concern. But the record shows otherwise. The RTC document expressly acknowledged Mr. Gebhart’s comments on this issue, stating the EPA recognizes that bias in model results – for example, the overestimates of sulfates and underestimates of nitrate as described by [Mr. Gebhart] here – can result in incorrect estimates of the relative benefits of SO2 and NOx emissions reductions if the modeling results are compared in absolute terms without correcting for model bias. Deferred Jt. App. at 949. EPA then discussed how the dispersion modeling Utah submitted reasonably addressed this concern. EPA explained: To address this concern, the EPA has developed guidance and software to help correct for bias in model results. The EPA’s recommended approach is . . . implemented in publicly available software called [SMAT-CE]. In the recommended approach, the model-simulated future concentrations of sulfate and nitrate are weighted by the amount that the model over or underestimated observed sulfate and nitrate concentrations in the base year simulation. Thus, in this application, the SMAT-CE software was used to reduce the model-simulated future sulfate benefits for each emissions scenario, proportional to the extent that the model overestimated sulfate in the baseline simulation, and to increase the model-simulated future nitrate benefits for each emissions scenario, proportional to the extent that the model underestimated nitrate in the baseline simulation. . . . While no model can perfectly simulate the measured concentrations, the EPA has determined that this is a reasonable approach to correct for systematic bias in model simulations of individual PM2.5 species, and this approach for correcting bias is used in most regulatory applications of photochemical air quality models, both for regional haze and for PM2.5 nonattainment state implementation plans. Id. at 949-50 (emphases added) (footnote omitted). 35 Appellate Case: 21-9509 Document: 010110902437 Date Filed: 08/14/2023 Page: 36 EPA later reiterated its assessment that the dispersion modeling included “measures to address uncertainty in the model simulation,” including “individual components of pollutants that contribute to haze.” Id. at 954. “[O]ther uncertainties in the CAMx modeling such as errors in wind speed, wind direction and atmospheric turbulence,” EPA observed, “apply to both the BART Benchmark and NOx BART Alternative modeling scenarios.” Id. at 954-55. EPA explained these uncertainties in the modeling would not jeopardize the comparison between the BART Benchmark and BART Alternative scenarios because they applied equally to both scenarios. See id. at 955. In its response, EPA disagreed with Mr. Gebhart that the dispersion modeling’s visibility results should be treated as essentially zero, reasoning it “has confidence in the finding of relatively greater visibility benefit in the . . . BART Alternative scenario even when the absolute visibility benefits are small.” Id. (emphasis added). For these reasons, the administrative record does not support Petitioners’ argument that EPA overlooked contrary record evidence or failed to consider an important aspect of the problem regarding the reliability of the dispersion modeling Utah submitted. See, e.g., State Farm, 463 U.S. at 42 (“[A] reviewing court may not set aside an agency rule that is rational, based on consideration of the relevant factors and within the scope of the authority delegated to the agency by the statute.”). Nor does 36 Appellate Case: 21-9509 Document: 010110902437 Date Filed: 08/14/2023 Page: 37 the record support Petitioners’ argument that EPA failed to explain its reliance on the small relative visibility improvement attributed to the BART Alternative. Petitioners’ reliance on National Parks Conservation Association v. EPA (NPCA), 788 F.3d 1134 (9th Cir. 2015), misses the mark. In NPCA, the Ninth Circuit found EPA failed to offer any explanation in its challenged rulemaking on an issue raised in petitioner’s comments. See id. at 1146-47 (describing petitioner’s objections to EPA’s rulemaking, EPA’s responses, and concluding “EPA simply offered no response” to one of the petitioner’s objections). Not so here. As we have described, EPA rationally explained its basis for disagreeing with the comments submitted by Petitioners. Thus, we are unconvinced EPA failed to “consider[] the relevant data and rationally explain[] its decision.” WildEarth Guardians, 770 F.3d at 927. 21 The parties appear to dispute whether Petitioners advanced a “margin of error” challenge to the dispersion modeling’s results during the rulemaking process for the Final Rule. Compare Final Opening Br. 28-31, and Final Reply Br. at 21-23, with, e.g., EPA Response Br. at 34-35. 21 Petitioners argued Mr. Gebhart’s comments raised a “margin of error” challenge, directed the court to pages 887 and 888 of the appellate record as support, and argued EPA failed to respond. See Oral Argument Recording at 43:47-44:25. Respondents argued neither Mr. Gebhart’s comments nor Petitioners’ comments raised a “margin of error” challenge to the dispersion modeling’s results. We note the phrase “margin of error” appears nowhere on the pages referenced by Petitioners’ counsel. Deferred Jt. App. at 887-88 (discussing 37 Appellate Case: 21-9509 Document: 010110902437 Date Filed: 08/14/2023 Page: 38 III EPA’s Final Rule approving Utah’s July 2019 revised SIP and concurrently withdrawing the 2016 FIP was not “arbitrary, capricious, an abuse of discretion, or otherwise not in accordance with law.” 5 U.S.C. § 706(2)(A). We therefore DENY the petition for review. “CAMx modeling uncertainty,” “modeling uncertainty,” “model errors,” “inherent modeling errors” and “modeled concentration”). Nor does “margin of error” appear anywhere in the approximately 30-page comment letter Petitioners submitted on the Proposed Rule. See id. at 840-69. Under the circumstances, we find no basis to determine Petitioners raised such a challenge with reasonable specificity during the rulemaking process. See 42 U.S.C. § 7607(d)(7)(B) (“Only 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 review.”). We therefore lack jurisdiction to consider Petitioners’ argument challenging the Final Rule based on EPA’s purported failure to respond to comments raising a potential “margin of error” issue. Oklahoma v. EPA, 723 F.3d at 1222 (determining, pursuant to § 7607(d)(7)(B), the court lacked jurisdiction to consider an argument asserted in a petition for review where petitioners failed to raise the argument during the underlying rulemaking process). 38
Some case metadata and case summaries were written with the help of AI, which can produce inaccuracies. You should read the full case before relying on it for legal research purposes.
This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.
