Sierra Club v. Envtl. Prot. Agency

United States Court of Appeals FOR THE DISTRICT OF COLUMBIA CIRCUIT Argued December 10, 2018 Decided May 31, 2019 No. 16-1158 SIERRA CLUB, PETITIONER v. ENVIRONMENTAL PROTECTION AGENCY AND ANDREW WHEELER, ADMINISTRATOR, U.S. ENVIRONMENTAL PROTECTION AGENCY, RESPONDENTS On Petition for Review of Final Action of the United States Environmental Protection Agency Tosh Sagar argued the cause for petitioner. With him on the briefs were Seth L. Johnson and David S. Baron. Phillip R. Dupré, Attorney, U.S. Department of Justice, argued the cause for respondents. With him on the brief were Jeffrey H. Wood, Acting Assistant Attorney General, Jonathan D. Brightbill, Deputy Assistant Attorney General, and Jonathan Skinner-Thompson, Counsel, U.S. Environmental Protection Agency. 2 Before: GRIFFITH and WILKINS, Circuit Judges, and WILLIAMS, Senior Circuit Judge. Opinion for the Court filed by Senior Circuit Judge WILLIAMS. WILLIAMS, Senior Circuit Judge: To implement the Clean Air Act, the Environmental Protection Agency oversees state procedures for creating and running air monitoring networks. In 2016, EPA adopted a rule, Revisions to Ambient Monitoring Quality Assurance and Other Requirements, 81 Fed. Reg. 17,248 (Mar. 28, 2016) (“Final Rule”), modifying its regulations on the subject, specifically Part 58 of Title 40 of the Code of Federal Regulations. The amendments (1) tightened procedures for state changes to annual monitoring network plans, (2) authorized limited reductions in required sampling frequency, and (3) proposed revisions to certain quality assurance requirements related to monitoring for Prevention of Significant Deterioration. Sierra Club raises three objections. Resting on EPA’s language in the preamble to the rule, it attacks the divergence between EPA’s procedures for reviewing SIPs and annual monitoring network plans—a divergence embodied in a 2006 EPA regulation that has long since passed the deadline for seeking judicial review. It challenges (on the merits) the new authority on sampling frequency reductions. And it sees a fatal procedural defect in the quality assurance adjustments in the form of EPA’s statement—plainly and concededly mistaken— that no commenter had criticized the changes. For the reasons below, however, we find that Sierra Club (1) is barred from seeking review of the claimed legal requirement that monitoring plans be assessed under the same procedures as SIPs because the new rule and EPA’s preamble 3 did no more than echo a prior EPA regulation, (2) lacks standing to attack the sampling frequency changes, and (3) has made no showing that the asserted non-response on quality assurance issues manifested any failure to consider factors relevant to the changes. Thus we dismiss the first two claims and deny the third. * * * The Clean Air Act, 42 U.S.C. §§ 7401–7671q, establishes a comprehensive system for regulating and improving the nation’s air quality, divvying up responsibility between the federal government and the states. First, EPA identifies air pollutants that endanger public health or welfare, and sets National Ambient Air Quality Standards, or NAAQS, that specify the maximum permissible concentration of those pollutants in the ambient air. 42 U.S.C. §§ 7408–09. Then, subject to EPA approval, states adopt State Implementation Plans, or SIPs, id. § 7410(a)(1), which are to bring areas into attainment with the NAAQS (if they are not already), see id. § 7502(a)(2)(A), and to “prevent significant deterioration of air quality,” id. § 7471. To make performance of these functions possible, EPA “promulgate[s] regulations establishing an air quality monitoring system throughout the United States.” 42 U.S.C. § 7619(a). Those regulations, among other things, require states to submit an “annual monitoring network plan” that documents “the establishment and maintenance of an air quality surveillance system that consists of a network of” state or local air monitoring stations. 40 C.F.R. § 58.10(a)(1). We now turn to Sierra Club’s three challenges to EPA’s recent revisions to its monitoring regulations. 4 * * * First and foremost, Sierra Club attacks EPA’s revised regulation governing the review and approval of annual monitoring network plans, 40 C.F.R. § 58.10(a), on the ground that it violates Sierra Club’s reading of the Clean Air Act. Because the act, in Sierra Club’s view, renders a state’s “monitoring network plan . . . part of a SIP,” such plans must be subjected to the review procedures applicable to SIPs. Sierra Club Br. 24. But no later than 2006 EPA’s regulations pursued the non- SIP path. See Revisions to Ambient Air Monitoring Regulations, 71 Fed. Reg. 61,236 (Oct. 17, 2006). A decade later, Sierra Club cannot force EPA back up the trail. The Clean Air Act requires that petitions for review be filed “within sixty days” of a challenged action appearing in the Federal Register. 42 U.S.C. § 7607(b)(1). Accordingly (absent EPA’s reopening the issue), Sierra Club’s time for challenging EPA’s adoption of a non-SIP approach to reviewing annual monitoring network plans has passed. And because the issue is jurisdictional, Sierra Club v. EPA, 895 F.3d 1, 16 (D.C. Cir. 2018), we must raise it ourselves, see, e.g., Gonzalez v. Thaler, 565 U.S. 134, 141 (2012), and dismiss the petition, Medical Waste Inst. & Energy Recovery Council v. EPA, 645 F.3d 420, 427 (D.C. Cir. 2011). EPA’s decision to place annual monitoring network plans outside the SIP-review process was evident. For example, while the statute requires EPA approval of SIP revisions to be preceded by notice and an opportunity for comment, see 42 U.S.C. § 7607(d)(1)(B), (3), (4)(B)(i), (5), (6)(B); see also Sierra Club Br. 8, the 2006 rulemaking provided that, for certain monitoring plans, “the Regional Administrator is not required to provide a separate opportunity for comment,” 40 5 C.F.R. § 58.10(a)(2) (2007) (emphasis added); see also 71 Fed. Reg. at 61,248/1. The 2006 rulemaking also embodied the same disconnect between state processes for formulating monitoring plans and for formulating SIPs—at least under Sierra Club’s reading of the statute. Sierra Club complains that the current provision on the subject is unlawful because it diverges from the statutory requirement applicable to SIP submissions—namely, that states act only after providing “reasonable notice and public hearings,” 42 U.S.C. § 7410(a)(1), (a)(2), (l). See Sierra Club Br. 37. But EPA created that divergence no later than the 2006 rulemaking, which similarly fell short of that standard, demanding only that a monitoring “plan must be made available for public inspection.” 40 C.F.R. § 58.10(a)(1) (2007). Thus, by at least 2006 EPA had necessarily concluded that annual monitoring network plans were not components of a SIP. In the rulemaking currently under review EPA simply continued the same approach. In 2014 it proposed two modest revisions to 40 C.F.R. § 58.10(a). See Revisions to Ambient Monitoring Quality Assurance and Other Requirements, 79 Fed. Reg. 54,356, 54,359/1–2 (Sept. 11, 2014) (“Proposed Rule”). The proposal gave no indication that EPA intended to address the relationship between annual monitoring network plans and SIPs, or the requirements applicable to SIPs, which are addressed (in great detail) elsewhere, see 40 C.F.R. pt. 51 (concerning the “Requirements for Preparation, Adoption, and Submittal of [SIPs]”). In proposing and adopting these tweaks, EPA never purported to close the gap in review procedures between the two types of plans. Rather, it maintained (with 6 slight edits) the non-SIP approach it adopted, at the latest, in 2006. Accordingly, if Sierra Club disagreed with EPA’s disjuncture between monitoring plans and SIPs, it should have raised its objection at the conclusion of the 2006 rulemaking, “within sixty days of EPA’s first use of the [non-SIP-style] approach.” Medical Waste Inst., 645 F.3d at 427; see also, e.g., Am. Iron & Steel Inst. v. EPA, 886 F.2d 390, 397 (D.C. Cir. 1989) (holding that the time for filing a petition started when “EPA first set out its understanding” of its authority). In an effort to tie the monitoring-plan-is-really-a-SIP issue to the 2016 rulemaking, Sierra Club points to a single statement EPA made in the preamble to the Final Rule: [S]ection 110(a)(2)(B) [of the Clean Air Act, 42 U.S.C. § 7410(a)(2)(B),] simply requires that monitoring agencies have the legal authority to implement 40 CFR part 58 [concerning monitoring network plans]; it does not treat annual monitoring network plans . . . as “integral parts” of a SIP subject to public participation whenever such network plans are established or modified. Final Rule, 81 Fed. Reg. at 17,251/3. But far from indicating that EPA intended to reconsider the separation of monitoring plans and SIPs, this statement merely responded (quite briefly) to a comment lodged by Sierra Club’s counsel, Earthjustice, in an attempt to reopen the issue. See Earthjustice & American Lung Association Comments, EPA-HQ-OAR-2013-0619-0034, at 2 (Nov. 10, 2014) (“Earthjustice Comments”), J.A. 96. Petitioners, however, cannot “comment on matters other than those actually at issue, goad an agency into a reply, and then sue on the grounds that 7 the agency had re-opened the issue.” United Transp. Union- Ill. Legislative Bd. v. Surface Transp. Bd., 132 F.3d 71, 76 (D.C. Cir. 1998) (quoting Massachusetts v. ICC, 893 F.2d 1368, 1372 (D.C. Cir. 1990)). Of course, Sierra Club’s submissions might be read as an invitation to EPA to reopen that issue, but agencies are free to decline such invitations. Given “the entire context of the rulemaking,” it is clear that EPA declined and did not reopen consideration of the SIP-monitoring-plan divide. Am. Road & Transp. Builders Ass’n v. EPA, 588 F.3d 1109, 1115 (D.C. Cir. 2009) (citation omitted). In sum, EPA’s rejection of Sierra Club’s extraneous comment did not give Sierra Club the right to challenge longstanding aspects of EPA’s regulations that the agency did not open for reconsideration. (We take no position on the merits of Sierra Club’s view that monitoring plans are a subset of SIPs, nor on whether Sierra Club may challenge EPA’s refusal to adopt SIP-style-review procedures in another context, such as in a petition for rulemaking.) * * * Sierra Club next challenges EPA’s decision to permit Regional Administrators to give case-by-case approval to reductions in the minimum required sampling frequency of monitoring for fine particulate matter. Known as PM2.5, fine particulate matter consists of airborne particles that are 2.5 micrometers in diameter or smaller—less than one-thirtieth the thickness of human hair. Air Quality Designations and Classifications for the Fine Particles (PM2.5) National Ambient Air Quality Standards, 70 Fed. Reg. 944, 945/2 (Jan. 5, 2005). Under prior regulations, certain air monitoring stations that track PM2.5 were required to operate on at least a 1-in-3 8 day sampling frequency. Proposed Rule, 79 Fed. Reg. at 54,360/2; see 40 C.F.R. § 58.12(d) (2015). On this, there is no immediate change. Rather, EPA’s revisions created the possibility of exceptions—enabling possible reductions from 1-in-3 days to 1-in-6 days (or for seasonal sampling). EPA sought to address the sort of situation where a particular monitor was “highly unlikely” to record an otherwise undetected violation of the PM2.5 NAAQS. Final Rule, 81 Fed. Reg. at 17,254/1. One example it noted was a monitor located in an area with “very low PM2.5 concentrations relative to the NAAQS.” Id. Another was a monitor in an urban environment surrounded by a superabundance of other monitors, all with higher readings. Id. Accordingly, EPA reasoned that in such instances the 1-in-3 sampling frequency might be unnecessary. Id. To counteract the possibility of excessive redundancy, EPA gave Regional Administrators a cautiously hedged authority to approve state requests to reduce specific monitors’ sampling frequency to 1-in-6 days or to seasonal sampling. 40 C.F.R. § 58.12(d)(1)(ii) (2018). Under the rule, the Regional Administrator must first conduct a case-by-case analysis, considering factors “including but not limited to the historical PM2.5 data quality assessments” and the location of other PM2.5 monitors. Id. He must also “determine[] that the reduction in sampling frequency will not compromise data needed for implementation of the NAAQS.” Id. Only then may approval be granted. Sierra Club, nevertheless, finds much to fear. Even with an EPA gatekeeper, it says, a reduction in mandatory sampling frequency “creates an increased risk that excessive daily PM2.5 levels will go undetected.” Earthjustice Comments at 4, J.A. 9 98. Sierra Clubs claims that EPA arbitrarily failed to consider this risk increase. But our jurisdiction to consider the issue requires that Sierra Club establish its standing. Sierra Club v. EPA, 292 F.3d 895, 898 (D.C. Cir. 2002). Here, it appears to assert only associational standing. See Sierra Club Br. 31; Sierra Club Reply Br. 25–26. In this context, it must demonstrate, not merely allege, that there is a “substantial probability” that one of its members will suffer an injury if the court does not take action, i.e., prevent EPA from allowing regional administrators to consider reductions in sampling frequency. Sierra Club v. EPA, 754 F.3d 995, 1001 (D.C. Cir. 2014) (quoting Natural Resources Defense Council v. EPA, 464 F.3d 1, 6 (D.C. Cir. 2006)). This demonstration must be made “by affidavit or other evidence.” Sierra Club, 292 F.3d at 899 (quoting Lujan v. Defenders of Wildlife, 504 U.S. 555, 561 (1992)). Sierra Club has failed to make the requisite showing. For a Sierra Club member to face an increased risk of harm, the following conditions would have to be fulfilled. (1) A state must request a reduction in sampling frequency; (2) the request must concern a monitor near one of Sierra Club’s members; (3) the request must be approved by the Regional Administrator; (4) there must be a likelihood that a spike in PM2.5 levels near that monitor will occur at a time when the monitor would have been sampling but for the approved reduction; (5) and conditions must be such that no nearby monitor would pick up the spike. To suggest even a minimally credible possibility of the above occurring, Sierra Club identifies three monitors that are (i) eligible for a reduction in sampling and (ii) placed near a Sierra Club member. One is in Texas (Houston); two are in Oregon (Oakridge and Klamath Falls). Sierra Club Reply Br. 10 26; see Joshua Berman Decl. ¶¶ 34–38 (Mar. 16, 2018). But is Texas or Oregon likely to request any reductions in sampling frequency? Courts are generally “hesitant” to base standing on a chain of events that “‘depends on the unfettered choices made by independent actors not before the courts,’” R.J. Reynolds Tobacco Co. v. FDA, 810 F.3d 827, 831 (D.C. Cir. 2016) (quoting Lujan, 504 U.S. at 562), such as state regulators, see, e.g., Masias v. EPA, 906 F.3d 1069, 1074 (D.C. Cir. 2018); Scenic Am., Inc. v. U.S. Dep’t of Transp., 836 F.3d 42, 50 (D.C. Cir. 2016); Miami Bldg. & Constr. Trades Council v. Secretary of Defense, 493 F.3d 201, 205–06 (D.C. Cir. 2007). In any case, even if Texas or Oregon were likely to request reductions, how likely is it that they would do so for monitors at the sites identified by Sierra Club as near specific members, to wit, sites 482011039, 410350004, or 410392013? Berman Decl. ¶¶ 35– 37. Sierra Club seeks to fill this gap in state motivation by pointing out that “states . . . lobbied for these changes to save money.” Sierra Club Reply Br. 26–27. The inference may be sound—for states that lobbied. But Sierra Club fails to point us to any evidence that Texas or Oregon was among the unspecified states that did so. See id. at 27 (citing Final Rule, 81 Fed. Reg. at 17,254/2, which simply states that all comments, save one, were supportive of the rule change); Berman Decl. ¶ 33 (same). And we need not scour the administrative record ourselves. See, e.g., Masias, 906 F.3d at 1080 (citing Fed. R. App. P. 28(a)(8)(A)). In any event, nothing suggests that the monitors at the three numbered sites are prime candidates for reduction, whatever Texas’s or Oregon’s general plans may be. Cf. Sierra Club v. EPA, 755 F.3d 968, 974 (D.C. Cir. 2014) (finding standing where EPA, effectively reinforcing petitioners’ assertions, pointed to specific refineries near Sierra Club’s members that were “expected to take advantage” of the rule). 11 Further, the eligible monitors appear to be located at rather low-risk sites. In 2016, not one of them recorded a violation of the 24-hour PM2.5 NAAQS—or even came particularly close to doing so. See Berman Decl. ¶¶ 31, 35–37. Nor did any come within even 10% of an annual PM2.5 NAAQS violation—for three reporting periods in a row. See Berman Decl. ¶¶ 31–32, 35–37. Far from it. As the table below indicates, the monitors have consistently—year after year—fallen well below the PM2.5 annual NAAQS. Monitor’s Design Annual Monitor Design % Value* NAAQS Location Value* Diff.** Years (μg/m3) (μg/m3) 482011039 2012-14 9.6 12.0 - 20% (Houston, 2013-15 9.6 12.0 - 20% TX) 2014-16 9.2 12.0 - 23% 410350004 2012-14 10.2 12.0 - 15% (Klamath 2013-15 10.0 12.0 - 17% Falls, OR) 2014-16 8.3 12.0 - 31% 410392013 2012-14 9.2 12.0 - 23% (Oakridge, 2013-15 9.6 12.0 - 20% OR) 2014-16 8.5 12.0 - 29% * “Design values” are “the 3-year average NAAQS metrics that are compared to the NAAQS levels to determine when a monitoring site meets or does not meet the NAAQS . . . .” The table references the annual NAAQS—the “3-year average of PM2.5 annual mean mass concentrations for each eligible monitoring site.” 40 C.F.R. pt. 50, app. N(1.0)(c). ** “The national primary ambient air quality standard[] for PM2.5 [is] 12.0 micrograms per cubic meter (μg/m3) annual arithmetic mean concentration . . . .” 40 C.F.R. § 50.18(a); see also id. pt. 50, app. N(4.4). The “% Diff.” is the difference between the design value calculated using the monitor’s data and the national standard, divided by the national standard. For the underlying data, see Berman Decl. ¶¶ 31, 35–37. 12 Sierra Club identifies no reason to believe that an abrupt reversal in PM2.5 fortunes near these sites is likely, much less “certainly impending.” Nat’l Ass’n of Home Builders v. EPA, 667 F.3d 6, 15 (D.C. Cir. 2011) (quoting Am. Chem. Council v. Dep’t of Transp., 468 F.3d 810, 819 (D.C. Cir. 2006)). Finally, Sierra Club does nothing to build into its theory of harm the analytical exercise that the Regional Administrator must undertake before granting approval, such as determining whether “continuous PM2.5 monitors” exist nearby, and whether an unexpected spike in fine particulate matter would really have registered at one of the sites (had it been kept at 1- in-3) and yet evaded all other monitors. Final Rule, 81 Fed. Reg. at 17,254/1; see also 40 C.F.R. § 58.12(d)(1)(ii). At bottom, Sierra Club’s claim to standing “stacks speculation upon hypothetical upon speculation.” Kansas Corp. Comm’n v. FERC, 881 F.3d 924, 931 (D.C. Cir. 2018) (quoting N.Y. Regional Interconnect, Inc. v. FERC, 634 F.3d 581, 587 (D.C. Cir. 2011)). In these circumstances, Sierra Club has failed to establish standing. Accordingly, the portion of the petition for review challenging EPA’s revisions of minimum sampling frequency is dismissed. * * * Finally, Sierra Club protests adjustments EPA made to four quality assurance requirements for Prevention of Significant Deterioration, or PSD, air monitoring. See Sierra Club Br. 55–56 & n.18; see also Final Rule, 81 Fed. Reg. at 17,271–75. As the name implies, PSD monitoring is designed to evaluate whether new or significantly modified sources of pollution will bring about significant deteriorations in air quality. 13 Until adoption of the Final Rule, the quality assurance requirements for PSD monitoring had generally been the same as the requirements for monitoring used to measure compliance with the NAAQS. Final Rule, 81 Fed. Reg. at 17,271/1. Compare 40 C.F.R. pt. 58, app. A (NAAQS), with 40 C.F.R. pt. 58, app. B (PSD). In 2014, however, EPA proposed some revisions relating to PSD monitoring. See Proposed Rule, 79 Fed. Reg. at 54,369–72. Earthjustice (Sierra Club’s counsel here) and the American Lung Association jointly objected to that proposal, saying that EPA should apply the same requirements to the PSD monitors as it does to monitors ensuring NAAQS compliance. The protest identified four specific ways in which the rule would make the PSD quality assurance requirements weaker than those for the NAAQS, and argued that such relaxations were wrong, primarily because PSD monitoring was “required for the purpose of determining whether the proposed facility will cause or contribut[e] to exceedances of . . . NAAQS.” Earthjustice Comments at 8, J.A. 102; see also Sierra Club Br. 57. EPA overlooked this comment. As the agency now admits, in discussing the Final Rule it inaccurately stated that it had received only favorable comments on its proposed changes. See EPA Br. 49–50; see also, e.g., Final Rule, 81 Fed. Reg. at 17,271/3. Sierra Club argues that EPA could not meaningfully have “respond[ed] to significant points raised by the public,” as EPA must, as it failed even to recognize that anyone made adverse comments. Sierra Club Br. 58 (quoting Lake Carriers’ Ass’n v. EPA, 652 F.3d 1, 11 (D.C. Cir. 2011)). But a “failure to respond to comments is significant only insofar as it demonstrates that the agency’s decision was not based on a consideration of the relevant factors.” Sierra Club 14 v. EPA, 353 F.3d 976, 986 (D.C. Cir. 2004) (Roberts, J.) (emphasis added) (quoting Thompson v. Clark, 741 F.2d 401, 409 (D.C. Cir. 1984)). The principle, of course, applies whether EPA expressly acknowledged Earthjustice’s comment or not. Here EPA plainly addressed the factors that the comment had said must be considered. See generally Final Rule, 81 Fed. Reg. at 17,271–75. Take Sierra Club’s first example—“waiving implementation of the National Performance Evaluation Program (‘NPEP’).” Sierra Club Br. 55 n.18. EPA in fact addressed the substance of Earthjustice’s NAAQS- requirements-must-meet-PSD-requirements concern in this context, saying that NPEP requirements could not be waived “if a PSD reviewing authority intended to use PSD data for any official comparison to the NAAQS beyond” some limited PSD uses. Final Rule, 81 Fed. Reg. at 17,271/2. And it explained, in detail, why PSD monitoring otherwise needed more “flexibility.” Id. For instance, because PSD monitoring is shorter term (usually a year or less), it may, EPA elaborated, “be more difficult” to arrange the specialized equipment, personnel, and relationships that would be needed to implement the NPEP. Id. at 17,271/1–2. This “explanation makes it evident that [EPA] did consider the relevant factors.” Sierra Club, 353 F.3d at 986. The same is true for each of the remaining changes to which Earthjustice objected. Compare Sierra Club Br. 55–56 & n.18, with Final Rule, 81 Fed. Reg. at 17,271–75. As detailed in the table below, for each of the changes identified by Sierra Club, EPA explained why it was altering the PSD requirements (relative to the NAAQS requirements): 15 Changes described Excerpt from EPA’s Explanation in Sierra Club Br. of each change 55–56 n.18 “(2) [E]liminating “Since PSD does not implement lead quality assur- NCore sites, the EPA proposed to ance requirements eliminate the [lead] [quality assur- for collocated sam- ance] language specific to non-source pling and lead per- oriented NCore sites from PSD while formance evalua- retaining the PSD [quality assurance] tion procedures for requirements for routine [lead] moni- non-source oriented toring.” 81 Fed. Reg. at 17,272/1. NCore sites.” “(3) [R]elaxing data “Realizing that PSD monitoring may quality objectives have different monitoring objectives, for PSD monitoring the EPA proposed to . . . allow deci- organizations.” sions on [data quality objectives] to be determined through consultation be- tween the appropriate PSD reviewing authority and PSD monitoring organ- ization.” 81 Fed. Reg. at 17,272/3. “(4) [W]aiving the “Due to the relatively short-term na- concentration valid- ture of most PSD monitoring, the like- ity threshold for im- lihood of measuring low concentra- plementation of the tions in many areas attaining the PM2.5 performance PM2.5 standard and the time required evaluation in the to weigh filters collected in perfor- last quarter of PSD mance evaluations, a PSD monitoring monitoring.” organization[] . . . [may waive the] threshold for validity of performance evaluations conducted in the last quarter of monitoring . . . .” 81 Fed. Reg. at 17,275/1. 16 To be sure, all these explanations may, as a substantive matter, suffer from some infirmity that renders them inadequate. But Sierra Club has not raised that argument, much less developed it. Rather, it steadfastly maintains that EPA “cannot identify any consideration” of Earthjustice’s concerns, Sierra Club Reply Br. 30, a claim that is transparently mistaken. * * * For the foregoing reasons, the petition for review is dismissed in part and denied in part. So ordered.