Background

[1] The Clean Air Act includes a program for the preven- tion of significant deterioration of air quality (“PSD” pro- gram), which applies to areas that are actually or potentially low in air pollution. Several of the program’s restrictions on emissions are triggered when a major stationary source (i.e., a major pollution emitter) submits an application for a permit for new construction or major modification within the area under the appropriate regulations. 40 C.F.R. § 52.21(b) (14)(ii). A central issue in the present appeal is whether the PSD restrictions were triggered, or must be deemed triggered, in area 61 by the actions of the Barrick Mine. Great Basin contends that the PSD restrictions were triggered, and that as a consequence EPA’s decision to allow division of area 61 was impermissible. We agree with the EPA, however, that the restrictions were not triggered, and that division of area 61 was not arbitrary, capricious, or contrary to law. Before we explain our reasons, we think it best to describe briefly the relevant framework of the Act.

1. The Clean Air Act’s PSD Program

[2] The Clean Air Act regulates air pollutants such as par- ticulate matter, sulfur dioxide, and nitrogen oxide. See 42 U.S.C. §§ 7408-7409. The Act requires the division of states into air quality planning areas (“baseline areas”). These base- line areas are assigned one of three labels—attainment, unclassifiable, or nonattainment—depending on the quality of their air. If an area fails to meet national air quality standards, it is classified as a nonattainment area. If an area meets national standards, it is classified as an attainment area. If it is unclear whether the area meets the standards, the area is denominated unclassifiable. The PSD program applies to the 3544 GREAT BASIN MINE v. EPA latter two categories, and it is undisputed that area 61 falls within them.1

[3] In attainment and unclassifiable areas, the PSD program attempts to maintain the relatively clean air by limiting the total pollution “increment” per year. The PSD restrictions are not automatic, however; they come into effect when an appli- cation is filed under the appropriate regulations for new con- struction of a major stationary source or for major modification of an existing major source within the area. 40 C.F.R. § 52.21(b)(14)(ii). It is important for our purposes to note that application for a permit is the trigger, not simply the existence of substantial pollution emissions. The filing of such an application establishes a “minor source baseline date” for which the EPA determines the ambient “baseline concen- tration” for the area. 40 C.F.R. § 52.21(b)(13), (14)(ii). There- after, the PSD program places strict limits on aggregate increases in pollution within the baseline area whether the increases come from minor or major sources.

[4] If no major source within a baseline area has applied for a permit, however, the PSD restrictions are not triggered. In that event, pollution emitters are free to operate under the less restrictive national air quality standards, rather than the stric- ter standards of the PSD program. See, e.g., Reno-Sparks Indian Colony v. EPA, 336 F.3d 899, 902-03 (9th Cir. 2003).

2. Discretion to Redesignate Baseline Areas

Because the PSD program is administered by baseline areas, the number and size of these areas has a very consider- able effect on potential polluters. The greater the number and the smaller the size of the baseline areas, “the less likely it is that a major source has applied for a permit within any one area, thereby establishing a baseline date.” Id. at 903. Thus a 1 Area 61 was classified as attainment for sulfur dioxide and as unclassi- fiable for particulate matter and nitrogen dioxide. GREAT BASIN MINE v. EPA 3545 minor source is more likely to “find an area in which to oper- ate where it is not subject to the requirements of the PSD pro- gram.” Id. On a smaller scale, the division of one baseline area into two might have a similar effect; if a major source were to seek a permit in only one of the two newly-divided areas, the other might escape PSD regulation that would have applied had the original area not been split.

[5] Despite this consideration, the EPA has broad discretion to grant a state’s request to divide an area for which no base- line date or baseline concentration has been established. The EPA may grant redesignation requests “on the basis of air quality data, planning and control considerations, or any other air quality-related considerations the Administrator deems appropriate.” 42 U.S.C. § 7407(d)(3)(A). This decision also must rely on “sufficient data.” 40 C.F.R. § 81.300(a).

[6] The situation is different when PSD restrictions have already been triggered in an area. Division of such an area into two new areas raises additional problems. One is the question whether one of the new areas may or should be “un- triggered” because the major source that triggered the PSD restrictions lies in the other new area. Perhaps in recognition of this and other problems, the discretion of the EPA is more limited when dealing with redesignation of an area for which PSD restrictions have been triggered. The EPA cannot redesignate, for example, if the new area would “intersect or be smaller than the area of impact of any major stationary source or major modification which . . . [e]stablishes a minor source baseline date.” 40 C.F.R. §§ 51.166(b)(15)(ii), 52.21(b)(15)(ii). Moreover, the EPA has indicated that it is more likely to deny a request to split an area in which the PSD caps apply because the redesignation may allow greater deterioration of the air quality.

The Present Controversy

This dispute arose when the State of Nevada submitted its request to the EPA to divide baseline area 61 (550 square 3546 GREAT BASIN MINE v. EPA miles) into two, lower 61 and upper 61. Nevada claimed that the split would aid its air quality management because the two new areas more accurately reflect the “local air transport pro- cesses,” industrial development, and the region’s topography, among other reasons. Much of the present dispute arises from the fact that, in Nevada’s view, the Barrick Mine, although a major source, has never applied or been required to apply for a permit for new construction or a major modification. This situation presumably results from the fact either that the Mine was a major source before the Clean Air Act’s requirements took effect, or that it became a major source by small incre- ments not subject to permits. Nevada reflected its view in its redesignation request, in which it asserted that “Area 61 does not contain any PSD sources, and it has not been significantly impacted by any major source or modification,” although it acknowledged that Barrick Goldstrike Mine was “a major source for PM10, NOx [nitrogen oxide], and SO2.”

In response to Nevada’s request, the EPA issued a proposed rule granting the request to split area 61 into lower 61 and upper 61. See 67 Fed. Reg. 21194, 21197 (Apr. 30, 2002). The EPA proposed to approve the request because of its pol- icy to “provide States a fair degree of autonomy to balance air quality management with economic planning” and because the redesignation would not interfere with Nevada’s manage- ment of air quality in the area. Id. In concluding that the change met the statutory and regulatory requirements of the Act, the EPA relied on its conclusion that “no PSD source has located in [area 61], . . . and the newly created baseline areas . . . do not intersect the area of impact of any major PSD source nor do they have boundaries that are smaller than such impact area.” Id.

In a footnote, the proposed rule acknowledged that the Bar- rick Mine was a major source, but disregarded it because “the source has not been subject to PSD review.” Id. at 21196 n.3. The proposed rule also stated that the redesignation did not result in “an untriggering of the baseline area,” and thus there GREAT BASIN MINE v. EPA 3547 was “no elimination of already consumed [pollution] incre- ment and no consumed increment would be added to the base- line for the area.” Id. at 21197. Moreover, the division did not “carve out small ‘postage stamp’ areas encompassing only the significant impact area around a major PSD source.” Id. at n.4. The EPA also solicited public comment on its proposed rule.

Relying on the reasons listed in the proposed rule, the EPA’s final rule granted Nevada’s request despite criticism from several commentators, including Great Basin. See 67 Fed. Reg. 68769, 68771, 68776 (Nov. 13, 2002). Great Basin petitioned for review under 42 U.S.C. § 7607(b)(1).

Standard of Review

“In reviewing a final action by the EPA, we reverse only if it is arbitrary, capricious, or contrary to law or if it exceeds the statutory jurisdiction, authority, or limitations.” Exxon Mobil Corp. v. EPA, 217 F.3d 1246, 1248 (9th Cir. 2000). We will overturn a final action of the EPA, “if the agency has relied on factors which Congress has not intended it to con- sider, 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).

Discussion

The critical question is whether Barrick Goldstrike Mine (hereafter “Barrick”) triggered the PSD pollution regulations in area 61. If not, the choice whether to approve the rede- signation is committed to the EPA’s discretion within broad limits that we have already described. We conclude that Barr- ick did not trigger the PSD restrictions, and that the EPA did not act arbitrarily, capriciously, or contrary to law in approv- 3548 GREAT BASIN MINE v. EPA ing division of area 61. We therefore deny the petition for review.

1. The Barrick Mine Did Not Trigger the PSD Restrictions

[7] Great Basin attacks the final rule on four fronts, and we address each in turn. Great Basin first argues that Barrick’s status as a major source triggered the minor source baseline date, thereby limiting the EPA’s discretion to redesignate area 61. The minor source baseline date is triggered by an applica- tion for a permit under the PSD program. See 40 C.F.R. § 52.21(b)(14)(ii); Reno-Sparks, 336 F.3d at 903. The PSD program requires a permit for (1) any new “major emitting facility” or (2) any modification of a major facility that increases the net emissions of regulated pollutants. See, e.g., Alaska Dep’t of Envtl. Conservation v. EPA, 540 U.S. 461, 472 (2004). Although the PSD program requires a permit for the construction of a major emitting facility built after August 7, 1977, see 42 U.S.C. § 7475(a)(1), the record does not indi- cate when Barrick was constructed, or whether it was initially constructed as a major emitting facility. Therefore, Great Basin’s first challenge fails.

[8] Barrick did undertake modifications, however, and they are the center of Great Basin’s final three challenges. There is no dispute that Barrick modified its facilities three times in 2001. There also is no dispute that the PSD program requires a permit for any “major modification” that causes a forty tons- per-year (“tpy”) increase in nitrogen oxides. 40 C.F.R. § 51.21(b)(23)(I); Alaska Dep’t of Envtl. Conservation, 540 U.S. at 472. Finally, it is undisputed that, in the aggregate, Barrick’s three 2001 modifications increased nitrogen oxide emissions over forty tpy. It is disputed, however, whether these totals should be accumulated, and the EPA argues that, when viewed separately, no single modification caused an increase of forty tpy or more. GREAT BASIN MINE v. EPA 3549 In April, August, and October, 2001, Barrick sought per- mits from the State of Nevada’s Bureau of Air Quality for each of the three modifications. For the first and largest emis- sions modification, Barrick claimed that nitrogen oxide emis- sions would increase by less than forty tpy.2 Barrick estimated that its starting nitrogen oxide emissions were 383.3 tpy, and its modified emissions would be 423.1 tpy. This figure (39.8) falls narrowly below the forty tpy trigger.3

Great Basin disputes the starting point, however. Great Basin contends that Barrick reported its starting emissions as 374.6, not 383.3, resulting in an increase of 48.5 tpy of nitro- gen oxides. The EPA responds that this is a “scrivener’s error on Barrick’s part,” and the accurate starting point is 383.3 tpy. The EPA and Nevada chose to rely on this higher starting fig- ure, and their reasonable reliance is supported by the record. We therefore view 383.3 as the correct starting point, and the first modification causes less than a forty tpy increase in nitro- gen oxides.

The next modification increased the estimated nitrogen oxide emissions from 423.1 to 426 tpy.4 Under either starting point, then, if the nitrogen oxide emissions increases from the two modifications are combined for purposes of the PSD trig- ger, the modifications exceed forty tpy (i.e., 42.9).5 Thus, Great Basin’s final two arguments contend that these totals should be accumulated. 2 This modification put twelve of Barrick’s emergency electric genera- tors into regular use. 3 Both the EPA and the State of Nevada noted the narrowness of this margin, and Nevada conditioned the issuance of its permit on Barrick’s agreement that the new generator use would not cause forty tpy or more of additional nitrogen oxide emissions. 4 This modification replaced “existing backfill crushing, screening, and shotcrete plants and [made] miscellaneous changes to related existing facilities.” It is important to note that this modification is unrelated to the previous modification. 5 The final modification did not change the nitrogen oxide emissions. 3550 GREAT BASIN MINE v. EPA [9] Great Basin first argues that the modifications should be accumulated because, in November 2001, Barrick submitted an operating permit application to Nevada “incorporat[ing]” all three “currently pending” modifications. Great Basin claims that the EPA should treat this operating permit applica- tion as a PSD permit application grouping the three modifica- tions. Title V operating permits, however, are not part of the Act’s PSD program. See 42 U.S.C. §§ 7661-7661f; United States v. Marine Shale Processors, 81 F.3d 1329, 1356 (5th Cir. 1996); New York v. Niagara Mohawk Power Corp., 263 F. Supp. 2d 650, 661 (W.D.N.Y. 2003). Therefore, the Act does not require the EPA to treat PSD permits and operating permits interchangeably.6

Finally, Great Basin argues that it should prevail because, under the EPA’s practice of accumulation, the April and August modifications exceed the forty tpy threshold. The final rules of the EPA—at the time of its decision—indicate that, for major sources, minor emissions increases are accumulated during a period “contemporaneous” with the modification, with “contemporaneous” basically meaning within the preced- ing five years. See Requirements for Preparation, Adoption, and Submittal of Implementation Plans; Approval and Pro- mulgation of Implementation Plans, 45 Fed. Reg. 52676, 52701-02 (Aug. 7, 1980). Obviously, the gap between April and August of the same year is less than five years. Therefore, according to Great Basin, the modifications should have trig- gered the PSD caps, and the EPA failed to consider this important factor in its decision.7 6 Our disposition of this issue makes it unnecessary for us to examine Great Basin’s assumption that incorporating three separate modifications in one permit application would require the effects of the modifications to be accumulated. 7 The EPA counters that Great Basin failed to show prejudice because the EPA could divide area 61 even if Barrick had triggered the PSD pro- gram. This argument is unavailing for two reasons. First, the EPA stated in its final rule that it relied on the conclusion that Barrick (or any other entity) had not triggered the PSD caps. GREAT BASIN MINE v. EPA 3551 [10] This argument fails for at least two reasons. First, Great Basin waited far too long to raise it. Great Basin raised the argument neither before the agency, nor in its petition for review, nor in its opening brief in this court; it presented the argument for the first time in its reply brief here. Second, and more important, even if we were to agree with Great Basin’s interpretation, vacating the EPA’s rule on that basis would serve no purpose. In 2003, the EPA clarified the regulation, and it now states that emission increases are not netted unless the modification, in and of itself, is significant (i.e., results in this case in a forty or more tpy increase of nitrogen oxides). Therefore, even if the EPA incorrectly read the regulation then8 to mean exactly what it states now,9 remanding the decision to the agency would not change the result because no individ- ual modification met the threshold.

Second, while the decision to redesignate is largely discretionary, see, e.g., 42 U.S.C. § 7407(d)(3)(A), it does have its limits, particularly when a major source has triggered the PSD program. The EPA is prohibited, for example, from dividing area 61 if the new area would “intersect or be smaller than the area of impact of any major stationary source or major modification which . . . [e]stablishes a minor source baseline date.” 40 C.F.R. §§ 52.21(b)(15)(ii), 51.166(b)(15)(ii). Operating under the assump- tion that Barrick did not trigger a minor source baseline date, the EPA did not analyze whether lower 61 intersects with, or is smaller than, Barrick’s area of impact. Counsel for the EPA analyzed it in his brief, and he thoughtfully concluded that it did not. The EPA itself, however, did not analyze the issue when it made its decision. 8 The regulation stated that a “ ‘major modification’ means any physical change in . . . a major stationary source that would result in a significant net emissions increase of [nitrogen oxides].” 40 C.F.R. § 52.21(b)(2)(I) (2002). 9 The regulation states that a “ ‘major modification’ means any physical change . . . that would result in: a significant emissions increase [of nitro- gen oxides]; and a significant net emissions increase of that pollutant from the major stationary source.” 40 C.F.R. § 52.21(b)(2)(I) (2004) (emphasis added). A “significant emissions increase” in nitrogen oxides means an increase that exceeds 40 tpy. 40 C.F.R. §§ 52.21(b)(23), (40). The EPA’s comments on this change assert that the new regulation “clarif[ies] what has always been [its] policy.” 67 Fed. Reg. 80186, 80190 (Dec. 31, 2002). 3552 GREAT BASIN MINE v. EPA 2. The EPA’s Rule Approving Redesignation Was Not Arbitrary, Capricious, or Contrary to Law

[11] Having determined that Barrick did not trigger the PSD program, we conclude that the EPA permissibly granted Nevada’s request to redesignate. As we pointed out earlier, the EPA has broad discretion to grant or deny requests. With the exception of barriers irrelevant here, the EPA may grant the request “on the basis of air quality data, planning and con- trol considerations, or any other air quality-related consider- ations the Administrator deems appropriate.” 42 U.S.C. § 7407(d)(3)(A); see also 40 C.F.R. § 81.300(a). The EPA’s decision complies with these statutory and regulatory criteria. The record adequately supports the EPA’s finding that the two new areas more accurately reflect the region’s “local air transport processes,” industrial development, and topography, among other reasons. The record also does not indicate that either Nevada or the EPA have a history of carving out “post- age stamp” baseline areas, and upper and lower 61 are no excep- tions.10 Therefore, EPA did not act arbitrarily or exceed its authority in granting Nevada’s request.11

PETITION FOR REVIEW DENIED.

10 The smaller of the two areas, lower 61, is over 200 square miles. Although it was pointed out in oral argument that, in the West, 200 square miles seems more like a postage stamp than it would elsewhere, we con- clude that lower 61 still does not qualify. 11 In light of this disposition, we deny Great Basin’s request for attor- neys’ fees. See Western States Petroleum Ass’n v. EPA, 87 F.3d 280, 286 (9th Cir. 1996).