Sierra Club v. Mosier

• Court: Supreme Court of Kansas
• Decided: March 17, 2017
• Docket: 112008
• Status: Published

Opinion

IN THE SUPREME COURT OF THE STATE OF KANSAS

No. 112,008

SIERRA CLUB, Appellant,

v.

SUSAN MOSIER, M.D., in her Official Capacity as Secretary of the KANSAS DEPARTMENT OF HEALTH AND ENVIRONMENT, and the KANSAS DEPARTMENT OF HEALTH AND ENVIRONMENT, an Agency of the STATE OF KANSAS, Appellees,

and

TRI-STATE GENERATION AND TRANSMISSION ASSOCIATION, INC., and SUNFLOWER ELECTRIC POWER CORPORATION, Intervenors.

SYLLABUS BY THE COURT

1. The Kansas Judicial Review Act, K.S.A. 77-601 et seq., governs the scope of judicial review of an agency's action, and K.S.A. 2015 Supp. 77-621(c) lists the circumstances under which a court may grant relief. Under the Act, the party appealing from an agency's decision bears the burden of establishing error.

2. Under the federal Clean Air Act, 42 U.S.C. § 7401 et seq. (2012), and the Kansas Air Quality Act, K.S.A. 65-3001 et seq., a permit under the prevention of significant deterioration construction permit program for an "anyway source"—that is, a source required to obtain a prevention of significant deterioration construction permit "anyway"

1 for pollutants other than greenhouse gases—must contain limitations on greenhouse gas emissions based on the application of the best available control technology, but only if the permit is issued on or after January 2, 2011.

3. A party abandons an argument by not briefing it.

4. An appellate court exercises de novo review over questions of law regarding whether its mandate has been followed and properly interpreted.

5. The Kansas Judicial Review Act, specifically K.S.A. 77-622(b), allows for a wide range of remedies. In granting relief, a court may order agency action required by law, order agency exercise of discretion required by law, set aside or modify agency action, enjoin or stay the effectiveness of agency action, remand a matter for further proceedings, render a declaratory judgment, or take any other action that is authorized and appropriate.

6. K.S.A. 77-616 and K.S.A. 77-622(b) of the Kansas Judicial Review Act authorize an administrative agency to enjoin or stay the effectiveness of an agency action on appropriate terms unless otherwise precluded by law.

7. In this case, appellant fails to establish that the Kansas Department of Health and Environment erred in granting a stay and limiting the scope of remand proceedings to

2 only those issues subject to the remand order in Sierra Club v. Moser, 298 Kan. 22, 310 P.3d 360 (2013).

8. Under the separation of powers doctrine, determination of appropriate policy must be left to the legislative and executive branches of Kansas government, and courts are limited to the exercise of judicial power in interpreting and applying the law.

9. When statutes and regulations are silent about a particular policy matter, courts are ill-equipped to fill such a gap. Often the wisest course is for courts to defer to the legislature to fill the gap.

10. Courts do not afford significant deference to an administrative agency's statutory interpretation, but where an agency possesses discretion, a court must presume the validity of the agency action and cannot substitute its judgment for that of the administrative agency unless the agency's action is unlawful, unreasonable, arbitrary, or capricious.

11. Before the Secretary of the Kansas Department of Health and Environment may approve a permit for an air contaminant emission source, K.S.A. 2015 Supp. 65-3008a requires the Secretary to provide a public comment period. All relevant comments must be considered in making a final decision on a proposed permit action.

3 12. As a general rule, a party may not raise a new argument in a motion for reconsideration, although some courts recognize an exception when the arguments could not have been presented earlier.

13. K.S.A. 2015 Supp. 77-617 of the Kansas Judicial Review Act limits when a person may obtain judicial review of an issue that was not raised before an administrative agency.

14. The Environmental Protection Agency has established recommended guidelines for air quality modeling to establish a level of consistency. However, it never intended for its guidelines to be a compendium of all acceptable modeling techniques. Use of an alternative model, or the modification of a preferred model, means the model must then be justified on a case-by-case basis. The Environmental Protection Agency permits appropriate reviewing authorities—which includes state agencies like the Kansas Department of Health and Environment—to act as its representatives in approving models.

15. Under K.S.A. 2015 Supp. 77-621(e), courts conducting judicial review of administrative action must consider the harmless error rule.

16. An electric utility steam generating unit permit application must include technical information about proposed emissions, emission controls, and estimated control

4 efficiency under 40 C.F.R. § 63.5(d)(2) (2016). But the Mercury and Air Toxics Standards, 77 Fed. Reg. 9304 (February 16, 2012), envision that even after construction a source will have some period of time to refine its emission controls and demonstrate initial compliance.

17. In reviewing the evidence in light of the record as a whole, the court shall not reweigh the evidence or engage in de novo review.

18. The impact of a written instrument is determined by giving words in the instrument their plain and ordinary meaning.

19. An appellant may not raise new issues in a reply brief.

Appeal from Kansas Department of Health and Environment. Opinion filed March 17, 2017. Affirmed.

Amanda W. Goodin, of Earthjustice, of Seattle, Washington, argued the cause, and Todd D. True and Anna M. Sewell, of the same office, and Robert V. Eye, of Robert V. Eye Law Office, L.L.C., of Lawrence, were with her on the briefs for appellant.

Steve R. Fabert, assistant attorney general, argued the cause, and Jeffrey A. Chanay, chief deputy attorney general, was with him on the brief for appellee.

William L. Wehrum, of Hunton & Williams, LLP, of Washington, D.C., argued the cause, and James D. Oliver, of Foulston Siefkin LLP, of Overland Park, and Howard Kenison, of Lindquist & Vennum, of Denver, Colorado, were on the brief for intervenor Tri-State Generation and Transmission

5 Association, Inc., and William L. Wehrum and Henry V. Nickel, of Hunton & Williams, LLP, of Washington, D.C., and Derek T. Teeter, of Husch Blackwell LLP, of Kansas City, Missouri, and Mark D. Calcara and Mark A. Rondeau, of Watkins Calcara, Chtd., of Great Bend, were on the brief for intervenor Sunflower Electric Power Corporation.

The opinion of the court was delivered by

LUCKERT, J.: In Sierra Club v. Moser, 298 Kan. 22, 310 P.3d 360 (2013) (Sierra Club I), this court reviewed the Kansas Department of Health and Environment's (KDHE's) decision to issue a prevention of significant deterioration (PSD) construction permit to Sunflower Electric Power Corporation (Sunflower). The permit, issued December 16, 2010, authorized Sunflower to build an 895-megawatt coal-fired electric generating unit, referred to as Holcomb 2, at a site near Holcomb where Sunflower already operates a coal-fired station, Holcomb 1.

In Sierra Club I, this court held KDHE had failed to comply with the federal Clean Air Act, 42 U.S.C. § 7401 et seq. (2006), because it had not applied federal Environmental Protection Agency (EPA) regulations setting 1-hour emission limits for nitrogen dioxide and sulfur dioxide. 298 Kan. at 26. We therefore remanded the permit to KDHE with directions to apply the 1-hour emission limits. We also noted that KDHE would have to apply some other regulations that had become effective during the period of judicial review in Sierra Club I; those regulations—known as the Mercury and Air Toxics Standards, 81 Fed. Reg. 24420 (April 25, 2016); 77 Fed. Reg. 9304 (February 16, 2012)—relate to the control of pollutants collectively referred to as hazardous air pollutants. 298 Kan. at 61. As to other issues, we determined that "the scope of the proceedings on remand must be determined by KDHE." 298 Kan. at 62.

6 On remand, KDHE issued an Addendum to the 2010 permit. Sierra Club now seeks judicial review of that action. It argues KDHE could not simply craft an addendum to the 2010 permit but was required to conduct an entirely new permitting process. Sierra Club also argues, for various reasons, the Addendum fails to incorporate newly promulgated regulations for greenhouse gases; fails to properly address the 1-hour emission limits for nitrogen dioxide and sulfur dioxide; fails to properly comply with the hazardous air pollutant limits and the Mercury and Air Toxics Standards; and fails to meet new source performance standards.

We reject Sierra Club's various arguments and affirm KDHE's action.

FACTS AND PROCEDURAL HISTORY

Our decision in Sierra Club I, 298 Kan. at 26-28, discusses the events leading to KDHE issuing the 2010 permit. Highly summarized, as a part of that process, KDHE conducted public hearings and a public comment period during which Sierra Club officials, experts, and members participated. After the hearings and the final public comment period, KDHE responded to the comments of Sierra Club and others and eventually issued the final PSD permit. Sierra Club sought judicial review, and Sunflower and Tri-State Generation and Transmission Association, Inc., which holds an option from Sunflower for rights to a portion of the new power to be generated, intervened.

The Sierra Club I decision also discussed the legal framework for the parties' dispute and explained the interrelationship between the federal Clean Air Act and the Kansas Air Quality Act, K.S.A. 65-3001 et seq., as well as applicable Kansas administrative regulations. See 298 Kan. at 43-47. In essence, the federal Clean Air Act sets up a regulatory system in which the federal EPA bears some responsibility, but the states hold primary responsibility for maintaining air quality within their borders. States

7 must adopt a state implementation plan aimed at assuring the state meets national ambient air quality standards set by Congress and the federal EPA. In Kansas, KDHE administers and enforces Kansas' state implementation plan. See K.S.A. 2015 Supp. 65- 3005(b)(1); K.A.R. 28-19-200(hhh) (2009).

We further explained:

"Kansas' [state implementation plan] follows the [Clean Air Act] by requiring a facility to obtain a permit under the PSD construction permit program, the purpose of which is well- described by its title—the prevention of significant deterioration; it is this type of permit that is at issue in this case. The program applies to the modification or construction of a 'major emitting facility' that is or will be located in an attainment or unclassified area. See 42 U.S.C. § 7475(a) (2006); 40 C.F.R. § 52.21 (2012); K.S.A. 2012 Supp. 65-3029; K.A.R. 28-19-300; K.A.R. 28-19-350 (2012 Supp.); see also 42 U.S.C. § 7479(1) (2006) (defining 'major emitting facility'); 42 U.S.C. § 7479(2)(C) ( 'construction' includes 'modification'); K.A.R. 28-19-200(p) (defining 'construction').

"Under the PSD regulations, the owner of a proposed source must prove that the construction will not cause violations of certain air quality standards. In the [Clean Air Act], Congress charged the EPA with promulgating PSD regulations that would 'provide specific numerical measures against which permit applications may be evaluated, a framework for stimulating improved control technology, protection of air quality values, and fulfill the goals and purposes set forth' in the PSD program. 42 U.S.C. § 7476(a), (c) (2006). In PSD regulations promulgated under this authority, the EPA sets maximum allowable increases, or 'increments,' for some pollutants based on a mathematical relationship to each pollutant's [national ambient air quality standards]. See, e.g., 75 Fed. Reg. 64,864, 64,885 (October 20, 2010). The EPA has also established de minimis thresholds which set specific values, in relation to each pollutant's [national ambient air quality standards], below which the pollutant is not considered to cause or contribute to a violation of the [national ambient air quality standards] or to an established increment. These increments and de minimis thresholds are then used by the states to calibrate their

8 [state implementation plan] requirements. See Environmental Defense Fund v. Adm'r of United States E.P.A., 898 F.2d 183, 185 (D.C. Cir. 1990).

"In addition, if the facility is located in an attainment area, the owner must prove the proposed operations are in compliance with the best available control technology (BACT) requirements or, if the facility is located in a nonattainment area, the lowest achievable emissions rate technology (LAER). See 42 U.S.C. §§ 7475(a)(4), 7479(3), 7503 (2006); 40 C.F.R. §§ 52.21(b)(12), (j)(2) (2012); K.A.R. 28-19-350(b) (2012 Supp.) (incorporating by reference 40 C.F.R. § 52.21); US Magnesium, LLC v. United States E.P.A., 690 F.3d 1157 (10th Cir. 2012). The LAER requirement is applied in nonattainment areas because the [Clean Air Act] seeks to offset emissions increases in those areas with emissions reductions from other sources in the area. New York v. E.P.A., 443 F.3d 880, 883 n.1 (D.C. Cir. 2006) (citing 42 U.S.C. § 7503).

"If a permitting authority determines an owner has met these burdens and issues a PSD permit, the permitting authority must include limitations or conditions to ensure that emissions from the permitted facility: (1) will not cause or contribute to violations of the [national ambient air quality standards] established by the [Clean Air Act] and (2) will be controlled sufficiently to maintain existing air quality in the surrounding region. See United States v. Pacific Gas & Elec., 776 F. Supp. 2d 1007, 1013 (N.D. Cal. 2011).

"It is undisputed that the proposed Holcomb 2 power plant is a facility subject to the PSD program. The proposed capacity makes Holcomb 2 a major emitting facility, and Holcomb is located in Finney County, which has been designated as an attainment area for numerous pollutants. See 40 C.F.R. § 81.317 (2012).

"Hence, the KDHE's permitting process included a [best available control technology] analysis, and the Holcomb 2 PSD permit established emission limits for Holcomb 2." Sierra Club I, 298 Kan. at 45-46.

9 Based on the plain language in 42 U.S.C. § 7475(a)(3) (2006), 40 C.F.R. § 52.21(k) (2012), and K.A.R. 28-19-350(a), (b) (2012 Supp.), in Sierra Club I we held that Sunflower had to demonstrate that the Holcomb 2 project would not cause "air pollution in excess of any [national ambient air quality standards] even if those standards have not been incorporated into Kansas' SIP, unless the federal regulatory requirements indicate otherwise." 298 Kan. at 58. That meant KDHE needed to apply the new 1-hour nitrogen dioxide and sulfur dioxide national ambient air quality standards and had erred in not doing so during the permitting process. 298 Kan. at 58.

That holding, we determined, meant we either did not need to resolve or could not resolve the parties' arguments (1) regarding the methodology used in modeling nitrogen dioxide and sulfur dioxide emission or (2) regarding whether KDHE erred in issuing the Holcomb 2 PSD permit without setting adequate hazardous air pollution emission limits. As to this latter point, the parties had disagreed as to whether the hazardous air pollution regulations in place at the time the PSD permit had been issued applied to Holcomb 2. But the parties agreed the Mercury and Air Toxics Standards, adopted after the issuance of the permit and while Sierra Club I was pending before this court, would apply to Holcomb 2 and would have to be considered if the PSD permit was remanded. 298 Kan. at 59-60.

We next considered whether these holdings rendered moot Sierra Club's contention that the Holcomb 2 PSD permit did not adequately reflect the best available control technology for each regulated pollutant. We concluded we could not determine whether the issue was moot on the record before us. We noted that 40 C.F.R. § 52.21(r)(2) and K.A.R. 28-19-301(c) (2009) provide that a PSD permit expires if construction does not begin within 18 months of issuance. This time limitation seeks to

10 ensure a new emitting source uses up-to-date technology. But during Sierra Club I proceedings, KDHE issued a stay of the construction deadline. 298 Kan. at 61-62.

The stay order was not part of the record in Sierra Club I, however, and the parties had not argued about the impact remand proceedings would have on the best available control technology assessment. We, therefore, concluded "the scope of the proceedings on remand must be determined by the KDHE." 298 Kan. at 62. Because the remand proceedings might rely on KDHE's 2010 best available control technology determinations, we considered and rejected Sierra Club's various arguments regarding those determinations. 298 Kan. at 64-78.

Shortly after the issuance of the Sierra Club I mandate, Sunflower contacted KDHE and asked whether the stay was still in place. KDHE responded: "The Stay Order continues in effect since the reversal and remand did not result in final disposition of the permit and the terms of that permit." Sierra Club challenged the lawfulness of the stay; nevertheless, KDHE continued the stay "until completion of the remand proceedings."

During the remand, Sunflower did not modify its application or propose the use of new technology, and KDHE therefore determined it need only address this court's remand directions in Sierra Club I. KDHE issued a draft "Addendum to Air Emission Source Construction Permit" in January 2014. KDHE did not issue new or updated monitoring data or other documentation with the Addendum. KDHE opened a 30-day public comment period and held a public hearing in Garden City near the Holcomb site.

Sierra Club and some of its members submitted written comments. Sierra Club's comments incorporated its 2010 comments and added other objections. Regarding the remand proceedings, Sierra Club asked for a longer public comment period and more

11 public hearings, arguing "the truncated permit 'addendum' process proposed by KDHE" was an inadequate "substitute for a full new permitting decision."

As to specifics of the draft Addendum, Sierra Club argued that, although the draft noted the new Mercury and Air Toxics Standards and other hazardous air pollutants regulations applied, it did not adjust limits from the 2010 permit and those limits were not as stringent as required by the new regulations. Sierra Club noted that KDHE had relied on outdated modeling conducted for the 2010 permitting process and the 1-hour nitrogen dioxide and sulfur dioxide limits were not sufficiently stringent. Finally, Sierra Club objected that the best available control technology determination for Holcomb 2 was dated and, during the intervening time since remand, similar facilities had consistently achieved emission limits lower than those set in the Holcomb 2 PSD.

Sierra Club reiterated its position that Sunflower would have to "substantially change the design" of Holcomb 2 and that "KDHE's alleged 'stay' is unlawful and without effect." Sierra Club commented that "Sunflower must apply for a new permit, or at the very least must update its [best available control technology] determination to ensure that the plant it constructs incorporates the most current and most protective pollution controls," including those for greenhouse gases governed by a regulation that became effective January 2, 2011. Sierra Club attached expert analysis to support its contentions.

Sunflower also provided public comments. Relevant to the issues on appeal, Sunflower provided updated modeling data—dated February 2014—regarding emissions at the Holcomb site.

12 On May 30, 2014, KDHE responded to the comments it had received. Responding to Sierra Club's comment about the best available control technology determination, KDHE wrote:

"[T]he December 16, 2010 permit is still under the initial 18 months from issuance of the permit to commence construction. Because of the ongoing litigation after issuance of the permit, the facility requested and was granted a 'stay' by the Secretary of KDHE's July 11, 2011 Stay Order effective June 1, 2011. The facility further requested confirmation on November 7, 2013 that the Stay was still in effect until completion of the remand decision. On November 12, 2013 the Secretary of KDHE's letter stated the Stay Order continues in effect since the reversal and remand did not result in final disposition of the permit and the terms of that permit. The facility will still have 12 months and 2 weeks to construct once the Stay is lifted before an application for an extension, and possible re- evaluation of [best available control technology], would be required."

On the same day, KDHE issued an "Addendum to the December 16, 2010 Construction Permit." In its Responsiveness Summary, KDHE indicated the Addendum "supplement[ed]" the already issued permit. The Addendum, unchanged from the draft, provided that "[e]xcept as specified in this Addendum, all provisions of the permit issued December 16, 2010 remain in effect." The Addendum provided:

"2a. (last paragraph) The owner or operator shall not emit or cause to be emitted [nitrogen oxides] emissions exceeding 1740 lbs/hour on a one hour block average basis, including during startup and shutdown.

"2b. (last paragraph) The owner or operator shall not emit or cause to be emitted [sulfur dioxide] emissions exceeding 4089 lbs/hour on a one hour block average basis, including during startup and shutdown.

13 "2j. The owner or operator shall comply with all applicable provisions of 40 CFR Part 63 Subpart UUUUU for an [electric utility steam generating unit] as defined per 40 CFR 63.9985. Applicable portions of Emissions and Operating Limits, Work Practice Standards, Performance Testing, Continuous Compliance, and Reporting sections from the 40 CFR Part 63 Subpart UUUUU in effect upon startup of [Holcomb 2] shall apply."

The first two paragraphs related to the 1-hour limits on nitrogen dioxide and sulfur dioxide and the third paragraph to the Mercury and Air Toxics Standards.

In June 2014, Sierra Club asked KDHE to reconsider its decision. For the first time, Sierra Club objected to KDHE's reliance on significant impact levels in reaching its conclusion that Holcomb 2 did not cause or contribute to violations of the national ambient air quality standards. KDHE denied Sierra Club's request, which ended the remand proceedings and prompted Sierra Club to file a petition for judicial review in the Court of Appeals. We granted KDHE's subsequent motion to transfer the case to this court. See K.S.A. 2015 Supp. 20-3017. In October 2014, Sunflower requested an extension of the 18-month deadline for beginning construction, and KDHE granted Sunflower's request. KDHE served Sierra Club with the order, and Sierra Club did not object.

ANALYSIS

In Sierra Club's initial brief before this court, it asks us to vacate the permit because of three alleged errors, which we have reordered as follows for purposes of our analysis: Whether KDHE erred in issuing the Addendum because the permit (1) does not contain greenhouse gas emissions limitations based on a best available control technology determination; (2) fails to ensure compliance with 1-hour standards for

14 nitrogen dioxide and sulfur dioxide; and (3) does not contain adequate emission limits under the Mercury and Air Toxics Standards for hazardous air pollutants. Within these three issues, Sierra Club asserts recurring arguments suggesting KDHE should have conducted a new permitting process on remand, should have required updated modeling and assessments, and should have applied regulations that became effective after December 16, 2010, when KDHE initially issued the Holcomb 2 PSD permit. In Sierra Club's reply brief it added a fourth issue, arguing Holcomb 2 cannot meet new source performance standards.

The Kansas Judicial Review Act, K.S.A. 77-601 et seq., governs the scope of our review of KDHE's action as to each of Sierra Club's issues and arguments. K.S.A. 2015 Supp. 77-621(c) lists eight circumstances under which a court may grant relief from an agency action. Of those eight circumstances, Sierra Club relies on three. Specifically, it argues KHDE (1) erroneously interpreted or applied the law; (2) made a fact determination not supported by substantial evidence; and (3) acted unreasonably, arbitrarily, or capriciously. K.S.A. 2015 Supp. 77-621(c)(4), (7), (8). As the party appealing from KDHE's decision, Sierra Club bears the burden of establishing error. K.S.A. 2015 Supp. 77-621(a)(1).

In response to Sierra Club's initial brief before this court, KDHE and the Intervenors argue Sierra Club failed to adequately argue how those three standards apply to each of Sierra Club's issues and failed to provide a pinpoint reference in the record where the issue was raised and ruled upon in the KDHE proceedings, as required by Supreme Court Rule 6.02(a)(5) (2015 Kan. Ct. R. Annot. 41). Indeed, Sierra Club's opening brief made only general references to K.S.A. 2015 Supp. 77-621 and did not always cite to points in the record where an issue had been raised below. Nevertheless, Sierra Club cured these problems in its reply brief. While not a model of appellate

15 practice, the reply fulfilled the purpose of Rule 6.02(a)(5) and resulted in no prejudice; KDHE, the Intervenors, and this court had the information before the parties argued to this court, at which time KDHE and the Intervenors had the opportunity to dispute and counter Sierra Club's positions.

KDHE and the Intervenors also generally argue Sierra Club may not seek relief on appeal because it failed to timely raise an argument before KDHE or because it waived, abandoned, or otherwise failed to preserve an argument for review. We will address these concerns within our discussion of the four issues raised by Sierra Club.

ISSUE 1: GREENHOUSE GAS EMISSION LIMITATIONS DID NOT APPLY.

The parties present several arguments relating to the validity of Sunflower's PSD permit in light of regulations imposing greenhouse gas emission limits. We first consider the regulatory context in which those arguments arise.

Under federal and Kansas law, a PSD permit for an "anyway source"—that is, a source required to obtain a PSD permit "anyway" for pollutants other than greenhouse gases—must contain limitations on greenhouse gas emissions based on the application of the best available control technology, but only if the permit is issued on or after January 2, 2011. 40 C.F.R. § 51.166(b)(48)(iv) (2016) (beginning January 2, 2011, greenhouse gas regulations apply to "new major stationary source[s]" and specified "existing major stationary source[s]"); 40 C.F.R. § 52.21(b)(49) (2016); Utility Air Regulatory Group v. EPA, 573 U.S. ___, 134 S. Ct. 2427, 2448, 189 L. Ed. 2d 372 (2014) ("EPA's decision to require [best available control technology] for greenhouse gases emitted by sources otherwise subject to PSD review is, as a general matter, a permissible interpretation of the statute"); K.A.R. 28-19-200a (2015 Supp.) (adopting definitions to implement federal

16 greenhouse gas tailoring rule); see, e.g., 75 Fed. Reg. 17004, pp. 17020, 17021 (April 2, 2010) ("EPA has concluded that [greenhouse gases] will not become subject to regulation [and hence the PSD best available control technology requirement will not apply to them] earlier than January 2, 2011"; "each PSD permit issued on or after January 2, 2011 would need to contain provisions that satisfy the PSD requirements that will apply to [greenhouse gases] as of that date"); see also 81 Fed. Reg. 68110 (Oct. 3, 2016) (public notice of proposed revisions to regulations); 75 Fed. Reg. 31514, p. 31527 (June 3, 2010) (Under tailoring rule, "[a]ny PSD permits issued to . . . Step 1 sources on or after January 2, 2011 will need to address [greenhouse gases].").

Because Sunflower's 2010 permit was issued weeks before the January 2, 2011, effective date of these regulations, KDHE did not include greenhouse gas emission limits in the 2010 permit. Nor did KDHE add limits when it issued the 2014 Addendum. Sierra Club argues we should set aside the permit under K.S.A. 2015 Supp. 77-621(c)(4), (7), and (8) because the omission of greenhouse gas emission limits constitutes an error of law, was based on facts not supported in the record, and was otherwise unreasonable, arbitrary, and capricious.

As a preliminary matter, KDHE argues Sierra Club failed to preserve this issue— more specifically, that it waived the argument—when it did not object to the 2010 PSD permit on the basis of deficient greenhouse gas emission limits. Further, KDHE argues the doctrine of res judicata bars consideration of this issue in light of Sierra Club v. Moser, 298 Kan. 22, 310 P.3d 360 (2013) (Sierra Club I). We reject KDHE's waiver arguments because the greenhouse gas emission limits did not apply to the December 16, 2010, PSD permit at issue in Sierra Club I. Hence, the greenhouse gas regulations were not ripe for decision and Sierra Club had no obligation to raise the issue in order to argue, as it does now, that the regulations applied. See State v. Kingsley, 299 Kan. 896, 901, 326

17 P.3d 1083 (2014) (for res judicata to apply, four elements must be met, including that a second action must raise the same claim as had been raised in a previous action and the previous action must have resulted in a final judgment on the merits of the duplicate claim); Shipe v. Public Wholesale Water Supply Dist. No. 25, 289 Kan. 160, 165-66, 210 P.3d 105 (2009) (Kansas courts do not consider issues unless the issues are ripe, meaning they have "taken fixed and final shape rather than remaining nebulous and contingent"); State v. Morton, 283 Kan. 464, 472-73, 153 P.3d 532 (2007) (holding that in remand proceedings before the district court, the law of the case doctrine did not bar reconsideration and reversal of a motion in limine at the new trial because the issue was not addressed in the first appeal).

Sierra Club's arguments in this current action became ripe only when KDHE issued the Addendum on remand, which was after the effective date of the regulations, without incorporating the greenhouse gas emission limitations. Sierra Club could not have waived or abandoned this argument before it became ripe, and the argument is properly before us. Consequently, we turn to the merits.

KDHE justified omitting the greenhouse gas emission limitations on the grounds that it had stayed the effect of K.A.R. 28-19-301(c) and had not issued a new permit. The regulation which KDHE effectively stayed, K.A.R. 28-19-301(c), incorporates 40 C.F.R. § 52.21(r)(2), which provides an "[a]pproval to construct shall become invalid if construction is not commenced within 18 months after receipt of such approval." See Sierra Club v. Franklin County Power of Illinois, 546 F.3d 918, 934 (7th Cir. 2008) (18-month limitation "ensure[s] that major emitting facilities comply with up-to-date [pollution control] . . . technology"). According to KDHE, by staying the effectiveness of the issued permit it had kept the 2010 permit in effect for approximately 4 years (now longer), but the 18-month limitation has been tolled and has not yet expired. In fact, as

18 previously quoted, KDHE explained that "[t]he facility will still have 12 months and 2 weeks to construct once the Stay is lifted before an application for an extension, and possible re-evaluation of [best available control technology], would be required." Because of the stay and because KDHE did not and does not view the Addendum as a new permit that triggered the greenhouse gas regulations, it determined nothing occurred on remand that changed the status from Sierra Club I—the permit had been issued in 2010 several weeks before the greenhouse gas rules became effective. Basically, in KDHE's view, the Addendum escaped the effective date of the greenhouse gas regulations.

Considering these circumstances, KDHE points out—validly, in our view—that Sierra Club has failed to properly raise any objection to the stay during the judicial review proceedings despite having argued to KDHE that the stay was illegal. A similar situation occurred in Sierra Club I, where Sierra Club raised objections to the stay before KDHE but did not continue these arguments during judicial review. See 298 Kan. at 61-62.

During the 2014 remand proceedings when Sierra Club renewed its objection to the stay, KDHE responded by presenting a memorandum issued by Stephen Page, director of the EPA's Office of Air Quality Planning and Standards, titled "Guidance on Extension of Prevention of Significant Deterioration (PSD) Permits under 40 CFR 52.21(r)(2)" and dated January 31, 2014 (Page Memorandum). The EPA intended through the Page Memorandum to clarify the agency's position on what constitutes an adequate justification for an extension of the 18-month period for beginning construction before the PSD permit expires. Although not specifically addressing a stay, as we are confronted with in this case, the Page Memorandum provides guidance regarding a state's discretion in handling construction deadlines. After quoting portions of the Page

19 Memorandum, KDHE concluded the stay had remained in place and, as a result, "there is no requirement for a new permit application or updated [best available control technology] determination."

Before us, Sierra Club now argues about the legal effect of the stay but it does not assert KDHE could not grant the stay. By not briefing those arguments, Sierra Club has abandoned them. See Superior Boiler Works, Inc. v. Kimball, 292 Kan. 885, 889, 259 P.3d 676 (2011) (an issue not briefed is abandoned). We, therefore, assume without deciding that KDHE had authority to issue the stay.

Nevertheless, Sierra Club has preserved and raised several other arguments about the legal effect of the stay. It primarily argues this court vacated the 2010 permit in Sierra Club I, which means the "2010 permit ceased to exist on the date of the Court's order, and the greenhouse gas requirement clock was accordingly reset." Sierra Club secondarily argues the greenhouse gas emission limits apply because: As a matter of law a remand acts as a vacatur; KDHE's modification of the PSD constituted a "final permit decision" that triggered application of the greenhouse gas emission limits; and KDHE needed to apply those limitations when modifying the permit, and the permit therefore failed to comply with new EPA rules announced August 3, 2015.

We turn to these arguments.

1.1. Sierra Club I did not mandate a new permitting process.

First, we examine Sierra Club's primary argument that this court vacated the 2010 permit, which means the 2014 permit was a "new permit" that had to comply with the greenhouse gas emission limitations. This argument places into issue the scope of our mandate in Sierra Club I. As such, it presents a question of law over which we exercise

20 unlimited review. See In re Estate of Einsel, 304 Kan. 567, 584, 374 P.3d 612 (2016) (compliance with mandate and the proper interpretation of a mandate are questions of law over which appellate courts exercise de novo review).

Sierra Club finds some support for its argument that Sierra Club I mandated a new permitting process on remand, including a new best available control technology determination relating to greenhouse gas emissions, from this court's isolated use of the word "vacate" at one point in the decision. The decision indicates we were "vacating the permit and ordering that the KDHE must apply new [hazardous air pollutant] regulations before issuing a new permit." (Emphasis added.) Sierra Club I, 298 Kan. at 82. This one- time use of the word "vacate" and the accompanying reference to a "new" permit certainly created ambiguity, which we regret.

Nevertheless, a lengthy discussion of the scope of the remand proceedings indicated our intent to leave the scope of the remand proceedings to KDHE's discretion. In particular, we discussed whether KDHE would need to perform a new best available control technology determination on remand. We stated:

"In their briefs, the parties do not discuss the potential for some arguments becoming moot if we find error on another issue. Nor do the parties suggest the scope of proceedings if the permit is remanded. Instead, we raise these issues sua sponte. In doing so, we recognize that the parties should be given an opportunity to present their positions regarding the scope of proceedings on remand and whether there must be a new [best available control technology] determination. [Citation omitted.]

"We also recognize there may be factual components to these arguments. For example, we are unable to determine whether additional modeling is necessary under the new 1-hour [nitrogen dioxide and sulfur dioxide standards]. In addition, we are unable to determine the potential effect of an order to stay that is referred to in the parties'

21 arguments but is not in the record before us. Based on the arguments, it appears that the KDHE stayed the effect of K.A.R. 28-19-301(c), which provides that a PSD permit will expire if construction has not begun within 18 months of its issuance. [Citations omitted.] We can only speculate as to the specifics of that order and its impact on remand proceedings, and the order is not before us for review.

"We, therefore, determine that the scope of the proceedings on remand must be determined by the KDHE. Because we express no opinion on the appropriate scope, we cannot rule that Sierra Club's [best available control technology] issue is moot. It is conceivable the KDHE will rely on some or all of its prior determinations." (Emphases added.) Sierra Club I, 298 Kan. at 61-62.

Recognizing the possibility KDHE would not conduct another best available control technology determination, we considered the parties' arguments regarding potential errors in KDHE's 2010 determination and held no error occurred. 298 Kan. at 65-78. In other words, the discussion of the issues in the Sierra Club I decision made clear that we were not mandating a new permitting process. Furthermore, in ordering the final disposition of the petition for review, we reversed and remanded—we did not state that we had vacated the permit. 298 Kan. at 88.

Following our direction that KDHE should determine the scope of proceedings on remand, KDHE concluded the 2010 PSD permit remained in effect under its stay order, a new best available control technology determination need not be conducted, a new permitting process was not required, and an addendum to the 2010 permit could be issued. Our decision in Sierra Club I granted KDHE the discretion to make such broad determinations regarding the scope of the remand proceedings, and our mandate did not vacate the 2010 Sunflower PSD permit.

22 1.2. Vacatur was not an inescapable result of remand.

Sierra Club also argues this court's remand decision in Sierra Club I essentially amounts to an order to vacate the permit and that KDHE necessarily had to treat the remand as a vacatur order. Sierra Club bolsters its argument by citing cases under the Federal Administrative Procedure Act (APA); these cases support the proposition that a federal reviewing court normally vacates an erroneous agency action. Indeed, 5 U.S.C. § 706(2) (2012) states a reviewing court "shall . . . set aside agency action" that is unlawful. But two important concessions accompany Sierra Club's general proposition: (1) Despite this clear statutory language, remand without vacating occurs in federal court, and (2) the strict language of the APA does not govern this issue in Kansas.

As Sierra Club recognizes in the first concession, federal courts reviewing administrative actions do occasionally remand without vacating, although some debate whether that is even a legitimate option in the federal system. See, e.g., Levin, "Vacation" at Sea: Judicial Remedies and Equitable Discretion in Administrative Law, 53 Duke L.J. 291 (2003). Even supporters of a judicial option to remand without vacating admit that the option is only available if section 706 of the APA is not read to literally mean what it says. Levin, 53 Duke L.J. at 292, 309-12.

Courts that remand without vacating often recognize pragmatic principles, noting that the rapidly evolving realm of administrative law sometimes calls for flexible remedies to avoid the disruptive effects of a complete redo—generally, when there would be irreparable harm or when remand alone seemed reasonable after balancing the significance of the agency's deficiencies with the disruptive consequences of vacatur. See, e.g., Cal. Communities Against Toxics v. United States E.P.A., 688 F.3d 989, 992 (9th Cir. 2012); Heartland Regional Medical Center v. Sebelius, 566 F.3d 193, 197 (D.C. Cir. 2009); see also Com. of Mass. v. United States Nuclear Regulatory Com'n, 924 F.2d

23 311, 336 (D.C. Cir. 1991) (specifically stating the court "will remand without vacating an agency's order where the reason for the remand is lack of reasoned decisionmaking" and to do otherwise would have "disruptive consequences"); Levin, 53 Duke L.J. at 345 ("[P]ragmatism and flexibility have been recurrent and durable themes in the jurisprudence of administrative law remedies.").

These cases create an inference that cuts against Sierra Club's position; even in the face of clear language requiring courts to set aside an agency's actions, courts have tailored a remedy to permit a remand without vacatur when such a flexible remedy is required, such as when vacatur would cause irreparable harm or be unduly disruptive. Recognizing these exceptions, Sierra Club urges a similar threshold requirement in Kansas and argues neither this court nor KDHE made the appropriate pragmatic findings. But Sierra Club's argument does not fit Kansas law.

Unlike the plain and mandatory language of section 706 of the APA, the Kansas Judicial Review Act allows a wide range of remedies. A court may

"grant other appropriate relief, whether mandatory, injunctive or declaratory; preliminary or final; temporary or permanent; equitable or legal. In granting relief, the court may order agency action required by law, order agency exercise of discretion required by law, set aside or modify agency action, enjoin or stay the effectiveness of agency action, remand the matter for further proceedings, render a declaratory judgment or take any other action that is authorized and appropriate." K.S.A. 77-622(b).

Only one of these many alternative remedies "set[s] aside" or vacates the administrative order. The Kansas Judicial Review Act provided this court with the authority to craft a remedy, and it did not compel KDHE, on receipt of the remand order, to treat the permit as having been vacated.

24 In sum, despite the one instance where we used the word "vacate" in Sierra Club I, this court left it to KDHE to determine how to comply with the EPA permitting process. Sierra Club makes no argument and cites to no Kansas authority that would prohibit a remand without vacating the 2010 permit. Consequently, we reject Sierra Club's claim that KDHE was required as a matter of law to consider the 2010 permit as having been vacated.

1.3. Sierra Club does not establish that the Addendum constitutes a new permit.

Sierra Club also argues that even if the remand proceedings were limited and the Addendum was a modification of an existing permit, it must be treated as a new final permit for purposes of the greenhouse gas regulations. As a result, according to Sierra Club, KDHE had to make a best available control technology determination for greenhouse gases. Again, it is Sierra Club's burden to show that a specific action taken by KDHE amounted to error. K.S.A. 2015 Supp. 77-621(a).

To meet that burden, Sierra Club first points to EPA's refusal to adopt grandfathering rules when it adopted the greenhouse gas emission limitations. While that is true, the EPA also stated:

"Final PSD permits issued before January 2, 2011 need not be reopened or amended to incorporate requirements for [greenhouse gases] that take effect after the permit is issued. A source that is authorized to construct under a PSD permit but has not yet begun actual construction on January 2, 2011 may begin actual construction after that date without having to amend the previously-issued PSD permit to incorporate [greenhouse gas] requirements, provided the permit has not expired." Prevention of Significant Deterioration and Title V Greenhouse Gas Tailoring Rule, 75 Fed. Reg. 31514, p. 31527 (June 3, 2010).

25 Here, as we have discussed, KDHE and the Intervenors take the position the stay tolled the expiration period and the permit had not expired. Consequently, according to KDHE, the EPA guidance does not apply.

Sierra Club also cites 40 C.F.R. § 124.15 (2016) to argue that a "final permit decision" includes a decision to modify a permit and to suggest that the date of a modification should be a permit's date of issuance. But 40 C.F.R. § 124.15 is part of the water program—not the air program—and the section discusses when a decision is final for appeal purposes. The provision simply does not apply to Sunflower's PSD permit, and Sierra Club does not cite any other federal regulation containing similar language that addresses the type of permit at issue.

Indeed, Sierra Club failed to brief an argument that KDHE's Addendum amounted to an unlawful amendment, revision, or modification of the 2010 permit. Moreover, Sierra Club's reply brief expressly rejected that position: The Addendum "was not a mere modification of an old permit which remained in effect." While Sierra Club shifted this position at oral argument and suggested the Addendum was a modification, simply presenting a new theory at oral argument is insufficient to present and preserve an argument never briefed before this court, especially where the written argument takes an opposite position. See State v. Tague, 296 Kan. 993, 1010, 298 P.3d 273 (2013) (refusing to consider new issue raised after brief filed).

In addition, Sierra Club cites to decisions of the EPA's Environmental Appeals Board in which intervening changes in the law have been applied during a remand, even if the new provisions were not the subject of the remand order. E.g., In re: Shell Gulf of Mexico, Inc., Shell Offshore, Inc., 15 E.A.D. 103, 2010 WL 5478647, at *35 (EAB Dec.

26 30, 2010); see also Ziffrin, Inc. v. United States, 318 U.S. 73, 78, 63 S. Ct. 465, 87 L. Ed. 621 (1943) (asserting that permitting and licensing decisions of regulatory agencies generally must reflect the law in effect at the time the agency makes a final determination on a pending application); State of Ala. ex rel. Baxley v. Environ. Pro. Agcy., 557 F.2d 1101, 1110 (5th Cir. 1977) (same). Yet Sierra Club fails to link that caselaw to any requirement that a state must do the same. And, significantly, federal procedure differs from Kansas'. Under the EPA-administered program, a PSD permit does not become effective if properly appealed to the Environmental Appeals Board. See 40 C.F.R. § 124.15(b). Sierra Club does not cite to any similar provision in Kansas law.

Significantly, under the Kansas Judicial Review Act, "[u]nless precluded by law, the agency may grant a stay on appropriate terms or other temporary remedies during the pendency of judicial review." K.S.A. 77-616(a); see K.S.A. 77-607 (requiring final agency action before seeking judicial review). And, as we have noted, Sierra Club has waived or abandoned any argument that the stay was precluded by law.

Finally, at oral argument, Sierra Club emphasized the effect of K.S.A. 2015 Supp. 65-3008a and K.A.R. 28-19-204 (2009) and argued these provisions treat post-permit proceedings like new permits. K.S.A. 2015 Supp. 65-3008a requires the opportunity for public comment and a public hearing before issuing, modifying, renewing, or reopening a permit. See K.S.A. 65-3008b(d) (allowing the Secretary of KDHE to "modify or reopen an approval or a permit for cause"). Likewise, K.A.R. 28-19-204 imposes public comment and hearing requirements. While these provisions clearly impose a hearing and public comment process, KDHE complied with that requirement with regard to those issues subject to the remand order or which KDHE deemed reopened. In general, these Kansas statutes and regulations grant KDHE wide discretion in handling proceedings that are modifications or reopenings, including the public comment period. Of course, this

27 discretion might be limited if provisions of the federal Clean Air Act or its regulations apply.

In determining what federal limitations needed to be considered, KDHE relied heavily on the January 31, 2014, Page Memorandum. Although the Page Memorandum addresses extensions of the construction deadline and not stays, it bolsters KDHE's decision to grant a stay and the resulting scope of the remand proceedings.

For example, Director Page noted that "[t]he text of 40 CFR 52.21(r)(2) does not provide any specific criteria or required process that must be satisfied before the EPA can exercise its discretion to determine that a permit extension is justified." Page Memorandum, p. 2. As a result, the EPA determined it would consider requests for extension on a case-by-case basis. It noted a number of factors that might justify an extension, including judicial review. Page Memorandum, p. 3. When explaining its position during the administrative proceedings, KDHE referred to the Page Memorandum for guidance and specifically noted the delay caused by judicial review.

Moreover, Director Page observed that even if a state incorporated 40 C.F.R. 52.21(r)(2) and the 18-month construction timeline into its regulations and plan, which Kansas has done, nothing requires a state to follow the EPA's memorandum regarding justifications and timeframes for extensions. Page Memorandum, p. 5. In other words, KDHE had broad discretion in how it complied with the construction deadline.

Significant to our analysis regarding the scope of post-permit proceedings, Director Page noted some specific requirements are in place for modifications. But he also noted that "the precise scope and meaning of the term 'modify' as applied to a PSD permit is not clear from the Part 124 regulations." Page Memorandum, p. 8. Sunflower

28 has not proposed changes or a modification to the permit, however, and Sierra Club has conceded we are not dealing with a modification. So these requirements do not come into play.

Rather, the remand proceedings are by direction of this court and KDHE. And Sierra Club I did not address the greenhouse gas regulations. For its part, KDHE may "reopen" a permit for cause in certain situations under K.S.A. 65-3008b(d) and "shall" reopen a permit if new air quality requirements become applicable within certain timeframes. These reopening provisions arguably would apply under the timeline of this case, but only as to new air quality limitations that would apply even after a permit has been issued. Here, the greenhouse gas regulations do not apply to permits issued before January 2, 2011, and KDHE was not compelled to reopen the permitting process.

Finally, the Page Memorandum concludes by stating: "[T]he EPA believes that a permitting authority has the discretion to evaluate on a case-by-case basis whether and under what circumstances it would be justified to issue a PSD permit extension without requiring the source to meet a new requirement that took effect during the term of the initial permit." Page Memorandum, p. 6. This conclusion supports KDHE's position that it could address the limited issues on remand without being compelled to require Sunflower to meet the greenhouse gas regulations.

In summary, Sierra Club has failed to cite any authority establishing that KDHE violated the law and has failed to meet its burden of establishing that KDHE erroneously interpreted or applied the law. See K.S.A. 2015 Supp. 77-621(c)(4).

29 1.4. Sierra Club fails to establish other grounds for relief.

Further, on this issue, Sierra Club has not pointed to any KDHE finding of fact that is not supported by substantial evidence. It has thus failed to establish a right to relief under K.S.A. 2015 Supp. 77-621(c)(7).

Finally, Sierra Club has failed to establish that KDHE's action was unreasonable, arbitrary, or capricious—the grounds for relief from agency action provided for in K.S.A. 2015 Supp. 77-621(c)(8). Rather, the statutes and regulations give KDHE discretion as the state plan administrator. Sierra Club's arguments rely in large part on policy considerations: Construction permits typically last 18 months because new technologies constantly emerge, and the EPA wants new facilities to adopt the latest technology. At this point, the stay and the permit extension in this case have resulted in a more than quadrupling of this 18-month period. One could thus conclude from the EPA's policy guidance that there is good reason for KDHE to engage in new substantive analyses on remand. But Sierra Club must do more than raise policy arguments.

Under the separation of powers doctrine, determination of the appropriate policy must be left to the legislative and executive branches of Kansas government. See State ex rel. Tomasic v. Unified Gov. of Wyandotte Co./Kansas City, 264 Kan. 293, 337-38, 955 P.2d 1136 (1998) ("Courts are limited to the exercise of judicial power in interpreting and applying the law and may not usurp the legislative power of determining policy matters or the executive power of implementing such policy."). Therefore, in order for us to grant Sierra Club's request for relief, Sierra Club must link its policy arguments to an erroneous KDHE action—that is, it must show that the greenhouse gas emission limitations and the accompanying best available control technology requirements apply

30 when a permit issued before January 2, 2011, has been remanded for further consideration. And it fails to do so.

Clearly, there exists a gap where the statutes and regulations are silent about the situation before us—the intersection of various provision of the federal Clean Air Act and Kansas provisions regarding the stay of administrative orders. But "courts are ill- equipped to fill such a gap. Often the wisest course is for courts to defer to the legislature to act to fill such a gap." State v. Keel, 302 Kan. 560, 573, 357 P.3d 251 (2015), cert. denied 136 S. Ct. 865 (2016); see Ft. Hays St. Univ. v. University Ch., Am. Ass'n of Univ. Profs., 290 Kan. 446, 464-65, 228 P.3d 403 (2010). And while courts do not afford significant deference to an administrative agency's statutory interpretation, Ft. Hays St. Univ., 290 Kan. at 457, where an agency possesses discretion, a court must presume the validity of the agency action and cannot substitute its judgment for that of the administrative agency unless the action is unlawful, unreasonable, arbitrary, or capricious. Pork Motel, Corp. v. Kansas Dept. of Health & Environment, 234 Kan. 374, 382, 673 P.2d 1126 (1983). Sierra Club fails to meet its burden to show KDHE's action fits that description.

In conclusion, we find no merit in Sierra Club's argument that KDHE erred in failing to determine greenhouse gas emission limitations based on the best available control technology.

ISSUE 2: KDHE DID NOT ERR IN SETTING 1-HOUR NITROGEN DIOXIDE AND SULFUR DIOXIDE EMISSION LIMITS.

We next consider an issue that was discussed in Sierra Club v. Moser, 298 Kan. 22, 310 P.3d 360 (2013) (Sierra Club I), specifically the application of EPA-promulgated regulations that became effective several months before the Holcomb 2 PSD permit was

31 issued. These regulations established 1-hour emission limits for nitrogen dioxide and sulfur dioxide. KDHE initially determined these new regulations did not apply to Sunflower's pending PSD permit because Kansas had not yet amended its state implementation plan to incorporate the regulations.

To provide the factual context for this argument, we first look back to the 2010 administrative proceeding. At that point, although KDHE did not impose limits consistent with the new federal regulations, it requested and received computer modeling from Sunflower. The information includes data reporting the emissions generated by Holcomb 1 and other air emissions sources in the region. The modeling projected instances in which pollutants in the ambient air in the region would exceed allowable limits.

Nevertheless, focusing on Holcomb 2's modeled emissions, KDHE concluded "[t]he results of the analyses indicated that for all modeled exceedances, the proposed project [Holcomb 2] contributes less than the [significant impact level] and therefore does not cause or contribute to any modeled exceedance." KDHE did, however, impose conditions on the 2010 permit that required Sunflower to notify KDHE if the total nitrogen oxides and sulfur dioxide emissions ever exceeded the pounds per hour amounts modeled in the permit application.

In the Sierra Club I action seeking judicial review, Sierra Club argued the permit needed to explicitly state emission limits rather than merely impose a requirement that required Sunflower to notify KDHE if emissions exceeded stated amounts. In the Sierra Club I decision, we agreed and held KDHE erred in its determination the hourly emission regulations did not apply. On that basis, we remanded the permit for application of the new regulations. 298 Kan. at 59.

32 On remand, KDHE did not request any new modeling. Instead, in the proposed Addendum, it incorporated specific 1-hour emission limitations identical to the amounts that triggered the 2010 notification requirements and that had been used in the 2010 modeling. Sierra Club requested more stringent 1-hour emission limits, but KDHE did not amend the 1-hour emission limitations, even after receiving updated modeling from Sunflower in the last days of the public comment period.

Then, in issuing the Summary Sheet that accompanied the Addendum, KDHE made clear that the nitrogen oxides and sulfur dioxide limits "contained in the Permit Addendum are . . . based on the 2010 modeling." And in the Responsiveness Summary, KDHE reiterated, in response to Sierra Club's comments, that the new 1-hour emissions limits were based on the 2010 modeling. According to KDHE, applying the permit limits to the 2010 dispersion modeling shows that Holcomb 2 "is not expected to cause or contribute to an exceedance" of the 1-hour emission limits.

In petitioning for review of the remand proceedings, Sierra Club claims Sunflower failed to demonstrate factually that emissions from Holcomb 2 will not cause or contribute to violations of any national ambient air quality standards. According to Sierra Club, modeling in the record shows just the opposite—i.e., Holcomb 2 will cause or contribute to thousands of violations of the 1-hour nitrogen dioxide and sulfur dioxide national ambient air quality standards. KDHE does not contest there were modeled 1-hour emission violations, but it and the Intervenors respond with a legal argument: Holcomb 2 will not "cause or contribute to" those modeled violations.

The question of whether Holcomb 2 will "cause or contribute to" a national ambient air quality standards violation arises from statutory language that prohibits the construction of any "major emitting facility," such as Holcomb 2, unless the owner or

33 operator, in this case Sunflower, demonstrates "that emissions from construction or operation of such facility will not cause, or contribute to, air pollution in excess of any . . . national ambient air quality standard in any air quality control region." (Emphasis added.) 42 U.S.C. § 7475(a)(3) (2012); see 40 C.F.R. § 52.21(k) (2016); K.A.R. 28-19-301(d) (2009).

In applying this standard, KDHE, along with the EPA, has historically set significant impact levels (SILs). In essence, SILs set numerical values for impact on ambient air quality below which the regulatory agency will consider a source to have a legally insignificant effect. In short, if the source's emissions are below the governing SIL, the source does not cause or contribute to the exceedance. Sierra Club I, 298 Kan. at 45-46; see Alabama Power Co v. Costle, 636 F.2d 323 (D.C. Cir. 1979) (seminal case discussing SILs).

KDHE has established SILs for 1-hour nitrogen dioxide and sulfur dioxide emissions and adopted a regulation providing that "[a] major source . . . shall be considered to cause or contribute to a violation of a national ambient air quality standard if the air quality impact of the source . . . would exceed the following levels at any locality . . . ." (Emphasis added.) K.A.R. 28-19-350(f)(1) (2015 Supp.). In a 2010 memorandum, KDHE explained:

"The EPA and KDHE consider a source whose individual impact falls below a SIL to have a de minimis impact on air quality concentrations.

....

34 ". . . If the [national ambient air quality standard] is exceeded, a new project has a significant contribution to a modeled [national ambient air quality standard] exceedance if its contribution exceeds the SIL at that receptor, day, and time . . . ."

In this case, applying the applicable SILs, KDHE recognized in the 2010 and 2014 Responsiveness Summaries that modeling reflected times when the nitrogen dioxide and sulfur dioxide in the air affected by the Holcomb station would exceed the national ambient air quality standards, but KDHE did not view Holcomb 2 as causing or contributing to those exceedances because the Holcomb 2 emissions would be below the SILs.

Sierra Club now challenges the legality of this approach, arguing the statutory language of the Clean Air Act bars any de minimis exemption and Sunflower must do more than show its emissions are below the SILs. In addition, Sierra Club argues that KDHE violated the Clean Air Act by failing to allow public comment on the 2014 modeling conducted by Sunflower and that the modeling made unlawful assumptions about emissions from Holcomb 1.

For purposes of our analysis, we begin with Sierra Club's argument about the illegality of KDHE's 2014 procedure—i.e., Sierra Club's argument the process needed to allow public comments on the 2014 modeling. We conclude Sierra Club fails to establish that KDHE erred in the way it handled the public comments. We then conclude Sierra Club failed to preserve the remaining arguments.

2.1. Sierra Club fails to establish that KDHE erred in its handling of public comments during the remand proceedings.

35 During the 2014 proceedings, KDHE provided notice of the draft Addendum, held a public hearing, and allowed a 30-day period for public comment. See K.S.A. 2015 Supp. 65-3008a(a) (mandating KDHE provide for public comment period before issuing, modifying, or reopening a permit); K.A.R. 28-19-350(k); K.A.R. 28-19-204; see also 42 U.S.C. § 7475(a)(2). Sierra Club submitted comments, as did Sunflower.

As relevant to the 1-hour nitrogen dioxide and sulfur dioxide emission limits, Sierra Club incorporated its 2010 comments, acknowledging that "[t]hese limits are based on modeling and analysis conducted in association with the 2010 permit." Sierra Club also asserted that KDHE "can and should include more stringent one-hour limits in the permit to protect against the many health risks posed by these pollutants." Finally, Sierra Club argued the lapse of time between the 2010 and 2014 proceedings meant a new analysis of best available control technology must be conducted.

In Sunflower's comments, which were submitted just a day or so before the end of the public comment period, Sunflower also referred to the 2010 modeling data. In addition, it submitted an analysis prepared by Shaw Environmental, Inc., dated in February 2014 that contained new modeling data.

In response to the public comments, KDHE noted that the new data had been received. It concluded: "The modeling results submitted indicate that there were modeled exceedances, however, for all modeled exceedances, the expansion project contributes less than the SIL and therefore does not cause or contribute to any [national ambient air quality standard] exceedance." On the same day, KDHE issued the Air Emission Source Construction Permit Addendum.

36 Sierra Club then filed a petition for reconsideration. It complained that KDHE had not made the February 2014 Shaw analysis public and did not offer the opportunity for public comment. Sierra Club continued:

"[I]t appears there are issues of both law and fact raised by KDHE's reliance on this new analysis and underlying data, including KDHE's improper reliance on significant impact levels (SILs) to excuse modeled exceedances of the one-hour [nitrogen dioxide] and [sulfur dioxide] standards and improper use of and conclusions drawn from emissions data from Holcomb 1."

KDHE denied the petition for reconsideration. In rejecting Sierra Club's assertion that KDHE had relied on the 2014 data, KDHE stated:

"KDHE did not rely on new analysis in the Permit Addendum decision. KDHE relied on the 2010 modeling analysis, which included the Holcomb 1 data and SILs developed pursuant to KDHE authority as an approved State under Section 110 of the Clean Air Act. KDHE responded to comments related to these issues in the December 16, 2010 Responsiveness Summary."

Sierra Club suggests KDHE needed to provide a period of public comment on the 2014 emission modeling. It further argues we should not accept KDHE's conclusion that it relied on the 2010 data in issuing the Addendum because KDHE noted the 2014 data in its Responsiveness Summary; KDHE even noted that the 2014 data substantiated its conclusion that Holcomb 2 would not cause or contribute to exceedances of the 1-hour emission limits.

As recognized by Sierra Club, KDHE clearly considered the 2014 data. Indeed, KDHE had to do so under K.A.R. 28-19-204 (2009), which requires the opportunity for public participation in the permitting process and mandates that all relevant comments

37 "be considered in making a final decision on the proposed permit action." Nevertheless, consideration does not equal reliance. Certainly, to rely on information means it must have been considered. See Webster's II New College Dictionary 937 (1995) (defining "rely" to mean "[t]o depend"). But information can be considered without depending— i.e., relying—upon it when making a determination. See Webster's II New College Dictionary 240 (defining "consider" as "[t]o think about seriously"; "to take into account"; "bear in mind"). For example, information may be deemed to be irrelevant or not persuasive. Thus, a decision can be made without relying on all information considered.

The entire 2014 administrative record verifies that KDHE considered the 2014 modeling but did not rely on the modeling in setting the nitrogen dioxide and sulfur dioxide emission limits. Indeed, KDHE's 2014 limits mirrored the emission levels which would have triggered the notice requirements in the 2010 permit. Furthermore, KDHE noted in the Addendum Summary Sheet that the nitrogen oxides and sulfur dioxide emission limits "contained in the Permit Addendum are . . . based on the 2010 modeling." Simply put, KDHE's overall approach to the Addendum reflects its reliance on the 2010 data—it did not view the Addendum as a new permit, it did not require a new permit application, and it did not require new modeling data.

Once again, KDHE's position that the 2014 proceedings did not relate to a new permit is significant to our consideration of Sierra Club's arguments regarding the timeliness and availability of data before public hearings. Sierra Club cites 42 U.S.C. § 7475(a)(2) (2012) and 42 U.S.C. § 7475(e)(2) (2012). Section 7475(a)(2) provides that a permit may not issue unless "the proposed permit has been subject to review . . . and a public hearing has been held with opportunity for interested persons . . . to appear and submit written or oral presentations on the air quality impact of such source, alternatives

38 thereto, control technology requirements, and other appropriate considerations." (Emphasis added.) And 42 U.S.C. § 7475(e)(2) sets a time limit for the data: "Such data shall be gathered over a period of one calendar year preceding the date of application for a permit under this part . . . . The results of such analysis shall be available at the time of the public hearing on the application for such permit." See 42 U.S.C. § 7475(e)(3)(C) (regulations must require that air quality analysis be available at the time of public hearing). But these provisions relate to a proposed permit.

As we have noted, in contrast, under K.S.A. 2015 Supp. 65-3008a(c) and K.A.R. 28-19-204 (2009), while the Secretary of KDHE must allow public comment, the Secretary otherwise has considerable discretion in handling the public comment period in post-permit situations, such as when modifying or reopening a permit. Sierra Club has not cited any authority imposing a timeframe for data collection for post-permit proceedings. Contrary to Sierra Club's position, nothing suggests that KDHE needed to expand the public comment period once Sunflower provided public comments which included the 2014 modeling data. And Sierra Club had ample opportunity to comment on the 2010 data upon which KDHE actually relied. In summary, Sierra Club fails to provide any authority indicating that KDHE erred in handling the public comment procedure in the context of post-permit proceedings related to what KDHE deemed to be a very limited remand.

In an alternative argument, Sierra Club argues KDHE acted arbitrarily when it failed to rely on the 2014 data, which Sierra Club suggests show a number of exceedances—meaning Holcomb 2 will cause or contribute to violations of the national ambient air quality standards. But, as we have described above, KDHE did not totally ignore the 2014 data. Instead, it noted the data and concluded: "The modeling results submitted indicate that there were modeled exceedances, however, for all modeled

39 exceedances, the expansion project contributes less than the SIL and therefore does not cause or contribute to any [national ambient air quality standard] exceedance." We next discuss the question of whether that conclusion was erroneous. But for purposes of considering error in the procedure utilized, we conclude Sierra Club fails to show that KDHE arbitrarily disregarded relevant information.

In conclusion, Sierra Club fails to establish that KDHE committed error in the procedures it followed on remand.

2.2. Sierra Club failed to preserve its argument regarding the legality of SILs.

Sierra Club presents a lengthy argument regarding the legality of KDHE's approach of using significant impact levels (SILs). For support, Sierra Club cites Sierra Club v. E.P.A., 705 F.3d 458 (D.C. Cir. 2013), which dealt with the SIL applicable to particulate matter; Sierra Club argues the decision establishes that both the EPA and a court have rejected reliance on SILs in the manner applied by KDHE in this case. KDHE counters, citing Utility Air Regulatory Group v. EPA, 573 U.S. ___, 134 S. Ct. 2427, 2449, 189 L. Ed. 2d 372 (2014). But KDHE also argues Sierra Club failed to preserve this argument, and the Intervenors add support to this contention. We agree that on this point Sierra Club failed to preserve its argument; Sierra Club failed to timely present the argument to KDHE—either in the 2010 or the 2014 administrative proceedings—and cannot assert it for the first time in a petition for reconsideration or a petition seeking review of an agency action.

In the 2010 administrative proceeding, Sierra Club, like KDHE, seemed to operate under the assumption that, even with modeled national ambient air quality standards violations, SILs could be used along with additional modeling to show that a source did

40 not cause or contribute to the violation. For example, during the public comment period, Sierra Club argued the numeric level of the SILs KDHE relied on had not been justified.

At the next stage of the 2010 proceedings, in its petition for judicial review, Sierra Club included an argument that could be read as the same argument it makes now. It argued in the 2010 petition that "the text of the Clean Air Act does not authorize a source to cause or contribute to any" exceedances of the national ambient air quality standards, "no matter how small the contribution." But Sierra Club abandoned this argument for purposes of the 2010 action by not briefing it in the 2010 proceedings before this court. Consequently, this court did not address the issue in Sierra Club I. See Superior Boiler Works, Inc. v. Kimball, 292 Kan. 885, 889, 259 P.3d 676 (2011) (an issue not briefed is abandoned); but see State v. Morton, 283 Kan. 464, 472-73, 153 P.3d 532 (2007) (in remand proceedings, law of the case doctrine did not bar reconsideration and reversal of the motion in limine at the new trial because issue was not addressed in first appeal).

In the remand proceedings, Sierra Club's public comments during the 2014 proceeding incorporated its 2010 comments. But Sierra Club did not comment on the legality of relying on SILs or of KDHE's method of considering SILs. Instead, the first mention of the arguments now raised came in Sierra Club's motion for reconsideration before KDHE. There, for the first time in any administrative proceeding, Sierra Club referred to new issues of law and fact arising from the 2014 data, "including KDHE's improper reliance on significant impact levels (SILs) to excuse modeled exceedances of the one-hour [nitrogen dioxide] and [sulfur dioxide] standards and improper use of and conclusions drawn from emissions data from Holcomb 1." Sierra Club did not cite any authority or provide any explanation of the basis for its argument. Cf. McCain Foods USA, Inc. v. Central Processors, Inc., 275 Kan. 1, 15, 61 P.3d 68 (2002) ("Simply pressing a point without pertinent authority, or without showing why it is sound despite a

41 lack of supporting authority, is akin to failing to brief an issue. 'Where the appellant fails to brief an issue, that issue is waived or abandoned.'"). As a result, the vague references without any supporting authority provided no indication of the specific argument that is now raised in this action.

KDHE merely denied the motion after stating it had not relied on the 2014 data. KDHE did not attempt to divine the specifics of the issues of fact and law referenced by Sierra Club.

The Intervenors suggest Sierra Club's arguments regarding the legality of KDHE's reliance on SILs came too late to preserve the argument for our review. We agree.

As a general rule, a party may not raise a new argument in a motion for reconsideration. But some courts recognize an exception when the arguments could not have been presented earlier. See 56 Am. Jur. 2d, Motions, Rules, and Orders § 40.

Similar rules bar raising new issues before a court when the issue has not been raised before an administrative agency. K.S.A. 2015 Supp. 77-617 limits when "[a] person may obtain judicial review of an issue that was not raised before the agency." One exception allows consideration of the issue if

"(d) the interests of justice would be served by judicial resolution of an issue arising from:

(1) A change in controlling law occurring after the agency action; or

(2) agency action occurring or first reasonably knowable to the person after the person exhausted the last feasible opportunity for seeking relief from the agency." K.S.A. 2015 Supp. 77-617.

42 Sierra Club invokes the exception and argues it raised the issue as soon as feasible—that is, once it knew of the 2014 data. See Supreme Court Rule 6.02(a)(5) (2015 Kan. Ct. R. Annot. 41) ("If the issue was not raised below, there must be an explanation why the issue is properly before the court.").

Yet, Sierra Club fails to explain why its legal arguments regarding KDHE's application of the SILs did not apply when it knew KDHE relied on the SILs to determine Holcomb 2's modeled emissions would not cause or contribute to the exceedances, using 2010 modeling. More simply, Sierra Club's alleged lack of access to the 2014 data does not explain why its legal argument, which did not turn on that data, could not have been raised sooner. For example, Sierra Club argues in its reply brief on this second round of judicial review that "both the 2010 modeling and 2014 modeling show the plant, as designed and proposed for operation, will cause unlawful exceedances of the [national ambient air quality standards]. These exceedances cannot properly be excused by the use of significant impact levels ('SILs')." Later, in the same brief, Sierra Club argued "[t]he 2010 modeling also predicts . . . exceedances" of the 1-hour national ambient air quality standards and "[a]llowing KDHE to rely on SILs to excuse these extensive violations of the [national ambient air quality standards] would undermine Congress' legislative design in the Act." And finally, Sierra Club asserted:

"KDHE incorrectly asserts that the many modeled [national ambient air quality standards] violations the air modeling from 2014 and 2010 reveal may all be categorically disregarded without any analysis just because the maximum contribution from Holcomb 2 to any [one of] these violations individually was less than a SIL threshold. KDHE Br. at 22-23. Such an automatic excusal of [national ambient air quality standards] violations is precisely the use of SILs that EPA and the D.C. Circuit have rejected."

43 Furthermore, the decision on which Sierra Club relies—Sierra Club v. E.P.A., 705 F.3d 458 (D.C. Cir. 2013)—was decided before Sierra Club made its comments during the 2014 administrative proceeding. And Sierra Club had recognized the argument in its 2010 petition for judicial review. Consequently, even at the point during the remand proceedings where Sierra Club knew only that KDHE had applied the SILs to conclude Holcomb 2 did not cause or contribute to the 2010 exceedances, it could have and should have made the argument it vaguely raised in the motion for reconsideration. Raising the issue in the motion for reconsideration was simply too late.

Moreover, Sierra Club's vague reference to the issue in the motion for reconsideration did not give KDHE an adequate opportunity to consider the question. The entire concept of judicial review contemplates that an agency must have had an adequate opportunity to consider the merits of an issue. See, e.g., United States v. L.A. Tucker Truck Lines, Inc., 344 U.S. 33, 37, 73 S. Ct. 67, 69, 97 L. Ed. 54 (1952); Nuclear Energy Institute, Inc. v. E.P.A., 373 F.3d 1251, 1297-98 (D.C. Cir. 2004); National Wildlife Federation v. E.P.A., 286 F.3d 554, 562 (D.C. Cir. 2002); National Ass'n of Mfrs. v. United States Dept. of Interior, 134 F.3d 1095, 1111 (D.C. Cir. 1998).

In essence, Sierra Club made an insufficient effort to raise the issue before KDHE. As a result, K.S.A. 2015 Supp. 77-617 prohibits Sierra Club from raising the issue before this court; there has not been a change of controlling authority, and Sierra Club could have raised the issue based on the 2010 data during the comment period. We thus conclude Sierra Club failed to preserve the issue for our review.

44 2.3. KDHE's assumptions about emissions from Holcomb 1 did not result in reversible error.

Finally, Sierra Club argues the modeling on which KDHE based its decision is flawed because it makes unlawful assumptions about emissions from Holcomb 1. Essentially, Sierra Club claims Sunflower erroneously modeled emissions from Holcomb 1 based on a 30-day average when it should have used the theoretical maximum 1-hour emissions based on Holcomb 1's design capacity. Alternatively, Sierra Club argues that if KDHE did not use 1-hour emission modeling data for Holcomb 1, it needed to impose 1-hour emission limits on Holcomb 1.

Sierra Club raised this issue in both the 2010 and 2014 administrative proceedings and in the Sierra Club I proceedings before this court, although we concluded we did not need to address the issue based on our holding. Sierra Club v. Moser, 298 Kan. 22, 58, 310 P.3d 360 (2013) (Sierra Club I). By repeatedly raising the issue, Sierra Club adequately preserved its arguments. Ultimately, however, we conclude that, even if [KDHE] error is assumed, we must consider it harmless because of Sierra Club's failure to preserve its previously discussed SILs-reliance argument. See K.S.A. 2015 Supp. 77- 621(e) (in conducting judicial review of administrative action, "due account shall be taken by the court of the rule of harmless error").

To explain this conclusion it helps to unravel the parties' arguments. In essence, Sierra Club's arguments on this point relate to modeling and control of emissions of Holcomb 1, not Holcomb 2. KDHE merely responded during the administrative proceeding by saying it did not have authority to change limits relating to Holcomb 1. And Sierra Club does not dispute this. Instead, before us, it focuses on the modeling of Holcomb 1 emissions and points to the federal EPA's comments about the need to have 1-hour emission modeling of Holcomb 1. In response, the Intervenors rely on a

45 memorandum issued by the EPA after its comments in this case. See Memorandum from Tyler Fox, Leader, Air Quality Modeling Grp., C439-0l, EPA, to Reg'l Air Div. Dirs., "Additional Clarification Regarding Application of Appendix W Modeling Guidance for the 1-hour N02 National Ambient Air Quality Standard" (Mar. 1, 2011) (Fox Memorandum). The Intervenors argue the guidance in the Fox Memorandum illustrates that KDHE had the discretion to allow the modeling as it was done here. But Sierra Club counters by disputing the applicability of the guidance in the memorandum.

We need not sort through the technical arguments about the Fox Memorandum, however. But we do note it does grant considerable discretion and makes allowances for application of SILs, stating, for example, that a cumulative analysis of ambient impacts will not be necessary if a facility can demonstrate its emission will not exceed the appropriate SILs. Nevertheless, we need not look beyond the authority cited by Sierra Club, 40 C.F.R. Part 51, App. W (2016) (Appendix W), which is titled "Guideline on Air Quality Models."

Section 3.0 of the Appendix is labeled "Recommended Air Quality Models." Sierra Club asserts these "guidelines" and "recommended" models amount to binding law in Kansas because Kansas incorporated them into its state implementation plan. See K.A.R. 28-19-350(d)(3) (2015 Supp.). But Sierra Club does not explain how Kansas' incorporation of nonbinding "recommendations" transforms those "recommendations" into legal mandates. Even from the EPA's guidance in the record it is clear that its recommended models are not mandatory. Moreover, the language in 40 C.F.R. Part 51, App. W is contrary to Sierra Club's absolutist position.

The language indicates the EPA never intended for the guidelines "to be a compendium of [all acceptable] modeling techniques." 40 C.F.R. Part 51, App. W,

46 (1.0)(a), (3.0)(d) ("It should not be construed that the preferred models identified here are to be permanently used to the exclusion of all others or that they are the only models available for relating emissions to air quality."). Indeed, "[t]he use of air quality measurements alone . . . could be preferable . . . when models are found to be unacceptable and monitoring data with sufficient spatial and temporal coverage are available." 40 C.F.R. Part 51, App. W(1.0)(b); see Sierra Club v. Wyoming Dept. of Environmental Quality, 2011 WY 42, ¶ 54, 251 P.3d 310 (2011) (modeling is the primary analytical tool, but it is not exclusive and not mandated in every instance).

Further, the EPA contemplates that it may be appropriate, depending on the case, to exercise discretion and rely on models it does not identify. See 40 C.F.R. Part 51, App. W(3.0)(b); 40 C.F.R. Part 51, App. W(3.1.2)(c); 40 C.F.R. Part 51, App. W(3.2.2)(b). Many variables—even unknown variables—affect the accuracy and certainty of modeling. See 40 C.F.R. Part 51, App. W(9.1.1). "[T]he process of conducting an air quality analysis is extremely complex." Andracsek & Lambrechts, Air Dispersion Modeling in the Major Source Construction Permitting Process, 28 Nat. Resources & Env't 29, 37 (2013). "Dispersion modeling is more of an art form than an actual science." 28 Nat. Resources & Env't at 30. It is up to the reviewing authority—here KDHE—to determine if a source has adequately demonstrated the appropriateness of its modeling. See 40 C.F.R. Part 51, App. W(8.2.3)(d). KDHE made that determination in this case.

These passages indicate KDHE has clear authority to, in its discretion, deviate from a recommended model given the inherent variability and complexity of dispersion modeling. Sierra Club fails to prove KDHE made a legal error by merely showing that KDHE did not rely on a "recommended" model. K.S.A. 2015 Supp. 77-621(c)(4).

47 Furthermore, KDHE had years of data before it and the data indicated there was a 99.9995% likelihood that the actual emission rate from Holcomb 1 is less than or equal to the modeled value. Given that and the other data in the record, we cannot conclude KDHE lacked a factual basis for reaching the conclusions it did about the modeling and for determining that the modeled emission rates for Holcomb 1 were representative and further modeling need not be performed. See Pork Motel, Corp. v. Kansas Dept. of Health & Environment, 234 Kan. 374, 382, 673 P.2d 1126 (1983) (where an agency possesses discretion, a court must presume the validity of the agency action and cannot substitute its judgment for that of the administrative agency unless the action is unlawful, unreasonable, arbitrary, or capricious). Sierra Club has failed to meet its burden.

Finally, as we previously noted, even if KDHE erred, we must consider it harmless. See K.S.A. 2015 Supp. 77-621(e) (in conducting judicial review of administrative action, "due account shall be taken by the court of the rule of harmless error"). Even if different modeling showed instances where the cumulative emissions of Holcomb 1, other background sources, and Holcomb 2 exceeded national ambient air quality standards, KDHE's reliance on the SILs to conclude Holcomb 2 did not cause or contribute to these exceedances would be unaffected. Therefore, Sierra Club's failure to preserve the SILs argument means we cannot find a basis for invalidating KDHE's action.

Consequently, we conclude that Sierra Club fails to establish a basis for invalidating KDHE's action because of the Holcomb 1 modeling.

ISSUE 3: THE 2014 ADDENDUM APPROPRIATELY ADDRESSES HAZARDOUS AIR POLLUTANTS EMISSION LIMITS.

Sierra Club's next issue relates to the Addendum's provisions regarding hazardous air pollutants emission limits. The Clean Air Act provides a list of hazardous air

48 pollutants. See 42 U.S.C. § 7412(b) (2012). And, with respect to hazardous air pollutants, the Clean Air Act requires:

"Emission standards promulgated under this subsection and applicable to new or existing sources of hazardous air pollutants shall require the maximum degree of reduction in emissions of the hazardous air pollutants subject to this section . . . that the Administrator, taking into consideration the cost of achieving such emission reduction, and any non-air quality health and environmental impacts and energy requirements, determines is achievable for new or existing sources . . . . 42 U.S.C. § 7412(d)(2).

The Clean Air Act also provides that "[t]he maximum degree of reduction in emissions that is deemed achievable for new sources in a category or subcategory shall not be less stringent than the emission control that is achieved in practice by the best controlled similar source, as determined by the Administrator." 42 U.S.C. § 7412(d)(3).

Under 40 C.F.R. § 63.9982(b) (2016), Holcomb 2 is considered a new source because construction will commence after May 3, 2011. Likewise, this court recognized in Sierra Club I that while these hazardous air pollutants emission limits were not in effect at the time KDHE issued the 2010 permit, these new limits would apply to Holcomb 2 on remand because the effective date keyed on the start of construction rather than the issuance of the permit. See Sierra Club I, 298 Kan. at 59-61. So after the remand, it was up to KDHE to ensure that Holcomb 2 would comply with the new hazardous air pollutants emission regulations.

The Clean Air Act provides a table that establishes the hazardous air pollutants emission limits depending on what type of electric utility steam generating unit is being constructed. See 40 C.F.R. Part 63, Subpart UUUUU, Table 1 (2016). Here, the Addendum issued on remand stated:

49 "The owner or operator shall comply with all applicable provisions of 40 CFR Part 63 Subpart UUUUU for an EGU as defined per 40 CFR 63.9985. Applicable portions of Emission and Operating Limits, Work Practice Standards, Performance Testing, Continuous Compliance, and Reporting sections from the 40 CFR Part 63 Subpart UUUUU in effect upon startup of H2 shall apply."

Now, in this appeal, Sierra Club argues KDHE failed to ensure compliance with the new hazardous air pollutants emission limits in 40 C.F.R. Part 63, Subpart UUUUU because KDHE merely stated that those new limits apply without requiring Sunflower to prove exactly how Holcomb 2 would achieve those limits. Sierra Club at times argues KDHE was required to conduct a pre-permit review and at other times a "pre- construction review" and "provide an[] analysis" demonstrating how the Mercury and Air Toxics Standards "will be met . . . by Holcomb 2" because Holcomb 2 will be a "major source" of hazardous air pollutants. Sierra Club claims that an applicant must prove exactly how it will comply with those new limits. At the core of Sierra Club's argument is a dispute over the degree of proof the owner of a proposed source must provide and a claim that Sunflower failed to provide enough. ("None of these vague, non-specific statements [about proposed control technologies] provide any design details or information of how either the older, less stringent limit or the newer, more stringent limit is to be met.")

This argument is unpersuasive for two reasons: (1) The hazardous air pollutants emission statutes anticipate it may take some time—even after construction—for a source to demonstrate compliance with limits, and (2) 42 U.S.C. § 7412(h)(1) indicates that setting enforceable limits is in most cases enough to determine compliance.

50 As to the first reason, Sierra Club references 42 U.S.C. § 7412(i)(1), claiming it requires—pre-permit—an applicant to "demonstrate with specificity the control technology that will be used to ensure that 'maximum achievable control technology' emission limits will be met." The overarching statute on which Sierra Club relies, section 7412(i)(1), provides:

"After the effective date of any emission standard, limitation, or regulation under subsection (d), (f) or (h) of this section, no person may construct any new major source or reconstruct any existing major source subject to such emission standard, regulation or limitation unless the Administrator (or a State with a permit program under subchapter V of this chapter) determines that such source, if properly constructed, reconstructed and operated, will comply with the standard, regulation or limitation."

See also 40 C.F.R. § 63.5(e)(1)(i) (2016).

Regulations issued pursuant to this provision provide that for a unit like Holcomb 2 the hazardous air pollutants emission limits are defined at 40 C.F.R. Part 63, Subpart UUUUU, Table 1. And importantly, subpart UUUUU deals with all coal-fired electric utility steam generating units, whether they are existing, new, or reconstructed units. See 40 C.F.R. § 63.9982(a)-(d). KDHE argues that nothing in the record establishes Holcomb 2 as a "major source," and the Intervenors note that the 2014 Addendum expressly sets the limits at the major source limits. See 40 C.F.R. § 63.2 (2016).

Significantly, in its reply brief, Sierra Club concedes that "whether Holcomb 2 is a 'major source' of hazardous air pollutants . . . would only become relevant if the [Mercury and Air Toxics Standards] rule is struck down and vacated" in litigation separate from this case. See Michigan v. EPA, 576 U.S. ___, 135 S. Ct. 2699, 2712, 192 L. Ed. 2d 674 (2015) (reversing EPA's interpretation of 42 U.S.C. § 7412[n][1][A] and requiring a consideration of cost when regulating power plants). Indeed, this court recognized in

51 Sierra Club I that whether Holcomb 2 was a "major" source was irrelevant because, regardless, the new hazardous air pollutants emission regulations would apply to Holcomb 2. 298 Kan. at 60. Given that those new hazardous air pollutants emission rules are still in force, the question of whether Holcomb 2 is a "major" source under the old law is currently moot under existing law and not ripe pending speculation regarding the resolution of the other litigation. See White Stallion Energy Center, LLC v. Environmental Protection Agency, No. 12-1100, 2015 WL 11051103 (D.C. Cir. 2015) (unpublished opinion) (order remanding the Mercury and Air Toxics Standards final rule to EPA without vacatur), cert. denied sub nom. Michigan v. EPA, 136 S. Ct. 2463 (2016); see also Shipe v. Public Wholesale Water Supply Dist. No. 25, 289 Kan. 160, 165-66, 210 P.3d 105 (2009) (Kansas courts do not consider issues unless the issues are ripe, meaning issues have "taken fixed and final shape rather than remaining nebulous and contingent.").

Instead, under the Mercury and Air Toxics Standards final rule, the category of source—existing, reconstructed, or new—determines when compliance with the hazardous air pollutants emission limits must be demonstrated. Specifically as to Holcomb 2, "[i]f your new . . . affected source commences construction . . . after April 16, 2012, you must demonstrate initial compliance with the promulgated emission limits no later than 180 days after startup of the source." (Emphasis added.) 40 C.F.R. § 63.10005(g)(2) (2016). While the permit application must include technical information about proposed emissions, emission controls, and estimated control efficiency under 40 C.F.R. § 63.5(d)(2), the Mercury and Air Toxics Standards envision that even after construction a source will have a period of time to refine its emission controls and demonstrate compliance.

52 As to the second reason Sierra Club's argument is unpersuasive, 42 U.S.C. § 7412(h)(1) seems contrary to the notion that the 2014 Addendum's enforceable emission limits are inadequate without specific design details.

"[I]f it is not feasible in the judgment of the Administrator to prescribe or enforce an emission standard for control of a hazardous air pollutant or pollutants, the Administrator may, in lieu thereof, promulgate a design, equipment, work practice, or operational standard, or combination thereof, which in the Administrator's judgment is consistent with the provisions of subsection (d) [emission standards] or (f) of this section." (Emphases added.) 42 U.S.C. § 7412(h)(1).

With respect to electric utility steam generating units like Holcomb 2, the Administrator determined it was feasible "to prescribe or enforce an emission standard for control of hazardous air . . . pollutants." See 40 C.F.R. Part 63, Subpart UUUUU, Table 1. And through its Addendum, KDHE imposed those hazardous air pollutants emission limits on Holcomb 2. If the prescription of those limits and their enforcement was inadequate, then the applicable statutes and regulations would identify a specific design, equipment, work practice, or operational standard with which Holcomb 2 must comply. Sierra Club has pointed to no such standards.

Moving past Sierra Club's legal argument that KDHE had to require Sunflower to explain with specificity the intricacies of its future pollution controls, Sierra Club also claims that substantial evidence does not support KDHE's finding that Holcomb 2 can meet the hazardous air pollutants emission limits. We disagree.

Sunflower submitted its proposed air pollution control technologies in 2010 in accordance with 40 C.F.R. § 63.5(d)(2), and, in creating the Addendum, KDHE accepted those same proposed air pollution control technologies as sufficient to meet the hazardous

53 air pollutants emission limits. In the remand proceedings, Sierra Club challenged Holcomb 2's ability to meet the emission limits. Basically, Sierra Club argues the original emission limits were higher in 2010; Holcomb 2 did not change any of its design plans to meet the new more stringent hazardous air pollutants emission limits in place under the 2014 Addendum; and, because there were no design changes, Holcomb 2 cannot meet the newer, stricter limits.

Granted, the 2014 hazardous air pollutants emission limits are stricter. But contrary to Sierra Club's claim, that does not necessarily mean the original design of Holcomb 2 could not also meet those lower limits (and this court in Sierra Club I did not conclude the original design could not possibly comply with new hazardous air pollutants emission limits; this court simply held those new limits would apply on remand). Sierra Club I, 298 Kan. at 60-61. During remand, KDHE found that

"[t]he control technology in the December 10, 2010[,] permit was, and still is, the technology that is used in PSD pollutants and hazardous air pollutants (HAPs) control alike. . . .

"The most effective control for mercury is a combination of fuel blending, selective catalytic reduction (SCR) to oxidize the gaseous mercury, carbon or sorbent injection to absorb the mercury and fabric filter (FF) or electrostatic precipitator to capture the absorbed mercury. The December 16, 2010 permit requires an SCR be used to control [nitrogen oxides] and an FF be used to control [particulate matter], lead and [sulfuric acid]. The permit states that emission limits will be met by blending various coals, or by the injection of powdered activated carbon (PAC) or other sorbent or both. PAC or sorbent injection equipment will be installed for the [Holcomb 2] steam generator."

KDHE accepted Sunflower's original design plan for Holcomb 2 as sufficient to meet the new hazardous air pollutants emission limits, and evidence in the record

54 supported that finding. Moreover, during the administrative proceedings, KDHE addressed the specific focus of Sierra Club's comments relating to hazardous air pollutants—that is, the standards for mercury, acid gases, and other toxic emissions from all coal- and oil-fired electric utility steam generating units of more than 25 megawatts. See 40 C.F.R. Part 63.10042 (2016); 77 Fed. Reg. 9304 (February 16, 2012) (EPA's rule known as the "Mercury and Air Toxics Standards"). KDHE noted: "Sunflower has provided information that the mercury control vendors will guarantee their control technology to meet the [Mercury and Air Toxics Standards] requirements for [Holcomb 2] before Sunflower will contractually obligate to the equipment."

In assessing these findings, this court does not reweigh the evidence or engage in de novo review. See K.S.A. 2015 Supp. 77-621(d). Thus, Sierra Club's attempt to point to conflicting evidence in the record, or to come up with hypothetical unanswered questions about how exactly Holcomb 2 will meet emission limits, is unavailing.

Alternatively, Sierra Club makes several other arguments. First, it argues the 2014 Addendum must specifically state the limits in the Mercury and Air Toxics Standards verbatim rather than by incorporating them by reference. But Sierra Club does not cite any legal authority and we find nothing in the Kansas state implementation plan or Kansas regulations to support that argument. See McCain Foods USA, Inc. v. Central Processors, Inc., 275 Kan. 1, 15, 61 P.3d 68 (2002) ("Simply pressing a point without pertinent authority, or without showing why it is sound despite a lack of supporting authority, is akin to failing to brief an issue. 'Where the appellant fails to brief an issue, that issue is waived or abandoned.'").

Second, Sierra Club argues KDHE needed to replace the older hazardous air pollutants emission limits in the 2010 permit because the Addendum does not specifically

55 say the 2010 limits are void. During remand, KDHE concluded that the more stringent limits of the 2014 Addendum superseded any lesser limits from the original permit. The impact of a written instrument is determined by giving words in the instrument their plain and ordinary meaning. See Amoco Production Co. v. Wilson, Inc., 266 Kan. 1084, 1088- 89, 976 P.2d 941(1999). The Addendum states: "Except as specified in this Addendum, all provisions of the permit issued December 16, 2010[,] remain in effect." (Emphasis added.) While the Addendum might phrase it in the converse, the Addendum clearly imparts that the new hazardous air pollutants emission limits render void any conflicting provisions from the 2010 permit.

In summary, we conclude that Sierra Club has failed to establish that KDHE erred in addressing hazardous air pollutants.

ISSUE 4: NEW SOURCE PERFORMANCE STANDARDS ARE NOT PROPERLY BEFORE THIS COURT.

Additionally—for the first time in its reply brief—Sierra Club argues Holcomb 2 cannot meet new source performance standards (NSPS). The EPA issued a final rule, effective October 23, 2015, regarding the NSPS for electric utility generating units like Holcomb 2. See 80 Fed. Reg. 64510 (October 23, 2015). Any source that commences construction after January 8, 2014, will be required to meet the NSPS. 80 Fed. Reg. 64510, p. 64512. And the EPA specifically mentioned Holcomb 2, noting that "it appears [Holcomb 2 has] not commenced construction for purposes of section 111(b) and therefore would likely be [a] new source[] should [it] actually be constructed." 80 Fed. Reg. 64510, p. 64543.

"An appellant may not raise new issues in a reply brief." State v. McCullough, 293 Kan. 970, 984, 270 P.3d 1142 (2012); see Supreme Court Rule 6.05 (2015 Kan. Ct. R.

56 Annot. 49). Thus, we will not consider this argument. Nevertheless, we note Sierra Club's concession that Holcomb 2 still needs federal approval before commencing construction. So whether Holcomb 2 can comply with the NSPS will be dealt with on another day.

CONCLUSION

Sierra Club failed to establish KDHE erred in adding an addendum to Holcomb 2's 2010 permit. It also failed to establish KDHE erred by not including best available control technology requirements for greenhouse gas emissions, in setting the 1-hour nitrogen dioxide and sulfur dioxide emission limits, in applying the hazardous air pollutants emission regulations and the Mercury and Air Toxics Standards, or in addressing the NSPS. We, therefore, conclude Sierra Club has failed to establish a basis for invalidating Sunflower's PSD permit.

Affirmed.

JOHNSON AND STEGALL, JJ., not participating. MICHAEL J. MALONE, Senior Judge, assigned.1 STEVEN L. HORNBAKER, District Judge, assigned.2

1 REPORTER'S NOTE: Senior Judge Malone was appointed to hear case No. 112,008 vice Justice Stegall under the authority vested in the Supreme Court by K.S.A. 20-2616. 2 REPORTER'S NOTE: District Judge Hornbaker was appointed to hear case No. 112,008 vice Justice Johnson under the authority vested in the Supreme Court by art. 3, § 6(f) of the Kansas Constitution.

57 ***

HORNBAKER, J., concurring: This case is not a quibble over a few stray particles. Sunflower Electric Power Corporation is seeking to construct the largest coal-fired electric generating unit in the United States. The proposed Holcomb 2 plant will— unquestionably—spew, each year, billions of pounds of pollutants into the air of Kansas. That, according to Sunflower's modeling in the record, will leave Kansans thousands of times a year with air that is more hazardous to their health than that allowed by the National Ambient Air Quality Standards (NAAQS).

In Sierra Club v. Moser, 298 Kan. 22, 57-61, 310 P.3d 360 (2013), after a thorough review, this court ordered the Kansas Department of Health and Environment (KDHE) to apply the Environmental Protection Agency's (EPA) newly adopted regulations providing limits for 1-hour emissions and hazardous air pollutants (HAPs). KDHE purported to comply with that mandate by creating an "Addendum." Basically, the addendum addressed this court's mandate by stating the following: 1-hour emission limits apply, and so do HAPs limits. I agree with the majority that Sierra Club failed in its burden to articulate how KDHE erred in issuing the Addendum. I also agree with the majority that Sierra Club failed to preserve its argument that Significant Impact Levels (SILs) cannot be used to excuse violations of the NAAQS.

KDHE and Sunflower relied on SILs to excuse these exceedances. I, however, see no legal basis by which a SIL could excuse an NAAQS violation. Sierra Club first thought to mention the issue in a two-page petition for reconsideration before KDHE, leaving our hands procedurally bound and unable to address the issue. See Kingsley v. Kansas Dept. of Revenue, 288 Kan. 390, 410, 204 P.3d 562 (2009) (failure to raise an issue before an administrative agency bars reviewing court from deciding that issue).

58 Were the court able to reach the merits of the SILs issue, I would hold that the plain language of the Clean Air Act (CAA) does not permit SILs to be used as they traditionally have been.

The owner or operator of a major emitting facility must demonstrate "that emissions from construction or operation of such facility will not cause, or contribute to, air pollution in excess of any [NAAQS]." 42 U.S.C. § 7475(a)(3) (2012). That seems plain enough to me, and we must adhere to the plain meaning of a statute. See Cady v. Schroll, 298 Kan. 731, 738-39, 317 P.3d 90 (2014). If the air quality in an area is compliant with the NAAQS, and a facility's construction would make the air quality violate any of the NAAQS, then federal law prohibits construction because construction would cause the exceedance. Or, if air quality already violates any of the NAAQS, then a new facility cannot be constructed because the new facility would contribute to the exceedance. Even Alabama Power Co. v. Costle, 636 F.2d 323, 360-61 (D.C. Cir. 1979), which recognized de minimis authority, held that such authority was not available when "Congress has been extraordinarily rigid." Congress could not have been more rigid— facilities cannot "cause, or contribute to, air pollution in excess of any [NAAQS]." 42 U.S.C. § 7475(a)(3). There is simply no provision suggesting that emissions could cause or contribute as long as the emissions are below a SIL.

In addition, Sierra Club argues on this appeal that a recent case makes reliance on SILs questionable. From my reading of 42 U.S.C. § 7475(a)(3), I agree. In Sierra Club v. E.P.A., 705 F.3d 458 (D.C. Cir. 2013), Sierra Club challenged the EPA's authority to utilize SILs in any case. In response, the EPA acknowledged that certain codifications of its SIL provisions regarding particulate matter less than 2.5 micrometers (PM2.5) were flawed. So the EPA asked the court to vacate those provisions and remand for the EPA to promulgate new SIL regulations. 705 F.3d at 463-64.

59 I understand SILs have long been used as a method of establishing that a source does not cause or contribute to an NAAQS violation. Moreover, it is true that Sierra Club v. E.P.A. did not ultimately reject this longstanding use. 705 F.3d at 464 (noting that the court would address "the universal disallowance of all de minimis authority" only if the EPA promulgated new SILs on remand relying on that authority). But after Sierra Club v. E.P.A., the EPA eliminated the SIL for PM2.5 and has yet to replace it. See 78 Fed. Reg. 73698, p. 73699 (December 9, 2013) (removing the PM2.5 SILs from its regulations).

KDHE claims that the United States Supreme Court rejected Sierra Club's argument regarding the use of SILs in Utility Air Regulatory Group v. EPA, 573 U.S. ___, 134 S. Ct. 2427, 189 L. Ed. 2d 372 (2014). Certainly, the Court did recognize the EPA's authority to characterize certain emissions as de minimis under Alabama Power. Utility Air Regulatory Group, 134 S. Ct. at 2435 n.1 ("EPA . . . has established pollutant- specific numerical thresholds below which a facility's emissions of a pollutant, and increase therein, are considered de minimis for those purposes.") (citing Alabama Power, 636 F.2d at 360-61, 400, 405). But the Court was simply not considering the issue here— whether a SIL could be used to authorize a source to cause or contribute to an NAAQS violation. The Court was considering whether greenhouse gas emissions triggered extensive Prevention of Significant Deterioration (PSD) and Title V requirements, and whether a source already regulated under the PSD program had to employ the best available control technologies for greenhouse gas emissions. 134 S. Ct. at 2438.

The Court in Utility Air Regulatory Group reaffirmed "the core administrative-law principle that an agency may not rewrite clear statutory terms to suit its own sense of how the statute should operate." 134 S. Ct. at 2446. Again, the statute is not equivocal in its prohibition of NAAQS violations: "[T]he owner or operator of such facility [must]

60 demonstrate[], as required pursuant to section 7410(j) of this title, that emissions from construction or operation of such facility will not cause, or contribute to, air pollution in excess of any [NAAQS.]" 42 U.S.C. § 7475(a)(3); see also Sierra Club v. E.P.A. 705 F.3d at 464 ("On remand the EPA may promulgate regulations that do not include SILs or do include SILs that do not allow the construction or modification of a source to evade the requirements of the Act."); Alabama Power, 636 F.2d at 360-61 (reliance on de minimis authority is improper if Congress has been "extraordinarily rigid"). The EPA conceded in Sierra Club v. E.P.A. that it never intended for a SIL to permit a CAA violation. The traditional view of SILs (as they were used in issuing the permit in question) permits that very thing, which amounts to an impermissible rewriting of clear statutory terms by courts and administrative agencies.

Despite my opinion on the merits of the SILs issue, as the majority holds, Sierra Club unfortunately neglected to raise this argument until this appeal. And because the argument's premise rests on statutory interpretation, Sierra Club had no reason to wait this long. Kingsley, 288 Kan. at 410 (failure to raise an issue before an administrative agency bars reviewing court from deciding that issue). Thus I reluctantly concur in the judgment affirming KDHE's decision to issue an Addendum to the 2010 Holcomb 2 permit.