Searching over 5,500,000 cases.


searching
Buy This Entire Record For $7.95

Download the entire decision to receive the complete text, official citation,
docket number, dissents and concurrences, and footnotes for this case.

Learn more about what you receive with purchase of this case.

Sierra Club v. United States Environmental Protection Agency

United States Court of Appeals, Fifth Circuit

October 3, 2019

SIERRA CLUB; NATIONAL PARKS CONSERVATION ASSOCIATION; ENTERGY LOUISIANA, L.L.C., Petitioners
v.
UNITED STATES ENVIRONMENTAL PROTECTION AGENCY; ANDREW WHEELER, Administrator, United States Environmental Protection Agency, Respondents

          On Petitions for Review of Final Administrative Action of the United States Environmental Protection Agency

          Before SMITH, WIENER, and ELROD, Circuit Judges.

          WIENER, CIRCUIT JUDGE

         In a December 21, 2017 Final Rule ("the Final Rule"), the United States Environmental Protection Agency ("the EPA") approved Louisiana's state implementation plan ("SIP") for controlling regional haze. Louisiana's regional haze SIP had two alleged problems. First, the SIP used an outdated air-pollution model called "CALPUFF" to measure the visibility impacts of powerplant emissions. Second, the SIP included a sparse explanation for how Louisiana weighed five mandatory statutory factors in determining the Best Available Retrofit Technology ("BART") for controlling emissions at Unit 6 of the Roy S. Nelson powerplant ("Nelson"). Despite the EPA's knowledge of these problems, it determined that Louisiana had fulfilled its obligations under the Clean Air Act.

         We consider two petitions for review of the Final Rule. One is from Petitioners-Appellants Sierra Club and National Parks Conservation Association (collectively, "Environmental Petitioners"). The other is from Petitioner-Appellant Entergy Louisiana, L.L.C., the owner of the Nelson powerplant, and Cleco Power, L.L.C., an intervenor in this case (collectively, "Industry Petitioners").

         Environmental Petitioners maintain that Louisiana's SIP does too little to curb regional haze at federally protected areas. They contend that (1) Louisiana's determination that "low-sulfur coal" was the BART for the Nelson powerplant was deficient in several respects and (2) the EPA acted arbitrarily and capriciously in approving Louisiana's SIP because it knew about those deficiencies.

         In contrast, Industry Petitioners insist that Louisiana's SIP overestimates the amount of pollution that their powerplants produce. In their challenge to the EPA's approval of Louisiana's "subject to BART" determinations, Industry Petitioners object to Louisiana's and the EPA's use of the "CALPUFF" model, which they maintain relies on several flawed technical assumptions.

         We deny Industry Petitioners' petition. We afford "significant deference" to agency decisions involving analysis of scientific data within the agency's technical expertise. The EPA's selection of a model to measure air pollution levels is precisely that type of decision. The EPA therefore did not act arbitrarily and capriciously in relying on the CALPUFF model to approve Louisiana's "subject to BART" determinations.

         Although Environmental Petitioners' challenge presents a closer question, we deny that petition as well. Louisiana's explanation of its BART determination for Nelson omitted two of the five mandatory factors and failed to compare-or even set out-the numbers for the costs and benefits of the control options Louisiana considered. Louisiana also failed to explain how its decision accounted for the EPA-submitted analyses that pointed out substantial flaws in other analyses in the administrative record. But applying the deferential standards of the Administrative Procedures Act to the facts of this case, we hold that the EPA's approval of Louisiana's SIP was not arbitrary and capricious.

         The petitions for review are denied.

         I. Background

         This case addresses the EPA's approval of Louisiana's SIP for controlling regional haze. The Clean Air Act "requires the states and the federal government to set and seek to achieve targets for visibility in protected national parks and wildlife areas by modifying regulations that control air pollutants in ambient air."[1] Under the Act, the federal government identifies air pollutants and sets standards, and the states have "the primary responsibility" for implementing those standards through SIPs.[2] After a state submits its SIP, the EPA reviews the SIP for compliance with the Clean Air Act.

         Powerplants that emit sulfur dioxide ("SO2") and oxides of nitrogen ("NOx") contribute to regional haze in protected federal areas.[3] Louisiana has five powerplants that cause or contribute to visibility impairments in such areas.[4] Since 2008, Louisiana has revised its SIP several times and established emission controls at some of those powerplants. In October 2017, Louisiana submitted its final SIP revisions, which addressed, inter alia, emission controls at Unit 6 of the Nelson powerplant.

         On December 21, 2017, the EPA promulgated a final rule approving Louisiana's SIP. 82 Fed. Reg. 60, 520 (Dec. 21, 2017) ("the Final Rule"). The two petitions for review of the Final Rule address the EPA's approval of (1) Louisiana's determination that Nelson and two units at the Cleco-owned Brame Energy Center ("Brame") are subject to BART for controlling emissions, (2) Louisiana's selection of low-sulfur coal as BART for controlling SO2 emissions at Nelson, and (3) Louisiana's reasonable progress goals and long-term strategy.

         Environmental Petitioners first address Louisiana's BART determination for Nelson. They contend that the Louisiana Department of Environmental Quality ("LDEQ") erred in three ways: (1) determining that low-sulfur coal was BART for Nelson, despite the EPA-submitted analyses that contradicted the analyses in the record, (2) failing to provide a rational basis for rejecting a more effective pollution control, and (3) not complying with the BART guidelines or considering all the mandatory BART factors.

         Environmental Petitioners next object to the Final Rule's approval of Louisiana's long-term strategy and reasonable progress goals. They maintain that Louisiana's 2017 SIP revisions did not fulfill the state's obligations to revise and resubmit its long-term strategy and reasonable progress goals after the EPA disapproved that strategy and those goals in 2012. Louisiana's 2017 SIP revisions did not impose additional controls at non-BART "reasonable progress" sources and did not address the state's reasonable progress goals or long-term strategy for achieving natural visibility conditions. According to Environmental Petitioners, the EPA improperly overlooked these omissions when it approved Louisiana's SIP.

         Environmental Petitioners object to the type of BART control Louisiana implemented. Industry Petitioners, in contrast, object to Louisiana's determination that Nelson and Brame are subject to BART at all. Industry Petitioners challenge the technical assumptions underlying the modeling methods on which Louisiana and the EPA relied. Louisiana relied on the "CALPUFF" model, and the EPA relied on both the CALPUFF and the "CAMx" models. According to Industry Petitioners, the CALPUFF model overstates the visibility effects of powerplant emissions. They also maintain that the EPA exceeded its authority by using its own CAMx modeling to support Louisiana's modeling results.

         Each petitioner in this case is also a respondent-intervenor. The Environmental Petitioners, in addition to petitioning for review of some parts of the Final Rule, also intervened and filed a responsive brief opposing Industry Petitioners' petition. Likewise, Industry Petitioners, in addition to petitioning for review of some parts of the Final Rule, also intervened and filed a responsive brief opposing Environmental Petitioners' petition.

         The EPA responded to both petitions, insisting that the Final Rule should be approved in full. The LDEQ filed an amicus brief supporting the EPA's position.

         A. Statutory and Regulatory Framework

         This case is governed by the Clean Air Act and the regulations implementing it. In 1977, "in response to a growing awareness that visibility was rapidly deteriorating in many places, such as wilderness areas and national parks," Congress amended the Act by enacting § 169A.[5]

         That amendment established as a national goal "the prevention of any future, and the remedying of any existing, impairment in visibility in mandatory class I Federal areas which impairment results from manmade air pollution."[6] Protected class I Federal areas include "all (1) international parks, (2) national wilderness areas which exceed 5, 000 acres in size, (3) national memorial parks which exceed 5, 000 acres in size, and (4) national parks which exceed 6, 000 acres in size."[7] Louisiana's Breton National Wildlife Refuge and the Caney Creek Wilderness Area in southwest Arkansas are the protected Class I Federal areas at issue here.[8]

         The Clean Air Act is "an experiment in cooperative federalism," in which the federal government identifies pollutants and sets visibility targets, and the states implement those standards though SIPs.[9] The Act directed the EPA to issue regulations requiring: (1) that states submit SIPs to the EPA, (2) the installation of the "best available retrofit technology, as determined by the State . . . for controlling emissions" at specified air-pollution sources, and (3) that each state adopt a long-term strategy "for making reasonable progress" toward the national visibility goal.[10]

         Based on the Act's directive, the EPA promulgated the Regional Haze Rule in 1999.[11] "The Regional Haze Rule established the guidelines for state compliance with the air visibility requirements of [the Clean Air Act]."[12] In 2005, the EPA revised that Rule and issued the BART guidelines, which set out the process for states to establish BART emissions limitations.[13]

         The Clean Air Act and the Regional Haze Rule impose five requirements for SIPs:

For each affected wilderness and national park, the plan must: (1) set "reasonable progress goals" toward achieving natural visibility conditions that ensure improvements in visibility on the most impaired days over the period of the implementation plan; (2) calculate baseline visibility and natural visibility conditions; (3) devise a long-term strategy with enforceable emissions limitations, compliance schedules, and other measures necessary to achieve the reasonable progress goals; (4) develop a monitoring strategy for measuring and reporting visibility; and (5) list the best available retrofit technology . . . that emission sources in the state will have to adopt to achieve the visibility goals, along with a schedule for implementing BART.[14]

         After a state submits its SIP to the EPA, the agency reviews the SIP for compliance with the Clean Air Act. If the EPA determines that a SIP does not comply with the Act, it must promulgate a "Federal implementation plan" that fixes the SIP's shortcomings, unless the state corrects the deficiency.[15] The EPA's role is confined "to the ministerial function of reviewing SIPs for consistency with the Act's requirements."[16]

         The parties focus on two of the Clean Air Act's and Regional Haze Rule's requirements: (1) the BART emission limits and (2) the reasonable progress goals and long-term strategy that each state must implement.

         1. BART Emission Limits

         The Regional Haze Rule defines BART as "an emission limitation based on the degree of reduction achievable through the application of the best system of continuous emission reduction for each pollutant which is emitted by an existing stationary facility."[17] The BART process has three steps. First, a state must identify all "BART-eligible sources." Second, it must determine which of those BART-eligible sources are "subject to BART." Third, for each source that is "subject to BART," the state must make a "BART determination" by analyzing and selecting the appropriate emission control for that source.[18]

         Step one is identifying all "BART-eligible sources." That definition includes all stationary facilities that (1) were in existence before August 7, 1977 but were not in operation before August 7, 1962, (2) have "the potential to emit 250 tons per year or more of any visibility-impairing air pollutant," and (3) fall within one of 26 listed source categories.[19] The parties agree that Nelson and Brame satisfy these requirements and are "BART-eligible" sources.

         Step two is making "subject to BART" determinations. The state must determine which of the BART-eligible sources emit air pollutants that "may reasonably be anticipated to cause or contribute to any impairment of visibility" in a Class I area.[20] The key words are "cause" and "contribute to." Sources that a state determines may reasonably be anticipated to "cause" or "contribute to" visibility impairment at a protected area are "subject to BART."[21] Under the BART Guidelines, a 1.0 deciview[22] change from an individual source "causes" visibility impairment, whereas a 0.5 deciview change from an individual source "contributes to" impairment.[23] In some instances, states may set a lower threshold for sources that "contribute to" visibility impairment.[24]

         At the "subject to BART" screening step, states have discretion to either (a) determine that all BART-eligible sources are "subject to BART" via "collective attribution" or (b) conduct individualized testing to determine that a source, or a group of sources, is exempt from BART.[25] Here, Louisiana opted for the second option.

         The third step is the "BART determination," in which a state must identify the appropriate emission control for each "subject to BART" source.[26]At each such source, the state must consider five statutory factors: (1) the costs of compliance, (2) the energy and nonair quality environmental impacts of compliance, (3) any existing pollution control technology in use at the source, (4) the remaining useful life of the source, and (5) the degree of improvement in visibility which may reasonably be anticipated to result from the use of such technology.[27]

         For powerplants with generating capacities greater than 750 megawatts, like those at issue here, a state's BART determination must comply with the BART guidelines, [28] which provide step-by-step instructions for making BART determinations.[29] In contrast, at the BART-eligibility and "subject to BART" steps, the BART guidelines are advisory only.[30]

         2. Reasonable Progress Goals and Long-Term Strategy

         The Regional Haze Rule requires a state to "establish goals (expressed in deciviews) that provide for reasonable progress towards achieving natural visibility conditions" for each Class I area within that state.[31] The process and considerations for calculating these goals are as follows:

A state begins by calculating the steady linear rate of decreasing emissions that would achieve natural visibility in the covered wildernesses and national parks by the year 2064. 40 C.F.R. § 51.308. If a state determines that the linear rate would result in unreasonable regulations, it must propose an alternative set of reasonable progress goals and demonstrate both that the linear rate is unreasonable and that the alternative goals are reasonable. Id. § 51.308(d)(1)(ii). The Clean Air Act and the Regional Haze Rule require a state to consider four factors when setting reasonable progress goals: "the costs of compliance, the time necessary for compliance, and the energy and nonair quality environmental impacts of compliance, and the remaining useful life of any existing source subject to such requirements." 42 U.S.C. § 7491(g)(1); see also 40 C.F.R. § 51.308(d)(1)(i)(A) (repeating the factors listed in § 7491(g)(1)).[32]

         The evaluation of the four statutory factors is often referred to as a "'four-factor analysis' or 'reasonable progress analysis.'"[33]

         In addition to setting reasonable progress goals, a state must establish a "long-term strategy" to make reasonable progress toward achieving natural visibility conditions.[34] This strategy must include "enforceable emissions limitations, compliance schedules, and other measures necessary to achieve the reasonable progress goals."[35] As part of its long-term strategy, a state may impose additional emission reduction measures at sources that are not "subject to BART."[36] Sources that are regulated through the reasonable progress goals-rather than through BART-are referred to as "reasonable progress sources."[37]

         B. Factual Background

         The procedural history of this case began when Louisiana submitted its regional haze SIP to the EPA in 2008. Since then, Louisiana has revised and resubmitted its SIP many times. Louisiana submitted its final SIP revisions in October 2017, and the EPA approved Louisiana's SIP in its December 2017 Final Rule.[38]

         1. Louisiana's 2008 Submittal

         Louisiana submitted a SIP for addressing regional haze in June 2008. In that SIP, Louisiana relied on an emissions-trading program called the Clean Air Interstate Rule ("CAIR") to satisfy its BART, reasonable-progress, and long-term strategy obligations. The EPA reviewed that SIP in two separate actions in 2012.[39]

         The first such action, a national rulemaking, was based on a D.C. Circuit decision that (1) held that states could not rely on CAIR to satisfy their BART obligations and (2) remanded CAIR to the EPA.[40] In that national rulemaking, the EPA finalized a "limited disapproval" of Louisiana's SIP because it had relied on CAIR.[41]

         The second such action "partially disapproved" Louisiana's SIP submission based on issues other than, and "go[ing beyond]," the SIP's reliance on CAIR. In addition to this "partial disapproval" of particular parts of Louisiana's SIP, that action also included a "partial limited approval" of other parts of the SIP.

         The "partial limited approval" part of that action was based on the EPA's conclusion that specified parts of the SIP, "as a whole, strengthen[ed] the State's SIP."[42] Although the EPA concluded that some of the SIP provisions- including Louisiana's reasonable progress goals and long-term strategy-did not comply with the Clean Air Act, those noncompliant provisions were included in the partial limited approval.[43]

         In the "partial disapproval" part of that action, the EPA disapproved the parts of the SIP related to CAIR, including the BART analyses for non-electric generating unit ("EGU") sources[44] and the BART determinations for four specific non-EGU sources.[45] Those deficiencies are not at issue here.

         A definitional note: Limited approvals and partial approvals are different. If a submittal does not meet all of the Act's requirements, but a "separable" part does meet those requirements, a "partial approval may be used to approve that part of the submittal and disapprove the remainder."[46] In contrast, if parts of a submittal do not meet the Clean Air Act's requirements, but "the submittal as a whole" has a "strengthening effect" on the SIP, the EPA may use a "limited approval" to enact the entire submittal.[47] The practical difference is

that under a limited approval the EPA's approval action goes to the entire [submittal]. In other words, although portions of a [submittal] prevent the EPA from finding that the [submittal] meets a certain requirement of the Act, the EPA believes that the [submittal], as a whole, strengthens the SIP. Therefore, the EPA approves the entire [submittal]--even those portions that prohibit full approval. Likewise, when the EPA issues the limited disapproval, the disapproval applies to the entire [submittal] as failing to meet a specific requirement of the Act. The [submittal] remains a part of the [enforceable] SIP, however, under the limited disapproval, because the [submittal] strengthens the SIP. The disapproval only applies to whether the submittal meets a specific requirement of the Act and does not affect incorporation of the rule into the approved, federally enforceable SIP.[48]

         In short, in a partial approval, the approved parts of the submittal go into effect and the disapproved parts do not. In a limited approval, the entire part that is "limitedly" approved-including the provisions within that part that do not comply with the Clean Air Act-goes into effect and are incorporated into the approved SIP.

         As a result of the partial disapproval, the EPA was required to promulgate a federal implementation plan within two years of that disapproval, unless it approved a corrected state plan.[49] The EPA did not promulgate a federal plan or approve a corrected state plan within that timeframe. Based on that, the Sierra Club sued the EPA, and a federal district court entered a consent decree requiring the EPA to issue a federal plan or approve a corrected state plan by December 15, 2017.[50]

         2. Louisiana's 2017 Revisions and the EPA's Approval

         To bring its SIP into compliance, Louisiana revised its SIP to, among other things, address BART for the Nelson and Brame powerplants. Nelson and Brame are two of the largest sources of SO2 and NOx emissions in Louisiana, are "BART-eligible," and have generating capacities greater than 750 megawatts.[51]

         In August 2016, Louisiana submitted a partial plan addressing BART at "non-EGU" sources. The EPA proposed to approve that partial plan in October 2016 and finalized the approval in the December 2017 Final Rule.[52] The approval of that partial plan is not at issue here.

         In February 2017, Louisiana submitted a revised plan addressing BART for EGU sources, including Nelson and Brame. In June 2017, Louisiana revised its SIP to make a BART determination for Nelson. Louisiana revised its SIP again in October 2017. The EPA's December 2017 Final Rule explains the relevant procedural history as follows:

On June 20, 2017, LDEQ submitted a SIP revision for parallel processing related to Entergy's Nelson facility. On July 13, 2017, we proposed to approve this SIP revision along with the remaining portion of the February 2017 SIP revision that addressed BART for the Nelson facility. Specifically, we proposed to approve the LDEQ BART determinations for Nelson Units 6 and 4, and the Unit 4 auxiliary boiler, and the [administrative order on consent] that makes the emission limits that represent BART permanent and enforceable for the purposes of regional haze. We also solicited comment with respect to any information that would support or refute the costs in Entergy's evaluation of SO2 controls for Unit 6. On June 21, 2017, Entergy submitted a comment to LDEQ on its proposed SIP revision requesting a three-year compliance deadline to achieve the proposed SO2 BART limit for Nelson Unit 6. Entergy's letter explained that the company has coal contracts in place for the next three years, so the revised compliance date would provide the company sufficient time to transition to new mines with lower sulfur coal. Additionally, Entergy stated that it did not have the necessary equipment to blend varying fuel supplies. On August 24, 2017, we received a letter from LDEQ explaining their intent to revise the compliance date in the SIP revision for Nelson Unit 6 based on Entergy's comment letter. On September 26, 2017, we supplemented our proposed approval of the SO2 BART determination for Nelson by proposing to approve the three-year compliance date. On October 26, 2017, we received LDEQ's final SIP revision addressing Nelson, including a final [administrative order of consent] with emission limits and a SO2 compliance date three years from the effective date of the EPA's final approval of the SIP revision.[53]

         In the October 2017 SIP revisions, Louisiana (1) determined that Nelson and Brame were "subject to BART" and (2) determined that low-sulfur coal was BART for controlling SO2 emissions at Nelson.

         In reaching those conclusions, Louisiana considered analyses submitted by Entergy, Cleco, and the EPA. The analyses submitted by the parties addressed: (1) the modeling methods for making "subject to BART" determinations and (2) the potential control options for BART at Nelson. The EPA's analyses contradicted and sharply criticized significant parts of Entergy's and Cleco's analyses. The LDEQ included all the analyses as appendices to its revised SIP.

         i. The "Subject to BART" Determinations

         The LDEQ concluded that Nelson and Brame were "subject to BART." The department reached that conclusion based on dispersion modeling that established that those powerplants produced emissions that surpassed the 0.5 deciview threshold for "contributing to" visibility impairment at the Breton National Wilderness Area and the Caney Creek Wilderness Area.

         The LDEQ relied on the "CALPUFF" model, which "predicts 24-hour average pollutant concentrations based on source emissions and how they disperse in the atmosphere" and converts those concentrations to daily deciview effects.[54] The BART guidelines provide that a state may "use CALPUFF or other appropriate model to estimate visibility impacts from a single source at a Class I area."[55] When the EPA promulgated those guidelines in 2005, CALPUFF was "the best regulatory modeling application currently available for predicting a single source's contribution to visibility impairment and [was then] the only EPA-approved model for use in estimating single source pollutant concentrations resulting from the long range transport of primary pollutants."[56] Although the EPA has recently removed CALPUFF as a "preferred" model for other air-quality modeling applications under the Clean Air Act, CALPUFF remains a recommended model for making "subject-to-BART" and BART emission control determinations.[57]

         Both Entergy's and the EPA's CALPUFF modeling showed that Nelson and Brame each had a greater than 0.5 deciview impact on visibility impairment at Breton and Caney Creek. Entergy, however, submitted reports showing that CALPUFF's reliance on flawed assumptions imposed a too-high margin of error.

         In addition to conducting CALPUFF modeling, Entergy and the EPA conducted another type of modeling called "CAMx." Entergy and the EPA submitted their CAMx analyses to the LDEQ.

         The EPA's CAMx modeling showed that Nelson and Brame exceeded the 0.5 deciview threshold. In contrast, Entergy's CAMx modeling showed that those powerplants' emissions did not exceed the 0.5 deciview threshold. The EPA, however, concluded that Entergy's CAMx modeling "was not conducted in accordance with the BART Guidelines and d[id] not properly assess maximum baseline impacts, so [the EPA] consider[ed] this CAMx modeling provided by Entergy to be invalid for supporting a determination of minimal visibility impacts."[58]

         When the LDEQ reviewed the models that the parties had submitted, it stated that it did "not have the expertise with which to review those [CAMx] model runs," and instead relied only on the CALPUFF modeling. The LDEQ did, however, include the CAMx analyses as appendices to its revised SIP.

         ii. BART Determinations for Brame and Nelson

         Louisiana's revised SIP also made BART determinations for Brame and Nelson. For the two Brame units, the LDEQ determined that "no additional controls constitute[d] BART." The LDEQ reached that determination based on those units' existing controls. At one of those units, BART was satisfied via the unit's earlier conversion from coal to natural gas. The other unit satisfied its BART obligations based on an earlier dry sorbent injection installation.[59]

         For Nelson, the LDEQ changed its BART determination between the February 2017 revision and the June and October 2017 revisions. In the February 2017 SIP revision, the LDEQ concluded that no additional controls were necessary for Nelson to satisfy BART. Although the LDEQ stated that "low sulfur coal presents the most feasible control based on economics and impacts to visibility," it also stated that it "believe[d] that the visibility improvement that would be achieved through the installation and operation of controls at each of the Nelson units would be negligible, therefore the facility's existing controls satisfy the BART requirements and no further controls are necessary."

         In its June 2017 SIP revision, however, the LDEQ changed course and determined that low-sulfur coal was BART for controlling SO2 at Nelson. The October 2017 SIP revision retained that determination and extended the compliance date for BART at Nelson by three years.

         In the October 2017 SIP revision, the LDEQ stated that it evaluated four different technologies for reducing emissions at Nelson: (1) low-sulfur coal, (2) dry sorbent injection ("DSI"), (3) dry flue-gas desulfurization ("FGD"), or a "dry scrubber," and (4) wet FGD, or a "wet scrubber." The LDEQ stated its reasoning for selecting low-sulfur coal as BART at Nelson as follows:

In the Entergy BART five-factor SO2 analysis for the Unit 6 Boiler, a number of emission reduction controls were reviewed. The reviewed controls included the use of a lower sulfur coal, DSI, enhanced DSI, dry flue-gas desulfurization (FGD) and wet FGD. LDEQ has reviewed and weighed the five factors carefully; after a review of the information that Entergy and EPA provided, LDEQ has concluded that the appropriate BART for this facility is to establish an emission limit of .6 lbs/MMBtu based on a 30 day rolling average as defined in the AOC (see Appendix D). While additional visibility benefits may be available through the use of FGD, the lower sulfur coal option results in visibility benefits at a lower annual cost. Along with the extra cost, FGD use results in additional waste spent due to spent reagent and has some power demands to run the equipment. LDEQ believes, at present, that the use of lower sulfur coal presents the appropriate SO2 control based on consideration of economics, energy impacts, non-air quality environmental impacts, and impacts to visibility.

         iii. Long-Term Strategy and Reasonable Progress Goals

         Louisiana's revised SIP did not specifically set out a long-term strategy or impose emission controls at additional "reasonable progress sources." Neither did it evaluate the four "reasonable progress factors" for determining whether any non-BART reasonable progress sources should be controlled.

         3. The EPA's Approval of Louisiana's SIP

         The EPA proposed two separate rules to approve Louisiana's revised SIP. In May 2017, the EPA issued a proposed rule to approve the entire February 2017 SIP revision except for the part about Nelson, on which the EPA deferred action.[60] This rulemaking included an approval of Louisiana's "subject to BART" determination for Brame.[61]

         In June 2017, Louisiana revised its SIP by changing its previous "no further controls" BART determination for Nelson to require the use of low-sulfur coal. In a July 2017 rulemaking ("the Proposed Rule"), the EPA proposed to approve the remaining part of Louisiana's SIP addressing Nelson.[62] The Proposed Rule (1) agreed with the LDEQ that Nelson was "subject to BART" and (2) stated that the LDEQ had adequately analyzed the five mandatory BART factors.[63] But the Proposed Rule also criticized several aspects of the SIP revision, including Entergy's modeling, cost estimates, and supporting documentation. The Proposed Rule stated that the EPA performed its own modeling and analyses to correct those errors.[64]

         After the EPA issued the Proposed Rule, Louisiana submitted a letter to the EPA explaining the state's intention to extend the compliance date in its draft SIP for Nelson by three years. In response to that letter, the EPA supplemented the Proposed Rule to reflect such change.[65]

         Louisiana revised its SIP again in October 2017. As discussed above, that revision extended the date for Nelson to comply with BART by three years. The EPA finalized its proposed approval of all of Louisiana's SIP revisions when it issued the Final Rule on December 21, 2017.[66]

         The Environmental and Industry Petitioners timely petitioned for review of the Final Rule. On the same day that this case was filed, the Environmental Petitioners also filed an administrative petition for reconsideration of the Final Rule. This court stayed the case pending the EPA's decision on the administrative petition, then lifted the stay after the EPA denied that petition.[67]

         II. Standard of Review

         We review the EPA's approval of Louisiana's regional haze SIP under the standards set out in the Administrative Procedure Act ("APA"), which require us to set aside an agency action that is "arbitrary, capricious, an abuse of discretion, or otherwise not in accordance with law."[68] An action is arbitrary and capricious "if the agency has relied on factors which Congress has not intended it to consider, entirely failed to consider an important aspect of the problem, offered an explanation for its decision that runs counter to the evidence before the agency, or is so implausible that it could not be ascribed to a difference in view or the product of agency expertise."[69]

         The arbitrary and capricious standard is "narrow," and we must "be mindful not to substitute [our] judgment for that of the agency."[70] We "must also ensure that the agency 'examined the relevant data and articulated a satisfactory explanation for its action.'"[71] "We consider whether the decision was based on a consideration of the relevant factors and whether there has been a clear error of judgment."[72]

         Additionally, federal agencies "are required to engage in 'reasoned decisionmaking.'"[73] The agency's process must be "logical and rational," and its decision "is lawful only if it rests 'on a consideration of the relevant factors.'"[74]The agency must "articulate a satisfactory explanation for its action including a 'rational connection between the facts found and the choice made.'"[75] Under this standard, courts have vacated agency decisions that created "unexplained inconsistencies in the rulemaking record."[76]

         Under the Clean Air Act's structure of cooperative federalism, Louisiana is the entity considering "relevant factors," and the EPA's role is confined to ensuring that Louisiana's determinations complied with the Clean Air Act.[77]Under this structure, we review whether the EPA was arbitrary and capricious in approving Louisiana's compliance with the Act.

         III. Analysis

         As a threshold matter, Industry Petitioners and Environmental Petitioners have standing. Industry Petitioners' standing is based on Entergy's and Cleco's ownership of the powerplants at issue. Environmental Petitioners have standing because one of the Sierra Club's members submitted a declaration adequately asserting that (1) he regularly visits the national parks in question, (2) he has concrete plans to return in the future, and (3) regional haze affects his visibility. Although the declarations submitted by the National Parks Conservation Association do not appear to give it standing in this case, "one party with standing is sufficient to satisfy Article III's case-or-controversy requirement."[78]

         Each petition raises two major issues. Environmental Petitioners challenge the EPA's approval of (1) Louisiana's selection of low-sulfur coal as BART for controlling SO2 emissions at Nelson and (2) Louisiana's reasonable progress goals and long-term strategy.

         Industry Petitioners challenge the EPA's approval of Louisiana's "subject to BART" determinations at Nelson and Brame. They contend that (1) the CALPUFF modeling that Louisiana relied on was technically flawed; (2) the CAMx modeling, on which the EPA relied but Louisiana did not, was technically flawed; and (3) even if the CAMx modeling were not technically ¶ awed, the EPA should not have considered it because Louisiana expressly did not consider it.

         We address each issue in turn.

         A. The Environmental Petitioners' Petition

         The first part of Environmental Petitioners' challenge addresses the EPA's approval of Louisiana's determination that low-sulfur coal was BART for Nelson. The second part of their challenge addresses the EPA's approval of Louisiana's long-term strategy and reasonable progress goals.

         We deny Environmental Petitioners' petition, and we explain our reasons below.

         1. The EPA's Approval of the BART Determination for Nelson

         In its October 2017 SIP revision, Louisiana selected low-sulfur coal as BART for controlling SO2 emissions at Nelson. The LDEQ explained its determination as follows:

In the Entergy BART five-factor SO2 analysis for the Unit 6 Boiler, a number of emission reduction controls were reviewed. The reviewed controls included the use of a lower sulfur coal, DSI, enhanced DSI, dry flue-gas desulfurization (FGD) and wet FGD. LDEQ has reviewed and weighed the five factors carefully; after a review of the information that Entergy and EPA provided, LDEQ has concluded that the appropriate BART for this facility is to establish an emission limit of .6 lbs/MMBtu based on a 30 day rolling average as defined in the AOC (see Appendix D). While additional visibility benefits may be available through the use of FGD, the lower sulfur coal option results in visibility benefits at a lower annual cost. Along with the extra cost, FGD use results in additional waste spent due to spent reagent and has some power demands to run the equipment. LDEQ believes, at present, that the use of lower sulfur coal presents the appropriate SO2 control based on consideration of economics, energy impacts, non-air quality environmental impacts, and impacts to visibility.

         Environmental Petitioners contend that the EPA's approval of this determination was arbitrary and capricious for three reasons. First, Louisiana's selection of low-sulfur coal as BART at Nelson was inconsistent with the evidence that the state considered. Second, Louisiana did not provide a "rational basis" for rejecting more effective pollution controls. Third, in its BART determination, Louisiana did not weigh the Clean Air Act's factors in compliance with the mandatory BART guidelines.

         i. The Evidence Louisiana Considered

         According to Environmental Petitioners, Louisiana's BART determination rested solely on analysis by Nelson's owner, Entergy. Environmental Petitioners point out that the EPA was "unable to verify any of the company's costs" because those costs were based on a proprietary database to which the EPA was not given access.[79] Similarly, the EPA could not verify Entergy's modeling analyses because Entergy did not provide the "inputs" that were used in the modeling.

         For the parts of Entergy's analyses that the EPA did review, the EPA concluded that Entergy's cost and visibility analyses, on which Louisiana relied, had many errors, including: (1) implementing improper costs, (2) inflated contingency estimates, and (3) modeling errors. The EPA pointed out these errors in the Proposed Rule: (1) "Entergy's control cost estimates included costs not allowed under our Control Cost Manual (e.g., escalation during construction and owner's costs)"; (2) "Entergy also assumed a contingency of 25%, which we note is unusually high"; and (3) "Entergy's CALPUFF modeling included errors in its estimates of sulfuric acid and PM emissions."[80]

         After noting these perceived errors, the EPA submitted its own BART analyses that reached markedly different cost-effectiveness estimates for each proposed control option at Nelson. The following table compares the EPA's and Entergy's cost-effectiveness calculations, in cost per ton of pollutant removed, for each control option.

Cost-Effectiveness Comparison[81]

Technology

Entergy's Calculation

the EPA's Calculation

Low-sulfur coal

$597

$2, 957

DSI

$5, 611

$3, 578 - $4, 302

Wet scrubber

$4, 413

$2, 743

Dry scrubber (SDA)

$4, 536

$2, 706

         Most striking is the difference between the EPA's and Entergy's calculations for low-sulfur coal. Entergy estimated that low-sulfur coal would cost $597 per ton of pollution removed; the EPA estimated that it would cost $2, 957 per ton. Entergy's estimate for what the parties agree would be the most effective option-a dry scrubber-was significantly higher than the EPA's estimate. Entergy's estimate was $4, 536 per ton of pollution removed; the EPA's was $2, 706 per ton.

         Similarly, the EPA's conclusions about the visibility improvements for each option differed from Entergy's conclusions. In the Proposed Rule, the EPA explained that the differences resulted from Entergy's model's failure to follow the BART guidelines.[82] The following chart compares the visibility improvement calculations (in deciviews) for each control option.

Expected Visibility Improvement Comparison for Caney Creek

Technology

Trinity Consultants' Calculation

the EPA's Calculation

Low-sulfur coal

0.164

0.411

DSI

0.302

0.511

Dry scrubber (SDA)

0.355

0.831

         According to Environmental Petitioners, Louisiana's reliance on Entergy's analyses-which the EPA's analyses concluded overestimated the costs and underestimated the benefits of more effective pollution controls-was irrational. After the EPA informed Louisiana of those errors, that state reached the same BART determination. In its SIP revision, Louisiana did not acknowledge the EPA's criticisms and did not attempt to reconcile the conflicting the EPA and Entergy analyses. Environmental Petitioners insist that Louisiana's failure to explain why its decision remained unchanged after reviewing the EPA's contradictory analyses created an "unexplained inconsistenc[y] in the rulemaking record" that made the EPA's approval arbitrary and capricious.[83]

         The EPA does not defend these discrepancies. In its brief, it concedes that the agency "was aware of these errors when it approved the SIP," acknowledges that the SIP revision contained "a number of errors," and describes Entergy's submissions as "Entergy's faulty analyses."

         The EPA contends that, despite these errors, Louisiana "may, to the extent supported by the record as a whole, reach the same conclusion both before and after reviewing a particular set of information."[84] The EPA points to the fact that its own analyses were in the record and that the LDEQ included those analyses in an appendix to the revised SIP. The EPA insists that, based on the agency's earlier review of the "entirety of" Louisiana's SIP submission, Louisiana considered all the information in its October 2017 SIP submission, weighed that information in arriving at its final BART determination, and explained the reasons for its decision.

         Industry Respondents defend the analyses they submitted to the LDEQ and insist that those analyses adequately supported Louisiana's BART determination. They criticize the analyses that the EPA submitted to the LDEQ and challenge the EPA's earlier determination that their analyses failed to comply with the BART guidelines.

         Environmental Petitioners have the better arguments on this point. Entergy and the EPA reached significantly different conclusions about the costs and visibility improvements of each emission control. But after the EPA submitted analyses pointing out flaws in Entergy's analysis, the LDEQ did not discuss the EPA-identified errors, and it reached the same BART determination for Nelson as it did before the EPA identified mistakes. Given the EPA's analysis showing that Entergy significantly underestimated the cost of low-sulfur coal and overestimated the cost of a scrubber, Louisiana's failure to address or reconcile the conflicting analyses appears to create an "unexplained inconsistenc[y]" in the rulemaking record.[85] Although that shortcoming in Louisiana's SIP is worthy of careful scrutiny, it does not fully resolve the matter.

         ii. Rational Explanation for Rejecting a More Effective Control

         Environmental Petitioners next contend that Louisiana did not provide a rational explanation for rejecting a more effective pollution control at Nelson. They point to the EPA's past actions that reviewed BART determinations for whether the costs and visibility benefits of a determination were within the range of the EPA's prior BART determinations.[86] They also cite the BART guidelines, which recommend that "[a] reasonable range would be a range that is consistent with the range of cost effectiveness values used in other similar permit decisions over a period of time."[87]

         The LDEQ's SIP revisions did not provide cost-effectiveness numbers for any of the control options it considered. Based on the LDEQ's failure to include those numbers in its SIP, the EPA was not able to compare the instant BART determination with the EPA's previously approved BART determinations. According to Environmental Petitioners, this lack of a comparison to prior EPA-approved BART determinations amounts to an "unexplained deviation from past practice."[88]

         To bolster this argument, Environmental Petitioners compared the EPA's cost-effectiveness numbers for a scrubber with all of the EPA's prior BART determinations. Environmental Petitioners included a graph of those results, which shows that the cost-effectiveness numbers for a scrubber align with the EPA's prior BART decisions.

         The EPA responds that, although in other actions it has stated that comparisons to prior BART determinations are helpful, the BART guidelines do not require such comparisons. The EPA points to the Final Rule's statements explaining that BART determinations depend on the unique circumstances of each source and that "[s]ome variation is to be expected because SIP actions are highly fact-dependent."[89] The EPA also cites several cases holding that states have broad authority to weigh the statutory factors and to pick appropriate emission controls.[90]

         The EPA has the better argument on this point. Louisiana has the authority to select the appropriate BART emission control. The fact that a scrubber more closely aligns with the EPA's prior BART determinations did not preclude Louisiana from choosing low-sulfur coal instead. The guideline that Environmental Petitioners cite does not require that BART determinations align with the EPA's prior BART determinations:

You should provide documentation of any unusual circumstances that exist for the source that would lead to cost-effectiveness estimates that would exceed that for recent retrofits. This is especially important in cases where recent retrofits have cost-effectiveness values that are within what has been considered a reasonable range, but your analysis concludes that costs for the source being analyzed are not considered reasonable. (A reasonable range would be a range that is consistent with the range of cost effectiveness values used in other similar permit decisions over a period of time.)[91]

         Although the parties do not explain how low-sulfur coal compares to the EPA's prior BART determinations, the record is devoid of any "unusual circumstances" indicating that low-sulfur coal's cost-effectiveness estimates are out of line with prior BART determinations. Therefore, the guideline addressing "similar permit decisions over a period of time" does not apply here.

         iii. Weighing the Statutory Factors and Compliance with the BART Guidelines

         In making a BART determination, a state "shall take into consideration" five factors: (1) the costs of compliance, (2) the energy and nonair quality environmental impacts of compliance, (3) any existing pollution control technology in use at the source, (4) the remaining useful life of the source, and (5) the degree of improvement in visibility that may reasonably be anticipated to result from the use of such technology.[92] States have discretion as to "the weight and significance" of each factor, as well as the appropriate emission control that qualifies as BART at a given source.[93]

         In addition to considering those statutory factors, for powerplants with generating capacities greater than 750 megawatts, like Nelson and Brame, the state's BART determination "shall be determined pursuant to [the BART] guidelines."[94] The BART guidelines state:

From the alternatives you evaluated . . . we recommend you develop a chart (or charts) displaying for each of the alternatives:
(1) Expected emission rate . . .;
(2) Emissions performance level . ...

Buy This Entire Record For $7.95

Download the entire decision to receive the complete text, official citation,
docket number, dissents and concurrences, and footnotes for this case.

Learn more about what you receive with purchase of this case.