# Approval of Air Quality Implementation Plans; Navajo Nation; Regional Haze Requirements for Navajo Generating Station > Briefs, arguments, decisions, and more. URL: https://www.frixlaw.com/law-library/documents/fr%3A2014-18228 ## Record - **Collection:** Federal Register - **Document type:** Rule - **Published:** August 8, 2014 - **Citation:** 79 FR 46514 ## Text ENVIRONMENTAL PROTECTION AGENCY 40 CFR Part 49 [EPA-R09-OAR-2013-0009; FRL-9914-62-Region 9] Approval of Air Quality Implementation Plans; Navajo Nation; Regional Haze Requirements for Navajo Generating Station AGENCY: Environmental Protection Agency (EPA). ACTION: Final rule. SUMMARY: The Environmental Protection Agency (EPA) is promulgating a source-specific Federal Implementation Plan (FIP) requiring the Navajo Generating Station (NGS), a coal-fired power plant located on the Navajo Nation near Page, Arizona, to achieve reductions in oxides of nitrogen (NO X ) required under the Best Available Retrofit Technology (BART) provisions of the Clean Air Act (CAA) and the Regional Haze Rule (RHR). On February 5, 2013, EPA issued a proposed BART determination for NGS and an alternative to BART. In a supplemental proposal on October 22, 2013, EPA proposed to approve a new alternative plan, based on an agreement developed by a group of stakeholders known as the Technical Work Group (TWG). EPA is finalizing the alternative to BART described in our supplemental proposal. This rule is consistent with the TWG Agreement, including a lifetime cap in total emissions of NO X from NGS over 2009-2044 (2009-2044 NO X Cap). Our final action will achieve greater emissions reductions than BART and is expected to significantly reduce the impact of NGS on visibility at 11 mandatory Class I Federal areas. The operator of NGS must implement one of several alternative operating scenarios to achieve the necessary emission reductions to comply with the 2009-2044 NO X Cap. DATES: Effective date: This rule is effective on October 7, 2014. FOR FURTHER INFORMATION CONTACT: Anita Lee, EPA Region 9, (415) 972-3958, lee.anita@epa.gov. SUPPLEMENTARY INFORMATION: EPA has established a docket for this action under Docket ID No. EPA-R09-OAR-2013-0009. The index to the docket for this action is available electronically at http://www.regulations.gov and in hard copy at EPA Region 9, 75 Hawthorne Street, San Francisco, California. While documents in the docket are listed in the index, some information may be publicly available only at the hard copy location (e.g. copyrighted material, voluminous or oversized documents, etc.), and some may not be publicly available in either location (e.g. Confidential Business Information (CBI)). To inspect the hard copy materials, please schedule an appointment during normal business hours with the contact listed in the FOR FURTHER INFORMATION CONTACT section. A reasonable fee may be charged for copies. Throughout this document, “we”, “us”, and “our” refer to EPA. Table of Contents I. Executive Summary II. Background for the Final Rule A. History of NGS B. Summary of Statutory and Regulatory Framework for Addressing Visibility and Sources Located in Indian Country C. Summary of Proposed Rule and Supplemental Proposal D. Summary of Legal Rationale for Compliance Flexibility III. Summary of Final FIP Provisions IV. Summary of Major Issues Raised by Commenters V. Summary of Final Action VI. Administrative Requirements A. Executive Order 12866: Regulatory Planning and Review and Executive Order 13563: Improving Regulation and Regulatory Review B. Paperwork Reduction Act C. Regulatory Flexibility Act D. Unfunded Mandates Reform Act E. Executive Order 13132: Federalism F. Executive Order 13175: Consultation and Coordination With Indian Tribal Governments G. Executive Order 13045: Protection of Children From Environmental Health Risks and Safety Risks H. Executive Order 13211: Actions Concerning Regulations That Significantly Affect Energy Supply, Distribution, or Use I. National Technology Transfer and Advancement Act J. Executive Order 12898: Federal Actions To Address Environmental Justice in Minority Populations and Low-Income Populations K. Congressional Review Act L. Petitions for Judicial Review I. Executive Summary EPA is taking final action pursuant to the CAA and the RHR to require Units 1, 2, and 3 at NGS to reduce emissions of NO X in order to reduce the impact NGS has on visibility at 11 mandatory Class I Federal areas. We are finalizing an alternative to BART based on agreed-upon recommendations developed by a group of diverse stakeholders known as the Technical Work Group (TWG). Our final action limits emissions of NO X from NGS by establishing a long-term facility-wide cap on total NO X emissions from 2009 to 2044 and requires the implementation of one of several alternative operating scenarios to ensure that the 2009-2044 cap is met. Generally, the alternative operating scenarios require the closure of one unit at NGS (or the curtailment of electricity generation by a similar amount) in 2019, and compliance with a NO X emission limit that is achievable with the installation of selective catalytic reduction (SCR) on two units in 2030. As part of our final action, EPA is also setting a source-specific BART Benchmark against which to compare the TWG Alternative to ensure that it will achieve greater reasonable progress than BART. The BART Benchmark is consistent with the BART determination we proposed on February 5, 2013, requiring all three units at NGS to meet an emission limit achievable with SCR within five years of a final rule. EPA is not finalizing our proposed BART determination for NGS in the regulatory requirements of this Final Rule. EPA's action to finalize an alternative to BART consistent with the TWG Agreement will achieve greater NO X emission reductions at lower cost than BART in exchange for flexibility in the timeframe for achieving NO X reductions. When fully implemented, this Final Rule requires over an 80 percent reduction in NO X emissions from NGS and is expected to significantly reduce the impact of NGS on visibility at 11 mandatory Class I Federal areas. II. Background for the Final Rule A. History of NGS NGS is a coal-fired power plant located on the Navajo Nation Indian Reservation near Page, Arizona. The facility consists of three 750 megawatt (MW) coal-fired electric utility steam generating units with a total capacity of 2250 MW constructed from 1974 to 1976. The three units at NGS are co-owned by six entities: The United States Bureau of Reclamation (Reclamation) (24.3 percent); Salt River Project (21.7 percent), which also serves as the facility operator; Los Angeles Department of Water and Power (21.2 percent); Arizona Public Service (14 percent); NV Energy (11.3 percent); and Tucson Electric Power (7.5 percent). Federal participation in NGS was authorized in the Colorado River Basin Project Act of 1968 as a preferred alternative to building hydroelectric dams in the Grand Canyon for the purpose of providing power to the Central Arizona Project (CAP). 1 The CAP is a 336-mile water distribution system that delivers about 1.5 million acre-feet (AF) per year of Colorado River water from Lake Havasu in western Arizona to non-Indian agricultural (NIA) water users in central Arizona, Indian tribes located in Arizona, and municipal water users in Maricopa, Pinal, and Pima Counties in Arizona. The CAP water is used to meet the terms of a number of Indian water-rights settlements in central Arizona and to reduce groundwater usage in the region. A portion of Reclamation's share of electricity from NGS powers the pumps that move CAP water to its destinations along the distribution system. 1 For more detail and for citations or references to the information provided in this Background section, please see the Proposed Rule at 78 FR 8274 (February 5, 2013). Several tribes located in Arizona, including the Gila River Indian Community, the Ak-Chin Indian Community, the Tohono O'odham Nation, the San Carlos Apache Tribe, the White Mountain Apache Indian Tribe, the Fort McDowell Yavapai Nation, the Salt River Pima-Maricopa Indian Community, the Navajo Nation, the Yavapai-Apache Nation, the Hopi Tribe, the Pascua Yaqui Tribe, the Yavapai-Prescott Tribe, and the Tonto Apache Nation, have CAP water allocations or contracts. In exchange for allocations of CAP water at reduced cost and access to funds for the development of water infrastructure, the tribes with water settlement agreements have released their claims to other water in Arizona. Excess NGS power owned by Reclamation that is not used by CAP is sold and profits are deposited into the Lower Colorado River Basin Development Fund (Development Fund) to support the tribal water settlement agreements. The U.S. Department of the Interior (DOI or Interior), through Reclamation, plays an important role in the implementation of these settlement agreements and the management of the Development Fund. The coal used by NGS is supplied by the Kayenta Mine, operated by Peabody Energy and located on reservation lands of both the Navajo Nation and the Hopi Tribe. Taxes and royalties from NGS and the Kayenta Mine are paid to the Navajo Nation and Hopi Tribe, contributing to the annual revenues for both governments. EPA understands that the process is underway to renew site leases for NGS and the Kayenta Mine, as well as associated rights of way agreements and contracts with the Navajo Nation and Hopi Tribe. Given the extent of federal and tribal interests in NGS, on January 4, 2013, EPA, DOI, and the Department of Energy (DOE) signed a joint federal agency statement (Joint Statement) committing to collaborate on several short- and long-term goals, including analyzing and pursuing strategies for providing clean, affordable, and reliable power, affordable and sustainable water, and sustainable economic development to key stakeholders who currently depend on NGS. 2 The Joint Statement also recognizes the trust responsibility of the Federal government to Indian tribes. 2 See document title “2013_0104 Joint Federal Agency Statement on NGS” within document number 0005 in the docket for this proposed rulemaking at EPA-R09-OAR-2013-0009, which can be found at www.regulations.gov. B. Summary of Statutory and Regulatory Framework for Addressing Visibility and Sources Located in Indian Country In our Proposed Rules, we provided a detailed discussion of the statutory and regulatory framework for addressing visibility impairment in the mandatory Class I Federal Areas, addressing sources located in Indian country under the statute and the Tribal Authority Rule (TAR), and developing BART determinations pursuant to the CAA and the BART Guidelines set forth in Appendix Y to 40 CFR Part 51. 3 Here, we provide a brief summary of the statutory and regulatory framework. 3 See 78 FR 8274 (February 5, 2013) and 78 FR 62509 (October 22, 2013). Title I, part C, subpart II of the CAA Amendments of 1977 establishes a visibility protection program that sets forth “as a national goal the prevention of any future, and the remedying of any existing, impairment of visibility in mandatory class I Federal areas which impairment results from man-made air pollution.” 4 EPA promulgated regional haze regulations implementing the program on April 22, 1999. 5 Consistent with the statutory requirement in 42 U.S.C. 7491(b)(2)(a), EPA's 1999 regional haze regulations include a provision that States must require certain major stationary sources to procure, install, and operate BART. This provision covers sources in listed industrial categories with the potential to emit 250 or more tons per year of an air pollutant that were “in existence on August 7, 1977, but which ha[ve] not been in operation for more than fifteen years as of such date.” These sources are considered to be “BART-eligible.” 6 NGS meets these criteria and is a BART-eligible source. 7 4 See 42 U.S.C. 7419A(a)(1). 5 See 64 FR 35765 (April 22, 1999). 6 See 40 CFR 51.308(e)(1). 7 See 74 FR 44314 (August 28, 2009); 78 FR 8279 (February 5, 2013); see also 56 FR 50172 (October 3, 1991) addressing BART for SO 2 based on Reasonably Attributable Visibility Impairment. BART-eligible sources that are reasonably anticipated to cause or contribute to visibility impairment are “subject” to the BART requirements. 8 Generally speaking, a BART-eligible source with a predicted visibility impact of 0.5 deciviews (dv) or more in a Class I area is considered to “contribute” to visibility impairment. 9 NGS contributes to visibility impairment at 11 surrounding Class I areas in excess of this threshold, and is thus subject to BART. 8 See 70 FR 39104 at 39161 (July 6, 2005). 9 Id. In determining BART, States are required to take into account five factors identified in the CAA and EPA's regulations. 10 Those factors are: (1) The costs of compliance, (2) the energy and non-air quality environmental impacts of compliance, (3) any pollution control equipment in use or in existence 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. 11 EPA's guidelines for evaluating BART provide more detail and are set forth in Appendix Y to 40 CFR Part 51. 10 42 U.S.C. 7491(g)(2) and 40 CFR 51.308(e)(1)(A)(ii)(A). 11 40 CFR 51.308(e)(1)(ii)(A). In 1998, EPA promulgated the Tribal Authority Rule (TAR) relating to implementation of CAA programs in Indian country. 12 In the TAR, EPA determined that it has the discretionary authority to promulgate “such federal implementation plan provisions as are necessary or appropriate to protect air quality” consistent with CAA sections 301(a) and 301(d)(4) when a tribe has not submitted or EPA has not approved a Tribal Implementation Plan (TIP). 13 EPA has previously promulgated FIPs under the TAR to regulate air pollutants emitted from NGS. 14 12 See 40 CFR part 49; see also 59 FR 43956 (August 25, 1994) (proposed rule); 63 FR 7254 (February 12, 1998) (final rule); Arizona Public Service Company v. EPA, 211 F.3d 1280 (D.C. Cir. 2000), cert. den., 532 U.S. 970 (2001) (upholding the TAR). 13 See 40 CFR 49.11(a). 14 See 56 FR 50172 (October 3, 1991). In 1999, EPA proposed a FIP for NGS to fill the regulatory gap that existed because Arizona State permits and SIP rules are not applicable or enforceable in the Navajo Nation, and the Tribe had not sought approval of a TIP covering the plant. 64 FR 48731 (September 8, 1999). EPA then re-proposed the FIP with some additional conditions in September 2006. 71 FR 53631 (September 12, 2006). EPA finalized that NGS FIP on March 5, 2010. 75 FR 10174. Under the CAA, compliance with emission limits determined to be BART must be achieved as expeditiously as practicable but not later than 5 years after the effective date of the final BART determination (See CAA 169A(b)(2)(A) and (g)(4)). As discussed in greater detail in our Proposed Rule, EPA recognizes that the circumstances related to NGS create unusual and significant challenges for a 5-year compliance schedule. 15 Based on those challenges and our discretion under the TAR for implementing CAA requirements in Indian country, we considered other options that are consistent with the CAA and RHR, and that provide for a more flexible, extended compliance schedule. 15 Because of its complicated history and its location on the Navajo Nation, NGS faces numerous unique complexities and the unusual requirement to comply with NEPA for lease and other rights-of-way approvals, which apply only to NGS and Four Corners Power Plant, the other coal-fired power plant located on the Navajo Nation. EPA also understands the importance of the continued operation of NGS and the Kayenta Mine to the Navajo Nation and Hopi Tribe as a source of direct revenues through lease payments and coal royalties, as well as the importance of Reclamation's share of NGS to supply water to many tribes located in Arizona in accordance with several water settlement acts. EPA also recognizes that Reclamation may have fewer options compared to the other owners for financing pollution control or other large capital improvement projects at NGS. SRP expressed concern that the owners of NGS may choose to retire the facility if faced with the financial risk of making a large capital investment within 5 years without also having certainty that the lease and contract re-negotiations would conclude in a timely and favorable manner. EPA's BART regulations allow an alternative in lieu of BART, provided the alternative results in greater reasonable progress than would have been achieved through installation of BART. 16 Generally, an alternative is considered to be approvable provided it results in greater emissions reductions and the geographic distribution in emissions from the alternative is not substantially different than the distribution of the emissions under BART. 17 For a state that is subject to the submittal deadlines in the RHR, the regulations provide that alternatives to BART must ensure that all necessary emission reductions occur within the period of the first long-term strategy for regional haze (i.e., by 2018) for states that were required to submit regional haze SIPs in December 2007. 18 Thus, if states had submitted timely regional haze SIPs in 2007 with BART compliance deadlines in 2012, the RHR provided more than 5 additional years for the implementation of alternatives to BART. 16 40 CFR 51.308(e)(2). 17 40 CFR 51.308(e)(3). 18 40 CFR 51.308(e)(2)(iii). C. Summary of Proposed Rule and Supplemental Proposal EPA published an advanced notice of proposed rulemaking (ANPR) concerning BART for NGS and the Four Corners Power Plant in August 2009. 19 On February 5, 2013, EPA's proposed BART determination for NGS was published in the Federal Register and provided a thorough discussion of the statutory and regulatory framework for addressing visibility through application of BART for sources located in Indian country, and of the factual background for our BART determination at NGS. 20 The proposal analyzed the five BART factors and proposed to find that BART for NGS was installation of emissions controls to meet a NO X emission limit of 0.055 lb/MMBtu based on a rolling average of 30 boiler operating days (30-BOD average). 21 However, in recognition of the important role that NGS and the Kayenta Mine play in providing employment and revenue to the Navajo Nation and Hopi Tribe, and the role of Reclamation's share of electricity generated by NGS in fulfilling water settlement agreements with numerous tribes located in Arizona, we proposed that the potential economic impacts to tribes argue for thoughtful consideration of how flexibility in the compliance timeframe could be provided consistent with the air quality goals of the CAA. 22 Therefore, as discussed in our Proposed Rule, EPA proposed to exercise our authority and discretion under section 301(d)(4) of the CAA and 40 CFR 49.11(a) to propose an appropriate timeframe for alternative measures to BART under the RHR for NGS. We provided a thorough discussion of the legal rationale for setting the compliance schedule for alternative measures in our Proposed Rule. 23 19 See 74 FR 44314 (August 28, 2009). 20 See 78 FR 8274 (February 5, 2013). 21 Id. at 8288. 22 Id. at 8284. 23 Id. at 8289. Our Proposed Rule included a framework for evaluating alternatives to BART. 24 As part of the framework, EPA proposed a NO X emission credit for the previous early and voluntary installation of low-NO X burners with separated over-fire air (LNB/SOFA) over the 2009-2011 timeframe (LNB/SOFA credit). We proposed that the LNB/SOFA credit supported setting a compliance timeframe based on the flexibility under section 301(d)(4) of the CAA and 40 CFR 49.11(a). 25 EPA proposed to find that an alternative is “better than BART” if the total emissions over 2009-2044 from the alternative measure, minus the LNB/SOFA credit, are less than the total emissions under our proposed BART determination for the same period (i.e., the BART Benchmark). Consistent with this framework, EPA proposed an alternative to BART, requiring compliance with an emission limit of 0.055 lb/MMBtu on one unit per year in 2021, 2022, and 2023 (Alternative 1). We calculated that total emissions under Alternative 1 over 2009-2044, minus the LNB/SOFA credit, would be less than emissions based on the BART Benchmark. Thus, we proposed to find that Alternative 1 was “better than BART”. EPA recognized that there may be interest in additional flexibility beyond the 2021-2023 timeframe. EPA evaluated two additional compliance schedules but did not propose to approve them as “better than BART” alternatives because total emissions over 2009-2044 under these compliance schedules exceeded the BART Benchmark. However, we noted that potential technologies or other options for achieving additional emission reductions could bridge the NO X emission reduction deficit for alternatives to BART with compliance schedules that do not, by themselves, meet the BART Benchmark. 26 We invited stakeholders to submit additional BART alternatives, consistent with our proposed framework, for EPA's consideration. 24 Id. at 8290-92. 25 78 FR 62509 at 62511 (October 22, 2013). 26 78 FR 8274 at 8291 (February 5, 2013). On July 26, 2013, a stakeholder group, known as the Technical Work Group on NGS (TWG), submitted an agreement that had been established among the seven diverse entities in the TWG. We refer to the July 26, 2013, document as the “TWG Agreement.” The TWG is composed of representatives from Central Arizona Water Conservation District (CAWCD), the Environmental Defense Fund (EDF), the Gila River Indian Community (Gila River or the Community), the Navajo Nation (Navajo), Salt River Project (SRP) on behalf of itself and the other non-federal owners, DOI, and Western Resource Advocates (WRA). Although EPA attended the opening session of a “kick-off” meeting for the TWG on March 21, 2013, at which we described our Proposed Rule, EPA did not otherwise participate in the TWG and was not involved in any of the discussions leading to submittal of the TWG Agreement. Appendix B to the TWG Agreement contained TWG's recommendation for an alternative to BART. In general, the alternative plan in the TWG Agreement included closure of one unit at NGS, or curtailment of net generating capacity by an equivalent amount, in 2019 and compliance with a NO X emission limit of 0.07 lb/MMBtu on two units at NGS beginning in 2030. The TWG Agreement also included a provision requiring the operator of NGS to cease conventional coal-fired generation at NGS by the end of 2044. EPA independently evaluated Appendix B to the TWG Agreement to determine whether it complied with the framework we put forth in our Proposed Rule, as well as the statutory and regulatory requirements in the CAA and the RHR. On October 22, 2013, EPA published a Supplemental Proposal describing the TWG Agreement and requesting comment. 27 Our Supplemental Proposal contained a detailed evaluation of Appendix B to the TWG Agreement along with a discussion of our legal rationale for proposing to approve requirements consistent with the TWG Agreement as meeting the requirements for an alternative to BART. Throughout this document, we refer to the regulations we proposed in our Supplemental Proposal that are consistent with Appendix B of the TWG Agreement as the “TWG Alternative.” Thus, in this document, the term TWG Alternative refers to EPA's independent regulatory requirements for NGS consistent with the TWG Agreement, rather than to Appendix B of the TWG Agreement. 27 See 78 FR 62509 (October 22, 2013). In our Supplemental Proposal, we proposed to revise the numerical value of the BART Benchmark from our Proposed Rule. We also proposed a 2009-2044 NO X Cap based on the revised numerical value of the BART Benchmark. In our Proposed Rule, we calculated the BART Benchmark to be 358,974 tons of NO X . As discussed in our Supplemental Proposal, we proposed three changes to the BART Benchmark: (1) Correction of a transcription error; (2) correction of the date that EPA anticipated would be 5 years following the effective date of the final rule (i.e., July 1, 2019 instead of January 1, 2018); and (3) application of the LNB/SOFA credit to the BART Benchmark, rather than alternatives to BART, to represent emissions under BART if LNB/SOFA had been installed concurrently with selective catalytic reduction (SCR) to reduce NO X emissions. 28 Based on these changes, EPA proposed a 2009-2044 NO X Cap of 494,899 tons. Although EPA revised our accounting method for the LNB/SOFA credit in our Supplemental Proposal, EPA provided a demonstration that the method EPA used in our Proposed Rule to compare our proposed BART determination against BART alternatives was equivalent to the method in the Supplemental Proposal. 29 The application of the LNB/SOFA credit to the BART Benchmark in the Supplemental Proposal represented what total emissions over 2009-2044 would have been under our proposed BART determination if the operator of NGS had elected to install LNB/SOFA concurrently with SCR, i.e., within 5 years of a final rule, rather than in 2009-2011. Calculation of the BART Benchmark and 2009-2044 NO X Cap in this manner is easier to apply and enforce in the context of a cap in NO X emissions because the LNB/SOFA credit is built into the BART Benchmark rather than subtracted each year from actual cumulative emissions. 30 28 Id. Tables 1 and 3 at 62515-62516. 29 Id. Table 2 and footnote 32 at 62515. 30 In contrast, in our Proposed Rule, we calculated the BART Benchmark and emissions under BART alternatives using the actual early installation dates for LNB/SOFA and then applied the LNB/SOFA credit to BART alternatives for comparison against the BART Benchmark. Although this method would have resulted in a lower numerical value for the 2009-2044 NO X Cap, the LNB/SOFA credit (representing the early emission reductions achieved over 2009-2018) would have instead been subtracted from the calculations of cumulative emissions under the BART alternative. Although this is functionally equivalent to the method used in the Supplemental Proposal, this method would make annual comparisons of actual cumulative emissions under the BART alternative against the BART Benchmark more complicated because it would have required adjustments every year to total emissions to subtract out the LNB/SOFA credit. By accounting for the LNB/SOFA credit in the BART Benchmark, the actual annual emissions from NGS can be directly compared to the BART Benchmark without any further adjustments. In addition to the enforceable 2009-2044 NO X Cap, our Supplemental Proposal defines the operating scenarios that would be required depending on the final outcome of NGS ownership after the expiration of the current lease term at the end of 2019. In the TWG Agreement, the owners of NGS committed to maintain emissions from NGS below the 2009-2044 NO X Cap regardless of post-2019 ownership of NGS and the applicable operating scenario. As a result, the operating scenarios in the TWG Alternative include specific actions for achieving emission reductions in 2019 and in 2030. The TWG Alternative also provides for an operating scenario that is less well-defined in terms of specific actions but establishes a second NO X emissions cap over the period of 2009-2029 (2009-2029 NO X Cap) that is equivalent to emission reductions that would be achieved by a more well-defined operating scenario. The 2009-2029 NO X Cap would apply in addition to the 2009-2044 NO X Cap. The Supplemental Proposal included requirements for annual emission reporting to EPA that would also be made publicly available as part of the compliance demonstration for the TWG Alternative. D. Summary of Legal Rationale for Compliance Flexibility In our February 5, 2013, proposal for NGS, EPA proposed an alternative to BART that we referred to asAlternative 1. EPA proposed to find that consideration of a compliance schedule beyond 2018 for Alternative 1 at NGS was appropriate for a number of reasons, including the importance of NGS to numerous Indian tribes located in Arizona and the federal government's reliance on NGS to meet the requirements of water settlements with several tribes. Providing this timeframe for compliance would not, in itself, avoid or mitigate increases in water rates for tribes located in Arizona; however, it would provide time for the collaborating federal agencies to explore options to avoid or minimize potential impacts to tribes, including seeking funding to cover expenses for the federal portion of pollution control at NGS. In developing this framework, EPA proposed to exercise its authority and discretion under section 301(d)(4) of the CAA, 42 U.S.C. 7601(d)(4) and the TAR, 40 CFR 49.11(a), and proposed an appropriate timeframe for an alternative measure under the RHR for NGS. EPA considered this timeframe to be consistent with the general programmatic requirements. Under the RHR, States and regulated sources had almost 20 years from the issuance of the rule in 1999 to design and implement alternative measures to BART. For numerous reasons, including the myriad stakeholder interests and complex governmental interests unique to NGS, we are only now addressing the BART requirements for NGS. Our proposal to require emission reductions beyond 2018 was supported by CAA section 301(d)(4) and the TAR codified at 40 CFR 49.11(a). The TAR reflects EPA's commitment to promulgate “such Federal implementation plan provisions as are necessary or appropriate to protect air quality” in Indian country where a tribe either does not submit a Tribal Implementation Plan (TIP) or does not receive approval of a submitted TIP ( emphasis added ). The use of the term “provisions as are necessary or appropriate” indicates EPA's determination that it may only be necessary or appropriate to promulgate a FIP of limited scope. The United States Court of Appeals for the Tenth Circuit has previously endorsed the application of this approach in a challenge to the FIP for the Four Corners Power Plant, stating: “[40 CFR 49.11(a)] provides the EPA discretion to determine what rulemaking is necessary or appropriate to protect air quality and requires the EPA to promulgate such rulemaking.” 31 The court went on to observe: “Nothing in section 49.11(a) requires EPA . . . to submit a plan meeting the completeness criteria of [40 CFR part 51] Appendix V.” 32 While the decision in the Tenth Circuit focused on 40 CFR part 51, Appendix V, EPA believes the same considerations apply to the promulgation of a FIP intended to address the objectives set forth in 40 CFR 51.308(e)(2). In particular, EPA has discretion to determine if and when a FIP addressing the objectives set forth in 40 CFR 51.308(e)(2) should be promulgated, which necessarily includes discretion to determine the timing for complying with the requirements of any such FIP. 31 See Ariz. Public Serv. Co. v. EPA, 562 F.3d 1116 (10th Cir. 2009). 32 Id. III. Summary of Final FIP Provisions EPA is finalizing our finding that it is necessary or appropriate to promulgate a source-specific FIP requiring NGS to achieve NO X emission reductions required by the BART provisions of the CAA and RHR. EPA is determining that our proposed NO X emission limit of 0.055 lb/MMBtu, based on our analysis of the relevant factors, establishes the appropriate BART Benchmark for determining “better than BART.” Further, we are finalizing our assessment that the TWG Alternative, which establishes an enforceable 2009-2044 cap on NO X emissions from NGS over the life of the facility is “better than BART.” Finally, we are finalizing the TWG Alternative as the FIP requirements for NGS. EPA is promulgating four possible operating scenarios under the TWG Alternative (see Table 1). The operator of NGS must implement one of the four enforceable operating scenarios in order to comply with the 2009-2044 NO X Cap. The applicable operating scenario will depend on the outcome of ownership changes related to LADWP, NV Energy, and Navajo Nation, as well as whether the operator of NGS can increase capacity (by no more than 189 MW) to accommodate ownership changes, without triggering New Source Review permitting requirements, as described in Table 1. Once the ownership outcomes are finalized, the operator of NGS must implement the applicable Alternative as shown in Table 1. For example, if LADWP and NV Energy both retire their ownership shares of NGS and the Navajo Nation does not elect to purchase an ownership share of NGS, TWG Alternative A1 applies and the operator of NGS must implement Alternative A1 and may not elect to implement Alternatives A2, A3, or B. By December 1, 2019, the operator of NGS must notify EPA of the applicable Alternative (i.e., TWG Alternative A1, A2, A3, or B). In addition to the enforceable 2009-2044 NO X Cap, Alternatives A1, A2, and A3 each has enforceable emission reduction measures in 2019 and 2030 (see Table 1). Under Alternative B, in addition to the enforceable 2009-2044 NO X Cap, the operator of NGS must also ensure that cumulative NO X emissions over 2009-2029 comply with the 2009-2029 NO X Cap. The 2009-2029 NO X Cap is calculated based on emissions that would have been emitted over that period under Alternative A1. Under all Alternatives, if, based on required annual reports submitted by the operator of NGS to EPA, cumulative emissions of NO X from NGS exceed the 2009-2044 NO X Cap at any time prior to December 31, 2044, the operator of NGS must permanently cease operation of NGS. In addition, under Alternative B, if cumulative emissions of NO X exceed the 2009-2029 NO X Cap prior to 2029, the operator of NGS must temporarily cease operation of all units at NGS. 33 Under all Alternatives, the operator must permanently cease operation of all units at NGS by December 22, 2044. 33 The combination of the 2009-2044 and 2009-2029 NO X Caps under TWG Alternative B means that if NGS exceeds the 2009-2029 NO X Cap prior to 2029 it must cease operation, but the operator may re-start operation after 2030 as long as cumulative emissions have not yet exceeded the 2009-2044 NO X Cap. Under all TWG Alternatives, the operator of NGS must report to EPA annual emissions and heat input data and must make this information publicly available on its Web site. In addition, under TWG Alternative B, the operator must also submit to EPA annual Emission Reduction Plans projecting year-by-year emissions covering the 2020-2029 and 2030-2044 periods so that there is a plan for operation of NGS that ensures that cumulative emissions of NO X do not exceed the 2009-2029 NO X Cap and the 2009-2044 NO X Cap. Although year-by-year emissions projected in the annual Emission Reduction Plans are not enforceable (i.e., emissions in a given year are not required to match projections for that year in an Emission Reduction Plan), the requirement to submit Emission Reduction Plans is enforceable, and provides the operator with a framework for planning for future emissions reductions. The requirement also provides EPA and the public the opportunity to monitor and evaluate progress of emission reductions under TWG Alternative B. Table 1—Summary of the Major Regulatory Provisions of the TWG Alternative Applicability (Step 1) • If LADWP and NV Energy both exit NGS without selling their ownership interests (i.e., retire shares), or both exit by selling to an existing NGS participant; or one retires shares and the other sells to an existing NGS participant; and • If LADWP or NV Energy sells to a 3rd party, or does not exit NGS; (Step 2) • If Navajo Nation does not purchase ownership share by 12/31/19; • If Navajo Nation purchases up to 170 MW by 12/31/19; and • If Navajo Nation purchases up to 170 MW by 12/31/19; and • n/a. (Step 3) • n/a • If Participants increase capacity without triggering permit requirements; • If Participants cannot increase capacity without triggering permitting); • n/a. Applicable Alternative Then TWG Alternative A1 applies Then TWG Alternative A2 applies Then TWG Alternative A3 applies Then TWG Alternative B applies. Applicable Requirements • Comply with 2009-2044 NO X Cap of 494,899 tons. • Permanently cease operation of all units if cumulative emissions before 2044 exceed 2009-2044 NO X Cap. • Permanently cease conventional coal-fired electricity generation by December 22, 2044. Additional Emission Cap • n/a • Comply with 2009-2029 NO X Cap of 416,865 tons. Specific Requirements * • By 12/31/19 permanently close 1 unit • By 12/31/19 permanently close 1 unit • By 12/31/19 reduce net generating capacity by no less than 561 MW • Temporarily cease operation if cumulative emissions before 2029 exceed 2009-2029 NO X Cap. • By 12/31/30 meet NO X limit of 0.07 lb/MMBtu on 2 units • By 12/31/19 operator may increase capacity by no more than 189 MW • By 12/31/30 meet NO X limit of 0.07 lb/MMBtu on 2 units • By 12/31/30 meet 0.07 lb/MMBtu on 2 units Reporting • By December 1, 2019, notify EPA of applicable Alternative (A1, A2, A3, or B). • Submit annual report summarizing heat input and annual and cumulative emissions of NO X . • Make annual report publicly available on Web Site. • Submit application to revise Part 71 Operating Permit by December 31, 2020. Additional Reporting • n/a • By 12/31/19 and annually thereafter submit Emission Reduction Plans to project year-by-year emissions to assure compliance with NO X Caps. * All units must comply with the existing NO X emission limit of 0.24 lb/MMBtu established in a 2008 permitting action. See discussion in Proposed Rule at 78 FR 8284 (February 5, 2013). This limit applies to each unit unless otherwise stated. In our final rule, EPA has included several revisions to the proposed regulatory text (40 CFR 49.5513(j)) put forth in the Supplemental Proposal. The substantive revisions include: 1. Revision to § 49.5513(j)(3) to clarify that EPA is finalizing a “better than BART” Alternative; 2. Additions to § 49.5513(j)(3) to specify that the operator must temporarily cease operation of NGS if cumulative emissions of NO X exceed the 2009-2029 NO X Cap of 416,865 tons at any time prior to December 31, 2029 (under Alternative B), and must permanently cease operation of NGS if cumulative emissions of NO X exceed the 2009-2044 NO X Cap of 494,899 tons at any time prior to December 31, 2044 (under all Alternatives); 3. Additions to § 49.5513(j)(3)(i)(A)(2), (B)(3), and (C)(2), to specify that the NO X emission limit of 0.07 lb/MMBtu is to be calculated based on a rolling average basis of 30 boiler operating days; 4. Correction to § 49.5513(j)(3)(ii)(D), to specify that Alternative B shall also apply if either of the Departing Participants (i.e., LADWP or NV Energy) remains as a participant in NGS; 5. Addition of § 49.5513(j)(3)(iii), consistent with the TWG Agreement, to require the owners of NGS to cease its operation of conventional coal-fired generation at NGS no later than December 22, 2044; 34 34 See page 14 of the TWG Agreement (section IV.F). This section of the TWG Agreement also states that “[a]t its election, consistent with the Lease Amendment, the Navajo Nation may continue plant operations at NGS after December 22, 2044 consistent with EPA approval.” EPA is not including this provision into the regulatory requirements at § 49.5513(j)(3)(iii), however, EPA expects that NGS would be substantially modified if the Navajo Nation elects to continue operation of the facility after NGS ceases conventional coal-fired generation in 2044, and that NGS must then meet all applicable regulatory and permitting requirements in existence at that time. 6. Addition to § 49.5513(j)(4)(ii), to change the annual reporting date to begin in 2015 instead of the specific date of January 31, 2015, and specify that the report must be submitted to EPA and also made publicly-available within 30 days of the submittal deadline associated with the annual emission inventory required by the Part 71 Operating Permit for NGS; 7. Addition to § 49.5513(j)(4)(iii), to clarify that the Part 71 Operating Permit for NGS shall incorporate practically enforceable limits for NO X of 0.24 lb/MMBtu, on a 30‐day rolling average basis, for each Unit equipped with LNB/SOFA, and 0.07 lb/MMBtu, on a rolling average basis of 30 boiler operating days, for each Unit equipped with SCR, as federally enforceable permit conditions; and 8. Addition of § 49.5513(j)(4)(iv)(C), to specify that the requirement to submit annual Emission Reduction Plans beginning no later than December 31, 2019, must be incorporated into the Part 71 Operating Permit for NGS as a federally enforceable permit condition. 9. Revision to § 49.5513(j)(7) to require the owner or operator of NGS to maintain records that document compliance with the NO X Cap (e.g., daily emissions and heat input data) for the life of the facility, rather than at least five years. 10. Deletion of § 49.5513(j)(7)(vi) that required record-keeping of all major maintenance activities conducted on emission units, air pollution control equipment, and CEMS because record-keeping of maintenance activities are not needed to ensure compliance with the 2009-2029 and 2009-2044 NO X Caps. 11. Revision to § 49.5513(j)(11) to state that the affirmative defense provisions of paragraphs § 49.5513 (c)(2) and § 49.5513(i) do not apply to paragraph § 49.5513(j). 35 35 We note that in our Supplemental Proposal, we reported the affirmative defense provisions as paragraphs (c)(1) and (g)(3) in error. The correct citations are to paragraph (c)(2) and paragraph (i) of 40 CFR 49.5513. Revision (1) above is necessary to clarify that EPA is finalizing a “better than BART” alternative in lieu of BART. The BART Benchmark used to assess the “better than BART” alternative is based on our proposed BART determination for NGS, and the “better than BART” alternative is consistent with our Supplemental Proposal of the TWG Alternative. Revision (3) above is necessary because EPA inadvertently did not specify the averaging period associated with the emission limits for NO X in our Supplemental Proposal. Revisions (2) and (4) through (10) above are in response to comments submitted to EPA on our Supplemental Proposal. Revision (11) above amends a proposed provision in our Supplemental Proposal that limited the applicability of the existing affirmative defense provisions for startups, shutdowns, and malfunctions (from the previous FIP for NGS codified at 40 CFR 49.5513(c)(2) and 40 CFR 49.5513(i)) to malfunctions. 36 In this Final Action, we are revising (j)(11) to make clear that the existing affirmative defense provisions do not apply to the emission limits established in the TWG Alternative. 36 See final action for the previous FIP for NGS at 75 FR 10179 (March 5, 2010). Following the close of the public comment period, the United States Court of Appeals for the DC Circuit issued a decision concerning various aspects of the NESHAP for Portland cement plants issued by EPA in 2013, including the affirmative defense provision of that rule. 37 The court found that EPA lacked authority to establish an affirmative defense for private civil suits and held that under the CAA, the authority to determine civil penalty amounts lies exclusively with the courts, not EPA. The court did not address whether such an affirmative defense provision could be properly included in a SIP. However, the court's holding makes it clear that the CAA does not authorize promulgation of such a provision by EPA. In particular, the court's decision turned on an analysis of CAA sections 113 (Federal enforcement) and 304 (Citizen suits). These provisions apply with equal force to a civil action brought to enforce the provisions of a FIP. The logic of the court's decision thus applies to the promulgation of a FIP and precludes EPA from including an affirmative defense provision in a FIP. Therefore, we are not including an affirmative defense provision in the final FIP. 37 NRDC v. EPA, 2014 U.S. App. LEXIS 7281 (D.C. Cir.), in the docket for this rulemaking. We note that, if a source is unable to comply with emission standards as a result of a malfunction, EPA may use case-by-case enforcement discretion, as appropriate. Further, as the DC Circuit recognized, in an EPA or citizen enforcement action the court has the discretion to consider any defense raised and determine whether penalties are appropriate. 38 38 Id. at 24 (arguments that violations were caused by unavoidable technology failure can be made to the courts in future civil cases when the issue arises). IV. Summary of Major Issues Raised by Commenters The public comment period for our Proposed Rule opened on February 5, 2013. On two occasions, we extended the comment period on our Proposed Rule at the request of stakeholders, with a final closing date of January 6, 2014. Although we posted the pre-publication version of our Supplemental Proposal to the docket and to our Web site on September 25, 2013, the public comment period for the Supplemental Proposal officially began when it was published in the Federal Register on October 22, 2013. 39 We accepted public comments on our Supplemental Proposal, concurrently with our Proposed Rule, until January 6, 2014. Our Supplemental Proposal also included notice of five public hearings, one on the Navajo Nation, one on the Hopi reservation and three in the State of Arizona. The public hearings occurred during the week of November 12, 2013. In all, 194 oral testimonies were presented at the public hearings. 39 See document number 0182 (Pre-publication version of Supplemental Proposal for NGS Signed on September 25, 2013), posted to docket on September 25, 2013 and publication of Supplemental Proposal in Federal Register at 78 FR 62509 (October 22, 2013). We received over 77,000 written comments. Of these, over 76,800 comments came from private individuals who submitted substantially similar comments by email or postcard. We received an additional 300 unique written comments (not including duplicates, requests for extension of the public comment period, or requests for additional hearings) from a variety of individuals and entities, including tribal governments, environmental or public interest advocacy groups, water interest groups, groups representing industry or commerce, the operator and participants in NGS and the Kayenta Mine, elected officials, and state and local governments. In this document, EPA is providing an abbreviated summary of the major comments and EPA's responses to those comments, grouped together by subject matter. The complete response to comments document (RTC) includes the full summary of all substantive comments and EPA's full responses to those comments. The RTC is included in the docket for this rulemaking. 40 We are not responding to comments unrelated to our Proposed Rule or Supplemental Proposal for NGS in this document or in the RTC. 40 See document titled “EPA Responses to Comments on Final Rule for NGS” in the docket for this rule. A. General Comments From Public Hearings Comment: Contribution of NGS to the local and state economy and support for TWG Alternative Many commenters at the public hearings preferred the TWG Alternative because they believe that EPA's proposed BART determination would force NGS and the Kayenta Mine to close, causing economic harm to an area where the majority of residents are low-income and where opportunities for employment are limited. Many commenters stressed that NGS employs over 500 people and the Kayenta Mine has over 400 employees, and the loss of these jobs would only exacerbate the unemployment rate in the area, which currently ranges from 47 percent to 60 percent. A number of commenters noted that NGS supplies more than 90 percent of the energy used by Central Arizona Water Conservation District (CAWCD), which operates the Central Arizona Project (CAP), which transfers water from the Colorado River throughout Arizona. A few commenters urged EPA to uphold its federal trust obligations and ensure that tribal communities continue to have access to affordable water, and advised EPA to make a decision consistent with the legal rights that the Gila River Indian Community and other stakeholders negotiated and that Congress granted under the Arizona Water Settlements Act of 2004. A few commenters support the TWG Alternative because they believe it is a fair compromise created by a diverse group of stakeholders that provides a path for future operation at NGS by allowing for potential ownership changes and by providing an extension to install SCR technology, while still ensuring that the total emission reductions of NO X will be greater than those achieved under EPA's proposed BART determination. Response: EPA recognizes the contribution of NGS and the Kayenta Mine to the economy of the Navajo Nation, the Hopi Tribe, the city of Page, and the state of Arizona. In our Proposed Rule, EPA discussed the history of NGS and the relationship between NGS, the Central Arizona Project, and numerous tribes located in Arizona. 41 EPA notes that NGS is a facility that is subject to the BART requirement of the RHR, and emissions from NGS affect visibility at 11 national parks and wilderness areas in the Southwest. The analyses in our Proposed Rule and Supplemental Proposal determined that additional controls at NGS are cost-effective, will significantly reduce the contribution of NGS to visibility impairment at numerous Class I areas, and should not cause NGS to retire. However, for a number of reasons, including the importance of NGS to numerous Indian tribes located in Arizona and the federal government's reliance on NGS to meet the requirements of water settlements with several tribes, EPA also outlined a framework for considering “better than BART” alternatives that ensures emission reductions while providing additional flexibility to the operator of NGS. 42 41 See 78 FR 8274, at 8275 (February 5, 2013). 42 Id. and 78 FR 62509 (October 22, 2013). EPA agrees with comments that the TWG Agreement represents a compromise between diverse stakeholders, although we recognize that the members of the TWG did not invite all affected stakeholders to participate in their discussions. The TWG Alternative provides certainty for future operation of NGS, flexibility in the compliance timeframe, and more emission reductions of NO X than would have been achieved under EPA's proposed BART determination. Based on our analysis in our Supplemental Proposal and consideration of all comments received, EPA is taking action to finalize requirements consistent with the TWG Agreement we put forth in our Supplemental Proposal, i.e., the TWG Alternative. Comment: Impact of air pollutants from NGS on public health and welfare and support for proposed BART determination . Several commenters favor EPA's proposed BART determination for NGS because they believe that emissions from NGS cause health problems in the area, including respiratory illness and heart disease. One commenter cited a Clean Air Task Force study which states that NGS is responsible for approximately $127 million in health costs every year. Many of these commenters urged EPA to conduct health studies to determine the actual impact to health in these communities. Some commenters favor stringent controls because they believe that emissions from NGS adversely affect native plant species and harm traditional dry land farming. Others assert that emissions from NGS can be linked to high levels of mercury found in fish species located in nearby lakes. Many commenters expressed concerns over the well-being of the Navajo Aquifer. A number of commenters favor stringent controls because they believe that emissions produced from NGS contribute to climate change. In contrast, a few commenters questioned the extent to which emissions from NGS impact public health and the environment, asserting that the haze is a result of emissions from natural sources (e.g., volcanoes, wind/dust storms, and forest fires) and pollution produced from nearby cities (i.e., Phoenix, Los Angeles, and Las Vegas). Another commenter asserted that EPA's Web site states that vehicles are the largest producers of NO X emissions in the country and concludes that EPA is ignoring mobile sources and unfairly targeting stationary sources. Some commenters preferred EPA's proposed BART determination over the TWG Alternative because they believe that the alternative is based on a false premise. They asserted that the closure of a single unit is not equivalent to cleaning up all three units because the reduction in capacity will ultimately require new electricity generation elsewhere because the demand for power does not change. Response: Protection of human health and the environment is EPA's mission and forms the basis for many Agency actions, including establishing the National Ambient Air Quality Standards (NAAQS), and promulgation of regulations such as the New Source Performance Standards (NSPS) and the National Emission Standards for Hazardous Air Pollutants (NESHAP). In addition to Clean Air Act requirements to protect human health, in the 1977 Clean Air Act Amendments, Congress declared as a national goal the prevention of any future, and the remedying of any existing, impairment of visibility in mandatory class I Federal areas which impairment results from manmade air pollution (See CAA § 169A). EPA agrees that visibility-impairing pollutants are among the same pollutants that affect human and ecosystem health; however, health studies are beyond the scope of this BART analysis. Similarly, hazardous air pollutants (HAPs), such as mercury, are not visibility-impairing pollutants and therefore are beyond the scope of this BART analysis. 43 43 Emissions of HAPs from various source categories are addressed generally through the NESHAP. EPA addressed mercury emissions from power plants specifically in the final Mercury and Air Toxics Standard (MATS). 77 FR 9304 (February 16, 2012). EPA agrees that climate change is an important issue. 44 However, the RHR addresses pollutants that impair visibility and is not intended to address pollutants that contribute to climate change. EPA has developed various programs and activities to address emissions of greenhouse gases. 45 On June 2, 2014, EPA signed a proposal to cut greenhouse gas emissions from coal-fired power plants by up to 30 percent by 2030. 46 Although regulation of greenhouse gases is conducted under separate statutory requirements from regional haze, EPA is mindful that this BART determination for NGS is not the only regulatory program that affects this facility and the region. 44 http://www.epa.gov/climatechange/basics/. 45 http://www.epa.gov/climatechange/EPAactivities.html. 46 See http://www2.epa.gov/carbon-pollution-standards. EPA agrees with comments that mining and combustion of coal affect the environment. EPA notes that Reclamation has started its process to develop an Environmental Impact Statement (EIS) required under the National Environmental Protection Act (NEPA) for activities resulting from the continued operation of NGS and the Kayenta Mine. 47 The on-going NEPA process provides numerous opportunities and the appropriate forum to raise concerns related to the impacts of mining and use of water from the Navajo Aquifer. We further note that representatives of DOI attended all the public hearings on NGS held by EPA and are aware of the issues raised by commenters during the BART process regarding mining and the Navajo Aquifer. 47 For more information, please see www.ngskmc-eis.net. EPA disagrees with the assertion that EPA is unfairly targeting stationary sources of emissions and ignoring the significant contribution of motor vehicle emissions. Consistent with title II of the CAA, the EPA Office of Transportation and Air Quality protects public health and air quality by, among other things, regulating air pollution from motor vehicles, engines, and the fuels to operate them. 48 New cars and sport utility vehicles sold today have emission levels of hydrocarbons, NO X , and carbon monoxide that are 98-99 percent lower than new vehicles sold in the 1960s on a per mile basis. 49 Similarly, standards established for heavy-duty highway and non-road sources require emission rate reductions on the order of 90 percent or more for particulate matter and NO X . In 2014, EPA finalized new vehicle emission standards and reduced the fuel sulfur content of gasoline to achieve additional reductions in tailpipe and evaporative emissions from passenger cars, light-duty vehicles, medium-duty passenger cars, and some heavy-duty vehicles starting in 2017. 50 48 http://www.epa.gov/otaq/. 49 See, e.g., 76 FR 74854, at 74900 (December 1, 2011). 50 See Fact Sheet for Tier 3 Standards available at http://www.epa.gov/otaq/documents/tier3/420f14010.pdf . EPA agrees that forest fires and volcanic eruptions, when they occur, can impact visibility to a greater extent than anthropogenic sources of emissions. However, Congress directed EPA to develop rules to address on-going emissions from stationary sources subject to BART to remedy the existing impairment of visibility in Class I areas and restore visibility to natural conditions. EPA disagrees with assertions that the TWG Alternative is based on a false premise because the closure or curtailment of one unit would just result in electricity being produced elsewhere. Closure of one unit at NGS or the curtailment of an equivalent amount of electricity generation is possible based on LADWP and NV Energy's intended divestiture from NGS. Consistent with state law in California and Nevada, additional electricity needed to replace lost generation from NGS, associated with LADWP and NV Energy's divestiture, would come from energy sources that emit less air pollution than a conventional coal-fired power plant operating with SCR on all units. 51 51 See RTC and references therein. Comments regarding specific aspects of the TWG Alternative are discussed in Section 9.0 of the RTC. Comment: Environmental and Social Justice. Several commenters consider the presence of NGS and several other power plants in and around the Navajo Nation to represent an environmental and economic justice issue. One commenter noted that a Navajo water hauler in Kaibeto, a Navajo community near Page, pays 10 to 20 times more for water, or $13,000 per acre foot, than municipal CAP water users in Glendale or a farmer in Tempe, who pay $551 and $41 per acre feet, respectively. Several commenters opined that the leaders of the Navajo Nation and EPA have not protected the interests of the local population. A few expressed concerns over how the alternatives were written, noting that many tribal residents do not understand the technical language used in the documents and therefore cannot adequately comment on the validity of the alternatives proposed. Some commenters argued that pollution can be controlled using existing technology and EPA should apply the same standard to NGS as other coal-burning power plants (e.g., Four Corners Power Plant). A few commenters argued that extending the compliance timeframe for NGS demonstrates that the federal government considers itself exempt from federal law. Several argued that tribal communities do not have the funds to develop proposals and/or conduct environmental assessments and urged that EPA uphold federal trust responsibilities and create an equal playing field. Response: EPA defines Environmental Justice as “the fair treatment and meaningful involvement of all people regardless of race, color, national origin, or income with respect to the development, implementation, and enforcement of environmental laws, regulations, and policies. EPA has this goal for all communities and persons across the country. It will be achieved when everyone enjoys the same degree of protection from environmental and health hazards and equal access to the decision-making process to have a healthy environment in which to live, learn, and work.” 52 52 http://www.epa.gov/environmentaljustice/. EPA takes fair treatment and meaningful involvement seriously and provided numerous opportunities for tribal governments, environmental and tribal non-governmental organizations, and other interested stakeholders to provide input in the development of our Proposed Rule, Supplemental Proposal, and Final Rule for NGS. EPA began our public involvement process for a BART determination for NGS in 2009, when we published an Advanced Notice of Proposed Rulemaking (ANPR). Although we initially provided a 30-day public comment period, at the request of tribal governments and other interested stakeholders, we extended the comment period for tribes another 30 days to October 28, 2009 and, to allow additional time for government-to-government consultation on NGS, agreed to accept comments from tribes until March 1, 2010. EPA received over 6,000 comments on the ANPR. 53 During 2009 through 2012, EPA met with various stakeholders, including tribal governments and tribal environmental groups, to discuss NGS and hear concerns related to a BART determination for this facility. 54 We initially provided a 90-day comment period for the Proposed Rule on February 5, 2013, and at the request of various stakeholders, we provided several extensions of the public comment period, which closed on January 6, 2014. During the 11-month comment period, EPA continued to meet with stakeholders, at their request, to discuss our proposed BART determination for NGS and our framework for “better than BART” alternatives. 55 53 See page 25 of the TSD to the February 5, 2013 Proposed Rule. 54 See, for example document number 0232 in the ANPR docket at EPA-R09-OAR-2009-0598, and document numbers 0008 and 0009 in the docket for this rule. 55 See, for example, document number 0150, 0152, 0166, 0173, 0302, and 0303 in the docket for this rule. On July 26, 2013, the TWG submitted Appendix B to the TWG Agreement to EPA for consideration. EPA posted the TWG Agreement to our docket on the same day to provide the public an opportunity to review it. 56 On September 25, 2013, EPA posted our Supplemental Proposal, along with supporting documents, to the docket to allow for pre-publication review by interested parties. 57 The Supplemental Proposal was published in the Federal Register on October 22, 2013. The comment period for the Supplemental Proposal closed on the same day as the BART proposal, on January 6, 2014. The Supplemental Proposal also included notice of five open house and public hearing events EPA scheduled throughout Arizona in November 2013. The open houses allowed members of the public an opportunity to talk with representatives from EPA and ask questions. EPA held events at the LeChee Chapter House, located on the Navajo Nation, as well as in Page, Arizona, and provided oral interpretation services between English and Diné (the Navajo language). EPA also held an event at the Hopi Day School, located in Kykotsmovi, the seat of the Hopi tribal government. 58 Finally, we also held events in Phoenix and in Tucson, Arizona, to allow stakeholders in central and southern Arizona, representing CAP water interests and several tribes receiving CAP water, the opportunity to provide comment and talk with representatives from EPA. Although EPA understands that the TSD and Federal Register notices include technical information that may be difficult to understand, EPA provided Fact Sheets and handouts, written in plain language, at the open house and public hearing events. 59 EPA representatives were also present at the events to discuss and explain our Proposals. EPA recognizes that many tribal communities do not have the funds to develop alternative proposals or hire experts on their behalf; however, this does not diminish such communities' ability to participate in the rulemaking process in a meaningful way as EPA takes seriously its responsibility to explain its proposal to all interested parties and assesses all comments, regardless of the form of the comment or whether or not the commenter has a technical background. 56 See document number 0122 in docket for this rule. 57 See document numbers 0182, 0183, and 0184 in the docket for this rule. 58 EPA engaged with the government of the Hopi Tribe to search for an oral interpreter between English and the Hopi language, but the Hopi Tribe was unable to locate anyone to provide those services. 59 See document 0219 in the docket for this rule. As stated in our Proposed Rule and Supplemental Proposal, EPA has determined that these proposed rules, if finalized, will not have disproportionately high and adverse human health or environmental effects on minority or low-income populations because they increase the level of environmental protection for all affected populations (i.e., require emission reductions from NGS). 60 EPA recognizes that some commenters may view the timeframe for compliance under EPA's framework for BART Alternatives as an environmental justice issue. We note that the LNB/SOFA credit, an important component of the extended timeframe, was based on real, actual emission reductions beginning in 2009 that were voluntary and not required by any rule or regulation. We also note that the TWG Alternative, which calls for closure of one unit in 2019 (or equivalent curtailment) will result not only in greater reductions of NO X than would have been achieved under BART, but also reductions of several other pollutants, including SO 2 , PM, CO 2 , and mercury. Thus, although the TWG Alternative includes a compliance timeframe for achieving additional reductions in 2030, over 2009-2044, the TWG Alternative will result in reductions of additional pollutants that affect visibility or human health, and will provide an enforceable mechanism to ensure that NGS ceases conventional coal-fired electricity generation at NGS by the end of 2044. 60 See discussions under Executive Order 12898 in 78 FR 8793 (February 5, 2013) and 78 FR 62520 (October 22, 2013). EPA recognizes that numerous commenters expressed frustration regarding social inequities related to costs and benefits of coal mining and combustion and water availability and cost. We recommend participating in the EIS process for NGS and Kayenta Mine to raise any concerns related to costs, benefits, and the environmental and social justice of coal mining and coal combustion at the Kayenta Mine and NGS. B. Comments on Factor 1—Cost of Controls Comment: EPA underestimated SCR costs. Several commenters asserted that EPA underestimated the cost of compliance by improperly reworking cost estimates developed for SRP by Sargent and Lundy (S&L) in 2010 and disregarding real costs that would be incurred. One commenter quoted the BART Guidelines and the final RHR to assert that although the use of the Control Cost Manual is encouraged, it is not mandated, and that EPA has discretion to use additional sources of cost information. The commenter believes, therefore, that the SRP estimates for the excluded cost items are appropriate to use because they are more precise than the generic statements that EPA relied upon in the Control Cost Manual. Response: EPA disagrees with the comment that we improperly reworked and underestimated the SCR cost estimates. We note, however, that even if we had relied only on the cost estimate provided by SRP, EPA still would have concluded that SCR is cost-effective at NGS. EPA used a hybrid approach for our cost analysis that relied primarily on the cost estimates provided by SRP, but also followed the BART Guidelines to determine whether S&L included cost estimates for services or equipment associated with SCR that were not allowed under the EPA Control Cost Manual. The BART guidelines state “[i]n order to maintain and improve consistency, cost estimates should be based on the OAQPS Control Cost Manual, where possible”. 61 The capital cost estimate EPA presented in the proposed rulemaking for SCR plus LNB/SOFA ($541 million total for Units 1-3) is only 8 percent lower than the SRP cost estimate ($589 million). SRP's cost estimate would not have changed our conclusion that SCR is cost-effective at NGS. 61 The OAQPS Control Cost Manual is now referred to as the EPA Control Cost Manual. As discussed in the TSD to the proposed rulemaking, EPA made four adjustments to SRP's cost estimates for SCR, namely, to exclude “Owners Construction Management, O&M Support and Contract Service,” “Owners Legal Support and Insurance,” and “Allowance for Funds Used During Construction,” and to use an interest rate of 7 percent. 62 Our detailed, line-by-line analysis was included in the docket for this proposed rulemaking and provided an explanation for why we retained, modified, or rejected each line item. 63 Please see the RTC for additional discussion of these four adjustments to the S&L cost analysis. 62 See Table 12 of the TSD to the February 5, 2013 Proposed Rulemaking, available as document number 0014 in the docket for this rule. 63 See MS Excel document titled “EPA cost analysis for NGS” within document number 0004 in the docket for this rule. In our proposed rule, we presented total capital and total annual cost estimates from EPA and SRP, as well as average and incremental cost-effectiveness values based on EPA and SRP assumptions for total annual cost and total annual NO X reductions. Based on SRP's analysis, average cost-effectiveness of SCR+LNB/SOFA at NGS was less than $3,000 per ton and incremental cost-effectiveness of SCR+LNB/SOFA (compared to SNCR+LNB/SOFA) was approximately $5,300 per ton. 64 EPA stated that the cost-effectiveness values calculated by both EPA and SRP for SCR+LNB/SOFA are lower than or within the range of other BART evaluations where EPA or a state has determined that SCR is BART (ranging from approximately $2,000 to $6,000 per ton). EPA has accordingly determined that SCR is cost-effective at NGS. 65 Therefore, even if EPA accepted the S&L cost estimates submitted by SRP, as commenters suggest, EPA would still have determined that SCR is cost-effective for NGS. 64 See Table 3 of our Proposed Rule, 78 FR 8281 (February 5, 2013). 65 See our Proposed Rule at 78 FR 8281 (February 5, 2013). Comment: EPA overestimated SCR costs. One commenter asserted that EPA overestimated the cost of installing SCR at NGS. Although the commenter supported EPA's adjustments to the S&L cost estimates, the commenter asserted that further revisions are appropriate. The commenter stated that EPA overestimated the following costs: Outage costs associated with installation and “preinstallation” work; catalyst costs; and auxiliary power. In addition, the commenter asserted that EPA overestimated annual costs by assuming 20 years as the basis for amortizing costs and using an inflated interest rate of 7 percent. Although the commenter concurs with EPA's conclusion that SCR plus LNB/SOFA is cost-effective at $2,240 per ton of NO X removed, the commenter re-calculated cost-effectiveness to be $1,412 per ton for Unit 1, $1,331 per ton for Unit 2, and $1,497 per ton for Unit 3. Response: EPA disagrees with the commenter that any revisions to EPA's estimate of SCR costs are necessary. Even if some of the costs projected by S&L and used by EPA may be overestimated (e.g., the commenter points primarily to capital recovery, catalyst replacement costs, and costs for lost power generation), EPA disagrees that we must correct every issue of concern raised by the commenters in order to support our determination of the BART Benchmark. EPA made four specific corrections to the estimates provided by S&L and SRP to make the cost calculation methodology consistent with methodologies used for BART cost calculations nationally. 66 As noted in other responses even if we consider the average and incremental cost effectiveness of SCR using SRP and S&L's full cost projections, EPA would still determine that SCR at NGS is cost-effective. The cost-effectiveness values cited by the commenter, below $1,500 per ton, certainly suggest that SCR could be even more cost-effective than the values we relied upon in our proposal, but this would not change our overall determination that SCR is cost-effective for NGS. 66 See, e.g., Final Regional Haze Plan for Arizona (Phase 1) at 77 FR 72512 at 72531 (December 5, 2012); Final Regional Haze Plan for North Dakota at 77 FR 20894 at 20916-17 (April 16, 2012); Final Regional Haze Plan for New Mexico at 76 FR 52388 at 52399-52400 (August 22, 2011); Final Regional Haze Plan for Wyoming at 79 FR 5032 at 5082 (January 30, 2014). Comment: Updated SCR cost estimate from SRP. SRP contracted with S&L in 2013 to review and update the SCR cost estimates that were prepared in 2010. S&L escalated costs for inflation, and incorporated other minor adjustments to reflect a lower NO X design target. SRP's revised capital cost estimates for SCR installation on all three units total $650 million (in 2013 dollars) compared to SRP's 2010 cost estimate of $544 million. Response: EPA reviewed the updated 2013 cost estimates developed by S&L and provided by SRP. 67 In its 2013 cost report, S&L explains that it escalated labor and material costs, and updated cost estimates based on a revised design target of 0.03 lb/MMBtu (so that the SCR system is deployed as a 3+1 system rather than a 2+2 catalyst layer system), and other design features, including a low-load temperature control system to operate SCR at lower loads. S&L escalated several costs at rates above 6.7 or 8 percent (e.g., freight, scaffolding). S&L did not make any revisions to the components of variable annual costs, including maintenance labor, auxiliary power, steam, and catalyst replacement. To be consistent with the cost estimates in our Proposed Rule, EPA accepted most of the line item costs as adjusted by S&L and made the same four adjustments to the 2013 cost estimates as we had applied to the 2010 cost estimates. These changes result in an 8 percent difference in total capital costs of SCR between EPA's 2013 estimate and SRP's 2013 estimate and a 21 percent difference in the total annual costs of SCR between the 2013 estimates from EPA and SRP (see Table 2). 67 See RTC and references therein. Table 2—Cost Estimates for SCR in 2010 and 2013 Dollars Total capital cost (million) in 2010$ Total capital cost (million) in 2013$ Total annual cost (million) in 2010$ Total annual cost (million) in 2013$ EPA Estimate $496 $598 $59 $69 SRP Estimate 544 650 75 88 In our proposed BART determination, EPA also presented the average and incremental cost-effectiveness of controls, based on the combination of combustion controls (LNB/SOFA) and post-combustion controls (i.e., SNCR or SCR). Therefore, cost-effectiveness values presented in our Proposed Rule were based on total annual cost of SCR in combination with annual cost of LNB/SOFA (SCR+LNB/SOFA), SNCR in combination with LNB/SOFA (SNCR+LNB/SOFA) or LNB/SOFA alone. 68 Based on the updated 2013 cost estimates for SCR, Table 3 shows the average and incremental cost-effectiveness of controls, in both 2010 and 2013 dollars, based on EPA and SRP assumptions for total annual cost and annual NO X reductions achieved by SCR. See RTC for further detail on cost-effectiveness of SNCR+LNB/SOFA and LNB/SOFA. 68 78 FR 8281, February 5, 2013. Table 3—Cost Effectiveness of Controls in 2010 and 2013 Dollars 2010 $ EPA SRP 2013 $ EPA SRP SCR+ LNB/SOFA: Total Annual Cost * $67.5 million $80.2 million $74.4 million $92.6 million. Annual NO X reduced (tpy) 28,573 26,180 28,573 26,180. NO X Limit (lb/MMBtu) 0.055 0.080 0.055 0.080. Average Cost Effectiveness ($/ton) $2,369 $3,069 $2,605 $3,537. Incremental Cost Effectiveness (vs. LNB/SOFA) ($/ton) $3,522 $4,889 $3,899 $5,695. Incremental Cost Effectiveness (vs. SNCR+LNB/SOFA) ($/ton) $3,239 $5,357 $3,798 $6,647. * EPA's cost effectiveness calculations represent SCR in combination with LNB/SOFA, rather than SCR alone. Based on the revised 2013 cost estimates for SCR+LNB/SOFA, the revised average cost-effectiveness of SCR+LNB/SOFA is roughly 10 percent higher (based on EPA's estimates) than the average cost-effectiveness values reported in our Proposed Rule, and roughly 15 percent higher based on SRP's estimates. 69 The 2013 values for average cost-effectiveness of SCR+LNB/SOFA based on EPA and SRP estimates are still comparable to the range of values determined cost-effective for SCR in other BART determinations. For these reasons, EPA continues to consider SCR+LNB/SOFA as cost-effective at NGS. 69 For informational purposes, EPA included the incremental cost-effectiveness values of SCR+LNB/SOFA (estimated in 2010 and 2013) compared to LNB/SOFA and SNCR+LNB/SOFA, but we note that a comparison of the percent change in incremental cost-effectiveness between 2010 and 2013 is not informative because SRP did not provide updated cost estimates (in 2013 dollars) for the other control technologies. Comment: Cost-Effectiveness of Presumptive BART. One commenter stated that in establishing presumptive limits in the BART Guidelines, EPA recognized that SCR is not cost-effective and that combustion controls such as LNB/SOFA represent the most cost-effective control options for most boiler types. The commenter pointed out that in establishing presumptive limits, EPA considered controls that cost less than $1,500 per ton to be cost-effective, and that the cost-effectiveness for SCR at NGS, which ranges from $3,000 to $6,000 per ton based on 2010 estimates, is well above this threshold. The commenter concluded that EPA should have rejected SCR and proposed LNB/SOFA as BART for NGS. Response: EPA disagrees with the assertion that the BART Guidelines established a threshold for cost-effectiveness against which all future BART determinations must compare. In developing the presumptive NO X limits for BART in 2005, EPA did not set the cost-effectiveness values estimated for combustion controls as the threshold for determining whether a given control technology was or was not cost-effective. If EPA had intended the cost-effectiveness values estimated in 2005 to represent a threshold for BART, it is reasonable to assume that the BART Guidelines would have included those cost-effectiveness values as thresholds in Appendix Y, and would have required future cost estimates to be presented in 2005 dollars for appropriate comparison to the thresholds. The BART Guidelines do not set a numerical definition for “cost-effective”, and the analysis of presumptive limits uses cost-effectiveness as a means to broadly compare control technologies, not as a threshold for rejecting controls for an individual unit or facility that exceed the average cost-effectiveness of combustion controls. In addition, as discussed in the RTC, a value of $1,500 per ton is not an appropriate or relevant value for determining cost-effectiveness. Comment: Indirect costs should also be considered under Factor 1. The Gila River Indian Community asserted that EPA conducted the analysis of cost-effectiveness incorrectly by not including the indirect costs of the requirements and only considering the direct cost of the requirements. The commenter stated that EPA did not give sufficient consideration to the high costs to tribes associated with indirect impacts of its proposed BART determination. Response: EPA disagrees with the assertion that our cost-effectiveness analysis was incorrect because it did not include indirect costs in the assessment of the costs of compliance. The BART Guidelines, which States and EPA must follow in BART determinations for fossil-fuel fired power plants greater than 750 MW, 70 focus on the direct costs of the pollution control equipment and other capital and annual costs associated with the control technology alternatives. The BART Guidelines do not require consideration of the cost of potential indirect effects of BART control options when assessing the costs of compliance. Therefore, EPA disagrees that our analysis for Factor 1 was incorrect or incomplete because it did not include indirect costs to tribes. EPA further notes that under Factor 2, the energy and non-air quality environmental impacts analysis, the BART Guidelines specifically require the energy impact analysis to consider direct energy impacts (e.g., parasitic load from certain control technologies) and to generally exclude indirect energy impacts of controls (e.g., energy to produce raw materials for construction of control equipment) unless the indirect impact is unusual or significant. 70 40 CFR 51.308(e)(1)(ii)(B). However, because of the unique relationship between NGS, tribes, and tribal water settlement agreements, and to inform our government-to-government consultation with tribes, EPA did consider potential indirect effects of control options to tribes under Factor 2. EPA quantified the impact to electricity rates and CAP water rates, and also assessed whether installation of SCR would result in electricity generation costs at NGS that exceed the cost to purchase power on the wholesale market. Therefore, although EPA appropriately did not consider indirect costs in our analysis of Factor 1, EPA did include consideration of indirect impacts to tribes and other entities in our analysis of Factor 2. C. Comments on Factor 2—Energy and Non-Air Quality Environmental Impacts, Including Economic Impacts Comment: EPA's Affordability Analysis relied on invalid assumptions. One commenter submitted a report, prepared by Management Information Services, Inc. (MISI report), asserting that EPA made several assumptions that underestimated the cost of continuing to operate NGS with additional controls, including the assumption that no new capital would be deployed at NGS over the next 25 years, the assumption that the increase in the annual NGS lease cost would be $15 million per year (which is lower than actual increase in lease cost of $43 million per year that was released after publication of our Proposed Rule), and the use of EPA's capital cost estimates for SCR instead of the cost estimated by S&L. Other commenters asserted that EPA underestimated the cost of closing NGS and purchasing power on the wholesale market, by not accounting for costs associated with stranded investments and decommissioning NGS. Response: EPA recognizes the economic importance of NGS to the State of Arizona, the Navajo Nation, and the Hopi Tribe. The purpose of the Affordability Analysis in our docket was to determine whether the control options for BART would have a detrimental impact on the competitiveness of NGS in the western power market, affecting whether the NGS owners would continue to operate NGS or replace NGS generation with less expensive market power. The Affordability Analysis indicated that, even if SCR installation was required on all three units at NGS, power produced at NGS would remain less expensive than the cost to replace power through wholesale purchases. Because utilities will generally provide power to their customers in a least-cost manner and because NGS, with the installation and operation of SCR, remained the less expensive option, EPA determined that the operation and installation of SCR, in and of itself, was not likely to force NGS to close. In response to multiple comments expressing concern related to simplifying assumptions or outdated data, EPA updated the Affordability Analysis with the most current power market price curves from the U.S. Energy Information Administration (EIA) and recent forward power market prices in March 2014 and other more current modeling variables. These revisions are discussed in more detail in the RTC as well as in additional supporting documents. 71 The updated model results, comparing the net present value (NPV) of electricity generation costs with air pollution controls installed compared to the costs to purchase an equivalent amount of power on the wholesale market, are summarized in the RTC. Overall, the combined changes do not change the conclusions from the original Affordability Analysis that installing and operating SCR at NGS would be less costly than closing NGS and purchasing replacement power from the wholesale market. 71 See RTC and references therein. Comment: EPA's failure to appropriately consider the impacts to non-Indian agricultural (NIA) water users renders its Factor 2 analysis arbitrary, capricious, and an abuse of discretion. One commenter stated that, as a result of errors and omissions, EPA's Factor 2 analysis is arbitrary, capricious, and an abuse of discretion. The commenter asserted that there are several problems with the EPA analysis related to NIA users of CAP water, including erroneous assumptions, insufficient support for conclusions, failure to consider decreased farming profitability and increased unemployment, failure to acknowledge the inability of NIA water users to pass along cost increases as compared to municipal users, and other factors. Response: EPA recognizes that CAP water is an important resource for NIA and other users of water in Arizona. As a result, as one of a number of discretionary analyses EPA conducted on the indirect impacts on major stakeholders, EPA calculated water rate increases to NIA users of CAP water and municipal and industrial users of CAP water. EPA disagrees that our discussion of impacts to NIA users of CAP water renders our Factor 2 analysis arbitrary, capricious, and an abuse of discretion. Neither the CAA nor the BART Guidelines require consideration of indirect costs or indirect impacts of controls in a BART analysis. EPA, nevertheless, included an evaluation of impacts to some of the major stakeholders in NGS in our BART analysis under Factor 2, including NIA users, as consistent with the statement in the BART Guidelines that “the energy impacts analysis may consider . . . whether a given alternative would result in significant economic disruption or unemployment” (emphasis added). 72 72 See BART Guidelines at 70 FR 39169 (July 6, 2005). EPA recognizes that the information we had available to us about NIA users of CAP water was limited, and we acknowledged in the TSD to our Proposed Rule that we had several questions about CAP and groundwater availability to NIA water users. EPA appreciates the clarifications and additional information provided by NIA users of CAP water during the comment period for our proposals. The additional information provided during the comment period about NIA users of CAP water does not change our conclusion under Factor 2, that the potential economic impacts to tribes argue for flexibility in the compliance timeframe for NGS, because this compliance flexibility also benefits other stakeholders, including the NIA users of CAP water. Comment: EPA must evaluate cumulative economic impact of other rulemakings. One commenter asserted that the BART proposal must take into account the context in which the regional haze rules are being implemented and conduct a cumulative impact analysis of all EPA rulemakings. The commenter noted that the two remaining copper smelters in Arizona are already subject to BART for SO 2 and they also have to make significant capital investments to comply with other regulatory programs and initiatives such as the revised SO 2 NAAQS. Response: EPA disagrees with the assertion that we must consider the total cost impact of all EPA regulatory requirements in a BART analysis. EPA recognizes that other facilities, whose water and electricity rates may be affected by our BART determination for NGS, may also be subject to BART for their own emissions of visibility-impairing pollutants. As a general matter, EPA is mindful that facilities may be affected by multiple regulatory and program activities. We note that BART is a case-by-case determination that is based on a source-specific analysis of five factors, which include considerations of the unique circumstances of each affected facility, as required under the CAA. Comment: Impact to the Development Fund. One commenter stated that the increased cost of electricity generation associated with SCR would reduce the competitiveness of the price of NGS power on the wholesale market and therefore reduce the revenue that flows into the Development Fund. Response: As discussed in our Proposal Rule and TSD, EPA recognizes that any electricity owned by Reclamation based on its 24.3 percent participation in NGS that is not used by CAP is sold and revenues are deposited into the Development Fund. 73 This fund is authorized to pay the delivery portion of the cost of CAP water for certain Indian tribes and to pay the cost of constructing delivery systems to bring CAP water to certain Indian tribes. 74 EPA considers the potential economic impacts to tribes, including potential impacts to the Development Fund, as part of BART factor 2 to support the appropriateness of flexibility in the compliance timeframe for NGS. 73 See Proposed Rule at 78 FR 8282 (February 5, 2013) and TSD at pages 71-72. 74 See Proposed Rule at 78 FR 8283 (February 5, 2013). Comment: No basis for public health claim. One commenter asserted that EPA has no basis for claiming that the NO X reductions from NGS would lead to a public health benefit. The commenter noted that EPA establishes NAAQS at levels that are protective of public health and welfare with an adequate margin of safety that accounts for sensitive populations such as children and the elderly, and that EPA has never found that any of the areas around NGS fail to attain the NAAQS. The commenter asserted that EPA must conduct a health risk evaluation that follows the four basic steps of the risk assessment process: Hazard identification, dose‐response, exposure assessment, and risk characterization. Response: EPA agrees that the purpose of this rule is to reduce visibility impairment caused by emissions of NO X from NGS. EPA has not conducted a health risk evaluation for this rulemaking that attempts to characterize or quantify a public health benefit. Because NO X is itself a criteria pollutant that affects public health and is also a precursor to ozone and fine particulate matter, which are also criteria pollutants that affect public health, we consider it reasonable to state that other benefits could exist. We also note that EPA does not agree that there are no health benefits from reductions in ozone and fine particulate matter below the level of the NAAQS. On the contrary, EPA's practice of quantifying these benefits in regulatory impact assessments has been strongly supported by peer-reviewed science. 75 75 See EPA, 2010, “Summary of Expert Opinions on the Existence of a Threshold in the Concentration-Response Function for PM 2.5 -related Mortality Technical Support Document.” Available from http://www.epa.gov/ttn/ecas/regdata/Benefits/thresholdstsd.pdf. D. Comments on Factor 3—Existing Controls at NGS Comment: EPA failed to consider existing controls. Based on EPA's statement in the Proposed Rule that the early installation of LNB/SOFA would not influence EPA's BART determination and EPA's use of a baseline scenario in the visibility modeling that did not include LNB/SOFA, the operator of the Kayenta Mine concluded that EPA failed to consider existing controls. Response: EPA disagrees with the assertion that we failed to consider existing controls. As described in our Proposed Rule and consistent with the BART Guidelines (directing BART determinations to conduct the five-factor analysis generally using a 2001-2003 baseline) EPA evaluated LNB/SOFA as a separate control technology in our BART analysis, as well as a technology that can be used in combination with post-combustion control technologies (i.e., SNCR and SCR). 76 We also discussed the voluntary installation of LNB/SOFA in 2009-2011 under Factor 3: Existing Controls at NGS. 77 76 See Proposed Rule at 78 FR 8280, 8284 and 8285 (February 5, 2013). 77 Id. at 8284. As discussed in section 8.5 of the RTC, EPA properly considered baseline emissions over the period 2001-2003 in our analysis of cost-effectiveness and anticipated visibility benefits of controls. Therefore, although we did not “consider existing controls” in the exact manner preferred by the commenter, we appropriately considered the existence of LNB/SOFA in Factor 3 of our BART analysis. In addition, the “better than BART” framework that we used to assess and finalize BART alternatives explicitly accounts for the existing LNB/SOFA. Comment: EPA should determine existing controls to be BART. Several commenters noted that NGS spent millions of dollars on LNB/SOFA to reduce NO X emissions to levels below the presumptive NO X emission levels in the BART Guidelines. One commenter stated that installing LNB/SOFA prior to a requirement to do so under the RHR or any other CAA requirement has resulted in greater total NO X emission reductions in the first regional haze planning period than would be required by the most stringent EPA BART determination. Response: EPA recognizes that the early and voluntary installation of LNB/SOFA on one unit per year in 2009-2011 at NGS resulted in significant emission reductions from NGS. EPA agrees that the early installation of LNB/SOFA on one unit per year was voluntary and resulted in significant NO X reductions in the first planning period for Regional Haze. However, based on our five-factor analysis, we have determined that SCR+LNB/SOFA is also cost-effective and would result in significant additional visibility improvement at a number of Class I areas. We therefore disagree that LNB/SOFA should be determined BART for NGS. E. Comments on Factor 5—Anticipated Visibility Benefits Comment: General Comments on Visibility. Numerous commenters questioned the extent to which NGS impacts visibility at Class I areas or disputed EPA's analysis that installation of SCR at NGS would improve visibility. Many commenters asserted that the haze is produced from emissions from other sources. Some commenters stated that the wind near and around the Grand Canyon blows predominantly west to east; thus, emissions from the NGS are pushed away from several Class I areas, not towards them. Response: We are aware of the studies cited by commenters purporting to show that controls on NGS would yield little visibility improvement, and we address them in section 7.0 of the RTC. We are also aware of work performed by the Western Regional Air Partnership (WRAP) suggesting that the relative contribution of nitrate from point sources to visibility impacts is relatively small. 78 The CAA and RHR require that BART be installed on certain old, large stationary sources as part of the overall approach to improving visibility at Class I areas. No control at an individual source will be sufficient to meet the goal of remedying existing impairment of visibility in mandatory class I Federal areas which result from manmade air pollution, as set out in section 169A of the CAA. 78 See, e.g. WRAP PM Source Apportionment Technology (PSAT) results, available on WRAP Technical Support System, Source Apportionment Web page at http://vista.cira.colostate.edu/TSS/Results/SA.aspx. On the issue of wind direction, we note that the CALPUFF modeling uses three years of hourly meteorological input, which is based on meteorological modeling as well as observational data from stations throughout a large area. The input includes wind speed and direction, and would include the particular wind direction patterns noted by the commenter. The more sophisticated meteorological treatment in CALPUFF enables it to track the pollutant plume from NGS, including its twists and turns over multiple days. We consider this approach to adequately account for variability in winds noted by the commenter. Comment: EPA underestimated visibility benefits of SCR. One commenter stated that the visibility benefits of SCR are greater than those modeled by EPA because EPA underestimated SCR performance and because EPA overestimated the potential increase in sulfate emissions that may come with the addition of SCR controls by assuming an SO 2 to SO 3 conversion rate that is too high and using an erroneous value for the coal sulfur content. The commenter stated that its own modeling shows greater visibility improvement than demonstrated by EPA. Response: We disagree that EPA underestimated the visibility benefits of SCR and we note that the commenter's assertion that the visibility benefits are even better would not change our proposed determination under Factor 5 that the anticipated visibility benefits of SCR+LNB/SOFA are significant and support our proposed BART limit for NO X , achievable with SCR+LNB/SOFA. Please see the RTC for a detailed discussion of EPA's responses to the commenter's specific assertions. Comment: EPA overestimated visibility impact of NGS by using background ammonia concentrations that were too high. Several commenters argued that EPA's assumed ammonia background concentration of 1 part per billion (ppb), the default value recommended by the Interagency Workgroup on Air Quality Modeling (IWAQM), is unrealistically high compared to measured values in the area, resulting in artificially high model projections of visibility impacts, particularly in the winter. 79 The commenter noted that the use of a constant value of 1.0 ppb for background ammonia concentration fails to account for known variations in monthly or seasonal ammonia concentration. 79 Interagency Workgroup On Air Quality Modeling (IWAQM) Phase 2 Summary Report And Recommendations For Modeling Long Range Transport Impacts (EPA-454/R-98-019), EPA OAQPS, December 1998, http://www.epa.gov/scram001/7thconf/calpuff/phase2.pdf. One commenter cited an analysis conducted on behalf of SRP by AECOM and Dr. Ivar Tombach. The commenter stated that the Tombach study compared modeled predictions of ammonium nitrates using both EPA's and AECOM's ammonia background concentrations to measured ammonia values, demonstrating that the EPA's assumptions over-predict actual measured values by a factor of 10 or more in some cases. One commenter noted that when the IWAQM guidance was issued 14 years ago, CALPUFF did not have the capability of accommodating monthly ammonia background concentrations as it has since been updated to do. The commenter asserted that EPA's reliance on a constant value is an outdated approach. Response: EPA has already considered and addressed the same arguments and data provided by commenters related to background ammonia concentrations in other rulemakings, including our final rule for Four Corners Power Plant. 80 As summarized briefly below, EPA disagrees that our use of the IWAQM default background ammonia concentration for arid areas of 1 ppb was inappropriate. Please see the RTC for the full response to this comment. 80 See RTC and references therein. We have carefully reviewed the comments and concluded that, on balance, the evidence does not support using lower values for background ammonia concentrations, as argued by the commenters, in estimating the visibility impacts from NGS. Much of the existing measured data cited by the commenters is from other states and may not be representative for evaluating visibility impacts from NGS. 81 Further, existing data sometimes represent ammonia alone rather than total ammonia and ammonium. Because ammonium represents part of the pool of ammonia that could be available to interact with the SO 2 and NO X emitted from stationary sources, it should be accounted for in the value for background ammonia concentrations used in the model. In several of the research papers cited by commenters, the amount of measured ammonium is comparable to and at times much greater than the amount of ammonia. 82 Measurements made by SRP closer to NGS over December 2009 to April 2010, which included ammonia and ammonium, showed that depending on time and location, typical ammonia concentrations ranged from 0.2 ppb to 0.8 ppb and the concentration of total ammonia and ammonium ranged from 0.6 to 1.2 ppb, which is considerably higher than the 0.2 ppb winter values used in SRP's modeling. 83 Although some of the ammonium may not be available to interact with pollutants from NGS, the sum of ammonia and ammonium provides an upper bound estimate of background ammonia concentrations, and represents a conservative estimate for modeling. 81 See e.g., SRP comments Appendix G, “Measurements of Ambient Background Ammonia on the Colorado Plateau and Visibility Modeling Implications”, Prepared by Salt River Project, Contributors: Ivar Tombach and Robert Paine, September 2010. Referred to here as “SRP monitoring report”, or Tombach & Paine 2010. 82 See RTC and references therein. 83 SRP monitoring report, or Tombach & Paine 2010, and SRP comments Appendix C. “Revised BART Analysis for the Navajo Generation Station Units 1-3” (January 2009) and Appendix I. “Visibility Modeling Using Upgraded CALPUFF Model” (February 2011). We further note that there are measurements of gaseous ammonia alone that show concentrations close to or greater than the concentration of 1 ppb, even in winter when ammonia concentrations are expected to be lowest. Winter measurements, representing 3-week averages, ranged from 1.1 ppb to 1.8 ppb at a monitor at the Farmington Airport in northwestern New Mexico. 84 Measurements from the winters of 2011-2013 from the AMoN network ranged from 1.1 to 1.3 ppb for Farmington, and 0.7-0.9 ppb for Chiricahua, in southeastern Arizona. 85 84 See RTC and references therein. 85 Id. We further note that there is significant variability in the concentrations of ammonia measured at different times and places. Even the SRP monitoring report (Tombach & Paine, 2010, cited above) describes a surprisingly high spatial variability in ammonia concentrations. Because of the variability and its unknown causes, the data collected for SRP did not lead to a clear picture of appropriate and representative background ammonia concentrations to use with CALPUFF. Finally, we note that using the background ammonia concentrations recommended by commenters does not change our conclusion under Factor 5 because CALPUFF modeling of SCR shows substantial visibility benefits even using the alternative assumptions. 86 Using a background ammonia concentration of 1 ppb ammonia, EPA modeled the greatest benefit from SCR+LNB/SOFA to be 5.4 deciviews at Capitol Reef NP, and modeled a visibility benefit exceeding 1 to 2 deciviews at ten additional Class I areas. Using the ammonia concentration recommended by some commenters (ranging from 0.2 ppb in winter to 1.0 ppb in summer), EPA modeled the greatest benefit of SCR to be 2.3 dv, and modeled a visibility benefit exceeding 1 deciview at nine Class I areas, with three of these nine areas having a benefit of approximately two deciviews. Even assuming a lower ammonia concentration, the modeling demonstrates that the installation of SCR+LNB/SOFA at NGS would have a significant beneficial impact on visibility at a number of Class I areas. Our conclusion as to the appropriate BART Benchmark for NGS would not accordingly change. 86 See RTC and references therein. Comment: EPA should have used an updated version of CALPUFF. Several commenters asserted that EPA erred in using CALPUFF version 5.8 in its modeling rather than the more recent CALPUFF version 6.42, released by TRC. One commenter argued that CALPUFF version 6.42 predicts lower visibility benefits than version 5.8. Response: We disagree with the commenters that a new CALPUFF version should be used for the BART determination. We relied on version 5.8 of CALPUFF because it is the version approved by EPA through a public notice and comment rulemaking, in accordance with the Guideline on Air Quality Models (“GAQM”, 40 CFR part 51, Appendix W, section 6.2.1.e). 87 CALPUFF version 6.4 is not approved by EPA for regulatory purposes, and we do not agree that the changes made to this most recent version of CALPUFF were simple model updates to address bugs. A full evaluation of a new model such as CALPUFF version 6.4 is needed before it should be used for regulatory purposes as errors that are not immediately apparent can be introduced along with new model features. 87 See RTC and references therein. Comment: Closure of Mohave Project did not improve visibility and shows CALPUFF is unreliable. One commenter discussed the findings of an analysis conducted after the closure of the Mohave Power Project (MPP) (a 1,580 MW coal-fired power plant) to evaluate whether the closure had resulted in improved visibility in Grand Canyon National Park. 88 The commenter indicated that although CALPUFF version 5.8 modeling predicted that the plant had a significant impact on visibility in the Grand Canyon, this study concluded that there was “virtually no evidence that the MPP closure improved visibility in the Grand Canyon.” The commenter asserted that this study raises questions about the reliability of CALPUFF. 88 See RTC and references therein. Response: We disagree that the Terhorst & Berkman (T&B) study cited by the commenters raises questions about CALPUFF's reliability. The conclusion in the T&B study on the effect of MPP closure is actually similar to that from earlier analyses, which also predicted improvements less than the human perceptibility threshold of 1 dv. A response to the T&B study written by White et al., stated that the T&B analysis is “misleadingly presented as discrediting previous studies and their interpretation by regulators. In reality the T&B analysis validates a consensus on MPP's visibility impact that was established years before its closure.” 89 89 W.H. White, R.J. Farber, W.C. Malm, M. Nuttall, M.L. Pitchford, B.A. Schichtel, Comment on “Effect of coal-fired power generation on visibility in a nearby national park (Terhorst and Berkman, 2010)”, Atmospheric Environment 55 (2012) 173-178. doi:10.1016/j.atmosenv.2012.02.076. Also available at: http://www.dri.edu/marc-pitchford?showall=&start=2. White et al., explicitly addressed the purported disagreement between the T&B methodology and results from CALPUFF, pointing out that the comparison was flawed in several ways. First, the ambient data relied upon by T&B are collected only every third day; this results in an insufficient number of days for a valid statistical comparison to the 98th percentile results reported from CALPUFF. Another important flaw is that when T&B translated visibility extinction into deciviews, they used recent polluted conditions as the background for comparison, whereas the BART Guidelines and the CALPUFF results use natural conditions as background. 90 When the T&B results are computed using natural background, they are substantially larger, and generally in agreement with CALPUFF results. 90 EPA considered and rejected comments on the proposed BART Guidelines that visibility impacts should be evaluated relative to current degraded visibility conditions and concluded that “[u]sing existing conditions as the baseline for single source visibility impact determinations would create the following paradox: The dirtier the existing air, the less likely it would be that any control is required.” (70 FR 39104 at 39124, July 6, 2005). F. Comments on BART Determination for NO X Comment: BART limit for NGS should be 0.04 lb/MMBtu. One commenter argued that the final BART emission limit should be more stringent and no higher than 0.04 lb/MMBtu. The comment noted that permitting authorities have required lower NO X limits than 0.055 lb/MMBtu in recent BACT determinations based on SCR in combination with combustion controls. Response: EPA disagrees with the commenter that the BART Benchmark for NGS should be 0.04 lb/MMBtu. We note that the commenter has not provided any specific information to show that NGS could demonstrate continuous compliance with an emission limit of 0.04 lb/MMBtu. The commenter generally argued that SCR systems are typically designed to achieve 90 percent removal. EPA notes that although an SCR system can be designed to a specific target, the design target is typically not equivalent to the actual emission limit. 91 EPA proposed a limit of 0.055 lb/MMBtu achievable with SCR+LNB/SOFA, and using a baseline emission rate of 0.35 lb/MMBtu, this represents a removal efficiency of 84 percent. 92 However, as noted elsewhere in the RTC, the limit of 0.055 lb/MMBtu, which accommodates startup, shutdown, and low-load operation, is based on a design target of 0.03 lb/MMBtu. This represents a design target removal efficiency of 91 percent for SCR+LNB/SOFA (from a baseline of 0.35 lb/MMBtu), or 88 percent for SCR alone (i.e., from 0.24 lb/MMBtu). 91 EPA's Environmental Appeals Board has recognized that PSD emission limits must be set to allow fluctuations in operations, stating: “To account for these possibilities, a permitting authority must be allowed a certain degree of discretion to set the emissions limitation at a level that does not necessarily reflect the highest possible control efficiency, but will allow the permittee to achieve compliance consistently.” In Re Masonite Corporation, 5 E.A.D. 551, 560-61 (1994). 92 See RTC and references therein. EPA disagrees with the commenter's assertion that emission limits associated with BART must meet BACT or the lowest emission rate ever achieved with that technology at any coal-fired power plant. The BART Guidelines state that: “[i]n assessing the capability of the control alternative, latitude exists to consider special circumstances pertinent to the specific source under review, or regarding the prior application of the control alternative”, (70 FR 39166) and that “[t]o complete the BART process, you must establish enforceable emission limits that reflect the BART requirements . . .” (70 FR 39172). The five-factor BART analysis described in the Guidelines is a case-by-case analysis that considers site specific factors in assessing the best technology for continuous emission controls. After a technology is determined as BART, the BART Guidelines require establishment of an emission limit that reflects the BART requirements, but does not specify that the emission limit must represent the maximum level of control achieved by the technology selected as BART. For these reasons, EPA is not using the lower limit recommended by the commenter in setting the BART Benchmark. Comment: BART limit for NGS should be in the range of 0.07-0.08 lb/MMBtu. Several commenters asserted that the NO X emission limit EPA proposed for NGS is unachievable. One commenter noted that the averaging period for the proposed limit of 0.055 lb/MMBtu includes periods when the SCR is unable to operate such as startup, shutdown, and periods of load-cycling. The commenter made the following arguments: (1) The S&L analysis submitted by the commenter shows that the proposed emission limit is unachievable on a continuous basis; (2) the NO X emissions achieved in other SCR retrofit situations do not justify the proposed emission limit. Response: EPA disagrees with the commenter that the limit used in setting the BART Benchmark for NGS should be higher than our proposed limit of 0.055 lb/MMBtu, in the range of 0.07 to 0.08 lb/MMBtu. 93 93 The response included in this Final Rule is abbreviated and excludes the graphs and tables EPA generated to support our response. For additional detail, please see the RTC. The S&L report generally argues that because the emission limit is established based on a 30-BOD average basis, the proposed emission limit of 0.055 lb/MMBtu is not consistently achievable at NGS. The S&L analysis is based on a design target of 0.03 lb/MMBtu and suggests an emission limit in the range of 0.07-0.08 lb/MMBtu would be required to accommodate periods of load-cycling operation, startups, and shutdowns. S&L is recommending a limit that is 2.3 to 2.7 times higher than the design target, or a compliance margin of 133 to 167 percent. The S&L report discusses the temperature limitations associated with SCR and explains that at temperatures below a specific minimum operating temperature, a component of the SCR system (i.e., ammonia injection) must cease to prevent ammonium salt formation on the catalyst. S&L asserts that a minimum operating temperature of 580 °F is typical for retrofit SCR control systems installed on coal-fired electric generating units with similar coal sulfur content and states that this temperature corresponds with a gross load of approximately 650 MW (650 gross MW, or MWg). S&L further assumes that SRP will likely modify the units to increase flue gas temperatures at lower operating loads by installing one of several options for low load temperature control. In their analysis, S&L assumes the low load temperature control would be achieved with a water-side bypass (to allow water to bypass the economizer tube bundles during low-load operation). The S&L report states “[b]ased on a preliminary review of the available systems, a water-side bypass system should be capable of increasing the temperature of the bulk flue gas by approximately 25 °F to 65 °F during low-load operation. For this evaluation, a low-load temperature control system capable of achieving a temperature increase of 65 °F during low-load operations was assumed for modeling purposes.” S&L further estimates that this would correspond to a minimum gross load of 450 MWg for the SCR to operate, or operation at 55 percent capacity. Using the assumption that the SCR would not operate at loads below 450 MWg, S&L used 2012 operations data at NGS to estimate emission rates at NGS assuming a design target of 0.03 lb/MMBtu with actual steady-state operations achieving 0.04 lb/MMBtu. S&L modeled eighteen different operating scenarios and identified seven scenarios, which included periods of low load cycling along with unit startup and shutdowns, that resulted in the maximum 30-BOD average for each unit and facility-wide, that exceeded 0.055 lb/MMBtu. The highest 30-BOD average S&L modeled was 0.077 lb/MMBtu for Unit 2, achieved under 3 different operating scenarios involving low-load cycling. SRP and S&L did not provide the underlying data used in the S&L analysis. Therefore, EPA evaluated the S&L report by reviewing emissions data from the EPA Air Markets Program Data (AMPD) for multiple years, as well as emissions data from other facilities that were constructed or retrofit with SCR. EPA sought to understand 2012 operations at NGS within the context of longer term operational trends at the facility, as well as understand the minimum operating load assumed by S&L for NGS within the context of minimum operating loads at other facilities with SCR. EPA evaluated the reported hourly gross load operating data for Units 1-3 at NGS for the years 2001, 2003, 2010, 2011, 2012, and 2013. 94 Emission data from AMPD show that NGS, and in particular, Unit 2, spent a higher percentage of operating hours at gross loads below 450 MWg in 2012 compared to other years. The 2012 gross load profiles for Unit 2 (as well as Units 1 and 3) are characteristic of load-cycling units, with significant periods of time below the purported SCR minimum operating load of 450 MWg, particularly in the spring. Please see the RTC for more detail. In 2010, Unit 2 also operated for significant periods of time at loads below 450 MWg. However, these periods in 2010 occurred following the major outage on Unit 2 (following installation of LNB/SOFA on that unit). Although Units 1-3 at NGS did appear to operate as load-cycling units and operated below 450 MWg for significant periods of time in 2012, this type of operation does not appear to be characteristic of typical operation at NGS, based on our evaluation of previous years, as well as 2013. 94 See RTC and references therein. Based on the gross load operating profiles for six years, EPA estimated the rolling 30-BOD averages for each BOD to determine whether the operating profiles (which included actual startup, shutdown, and load-cycling in each year) would result in 30-BOD averages that would exceed 0.055 lb/MMBtu. Based on our analysis, EPA projected the highest 30-BOD average to be 0.079 lb/MMBtu (Unit 2 in 2010). Using 2012 data, representative of load-cycling operation, EPA projected the highest 30-BOD average to also occur on Unit 2 (0.075 lb/MMBtu). Similarly, S&L projected the highest 30-BOD average in 2012 was from Unit 2, at 0.077 lb/MMBtu. Therefore, although the scenarios modeled by S&L and EPA were not identical, the highest 30-BOD averages projected by EPA and S&L, using similar starting assumptions, were comparable. Our analysis, of projected SCR performance, which included emission and operating profiles of actual startup and shutdown events, and load-cycling in various years, showed that Unit 3 was not projected to exceed 0.055 lb/MMBtu in any of the evaluated years, and that there were several years within these six selected years that Units 1 and 2 would also not exceed 0.055 lb/MMBtu. The analysis of projected 30-BOD average emission rates assumes that S&L's value of 450 MWg (or 55 percent capacity) for the minimum operating load to operate SCR at NGS is correct. EPA notes that 450 MWg was a value that S&L assumed based on preliminary analysis of available low load temperature control systems. SRP submitted a similar S&L analysis to EPA for Units 1 and 3 at Coronado Generating Station (CGS). 95 Units 1 and 2 at CGS are 430 MWg Riley-Turbo units that typically operate as load-cycling units. CGS burns low-sulfur coal from the Powder River Basin (PRB coal). With the application of low-load temperature controls on these units, S&L's analysis suggests that the minimum operation load for SCR on Units 1 and 2 at CGS would be 138 MWg (or 32 percent capacity). This is significantly lower than the 55 percent capacity S&L assumed for NGS. S&L stated that the coal sulfur content will affect the minimum operating load for SCR. NGS does not burn PRB coal; however, NGS does burn low-sulfur coal from the Kayenta Mine. AECOM, SRP's consultant for visibility modeling, reported the maximum sulfur content of the coal as 0.593 percent based on daily data for the 2001-2003 period. For comparison, various sources reference PRB coal as generally low-sulfur coal with a sulfur content of less than 1 percent, or a mean of 0.5 percent. 96 In contrast, high sulfur coal is typically above 3 percent. 97 95 See RTC and references therein. 96 See, for example, publication from the U.S. Geological Survey, figure PQ-4 and Table PQ-1, available at http://pubs.usgs.gov/pp/p1625a/Chapters/PQ.pdf. 97 Id. EPA evaluated emission data of eight well-performing units burning PRB coal and generated empirical estimates for minimum operating loads and capacity requirements for SCR operation at those facilities. Based on this analysis (see RTC for further detail), EPA estimated capacity requirements for SCR operation that ranged from 35 percent to 46 percent, with an average value of 40 percent. Using the average (40 percent) and the maximum (46 percent) capacity requirement to operate SCR, EPA projected that NGS would meet a limit of 0.055 lb/MMBtu (on a 30-BOD average) for all but 3 cases (i.e., Units 1 and 2 in 2012, and Unit 2 in 2010) under the 46 percent capacity requirement. Under the 40 percent capacity requirement to run SCR, Units 1 and 2 in 2012 would remain below 0.055 lb/MMBtu and for Unit 2 in 2012 the highest 30-BOD average was projected to be exactly 0.055 lb/MMBtu. Operation of Unit 2 in 2010 was not typical of normal operation. Please see RTC for more detail on this analysis [Text truncated at 120,000 characters. The full text is on the page linked above.] --- Source: Frix Law Library, https://www.frixlaw.com/law-library/documents/fr%3A2014-18228. Public record. Not legal advice.