# Air Plan Approval; OR; Regional Haze Plan for the Second Implementation Period > Briefs, arguments, decisions, and more. URL: https://www.frixlaw.com/law-library/documents/fr%3A2024-22603 ## Record - **Collection:** Federal Register - **Document type:** Rule - **Published:** October 8, 2024 - **Citation:** 89 FR 81361 ## Text ENVIRONMENTAL PROTECTION AGENCY 40 CFR Part 52 [EPA-R10-OAR-2023-0600, FRL-11593-02-R10] Air Plan Approval; OR; Regional Haze Plan for the Second Implementation Period AGENCY: Environmental Protection Agency (EPA). ACTION: Final rule. SUMMARY: The Environmental Protection Agency (EPA) is approving the regional haze state implementation plan revision submitted by Oregon on April 29, 2022, as supplemented on November 22, 2023, as satisfying applicable requirements under the Clean Air Act and the EPA's Regional Haze Rule for the program's second implementation period. The Oregon submission addressed the requirement that states must periodically revise their long-term strategies for making reasonable progress towards the national goal of preventing any future, and remedying any existing, anthropogenic impairment of visibility, including regional haze, in mandatory Class I Federal areas. The Oregon submission also addressed other applicable requirements for the second implementation period of the regional haze program. DATES: This final rule is effective November 7, 2024. ADDRESSES: The EPA has established a docket for this action under Docket ID No. EPA-R10-OAR-2023-0600. All documents in the docket are listed on the https://www.regulations.gov website. Although listed in the index, some information is not publicly available, e.g., Confidential Business Information or other information the disclosure of which is restricted by statute. Certain other material, such as copyrighted material, is not placed on the internet and will be publicly available only in hard copy form. Publicly available docket materials are available at https://www.regulations.gov, or please contact the person listed in the FOR FURTHER INFORMATION CONTACT section for additional availability information. FOR FURTHER INFORMATION CONTACT: Jeff Hunt, EPA Region 10, 1200 Sixth Avenue, Suite 155, Seattle, WA 98101, at (206) 553-0256 or hunt.jeff@epa.gov. SUPPLEMENTARY INFORMATION: Throughout this document, wherever “we” or “our” is used, it means the EPA. Table of Contents I. Background II. EPA Responses to Comments Received A. National Park Service Comments 1. Federal Land Manager Consultation 2. Use of the Four Statutory Factors in Determining Reasonable Progress 3. Use of Permitted Emissions Limits To Align Allowable Emissions With Actual Emissions 4. Use of a Stipulated Agreement and Final Order (SAFO) Versus a Unilateral Order 5. Compliance Deadlines 6. Standards for Emissions Unit Replacement 7. Wauna Facility—Biomass Fired Fluidized Bed Boiler 8. Georgia-Pacific—Toledo LLC—Final Control Determination 9. Georgia-Pacific—Toledo LLC—Emission Limit 10. Georgia-Pacific—Toledo LLC—Emissions Unit Replacement 11. Cascade Pacific Pulp, LLC—Halsey Pulp Mill Power Boilers—SCR 12. Cascade Pacific Pulp, LLC—Halsey Pulp Mill Power Boilers—LNB 13. Cascade Pacific Pulp, LLC—Halsey Pulp Mill Power Boilers—Compliance Deadline and Emission Limit 14. International Paper—Springfield—Emission Limit 15. Owens-Brockway Glass Container Inc. B. Environmental Organizations' Comments 1. Stationary Source Contribution 2. Oregon's Regional Haze Rule Is Inconsistent With the CAA 3. Oregon's Use of PSEL Reductions as a Source Selection Method 4. Oregon's “Alternative Compliance” Pathways 5. Documentation of Oregon's Four-Factor Analysis Process 6. Owens-Brockway Glass Container Inc. 7. Reasonable Progress Goals 8. Robust Demonstration Requirements 9. Environmental Justice III. Final Action IV. Incorporation by Reference V. Statutory and Executive Order Reviews I. Background On February 23, 2024, the EPA proposed to approve the regional haze state implementation plan (SIP) revision submitted by Oregon on April 29, 2022, as supplemented on November 22, 2023, as satisfying applicable requirements under the Clean Air Act (CAA) and the EPA's Regional Haze Rule (RHR) for the program's second implementation period (89 FR 13622). The public comment period for our proposed action was originally scheduled to close on March 25, 2024. However, on February 28, 2024, we received a request to extend the public comment period an additional 30 days. 1 On March 14, 2024, we published a document in the Federal Register extending the public comment period end date from March 25, 2024, to April 24, 2024 (89 FR 18866). 1 This extension request letter may be found in the docket for this action. We received four comments. We determined that two comments were not germane to our action, for the following reasons. One commenter expressed opposition to the cultivation of cannabis, asserting general air pollution concerns. The commenter did not provide any tangible connection to the regional haze requirements or the Oregon submission. The EPA acknowledges the commenter's concerns; however, the comment is outside the scope of this action and does not indicate that the EPA's approval of the SIP submission is inconsistent with the CAA. Oversight of cannabis farms is unrelated to this regional haze action. A second commenter cited some details from the Oregon regional haze plan, asserting a connection to transmission of the coronavirus. However, the commenter provided no logical basis for this assertion. The EPA acknowledges the commenter's concerns; however, the comment is outside the scope of this action and does not indicate that the EPA's approval of the SIP submission is inconsistent with the CAA. Potential connections between air pollution and respiratory viruses on public health is unrelated to this regional haze action. We also received two germane comments. One was submitted by the National Park Service (NPS). The second was submitted by Earthjustice on behalf of a coalition of environmental organizations consisting of Cully Air Action Team, National Parks Conservation Association, Neighbors for Clean Air, Northwest Environmental Defense Center, Oregon Environmental Council, Sierra Club, and Verde (Environmental Organizations). The full text of the comments may be found in the docket for this action. We have summarized the comments and provided our responses in section II. of this preamble. Under the Clean Air Act, the Administrator is required to approve a SIP submission that complies with the provisions of the Clean Air Act and applicable Federal regulations. 42 U.S.C. 7410(k); 40 CFR 52.02(a). Thus, in reviewing SIP submissions, the EPA's role is to approve state choices, provided that they meet the criteria of the Clean Air Act. II. EPA Responses to Comments Received A. National Park Service Comments 1. Federal Land Manager Consultation Comment: “[W]e would like to make the EPA aware that the Oregon SIP process did not meet the requirements for Federal Land Manager (FLM) consultation . . . The NPS participated in early, informal engagement with the Oregon Department of Environmental Quality (ODEQ) regarding SIP development beginning in January of 2020. This productive collaboration included a meeting, subsequent written documentation, and staff-to-staff technical feedback on individual facility four-factor analyses as documented in the Oregon SIP. The NPS appreciates the extensive efforts that Oregon invested in early communication. However, many of the draft conclusions and determinations presented to the NPS during early engagement were not incorporated into the draft SIP released for public review in late August 2021. In fact, the public comment draft included substantial changes to the facility-specific control determinations in comparison with what the NPS had reviewed previously.” Response: The EPA disagrees with the commenter's assertion that Oregon did not meet the requirements for Federal Land Manager (FLM) consultation in 40 CFR 51.308(i). As described below, ODEQ met all of the FLM consultation statutory and regulatory requirements. Chapter 6.3 Consultations with Federal Land Managers of the April 29, 2022, SIP revision contained documentation of the extensive outreach with the NPS. This included providing a May 5, 2021, draft of the regional haze plan explicitly for the purpose of FLM consultation (May 2021 FLM draft). A key element of 40 CFR 51.308(i)(2) is that consultation occur early enough in a state's policy analyses of its long-term strategy so that information and recommendations provided by the FLMs can meaningfully inform a state's decisions on the long-term strategy. Chapters 6.3.3 Federal Land Manager Review of Draft State Implementation Plan and 6.3.4 Federal Land Manager Comments and DEQ Responses contained Oregon's responses to all comments received as part of the May 2021 FLM draft review process and prior consultation outreach. The NPS's characterization of this effort as “informal consultation” is not consistent with the EPA's regulations. The requirements of 40 CFR 51.308(i)(2) contain no bifurcation of “informal” versus “formal” consultation. Consistent with the preamble of the EPA's 2017 Regional Haze Rule, the Oregon Department of Environmental Quality (ODEQ) made a good faith effort to involve the NPS early in development of the long-term strategy. The RHR preamble specifically states that consultation should be used to inform “decisions that are about to be made by the state on its long-term strategy . . .” (emphasis added). 2 ODEQ used the comments and feedback from the May 2021 FLM consultation draft to inform the final control determinations contained in the draft provided for public notice and comment period starting on August 27, 2021 (August 2021 public comment draft). The interpretation that all control determinations must be finalized before initiating FLM consultation is not supported by the text of 40 CFR 51.308(i), nor is it in keeping with the intent of 40 CFR 51.308(i) to foster early engagement. 2 See 82 FR 3078 (January 10, 2017) at page 3116. In addition to the May 2021 FLM consultation draft process, ODEQ provided opportunity for review and comment on the August 2021 public draft. In response to public interest, ODEQ extended the public comment period for an additional 30 days, going from August 27, 2021, to November 1, 2021, so that all parties had adequate time to review the technical determinations. The NPS used this opportunity to provide additional comments which are included in Chapter 6.6 Public Comments and Responses of the April 29, 2022, SIP revision, along with ODEQ's responses to the comments. 3 3 See Oregon's April 29, 2022, submission, pages 136-140 (comment numbers 6-9). Also, in response to NPS and Environmental Organizations' concerns, ODEQ supplemented the regional haze SIP on November 22, 2023, with appendices 1 through 6 (2023 supplement). These appendices contained additional correspondence used by ODEQ in making control determinations under the statutory four factors for Boise Cascade Wood Products, LLC—Elgin Complex, Boise Cascade Wood Products, LLC—Medford, Georgia Pacific—Wauna Mill, Georgia Pacific—Toledo LLC, Cascade Pacific Pulp, LLC—Halsey Pulp Mill, and International Paper Company—Springfield Mill. As acknowledged by the NPS, ODEQ conducted FLM consultation on the 2023 supplement and included NPS comments in Section 6: Response to Federal Land Manager Review Comments. For the reasons stated above, it is our determination that ODEQ adequately conducted FLM consultation and has thus fulfilled the requirements of 40 CFR 51.308(i). 2. Use of the Four Statutory Factors in Determining Reasonable Progress Comment: “The NPS recommends that states, including Oregon, base reasonable progress control determinations on the four statutory factors identified in § 7491(g)(1) of the Clean Air Act (CAA). The NPS suggests that it may not be sufficient to consider the factors and then select a less protective control measure (or permit reduction) that is unrelated to the four-factor analysis. We recommend that determinations not clearly based on the four factors should demonstrate how the alternative measure is reasonable and/or equivalent to the outcome of the four-factor analysis.” Response: The EPA agrees with the NPS that reasonable progress control determinations must be grounded in the state's consideration of the four statutory factors identified in CAA section 169A and 40 CFR 51.308(f)(2). In section IV.E.b of our proposed rulemaking, The EPA's Evaluation of the Oregon Long-Term Strategy, we explained how the Oregon process was grounded in the four-factor analysis (FFA) process. 4 However, the NPS's characterization that ODEQ's January 2021 “Preliminary Determination of Cost Effective Controls for Regional Haze” letters constituted final four-factor determinations, and all subsequent correspondence and action was outside the four-factor framework is an inaccurate characterization of the Oregon process. 5 The EPA will reiterate key aspects from our proposed rulemaking as well as details from Chapter 3.4 Four Factor Analysis of Oregon's April 29, 2022, regional haze plan. 4 See 89 FR 13622 (February 23, 2024) at page 13637. 5 Complete copies of the “Preliminary Determination of Cost Effective Controls for Regional Haze” letters are included in the docket for this action. Oregon Administrative Rules (OAR) Division 223 Regional Haze Rules dictated the regulatory processes for determining the controls necessary for reasonable progress. OAR 340-223-0110(1) required all affected facilities to submit four-factor analyses. The required contents of the four-factor analyses were specified in OAR 340-223-0120, which mirrored the four statutory factors of CAA section 169A(g)(1). Of the 17 facilities that submitted four-factor analyses in accordance with this rule, nearly all affected facilities submitted detailed demonstrations developed by independent consulting firms and/or certified professional engineers asserting that no feasible or cost-effective controls were available in applying the four-factors. 6 Only 2 facilities, Owens-Brockway Glass Container Inc. and Gilchrist Forest Products found cost effective controls under the four factors. 6 Complete copies of the four-factor analyses are included in the docket for this action. Rather than simply relying on these submissions to satisfy 40 CFR 51.308(f)(2), Oregon engaged with the affected facilities to identify feasible control options. Specifically, ODEQ issued “Preliminary Determination of Cost Effective Controls for Regional Haze” letters proposing more stringent controls unless facilities could further demonstrate that the measures were truly not cost effective or technically feasible. This initiated the process between January 2021 and August 2021 when ODEQ assessed and determined final control determinations under the four factors, primarily based on the technical feasibility and cost correspondence documented in appendices 1-6. 7 7 Complete copies of appendices 1-6 were included in the November 2023 supplement to the regional haze plan and are also included in the docket for this action (document numbers 246-251). Thus, these preliminary letters did not replace the four factor analyses submitted by the affected facilities. Nor were the preliminary letters final determinations on what controls are feasible and necessary for reasonable progress under OAR 340-223-120(4) or 40 CFR 51.308(f)(2). Rather, these preliminary determinations letters were a result of ODEQ's initial adjustment of the four-factor analyses submitted by the affected facilities based on additional information and to aid in a consistent review across all four-factor analyses. 8 Consistent with 40 CFR 51.308(f)(2) and OAR 340-223-120(4), ODEQ determined the controls necessary for reasonable progress after issuing the preliminary determinations and collecting and analyzing additional information from the facilities regarding the four statutory factors. As ODEQ explained in its response to comments on the initial SIP submission, in some cases ODEQ agreed with facilities that controls it preliminarily proposed were not technically feasible or cost effective. 9 Ultimately, Oregon's submission demonstrates that it determined the controls necessary for reasonable progress based on its consideration of the four statutory factors and thus met the requirements of 40 CFR 51.308(f)(2). 8 See OAR 340-223-0120(2) and (3). 9 See Oregon Environmental Quality Commission Meeting, February 3-4, 2022, 1001_1.1_StaffReport_wAttachments Comment #17 at Item C 000035. Table 1 of this preamble provides a comparison of the controls evaluated in the four factor analyses submitted by the sources and the controls Oregon ultimately included in its long-term strategy. The exact facility-by-facility determinations are described in more detail in the Proposal and in subsequent responses to facility-specific comments in section II of this preamble. 10 This table was created using the EPA's 2019 Guidance recommendation to use projected actuals or recent actuals in cost-effectiveness calculations. 11 The four-factor analysis raised technical feasibility and cost barriers in the determination that, “Based on the Four Factor analysis presented above, no additional controls were determined to be cost effective for the biomass boilers at the Elgin Mill.” See 106_SAFOBCWoodProducts31-0006Elgin.pdf, at page 2-25. Correspondence related to these issues is included in 246_3.3.2_Appendix1_Boise.Cascade.Elgin_Correspondence.pdf, included in the docket for this action. 12 ODEQ did not pursue this control option, presumably because associated potential PM 10 emissions were low (69 tons per year) relative to potential NO X emissions (367 ton per year) and the significant difference between allowable PSEL emissions (Q/d = 11.92) and actual emissions (Q/d = 3.57). ODEQ instead proposed NO X controls in the January 2021 preliminary determination letter. See 120_haze-EVRAZ.pdf, included in the docket for this action. 13 The four-factor analysis argued, “A formal engineering analysis would be required to ultimately determine if SNCR would be effective on the boilers. This type of analysis would include obtaining temperature and flow data, developing a model of each boiler using computational fluid dynamics, determining residence time and degree of mixing, determining placement of injectors, and testing.” See 132_haze-GeorgiaPacific-Toledo-FFA.pdf, at page 2-19. 14 In a letter dated September 11, 2020, Gilchrist agreed that installation of an Electrostatic Precipitator on boilers B-1 and B-2 would be cost-effective, and provided a letter from a boiler vendor indicating that retrofitting those boilers with Selective Non-Catalytic Reduction was not technically feasible. See April 22, 2022 regional haze SIP at page 77. 15 Catalytic ceramic filter imposed on furnace D by separate enforcement action. See 701_OwensBrockway2020-208MAO.pdf included in the docket for this action. Table 1—Comparison of Four-Factor Analysis Controls and Final Controls [Please see the footnotes for technical feasibility issues identified in the four-factor analyses] Facility Controls below $10K/per ton reduction threshold in four-factor analyses submitted pursuant to OAR 340-223-0110(1) using recent actual or projected actual emissions (if provided) 10 Final controls imposed by ODEQ Biomass One, L.P None Installation of continuous emissions monitoring system (CEMS) and NO X optimization plan on North and South boilers. If the permittee is able to finalize a new power purchase agreement, the permitted must evaluate installation of selective catalytic reduction (SCR). Boise Cascade Wood Products, LLC—Elgin Complex Selective non-catalytic reduction (SNCR) ($9,523) and SCR ($9,538). 11 Installation of CEMS and NO X combustion improvement project. Cascade Pacific Pulp, LLC—Halsey Pulp Mill None Fuel restrictions and power boiler emissions unit replacement. EVRAZ Inc. NA Baghouse for slab cutting operations ($7,301). 12 Low NO X burners (LNB) on reheat furnace. Gas Transmission Northwest LLC—Compressor Station 13 None SCR or emissions unit replacement on turbines 13C and 13D. Georgia-Pacific—Toledo LLC Low NO X burners with flue gas recirculation (LNB/FGR) on boiler 1 ($7,083). SNCR on boilers 1 ($7,706) and 4 ($7,630). 13 LNB/FGR and CEMS on boilers 1, 3, and 4, or unit replacement on one or more boilers. Georgia Pacific—Wauna Mill (LNB/FGR) on power boiler ($9,223) LNB/FGR and CEMS on power boiler, LNB for paper machine 5, and emissions limits for paper machines 6 and 7. Gilchrist Forest Products None 14 Installation of electrostatic precipitator on units B-1 and B-2. International Paper—Springfield None Fuel restrictions, installation of CEMS on power boiler, and emissions limits. Northwest Pipeline LLC—Oregon City Compressor Station Low Emission Combustion Retrofit ($8,809) Emissions unit replacement and emissions limit. Owens-Brockway Glass Container Inc Catalytic ceramic filters for furnaces A ($5,256) & D ($5,035) Furnace A shut down and PSEL limit imposed. 15 Pacific Wood Laminates, Inc None ODEQ determined no controls <$10K. Roseburg Forest Products—Dillard SNCR on boilers 1 ($4,363), 2 ($4,170), and 6 ($3,635) Installation of CEMS and imposition of emissions limits. Permittee must install SNCR by June 30, 2025, if emissions limits are not met. Willamette Falls Paper Company None Fuel restrictions and Plantsite Emissions Limit (PSEL) reduction. Woodgrain Millwork LLC—Particleboard None ODEQ determined no controls <$10K. 3. Use of Permitted Emissions Limits To Align Allowable Emissions With Actual Emissions Comment: “We note that the Oregon SIP process allowed some facilities to accept permitted emission reductions to lower their surrogate visibility impact (emissions over distance or Q/d) to just below the threshold for selection rather than requiring implementation of cost-effective emission controls that were identified through a four-factor analysis . . . The NPS agrees that in cases where recent actual emissions would not have triggered source selection, permit adjustments may be an appropriate anti-backsliding measure. However, in cases where recent actual emissions exceed the established selection criteria ( e.g., Kingsford Manufacturing Company and Owens-Brockway Glass Container Inc.), we recommend that facilities not be allowed to back out of selection by accepting permitted emission reductions in lieu of implementing cost-effective emission controls identified through four-factor analyses.” Response: The EPA agrees that in cases where recent actual emissions would not have triggered source selection, permit adjustments are an appropriate anti-backsliding measure. As demonstrated in table 4 of our proposed rulemaking, this was the case for all facilities that accepted Plant Site Emission Limits (PSELs) for nitrogen oxides (NO X ), sulfur dioxide (SO 2 ), and/or coarse particulate matter (PM 10 ), except the two cases noted by the NPS, Kingsford Manufacturing Company and Owens-Brockway Glass Container Inc. 16 Information on Kingsford Manufacturing was included as a footnote to table 4, “ODEQ reviewed Kingsford Manufacturing Company which originally screened into analysis with a Q/d = 8.39 based on actual emissions as reported to the 2017 National Emissions Inventory (NEI) because a 2017 PSEL was not available at that time. However, in a letter dated May 22, 2020, ODEQ acknowledged a 2019 permit modification that had already lowered PSELs for NO X , SO 2 , and PM 10 to a Q/d = 4.02.” ODEQ used this 2019 information, and the accompanying PSEL conditions submitted in the Title V permit for approval into Oregon's SIP, in making its May 22, 2020, determination that “that Kingsford is not required to perform a four factor analysis for their Springfield facility during this round of the Regional Haze program.” 17 In a case like Kingsford Manufacturing Company, where contemporaneous information shows a significant, permanent change in emissions, we believe it was reasonable for Oregon to reassess the agency's source selection and control determination to incorporate more recent information. 16 See 89 FR 13622 (February 23, 2024) at page 13639. 17 See 142_haze-KingsfordManufCo.pdf in the docket for this action. Similarly, ODEQ's evaluation of control for the Owens-Brockway facility was influenced by contemporaneous events. Owens-Brockway was one of the few facilities that identified feasible cost-effective controls in the four-factor analysis process. 18 As noted by the commenter, in an October 27, 2020, letter ODEQ concurred with the findings that combined control of NO X , SO 2 and PM by catalytic ceramic filters (CCF) was cost-feasible for glass-melting furnaces A and D at the Owens-Brockway Portland facility. 19 Subsequently, in June 2021, ODEQ initiated an enforcement response for air quality violations unrelated to the regional haze program. 20 The enforcement response was still in progress during the summer of 2021, therefore ODEQ could not rely on the remedy being negotiated to resolve the human health violations (shutdown or imposition of pollution controls on Furnaces D). 21 Instead, ODEQ negotiated and submitted a separate August 9, 2021, order focused specifically on reasonable progress for regional haze program. 22 This order enshrined the shutdown of Furnace A and associated emissions reductions. 18 See 149_haze-Owens-Brockway-FFA.pdf in the docket for this action. 19 See 150_haze-Owens-Brockway.pdf in the docket for this action. 20 See https://www.oregon.gov/deq/programs/pages/owensbrockway.aspx. 21 See 701_OwensBrockway2020-208MAO.pdf in the docket for this action. 22 See 151_SAFOOwensBrockway0840001.pdf in the docket for this action. The EPA disagrees with the commenter that the Regional Haze Rule prohibits a state which selects sources based on allowable emissions from refining its source selection based on permanent and enforceable reductions in allowable emissions. ODEQ determined that sources with a Q/d <5 based on PSELs are not significant contributors to visibility impairment in Class I areas. Hence, controls on these sources are not necessary for reasonable progress. This holds true regardless of when in its SIP development process ODEQ made the determination. Most importantly, ODEQ ensured the PSEL reductions upon which it relied to determine that controls on Owen-Brockway were not necessary and were permanent and enforceable by submitting a source-specific SAFO and conditions from Owens-Brockway's title V permit. 4. Use of a Stipulated Agreement and Final Order (SAFO) Versus a Unilateral Order Comment: “. . . Oregon proposed alternative compliance options for several facilities in lieu of reasonable, cost-effective controls identified through the four-factor review process. In general, the NPS has concerns with this approach and previously shared this view with Oregon in staff-to-staff meetings between 2020 and 2023, and in writing via October 2021 public comments on the draft SIP and August 2023 consultation comments on the Oregon SIP supplement . . . The NPS also agrees that alternative compliance measures can be considered reasonable when accompanied by a technical demonstration that the emission reductions achieved will be equivalent to or better than those that would have resulted from requiring the controls identified through four-factor analysis. The NPS recommends that EPA require a technical demonstration detailing the actual emission reductions that will be achieved through alternative compliance and why the alternative compliance options are reasonable in light of the four statutory factors.” Response: The EPA disagrees that additional technical demonstrations are required to justify ODEQ's determinations of the controls necessary for reasonable progress. As explained in our response to comment in section II.A.2 “Use of the Four Statutory Factors in Determining Reasonable Progress” of this preamble, characterizing the January 2021 “Preliminary Determination of Cost Effective Controls for Regional Haze” letters as final four-factor control determinations and all subsequent correspondence and decisions after the preliminary letters as being outside the four-factor process is not an accurate portrayal of the Oregon process. Under OAR 340-223-0110(1), each affected facility was required to conduct a four-factor analysis compliant with OAR 340-223-0120 Four Factor Analysis. Using its authority under OAR 340-223-0120(3), ODEQ adjusted the four-factor analyses for consistency with basic inputs such as interest rates, equipment lifetime, and potential to emit (PSEL) in determining the proposed cost-effective controls. 23 However, it is clear from the text that the January 2021 preliminary letters are not final determinations nor independent four-factor analyses in themselves, “Based on the information provided in the four factor analysis, the cost information that you submitted, the additional information you provided, and the process DEQ is proposing to use to screen facilities, DEQ estimates the following controls are likely to be required at your facility . . . If you disagree with or would like to discuss DEQ's preliminary determination as outlined in this letter, we encourage you to reach out to the DEQ now.” 24 23 See 89 FR 13622 (February 23, 2024) at page 13641. 24 See 108_haze-BosieCascade-Medford.pdf . Under OAR 340-223-0110(1), if a source accepted ODEQ's preliminary determination, ODEQ could finalize the determinations in a unilateral order under OAR 340-223-0130, SAFO under OAR 340-223-0110(2)(b)(B), or other enforceable mechanism such as a permit modification. 25 However, since nearly all the affected facilities asserted no feasible cost-effective controls in the four-factor analyses, this initiated a process from January 2021 to August 2021 to review additional information regarding the technological feasibility and cost of controls pursuant to OAR 340-223-0120(2), to determine the controls necessary to select sources and for reasonable progress, and impose these controls either through a unilateral order or SAFO. 25 See 136_GilchristNoticeofApplicationforESPinstall20210608.pdf. Each of the compliance options in OAR 340-223-0110(2) are either part of Oregon's source selection methodology or grounded in the four-factor analysis required by OAR 340-223-0110(1) and 0120. We do not interpret OAR 340-223-0110(2) as permitting alternatives to the requirements of the CAA or Regional Haze Rule. Rather, entering into a SAFO (agreed order) is an alternative administrative mechanism to impose controls necessary for reasonable progress that would have been contained in a unilateral order. Our review of Oregon's Regional Haze SIP submission indicates that ODEQ continued to consider the four factors in its engagement with each of the sources after issuance of the preliminary determination letters. This is documented in appendices 1 through 6 of Oregon's November 22, 2023, SIP supplement. 26 After considering the additional information regarding technological feasibility, cost of controls, energy and non-air quality impacts, and time necessary to impose the controls, ODEQ determined the appropriate administrative mechanism to impose the enforceable emission limitations. In most cases the most efficient and effective mechanism was a SAFO issued under OAR 340-223-0120(2). 26 See documents 246 through 251b included in the docket. Based on our review of the administrative record, the EPA does not believe that the differences between the January 2021 preliminary control determinations and the final four-factor control determinations in the August 2021 SAFOs are a function of the enforceable mechanism used (a unilateral order under OAR 340-223-0130 versus a SAFO issued under OAR 340-223-0110(2)). Instead, as discussed in our facility-specific responses to comment, the differences appear to be a result of ODEQ's consideration of technical feasibility and cost as documented in appendices 1-6 of the November 2023 supplement. Given that these SAFOs are outgrowths of ODEQ consideration of the four factors, rather than other factors, we disagree that additional evaluation is necessary. 5. Compliance Deadlines Comment: “In the 2023 SIP supplement, Oregon extended the compliance deadlines for emission unit replacements (associated with alternative compliance) from July 31, 2026, to July 31, 2031. This extended deadline is well beyond the end of the current regional haze planning period and will allow current emissions from affected facilities to continue without mitigation for an additional five years. In their 2019 regional haze guidance document, the EPA states that the reasonable progress goals “for the second implementation period are to be based only on the combined effect of the LTS measures with compliance dates on or before December 31, 2028.” Response: The citation to the EPA's 2019 Guidance provided by the commenter deals with modeling, and notes that states cannot claim projected 2028 emissions reductions in the modeling if those control measures are not in effect by 2028. 27 This is not a regulatory prohibition on controls outside the implementation period. The relevant regulatory citation is 40 CFR 51.308(f)(2)(i), which requires that if a state concludes that a control measure cannot reasonably be installed and become operational until after the end of the implementation period, the state may not consider this fact in determining whether the measure is necessary to make reasonable progress. The clear implication is that controls after the end of the implementation period are allowable under the RHR if the determinations are reasonable. An example is Oregon's regional haze plan for the first implementation period which adopted regulatory provisions to cease coal-fired electricity generation at the Boardman facility, however implementation of the measures (closure of the coal-fired operations) would not occur until the second planning period, in 2020. 28 Another example is Washington's regional haze plan for the first implementation period which required closure of the coal-fired units at the TransAlta facility, however closure of units was phased in 2020 and 2025, during the second implementation period. 29 27 See 2019 Guidance, at page 46, Regional scale modeling of the LTS to set the RPGs for 2028. 28 See 77 FR 50611 (August 22, 2012). 29 See 79 FR 33438 (June 11, 2014). With respect to Oregon's determinations under the four factors, some pollution controls were imposed under 340-223-0110(2)(b)(B) and (C), which required installation of identified controls no later than July 31, 2026. For a subset of units, ODEQ used authority under OAR 340-223-0110(2)(b)(E), which allowed replacement of an emissions unit by no later than July 31, 2031. The comment, “[i]n the 2023 SIP supplement, Oregon extended the compliance deadlines for emission unit replacements (associated with alternative compliance) from July 31, 2026 to July 31, 2031” is a misreading of the Oregon regional haze rules. The July 31, 2026, compliance deadlines under OAR 340-223-0110(2)(b)(B) and (C) apply to retrofit options. Instead, ODEQ followed the compliance deadline in OAR 340-223-0110(2)(b)(E) which applies to emissions unit replacement. These regulatory provisions were adopted by the Oregon Environmental Quality Commission after a full public comment period from May 28, 2021, to June 30, 2021, and a hearing conducted on June 28, 2021. 30 30 See 702_staff report EQC meeting_072321_ItemJ_RegionalHaze.pdf. We also believe that the comment, “[i]n the 2023 SIP supplement, Oregon extended the compliance deadlines for emission unit replacements (associated with alternative compliance) from July 31, 2026 to July 31, 2031” is a misreading of the record. An example is the Northwest Pipeline Baker City facility. The original SAFO, effective August 9, 2021, did not include a concrete deadline for emissions unit replacement. 31 In response to EPA comment, as indicated in the amended SAFO, “DEQ received comments from the U.S. Environmental Protection Agency on the Regional Haze State Implementation Plan, requiring amendments to the SAFO,” ODEQ added a concrete compliance deadline for unit replacement and submitted the amended SAFO as part of the 2023 supplement. 32 In our review of the record we see no evidence to suggest ODEQ modified compliance deadlines without a clear basis under the four factors. 31 See 146_SAFONorthwestPiplineBaker.pdf, included in the docket for this action. 32 See 215_3.3.1_Attachment1.7.1_NWPipeline_Baker_01-0038-A1_SAFO.pdf, included in the docket for this action. Lastly, with respect to ODEQ's application of OAR 340-223-0110(2)(b)(E), this must be viewed in the context of the overall mix of timelines (most before 2028) and the other controls imposed (primarily by July 31, 2026). ODEQ evaluated 43 emissions units and a total of 62 control devices. 33 Of this universe, ODEQ determined that unit replacement may be a reasonable control option for 10 units. Given the complexity and logistical challenges of complete emissions unit replacement, we believe ODEQ's selective use of the full compliance deadline allowable under OAR 340-223-0110(2)(b)(E) is reasonable under the four factors of CAA section 169A(g)(1) and 40 CFR 51.308(f)(2), including the time necessary for compliance. 33 April 29, 2022, regional haze SIP, at page 17. 6. Standards for Emissions Unit Replacement Comment: “New emission units generally have lower emissions than older units. However, a wide variety of emissions are possible from new units. In several places, the Oregon SIP requires that new units “shall meet the most recent permitting standards and requirements for new emission units (including but not limited to New Source Performance Standards) in place at the time of submitting a permit application.” As the NPS shared with ODEQ during SIP supplement consultation, this may not be adequately protective because new source performance standards (NSPS) are frequently less stringent than best available control technology (BACT)-level controls or those that may be deemed reasonable through a four-factor analysis.” Response: As a fundamental matter, the EPA disagrees that best available control technology (BACT) is an appropriate threshold for evaluating Oregon's determinations of the controls necessary for reasonable progress for the second planning period. For a source recently permitted to BACT standards, it may be reasonable for the state to argue that these controls are equivalent to or more stringent than controls that would be derived under the regional haze four-factor process. 34 However, the inverse is not true. It is not the EPA's expectation that controls derived under the regional haze four-factor process necessarily meet the stringency level of BACT. 34 See 2019 Guidance, at page 22-23. With respect to the comment that Oregon's control determinations may not be adequately protective because new source performance standards (NSPS) are frequently less stringent than those that may be deemed reasonable through a four-factor analysis, we disagree. Many of the existing units that Oregon reviewed emit significantly more NO X , carbon monoxide, and volatile organic compounds than new units meeting the emission limits in 40 CFR part 60, subpart Subpart JJJJ—Standards of Performance for Stationary Spark Ignition Internal Combustion Engines. 35 Thus, replacement of these existing units with new units meeting the NSPS will result in substantial emissions reductions. Therefore, we disagree that these standards are not adequately protective. 35 Most recently updated August 10, 2022 (87 FR 48606). Finally, under OAR 340-223-0110(1) all affected facilities submitted four-factor analyses. These analyses were conducted by independent consultants and/or certified professional engineers on behalf of the sources. In all instances that ultimately resulted in unit replacement, these independent consultants and/or certified professional engineers provided four-factor demonstrations that there were “no feasible cost-effective” controls. Rather than accepting these “no feasible cost-effective” control demonstrations pro forma, ODEQ used its authority under OAR 340-223-0110(2)(b)(E) to compel significant future reductions (emission unit replacement) beyond the initial four-factor analyses. The example of Northwest Pipeline, Baker City is illustrative. This facility has three natural gas-fired reciprocating engines dating from 1956 (EU1) and one engine dating from 1981 (EU2). The four-factor analysis asserted the only feasible technology was low emission combustion retrofit with calculated cost-effectiveness of $25,850 for EU1 and $24,243 for EU2. Considering the significant emissions reductions from replacing these old engines, we believe this is a reasonable approach to considering the four statutory factors in determining the controls necessary for reasonable progress. 7. Wauna Facility—Biomass-Fired Fluidized Bed Boiler Comment: “The Georgia Pacific—Wauna Mill and Roseburg Forest Products—Dillard have the highest cumulative impact on NPS Class I areas. The NPS is generally satisfied with the outcome of the control determinations for these facilities. However, we note that ODEQ has not addressed the NPS recommendation to evaluate addition of low NO X burners and flue gas recirculation to reduce NO X emissions from the Georgia Pacific—Wauna Mill biomass-fired fluidized bed boiler which could further reduce haze-causing emissions from that facility.” Response: The EPA reviewed Oregon's April 2022 and November 2023 submissions and associated documents. We found citations related to the biomass-fired fluidized bed boiler at the Wauna facility. However, we found no record of a prior comment by the NPS directly related to the “NPS recommendation to evaluate addition of low NO X burners and flue gas recirculation to reduce NO X emissions from the Georgia Pacific—Wauna Mill biomass-fired fluidized bed boiler.” Below is a summary of the administrative record reviewed by the EPA. On June 15, 2020, Georgia-Pacific submitted a four-factor analysis that stated, “LNB [low NO X burners] are not feasible for GP Wauna's Fluidized Bed Boiler. The natural gas burners are only for auxiliary use and do not drive NO X emissions from the unit. The boiler already employs SNCR to reduce NO X emissions from the bubbling fluidized bed.” 36 As part of the May 2021 FLM consultation draft process, the NPS comments focused entirely on selective catalytic reduction (SCR) cost calculations for this unit and made no mention of LNB. 37 In its October 29, 2021, comments NPS did state, “We recommend that ODEQ's draft SIP more thoroughly address emissions from GP Wauna by including an analysis of emissions from the Fluidized Bed Boiler.” 38 However, the contents of the comments again focused exclusively on SCR costs, with no specific mention of LNB at this unit. Additional comments submitted on August 29, 2023, as part of the FLM consultation process for the November 2023 regional haze supplement, make no mention of SCR or LNB at this specific unit. 39 In the absence of more concrete information, we believe it was reasonable for ODEQ to rely on the determination in the four-factor analysis that LNB was not feasible for the Fluidized Bed Boiler (FBB) because “natural gas burners are only for auxiliary use and do not drive NO X emissions from the unit” 40 and was, therefore, not put forward by ODEQ as a potential control measure in the January 2021 “Preliminary Determination of Cost Effective Controls for Regional Haze” letter for this facility. 41 The comment does not present information that clearly refutes the determination in the four-factor analysis that LNB/FGR is not feasible for the FBB at the Wauna Mill. 36 See 129_haze-GeorgiaPacific-WaunaMill-FFA.pdf, at page 2-10. 37 See Chapter 6.3.4 Federal Land Manager Comments and DEQ Responses of Oregon's April 2022 submission, at page 126. 38 See 001_1.1_StaffReport_wAttachments, Attachment C.pdf, at page 15 of 58. 39 See 703_NPS Oregon Regional Haze SIP Supplement Consultation.pdf. 40 See 129_haze-GeorgiaPacific-WaunaMill-FFA.pdf, at page 2-10. 41 130_haze-GeorgiaPacificWauanMill.pdf. 8. Georgia-Pacific—Toledo LLC—Final Control Determination Comment: “We believe that selective catalytic reduction (SCR) remains a feasible and likely more rigorous NO X emission control option for this facility than either of the options proposed . . . The NPS continues to recommend that ODEQ and EPA evaluate the incremental cost-effectiveness of SCR versus the proposed low NO X burners and flue gas recirculation control option.” Response: The June 2020 four-factor analysis prepared by ALL4 on behalf of the Georgia-Pacific Toledo facility calculated cost effectiveness for LNB with flue gas recirculation (LNB/FGR), selective non-catalytic reduction (SNCR), and SCR, as shown in table 2. 42 42 See 249_3.3.2_Appendix4_Georgia.Pacific.Toledo_Correspondence.pdf., at page 349 of the PDF. Table 2—Cost-Effectiveness of Controls ($/Ton NO X ) Georgia-Pacific—Toledo LLC Control technology Unit Calculated using PSEL Calculated using 2017 actuals LNB/FGR EU-11 No. 4 Boiler $9,717 $10,042 LNB/FGR EU-13 No. 1 Boiler 4,769 7,083 LNB/FGR EU-18 No. 3 Boiler 14,822 21,024 SNCR EU-11 No. 4 Boiler 6,613 7,630 SNCR EU-13 No. 1 Boiler 5,191 7,706 SNCR EU-18 No. 3 Boiler 8,569 12,126 SCR EU-11 No. 4 Boiler 11,067 12,173 SCR EU-13 No. 1 Boiler 8,623 12,681 SCR EU-18 No. 3 Boiler 13,579 19,057 Georgia-Pacific's June 2020 four-factor analysis indicates that SCR may be cost-effective for the No. 1 Boiler when calculated using permitted allowable emissions (PSELs) under ODEQ's methodology. 43 However, as described in Section 5 of the November 2023 regional haze supplement, Oregon found Georgia-Pacific's April 30, 2021, follow-up four-factor analysis correspondence compelling with respect to both cost of compliance and energy and nonair quality environmental impacts of compliance. 44 Therefore, Oregon issued its final control determination to require LNB/FGR or unit replacement for all three boilers under order number 21-0005, amendment A1. 45 43 See 132_haze-GeorgiaPacific-Toledo-FFA.pdf, at page 2-21. 44 See 201_RH_Round2_Supplement_Final.pdf, at page 17 and 249 _ 3.3.2_Appendix4_Georgia.Pacific.Toledo_Correspondence.pdf, at page 394 of the PDF. 45 See 241_3.3.1_Attachment5.8_21-0005_SAFO_A1_GeorgiaPacific_Toledo (final signed).pdf. We believe that it was reasonable for ODEQ to appropriately weigh the “energy and nonair quality environmental impacts of compliance” as well as the cost of compliance considerations raised in the April 30, 2021, four-factor correspondence when determining the controls necessary for reasonable progress. 9. Georgia-Pacific—Toledo LLC—Emission Limit Comment: “[I]t is unclear how the emission limit associated with compliance option 1 was derived. We suggest that a four-factor analysis or technical demonstration justifying the 0.09 lb/MMBtu emission limit for NO X associated with the proposed control option would improve the SIP.” Response: The emissions limit associated with compliance option 1 was discussed in the April 30, 2021, four-factor analysis correspondence, included in Appendix 4 of ODEQ's 2023 supplement. 46 As stated in the April 30, 2021 letter, this limit was based on Georgia Pacific's internal engineering experience and discussions with outside vendors. In our review of the four-factor analysis, the 0.09 lb/MMBtu emission limit for NO X represents a 68% reduction for boiler 1, a 45% reduction for boiler 3, and a 68% reduction for boiler 4. 47 These reductions are generally comparable to the estimated emissions reductions in the four-factor analyses (79% for boiler 1, 47% for boiler 3, and 53% for boiler 4) which were calculated on a tons per year basis. 48 As calculated in our supporting memo included in the docket for this action, the 0.09 lb/MMBtu emission limit for NO X is comparable and slightly more stringent than the EPA's emissions factors contained in AP-42: Compilation of Air Emissions Factors from Stationary Sources for large wall-fired boilers controlled with flue gas recirculation. 49 Therefore, we believe ODEQ's selection of the final emissions limit is adequately justified, documented, and an acceptable means of refining the estimated emission rate contained in the four-factor analysis for the purposes of characterizing the cost of compliance. 50 46 See 249_3.3.2_Appendix4_Georgia.Pacific.Toledo_Correspondence.pdf, included in the docket for this action. 47 See 713_GP Toledo_supporting memo.pdf, included in the docket for this action. 48 Id. 49 See https://www.epa.gov/air-emissions-factors-and-quantification/ap-42-compilation-air-emissions-factors-stationary-sources and 711_AP42_1.4_natural_gas_combustion.pdf, included in the docket for this action. 50 See 2019 Regional Haze Guidance at pp. 29-32. 10. Georgia-Pacific—Toledo LLC Emissions—Unit Replacement Comment: “Nevertheless, the NPS supports compliance option 1. Installation of low NO X burners and flue gas recirculation will secure a 64% NO X reduction from Georgia-Pacific—Toledo LLC during the second implementation period (2018-2028). In contrast, compliance option 2 would defer emission reductions for an additional five years, beyond the end of the planning period.” Response: In the April 30, 2021, response letter to ODEQ, Georgia-Pacific stated, “The GP Toledo Mill has three affected power boilers (Nos. 1, 3, and 4 Power Boilers) and needs flexibility in determining if burners will be replaced in each unit or whether one or two new boilers will be constructed to replace these three units . . . Steam supply is a significant operational consideration for any pulp and paper manufacturing facility. Each GP mill requires steam in the pulp production process as well as the papermaking process. As such, changes to steam producing assets require substantial consideration of and planning for the assets themselves as well as the entire pulp and paper manufacturing process to minimize disruptions to overall mill operations. Both mills will need sufficient time to plan the boiler projects with both internal and external engineering resources, and then implement the changes with as little interruption to mill operations as possible. Therefore, GP is requesting an extended timeframe for implementation of these boiler projects.” As noted in the 2023 supplement to the regional haze plan, ODEQ considered this correspondence in determining under CAA section 169A(g)(1) and 40 CFR 51.308(f)(2) “the time necessary for compliance” that a deadline of July 31, 2031, was appropriate should complete emission unit replacement be necessary. 51 51 See 201_RH_Round2_Supplement_Final.pdf, at page 17. The EPA has reviewed SAFO 21-0005, the subsequent amendment effective December 5, 2022, and the associated four-factor analysis. Based on the four-factor analysis, installing LNB and flue gas recirculation based on 2017 actual emissions had cost effectiveness figures of EU 11 = $10,042, EU 13 = $7,083, and EU18 = $21,024. 52 Considering that for two of the boilers the cost effectiveness figure exceed $10,000/ton, we believe it was reasonable for ODEQ to provide flexibility on a unit-by-unit basis in providing the two compliance options: (1) full unit replacement by 2031; or (2) installation of LNB with flue gas recirculation by 2026. 52 See 132_haze-GeorgiaPacific-Toledo-FFA.pdf, at page 2-17. 11. Cascade Pacific Pulp, LLC—Halsey Pulp Mill Power Boilers—SCR Comment: “The Oregon SIP requires this facility to eliminate use of no. 6 fuel oil by June 30, 2024, replace power boiler #2 with a new emissions unit that will achieve a limit of 0.036 lbs NO X /MMBtu as a 30-day rolling average no later than July 31, 2031, and, upon replacement of power boiler #2, limit emissions from power boiler #1 to no more than 27 tons of NO X per year . . . SCR may still be a feasible and more rigorous NO X emission control option for the power boiler #1 than the control determination requires. In its Good Neighbor Plan, EPA recently determined that SCR is technically feasible to control NOx emissions from natural gas-fired industrial boilers at pulp and paper mills.” Response: On June 15, 2020, Cascade Pacific Pulp, LLC submitted a four-factor analysis prepared by the environmental consulting service ALL4 Inc. 53 With respect to SCR, ALL4 calculated the cost effectiveness of SCR ($/ton NO X reduced) on CPP Halsey power boiler #1 to be $16,029 based on 2017 PSEL and $38,292 based on 2017 actual emissions. ALL4 calculated the cost effectiveness for CPP Halsey power boiler #2 to be $28,349 based on 2017 PSEL and $204,083 based on 2017 actual emissions. 54 Using its authority under OAR 340-223-0120(3), ODEQ preliminarily adjusted the four-factor analyses using conservative inputs such as interest rate (3.25%), equipment lifetime (30 years), and potential to emit (PSEL). However, after these adjustments, ODEQ did not find SCR cost-effective at the $10,000 threshold as evidenced by the agency's determination to propose LNB with flue gas recirculation instead of SCR for power boiler #1 as part of the “Preliminary Determination of Cost Effective Controls” letter. 55 53 See 110_haze-CascadePacificPulp-HalseyMill-FFA.pdf. 54 See 110_haze-CascadePacificPulp-HalseyMill-FFA.pdf, at page 2-21. 55 See 111_haze-CascasePacificPulp.pdf. In our 2019 guidance, we recommend the use of recent actuals or projected actuals rather than allowable emissions (PSELs) in calculating cost-effectiveness. 56 Considering the SCR cost effectiveness at these units based on the recent actual emissions contained in the four-factor analysis ($38,292 for power boiler #1 and $204,083 for power boiler #2), we have no reasonable basis to dispute Oregon's determination that SCR was not cost effective for these units. 56 See 2019 Guidance, Selection of emissions information for characterizing emissions-related factors, at page 30 and Use of actual emissions versus allowable emissions, at page 17. 12. Cascade Pacific Pulp, LLC—Halsey Pulp Mill Power Boilers—LNB Comment: “It is unclear how the future emission limit associated with the replacement of power boiler #2 was derived. We believe that a four-factor analysis or technical demonstration justifying the 0.036 lbs NO X /MMBtu emission limit would improve the SIP.” Response: The EPA acknowledges that ODEQ's decision-making for the August 25, 2023, final control determination under SAFO 22-3501-A2 could have been clearer in the SIP submissions. However, ODEQ's SIP submissions ultimately meet the requirement in 40 CFR 51.308(f)(2)(iii) to document the technical basis, including modeling, monitoring, cost, engineering, and emissions information, on which the state relied to determine the emission reduction measures that are necessary for reasonable progress. With respect to Cascade Pacific Pulp, LLC—Halsey Pulp Mill Power Boilers, ODEQ's November 2023 regional haze supplement documents ODEQ's process for its reasonable progress determination. 57 The 2023 supplement states that, on January 21, 2021, ODEQ proposed that LNB with flue gas recirculation on power boiler #1 could be cost effective and included this control as part of the “Preliminary Determination of Cost Effective Controls” letter. 58 On January 27, 2021, Cascade Pacific Pulp responded by questioning ODEQ's cost analysis and submitting a revised cost analysis performed by ALL4 consulting service for power boiler #1. 59 On August 9, 2021, Cascade Pacific Pulp and ODEQ entered into SAFO 22-3501, establishing installation of a LNB on power boiler #1. 60 On February 1, 2022, the parties agreed to amend the order to allow the option of unit replacement for power boiler #1. 61 On August 25, 2023, the parties again amended the order to allow the unit replacement of power boiler #2 instead of power boiler #1. 62 57 See 201_RH_Round2_Supplement_Final.pdf, at page 18. 58 See 248_3.3.2_Appendix3_Cascade.Pacific.Pulp_.Halsey_Correspondence.pdf at page 579 of the PDF. 59 I.d., at page 581 of the PDF. 60 See 112_SAFO22-3501CPPHalsey.pdf. 61 See 201_RH_Round2_Supplement_Final.pdf, at page 18. 62 See 243_3.3.1_Attachment6.1_22-3501_A2_SAFO_CPP_Halsey_Final_signed.pdf. While ODEQ's documentation could have been more robust, the commenter does not provide information to indicate that ODEQ's determination was unreasonable or inadequate. ODEQ's January 21, 2021, preliminary control determination and subsequent SAFO modifications are direct outgrowths of ODEQ's review of and action upon the June 15, 2020, four-factor analysis. This four-factor analysis (as revised on January 27, 2021) asserted that LNB with flue gas recirculation was not cost-effective for power boiler #1 ($10,559 per ton reduced based on PSEL and $26,446 per ton reduced based on 2017 actual emissions). 63 To the extent that LNB with flue gas recirculation (as proposed in ODEQ's preliminary determination) may be above Oregon's $10,000 per ton cost effectiveness threshold, as asserted by the ALL4 analysis, or may be below Oregon's $10,000 per ton threshold with a different assumption set, the EPA does not see a compelling basis to dispute ODEQ's final control determination. 63 See 248_3.3.2_Appendix3_Cascade.Pacific.Pulp_.Halsey_Correspondence.pdf, at page 581 of the pdf. First, it is clear from our review of the administrative record that ODEQ conducted a multi-year, extensive effort to evaluate control options under the four statutory factors of CAA section 169A(g)(1). 64 Second, Oregon's $10,000 per ton cost effectiveness threshold is one of the highest in the nation, if not the highest, applied specifically under the regional haze program. If the EPA were to conduct its own independent cost analysis, the EPA would not necessarily use a $10,000 threshold for determining reasonable progress controls. Third, ODEQ chose a more stringent methodology than the EPA's 2019 guidance recommends in calculating cost effectiveness using allowable emissions (PSELs). Use of recent actuals or projected actuals in accordance with the 2019 guidance 65 would almost certainly result in a less stringent outcome than ODEQ's methodology. Lastly, as noted in a previous response to comment, Oregon engaged in a rigorous process to improve the accuracy of the facility submitted four-factor analyses, rather than accepting the initial conclusions pro forma. In the case of Cascade Pacific Pulp Halsey, Oregon's process resulted in significant future emissions reductions (unit replacement) well beyond the four-factor analysis submitted pursuant to OAR 340-223-0110(1) which concluded there were no feasible cost-effective controls. More details on the 0.036 lbs NO X /MMBtu emission limit imposed by ODEQ are discussed in section II.A.13 of this preamble. 64 See 248_3.3.2_Appendix3_Cascade.Pacific.Pulp_.Halsey_Correspondence.pdf. 65 See 2019 Guidance, Selection of emissions information for characterizing emissions-related factors, at page 30 and Use of actual emissions versus allowable emissions, at page 17. 13. Cascade Pacific Pulp, LLC—Halsey Pulp Mill Power Boilers—Compliance Deadline and Emission Limit Comment: “The NPS supports the elimination of #6 fuel oil. However, replacing power boiler #2 on the identified schedule and requiring emission reductions from power boiler #1 following power boiler #2's replacement will defer emission reductions beyond the end of the planning period (see above for additional discussion).” Response: In a discussion with the EPA, ODEQ explained how the agency's perspective regarding emissions unit replacement evolved through the four-factor analysis process. 66 ODEQ found that new, purpose-built units with controls like LNB built in offered superior emissions reductions compared to the limitations of retrofitting an older unit. For units like the Halsey power boilers built in 1968, this was particularly notable. ODEQ weighed the superior emissions reductions of emissions unit replacement against the additional time necessary for compliance (2031 for unit replacement versus 2023 for retrofit in the original SAFO) and determined that this was a reasonable trade-off in considering the significantly improved emissions reductions. 66 See 712_CPP Halsey_supporting memo.pdf, included in the docket for this action. The EPA reviewed the four-factor analysis 67 and the emissions reductions expected to result from the LNB/FGR retrofit of power boiler #1 as required in SAFO 22-3501 as well as the emissions reductions resulting from replacement of power boiler #2 as required in SAFO 22-3501-A2. As described in our supporting memo to the docket, and based on our calculations, the LNB/FGR retrofit of power boiler #1 could potentially be expected to result in a reduction in NO X emissions of 0.07 lb/MMBtu, while complete emission unit replacement of power boiler #2 as the primary steam production will result in a reduction in NO X emissions of between 0.145-0.185 lb/MMBtu. Replacement of power boiler #2 will therefore result in significantly more reductions in NO X emissions than LNB/FGR retrofit of power boiler #1. Therefore, the EPA believes this is a credible rationale and indicates that the state appropriately considered the four factors in determining the controls necessary for reasonable progress. 67 See 110_haze-CascadePacificPulp-HalseyMill-FFA.pdf, included in the docket for this action. 14. International Paper—Springfield—Emission Limit Comment: “According to the four-factor analysis provided in the Oregon SIP, the recent actual emission rate achieved by the International Paper—Springfield power boiler is 0.22 lb NO X /MMBtu. Therefore, this control determination, requiring an emission limit of 0.25 lb NO X /MMBtu, may allow an increase in emissions from the primary emission unit at the facility. The Good Neighbor Plan limits NO X emissions from natural gas-fired boilers like the power boiler to 0.08 lb/mmBtu. The NPS recommends that EPA and ODEQ set a NO X emission limit consistent with the Good Neighbor Plan. The current control determination for this facility lowers the allowable permitted emissions but will not actually reduce haze-causing emissions.” Response: The EPA disagrees. The estimated recent actual emission rate is not directly comparable to the prescribed emission limit. The 0.22 lb NO X /MMBtu emission rate cited by the commenter is described in Appendix 6 of Oregon's November 2023 supplement, “All emissions used in the 4FA Report for 2017 were previously reported in the 2017 Annual report to Lane Regional Air Protection Agency (LRAPA) with one notable exception. The Power Boiler NO X emissions for the 4FA Report were determined by the Continuous Parameter Monitoring System Formula per Title V, permit condition 186.g. The NO X reported in the Annual report was based upon the maximum emission factor of 0.46 lb/MMBtu. The weighted average emission factor determined from the Continuous Parameter Monitoring System Formula is 0.2195 lb/MMBtu which was used to determine the actual NO X tons for 2017 from the Power Boiler.” 68 68 See 251b_Appendix6_InternationalPaper.SpringfieldMill_Correspondence.pdf, at page 4 of the PDF. The EPA notes that the 0.2195 lb/MMBtu figure used in the 2017 Annual report is an annual average emission factor whereas the 0.25 lb NO X /MMbtu emission limit is based on a 7-day rolling average. 69 Thus, ODEQ was reasonable in considering the emission rate the Power Boiler could achieve averaged over a rolling 7-day period rather than an annual period. Moreover, there are numerous variables and assumptions inherent in the formula used in the prior Title V permit to derive the emissions factor. In particular, Condition 186.g of the prior Title V permit included two emission factor formulas: one for natural gas flow rate less than or equal to 380MSCF/Hr and one for greater than 380MSCF/Hr. Each of these formulas contains a fixed multiplier and fixed correction factor. Any variation in each of these variables would yield a different emission rate. ODEQ was reasonable in taking these circumstances into consideration when setting the emission rate the company must achieve on a 7-day average basis. The EPA disagrees with the assertion that this short-term emission limit will lead to long-term emissions increases compared to recent actuals. Moreover, ODEQ is requiring CEMS—a reliable method of monitoring and recording emissions data. 70 This data will assure compliance with the emission rate and also inform later planning periods. 69 See 139_SAFODEQ-LRAPA-IP.pdf. 70 Id. With respect to the comments that ODEQ's reasonable progress determination will not reduce emissions, we note that reasonable progress has two prongs: the prevention of any future anthropogenic visibility impairment and the remedying of any existing anthropogenic visibility impairment. 71 The commenter is assuming that recent actuals are necessarily determinative of projected future actuals through 2028. This is not necessarily the case. Without lower PSELs, Springfield could ramp up production and emissions. Thus, ODEQ decision to lower PSELs to align with recent actuals is consistent with the Regional Haze Rule and CAA. 71 Clarifications Regarding Regional Haze State Implementation Plans for the Second Implementation Period, July 8, 2021, at page 8. Setting aside this meaningful difference in the monitoring and compliance method, the process ODEQ used to determine the controls necessary for reasonable progress for International Paper underscores the ODEQ's reasonableness. Importantly, ODEQ calculated cost thresholds based on allowable emissions (PSELs) versus recent actual emissions (2017). This decision was a driving force behind ODEQ's preliminary control determinations and enabled the state to adjust the initial four-factor analyses to ultimately determine the controls necessary for reasonable progress. As noted in our response to comment in section II.A.11 Cascade Pacific Pulp, LLC — Halsey Pulp Mill Power Boilers—SCR of this preamble, Oregon's decision to calculate cost thresholds based on allowable emissions was much more stringent than the EPA's recommendation in the 2019 guidance. 72 For a facility like International Paper, the difference between 2017 actuals (724 tons per year NO X ) 73 and allowable PSEL emissions (1692 tons per year NO X ) 74 resulted in dramatic differences in the cost effectiveness of control calculations ($ per ton of NO X reduced) as shown in table 3 of this preamble. On January 21, 2021, ODEQ used PSEL cost effectiveness of controls to propose SCR for the power boiler in the agency's “Preliminary Determination of Cost Effective Controls” letter. 75 On February 2, 2021, International Paper objected to ODEQ using allowable PSEL emissions in determining the cost effectiveness of controls. 76 International Paper also raised this issue in its September 18, 2020, letter to ODEQ stating, “In addition, we are concerned by DEQ's misdirected focus on reducing Plant Site Emission Limits (PSEL) rather than focusing upon the impact to visibility impairment of actual emissions. Focusing on PSEL in the evaluation of cost effectiveness for controls compounds the inequity of DEQ's approach to this process compared to other Western States. The Springfield Mill's cost effectiveness for actual emission reduction is well above the previously discussed threshold of $10,000/ton for all of the pollution control units listed by DEQ.” 77 72 See 2019 Guidance, Selection of emissions information for characterizing emissions-related factors, at page 30 and Use of actual emissions versus allowable emissions, at page 17. 73 See 251b_Appendix6_InternationalPaper.SpringfieldMill_Correspondence.pdf, at page 2 of the PDF. 74 Id. 75 See 138_haze-InternationalPaper.pdf. 76 See 251b_Appendix6_InternationalPaper.SpringfieldMill_Correspondence.pdf, at page 573 of the PDF. 77 Id, at page 557 of the PDF. International Paper then provided a March 15, 2021, memorandum from the ALL4 environmental consulting firm providing updated costs of controls, mirroring the parameters used in ODEQ's preliminary control determination (3.25% interest rate and 30-year equipment life). 78 In the same memorandum, ALL4 recommended that International Paper request a 179 ton per year NO X PSEL and 0.25 lb NO X /MMBtu emissions limit for the power boiler so that the calculation of cost effectiveness based on PSEL will more closely align with cost calculations based on actual emissions, yielding cost effectiveness of controls calculated to be $10,956 (LNB/FGR), $10,239 (SNCR), and $14,237 (SCR). 78 Id, at page 576 of the PDF. ODEQ's consideration of imposing SCR as part of the agency's preliminary control determination was only possible by using allowable emissions well above actual emissions, PSEL emissions (1692 tons per year NO X ) 79 versus actual 2017 emissions (724 tons per year NO X ). 80 In addition to the important fuel restriction requirements noted by the NPS, SAFO 208850 (effective August 9, 2021) was intended by Oregon as an anti-backsliding measure to prevent International Paper from future emissions growth during the second implementation period that may jeopardize reasonable progress. 79 Id. 80 See 251b_Appendix6_InternationalPaper.SpringfieldMill_Correspondence.pdf, at page 2 of the PDF. Lastly, with respect to the commenter's recommendation that EPA and ODEQ set a NO X emission limit consistent with the Good Neighbor Plan, we note that Oregon is not subject to the Good Neighbor Plan. This regulation was published on June 5, 2023 (88 FR 36654) to address the specific issue of human health impacts from ozone nonattainment, which has a different regulatory structure and requirements than the regional haze program. The EPA already determined that Oregon does not cause or contribute to ozone nonattainment in any other state. 81 For the specific set of states subject to the Good Neighbor Plan,the rule established emission limits for a broad suite of source categories including boilers in Iron and Steel Mills and Ferroalloy Manufacturing, Metal Ore Mining, Basic Chemical Manufacturing, Petroleum and Coal Products Manufacturing, and Pulp, Paper, and Paperboard Mills. 82 This is distinctively different than the regional haze four-factor analysis process which often focuses on source-specific factors in the evaluation. Another difference is that the emissions limit cited by the NPS applies only during the ozone season, directly for the purpose of addressing ozone nonattainment. Lastly, the Good Neighbor Plan for ozone estimated average cost-effectiveness per ton for pulp and paper facilities at $14,134, 83 which is not necessarily comparable to the threshold for determining the controls necessary for reasonable progress toward natural visibility conditions. Therefore, the NPS would need to provide greater detail to demonstrate the site-specific assumptions used to assert that a 0.08 lb/mmBtu limit is technically feasible and cost-effective under the four-factor regional haze analysis process, especially in light of the information in table 3 of this preamble showing that LNB/FGR, SNCR, and SCR were only possible for preliminary cost-effectiveness consideration using allowable 2017 PSEL emissions (1,692 tpy NO X ), well above actual 2017 emissions (724 tpy NO X ). 84 81 See 84 FR 22376 (May 17, 2019). 82 On June 27, 2024, the Supreme Court of the United States issued a stay of the rule pending review in the United States Court of Appeals for the District of Columbia Circuit Ohio et al. v. EPA, 603 U.S. __ (2024), available at https://www.supremecourt.gov/opinions/23pdf/23a349_0813.pdf. 83 See 88 FR 36654 (June 5, 2023), at page 36740. 84 April 22, 2022, regional haze SIP submission, at page 172. Table 3—Comparison of Cost-Effective Controls ($/Ton NO X ) International Paper—Springfield Power Boiler Control technology June 2020 FFA (PSEL) June 2020 FFA (2017 actual) March 2021 memorandum (179 ton per year NO X PSEL to align with recent actual emissions) LNB and FGR $2,928 $18,228 $10,956 SNCR 3,483 16,103 10,239 SCR 4,606 22,924 14,237 15. Owens-Brockway Glass Container Inc. Comment: “In this specific case, the NPS is aware that ODEQ is requiring the installation of controls outside of the regional haze process because of violations of the facility's particulate matter and opacity limits. The ODEQ issued a construction permit in November 2022 requiring installation of a new catalytic ceramic filter pollution control system that must be installed by June 30, 2024. The system will control multiple pollutants, including particulate matter, NO X , and SO 2 . A draft title V operating permit, currently undergoing public review, would impose new PSELs that will limit the facility's Q to 127 tons after the controls are installed, resulting in a Q/d of about 0.9 for the nearest NPS Class I area, Mount Rainier National Park in Washington . . . This control technology was also identified as reasonable based on evaluation of the four factors. The NPS agrees that installation of the ceramic filter system is reasonable and will result in meaningful reductions in haze-causing emissions. The NPS recommends EPA require incorporation of this control requirement into the regional haze SIP to ensure realization of emission reductions from control installation in this planning period.” Response: The special case of the Owens-Brockway facility is discussed in section II.A.3 this preamble. Permit modifications to implement the human-health enforcement response are still ongoing. 85 We see no basis for disapproval or continued delay of the regional haze SIP action while Oregon completes its human health enforcement response, especially considering the 75% emissions reductions from 2017 actuals and permanent shutdown of Furnace A imposed by ODEQ's August 9, 2021, regional haze-specific order. 86 85 See https://www.oregon.gov/deq/programs/pages/owensbrockway.aspx. 86 See 151_SAFOOwensBrockway0840001.pdf. B. Environmental Organizations' Comments Complete copies of the Environmental Organizations' comments and supporting attachments are included in the docket for this action. For readability, we arranged the responses to generally mirror the timeline of the Oregon process from site selection, review of controls, and imposition of controls. 1. Stationary Source Contribution Comment: “We submitted public comments to Oregon's Department of Environmental Quality (“DEQ”) on the state's draft SIP Revision on November 1, 2021, and on October 21, 2023, raising several of the same issues with Oregon's proposed regulation of stationary sources that collectively contribute 80% of the state's regional haze-forming emissions.” Response: Stationary sources do not contribute 80% of the state's regional haze-forming emissions. The emissions inventory analysis in Chapter 2.3 of Oregon's 2022 submission shows data from the EPA's 2017 National Emission Inventory (NEI). 87 Fuel combustion and process emissions associated with stationary sources account for 11% and 6%, respectively, of Oregon's PM 10 emissions. Fuel combustion and process emissions account for 14% and 4%, respectively of NO X emissions, with mobile sources accounting for 79% of NO X emissions. Fuel combustion and process emissions account for 57% and 13% of the 2017 SO 2 emissions inventory. However, as noted by Oregon, “The 2017 SO 2 inventory is largely overwhelmed by PGE Boardman's coal-fired power plant in Morrow County. With the closing of the plant in October 2020, those emissions have largely been eliminated, and the remainder of the emissions come from fuel combustion and prescribed fires.” 88 87 See April 22, 2022, regional haze SIP, at page 22-27. 88 See April 22, 2022, regional haze SIP, at page 24. The Environmental Organizations cite to Oregon's August 27, 2021, Notice of Proposed Rulemaking as the basis for the statement that stationary sources contribute 80% of the state's regional haze-forming emissions. However, the actual wording of the Notice of Proposed Rulemaking states, “Federally enforceable emission reductions and pollution controls at Title V stationary sources that collectively contribute 80% of Oregon regional haze-forming emissions from stationary sources. ” (Emphasis added) 89 The Environmental Organizations' adaptation of this quote omits the important qualifier “from stationary sources.” In intent and practice, ODEQ was referring to the EPA's draft regional haze guidance that recommended states set a source screening level such that 80% of the stationary source emissions inventory was captured. This formed the basis of Oregon's decision to set the source screening level at a quantity over distance (Q/d) = 5. This was not a statement that stationary sources contribute 80% the state's regional haze-forming emissions. Based on the most recent 2023 National Emissions Inventory trends data, 90 emissions categories associated with stationary sources contribute at most 18% of the cumulative anthropogenic PM 10, NO X , and SO 2 inventory. 91 While up to 18% is still a meaningful percentage of the overall regional haze precursor inventory, there is no evidence to support the claim that stationary sources collectively contribute 80% of the state's regional haze-forming emissions. 89 See 024_RHSIP2021.notice.pdf, at page 3, included in the docket for this action. 90 See https://www.epa.gov/air-emissions-inventories/air-pollutant-emissions-trends-data. 91 See 704_Oregon NEI data.xlsx and 705_Original NEI source data.xlsx, included in the docket for this action. For the purpose of this analysis, we conservatively assumed that all fuel combustion was attributable to stationary sources, which likely overestimates the contribution from stationary sources. 2. Oregon's Regional Haze Rule Is Inconsistent With the CAA Comment: The Environmental Organizations assert that Oregon's Regional Haze Rule codified at OAR Chapter 340, Division 223 is inconsistent on its face with the CAA and 40 CFR 51.308(f). The Environmental Organizations make four arguments in support of this comment: (1) neither the CAA nor RHR allow Oregon to unilaterally grant itself broad authority to establish an alternative compliance process that operates outside the Federal regional haze framework; (2) Oregon's rules would prevent the state from fulfilling its Federal Regional Haze obligations because they allow Oregon to provide a source with alternative compliance options that the state has not assessed through the four-factor analysis process; (3) Oregon's rules do not require ODEQ to document the technical basis for its alternative compliance decisions; and (4) Oregon's regional haze rule gives ODEQ the authority to reevaluate and reject controls deemed necessary for reasonable progress. Response: For the reasons stated below, we disagree with each of these comments. Before turning to each of the Environmental Organizations' points, we note that these comments conflict with these same Organizations' prior comments on Oregon's Regional Haze Rule. Oregon's Regional Haze Rule was adopted by the Oregon Environmental Quality Commission after a full public comment period from May 28, 2021, to June 30, 2021, and a hearing conducted on June 28, 2021. 92 Earthjustice, on behalf of the Cully Air Action Team, Earthjustice, Friends of the Columbia Gorge, Green Energy Institute, Oregon Environmental Council, National Parks Conservation Association, Neighbors for Clean Air, Northwest Environmental Defense Center, and Verde submitted comments supportive of the rulemaking stating, “We write in support of DEQ's proposed revisions to Oregon's Regional Haze rules. The revised rules reflect a reasoned, well-grounded, and pragmatic plan for implementing the Clean Air Act's visibility requirements. They will also benefit many communities in Oregon that are disproportionately burdened by pollution from emissions of PM, SO 2 , and NO X and communities that are most vulnerable to the most harmful effects of climate change. [The Clean Air Act requires] each state's strategy must be based on an analysis of emission control measures that are necessary to make “reasonable progress” towards the goal of restoring natural visibility to Class I areas. The emissions-reducing strategies in DEQ's revised Division 223 rules are consistent with EPA requirements for round II state implementation plans. The revised rules provide a strong foundation for Oregon's long-term strategy for reducing anthropogenic pollutants that impair visibility.” 92 See 702_staff report EQC meeting_072321_ItemJ_RegionalHaze.pdf. Thus, the Environmental Organizations took full advantage of their opportunity to raise concerns with Oregon's Regional Haze Rule during the state public comment process. At that time, the Organizations were highly supportive of the rule and gave ODEQ the clear impression that their rule was consistent with the CAA. The Environmental Organization do not address their stark change in position in their current comments on EPA's proposal nor repudiate their prior position. This gives the impression that the Environmental Organizations are concerned with ODEQ's application of Oregon's Regional Haze Rule, rather than the rule itself. Nevertheless, we address each of the Environmental Organizations' arguments against Oregon's Regional Haze Rule. First, we disagree that Oregon's Regional Haze Rule is disconnected from or inconsistent with the Federal Regional Haze Rule. Oregon adopted rules to implement the regional haze program at OAR Chapter 340, Division 223. The Division includes sections on source screening, four-factor analysis, options for compliance, and final orders requiring compliance. The source screening section establishes which sources are subject to Oregon's regional haze rules. Under the rule, all sources with a Title V operating permit and with a Q/d greater than or equal to 5 based on PSELs are subject to the regional haze rule. All sources subject to Oregon's regional haze rule must submit a four-factor analysis to ODEQ in accordance with OAR 340-223-0110(1) that meets the requirements of OAR 340-223-0120. The factors in OAR 340-223-0120 mirror those in the CAA section 169A(g)(1) and 40 CFR 51.308(f)(2)(i). As discussed in section II.A of this preamble, under OAR 340-223-0120(2) and (3), ODEQ is authorized to adjust the four-factor analysis to account for inaccuracies or insufficient information, and for consistency purposes. The rule further authorizes ODEQ to determine which controls would be cost effective and the time period the controls can be implemented. The regulations at OAR 340-223-0110 lay out the administrative mechanisms for imposing regional haze controls. Under this section and OAR 340-223-0130, ODEQ has the authority to order the source to install controls that ODEQ determines are cost effective on a timeline that ODEQ prescribes. Such orders are subject to appeal by the source. Alternatively, ODEQ may offer sources subject to the regional haze program the opportunity to enter into a SAFO. The rule provides five compliance options if ODEQ elects to enter into a SAFO: (1) lower PSELs to below Q/d equal to 5; (2) install controls identified by the source in a four factor analysis as cost effective for that source, provided ODEQ agrees that the controls will result in the greatest cost effective reductions; (3) install controls or reduce emissions that ODEQ determines, in its sole discretion, provide equivalent emissions reductions to controls that would be identified as cost effective for that source; (4) maintain controls that the source has already installed or maintain reduced emissions that ODEQ determines in its sole discretion have provided and will continue to provide equivalent reductions to controls that would be identified as cost effective for that source; and (5) replace emission unit with a new emission unit that meets the emission limits and requirements of the most recent applicable standard in place at the time of the permitting of the new emissions unit. Conceptually, nothing in the CAA nor the Federal Regional Haze Rule requires that states promulgate a regional-haze-specific state rule at all nor the form such a rule must take if a state elects to do so. Rather, the CAA and Regional Haze Rule provide states discretion on the manner in which they implement the regional haze program so long as the state's long-term strategy includes the enforceable emissions limitations, compliance schedules, and other measures that are necessary to make reasonable progress as determined based on a consideration of the four statutory factors and the state documents the technical basis for its decisions on the controls necessary for reasonable progress. As we discussed in the proposal and herein, Oregon's SIP submissions demonstrate that the state has done so. Moreover, each of the compliance options in OAR 340-223-0110(2) are either part of Oregon's source selection methodology or grounded in the four-factor analysis required by OAR 340-223-0110(1) and 0120. Throughout their comments, the Environmental Organizations reflect concerns with the term “alternative compliance” used to describe the administrative mechanism in OAR 340-223-0110(2) for ODEQ to enter into a SAFO with a source rather than a unilateral order. We do not interpret this as an alternative to the requirements of the CAA or Regional Haze Rule. Rather, entering into a SAFO is an alternative administrative mechanism to imposing controls necessary for reasonable progress. Our review of the subsections of OAR 340-223-0110(2) shows they are consistent with the CAA and Regional Haze Rule. The option to lower PSELs is discussed at length in sections II.A and II.B.3 of this preamble. This option is part of Oregon's method for selecting sources to undergo review and is consistent with the Regional Haze Rule. Each of the options in OAR 340-223-0110(2)(b)(B)-(D) make clear that ODEQ references the four-factor analysis as the basis to determine the acceptability of those options. For the option in OAR-340-223-0110(2)(b)(E) regarding emission unit replacement, ODEQ reasonably anticipated that sources would not evaluate unit replacement as a control option in a four-factor analysis, 93 but that unit replacement may be more cost effective or provide significantly greater emissions reductions than certain add-on controls or emissions limitations in existing emission units. Therefore, contrary to the Environmental Organizations' contention, these compliance options are grounded in the Regional Haze Rule. 93 2019 Regional Haze Guidance at p. 29. The Environmental Organizations appear to center their concerns on OAR 340-223-0110(2)(b)(C) and (D), which allow ODEQ to issue a SAFO that requires the source to install or maintain controls that achieve, in ODEQ's sole discretion, controls that provide equivalent emission reductions to controls that would be identified as cost effective for that source following the adjustment and review of the four-factor analysis. Oregon is subject to state administrative procedural requirements that require public review of the basis for its decisions. 94 In addition, we interpret Oregon's inclusion of the phrase “in its sole discretion” in OAR 340-223-0110(2)(b)(C) and (D) as necessary to preserve the durability of its SAFOs. Under OAR 340-223-0110(2), ODEQ has discretion to offer sources the option to impose controls necessary for reasonable progress through a SAFO rather than a unilateral order. A benefit of the SAFO option is avoiding an appeal under OAR 340-223-0130. Given this, Oregon was reasonable in foreclosing the possibility of a source, after having signed a SAFO agreeing to install controls, challenging whether the agreed upon control was equivalent to the controls identified as cost effective under four-factor analysis. Finally, we do not interpret OAR 340-223-0110(2)(b) as overriding EPA's authority under CAA Section 110 to determine whether the SIP submission meets CAA requirements nor the requirement in 40 CFR 51.308(f)(2)(iii) to document the basis for its decisions. As discussed in sections II.A.1, II.A.2, II.A.4, and II.B.5 of this preamble, in practice, ODEQ included in its SIP submission all of the correspondence that formed the basis for its determinations of what controls are necessary for reasonable progress. 94 See 001_1.1_StaffReport_wAttachments.pdf at p. 19; 015_4.1.1 SOS.Notice.FilingReceipt.8.27.21.pdf; Oregon Revised Statutes sections 183.310-183.690; OAR 340-011-0009. Second, with respect to the Environmental Organizations' argument that Oregon's regional haze rule prevents the state from fulfilling its regional haze obligations, ODEQ chose the compliance options in OAR 340-223-0110(2)(b) as the regulatory mechanism to effectuate its determinations of the controls necessary for reasonable progress based on the four factor analyses conducted under OAR 340-223-0120. As we stated in the proposed rulemaking for this action, reasonable progress analysis, including source selection, information gathering, characterization of the four statutory factors (and potentially visibility), balancing of the four factors, and selection of the emission reduction measures that represent reasonable progress, is a technically complex exercise, but also a flexible one that provides states with bounded discretion to design and implement approaches appropriate to their circumstances. 95 95 See 89 FR 13622 at 13629 (February 23, 2004). Accordingly, Oregon's regional haze rule requires ODEQ to make its equivalency determination based on the outcome of the four-factor analysis. Thus, we do not view Oregon's rules as permitting ODEQ to determine the controls necessary for reasonable progress without considering the four statutory factors but rather recognizing that in practice a four-factor analysis may not always yield a single, obvious control determination. As discussed in section II.A of this preamble, in practice, ODEQ carefully considered the four factors in determining the controls necessary for reasonable progress and the appropriate regulatory mechanism under OAR 340-223-0110(2)(b). Practically, the goal of the regional haze program is to impose enforceable emission limits, where possible expressed as a numerical emission limit. 96 Oregon's rules allow it to impose such a limit without rigidly adhering to a specific control technology. Nothing in the CAA nor regional haze rule prohibits this approach to achieving reasonable progress. 96 Clarifications Memo at pp. 11-12. Third, regarding documentation requirements, we do not interpret OAR 340-223-0110(2)(b) as circumventing ODEQ's state administrative procedural requirements to include in its public record the basis for its regulatory decisions. 97 OAR 340-223-0120 requires ODEQ to include in its record the additional information it uses to adjust the initial four factor analysis. Moreover, Oregon is subject to the Regional Haze Rule requirement to include in it SIP submission documentation of the technical basis, including modeling, monitoring, cost, engineering, and emissions information, on which the state is relying to determine the emission reduction measures that are necessary to make reasonable progress in each mandatory Class I Federal area it affects. 98 In recognition of this requirement, ODEQ supplemented its initial SIP submission with considerable documentation that informed the state's determination of the controls necessary for reasonable progress. 97 See OAR 340-011-0010 and 0024; See also 40 CFR 51.102. 98 See 40 CFR 51.308(f)(2)(iii). Finally, the EPA disagrees with the Environmental Organizations' argument that Oregon's regional haze rule allows the state to reevaluate and reject control measures deemed necessary for reasonable progress. This comment is predicated on the Environmental Organizations' incorrect interpretation of ODEQ's process for determining the controls necessary for reasonable progress. The Environmental Organizations presume that ODEQ's preliminary control determinations were its final control determinations. This is incorrect. See sections II.A, II.B.4, II.B.5 and II.B.6 of this preamble for EPA's interpretation and explanation of ODEQ's process. Under the Clean Air Act, the Administrator is required to approve a SIP submission that complies with the provisions of the Clean Air Act and applicable Federal regulations. 42 U.S.C. 7410(k); 40 CFR 52.02(a). Thus, in reviewing SIP submissions, the EPA's role is to approve state choices, provided that they meet the criteria of the Clean Air Act. Accordingly, this action approves the submitted state regulations as meeting Federal requirements. 3. Oregon's Use of PSEL Reductions as a Source Selection Method Comment: The Environmental Organizations also commented that ODEQ's application of Oregon's regional haze rule was inconsistent with the CAA and the Regional Haze Rule. The Environmental Organizations took issue with ODEQ's use of PSEL reductions stating: “DEQ used Oregon's alternative compliance process to offer facilities that screened into the Regional Haze program an option to screen back out from the program by agreeing to measures that would reduce their plant site emission limits (“PSEL”) so that Q/d would be below 5.00. This resulted in only 23 of the 32 screened-in sources completing the required four-factor analyses and allowed four of those 23 sources to belatedly screen back out from the program by reducing their PSELs so that Q/d is below 5.00.” Response: The regulatory provision cited by the Environmental Organizations is ODEQ's application of Oregon Administrative Rules (OAR) 340-223-0110(2)(b)(A) which allows a source to accept federally enforceable reductions of combined plant site emission limits (PSELs) of regional haze pollutants to bring the source's Q/d below 5. As stated in section II.B.2 of this preamble, this regulatory provision was adopted by the Oregon Environmental Quality Commission after a full public comment period from May 28, 2021, to June 30, 2021, and a hearing conducted on June 28, 2021. 99 The Environmental Organizations expressed support for Oregon's Regional Haze Rule, including the PSEL reduction option at that time. We also note that the option to limit PSELs aligns with Oregon's use of PSELs to initially select sources. Given that the state based initial source selection on PSELs ( i.e. allowables), Oregon offered the option for sources to lower PSELs below the significance threshold to satisfy reasonable progress (prevention of future impairment) under the regional haze program. 99 See 702_staff report EQC meeting_072321_ItemJ_regionalhaze.pdf. For the following reasons, we disagree with the Environmental Organizations' comments with respect to ODEQ's use of enforceable and permanent PSEL reductions as a means of refining source-screening or as a means of addressing reasonable progress for facilities with actual emissions below the screening threshold. As discussed in section II.A.3 of this preamble regarding similar comments submitted by the NPS, Oregon chose to use a more stringent methodology than the EPA's 2019 guidance for source screening and cost analysis based on allowable PSEL emissions rather than recent actual emissions or 2028 projected emissions. 100 Oregon intended this as (1) a method of initially capturing a broad selection of sources potentially impacting visibility in Class I areas and (2) as an anti-backsliding measure to ensure that facilities which had a Q/d less than 5 based on 2017 actual emissions (and would otherwise not be screened into analysis) do not have future emissions growth (based on allowable PSEL emissions) that could jeopardize reasonable progress. Pursuant to OAR 340-223-0110(2)(b)(A), Oregon entered into SAFOs to reduce allowable PSEL emissions to align with 2017 actual emissions. None of the facilities listed in table 4 would have been screened into review based on 2017 actual emissions. 100 2019 Guidance at page 17. Table 4—Facilities Screened in Using 2017 PSEL Q/ d 101 Facility 2017 Actual Q/d 2017 PSEL Q/d Outcome Cascades Tissue Group: A Division of Cascades Holding US Inc 3.02 63.72 No FFA—lowered PSEL to Q/d < 5.00. Timber Products Co. Limited Partnership 1.63 6.07 No FFA—lowered PSEL to Q/d < 5.00. PGE Beaver Plant/Port Westward I Plant 3.24 34.60 No FFA—lowered PSEL to Q/d < 5.00. Roseburg Forest Products—Riddle Plywood 2.10 5.29 No FFA—lowered PSEL to Q/d < 5.00. Roseburg Forest Products—Medford MDF 2.91 8.84 No FFA—lowered PSEL to Q/d < 5.00. Boise Cascade Wood Products, LLC—Medford 4.19 7.02 Conducted FFA—then lowered PSEL to Q/d < 5.00. Gas Transmission Northwest LLC—Compressor Station 12 2.33 14.13 Conducted FFA—then lowered PSEL to Q/d < 5.00. JELD-WEN 2.13 6.30 Conducted FFA—then lowered PSEL to Q/d < 5.00. Northwest Pipeline LLC—Baker Compressor Station 4.02 14.81 Conducted FFA—then lowered PSEL to Q/d < 5.00. The Environmental Organizations provided no compelling basis to demonstrate that aligning allowable PSEL emissions with actual emissions was a violation of regional haze requirements, especially when 2017 actual emissions are below the Q/d = 5 screening threshold. 101 April 29, 2022 Oregon SIP submission, Chapter 3.7 Facility-specific findings and results. For the two special cases where 2017 actual emissions were above Q/d = 5, Kingsford Manufacturing and Owens-Brockway, ODEQ had a reasoned basis for imposing permanent and enforceable emissions reductions such that the source's Q/d is less than 5. In the case of Kingsford Manufacturing, the facility already had a 2019 permit modification lowering emissions below Q/d = 5 prior to the development of four-factor analyses. 102 It was reasonable for ODEQ to consider this contemporaneous 2019 emissions information in updating the agency's source screening in 2020. The case of Owen-Brockway is more complex and described in our response to comment in sections II.A.3, II.A.15, and II.B.6 of this preamble. 102 See 142_haze-KingsfordManufCo.pdf included in the docket for this action. The Environmental Organizations focus on Boise Cascade Wood Products, LLC—Medford, Gas Transmission Northwest LLC—Compressor Station 12, JELD-WEN, and Northwest Pipeline LLC—Baker Compressor Station. These facilities all conducted four-factor analyses under OAR 340-223-0120 and then subsequently took a PSEL limit under OAR 340-223-0110(2)(b)(A). As we stated in section II.A of this preamble, Oregon determined that controls on sources with a Q/d of less than 5 based on PSELs are not necessary to make reasonable progress in the second planning period. The Environmental Organizations do not challenge this in their comments. Indeed, this is a particularly conservative source-selection method. Thus, Oregon was reasonable in not imposing controls based on a four-factor analysis for sources that have permanent and enforceable emissions limits such that their Q/d values are less than 5 based on PSELs. Moreover, as shown in table 5 of this preamble, there was only one control identified in the submitted four-factor analyses that was below the $10,000 cost per ton reduced threshold when calculated using PSEL (SCR at Gas Transmission Northwest LLC—Compressor Station 12). However, when calculated using 2017 actual emissions or projected actual emissions, the cost per ton reduced of SCR increased to $32,071 and $15,386, respectively. Considering the EPA's guidance that recommends the use of recent actuals or projected actuals in calculating cost-effectiveness, it was reasonable for ODEQ to offer the facility a PSEL reduction under OAR 340-223-0110(2)(b)(A) to align with actual emissions, especially when 2017 actual emissions at the facility were so far below the screening threshold (Q/d = 2.33). Table 5—Cost-Effectiveness of Controls ($/Ton NO X Reduced) 103 Facility Control technology Calculated using 2017 PSEL Calculated using recent actuals Calculated using projected actuals Boise Cascade Wood Products, LLC—Medford 104 SNCR $10,196 SCR 13,373 Gas Transmission Northwest LLC—Compressor Station 12 105 SCR (Unit A) 6,719 32,071 15,386 SCR (Unit B) 11,449 51,869 26,514 JELD-WEN 106 SCR—urea 19,969 SNCR—ammonia 18,135 Northwest Pipeline LLC—Baker Compressor Station 107 Low emission control (EU1) 25,850 Low emission control (EU2) 24,243 Finally, the lowering PSELs increases the likelihood that certain sources will be subject to NSR. ODEQ uses PSELs to manage emissions increases and decreases throughout the state to maintain the NAAQS and protect visibility. 108 Accordingly, changes to PSELs trigger Oregon's state and Federal new source review programs. 109 The applicability trigger often hinges on the increase in emissions over the netting basis. 110 The regulations also allow for deduction of certain unassigned emissions when determining whether an emission change requiring NSR occurs. 111 In several cases, ODEQ ordered the reduction of PSELs, the zeroing out of unassigned emissions, and reduction of the netting basis. 112 This increases the likelihood that the source will be subject to NSR and associated control technology review in the future. 103 Under OAR 340-223-0120, ODEQ required calculation of cost-effectiveness based on PSEL. However, Gas Transmission Northwest LLC was one of several facilities to provide cost-effectiveness based on recent actuals or projected actuals. To the extent supplementary information was available for a facility, the EPA added it to our analyses and tables. 104 See 107_haze-BoiseCascade-Medford-FFA.pdf in the docket for this action. 105 See 122_haze-GasTransmissionNW-Station12-FFA.pdf in the docket for this action. 106 See 140_haze-JELD-WEN-FFA.pdf in the docket for this action. OAR 340-223-0120 required cost calculation based on PSEL. 107 See 144_haze-NorthwestPipeline-Baker-FFA.pdf in the docket for this action. 108 See OAR 340-222-0010; 89 FR 22363, at page 22367 (April 1, 2024). 109 See OAR 340-224-0025. 110 Id. 111 See OAR 340-222-0055. 112 See 114 SAFO05-1849CascadesTissueGroup.pdf; 141a_Jeld wen permit mod_18-0006-TV-01-PM_2022_1.pdf; 151_SAFOOwensBrockway0840001.pdf; OAR 340-224-0070, 0270. 4. Oregon's “Alternative Compliance” Pathways Comment: The Environmental Organizations asserted generally that ODEQ's SAFOs were not outgrowths of ODEQ's considerations of the four-factor analysis. In the introduction portion of their comments, the Environmental Organizations asserted: “And instead of ordering all 17 facilities that completed four-factor analyses to implement the reasonable progress controls identified through those analyses, DEQ chose to offer agreements to all but one of the facilities—enabling them to evade the regional haze process. These agreements allowed sources to accept alternative emission reduction measures that will achieve far fewer reductions in haze-forming emissions than the highly effective pollution controls that DEQ originally identified in its 2021 control letters. The emission reductions measures in the agreements were not vetted through the four-factor analysis process. DEQ entered into the agreements without analyzing, determining, or demonstrating that they would result in emissions reductions equivalent to those reductions that would have occurred had the sources been required to install the controls identified through four-factor analyses.” Response: The Environmental Organizations argue that all changes from the January 2021 preliminary control determination letters to the final August 2021 control determinations are attributable to considerations other than regional haze. The EPA acknowledges that ODEQ's process was challenging to follow. However, in our review of the record, we have determined that ODEQ established these agreements within the framework of the Regional Haze Rule and the four statutory factors. Under OAR 340-223-0110(1) all affected facilities were required to submit four-factor analyses that comply with OAR 340-223-0120, which mirrors the Federal statutory requirement to consider the four statutory factors as outlined in CAA section 169A(g)(1) and 40 CFR 51.308(f)(2). Except for Owens-Brockway and Gilchrist Forest Products, twenty-one facilities provided four-factor demonstrations asserting “no feasible, cost-effective” controls were available. In issuing the January 2021 preliminary determination letters, ODEQ began the process of disputing the claims of “no feasible, cost- effective” controls in the four-factor analyses. As discussed in section II.A.2 of this preamble, these letters were not four-factor analyses themselves or determinations of the controls necessary for reasonable progress under OAR 340-223-120(4) or 40 CFR 51.308(f)(2). Rather, a plain reading of the letters and documentation provided by ODEQ indicates these letters were interim steps in ODEQ's refinement of the initial four-factor analyses. Given that ODEQ invited the recipients of the letters to discuss the preliminary findings with ODEQ, ODEQ clearly anticipated further refinements to the analyses. In this context, ODEQ appropriately initiated this interim process by asserting the most stringent measures that might be possible. Furthermore, the Environmental Organizations appear to ignore or mischaracterize the important correspondence included in appendices 1 through 6 included in both the state's docket for the 2023 regional haze supplement and the docket for the EPA's proposed rulemaking. 113 The initial four-factor analyses, ODEQ's refinement and preliminary letters, and this supplemental information collectively formed the basis for ODEQ's determination of the controls necessary for reasonable progress. Each of these steps in the process and associated documentation evince ODEQ's consideration of the four statutory factors consistent with 40 CFR 51.308(f)(2). 113 Documents 246 through 251 of the docket. The Environmental Organizations argue that because ODEQ's determination of the controls necessary for reasonable progress differed in some cases from the preliminary determinations that its final determinations could not have been based on a consideration of the four statutory factors. The EPA disagrees. As detailed in our facility-specific responses to NPS's comments, we have reviewed the correspondence and confirmed that ODEQ considered the four factors in making final control determinations. The supplemental correspondence indicates that ODEQ focused extensively on the technological feasibility of controls, cost of controls, and the time necessary for compliance. As discussed in our responses to NPS's comments in section II.A. of this preamble, ODEQ not choosing BACT-level controls or other controls advocated by the NPS and the Environmental Organizations does not mean that Oregon did not consider the four statutory factors. Section II.A of this preamble details our facility-specific findings under the four factors. In each case, Oregon had a rational basis under the four factors in making final determinations. Comment: In the analysis section of their comments, the Environmental Organizations asserted: “Neither the Clean Air Act nor Regional Haze Rule allow EPA or Oregon to reject viable controls identified through a four-factor analysis and offer sources alternative compliance measures that have not been analyzed against the four statutory factors, and which will not yield equivalent emission reductions. EPA's proposal to approve Oregon's alternative compliance agreements violates the principle that state determinations concerning the selection and implementation of controls necessary to meet reasonable progress requirements must be `reasonably moored' to the Clean Air Act, including the four factors listed in the statute.” Response: We agree with the general principle that a state's reasonable progress determinations must be based on consideration of the four statutory factors. As we stated in the proposal, 40 CFR 51.308(f)(2)(i) requires states to determine the emission reduction measures for sources that are necessary for reasonable progress by considering the four statutory factors. We disagree with the implication that Oregon did not do so. As we state in response to similar comments, the Environmental Organizations' argument rests on the premise that ODEQ's preliminary determination letters represented the culmination of the ODEQ's consideration of the four statutory factors and foreclosed any further consideration of those factors. This is incorrect. ODEQ's SIP submission makes clear that ODEQ concluded its consideration of the four factors subsequent to these letters, after the sources provided additional information regarding the availability of controls, cost of compliance, energy and non-air quality impacts of the controls, and time necessary to install the controls. 114 The commenters do not explain how ODEQ's consideration of the four factors prior to the preliminary determination letters is acceptable, but its consideration of the four factors after the letters is unacceptable. 114 See appendices 1-6 of Oregon's 2023 supplement. Our review of the information ODEQ included in the SIP submission indicates that ODEQ's determinations of the controls necessary for reasonable progress, particularly where its final determinations differed from its preliminary determinations, reflect ODEQ's careful consideration of technical feasibility and cost of controls—not an attempt to circumvent the requirements of 40 CFR 51.308(f)(2) as the commenters suggest. The EPA recognized in the proposal that reasonable progress analysis, including source selection, information gathering, characterization of the four statutory factors (and potentially visibility), balancing of the four factors, and selection of the emission reduction measures that represent reasonable progress, is a technically complex exercise, but also a flexible one that provides states with bounded discretion to design and implement approaches appropriate to their circumstances. 115 ODEQ's process of considering the four factors and for determining the controls necessary for reasonable progress reflect the technical challenges associated with installing retrofit controls on diverse industrial processes. For each source, the EPA is satisfied that ODEQ has done so. 115 89 FR 13622 at 13629 (February 23, 2004). Comment: The Environmental Organizations further asserted that ODEQ did not follow its own rules in entering into SAFOs. The Environmental Organizations asserted that ODEQ did not adequately determine whether the emissions reductions expected from each SAFO were equivalent to the emission reductions projected from the controls ODEQ preliminarily determined were cost effective in its letters to the sources. The Organizations stated: “But nothing in the record suggests that DEQ actually analyzed the emission reductions that would result from the alternative compliance agreements or compared them to the emission reductions that would result from installing controls identified through four-factor analyses. Many of the agreements contain several compliance options for the source that will not deliver equal emissions reductions. But rather than analyze the emissions expected from each of the compliance pathways, it appears that DEQ abandoned any effort to quantify the reductions expected from the agreements, stating in its response to comments that it did not have adequate information to allow it to determine equivalency with precision.” The Environmental Organizations further asserted that based on their own analysis, ODEQ's SAFOs will achieve far less emission reductions than the controls ODEQ initially determined were cost effective in its preliminary control letters to sources. Response: First, the Regional Haze Rule requires that the state determine the controls necessary for reasonable progress based on a consideration of the four statutory factors. As we explained in the Proposal and herein, Oregon's submission clearly demonstrates that it considered the four statutory factors in determining the controls necessary for reasonable progress. Second, Oregon followed its own rules in determining the controls necessary for reasonable progress. As stated above, the EPA disagrees that the preliminary control determinations represented ODEQ's final four factor analysis. Therefore, these preliminary determinations are not the correct barometer to measure whether an emission control will achieve equivalent emission reductions under OAR 340-223-0110(2)(C) or (D). A careful review of the SIP submission indicates that ODEQ invoked OAR 340-223-0110(2)(C) once in the case of the International Paper—Springfield Mill and OAR 340-223-0110(2)(D) once in the case of Roseburg Forest Products—Dillard. We discuss at length the appropriateness of ODEQ's reasonable progress determination for the International Paper—Springfield Mill in section II.A of this preamble. As documented in appendix 6 of ODEQ's 2023 regional haze supplement, International Paper responded to ODEQ's preliminary determination in a letter dated February 2, 2021, and a supporting memorandum dated March 15, 2021. 116 This information was cited in ODEQ's 2023 supplement as the basis for revising the preliminary determination, “On February 2, 2021, IP Springfield submitted a letter in response to DEQ's preliminary determination, explaining that the cost effectiveness of SCR installation was above the $10,000 per ton threshold for consideration.” 117 The Environmental Organizations provided no analysis or review of this follow-up correspondence to support the claim that ODEQ failed to “provide equivalent emissions reductions to controls that would be identified as cost effective for that source following the adjustment and review of a four-factor analysis.” SAFO 208850, effective August 9, 2021, requiring PSEL reductions, installation of CEMS, and fuel restrictions is precisely what ODEQ identified as cost effective for that source following the adjustment and review of a four-factor analysis. 116 See 251b_Appendix6_InternationalPaper.SpringfieldMill_Correspondence.pdf, included in the docket for this action. 117 See 2023 regional haze supplement, at page 19. With respect to Roseburg Forest Products, ODEQ did not issue a preliminary determination letter to the source, but indicated in its SIP submission that it initially approached the source with installation of SNCR on Boilers Nos. 1-3. The SIP submission indicates that the source and ODEQ then considered whether the Boiler Nos. 1, 2 and 6 could achieve a similar emission reduction by optimizing operations of the boilers. This engagement culminated in a SAFO that imposes an emission limit of 0.27 lb. NO X /mmBTU (7-day rolling average) for Boiler No. 1 and 0.26 lb. NO X for Boiler Nos. 2 and 6 (7-day rolling average). 118 The SAFO gave the facility the choice to achieve the emission limit either through installing SNCR or through boiler optimization. 119 As discussed in section II.B.2 of this preamble, the goal of the regional haze program is to impose enforceable emission limits, where possible expressed as a numerical emission limit. 120 Oregon's rules allow it to impose such a limit without rigidly adhering to a specific control technology. Nothing in the CAA nor Regional Haze Rule prohibits such a pragmatic approach to achieving reasonable progress. 118 See 157_SAFO20210809RFPDillard.pdf, included in the docket for this action. 119 Id. 120 Clarifications Memo at pp. 11-12. 5. Documentation of Oregon's Four-Factor Analysis Process Comment: “Additionally, the portion of Oregon's SIP Revision that EPA points to as supporting EPA's conclusion that Oregon adequately considered the four statutory factors does not contain any analysis of the alternative compliance measures. In this section of the SIP Revision, [O]DEQ merely explains that it sent sources control letters identifying cost-effective controls but later entered alternative compliance agreements without explaining its decision to include different and weaker controls in those agreements or how the agreements reflect the four statutory factors. For some sources, DEQ generally explains that the sources sent DEQ memoranda claiming that controls identified in the 2021 control letters were not technically feasible or cost-effective but DEQ does not include those letters in the SIP Revision, preventing EPA and the public from reviewing the source analyses. Nothing in the record supports a finding that Oregon analyzed these alternative compliance measures based on the four statutory factors.” Response: Section 5 of Oregon's 2023 regional haze supplement was added to explain changes from the January 2021 “Preliminary Determination of Cost Effective Controls for Regional Haze” letters (preliminary determinations) to the final four-factor determinations imposed by the August 2021 SAFOs. The 2023 supplement also contained appendices 1 through 6 that included the four-factor analyses submitted pursuant to OAR 340-223-0110(1), ODEQ's January 2021 “Preliminary Determination of Cost Effective Controls for Regional Haze” letters, and the correspondence from facilities in response to the preliminary determinations. As discussed in our facility-specific responses to NPS comments, our review showed that ODEQ's consideration of the correspondence in appendices 1 through 6 was grounded in the four factors in CAA section 169A(g)(1) and 40 CFR 51.308(f)(2) in making final control determinations. Contrary to the commenters' assertions, the documentation provided in Oregon's SIP submissions provide important context for understanding the Oregon process in comparing the final control determinations imposed by the August 2021 SAFOs to the four-factor analyses submitted to pursuant to OAR 340-223-0110(1). Except for Gilchrist Forest Products and Owens-Brockway, all four-factor analyses developed by environmental consulting firms and/or professional engineers on behalf of the sources determined that no feasible, cost-effective controls were available, or that further site-specific engineering analysis would be necessary. Examples are the June 2020 four-factor analyses for the Cascade Pacific Pulp—Halsey, Georgia-Pacific—Wauna, Georgia-Pacific—Toledo, and International Paper—Springfield facilities. 121 Chapter 2.3.1 Site-Specific Factors Limiting Implementation of the four-factor analysis states, “Currently known, site-specific factors that would limit the feasibility and increase the cost of installing additional controls include space constraints. A detailed engineering study for each of the controls evaluated in this report would be necessary before any additional controls were determined to be feasible or cost effective.” 122 As documented in appendices 1-6 of the 2023 regional haze supplement, it was precisely these types of technical feasibility and cost concerns that ODEQ considered both before and after issuing the agency's preliminary determinations in determining the controls necessary for reasonable progress. 123 121 See 110_haze-CascadePacificPulp-HalseyMill-FFA.pdf in the docket for this action. 122 Id, at page 2-14. 123 See document numbers 246 to 251 in the docket for this action. 6. Owens-Brockway Glass Container Inc. Comment: The Environmental Organizations referenced ODEQ's evaluation of the four-factor analysis and SAFO for the Owens-Brockway Glass Container facility in support of its [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%3A2024-22603. Public record. Not legal advice.