2024-17053. Air Plan Limited Approval and Limited Disapproval; Texas; Attainment Plan for the Rusk and Panola Counties 2010 Sulfur Dioxide Primary National Ambient Air Quality Standard Nonattainment Area; Finding of Failure To Attain the Primary ...
Table 1—2019-2022 One-Hour SO 2 Design Values for the Rusk-Panola Area
Years Martin Creek design value (ppb) 2019-2021 93 2020-2022 81 The attainment date for the area was January 12, 2022. In order for the EPA to determine that the area attained by the January 12, 2022, attainment date based solely on air quality monitoring data, the design value must be based upon complete, quality-assured monitored air quality data from three consecutive years (2019-2021) at each eligible monitoring site and equal to or less than the 75 ppb standard.
The one-hour SO2 design value at the Martin Creek monitoring site located within the Rusk-Panola area shows a violation of the 1-hour primary SO2 NAAQS with a concentration greater than 75 ppb for the 2019-2021 design value, and thus, EPA is making the determination that the Rusk-Panola area did not attain by its January 12, 2022, attainment date. We also note that the 2020-2022 design value also shows a violation of the NAAQS.
Under CAA section 179(d)(2), if the EPA determines that an area did not attain the NAAQS by the applicable deadline, the responsible air agency has up to 12 months from the effective date of the determination to submit a revised SIP for the area demonstrating attainment and containing any additional measures that the EPA may reasonably prescribe that can be feasibly implemented in the area in light of technological achievability, costs, and any non-air quality and other air quality-related health and environmental impacts as required. According to CAA section 179(d)(3), this revised SIP is to achieve attainment of the one-hour SO2 NAAQS as expeditiously as practicable, but no later than 5 years from the effective date of the area's failure to attain ( i.e., 5 years after the EPA publishes a final action in the Federal Register determining that the nonattainment area failed to attain the SO2 NAAQS). In addition to triggering requirements for a new SIP submittal, a final determination that a nonattainment area failed to attain the NAAQS by the attainment date would trigger the implementation of contingency measures adopted under 172(c)(9).
III. Limited Approval/Limited Disapproval
Under CAA sections 110(k)(3) and 301(a) and EPA's long-standing guidance,[25] the EPA is proposing a Limited Approval/Limited Disapproval action. A limited approval is appropriate when a SIP contains provisions that are SIP strengthening, but also contains a non-severable deficiency that prevents EPA from granting a full approval of the SIP. EPA's limited approval action allows the EPA to codify SIP requirements, in this case, that would meet all requirements of the CAA but for the noted force majeure and timely compliance deficiencies. Under this limited approval, the area would make progress toward attaining the NAAQS, even if the SIP cannot be fully approved as meeting all applicable requirements for demonstrating NAAQS attainment by the attainment date. EPA's limited disapproval action will ensure that the deficient portions of the SIP submittal will be addressed, either through an EPA approved SIP or a FIP. This subsection will discuss the deficiencies identified in the SIP, the reasoning for and impact of a limited approval and limited disapproval, and EPA's plan to cure the deficiency.
Section 172(c)(6) of the CAA requires that nonattainment area SIP's “include enforceable emission limitations, and such other control measures means or techniques . . . . as well as schedules and timetables for compliance, as may be necessary or appropriate to provide for attainment of such standard in such area by the applicable attainment date . . .”. Further, CAA section 302(k) defines “emission limitation” to mean a requirement which limits the quantity, rate, or concentration of air pollutant emissions on a continuous basis.
For an SO2 attainment plan to be fully approvable, a modeled attainment demonstration must be based on the maximum allowable emissions permitted under the SIP's emission limitations and under 172(c)(6) those limitations must be practically and legally enforceable and under 302(k) must be continuous. The same is true for the demonstration of RACM/RACT, RFP, and contingency measures. Satisfying the enforceability criteria ensures that NAAQS attainment will be achieved via compliance with the SIP as adopted.
A. Force Majeure Provision Deficiency
The control strategy for the Rusk-Panola area is found in the February 14, 2022, Agreed Order between TCEQ and Luminant for the Martin Lake Facility incorporated into the SIP submittal. The Agreed Order establishes emissions limits and control requirements for the source which are necessary for the area to attain the NAAQS. However, the Agreed Order also includes a force majeure provision which states that, under a triggering event, the facility's failure to comply with an emissions limitation or other provision is not a violation of the Agreed Order.[26] This provision allows exceedances of emission limitations of unknown frequency, duration, and magnitude, and thus impermissibly interferes with the ability to continuously enforce the emissions limitations relied upon to provide for attainment. The provision is not contemplated in the attainment modeling which relies on the emissions limits being continuously and permanently applied, and, therefore, makes the modeling not representative of actual air quality in the area should this provision of the SIP be triggered. In all cases, the emission limits and control measures must be continuously applicable and accompanied by appropriate methods and conditions to determine compliance and must be quantifiable ( i.e., a specific amount of emission reduction can be ascribed to the measures), fully enforceable (specifying clear, unambiguous and measurable requirements for which compliance can be practicably determined), replicable (the procedures for determining compliance are sufficiently specific and non-subjective so that two independent entities applying the procedures would obtain the same result), and accountable (source specific limits must be permanent and must reflect the assumptions used in the SIP demonstrations).[27] As written, emissions associated with a force majeure event may increase with unknown frequency, duration, and magnitude notwithstanding the emission limitations because if the force majeure provision is triggered it is no longer a violation to emit above the limitations set in the SIP.
Emissions reductions under this SIP submission are quantifiable, except if the force majeure provision is triggered and the facility is exempted from complying with emissions limitations to an unknown extent. The emissions limitations outlined in the SIP are enforceable by the state, EPA, and citizens, except in the case of a force majeure event when lack of compliance with the emission limitations does not constitute a violation and therefore cannot be enforced. There are no clear, unambiguous, and measurable requirements for emissions that occur once the force majeure provision is triggered. This could result in excess emissions and periods of non-compliance which are “not a violation of [the] Agreed Order.” [28] These unaccounted emissions could jeopardize the ability of the area to attain and maintain the NAAQS while also cutting off the ability to enforce emissions limitations necessary to attain. The provisions of this SIP are replicable and are written sufficiently specific and non-subjective, except for the force majeure provision that does not provide specific procedures on how the provision should be interpreted, when compliance should be exempt, or for how long compliance should be exempt. Accountability is also an issue as the impact of triggering the force majeure provision may exempt compliance with SIP requirements and lead to unknowable, unaccounted for emissions associated with that event.
If the control strategy in the SIP fails to adhere to these principles, then the attainment demonstration relying on that control strategy, that contemplates no such force majeure event, may no longer be representative of the nonattainment area when the provision is triggered. The force majeure provision impacts the enforceability of the agreed order and thus, cannot be severed from the emissions limitations contained in the Agreed Order and consequently impacts the entirety of the SIP revision. The provision could interfere with the SIP revision's ability to provide for attainment and maintenance of the NAAQS, RFP, implementation of RACM/RACT, enforceable emission limitations as necessary to provide for timely attainment, and implementation of contingency measures. Therefore, the force majeure provision included in the SIP submission warrants a limited disapproval.
B. Compliance Date Deficiency
Second, EPA proposes a limited disapproval on the basis that the date for compliance with the emission limitations as written in the Agreed Order and SIP submittal is several months after the attainment date for the area. Under CAA section 172(c)(6), the nonattainment plan must include provisions as necessary or appropriate to provide for the attainment by the applicable attainment date. Here, the applicable attainment date for the Rusk-Panola Nonattainment area was January 12, 2022, but the Martin Lake Facility was not required to comply with all of the emissions limitations set forth in the SIP submission's control strategy until 180 days later, July 11, 2022. Therefore, because the compliance date for the full control strategy is not until after the attainment date for the area, EPA proposes a limited disapproval for this SIP submission.
C. Limited Approval
Despite these deficiencies, and as further elaborated on in section V with the full analysis of the attainment plan, EPA proposes that absent this force majeure provision and the compliance date deficiency, the SIP's attainment modeling, controls, emissions limitations, and other requirements would otherwise be adequate to provide the needed emission reductions to provide for attainment in the Rusk-Panola area. Currently, there are no federally enforceable requirements that will bring the Rusk/Panola area into attainment for the 2010 SO2 NAAQS. Therefore, despite the enforceability concerns, EPA is proposing a limited approval to make these new requirements federally enforceable as a SIP strengthening measure that will result in emissions reductions and provide for progress towards attainment of the 1-hour primary SO2 NAAQS.
The Limited Approval encompasses the entire submittal, both the SIP strengthening and deficient provisions, making all provisions federally enforceable. The major source in the area will have to additionally incorporate the control requirements and emissions limits prescribed in the SIP into their CAA title V operating permit, which will also be subject to federal enforcement.
D. Limited Disapproval and Consequences
On the basis of the deficiencies noted above, EPA is also proposing a Limited Disapproval, which carries the same consequences as a full disapproval. In accordance with CAA section 179, this Limited Disapproval triggers an 18-month NSR emissions offset sanction clock and a 24-month federal highway sanction clock. This action also establishes a requirement under CAA section 110(c) for the EPA to promulgate a FIP within two years. However, because of EPA's previous Finding of Failure to Submit, EPA is past due to issue a FIP revision. The sanctions are terminated when EPA fully approves a corrective SIP revision. The FIP clock obligation is addressed when EPA issues a FIP or fully approves the required SIP revision. Issuing a Limited Approval/Limited Disapproval ensures that the area is subject to federally enforceable requirements that will provide for progress toward attainment, while simultaneously providing for the correction of the deficient portion of the SIP submittal.
As stated previously in this section, the force majeure provision, exempts enforcement of the emissions limitations and controls during a specific type of event. To remedy this deficiency in the SIP, the EPA plans to promulgate a FIP that reflects the control strategy included in TCEQ's SIP submission but does not include the force majeure provision.
IV. Attainment Demonstration and Longer-Term Averaging
In accordance with CAA section 172(c)(1), nonattainment SIPs must include provisions that provide for attainment of the NAAQS. Please see section I. Background, C. Attainment Demonstration for SO2 Nonattainment Areas subsection for a more detailed discussion of the Attainment Demonstration requirements. An area can achieve attainment by implementing the appropriate control strategy identified to reduce pollution at the requisite sources. 40 CFR part 51, subpart G further delineates the control strategy requirements that SIPs must meet, and EPA has long required that all SIPs and control strategies reflect the four fundamental principles of quantification, enforceability, replicability, and accountability. See General Preamble, at 13567-68. Generally, for 1-hour standards control strategies include requirements that are based on 1-hour averaging times as this is the most straight forward way to ensure variability in the emission rate will not interfere with attainment of the standard. However, EPA does allow states to adopt requirements using longer-term averaging limits as long as they can demonstrate they are comparably stringent to modeled 1-hour critical emissions values (CEV) that would, if adopted as emission limits, provide for attainment of the one-hour standard.
Texas' plan applies a 24-hour block average emission limit to Martin Lake. Therefore, EPA is providing the following discussion of its rationale for approving the use of longer-term average limits in plans designed to provide for attainment. EPA's April 2014 SO2 Guidance recommends that the emission limits be expressed as short-term average limits ( e.g., addressing emissions averaged over one or three hours), but also allows for emission limits with longer averaging times, up to 30 days, if certain criteria are met. See April 2014 SO2 Guidance, pp. 22 to 39. The guidance recommends that, should states and sources utilize a longer-term average limit, the limit should be set at an adjusted level that reflects a comparable degree of stringency as the modeled 1-hour CEVs (lb/hr and lb/MMBtu limits) {Note MMBtu is million British Thermal Units}.
In evaluating this option, EPA considered the nature of the standard, conducted detailed analyses of the impact of the use of up to 30-day average limits on the prospects for attaining the standard, and carefully reviewed how best to achieve an appropriate balance among the various factors that warrant consideration in judging whether a state's plan provides for attainment. See April 2014 SO2 Guidance at appendices B, C and D.
As stated above and specified in 40 CFR 50.17(b), the 1-hour primary SO2 NAAQS is met at an ambient air quality monitoring site when the 3-year average of the annual 99th percentile of daily maximum 1-hour average concentrations is less than or equal to 75 ppb. In a year with 365 days of valid monitoring data, the 99th percentile would be the fourth highest daily maximum 1-hour value. For longer-term average limits, States must demonstrate with adequate assurance that a longer-term averaging limit will have comparable stringency as the one-hour average CEV and provide for attainment. Critical to this demonstration is the frequency and magnitude of hourly exceedances occurring under a longer-term average limit and the control level needed to constrain those occurrences to provide for attainment comparable to a strategy based on a one-hour emission standard. The following is a synopsis of EPA's review of whether such plans provide for attainment based on modeling of the one-hour CEV (1-hour CEV emission rates lb/hr and lb/MMBtu) and in light of the NAAQS form for determining attainment.
For plans relying on longer-term averaging limits, EPA's guidance recommends that States establish a CEV based off of a continuously applicable 1-hour emissions limit before determining their longer-term averaging period and limits using fixed emission rates. The maximum emission rate that would be modeled to result in attainment ( i.e., in an “average year” [29] shows three, not four days with maximum hourly levels exceeding 75 ppb, over three consecutive years) is labeled the “critical emission value.” The modeling process for identifying this critical emission value considers the numerous variables that affect ambient concentrations of SO2, such as meteorological data, background concentrations, and topography. In the standard approach, the state would then provide for attainment by setting a continuously applicable 1-hour emission limit at this critical emission value.
EPA recognizes that some sources have highly variable emissions, for example, due to variations in fuel sulfur content and operating rate, that can make it extremely difficult, even with a well-designed control strategy, to ensure in practice that emissions for any given hour do not exceed the critical emissions value. EPA also acknowledges the concern that longer-term emission limits can allow short periods with emissions above the critical emissions value, which, if coincident with meteorological conditions conducive to high SO2 concentrations, could in turn create the possibility of a NAAQS level exceedance occurring on a day when an exceedance would not have occurred if emissions were continuously controlled at the level corresponding to the critical emissions value. However, for several reasons, EPA believes that the approach recommended in its April 2014 SO2 Guidance document suitably addresses this concern. First, from a practical perspective, EPA expects the actual emission profile of a source subject to an appropriately set longer-term average limit to be like the emission profile of a source subject to an analogous 1-hour average limit. EPA expects this similarity because it has recommended that the longer-term average limit be set at a level that is comparably stringent to the otherwise applicable 1-hour limit (reflecting a downward adjustment from the critical emissions value) and that takes the source's emissions profile into account. As a result, EPA expects either form of emissions limit to yield comparable air quality.
Second, from a more theoretical perspective, EPA has compared the likely air quality from a source that has maximum allowable emissions under an appropriately set longer-term limit, to the likely air quality from a source that has maximum allowable emissions under the comparable 1-hour limit. In this comparison, in the 1-hour average limit scenario, the source is presumed at all times to emit at the critical emissions level. In the longer-term average limit scenario, the source is presumed occasionally to emit more than the critical emissions value but on average, and presumably at most times, to emit well below the critical emissions value. In an “average year,” compliance with the 1-hour limit is expected to result in three exceedance days ( i.e., three days with an hourly value above 75 ppb) and a fourth day with a maximum hourly value at 75 ppb. By comparison, for the source complying with a longer-term limit, it is possible that additional exceedances would occur that would not occur in the 1-hour limit scenario (if emissions exceed the critical emissions value at times when meteorology is conducive to poor air quality). However, this comparison must also factor in the likelihood that exceedances that would be expected in the 1-hour limit scenario would not occur in the longer-term limit scenario. This result arises because the longer-term limit requires lower emissions most of the time since the limit is set well below the critical emissions value, so a source complying with an appropriately set longer term limit is likely to have lower emissions at critical times than would be the case if the source were emitting as allowed with a 1-hour limit.
As a hypothetical example to illustrate these points, suppose a source that always emits 1,000 pounds of SO2 per hour and this results in air quality at the level of the NAAQS ( i.e., results in a design value of 75 ppb). Suppose further that in an “average year,” these emissions cause the 5 highest maximum daily average 1-hour concentrations to be 100 ppb, 90 ppb, 80 ppb, 75 ppb, and 70 ppb. Then suppose that the source becomes subject to a 30-day average emission limit of 700 pounds per hour. It is theoretically possible for a source meeting this limit to have emissions that occasionally exceed 1,000 pounds per hour, but with a typical emissions profile, emissions would much more commonly be between 600 and 800 pounds per hour. This simplified example assumes a zero-background concentration, which allows one to assume a linear relationship between emissions and air quality. A nonzero background concentration would make the mathematics more difficult but would give similar results. Air quality will depend on what emissions happen at what critical hours but suppose that emissions at the relevant times on these 5 days are 800 pounds per hour, 1,100 pounds per hour, 500 pounds per hour, 900 pounds per hour, and 1,200 pounds per hour, respectively. This is a conservative example because the average of these emissions, 900 pounds per hour, is well over the 30-day average emission limit. These emissions would result in daily maximum 1-hour concentrations of 80 ppb, 99 ppb, 40 ppb, 67.5 ppb, and 84 ppb. In this example, the fifth day would have an exceedance that would not otherwise have occurred, but the third day would not have an exceedance that otherwise would have occurred, and the fourth day would have been below, rather than at, 75 ppb. In this example, the fourth highest maximum daily concentration under the 30-day average would be 67.5 ppb.
This simplified example encapsulates the findings of a more complicated statistical analysis that EPA conducted using a range of scenarios using actual plant data. As described in appendix B of EPA's April 2014 Guidance, EPA found that the requirement for a lower long term average emission limit is highly likely to yield better air quality than is required with a comparably stringent 1-hour limit. Based on analyses described in appendix B of its 2014 Guidance, EPA expects that an emissions profile with maximum allowable emissions under an appropriately set, comparably stringent 30-day average limit is likely to produce the net effect of having a lower number of hourly exceedances of the NAAQS level and better air quality than an emission profile with maximum allowable emissions under a 1-hour emission limit at the critical emissions value.[30] This result provides a compelling policy rationale for allowing the use of a longer averaging period, in appropriate circumstances where the facts indicate this result can be expected to occur.
The question then becomes whether this approach—which is likely to produce a lower number of overall hourly NAAQS level exceedances even though it may produce some unexpected exceedances above the critical emission value—meets the requirement in section 110(a)(1) and 172(c)(1) for state implementation plans to “provide for attainment” of the NAAQS. For SO2, a variety of factors can cause a well-designed attainment plan to fail and unexpectedly not result in attainment. For example, this can occur if meteorology occurs that is more conducive to poor air quality than was anticipated in the plan. Therefore, the plan must provide an adequate level of confidence that it will provide for attainment of the NAAQS. Additionally, when evaluating longer-term average limits, EPA must weigh the likely net effect on air quality. This evaluation must consider the risk that occasions with meteorology conducive to high concentrations will have elevated emissions leading to NAAQS level exceedances that would not otherwise have occurred and must also weigh the likelihood that the requirement for lower emissions on average will result in days not having hourly exceedances that would have been expected with emissions at the critical emissions value. Additional policy considerations, including the desirability of accommodating real world emissions variability without significant risk of NAAQS violations, are also appropriate factors for EPA to weigh in judging whether a plan provides for attainment with a reasonable degree of confidence. Based on these considerations, especially given the high likelihood that a continuously enforceable limit averaged over as long as 30 days, determined in accordance with EPA's guidance, will result in attainment, EPA believes as a general matter that such limits, if appropriately determined, can reasonably be considered to provide for attainment of the 2010 SO2 NAAQS.
EPA's April 2014 SO2 Guidance appendix B prescribes how a state's SIP should develop and demonstrate that an appropriate longer-term average limit provides for attainment. Development of longer-term average limits starts with a determination of the 1-hour emission limit that would provide for attainment ( i.e., the critical emissions value), establishment of an adjustment factor to determine the (lower) level of the longer-term average emission limit that would be estimated to have a stringency comparable to the otherwise necessary 1-hour emission limit, and application of the adjustment factor to the emissions limits. The method for deriving an appropriate adjustment factor uses a database of continuous emission data reflecting the type of control that the source will be using to comply with the SIP emission limits, which (if compliance requires new controls) may require use of an emission database from another source. The recommended method involves using this data to compute a complete set of emission averages, computed according to the averaging time and averaging procedures of the prospective emissions limit. In this recommended method, the ratio of the 99th percentile among these long-term averages to the 99th percentile of the 1-hour values represents an adjustment factor that may be multiplied by the candidate 1-hour emission limit to determine a longer-term average emission limit that may be considered comparably stringent.[31] The guidance also addresses a variety of related topics, such as the potential utility of setting supplemental emission limits, such as mass-based limits, to reduce the likelihood and/or magnitude of elevated emission levels that might occur under the longer-term emission rate limit.
V. Review of Modeled Attainment Plan
This section discusses EPA's review and analysis of the modeled attainment plan, including model selection, meteorological data, emissions data, receptor grid, emissions limits, and background concentrations. As discussed in detail in prior section III., EPA is proposing that a limited approval and limited disapproval action is necessary because, but for the presence of the force majeure provision in the SIP submission, the modeled attainment plan satisfies the EPA's CAA requirements and would adequately demonstrate the SIP requirements will provide for attainment absent the force majeure provision.
TCEQ's SIP submittal relies on an alternative model, which modifies the version of AERMOD approved for regulatory modeling. EPA reviewed the TCEQ's alternative model and also performed its own additional modeling analysis to determine whether the emission limits and control measures in the State's SIP, absent the force majeure provision, would provide for attainment in the Rusk-Panola area. The EPA's additional modeling analysis used the Alternative Model AERMOD v.21112 with Highly Buoyant Plume (TCEQ's alternative model AERMOD-HBP) and reflecting what was done in the TCEQ modeling. EPA modeled two of the highest modeled concentration scenarios utilizing the AERMOD-HBP v.21112 with HBP. Those scenarios produced a modeled maximum design value, with background concentration included, of 73.6 ppb of SO2, confirming TCEQ's modeling results for these two scenarios. EPA also ran these same two scenarios with the most recent version of AERMOD v.23132 with HBP code. EPA has included the HBP code for scientific testing and investigation as an alpha option in AERMOD v.23132, but it is important to note that alpha options are for scientific investigation and not approved for regulatory use. EPA compared the HBP code in AERMOD v.23132 with TCEQ's AERMOD-HBP, and it appears to be the same exact code. This modeling was performed by EPA to confirm that any other changes in AERMOD between v.21112 and v.23132 would not result in significant changes to TCEQ's attainment demonstration modeling, and the maximum modeled results for these two scenarios were the same. This modeling also confirms that the HBP code included in TCEQ's AERMOD-HBP and the alpha option implemented in AERMOD v.23132 resulted in the same maximum modeled concentrations. Additional, more detailed discussion of the State's modeling and EPA's modeling of these two highest concentration scenarios with both versions of AERMOD (TCEQ's v.21112 with HBP and EPA's v.23132 with non-regulatory alpha option HBP code) are contained in the Technical Support Document (TSD) for this proposed action. For EPA's alternative model review of the TCEQ's AERMOD-HBP see the AERMOD-HBP TSD.
A. Model Selection
According to the Modeling Guideline, alternative models may only be used instead of AERMOD with EPA review and approval. Texas' attainment demonstration used an alternative model instead of the Modeling Guideline preferred model, AERMOD v.21112.[32] TCEQ's alternative model modifies AERMOD's treatment of penetrated plumes which affects the resultant modeled concentrations by delaying mix down of the penetrated plume component under certain circumstances resulting in less emissions mixing down to add to surface level concentrations. The modified code added is referred to as the Highly Buoyant Plume Model Code [33] and the resultant alternative model is called AERMOD-Highly Buoyant Plume (AERMOD-HBP). Along with the AERMOD-HBP, Texas used the regulatory versions of AERMOD preprocessors (AERMET, AERMINUTE, AERSURFACE, AERMAP, and Building Profile Input Program for PRIME (BPIPPRM)), and where applicable, used the preprocessor's regulatory default parameters.
Texas requested use of an alternative model in a letter dated May 24, 2021. Vistra's consultant, AECOM, performed initial dispersion modeling for the Rusk-Panola area using AERMOD v. 19191. AECOM asserted its initial modeling showed that AERMOD v. 19191 can overpredict SO2 concentrations relative to available SO2 observations at the Longview and Martin Creek monitors some of the time.[34] Based on AECOM's initial modeling, TCEQ and Vistra approached EPA Region 6 in Fall 2020, that led to TCEQ requesting and having discussions with EPA, TCEQ's contractor, Vistra, and Vistra's contractor AECOM to develop an understanding of what information and analysis were needed to support a potential alternative model request with modified treatment of penetrated plumes. TCEQ formally requested EPA's review and approval of an alternative model in a letter dated May 24, 2021.[35] TCEQ's request did not include all of the necessary components previously discussed between October 2020 through April 2021 for the EPA to complete an alternative model review. The EPA continued to receive materials from TCEQ through August 2021, and EPA provided some feedback and clarification on some technical analyses that were needed for EPA to conduct its review of the alternative model request. EPA did not receive all necessary components until August 2021. Those components included the necessary information and modeling analysis to enable EPA to perform a full review in accordance with the alternative model review and approval guidelines. Several of these technical analyses are based on the available data including: AERMOD v.21112 regulatory version and AERMOD-HBP modeling results comparisons using actual emissions and meteorological data for the 2016-2020 period coupled with available SO2 monitoring data at the Martin Creek and Longview monitors.[36]
EPA Region 6 performed a detailed analysis of the alternative model request materials, including evaluating the theoretical rationale for modifying how penetrated plumes are treated in AERMOD. TCEQ and EPA's analysis centered on evaluation of the modeling results centered at the two monitors in the area, the Martin Creek monitor and Longview monitor, in order to compare modeled concentrations to monitored values. The evaluation and technical analysis of regulatory AERMOD results and the alternative model AERMOD-HBP results at receptors placed at the two monitor locations were compared with the monitored data for these two monitors using actual emissions, meteorology, and monitored concentrations in the 2016-2020 timeframe for the Longview monitor and 2018-2020 for the Martin Creek monitor. In addition to TCEQ's submittal materials, EPA also performed several technical analyses, including its own Cox-Tikvart statistical analysis to confirm the conclusions. Overall, EPA's review of TCEQ's submittal materials and EPA's own analyses confirmed TCEQ's conclusion that based on the data available, AERMOD-HBP performed better than AERMOD in this one case-specific and location-specific situation in the area around the Martin Lake Electric Generating Facility (EGF) facility. Full details of EPA's review and conclusions related to the alternative model approval are provided in the “EPA's Review of TCEQ's Alternative Model Request of AERMOD with Highly Buoyant Plume Treatment (HBP)” (AERMOD-HBP TSD), included in the docket for this action.
Once EPA Region 6 modelers and Office of Air Quality Planning and Standards' Air Quality Modeling Group (OAQPS-AQMG) modelers were in agreement that the Alternative Model could be approved in accordance with the Modeling Guideline section 3.2.2, EPA Region 6 sent a memorandum with the detailed AERMOD-HBP TSD attached (dated July 11, 2024) to the Model Clearinghouse (part of OAQPS-AQMG) that proposed approval of the Alternative Model and requested concurrence from the Model Clearinghouse. The Model Clearinghouse sent a memorandum (dated July 24, 2024) to EPA Region 6 concurring with the case specific approval of the Alternative Model in this case-specific situation. The AERMOD-HBP approval is limited specifically to only allow AERMOD-HBP to be used in the attainment demonstration modeling for this Rusk-Panola attainment demonstration. This approval limited to the specific attainment demonstration at Martin Lake is based on the location and situation-specific factors, including available monitoring data, that were considered when evaluating this alternative model for this specific case. The EPA notes that attempting to use this alternative model for any other purposes at the Martin Lake facility, or any other facility, would require a separate, stand-alone evaluation and approval in accordance with EPA's alternative model requirements. Please see the TSD, the Model Clearinghouse Memorandums, and EPA's AERMOD-HBP TSD for more details.
B. Meteorological Data
The State's modeling utilized surface meteorological data obtained from the Longview East Texas Regional Airport and upper air data from the Shreveport, Louisiana station from 2015-2019. The Longview East Texas Regional Airport is the closest National Weather Service (NWS) site to the Martin Lake facility and monitor at Tatum County Road, 2181d Martin Creek Lake site (Air Quality System (AQS) 484011082), approximately 19 kilometers (km) away and is representative of the meteorology in the Rusk-Panola area due to its proximity. The Shreveport, Louisiana NWS upper air site is the closest site of upper air data. TCEQ processed the surface and upper air data using the meteorological processing tools: AERMINUTE (v.15272), AERMET (v.21112), and AERSURFACE (v.20060). AERMINUTE was used to include measured one-minute wind averages, AERMET was used to generate meteorological data files, and AERSURFACE was used to determine the surface characteristics for the meteorological station. The current version of each preprocessor at the time the modeling demonstration was performed was AERMINUTE v.15272, AERMET v.21112, and AERSURFACEv. 20060. While the most recent versions of AERMINUTE and AERSURFACE were used, AERMET has been updated since the State conducted its modeling. However, based on the changes that the EPA made to AERMET, we would not expect to see any significant changes to modeling results if the data were processed with the latest version of AERMET, and EPA finds that TCEQ's data are still representative. EPA recommends using the closest NWS sites for surface and upper air data if they are considered representative of the area being modeled. In this situation, EPA concurs with the use of these two sites for this modeling as meeting EPA's criteria as being nearby and representative. EPA also finds that TCEQ adequately processed the data in accordance with the Modeling Guideline and EPA's Guidance to generate the necessary modeling data to be used in the AERMOD model runs. Therefore, the EPA is proposing to find the selection and processing of this data to be acceptable.
C. Emissions Data
The SIP revision identifies the Martin Lake facility as the primary SO2 source in the Rusk-Panola area. As there were no monitors in the area at the time of designation, EPA relied on modeling to designate the area and found that Martin Lake is likely producing almost all, if not 100%, of the emissions causing the maximum modeled design values that were above the NAAQS.[37] This information is also confirmed by review the of SO2 sources in the Rusk-Panola area provided in the SIP revision's emission inventory analysis. The emissions inventory analysis shows that there are no other major sources of SO2 within the boundary of the Rusk-Panola area. TCEQ appropriately modeled the maximum hourly emission rate for the Martin Lake facility emission sources. The only nearby SO2 source with emissions greater than 100 tons per year within 50 km of Martin Lake is the American Electric Power Pirkey Power Plant (Pirkey) located approximately 17 km outside of the Rusk-Panola area boundary in Harrison County. Since Pirkey is a background source outside of the Rusk-Panola NAA that could potentially contribute to concentrations in the Rusk-Panola NAA, pursuant to the Modeling Guideline table 8-1, a hybrid of actual and allowable emission factors was used for emissions from Pirkey in TCEQ's modeling for the Rusk-Panola area. The remainder of the sources are captured by using monitoring data that is representative of background concentrations. The inclusion of Pirkey assures that Texas incorporated all sources in the modeling that are considered to possibly create SO2 concentrations and/or concentration gradients anywhere in the Rusk-Panola NAA that are not represented by the background monitoring data.
The other facility, Pirkey, that is located outside of the Rusk-Panola area included in the modeling, is not located in a direction such that it can contribute to the maximum SO2 concentrations in the Rusk-Panola area (not upwind), and thus, would have a negligible impact on maximum modeled concentrations within the Rusk-Panola area. Therefore, TCEQ did not require new SO2 emission limits on Pirkey. EPA has reviewed the facility's data and notes that the Pirkey facility is 17 km away from Martin Lake and the nearby Martin Creek monitor near Tatum County Road, and thus adding emission limits to Pirkey are not critical to demonstrating attainment in the area. EPA concurs with TCEQ's approach of including Martin Lake (allowable emissions modeled) and Pirkey (hybrid of actual and allowable emission related factors to generate emissions modeled) emission sources, which comports with EPA's Modeling Guideline, including table 8-1, that provides guidance on what sources to include in the modeling and whether to model actuals or allowable emissions.
TCEQ used site specific building and stack data and modeled all stacks in Martin Lake at the lesser of actual stack height or Good Engineering Practice (GEP) stack height. The State's modeling included building downwash influences for all sources except for Pirkey because the effects of downwash from Pirkey are localized and would not affect modeled concentrations in the Rusk-Panola area. The EPA has determined that the SIP's selection of sources and preprocessing of that source data satisfies the requirements of the Modeling Guideline. For a more detailed analysis and conclusions on what sources were included in the modeling and how they were modeled, see the TSD.
As discussed in the TSD, Martin Lake was identified as the primary contributor to NAAQS violations in the Rusk-Panola area. Martin Lake is an EGF, with four point sources and one fugitive area source for SO2; the four point sources consist of three EGF boiler unit stacks and one combined stack for two auxiliary boilers. Modeling indicated emission reductions for Martin Lake were necessary to provide for attainment, and based on the modeling emission limitations were developed and included in the Agreed Order between TCEQ and Luminant that was submitted as part of this SIP revision. The modeling covers 42 operating scenarios in total. These different scenarios were developed based on a combination of an emissions limit specific to each EGF boiler, an overall emission cap on emissions from the three main EGF boiler stacks, and four different operating loads. There are four different operating loads for the EGF boilers: (1) high load, (2) medium load, (3) low load, and (4) a maintenance, startup and shutdown (MSS) load. TCEQ modeled control measures using the critical emission value of 8,208 lbs/hour as the one-hour averaging period emission cap for the three EGF boilers that correlates with the Agreed Order's longer-term averaging emissions limit of 7,469 lb/hr over a 24-hour block averaging period. TCEQ's modeling also modeled the critical emission value of 0.33 lb/MMBtu limit for each boiler that correlates to the agreed order limit of 0.32 lb/MMBtu limit on a 24-hour block averaging period for each boiler. TCEQ developed the 42 different emission scenarios based on these limits as discussed in the TSD. For more detailed discussion and evaluation of the Agreed Order 24-hour block averaging limits and the critical emission value comparably stringent 1-hour averaging limitation values used in the modeling, see section V.E. In conclusion, the EPA is proposing to find that Texas' choice of included sources and the scenarios modeled to cover the potential range of operating scenarios that could occur with the new limits in place to be appropriate and inclusive of worst-case scenarios, in the absence of the force majeure provision.
D. Receptor Grid
Within AERMOD, air quality concentration results are calculated at discrete locations identified by the user; these locations are called receptors. Receptors are placed in areas and outside the plant boundaries and areas within the plant boundary where the public has access. Areas within the plant where public access is restricted, are not considered ambient air for the purposes of compliance with NAAQS. TCEQ's modeling domain for this demonstration consisted of a 25.5 km by 24.5 km rectangular area centered around Martin Lake with three nested receptor grids. TCEQ placed receptors within and outside the Rusk-Panola nonattainment area: (1) receptors at 25 meter (m) spacing along the non-ambient air fence/boundary lines, (2) the innermost grid spanning 0 to 3 km from the center point, encompassing Martin Lake, with 50m spacing between receptors; (3) the middle-nested grid extended from 3 km to 9 km, with 100 m spacing between receptors; and (4) the outermost grid, which extends beyond the nonattainment boundary covers the rest of the modeled domain, had 500 m spacing. The TCEQ, after discussions with EPA and Vistra, removed receptors from the grid found within the property owned and controlled (public access is restricted) by Vistra,[38] Vistra restricts public access to this area through fencing, posting, and patrolling. Again, the air in the area controlled by Vistra is considered non-ambient air relative to its own emissions, and thus, they are not required to place receptors within these boundaries. Receptors with 25m spacing were also added along a section of public road within Vistra's property, and an additional receptor was placed at the location of the Martin Creek monitor. TCEQ determined receptor elevations using AERMAP in its modeling. EPA proposes that the receptor grid is consistent with EPA's Modeling Guideline and is adequate for demonstrating attainment within the NAA and the immediately surrounding area in this attainment demonstration modeling.
E. Emission Limits
As part of its control strategy for the Rusk-Panola area, Texas entered an Agreed Order with Luminant set emissions limitations for the Martin Lake facility, adopted on February 14 2022, pursuant to §§ 382.011, 382.012, 382.023, and 382.024 of the Texas Clean Air Act, Texas Health & Safety Code, Chapter 382, and the CAA. TCEQ incorporated the Agreed Order as part of its SIP revision submittal as a source-specific SIP revision seeking to establish federally enforceable emission limits. The limits in table 2 are hourly limits, and compliance with the limits is determined using the longer-term 24-hour block averaging period.
As stated in subsection C of this notice, there are no other major sources of SO2 within the nonattainment area that could contribute to nonattainment in the Rusk-Panola area. The Agreed Order set the compliance date for emission limits as “the date by which the State of Texas is required to demonstrate compliance with the 2010 1-hour SO2 NAAQS for the Rusk-Panola SO2 Nonattainment Area.” This means that the compliance date for the Agreed Order would be the attainment date, January 12, 2022 for limits other than the lb/MMBtu limits, which require compliance 180 days later (July 11, 2022).[39] EPA proposes to find that the source specific emissions limits as laid out in the Agreed Order submitted with this SIP revision would be sufficient, based on the above described modeling and recognizing the longer than 1 hour averaging period, to provide for attainment in the Rusk-Panola area absent the force majeure provision.
Table 2—Martin Lake Agreed Order Emission Limitations
Source ID Source description Sulfur Dioxide (SO 2 ) limit S-1, S-2, S-3 EGF Boiler Units II.a.(i) (1) Burn only subbituminous coal, No. 2 fuel oil, or natural gas; (2) limit the firing rate (when fired) for all three EGF boiler units to a combined rate not to exceed 27,000 million British thermal units (MMBtu) per hour (the firing rate is an operating cap for all three EGF boiler units combined); and (3) Optimize the FGD systems to ensure compliance with a combined SO 2 emission rate not to exceed 7,469 lb/hr on a block 24-hour average basis for the three EGF boiler units (the 7,469 lb/hr is an emission cap for all three EGF boiler units combined). The emission cap of 7,469 lb/hr applies at all times when fuel of any type is fired in any EGF boiler unit. (ii) . . . the Company shall ensure compliance with an SO 2 emission rate not to exceed 0.32 lb/MMBtu on a block 24-hour average basis for each EGF boiler unit. This emission rate applies at all times when fuel of any type is fired in any EGF boiler unit. S-1A and B Auxiliary Boilers II.b. (i) Fire only No. 2 fuel oil with a sulfur content of 0.10% by weight or less; (ii) Not exceed an SO 2 emission rate of 51.46 lbs/hr on a one-hour basis and 22.54 tpy on an annual basis, combined for the two Auxiliary Boilers (the 51.46 lbs/hr and 22.54 tpy are emission caps for the two auxiliary boilers combined); and (iii) Comply with a 10 percent annual capacity factor for each of the two Auxiliary Boilers. Annual capacity factor is the ratio between the actual heat input from all fuels burned during a calendar year and the potential heat input had the boiler been operated for 8,760 hours during a year at the maximum steady state design heat input capacity. The 10 percent annual capacity factor limit corresponds to a heat input of 219,000 MMBtu per calendar year, per Auxiliary Boiler. (iv) The Company shall monitor the sulfur content of the liquid fuel in accordance with fuel sampling requirements specified in 40 CFR part 75, appendix D, 2.2 Oil Sampling and Analysis. 1. Enforceability
An attainment plan must include emission limits that provide for attainment and that are: quantifiable, fully enforceable, replicable, and accountable.[40] Full enforceability includes the ability to enforce emissions limitations by the state, the EPA, or by private citizens through a citizen suit.[41] As discussed in detail in section III. of this notice, EPA proposes to find that the force majeure provision included in the SIP submission interferes with enforceability such that the Agreed Order and attainment plan may not be fully approved as meeting the requirements of CAA sections 110, 172, 191 and 192.
2. Longer-Term Average Emission Limits
As noted in section IV. and section V.C, the Texas SIP is using longer-term lb/hr and lb/MMBtu emission limits with 24-hour block averaging compliance limits for three EGF boilers at the Martin Lake facility. Therefore, the critical emissions values are the modeled emission rates/limits based on 1-hour averaging period and serve as the basis for developing emission rate limits for longer averaging period and the limits used in the attainment modeling for the area. Modeled emission rates of 8,208 lb/hr and 0.33 lbs/MMBtu (one hour averaging period) were calculated by Vistra to convert to 7,469 lbs/hour and 0.32 lbs/MMBtu averaged over a 24-hour block period. Martin Lake will be switching fuels from lignite and lignite blend operations to subbituminous coal, and therefore, Vistra and Texas evaluated other similar subbituminous coal fired units elsewhere in Texas with similar SO2 control devices that would be more representative of operating conditions rather than comparing to the past performance of the Martin Lake units operating on lignite coal. As such, Vistra and Texas determined that the NRG Limestone units would be appropriate sources to derive an adjustment factor to apply to the EGF boilers at Martin Lake due to the use of subbituminous coal at the NRG Limestone units and both facilities have wet FGD controls. In the SIP, Texas utilized three years of NRG Limestone (located near Jewett, Texas) emissions data from October 2018 through September 2021 to conduct the variability analysis, which coincides with when NRG Limestone burned only subbituminous coal. Texas employed the method detailed in our 2014 SO2 Guidance appendices B, C, and D for deriving an appropriate adjustment factor to adjust the lb/hr modeled emission rates and the lb/MMBtu emission limit, to result in equivalent 24-hour block averaging lb/hr and lb/MMBtu emission limits. Texas followed EPA's guidance and evaluated the historic 1-hour 99th percentile of SO2 emissions information (lb/hr and lb/MMBtu) against the 99th percentile 24-hour block average for the lb/hr and lb/MMBtu data to derive the appropriate adjustment factors of 0.91 for lb/hr and 0.97 lb/MMBtu. These factors were used by Texas to derive the emission limits of 7,469 lbs/hour and 0.32 lbs/MMBtu averaged over a 24-hour block period that were included in the Agreed Order for Martin Lake. EPA has reviewed TCEQ's information supporting the 24-hour block averaging limits and is proposing to find the analysis acceptable and that these represent comparably stringent limits that would, absent the force majeure clause be protective of the NAAQS. For a more detailed discussion of our analysis please see the TSD prepared for this action.
F. Background Concentrations
To develop background concentrations for the NAA, Texas relied on 2015-2019 SO2 data from the Midlothian OFW monitor in Ellis County (CAMS C52), approximately 220 km west of the NAA.[42] Texas determined that there were no representative nearby monitors to capture background concentrations, as the nearby monitors were all SO2 Data Requirements Rule (DRR) monitors sited to capture the impacts of major SO2 sources or other monitors with data significantly impacted by large SO2 sources ( e.g., Longview monitor in Gregg County is 19 km from Martin Lake but Martin Lake's emissions have historically had a large impact on this monitor's data). Therefore, Texas used a monitor located away from the source but still considered to be representative of background concentrations in the area. The Midlothian OFW monitor in Ellis County, Texas was chosen as it had complete SO2 Design Values (DVs) for the 2015 through 2019 period and had a more stable DV across recent years.
A fixed background concentration of 15.72 µg/m3 was added to modeled concentrations to result in maximum modeled concentrations for all 42 scenarios. These background values are representative of the contribution due to other sources within the Rusk-Panola area and surrounding areas that were not explicitly modeled combined with regional continental background in this area. See the TSD for additional information. Using this approach, the EPA is proposing to find the State's treatment of SO2 background levels to be acceptable for adding to modeled concentrations to represent background SO2 levels in this attainment demonstration modeling.
G. Summary of Results
The State's alternative modeling demonstration, which incorporates emissions scenarios based on the February 14, 2022, Agreed Order emission limits for the Martin Lake facility but recognizing the longer than 1 hour averaging time, resulted in modeled concentrations below the 1-hour primary SO2 NAAQS using the alternative model AERMOD-HBP. As noted, EPA Region 6 proposed approval and obtained concurrence from the Model Clearinghouse for the use of the alternative model for this specific application. TCEQ modeled 42 different scenarios representing the range of operations, emissions, and dispersion that could occur, incorporating the Agreed Order's required emission limits. These 42 modeled scenarios had maximum ambient air modeled DVs ranging from 40 ppb to 73.6 ppb (104.8 µg/m[3] to 192.8 µg/m[3] ) that all demonstrated attainment of the 1-hour SO2 NAAQS, without accounting for the possible excused non-compliant emissions periods under the force majeure provision.[43]
As part of EPA's modeling review, EPA modeled the two scenarios that resulted in the highest maximum DV of 73.6 ppb using the AERMOD-HBP alternative model (based on AERMOD v.21112), and the results duplicated TCEQ's results with the same maximum modeled design values. The only differences between EPA's model runs and TCEQ's model runs for these two scenarios were that EPA modeled all of the receptors in one AERMOD run using a single CPU processor, and EPA relied on AERMOD to generate the maximum DVs with inclusion of the background concentrations instead of all of the post-processing steps that TCEQ performed. TCEQ modified a prior version of AERMOD that was the current version at the time TCEQ developed their SIP, version v.21112. Since TCEQ submitted their SIP, there have been two updates to AERMOD, and the most recent version is AERMOD v.23132. EPA ran these same two scenarios using the current version of AERMOD, v.23132, with the non-regulatory alpha option HBP code (same as TCEQ's HBP code). The EPA has included the HBP code in AERMOD as an alpha option that is only for scientific testing and investigation, and the HBP code is not approved for any regulatory modeling. EPA did these model runs to see if the current version of AERMOD with the alternative HBP model code (not approved for use in regulatory modeling) would still result in the same maximum DV values obtained using the AERMOD HBP Alternative Model (AERMOD v.21112 with HBP code). This test is to verify that the code changes in the regulatory version of the model (non-HBP code) did not result in any model concentration changes. The more recent version of AERMOD with the non-regulatory HBP code included resulted in the same modeled results as TCEQ's AERMOD v.21112 with HBP code. This test confirms that the updates in the regulatory version of AERMOD between v.21112 and v.23132 do not result in any differences in the maximum design value when both AERMOD versions were run with the non-regulatory alternative model code HBP.
With the exception of the HBP code that is an alternative model, EPA's review of the rest of TCEQ's modeling components indicated that TCEQ used the regulatory AERMOD preprocessors (AERMET, AERMINUTE, AERSURFACE, AERMAP, and Building Profile Input Program for PRIME (BPIPPRM) for building/structure downwash). TCEQ's submitted modeling, where applicable, used the regulatory default parameters, and the options and settings for AERMOD and the processors used are acceptable. TCEQ broke up the receptor grid into multiple runs and post-processed the results outside of AERMOD, which complicated review. The EPA reviewed model input and output files for all 42 modeling runs and modeled two of the scenarios that had the highest design value and confirmed that TCEQ's approach resulted in the same maximum modeled concentration results. For more detailed information, explanation, and analysis of TCEQ's modeling please see the following documents included in this docket: TSD, AERMOD-HBP TSD, Model Clearing House memoranda.
After reviewing Texas' attainment demonstration and conducting additional modeling runs, the EPA agrees that Texas' submittal and supplemental materials, along with the Agreed Order (February 14, 2022) limits, constitute an attainment plan that would strengthen the SIP and sufficiently reduce emissions to meet the NAAQS in the Rusk-Panola area, but for the force majeure provision. However, due to the untimely compliance date for aspects of the attainment plan and the impacts of the force majeure provision of the Agreed Order on all of the Agreed Order's emission limits for the Martin Lake facility, Texas' attainment plan is insufficient to fully provide for attainment in the Rusk-Panola area or fully meet the requirements of CAA sections 110, 172, 191 and 192. EPA therefore proposes to issue a limited approval and limited disapproval for this attainment plan SIP revision. See section III. Limited Approval/Limited Disapproval for additional discussion. We therefore propose to determine that, absent the force majeure provision, Texas' plan would provide for attaining air quality under the 2010 1-hour primary SO2 NAAQS in the Rusk-Panola nonattainment area.
VI. Review of Other Plan Requirements
Section VI includes EPA's review of the following SIP elements: Emissions inventory, RACM and RACT, NSR, RFP, contingency measures, and conformity. EPA proposes that the SIP adequately satisfies the requirements for a baseline emissions inventory and nonattainment NSR, but due to the presence of the force majeure provision affecting the enforceability of the limits relied upon in the attainment demonstration, cannot otherwise meet the requirements of CAA sections 110, 172, 191 and 192, particularly for RACM/RACT, RFP, emissions limits necessary to provide for attainment, and contingency measures.
A. Emissions Inventory
The emissions inventory and source emission rate data for an area serve as the foundation for air quality modeling and other analyses that enable states to: (1) estimate the degree to which different sources within a nonattainment area contribute to violations within the affected area; and (2) assess the expected improvement in air quality within the nonattainment area due to the adoption and implementation of control measures. A nonattainment SIP must include a comprehensive, accurate, and current inventory of actual emissions from all sources of SO2 in the nonattainment area as well as any sources located outside the nonattainment area which may affect attainment in the area. See CAA section 172(c)(3). In its submittal, Texas included a current emissions inventory for the Rusk-Panola area covering the 2017-2022 period, which can be found below at table 3.
The State of Texas compiles a statewide emissions inventory for stationary sources in accordance with Texas regulations at 30 Texas Administrative Code (TAC) § 101.10, CAA requirements, and EPA guidance. The submitted data is then reviewed for quality assurance purposes and stored in the State of Texas Air Reporting System (STARS) database. In its submittal, Texas confirmed that stationary point sources ( i.e., Martin Lake) comprised over 99% of the SO2 emissions in the Rusk-Panola area. Texas determined the forecasted 2022 emissions for Martin Lake through historical point source heat input and a future year emissions limit that accounts for enforceable emissions reductions as required in the Agreed Order. According to the 2014 SO2 Guidance, nearby sources outside the NAA (Pirkey) should also be included in the emissions inventory.
TCEQ chose the year 2017 as the base year for its analyses as the most complete and representative record of annual SO2 emissions because (1) it was the most recent periodic inventory year available and (2) it was also the year that the EPA designated the Rusk-Panola area as nonattainment for the 2010 SO2 NAAQS.
The 2017 baseline area source emissions inventories were developed in accordance with the requirements of the Air Emissions Reporting Requirements (AERR) rule and developed using EPA-generated EIs, TCEQ-contracted projects, TCEQ staff projects, and 2014 EIs by applying growth factors derived from different sources.[44] TCEQ also developed non-road and on-road mobile source emissions inventories by using EPA's mobile source emissions models, Texas-specific utility of the EPA mobile source models, and EPA-approved methods and guidance.
A summary of the State's submitted emissions inventory is provided in the following table:
Table 3-1—Rusk-Panola Nonattainment Area SO 2 Emissions in Tons per Year (TPY)
Source category 2017 Base year reported emissions 2018 Reported emissions 2019 Reported emissions 2022 Attainment year emissions Agreed order federally enforceable maximum emissions Point—Martin Lake 36,441.46 56,198.55 46,549.50 22,269.31 32,736.76 Non-point 0.31 N/A N/A 0.43 N/A On-road Mobile 0.14 N/A N/A 0.14 N/A Non-road Mobile 0.02 N/A N/A 0.02 N/A Total 36,441.93 56,198.55 46,549.50 22,269.90 32,736.76 Table 3-2—AEP Pirkey Power Plant SO 2 EI Annual and Permitted Emissions in TPY 45
Source 2017 Reported emissions 2017 Reported emissions 2017 Reported emissions 2022 Attainment year emissions Permitted emissions Point—AEP Pirkey 3,959.80 5,084.80 3,073.00 4,039.20 35,820.00 The EPA agrees that the State's emissions inventories for point, nonpoint, and mobile sources are appropriate because they have been accumulated and reported in accordance with established methods and criteria. EPA proposes that the base year emissions inventory is representative and satisfies the EI requirement, however, EPA cannot fully approve the future year emission inventory due to enforceability concerns arising from the force majeure provision included in the Agreed Order.
B. Reasonably Available Control Measures and Reasonably Available Control Technology (RACM/RACT)
Section 172(c)(1) of the CAA requires states to adopt and submit all RACM, including RACT, as needed to attain the standards as expeditiously as practicable. Section 172(c)(6) requires the SIP to contain enforceable emission limits and control measures necessary to provide for timely attainment of the standard. The plan relies on ambient SO2 concentration reductions achieved by implementation of the Agreed Order's control requirements and emissions limits at Martin Lake. Martin Lake plans to implement SO2 emission limits (lb/hr and lb/MMBtu) for the three EGF boilers.
The control strategy at Martin Lake incorporates pre-combustion and post- combustion controls for the three EGF boilers and sets SO2 emission limits for the two auxiliary boilers. The EGF boilers will be limited to burning subbituminous coal, No. 2 fuel oil, and natural gas during operations and additionally have a combined cap on their firing rate (MMBtu/hr). Martin Lake intends for the subbituminous coal to be the primary fuel burned, which is lower in sulfur content compared to the lignite and lignite-blended mix of coals historically used by the facility. TCEQ additionally provides that the existing SO2 wet limestone scrubber system for the EGF boilers will be optimized to increase efficiency to meet the limits in the Agreed Order.
The final emission limitations as included in the February 14, 2022, Agreed Order are provided earlier in this document in section V.E., Emission Limitations. Texas has provided modeling which demonstrates that these measures for Martin Lake provide for timely attainment and meet the RACM and RACT requirements, without accounting for excused emissions not in compliance with the limits during force majeure periods. The EPA proposes that, but for the presence of the force majeure provision, the state would satisfy the requirements in section 172(c)(1) to adopt and submit all RACM, including RACT, as needed to attain the standard as expeditiously as practicable and in section 172(c)(6) to include emission limits as necessary to attain. However, due to the presence of the force majeure provision, at this time EPA can only propose a limited approval of the emission limits for SIP strengthening purposes.
C. New Source Review (NSR)
In its submittal, TCEQ provided a certification statement that Texas already has EPA-approved rules that address nonattainment NSR requirements. EPA initially approved Texas' nonattainment NSR regulations for SO2 on November 27, 1995 (60 FR 49781). TCEQ determined that because previously approved revisions to the Texas SIP already includes 30 TAC section 116.12 (Nonattainment and Prevention of Significant Deterioration Review Definitions) and 30 TAC section 116.151 (New Major Source or Major Modification in Nonattainment Area Other Than Ozone), Texas has satisfied the nonattainment NSR SIP requirements for the Rusk-Panola nonattainment area. Further, TCEQ already certified that Texas has EPA-approved rules that cover nonattainment NSR requirements with the timely-submitted 2010 SO2 NAAQS Infrastructure and Transport SIP Revision. Therefore, EPA concludes that the SIP satisfies the CAA's NSR requirements.
D. Reasonable Further Progress (RFP)
Section 171(1) of the CAA defines RFP as “such annual incremental reductions in emissions of the relevant air pollutant as are required by [part D] or may reasonably be required by the [EPA] for the purpose of ensuring attainment of the applicable [NAAQS] by the applicable attainment date.” For purposes of SO2 , the EPA issued guidance prescribing how states could satisfy this requirement when developing their nonattainment SIPs.[46] Since pollutants like SO2 usually have a limited number of sources affecting areas of air quality that are relatively well defined, and emissions control measures for such sources generally provide significant and immediate improvements in air quality, there is usually a single “step” between pre-control nonattainment and post-control attainment. Therefore, due to the discernible relationship between emissions and air quality, EPA interprets RFP in the SO2 context as “adherence to an ambitious compliance schedule” which “ensures that affected sources implement appropriate control measures as expeditiously as practicable” to ensure attainment by the applicable attainment date. See General Preamble, 74 FR 13498, 13547 (April l6, 1992).
In its submittal, TCEQ provided its rationale for concluding that the plan met the requirement for RFP in accordance with EPA guidance. According to TCEQ, the Rusk-Panola area contains a single source with well-defined emissions, such that emissions controls for this source should result in a “swift and dramatic improvement in air quality.” TCEQ further explained that enforceable emission limitations would be implemented for the source (Martin Lake) in this area and, therefore, this compliance schedule fulfills the RFP requirement for the Rusk-Panola area. In its submittal, TCEQ sets two compliance deadlines for Vistra to meet its emissions limits from the Agreed Order. For limits expressed in lbs/hr, compliance is required no later than the date by which Texas is required to demonstrate compliance with the 2010 SO2 NAAQS, which would be the area's attainment date of January 12, 2022. For limits expressed in lbs/MMBtu, compliance is required by July 11, 2022. EPA has determined that once control requirements and emissions limits have been implemented, these measures will provide for attainment in the area. This meets the requirement for RFP for the Rusk-Panola area. EPA proposes a limited approval/limited disapproval for this SIP submission in part because the compliance date is several months after the attainment date for this area. EPA proposes that, but for the presence of the force majeure provision, the SIP submittal would provide for RFP. However, due to the force majeure provision and untimely compliance date, EPA proposes a limited approval and limited disapproval of the SIP limits Texas relied upon for its RFP demonstration for SIP strengthening purposes and to apply federally enforceable limits to the area as expeditiously as possible.
E. Contingency Measures
As discussed in our 2014 SO2 Guidance, section 172(c)(9) of the CAA defines contingency measures as such measures in a SIP that are to be implemented in the event that an area fails to make RFP, or fails to attain the NAAQS, by the applicable attainment date. Contingency measures are to become effective without further action by the state or the EPA, where the area has failed to (1) achieve RFP or (2) attain the NAAQS by the statutory attainment date for the affected area. These control measures are to consist of other available control measures that are not included in the control strategy for the nonattainment area SIP. EPA guidance describes special features of SO2 planning that influence the suitability of alternative means of addressing the requirement in section 172(c)(9) for contingency measures. Because SO2 control measures are by definition based on what is directly and quantifiably necessary for emissions controls, any violations of the NAAQS are likely related to source violations of a source's permit or agreed order terms. Therefore, an appropriate means of satisfying this requirement for SO2 is for the state to have a comprehensive enforcement program that identifies sources of violations of the SO2 NAAQS and to undertake an aggressive follow-up for compliance and enforcement.
For its contingency plan, Texas stated that TCEQ's comprehensive program to (1) identify sources of violations of the NAAQS is satisfied through its monitoring network and (2) follow-up for compliance and enforcement is satisfied through TCEQ's enforcement programs authorized under the Texas Water Code and Texas Health and Safety Code. If EPA makes the determination that that the Rusk-Panola Area has failed to attain, TCEQ will notify Martin Lake and upon notification the owner or operator will be required to do a full system audit of all SO2 emissions from Martin Lake within 90 days. The owner or operator of Martin Lake must conduct a root cause analysis for the reason why the area failed to attain and recommend provisional SO2 emission controls as necessary. Additionally, Texas has the authority to issue orders pursuant to the Texas Clean Air Act and Texas Health and Safety Code for the purpose of supporting attainment and maintenance of the 2010 SO2 NAAQS. EPA believes that this approach generally continues to be a valid approach for the implementation of contingency measures to address the 2010 SO2 NAAQS. However, as previously discussed, the presence of the force majeure provision undermines the enforceability of the emission limits in the SIP submission, and consequently undermines the utility of Texas' enforcement authority to address periods of non-compliance with the limits. Therefore, EPA is proposing that but for the presence of the force majeure provision, Texas' plan would adequately provide for contingency measures as required by the CAA. As a result, EPA can only propose limited approval of the limits upon which the SIP relies for SIP strengthening purposes.
F. Conformity
Generally, as set forth in section 176(c) of the CAA, conformity requires that actions by federal agencies do not cause new air quality violations, worsen existing violations, or delay timely attainment of the relevant NAAQS. General conformity applies to federal actions, other than certain highway and transportation projects, if the action takes place in a nonattainment area or maintenance area ( i.e., an area which submitted a maintenance plan that meets the requirements of section 175A of the CAA and has been redesignated to attainment) for ozone, particulate matter, nitrogen dioxide, carbon monoxide, lead, or SO2. EPA's General Conformity Rule (40 CFR 93.150 to 93.165) establishes the criteria and procedures for determining if a federal action conforms to the SIP. With respect to the 2010 SO2 NAAQS, federal agencies are expected to continue to estimate emissions for conformity analyses in the same manner as they estimated emissions for conformity analyses under the previous SO2 NAAQS. EPA's General Conformity Rule includes the basic requirement that a federal agency's general conformity analysis be based on the latest and most accurate emission estimation techniques available (40 CFR 93.159(b)). When updated and improved emissions estimation techniques become available, EPA expects the federal agency to use these techniques. EPA finds that the Rusk-Panola SO2 Attainment Plan SIP Revision submission meets these conformity requirements.
Transportation conformity determinations are not required in SO2 nonattainment and maintenance areas. EPA concluded in its 1993 transportation conformity rule that highway and transit vehicles are not significant sources of SO2 . Therefore, transportation plans, transportation improvement programs and projects are presumed to conform to applicable implementation plans for SO2. ( See58 FR 3776, January 11, 1993.)
VII. Proposed Action
For Texas' February 28, 2022 SIP revision submittal, we are proposing a limited approval which will incorporate all of the submissions requirements, including the emission limits and associated control requirements such as monitoring, recordkeeping and reporting requirements into the State Implementation Plan. We have determined that the revision provides for emissions controls and limits that strengthen the existing EPA-approved Texas SIP and would satisfy the applicable CAA requirements of sections 110, 172, 191 and 192. We are proposing limited disapproval due to the enforceability deficiency caused by the force majeure provision in the SIP and the timing of the compliance date for the emissions limitations several months after the attainment date for this nonattainment area, as is necessary under the CAA and associated regulations. The State has demonstrated that its current Nonattainment NSR program covers this NAAQS; therefore, no revision to the SIP is required for the Nonattainment NSR element. Under CAA section 179(c)(1), EPA also proposes to determine that the Rusk-Panola SO2 NAA failed to attain the 2010 1-hour SO2 standard by the applicable attainment date of January 12, 2022 based on monitored data from 2019-2021.
VIII. Incorporation by Reference
In this action, we are proposing to include in a final rule regulatory text that includes incorporation by reference. In accordance with the requirements of 1 CFR 51.5, we are proposing to incorporate by reference revisions to the Texas source-specific requirements for Vistra's Martin Lake Electrical Station (Martin Lake) as described in section VII of this preamble, Proposed Action. These source-specific requirements for Martin Lake include SO2 emission limits and fuel limitations for the facility as well as other monitoring, recordkeeping, and reporting requirements. We have made, and will continue to make, these documents generally available electronically through www.regulations.gov (please contact the person identified in the FOR FURTHER INFORMATION CONTACT section of this preamble for more information).
IX. Environmental Justice Considerations
Information on Executive Order 12898 (Federal Actions To Address Environmental Justice in Minority Populations and Low-Income Populations, 59 FR 7629, February 16, 1994), and how EPA defines environmental justice (EJ) can be found in the section titled “VII. Statutory and Executive Order Reviews,” in this proposed rulemaking. EPA is providing additional analysis of environmental justice associated with this action. The results of this analysis are being provided for informational and transparency purposes only, not as a basis of our proposed action.
The EPA conducted a screening analysis using EJSCREEN, an environmental justice mapping and screening tool that provides EPA with a nationally consistent dataset and approach for combining various environmental and demographic indicators.[47] The EJSCREEN tool presents these indicators at a Census block group (CBG) level or a larger user-specified “buffer” area that covers multiple CBGs.[48] An individual CBG is a cluster of contiguous blocks within the same census tract and generally contains between 600 and 3,000 people. EJSCREEN is not a tool for performing in-depth risk analysis, but is instead a screening tool that provides an initial representation of indicators related to environmental justice and is subject to uncertainty in some underlying data ( e.g., some environmental indicators are based on monitoring data which are not uniformly available; others are based on self-reported data).[49] To help mitigate this uncertainty, we have summarized EJSCREEN data within larger “buffer” areas covering multiple block groups and representing the average resident within the buffer area surrounding Martin Lake. We present EJSCREEN environmental indicators to help screen for locations where residents may experience a higher overall pollution burden than would be expected for a block group with the same total population. These indicators of overall pollution burden include estimates of ambient particulate matter (PM2.5 ) and ozone concentration, a score for traffic proximity and volume, percentage of pre-1960 housing units (lead paint indicator), and scores for proximity to Superfund sites, risk management plan (RMP) sites, and hazardous waste facilities.[50] EJSCREEN also provides information on demographic indicators, including percent low-income, communities of color, linguistic isolation, and less than high school education.
The EPA prepared an EJSCREEN report covering a buffer area of approximately a 6-mile radius around the Martin Lake facility. Table 4 presents a summary of results from the EPA's screening-level analysis for Martin Lake compared to the U.S. as a whole. From that report, Martin Lake did not show EJ indices greater than the 80th percentiles. The full, detailed EJSCREEN report is provided in the docket for this rulemaking.
Table 4—EJSCREEN Analysis Summary for Martin Lake
Variables EJSCREEN values for buffer area (radius) for Martin Lake and the U.S. (percentile within U.S. where indicated) Martin Lake (Rusk-Panola Area, 6 miles) U.S. Pollution Burden Indicators: Particulate matter (PM 2.5 ), annual average 9.57 µg/m3 (77th %ile) 8.67 µg/m3 (—). Ozone, summer seasonal average of daily 8-hour max 40.1 ppb (32nd %ile) 42.5 ppb (—). Traffic proximity and volume score * 0.72 (2nd %ile) 760 (—). Lead paint (percentage pre-1960 housing) 0.12% (37th %ile) 0.27% (—). Superfund proximity score * 0.048 (42nd %ile) 0.13 (—). RMP proximity score * 0.17 (32nd %ile) 0.77 (—). Hazardous waste proximity score * 0.059 (11th %ile) 2.2 (—). Demographic Indicators: People of color population 31% (52nd %ile) 40% (—). Low-income population 25% (46th %ile) 30% (—). Linguistically isolated population 2% (62nd %ile) 5% (—). Population with less than high school education 13% (65th %ile) 12% (—). Population under 5 years of age 9% (82nd %ile) 6%. Population over 64 years of age 14% (44th %ile) 16% (—). * The traffic proximity and volume indicator is a score calculated by daily traffic count divided by distance in meters to the road. The Superfund proximity, RMP proximity, and hazardous waste proximity indicators are all scores calculated by site or facility counts divided by distance in kilometers.
Document Information
- Published:
- 08/02/2024
- Department:
- Environmental Protection Agency
- Entry Type:
- Proposed Rule
- Action:
- Proposed rule.
- Document Number:
- 2024-17053
- Dates:
- Written comments must be received on or before September 3, 2024.
- Pages:
- 63117-63134 (18 pages)
- Docket Numbers:
- EPA-R06-OAR-2022-0311, FRL-12123-01-R6
- Topics:
- Air pollution control, Environmental protection, Incorporation by reference, Intergovernmental relations, Reporting and recordkeeping requirements, Sulfur oxides
- PDF File:
- 2024-17053.pdf
- Supporting Documents:
- » TX240.19 Rusk-Panola February 25, 2022, SIP Submission - Appendix O Excel Document (Luminant Limestone 1 and 2 Discount Factor Analysis)
- » TX240.18 EJScreen Report - Martin Lake
- » TX240.17 List of documents redacted due to copyright concerns
- » TX240.16 Availability of Modeling Files and Alternative Model Analysis Files
- » TX240.15 TCEQ Updated Modeling Protocol for the Rusk-Panola (Martin Lake) Nonattainment Area Attainment Demonstration SIP Revision - 8-11-2021
- » TX240.14 EPA Response to TCEQ Request of Use of AERMO-HBP 7-7-2021
- » TX240.13 AECOM HBP Formulation Documentation - 5-1-2024
- » TX240.12 TCEQ Letter - Request of Use of AERMOD-HBP Alernative Model Approach 5-24-2021
- » TX240.11 EPA Model Evaluation Protocol September 1992
- » TX240.10 EPA Guidance Documents
- CFR: (1)
- 40 CFR 52