The URL can be used to link to this page

Home My WebLink AboutDAQ-2025-000613 UTAH STATE IMPLEMENTATION PLAN PM2.5 Maintenance Provisions for the Provo, UT Nonattainment Area SECTION IX.A.27 Provo, UT Maintenance Plan Section IX.A.27 i Table of Contents IX.A.27.a Introduction ..................................................................................................................................1 1) Background .......................................................................................................................................1 IX.A.27.b Redesignation Requirements ........................................................................................................2 (1) The Area Has Attained the PM2.5 NAAQS ..........................................................................................3 a) Ambient Air Quality Data (Monitoring) and Utah’s Monitoring Network ...................................3 i. Modeling Element .....................................................................................................................5 ii. EPA Acknowledgement ............................................................................................................5 b) Fully Approved Attainment Plan for PM2.5 ...................................................................................5 c) Improvements in Air Quality Due to Permanent and Enforceable Reductions in Emissions ..... 10 i. Improvement in Air Quality .................................................................................................... 10 ii. Reduction in Emissions ........................................................................................................... 12 d) State has Met Requirements of Section 110 and Part D .............................................................. 17 i. Section 110 .............................................................................................................................. 17 ii. Part D Subpart 1 and 4 ............................................................................................................. 17 e) Maintenance Plan for PM2.5 Areas ............................................................................................... 18 IX.A.27.c Maintenance Plan ....................................................................................................................... 18 (1) Demonstration of Maintenance - Modeling Analysis ........................................................................ 19 (a) Introduction .................................................................................................................................... 19 (b) Photochemical Modeling ............................................................................................................... 20 i. Emissions Preparation ............................................................................................................. 20 ii. Photochemical Modeling Domains and Grid Resolution ........................................................ 20 iii. Meteorological Data ............................................................................................................ 21 iv. Episode Selection ................................................................................................................ 22 (c) Photochemical Model Performance Evaluation ............................................................................. 31 (d) Modeled Attainment Test .............................................................................................................. 33 (2) Attainment Inventory ......................................................................................................................... 36 (3) Additional Controls for Future Years ................................................................................................ 38 (4) Mobile Source Budget for Purposes of Conformity .......................................................................... 3 8 (a) Mobile Source PM2.5 Emissions Budgets ....................................................................................... 39 i. Direct PM2.5, NOx, and VOC ................................................................................................... 39 Provo, UT Maintenance Plan Section IX.A.27 ii ii. Trading Ratios for Transportation Conformity ....................................................................... 41 (5) Nonattainment Requirements Applicable Pending Plan Approval .................................................... 41 (6) Revise in Eight Years ........................................................................................................................ 41 (7) Verification of Continued Maintenance and Monitoring ................................................................... 41 (8) Contingency Plan ............................................................................................................................... 42 (a) List of Potential Contingency Measures ........................................................................................ 42 (b) Tracking ......................................................................................................................................... 43 (c) Triggering ...................................................................................................................................... 43 List of Tables Table IX.A.27. 1 Prerequisites to Redesignation in the Federal Clean Air Act ............................................3 Table IX.A.27. 2 Monitored Ambient 24-hr PM2.5 Data ...............................................................................5 Table IX.A.27. 3 Provo, UT Serious SIP Approval Status ...........................................................................6 Table IX.A.27. 4 Area Source Rules Implementation Schedule and EPA Approval Status ....................... 10 Table IX.A.27. 5 Point Source Emission Control Measure Implementation Schedule and Compliance Mechanism .................................................................................................................................................. 14 Table IX.A.27. 6 Area Source Rule Emissions Reduction in Provo NAA ................................................. 16 Table IX.A.27. 7 CAA Maintenance Plan Requirements ............................................................................ 19 Table IX.A.27. 8 Atypical Event Values Excluded from Baseline Design Value at the Spanish Fork Monitor ........................................................................................................................................................ 35 Table IX.A.27. 9 Baseline and Future Design Values (ug/m3) at Monitors in Provo NAA ........................ 36 Table IX.A.27. 10 Emissions Inventory in Tons Per Average Episode Day by Source Category and Year ..................................................................................................................................................................... 38 Table IX.A.27. 11 2035 Mountainland Association of Government’s Motor Vehicle Emission Budget ... 40 Table IX.A.27. 12 Comparison of 2026 On-Road Mobile Inventory to Total Emissions Inventory .......... 40 List of Figures Figure IX.A.27. 1 Utah's PM2.5 Air Monitoring Network .............................................................................4 Figure IX.A.27. 2 Provo NAA PM2.5 Annual Mean Concentration ............................................................ 12 Figure IX.A.27. 3 Provo NAA PM2.5 98th Percentile of 24-hr Concentration ............................................. 12 Figure IX.A.27. 4 CAMx Photochemical Modeling Domains in Two-Way Nested Configuration ........... 21 Figure IX.A.27. 5 Measured and Modeled 24-hr PM2.5 Concentrations During January 1-10 2011 at Lindon Monitoring Station in Provo NAA .................................................................................................. 24 Figure IX.A.27. 6 a) Measured and b) Modeled Species Contribution (in %) to PM2.5 at Lindon Monitoring Station in the Provo NAA on a Typical 24-hr PM2.5 Exceedance Day .................................... 25 Figure IX.A.27. 7 Measured and Modeled 24-hr PM2.5 Concentrations During December 7-19, 2013, at Lindon Monitoring Station in the Provo NAA ............................................................................................ 26 Provo, UT Maintenance Plan Section IX.A.27 iii Figure IX.A.27. 8 a) Measured and b) Modeled Chemical Composition of 24-hr PM2.5 in ug/m3 and % of PM2.5 at Lindon Monitoring Station in Provo NAA on December 12, 2013 ............................................... 27 Figure IX.A.27. 9 Measured and Modeled 24-hr PM2.5 Concentrations During February 1-16, 2016, at Lindon Monitoring Station in the Provo NAA. FRM data was missing for all episode days. Reported measurements correspond to data collected with a continuous PM2.5 instrument. ...................................... 28 Figure IX.A.27. 10 a) Measured and b) Modeled Chemical Composition of PM2.5 in ug/m3 and % of PM2.5 at Lindon in the Provo NAA on February 12, 2016 .................................................................................... 29 Figure IX.A.27. 11 Modeled vs. Measured 24-hr PM2.5 at Lindon Monitoring Station for Each of the Three Modeling Episodes: January 2011, December 2013, and February 2016. Dots represent each individual day of the modeling episode. Linear regression fits (dashed line) and equation are shown for each episode. ............................................................................................................................................... 30 Figure IX.A.27. 12 Ten-day Time Series of Observed (black) and Modeled (red) 24-hr Average PM2.5 Concentrations During January 1-10, 2011, at Lindon Monitoring Station in the Provo NAA. Dashed Red Line is NAAQS for 24-hr PM2.5. ................................................................................................................. 32 Figure IX.A.27. 13 a) Measured and b) Modeled Species Contribution (in ug/m3 and %) to PM2.5 at Lindon Monitoring Station in the Provo NAA during a typical 24-hr PM2.5 exceedance day .................... 33 Provo, UT Maintenance Plan Section IX.A.27 iv List of Acronyms and Abbreviations BACM Best Available Control Measure BACT Best Available Control Technology CAA Clean Air Act CDD Clean Data Determination CFR Code of Federal Regulations CAMx Comprehensive Air Quality Model with Extensions DAQ Utah Division of Air Quality (also UDAQ) EPA Environmental Protection Agency FR Federal Register MOVES Motor Vehicle Emission Simulator MPO Metropolitan Planning Organization MVEB Motor Vehicle Emissions Budget μg/m3 Micrograms Per Cubic Meter Micron One Millionth of a Meter NAAQS National Ambient Air Quality Standards NH3 Ammonia NOx Nitrogen Oxides NNSR Nonattainment New Source Review PM Particulate Matter PM10 Particulate Matter Smaller Than 10 Microns in Diameter PM2.5 Particulate Matter Smaller Than 2.5 Microns in Diameter R-307 Utah Administrative Code Air Quality Rules Provo NAA Provo Nonattainment Area RACM Reasonably Available Control Measures RACT Reasonably Available Control Technology RFP Reasonable Further Progress SIP State Implementation Plan SMAT Software for Model Attainment Test SMOKE Sparse Matrix Operator Kernal Emissions SO2 Sulfur Dioxide SOx Sulfur Oxides TPY Tons Per Year TSD Technical Support Document UAC Utah Administrative Code UT Utah VMT Vehicle Miles Travelled VOC Volatile Organic Compounds WRF Weather Research and Forecasting Provo, UT Maintenance Plan Section IX.A.27 1 Section IX.A.27 1 PM2.5 Maintenance Provisions for the Provo, UT 2 Nonattainment Area 3 IX.A.27.a Introduction 4 The Provo Nonattainment Area (Provo NAA) has attained the 2006 PM2.5 24-hour National Ambient Air 5 Quality Standard (NAAQS). As a result, this Section has been added to the State Implementation Plan 6 (SIP) to demonstrate that the Provo NAA is eligible for redesignation to attainment. Under Section 7 107(d)(3)(E) of the Clean Air Act (CAA or the Act), a nonattainment area is eligible for redesignation 8 when the area has met the following requirements: (1) the area has attained the national ambient air 9 quality standard, (2) the area has an Environmental Protection Agency (EPA) approved SIP, (3) the 10 improvement in air quality is due to permanent and enforceable reductions in emissions resulting from 11 implementation of the SIP, (4) the state has met the SIP requirements of Section 110 and Part D of the 12 Act, and (5) the area has an EPA approved Maintenance Plan. 13 As demonstrated in Subsection IX.A.27.b, the Provo NAA has satisfied the redesignation requirements of 14 Section 107 and is eligible for redesignation pending the EPA’s approval of the Provo NAA Maintenance 15 Plan. The maintenance plan is included in Subsection IX.A.27.c and was written in compliance with 16 Section 175A of the Act. The maintenance plan demonstrates that the Provo NAA will continue to 17 maintain the 24-hr PM2.5 NAAQS through at least the year 2035. The maintenance plan also includes 18 contingency measures to assure that the State will promptly correct any violation of the standard that may 19 occur after redesignation. Upon the EPA’s approval of the maintenance plan, the State is requesting that 20 the Provo NAA be redesignated to attainment for the 2006 PM2.5 24-hour NAAQS.1 21 1) Background 22 In October of 2006, EPA revised the 1997 NAAQS for PM2.5. While the annual standard remained 23 unchanged at 15 µg /m3 until 2012, the 24-hr standard was lowered from 65 µg /m3 to 35 µg /m3. The 24 Utah Division of Air Quality (UDAQ) has monitored PM2.5 since 2000 and found that all areas have 25 complied with the 1997 standards. Since the promulgation of the 2006 standard, all or parts of seven Utah 26 counties have recorded monitoring data that was not in compliance with the new 24-hr standard. In 2012, 27 EPA lowered the annual standard to 12 µg /m3, and all areas of the state meet this new standard. 28 On November 13, 2009, EPA designated the Provo NAA as nonattainment for the 2006 24-hour PM2.5 29 NAAQS under the Act’s general provisions for nonattainment areas. On January 4, 2013, the D.C. Circuit 30 Court of Appeals issued a decision holding that the specific provisions for PM10 nonattainment areas, 31 which are found in Part D, Subpart 4 of the Act, also apply to PM2.5 nonattainment areas. These 32 provisions require EPA to classify a PM nonattainment area as “moderate” at the time it is designated 33 nonattainment. If the area cannot attain the NAAQS by the attainment date, then EPA is required to 34 1 Concurrent with the State’s submittal of SIP Section IX.A.27 to the EPA, Governor Gary Herbert will submit a letter to EPA requesting that EPA approve the maintenance plan and redesignate the Provo NAA to attainment. Provo, UT Maintenance Plan Section IX.A.27 2 reclassify the area as “serious.” On June 2, 2014, the EPA classified the Provo NAA as a moderate 1 nonattainment area with an attainment date of December 31, 2015. 2 The Act requires areas failing to meet the federal ambient PM2.5 standard to develop a SIP with sufficient 3 control requirements to expeditiously attain and maintain the standard. On December 22, 2014, UDAQ 4 submitted a moderate area nonattainment SIP for the Provo NAA.2 The modeled attainment 5 demonstration underlying the moderate SIP assessed the likelihood of attainment by the applicable 6 attainment date of December 31, 2015 and concluded that it would be impracticable to do so. 7 After reaching the statutory attainment date, the EPA was compelled to determine whether the area had or 8 had not achieved compliance with the standard by evaluating the prior three years of quality assured data. 9 On May 10, 2017, EPA determined that the Provo NAA did not reach attainment of the 2006 24-hour 10 standard by the attainment date (89 FR 21711). EPA subsequently reclassified the Provo NAA from a 11 moderate PM2.5 nonattainment area to a serious PM2.5 nonattainment area on June 9, 2017. 12 Under Subpart 4 of the Act, serious PM nonattainment areas require, in addition to the provisions 13 submitted to meet the moderate area planning requirements, the submittal of a SIP revision that: 1) 14 provides for attainment of the applicable NAAQS no later than the end of the 10th calendar year after the 15 area’s designation as nonattainment (December 31, 2019, for the Provo NAA), and 2) includes provisions 16 to assure that the best available control measures (BACM) for the control of PM2.5 and its precursors shall 17 be implemented no later than four years after the date the area is re-classified as a serious area (June 9, 18 2021, for the Provo NAA). To fulfill the subpart 4 requirements, UDAQ submitted serious SIP elements 19 to EPA on February 4, 2019, including BACM analysis. SIP approval is discussed in more detail in 20 IX.A.27.b(2). 21 The statutory attainment date for the Provo NAA is December 31, 2019. Under the 24-hour PM2.5 22 NAAQS, compliance is determined by the average of three years of 98th percentile values. On April 10, 23 2019 (84 FR 14267), the EPA published a final determination that based on the validated data from 2015-24 2017, the Provo, UT nonattainment area attained the 2006 primary and secondary 24-hour PM2.5 NAAQS. 25 The purpose of this SIP submittal is to demonstrate that the Provo NAA is eligible for redesignation to 26 attainment (IX.A.27.b) and document a ten-year maintenance plan (IX.A.27.c). 27 IX.A.27.b Redesignation Requirements 28 Section 107(d)(3)(E) of the Act outlines five requirements that a nonattainment area must satisfy before 29 an area may be eligible for redesignation from nonattainment to attainment status. Table IX.A.27.1 30 identifies the redesignation requirements as they are stated in Section 107(d)(3)(E) of the Act. Each 31 element will be addressed in turn, with the central element being the maintenance plan found in 32 Subsection IX.A.27.c below. 33 2 UDAQ. December 3, 2014. Utah State Implementation Plan. Control Measures for Area and Point Sources, Fine Particulate Matter, PM2.5 SIP for the Provo, UT Nonattainment Area. Section IX. Part A.22. https://deq.utah.gov/legacy/laws-and-rules/air- quality/sip/docs/2014/12Dec/SIP%20IX.A.22_PROVO_FINAL_Adopted%2012-3-14.pdfd%2012-3-14.pdf Provo, UT Maintenance Plan Section IX.A.27 3 Category Requirement Reference Addressed in Section Attainment of Standard Three consecutive years of PM2.5 monitoring data must show that violations of the standard are no longer occurring CAA §107(d)(3)(E)(i) IX.A.27.b(1) Approved SIP The attainment SIP for the area must be fully approved CAA §107(d)(3)(E)(ii) IX.A.27.b(2) Permanent and Enforceable Emissions Reductions The State must be able to reasonably attribute the improvement in air quality to emission reductions that are permanent and enforceable CAA §107(d)(3)(E)(iii), Calcagni memo (Sect 3, para 2) IX.A.27.b(3) Section 110 and Part D requirements The State must verify that the area has met all requirements applicable to the area under section 110 and Part D CAA: §107(d)(3)(E)(v), §110(a)(2), Sec 171 IX.A.27.b(4) Maintenance Plan The Administrator has fully approved the Maintenance Plan for the area as meeting the requirements of CAA §175A CAA: §107(d)(3)(E)(iv) IX.A.27.b(5) and IX.A.27.c Table IX.A.27. 1 Prerequisites to Redesignation in the Federal Clean Air Act 1 (1) The Area Has Attained the PM2.5 NAAQS 2 CAA 107(d)(3)(E)(i) – The Administrator determines that the area has attained the national ambient air 3 quality standard. To satisfy this requirement, the State must show that the area is attaining the applicable 4 NAAQS. According to EPA’s guidance3 concerning area redesignations, there are generally two 5 components involved in making this demonstration. The first relies upon ambient air quality data which 6 should be representative of the area of highest concentration and should be collected and quality assured 7 in accordance with 40 CFR 58. The second component relies upon supplemental air quality modeling. 8 Each component will be addressed in turn. 9 a) Ambient Air Quality Data (Monitoring) and Utah’s Monitoring Network 10 The NAAQS for PM2.5 are listed in 40 CFR 50.13. The 2006 24-hour NAAQS is 35 micrograms per cubic 11 meter (µg/m3) for a 24-hour period and is met when the 98th percentile 24-hr concentration is less than or 12 equal to 35 µg/m3. Each year’s 98th percentile is the daily value beneath which 98% of all daily values 13 would fall. The procedure for evaluating PM2.5 data with respect to the NAAQS is specified in Appendix 14 N of 40 CFR Part 50. Generally speaking, the 24-hr PM2.5 standard is met when a three-year average of 15 98th percentile values is less than or equal to 35 µg/m3. 16 PM2.5 has been monitored in Utah since 2000, following the promulgation of the 1997 PM2.5 NAAQS. 17 UDAQ’s monitors are appropriately located to assess concentration, trends, and changes in PM2.5 18 concentrations. During Utah’s wintertime temperature inversions, daily sampling and real time 19 monitoring are necessary for both public notification and to provide data for the air quality models. 20 The UDAQ Air Monitoring Section maintains an ambient air monitoring network in Utah in accordance 21 with 40 CFR 58 that collects both air quality and meteorological data. Figure IX.A.27.1 on the following 22 3 John Calcagni. September 4, 1992. EPA Memorandum “Procedures for Processing Requests to Redesignate Areas to Attainment.” Provo, UT Maintenance Plan Section IX.A.27 4 page shows the location of sites along the Wasatch Front and in the Cache Valley that collect PM2.5 data. 1 The ambient air quality monitoring network along Utah’s Wasatch Front and in the Cache Valley is 2 routinely audited by the EPA, and meets the agency’s requirements for air monitoring networks. 3 4 Figure IX.A.27. 1 Utah's PM2.5 Air Monitoring Network 5 6 Provo, UT Maintenance Plan Section IX.A.27 5 Table IX.A.27.2 below shows the 98th percentile values in µg /m3 for 2015, 2016, and 2017 as well as the 1 three-year average of these values. The three-year average, or design value from 2015-2017 was used by 2 EPA in their final clean data determination for the Provo NAA (84 FR 14267). 3 Location 2015 2016 2017 3-Year Average of 98th percentiles North Provo 25.0 36.6 21.9 27.8 Lindon 27.3 36.3 28.9 30.8 Spanish Fork 28.1 29.2 27.6 28.3 Table IX.A.27. 2 Monitored Ambient 24-hr PM2.5 Data 4 i. Modeling Element 5 EPA guidance4 concerning redesignation requests and maintenance plans discusses the requirement that 6 the area has attained the standard and notes that air quality modeling may be necessary to determine the 7 representativeness of the monitored data. Areas that were designated nonattainment based on modeling 8 will generally not be redesignated to attainment unless an acceptable modeling analysis indicates 9 attainment. The Provo NAA was not designated based on modeling; therefore, additional modeling is not 10 necessary to determine the representativeness of the monitored data. The Provo NAA clean data 11 determination was made based on validated ambient monitored values. Consequently, modeling is not 12 necessary to show attainment. However, modeling was conducted for the purpose of this maintenance 13 demonstration to show continued compliance with the PM2.5 NAAQS through the year 2035 (see section 14 IX.A.27.c). 15 ii. EPA Acknowledgement 16 The data presented in the preceding paragraphs demonstrates that the Provo NAA is attaining the 24-hr 17 PM2.5 NAAQS. On April 10, 2019, EPA published notice in the Federal Register (84 FR 14267) that 18 pursuant to CAA section 199(b)(2), “the EPA is finalizing a clean data determination (CDD) for the 2006 19 24-hour fine particulate matter (PM2.5) Provo, Utah (UT) nonattainment area (NAA).” This determination 20 was based on quality-assured, quality-controlled and validated ambient air monitoring data for 2015-21 2017. 22 23 b) Fully Approved Attainment Plan for PM2.5 24 CAA 107(d)(3)(E)(ii) - The Administrator has fully approved the applicable implementation plan for the 25 area under section 110(k). 26 27 Areas designated as nonattainment that attain the standard prior to the SIP submittal deadline, or prior to 28 an area’s approved attainment date, are eligible for reduced regulatory requirements as described in 29 EPA’s “Clean Data Policy.”5 Under the Clean Data Policy, the EPA issued a clean data determination on 30 April 10, 2019 (84 FR 14267) for the Provo NAA. The approval status of both the moderate and serious 31 Provo SIPs is dependent on the clean data determination requirements as detailed in 81 CFR 51.1015. For 32 4 Ibid 5 Steve Page, Director, EPA Office of Air Quality Policy Planning and Standard. December 14, 2004. EPA Memorandum to Air Division Directors, “Clean Data Policy for the Fine Particle National Ambient Air Quality Standards.” Provo, UT Maintenance Plan Section IX.A.27 6 a serious PM2.5 nonattainment area, a clean data determination suspends the requirements for the state to 1 submit an attainment demonstration, reasonable further progress (RFP) plans, quantitative milestones, and 2 contingency measures until such time as: (1) the area is redesignated to attainment, after which such 3 requirements are permanently discharged; or (2) the EPA determines that the area has re-violated the 4 PM2.5 NAAQS, at which time the state shall submit such attainment plan elements for the serious 5 nonattainment area by a future date to be determined by the EPA. Table IX.A.27.3 details the EPA SIP 6 approval status. 7 On February 4, 2019, Utah submitted the required serious SIP elements for the Provo NAA. Additionally, 8 EPA guidance6 states that approval action on SIP elements and the redesignation request may occur 9 simultaneously. Requirements listed in Table IX.A.27.3 that show pending approval may fall into this 10 category. 11 Requirement EPA Action & Date FR Citation Base Year and Projection Year Emission Inventories Approval Pending -- Modeled Attainment Demonstration Clean Data Determination 4/10/2019 84 FR 14267 BACT Approval Pending -- On-Road Mobile BACM Approval Pending -- Non-Road Mobile BACM Approval Pending -- Area Source BACM See Table IX.A.27.4 -- MVEB Clean Data Determination 4/10/2019 84 FR 14267 Nonattainment New Source Review (R307-403) Approved on 7/25/2019 84 FR 35832 Reasonable Further Progress Clean Data Determination 4/10/2019 84 FR 14267 Quantitative Milestones Clean Data Determination 4/10/2019 84 FR 14267 Contingency Measures Clean Data Determination 4/10/2019 84 FR 14267 Table IX.A.27. 3 Provo, UT Serious SIP Approval Status 12 13 The SIP elements still required under the clean data policy7 include emission inventories, NNSR 14 requirements, and BACM/BACT. The EPA approved R307-403, Permits: New and Modified Sources in 15 Nonattainment Areas and Maintenance Areas on July 25, 2019 (84 FR 35832), which covers the NNSR 16 requirement for the PM2.5 attainment plans. The State has submitted the emission inventories, and 17 BACM/BACT elements to the EPA, including the R307-300 series amendments and the point source 18 6 Calcagni (n 3) 7 Environmental Protection Agency. August 24, 2016. Fine Particulate Matter National Ambient Air Quality Standards: State Implementation Plan Requirements; Final Rule. 82 FR 58128. Provo, UT Maintenance Plan Section IX.A.27 7 BACT emission limitation and operating practices (Utah SIP Section IX.H). These SIP elements are still 1 pending EPA approval. 2 While many of the moderate and serious SIP elements are suspended under the clean data determination, 3 many of the moderate SIP element have been approved. As part of the Utah moderate SIPs, 24 area 4 source rules were either introduced or augmented to control PM2.5 and PM2.5 precursors. On February 25, 5 2016 (81 FR 9343), and October 19, 2016 (81 FR 71988), the EPA approved area source rule revisions 6 and reasonably available control measure (RACM) analyses (where appropriate) for the majority of the 7 R307-300 series. See Table IX.A.27.4 for details on rules, approval dates, and implementation schedules. 8 For the SLC and Provo NAAs, the BACM analysis resulted in revisions to 13 different area source rules 9 which affect surface coating, graphic arts, and aerospace manufacture and rework facilities. 10 EPA-Approved/Conditionally Approved Control Measures for UT Moderate PM2.5 SIPs Implementation Schedule R307-302 Solid Fuel Burning Devices 1 EPA conditionally approved* October 19, 2016 (81 FR 71988). February 1, 2017 R307-303 Commercial Cooking 1 EPA approved February 25, 2016 (81 FR 9343). December 15, 2015 R307-304 Solvent Cleaning 1 December 6, 2017 R307-307 Road Salting and Sanding EPA approved February 25, 2016 (81 FR 9343). January 1, 2014 R307-309 Nonattainment and Maintenance Areas for PM10 and PM2.5: Fugitive Emissions and Fugitive Dust 1 EPA proposed for approval September 14, 2017 (82 FR 43205). Salt Lake County, Utah County, and the City of Ogden – January 1, 2013. Remaining NAAs – April 1, 2013. Amended August 4, 2017 R307-312 Aggregate Processing Operations for PM2.5 Nonattainment Areas. EPA approved October 19, 2016 (81 FR 71988). February 4, 2016 Provo, UT Maintenance Plan Section IX.A.27 8 EPA-Approved/Conditionally Approved Control Measures for UT Moderate PM2.5 SIPs Implementation Schedule R307-335 Degreasing and Solvent Cleaning Operations 1 EPA approved February 25, 2016 (81 FR 9343). All sources within Salt Lake and Davis Counties R307-335-3 through R307-335-6 – January 1, 2013. All other sources defined in R307-335-2 – September 1, 2013. All sources within Box Elder, Cache, Utah, Weber, and Tooele Counties R307-335-7 – August 1, 2014 Amended October 29, 2017, by removing sections 6 & 7 to for rule R307-304 R307-342 Adhesives & Sealants 1 EPA approved February 25, 2016 (81 FR 9343). December 1, 2014 R307-343 Emissions Standards for Wood Furniture Manufacturing Operations 1 EPA approved February 25, 2016 (81 FR 9343) Sources in Salt Lake and Davis Counties – September 1, 2013. Sources in Box Elder, Cache, Tooele, Utah, and Weber Counties – January 1, 2014. Amended December 6, 2017 R307-344 Paper, Film & Foil Coatings 1 EPA approved February 25, 2016 (81 FR 9343) Sources in Salt Lake and Davis Counties – February 1, 2013. Sources in Box Elder, Cache, Tooele, Utah, and Weber Counties – January 1, 2014. Amended December 6, 2017 R307-345 Fabric & Vinyl Coatings 1 EPA approved February 25, 2016 (81 FR 9343) Sources in Salt Lake and Davis Counties – February 1, 2013. Sources in Box Elder, Cache, Tooele, Utah, and Weber Counties – January 1, 2011. Amended December 6, 2017 R307-346 Metal Furniture Surface Coatings 2 EPA approved February 25, 2016 (81 FR 9343) Sources in Salt Lake and Davis Counties – February 1, 2013. Sources in Box Elder, Cache, Tooele, Utah, and Weber Counties – January 1, 2014. Amended December 6, 2017 R307-347 Large Appliance Surface Coatings 2 EPA approved February 25, 2016 (81 FR 9343) Sources in Salt Lake and Davis Counties – February 1, 2013. Sources in Box Elder, Cache, Tooele, Utah, and Weber Counties – January 1, 2014. Amended December 6, 2017 Provo, UT Maintenance Plan Section IX.A.27 9 EPA-Approved/Conditionally Approved Control Measures for UT Moderate PM2.5 SIPs Implementation Schedule R307-348 Magnet Wire Coatings 2 EPA approved February 25, 2016 (81 FR 9343) Sources in Salt Lake and Davis Counties – February 1, 2013. Sources in Box Elder, Cache, Tooele, Utah, and Weber Counties – January 1, 2014. Amended December 6, 2017 R307-349 Flat Wood Panel Coatings 1 EPA approved February 25, 2016 (81 FR 9343) Sources in Salt Lake and Davis Counties – February 1, 2013. Sources in Box Elder, Cache, Tooele, Utah, and Weber Counties – January 1, 2014. Amended December 6, 2017 R307-350 Miscellaneous Metal Parts and Products Coatings 1 EPA approved February 25, 2016 (81 FR 9343) Sources in Salt Lake and Davis Counties – September 1, 2013. Sources in Box Elder, Cache, Tooele, Utah, and Weber Counties – January 1, 2014. Amended December 6, 2017 R307-351 Graphic Arts 1 EPA approved February 25, 2016 (81 FR 9343) Sources in Salt Lake and Davis Counties – February 1, 2013. Sources in Box Elder, Cache, Tooele, Utah, and Weber Counties – January 1, 2014. Amended December 6, 2017 R307-352 Metal Containers, Closure, and Coil Coatings 2 EPA approved February 25, 2016 (81 FR 9343) January 1, 2014 Amended December 6, 2017 R307-353 Plastic Parts Coatings 1 EPA approved February 25, 2016 (81 FR 9343) January 1, 2014 Amended December 6, 2017 R307-354 Automotive Refinishing Coatings 1 EPA approved February 25, 2016 (81 FR 9343) January 1, 2014 Amended December 6, 2017 R307-355 Control of Emissions from Aerospace Manufacture and Rework Facilities 1 EPA approved February 25, 2016 (81 FR 9343) January 1, 2014 Amended March 8, 2018 R307-356 Appliance Pilot Light 1 EPA approved February 25, 2016 (81 FR 9343) January 1, 2013 Provo, UT Maintenance Plan Section IX.A.27 10 EPA-Approved/Conditionally Approved Control Measures for UT Moderate PM2.5 SIPs Implementation Schedule R307-357 Consumer Products 1 EPA approved February 25, 2016 (81 FR 9343) May 8, 2014 R307-361 Architectural Coatings 1 EPA approved February 25, 2016 (81 FR 9343) October 31, 2013 Table IX.A.27. 4 Area Source Rules Implementation Schedule and EPA Approval Status 1 1 control measure implementation schedule and confirmation that measures have been implemented 2 2 control measure implementation schedule and review if any new sources located in the NAA 3 *UDAQ submitted the committed revisions on February 1, 2017, within the one-year conditional 4 approval window 5 6 The clean data determination has suspended all other elements of the Provo NAA PM2.5 attainment plan, 7 including reasonable further progress (RFP) plans, quantitative milestones, and contingency measures at 8 this time. Considering the suspended SIP elements through the clean data policy and the approval or 9 expected approval of required elements, Utah has met requirement 107(d)(3)(E)(ii) for the Provo NAA. 10 c) Improvements in Air Quality Due to Permanent and Enforceable Reductions in 11 Emissions 12 CAA 107(d)(3)(E)(iii) - The Administrator determines that the improvement in air quality is due to 13 permanent and enforceable reductions in emissions resulting from implementation of the applicable 14 implementation plan and applicable Federal air pollutant control regulations and other permanent and 15 enforceable reductions. Speaking further on the issue, EPA guidance8 reads that the State must be able to 16 reasonably attribute the improvement in air quality to emission reductions which are permanent and 17 enforceable. In the following sections, both the improvement in air quality and the emission reductions 18 themselves will be discussed. 19 i. Improvement in Air Quality 20 The improvement in air quality with respect to PM2.5 can be shown in a number of ways. Improvement, in 21 this case, is relative to the various control strategies that affected the airshed. For the Provo NAA, these 22 control strategies were implemented as the result of both the moderate SIP and the serious designation 23 BACM/BACT requirements, submitted to EPA in December 2014 and February 2019, respectively. The 24 various control measure effective dates are detailed in Tables IX.A.27.4 and IX.A.27.5. 25 An assessment of the ambient air quality data collected at monitors in the NAA from the year monitoring 26 began to 2018 (the last year of validated data) shows an observable decrease in monitored PM2.5 (see 27 Figure IX.A.27.2 and Figure IX.A.27.3). The Provo NAA is designated nonattainment only for the 24-28 hour health standard, not for the annual standard. However, it is useful to observe both the 98th percentile 29 average of 24-hr data as well as the annual arithmetic mean to understand trends (see Figure IX.A.27.2). 30 8 Calcagni (n 3) Provo, UT Maintenance Plan Section IX.A.27 11 Ambient concentrations in excess of the 24-hr standard are typically only incurred during winter months 1 when cold-pool conditions drive and trap secondary PM2.5. The actual cold-pool temperature inversions 2 vary in strength and duration from year to year, and the PM2.5 concentrations measured during those times 3 reflect this variability far more than they reflect gradual changes in the emissions of direct PM2.5 and 4 PM2.5 precursors. This variability is apparent in Figure IX.A.27.3. Despite the variability, if a line is fit 5 through the 24-hr data, the trend is noticeably downward and indicates an improvement of a little less 6 than one µg /m3 per year. 7 This episodic variability is reduced by looking at annual mean values of PM2.5 concentrations shown in 8 Figure IX.A.27.2. The data is still skewed more by winter data than summer data. It includes all of the 9 high values identified as the 98th percentiles, as well as the values ranked even higher. Still, the trend is 10 downward. Fitting a line through the data collected at the Lindon site (chosen because the monitor 11 consistently records the highest values in the NAA) reveals a trend that noticeably decreases and indicates 12 an improvement of approximately 3.0 µg /m3, over the 18-year span. 13 Provo, UT Maintenance Plan Section IX.A.27 12 1 Figure IX.A.27. 2 Provo NAA PM2.5 Annual Mean Concentration 2 3 Figure IX.A.27. 3 Provo NAA PM2.5 98th Percentile of 24-hr Concentration 4 ii. Reduction in Emissions 5 As stated above, EPA guidance9 says that the State must be able to reasonably attribute the improvement 6 in air quality to emission reductions that are permanent and enforceable. In making this showing, the State 7 9 Ibid Provo, UT Maintenance Plan Section IX.A.27 13 should estimate the percent reduction (from the year that was used to determine the design value) 1 achieved by Federal measures such as motor vehicle control, as well as by control measures that have 2 been adopted and implemented by the State. 3 As mentioned, the ambient air quality data presented in Subsection IX.A.27.b(3)(a) includes values prior 4 to the nonattainment designation through 2018 to illustrate the lasting effect of the implemented control 5 strategies. In discussing the effect of the controls, as well as the control measures themselves, however, it 6 is important to keep in mind the time necessary for their implementation. 7 The moderate nonattainment SIP for the Provo NAA included a statutory date for the implementation of 8 RACM/RACT of December 31, 2014. Thus, 2015 marked the first year in which RACM/RACT was 9 reflected in the emissions inventories for the Provo NAA. Section 189(c) of the CAA identifies, as a 10 required plan element, quantitative milestones which are to be achieved every three years, and which 11 demonstrate reasonable further progress (RFP) toward attainment of the standard by the applicable date. 12 As defined in CAA Section 171(1), the term reasonable further progress means “such annual incremental 13 reductions in emissions of the relevant air pollutant as are required by this part or may reasonably be 14 required by the Administrator for the purpose of ensuring attainment of the applicable national ambient 15 air quality standard by the applicable date.” Hence, the milestone report must demonstrate that the control 16 strategy is achieving reasonable progress toward attainment. 17 The RACM prescribed by the moderate nonattainment SIP and the subsequent implementation by the 18 State is discussed in more detail in a milestone report submitted for the Provo NAA to EPA on March 23, 19 2018, within the 90 day post-milestone date required by CAA 189(c)(2) and 51.1013(b). On October 24, 20 2018, EPA sent Governor Gary Herbert a letter stating “The Environmental Protection Agency has 21 determined that the 2017 Quantitative Milestone Reports are adequate. The basis for this determination is 22 set forth in the enclosures. This determination is based on the EPA’s review of information contained in 23 the Moderate Area Plans and additional information provided in the 2017 Quantitative Milestone 24 Reports.” This approval letter is included in the TSD for this SIP submittal. Much of the downward trend 25 in the ambient data as seen in Figures IX.A.27.2 and IX.A.27.3 is attributable to the controls implemented 26 through the moderate SIP. 27 40 CFR 51.1011 establishes that control measures must be implemented no later than the beginning of the 28 year containing the applicable attainment date, January 1, 2019, for the Provo NAA. Any control 29 measures implemented beyond such date are instead regarded as additional feasible measures. 30 Implementation schedules for point source control measures are included in Table IX.A.27.5. Emission 31 reductions leading to lower ambient values can be observed in Figures IX.A.27.2 and IX.A.27.3, with 32 further improvements expected beyond 2019 as a result of the more stringent BACM/BACT 33 requirements. 34 Company RACT Equipment Updates BACT Requirements Implementation Schedule Quantity Reduction (tons/yr) Compliance Mechanism PacifiCorp Lake Side N/A (currently at RACT) Retention of NOx limits from existing permit: Already Implementing (use of SCR) N/A identical to previous existing RACT NOx CEM Provo, UT Maintenance Plan Section IX.A.27 14 Block 1: 14.9 lb/hr Block 2: 18.1 lb/hr controls and permit as written McWane Ductile Limiting VOC emissions 118.16 tons/yr Replaced RACT limit w/ VOCs from finishing paint line < 1 ton/day Already implementing N/A Recordkeeping requirement outlined in IX.H.13.c.i.a Heat input limit on annealing ovens (63.29 MMBtu/hr) Retained RACT limit as BACT Already implementing N/A General requirements of IX.H.11.c Table IX.A.27. 5 Point Source Emission Control Measure Implementation Schedule and 1 Compliance Mechanism 2 3 As part of the Utah moderate SIPs, 24 area source rules were either introduced or augmented to control 4 PM2.5 and PM2.5 precursors. For the serious SIP area source BACM review, each of UDAQ’s existing area 5 source rules were re-evaluated to ensure that all appropriate source categories were addressed in 6 rulemaking and that the level of control required is consistent with BACM. For newly identified controls 7 or enhancement of existing controls, an evaluation was made to determine technological and economic 8 feasibility. The BACM review resulted in revisions to 13 different area source rules which affect surface 9 coating (for a variety of different surfaces), graphic arts, and aerospace manufacture & rework facilities. 10 The rules and amendments are listed in Table IX.A.27.4. Table IX.A.27.6 shows the effectiveness of the 11 area source rules within Provo NAA. 12 13 14 15 16 17 18 19 20 21 22 23 Provo, UT Maintenance Plan Section IX.A.27 15 1 2017 PROVO NAA lb/day Area Source Rule Name NOX VOC NH3 SO2 PM2.5 R307-342 adhesive/sealants 0.0 393.5 0.0 0.0 0.0 R307-355 aerospace manufacture & rework 0.0 0.0 0.0 0.0 0.0 R307-312 aggregate processing 0.0 22.8 0.0 0.0 0.0 R307-347 appliance surface coating 0.0 0.0 0.0 0.0 0.0 R307-354 automotive refinishing 0.0 154.6 0.0 0.0 0.0 R307-352 metal container, closure, & coil coating 0.0 0.0 0.0 0.0 0.0 R307-303 commercial cooking 0.0 282.4 0.0 0.0 0.0 R307-357 consumer products 0.0 1478.3 0.0 0.0 0.0 R307-335 degreasing & solvent cleaning 0.0 0.0 0.0 0.0 0.0 R307-349 flat wood panel coatings 0.0 0.0 0.0 0.0 0.0 R307-309 fugitive dust 0.0 0.0 0.0 0.0 916.3 R307-351 graphic arts 0.0 0.0 0.0 0.0 0.0 R307-208 outdoor wood boilers 3.4 111.4 2.8 3.4 105.8 R307-221 landfill controls 0.0 63.2 0.0 0.0 0.0 R307-348 magnet wire coatings 0.0 0.0 0.0 0.0 0.0 R307-346 metal furniture surface coating 0.0 0.0 0.0 0.0 0.0 R307-350 misc metal parts & product coating 0.0 0.0 0.0 0.0 0.0 R307-361 architectural coating 0.0 2178.6 0.0 0.0 0.0 R307-344 paper/film/foil coating 0.0 0.0 0.0 0.0 0.0 R307-356 appliance pilot light 494.3 28.9 0.0 3.2 2.3 R307-353 plastic parts coating 0.0 0.0 0.0 0.0 0.0 R307-302 residential wood burning ban 321.7 2659.7 100.0 38.0 2330.4 R307-230 water heaters 0.0 0.0 0.0 0.0 0.0 R307-343 wood furniture manufacturing 0.0 0.0 0.0 0.0 0.0 Total emissions reduced lb/day 819.4 7373.5 102.8 44.6 3354.8 Provo, UT Maintenance Plan Section IX.A.27 16 1 Table IX.A.27. 6 Area Source Rule Emissions Reduction in Provo NAA 2 3 In reality, the NAAs should expect to see continued improvement in the next five to ten years as a result 4 of the phase-in period of a number of the area source rules and some additional feasible measures 5 installed at point sources. For example, the gas-fired water heater rule R307-230 requires that only ultra-6 low NOx gas-fired water heaters to be sold or installed after July 1, 2018, but it takes years for water 7 heater turnover to occur. In addition, the 13 rules that were revised during the serious SIP BACM review 8 were implemented at the state level in 2018 and have a five-year phase-in period, resulting in full 9 emission reduction by 2023. Therefore, additional emissions reductions will be seen. These phase-in 10 periods were considered in the inventories used for modeling in this SIP. 11 Existing controls not implemented through the SIP process also affect the emission rates from non-12 stationary source categories. The federal motor vehicle control program has been one of the most 13 significant control strategies affecting emissions that produce PM2.5. Tier 1 and 2 standards were 14 implemented by 1997 and 2008 respectively. Tier 3 vehicle/engine standards were initiated with new 15 vehicles coming to market in 2017 (25% of new sales) with full phase in by 2021 (100% of new sales). 16 For gasoline, the five Wasatch Front refineries and the Sinclair refinery in Wyoming that also supplies 17 gasoline to the Wasatch Front market, are considered small refineries by EPA’s rule. As such, these 18 2026 PROVO NAA lb/day Area Source Rule Name NOX VOC NH3 SO2 PM2.5 R307-342 adhesive/sealants 0.00 608.95 0.00 0.00 0.00 R307-355 aerospace manufacture & rework 0.00 0.22 0.00 0.00 0.00 R307-312 aggregate processing 0.00 23.49 0.00 0.00 0.00 R307-347 appliance surface coating 0.00 48.12 0.00 0.00 0.00 R307-354 automotive refinishing 0.00 479.11 0.00 0.00 0.00 R307-352 metal container, closure, & coil coating 0.00 0.00 0.00 0.00 0.00 R307-303 commercial cooking 0.00 18.23 0.00 0.00 0.00 R307-357 consumer products 0.00 1829.57 0.00 0.00 0.00 R307-335 degreasing & solvent cleaning 0.00 483.36 0.00 0.00 0.00 R307-349 flat wood panel coatings 0.00 16.68 0.00 0.00 0.00 R307-309 fugitive dust 0.00 0.00 0.00 0.00 1118.72 R307-351 graphic arts 0.00 315.66 0.00 0.00 0.00 R307-208 outdoor wood boilers 3.20 108.40 2.80 3.20 103.00 R307-221 landfill controls 0.00 78.17 0.00 0.00 0.00 R307-348 magnet wire coatings 0.00 0.31 0.00 0.00 0.00 R307-346 metal furniture surface coating 0.00 99.92 0.00 0.00 0.00 R307-350 misc metal parts & product coating 0.00 64.83 0.00 0.00 0.00 R307-361 architectural coating 0.00 2696.22 0.00 0.00 0.00 R307-344 paper/film/foil coating 0.00 350.49 0.00 0.00 0.00 R307-356 appliance pilot light 2992.74 139.35 0.00 15.20 10.89 R307-353 plastic parts coating 0.00 125.97 0.00 0.00 0.00 R307-302 residential wood burning ban 312.68 2589.01 98.03 36.62 2268.45 R307-230 water heaters 2620.83 0.00 0.00 0.00 0.00 R307-343 wood furniture manufacturing 0.00 408.39 0.00 0.00 0.00 Total emissions reduced lb/day 5929.4 10484.4 100.8 55.0 3501.1 Provo, UT Maintenance Plan Section IX.A.27 17 refineries have a tier 3 delayed implementation date of January 1, 2020 to produce a tier 3 (10 ppm sulfur) 1 gasoline product or produce a gasoline product (greater than 10 ppm sulfur) with compensating sulfur 2 credits. Similarly, the Heavy-Duty Engine and Vehicle Standards took effect in 2007 and were fully 3 phased in by 2010. Air quality benefits, particularly those stemming from the light-duty and heavy-duty 4 vehicle standards, continue to be realized as older, higher-polluting vehicles are replaced by newer, 5 cleaner vehicles. 6 To supplement the federal motor vehicle control program, an Inspection and Maintenance Program was 7 implemented in Utah County. This program has been effective in identifying vehicles that no longer meet 8 the emission specifications for their respective makes and models and in ensuring that those vehicles are 9 repaired in a timely manner. 10 Emissions from non-road mobile emission sources also benefit from several significant regulatory 11 programs enacted at the federal level. This category of emitters includes airplanes, locomotives, hand-12 held engines, and larger portable engines such as generators and construction equipment. The 13 effectiveness of these controls has been incorporated into the “NONROAD” model UDAQ uses to 14 compile the inventory information for this source category. 15 The emissions reductions resulting from federal programs and the RACM/RACT plus BACM/BACT 16 controls incorporated into the Utah SIP or promulgated at the State level, result in emissions reductions 17 that are consistent with the notion of permanent and enforceable improvements in air quality. Taken 18 together with the trends in ambient air quality illustrated in the previous paragraph, along with the 19 continued implementation of the nonattainment SIP elements for the Provo NAA, they provide a reliable 20 indication that these improvements in air quality reflect the application of permanent steps to improve the 21 air quality in the region. 22 d) State has Met Requirements of Section 110 and Part D 23 CAA 107(d)(3)(E)(v) - The State containing such area has met all requirements applicable to the area 24 under section 110 and part D. Section 110 of the Act deals with the broad scope of state implementation 25 plans and the capacity of the respective state agency to effectively administer such a plan. Part D deals 26 specifically with plan requirements for nonattainment areas, including those requirements that are specific 27 to PM2.5. 28 i. Section 110 29 The State has met all requirements applicable to the Provo NAA under Section 110 of the Act. Section 30 110(a)(2) contains the general requirements or infrastructure elements necessary for EPA approval of the 31 SIP. On September 21, 2010, the State submitted an Infrastructure SIP to EPA demonstrating compliance 32 with the requirements of Section 110 that are applicable to the 2006 PM2.5 NAAQS. EPA approved the 33 State’s Infrastructure SIP on November 25, 2013 (78 FR 63883) for all Section 110 requirements that are 34 applicable to redesignation. 35 ii. Part D Subpart 1 and 4 36 Part D of the Act addresses “Plan Requirements for Nonattainment Areas.” Subparts 1 and 4 of Part D 37 contain planning elements that must be included in the SIP. This includes the requirement to submit an 38 Provo, UT Maintenance Plan Section IX.A.27 18 attainment demonstration, reasonable further progress plans, quantitative milestones and milestone 1 reports, a motor vehicle emission budget for the attainment year for the purposes of transportation 2 conformity, and contingency measures for the area. However, upon EPA’s issuance of a final clean data 3 determination demonstrating that the Provo NAA has attained the standard, these requirements are 4 suspended (40 C.F.R. § 51.1015(b) and 84 FR 26054). 5 The remaining Part D requirements that are relevant to redesignation are requirements that are 6 independent of helping the area achieve attainment. This includes the requirement to have a 7 nonattainment new source review (“NNSR”) program, emissions inventory submission, and 8 implementation of BACM/BACT. The State has satisfied these remaining requirements. Utah’s NNSR 9 program can be found in Utah Administrative Rule R307-403, Permits: New and Modified Sources in 10 Nonattainment Areas and Maintenance Areas. EPA fully approved the current version of the NNSR 11 program on July 25, 2019 (84 FR 35832). The BACM/BACT requirements and the emissions inventory 12 were included in the serious SIP element submittal for the Provo NAA that the State submitted to the 13 EPA on February 4, 2019. Upon EPA’s approval of these elements prior to or concurrently with EPA’s 14 action on the maintenance plan/redesignation request, Utah will have complied with all applicable Part D 15 requirements. 16 e) Maintenance Plan for PM2.5 Areas 17 As stated in the Act, an area may not be redesignated to attainment without first submitting and receiving 18 EPA approval of a maintenance plan. The maintenance plan is a quantitative showing that the area will 19 continue to attain the NAAQS for an additional 10 years (from EPA approval), accompanied by sufficient 20 assurance that the terms of the numeric demonstration will be administered by the State and by the EPA 21 in an oversight capacity. The maintenance plan is the central criterion for redesignation. It is contained in 22 the following subsection. 23 IX.A.27.c Maintenance Plan 24 CAA 107(d)(3)(E)(iv) - The Administrator has fully approved a maintenance plan for the area as meeting 25 the requirements of section 175A. An approved maintenance plan is one of several criteria necessary for 26 area redesignation as outlined in Section 107(d)(3)(E) of the Act. The maintenance plan itself, as 27 described in Section 175A of the Act and further addressed in EPA guidance10 has its own list of required 28 elements. The following table is presented to summarize these requirements. Each will then be addressed 29 in turn. 30 Category Requirement Reference Addressed in Section Maintenance demonstration Provide for maintenance of the relevant NAAQS in the area for at least 10 years after redesignation. CAA: 175A(a) IX.A.27.c (1) Revise in 8 Years The State must submit an additional revision to the plan, 8 years after CAA: 175A(b) IX.A.27.c (6) 10 Calcagni (n 3) Provo, UT Maintenance Plan Section IX.A.27 19 redesignation, showing an additional 10 years of maintenance. Continued Implementation of Nonattainment Area Control Strategy The Clean Air Act requires continued implementation of the NAA control strategy unless such measures are shown to be unnecessary for maintenance or are replaced with measures that achieve equivalent reductions. CAA: 175A(c), 110(l), Calcagni memo IX.A.27.c (5) Contingency Measures Areas seeking redesignation from nonattainment to attainment are required to develop contingency measures that include State commitments to implement additional control measures in response to future violations of the NAAQS. CAA: Sec 175A(d) IX.A.27.c (8) Verification of Continued Maintenance The maintenance plan must indicate how the State will track the progress of the maintenance plan. Calcagni memo IX.A.27.c (7) Table IX.A.27. 7 CAA Maintenance Plan Requirements 1 (1) Demonstration of Maintenance - Modeling Analysis 2 CAA 175A(a) - Each State which submits a request under section 107(d) for redesignation of a 3 nonattainment area as an area which has attained the NAAQS shall also submit a revision of the 4 applicable implementation plan to provide for maintenance of the NAAQS for at least 10 years after the 5 redesignation. The plan shall contain such additional measures, if any, as may be required to ensure 6 such maintenance. The maintenance demonstration is discussed in EPA guidance11 as one of the core 7 provisions that should be considered by states for inclusion in a maintenance plan. 8 According to the EPA guidance, a State may generally demonstrate maintenance of the NAAQS by 9 either showing that future emissions of a pollutant or its precursors will not exceed the level of the 10 attainment inventory (discussed below) or by modeling to show that the future mix of sources and 11 emission rates will not cause a violation of the NAAQS. Utah has elected to make its demonstration 12 based on air quality modeling. 13 (a) Introduction 14 The following chapter presents an analysis using observational datasets to detail the chemical regimes of 15 Utah’s NAAs. Prior to the develop of this maintenance plan, UDAQ conducted a technical analysis to 16 support the development of the serious SIP for the SLC NAA. The analysis included preparation of 17 emissions inventories and meteorological data, and the evaluation and application of a regional 18 photochemical model. Part of this process included episode selection to determine the episode that most 19 accurately replicates the photochemical formation of ambient PM2.5 during a persistent cold air pool 20 episode in the airshed. For this maintenance plan, UDAQ is using the same episode that was used for the 21 serious SIP modeling. 22 11 Ibid Provo, UT Maintenance Plan Section IX.A.27 20 (b) Photochemical Modeling 1 UDAQ used the Comprehensive Air Quality Model with Extensions (CAMx) version 6.30 for air quality 2 modeling. CAMx v6.30 is a state-of-the-art air quality model that includes State of Utah funded 3 enhancements for wintertime modeling. These enhancements include snow chemistry, topographical and 4 surface albedo refinements. CAMx is an EPA approved model for use in SIP modeling. Its configuration 5 for use in this SIP, with respect to model options and model adjustments, is discussed in the Technical 6 Support Document. 7 i. Emissions Preparation 8 The emissions processing model used in conjunction with CAMx is the Sparse Matrix Operator Kernel 9 Emissions Modeling System (SMOKE) version 3.6.512. SMOKE prepares the annual emissions inventory 10 for use in the air quality model. There are three aspects to the preparation of an annual emissions 11 inventory for air quality modeling: 12 ● Temporal: Convert emissions from annual to daily, weekly and hourly values. 13 ● Spatial: Convert emissions from a county-wide average to gridded emissions. 14 ● Speciation: Decompose PM2.5 and VOC emissions estimates into individual subspecies using the 15 latest Carbon Bond 6 speciation profiles. 16 The process of breaking down emissions for the air quality model was done with sets of activity profiles 17 and associated cross reference files. These are created for point or large industrial source emissions, 18 smaller area sources, and mobile sources. Direct PM2.5 and PM2.5 precursor estimates were modified via 19 temporal profiles to reflect wintertime conditions. 20 Activity profiles and their associated cross reference files from the EPA’s 2011v613 modeling platform 21 were used. For stationary non-point and mobile sources, spatial surrogates from the EPA Clearinghouse 22 for Inventories and Emissions Factors (CHIEF14) were used to distribute emissions in space across the 23 modeling domain. Emissions from large industrial sources (point sources) were placed at the location of 24 the source itself. Where reliable local information was available (population density, traffic demand 25 modeling, residential heating), profiles and surrogates were modified or developed to reflect that 26 information. 27 ii. Photochemical Modeling Domains and Grid Resolution 28 The UDAQ CAMx v6.30 modeling framework consists of two spatial domains: a high-resolution 1.33 km 29 domain nested inside of a coarser 4 km domain (see Figure IX.A.27.4). This configuration allows one to 30 efficiently integrate regional effects with local impacts within the Provo NAA. Vertical resolution in the 31 model consists of 41 layers extending to the top of the atmosphere. 32 12 https://www.cmascenter.org/smoke/ 13 https://www.epa.gov/air-emissions-modeling/2011-version-6-air-emissions-modeling-platforms 14 https://www.epa.gov/chief Provo, UT Maintenance Plan Section IX.A.27 21 1 Figure IX.A.27. 4 CAMx Photochemical Modeling Domains in Two-Way Nested 2 Configuration 3 4 The UDAQ 4 km coarse domain covers the entire state of Utah, a significant portion of Eastern Nevada 5 (including Las Vegas), as well as smaller portions of Idaho, Wyoming, Colorado, and Arizona. The fine 6 1.33 km domain covers all of Utah’s three PM2.5 nonattainment areas, including the Provo NAA. 7 Throughout this document, we will refer to the fine 1.33 km domain as the “modeling domain” when the 8 coarse domain is not specified. 9 iii. Meteorological Data 10 Meteorological modeling was carried out by the University of Utah (University) with financial support 11 from UDAQ. 12 Meteorological inputs were derived using the Weather Research and Forecasting15 (WRF) Advanced 13 Research WRF (WRF-ARW) model to prepare meteorological datasets for our use with the 14 photochemical model. WRF contains separate modules to compute different physical processes such as 15 surface energy budgets and soil interactions, turbulence, cloud microphysics, and atmospheric radiation. 16 Within WRF, the user has many options for selecting the different schemes for each type of physical 17 process. There is also a WRF Preprocessing System (WPS) that generates the initial and boundary 18 15 https://www.mmm.ucar.edu/weather-research-and-forecasting-model Provo, UT Maintenance Plan Section IX.A.27 22 conditions used by WRF, based on topographic datasets, land use information, and larger-scale 1 atmospheric and oceanic models. 2 Model performance of WRF was assessed against observations at sites maintained by the University. 3 WRF has reasonable ability to replicate the vertical temperature structure of the boundary layer (i.e., the 4 temperature inversion), although it is difficult for WRF to reproduce the inversion when the inversion is 5 shallow and strong (i.e., an 8-degree temperature increase over 100 vertical meters). A summary of the 6 performance evaluation results for WRF is included in the TSDs. 7 iv. Episode Selection 8 Part of the modeling exercise involves a test to see whether the model can successfully replicate the PM2.5 9 mass and composition that was observed during prior episode(s) of elevated PM2.5 concentration. The 10 selection of an appropriate episode, or episodes, for use in this exercise requires some forethought and 11 should determine the meteorological episode that helps produce the best air quality modeling 12 performance. 13 EPA Guidance16 identifies some selection criteria that should be considered for SIP modeling, including: 14  Select episodes that represent a variety of meteorological conditions that lead to elevated PM2.5. 15  Select episodes during which observed concentrations are close to the baseline design value. 16  Select episodes that have extensive air quality data bases. 17  Select enough episodes such that the model attainment test is based on multiple days at each 18 monitor violating NAAQS. 19 20 After careful consideration, the following meteorological episodes were selected as candidates for Utah’s 21 SIP modeling: 22  January 1-10, 2011 23  December 7-19, 2013 24  February 1-16, 2016 25 26 In addition to the criteria identified in the modeling guidance, each of these candidate episodes may be 27 characterized as having the following atmospheric conditions: 28  Nearly non-existent surface winds 29  Light to moderate winds aloft (wind speeds at mountaintop < 10-15 m/s) 30  Simple cloud structure in the lower troposphere (e.g., consisting of only one or no cloud layer) 31  Singular 24-hour PM2.5 peaks suggesting the absence of weak intermittent storms during the 32 episode 33 34 16 Environmental Protection Agency. April 2007. Guidance on the Use of Models and Other Analyses for Demonstrating Attainment of Air Quality Goals for Ozone, PM2.5, and Regional Haze. Provo, UT Maintenance Plan Section IX.A.27 23 Previous work conducted by the University of Utah and UDAQ showed the four conditions listed above 1 improve the likelihood for successfully simulating wintertime persistent cold air pools in the WRF 2 model17. A comprehensive discussion of the meteorological model performance for all three episodes may 3 be found in the Technical Support Document for the meteorological modeling18. 4 a) Model Adjustments and Settings 5 In order to better simulate Utah’s winter-time inversion episodes six different adjustments were made to 6 CAMx input data: 7 1. Increased vertical diffusion rates (Kvpatch). 8 2. Lowered residential wood smoke emissions to reflect burn ban compliance during forecasted high 9 PM2.5 days (burn ban). 10 3. Ozone deposition velocity set to zero and increased urban area surface albedo (snow chemistry) 11 4. Cloud water content reduced during certain days (cloud adjustment). 12 5. NH3 injection to account for missing NH3 sources in UDAQ’s inventory. This is defined as 13 artificially adding non-inventoried NH3 emissions to the inventoried emissions that are input into 14 CAMx. 15 6. Reduced the dry deposition rate of NH3 by setting NH3 Rscale to 1. Rscale is a parameter in 16 CAMx that reflects surface resistance. 17 7. Applied a 93% reduction to paved road dust emissions. 18 19 Depending on the episode, different adjustments were applied. All adjustments were applied to the 20 January 2011 episode while select adjustments were applied to the other two episodes. 21 Kvpatch improved overall model performance by enhancing vertical mixing over urban areas. Snow 22 chemistry modifications, which included reducing ozone deposition velocity and increasing surface 23 albedo over urban areas, helped improve the model performance by better representing secondary 24 ammonium nitrate formation during winter-time inversion episodes in Utah. 25 Cloud adjustments were only applied to the January 2011 episode, which was characterized by cloud 26 cover on January 6-8 over the Salt Lake and Utah valleys. This cloud cover led to a high bias in sulfate 27 due to the effect of NH3 on the gas-to-particle partitioning of sulfate in clouds. Application of the cloud 28 adjustment scheme helped reduce this bias. 29 Rscale modification and burn ban adjustments were also only applied to the January 2011 episode. The 30 burn ban adjustments reflect the compliance rate with the state’s two-stage policy ban on wood-burning. 31 A 93% reduction in paved road dust emissions was only applied to the January 2011 emissions. This 32 adjustment helped improve the model performance for crustal material. 33 17 https://www.mmm.ucar.edu/weather-research-and-forecasting-model 18 https://documents.deq.utah.gov/air-quality/planning/technical-analysis/research/model-improvements/3- wintertime-episodes/DAQ-2017-014342.pdf Provo, UT Maintenance Plan Section IX.A.27 24 b) Episodic Model Performance 1 2 Shown below for each of three episodes are the CAMx performance results for total 24-hour PM2.5 mass 3 and PM2.5 chemical species, including nitrate (NO3), sulfate (SO4), ammonium (NH4), organic carbon 4 (OC), elemental carbon (EC), chloride (Cl), sodium (Na), crustal material (CM) and other species (other 5 mass). 6 January 1-10, 2011 7 A comparison of 24-hr modeled and observed PM2.5 during January 1-10, 2011 at the Lindon monitoring 8 station in the Provo NAA showed that overall the model captures the temporal variation in PM2.5 well 9 (Figure IX.A.27.5). The gradual increase in PM2.5 concentration and its transition back to low levels are 10 generally well reproduced by the model. Moreover, with the exception of January 3-5, the bias between 11 measured and modeled PM2.5 is overall relatively small, particularly on PM2.5 exceedance days. The large 12 bias on January 3-5 can be mainly related to the meteorological model performance on these days where 13 jet wind speeds were overestimated in the WRF model19. 14 15 Figure IX.A.27. 5 Measured and Modeled 24-hr PM2.5 Concentrations During January 1-10 16 2011 at Lindon Monitoring Station in Provo NAA 17 18 The model performance for PM2.5 chemical species was also good for this episode as indicated by a 19 comparison of measured and modeled PM2.5 chemical composition at Lindon monitoring station on a 20 PM2.5 exceedance day (Figure IX.A.27.6). Given that measurements of PM2.5 chemical species were not 21 available for a PM2.5 exceedance day during the January 1-10, 2011, modeling episode, this analysis is 22 based on a comparison of the fraction of individual PM2.5 chemical species in total PM2.5 mass between 23 2011 model outputs and measurements from 2013. Measurements correspond to filter speciation data 24 collected at Lindon during a typical winter-time inversion event in 2013. 25 19 https://documents.deq.utah.gov/air-quality/planning/technical-analysis/research/model-improvements/3- wintertime-episodes/DAQ-2017-014342.pdf 0 10 20 30 40 50 60 12345678910 Jan. 2011 24 ‐ h r P M 2 . 5 c o n c e n t r a t i o n ( u g / m 3 ) Lindon Measured Modeled Provo, UT Maintenance Plan Section IX.A.27 25 1 Figure IX.A.27. 6 a) Measured and b) Modeled Species Contribution (in %) to PM2.5 at 2 Lindon Monitoring Station in the Provo NAA on a Typical 24-hr PM2.5 Exceedance Day 3 4 As can be seen, the chemical composition of modeled PM2.5 is similar to that of measured PM2.5, with 5 modeled secondary species, NO3, NH3, and SO4, accounting for over 50% of PM2.5 mass, in agreement 6 with measurements. The model also performed well for OC while it overestimated the percent 7 contributions of EC and crustal material to PM2.5. This overprediction could be related to an 8 overestimation in source emissions. Speciation measurements specific to this episode are needed for 9 further confirmation. 10 Overall, the model simulated well the timing and strength of the capping inversion during this January 11 episode. PM2.5 chemical species, particularly NO3 and ammonium, are also well simulated in the model, 12 suggesting that this episode is suitable for modeling. 13 December 7-19, 2013 14 The model performance for the December 7-19, 2013, episode was first evaluated for 24-hr PM2.5 mass. A 15 comparison of modeled and measured 24-hr PM2.5 during this period showed that, while the model 16 generally represented well the temporal variation in PM2.5, the model simulated low PM2.5 concentrations 17 compared to measurements (Figure IX.A.27.7). This is likely related to a warm model temperature bias in 18 the Utah Valley between December 10-14 due to inadequate simulation of stratus cloud formation during 19 NO3 49% NH4 15% SO4 2% OC 10% EC 4% Crustal Material 2% Na 0.13% Cl 1% Other Mass 17% Lindon December 12 2013 Measured PM2.5 = 45.9 ug/m3 NO3 50% NH4 15% SO4 3% OC 10% EC 8% Crustal Material 10% Na 0.25% Cl 0.49% Other mass 4% Lindon January 07 2011 Modeled PM2.5 = 45.02 ug/m3 b)a) Provo, UT Maintenance Plan Section IX.A.27 26 December 12-14 and inadequate representation of the surface of the Utah Lake. Although frozen in reality 1 during this December episode, the surface of the Utah lake was not represented as frozen in the model20. 2 3 Figure IX.A.27. 7 Measured and Modeled 24-hr PM2.5 Concentrations During December 7-4 19, 2013, at Lindon Monitoring Station in the Provo NAA 5 * Federal Reference Monitor (FRM) data is missing for this day. Reported measurement corresponds to 6 data collected with a continuous PM2.5 instrument. 7 To further evaluate the model performance during this episode, modeled and measured PM2.5 chemical 8 species on December 12, which corresponds to a PM2.5 exceedance day with available speciation 9 measurements, were compared (Figure IX.A.27.8). NO3, ammonium, and OC are all underpredicted in the 10 model, which is possibly related to the meteorological model performance. The WRF model 11 overpredicted surface temperatures, leading to increased mixing and therefore reduction in 12 concentrations. Moreover, similarly to the model performance for the January 2011 episode, crustal 13 material is overpredicted in the model. An adjustment to paved road dust emissions was not applied for 14 the December 2013 simulations. 15 Given that the strength of the capping inversion was not well simulated in the meteorological model, 16 selection of the December 2013 episode as modeling episode for modeling demonstration is not desirable. 17 20 https://documents.deq.utah.gov/air-quality/planning/technical-analysis/research/model-improvements/3- wintertime-episodes/DAQ-2017-014342.pdf 0 10 20 30 40 50 60 70 80 7 8 9* 10 11 12 13 14 15 16 17 18 19 December 2013 24 ‐ h r P M 2 . 5 c o n c e n t r a t i o n ( u g / m 3 ) Lindon Measured Modeled Provo, UT Maintenance Plan Section IX.A.27 27 1 Figure IX.A.27. 8 a) Measured and b) Modeled Chemical Composition of 24-hr PM2.5 in 2 ug/m3 and % of PM2.5 at Lindon Monitoring Station in Provo NAA on December 12, 2013 3 4 February 1-16, 2016 5 A comparison of modeled and measured 24-hr PM2.5 at the Lindon monitoring station in the Provo NAA 6 during February 1-16, 2016 showed that peak PM2.5 concentrations are not well simulated in the model 7 (Figure IX.A.27.9). The increase in PM2.5 is not well represented in the model, with PM2.5 concentrations 8 building up then dropping prematurely in the model. The model also failed at capturing the observed 9 PM2.5 peak on February 14. These results can be attributed to the meteorological model performance. A 10 warm modeled temperature bias in the Utah Valley due to early snow melt-out in the model as well as 11 premature dissipation of simulated clouds likely contributed to increased mixing and early dispersion of 12 PM2.5 in the model21. 13 21 https://documents.deq.utah.gov/air-quality/planning/technical-analysis/research/model-improvements/3- wintertime-episodes/DAQ-2017-014342.pdf NO3, 22.4, 49% NH4, 6.8, 15% SO4, 1.0, 2% OC, 4.5, 10% EC, 2.0, 4% Crustal Material, 0.9, 2% Na, 0.1, 0.13% Cl , 0.7, 1% Other Mass, 7.6, 17% Lindon December 12 2013 PM2.5 measured = 45.9 ug/m3 a) NO3, 10.7, 32% NH4, 3.5, 10% SO4, 1.0, 3% OC, 3.2, 10%EC, 2.1, 7% Crustal Material, 2.9, 9% Na, 0.0, 0% Cl , 0.1, 0% Other Mass, 9.6, 29% Lindon December 12 2013 PM2.5 modeled = 33 ug/m3 b) Provo, UT Maintenance Plan Section IX.A.27 28 1 Figure IX.A.27. 9 Measured and Modeled 24-hr PM2.5 Concentrations During February 1-2 16, 2016 at Lindon Monitoring Station in the Provo NAA. FRM data was missing for all 3 episode days. Reported measurements correspond to data collected with a continuous 4 PM2.5 instrument. 5 6 The model performance for this episode was further assessed for PM2.5 bulk chemical species on February 7 12, which corresponds to a PM2.5 exceedance day (Figure IX.A.27.10). NO3, a major component of PM2.5, 8 was underpredicted by about 25% in the model. Moreover, similarly to the model performance for the two 9 other meteorological episodes, EC and crustal material were overestimated in the model. The model 10 performance for all other species was overall acceptable. 11 Although the chemical composition of PM2.5 on February 12 is overall well reproduced by the model, the 12 timing in PM2.5 peaks was generally poorly represented, suggesting that this episode not suitable for 13 modeling. 14 15 0 10 20 30 40 50 60 70 80 90 12345678910111213141516 Feb. 2016 24 ‐ h r P M 2 . 5 c o n c e n t r a t i o n ( u g / m 3 ) Lindon Modeled Measured Provo, UT Maintenance Plan Section IX.A.27 29 1 Figure IX.A.27. 10 a) Measured and b) Modeled Chemical Composition of PM2.5 in ug/m3 2 and % of PM2.5 at Lindon in the Provo NAA on February 12, 2016 3 4 Conclusion 5 Examining the PM2.5 model performance for all three episodes, it is clear that CAMx performed best 6 when using the January 2011 WRF output, which was specifically calibrated to the meteorological 7 conditions experienced during January 2011, a period that coincided with an exhaustive field campaign 8 (Persistent Cold Air Pool Study (PCAPS)22). This was further confirmed by a linear regression analysis 9 that showed that modeled and measured PM2.5 at Lindon monitoring station were more strongly correlated 10 during the January 2011 episode (R2 = 0.89) compared to the other episodes (R2 = 0.05 and 0.81) (Figure 11 IX.A.27.11). They also displayed a slope that is close to unity (0.87) for the January 2011 episode, further 12 indicating their close agreement and good model performance when using the 2011 WRF output. 13 14 22 http://www.pcaps.utah.edu/ NO3, 21.3, 60% NH4, 5.7, 16% SO4, 1.5, 4% OC, 2.9, 8% EC, 1.6, 4% Crustal Material, 1.0, 3% Na, 0.1, 0.18% Cl, 0.2, 1%Other Mass, 1.6, 4% Lindon February 12 2016 PM2.5 Measured = 35.5 ug/m3 a) NO3, 16.0, 43% NH4, 5.0, 14%SO4, 1.3, 3% OC, 2.5, 7% EC, 4.1, 11% Crustal Material, 3.9, 11% Na, 0.1, 0.17% Cl, 0.0, 0.13% Other Mass, 4.1, 11% Lindon February 12 2016 PM2.5 Modeled = 37.0 ug/m3 b) Provo, UT Maintenance Plan Section IX.A.27 30 1 2 Figure IX.A.27. 11 Modeled vs. Measured 24-hr PM2.5 at Lindon Monitoring Station for 3 Each of the Three Modeling Episodes: January 2011, December 2013, and February 2016. 4 Dots represent each individual day of the modeling episode. Linear regression fits 5 (dashed line) and equation are shown for each episode. 6 7 Given that the January 2011 WRF data produced superior model performance when compared with the 8 other two episodes, UDAQ selected the January 2011 episode to conduct its modeled maintenance 9 demonstration work. A more thorough discussion is provided in the TSD. 10 11 y = 0.87x ‐ 2.70 R² = 0.89 0 5 10 15 20 25 30 35 40 45 50 0204060Mo d e l e d P M 2 . 5 c o n c e n t r a t i o n ( u g / m 3 ) Measured PM2.5 concentration (ug/m3) January 2011 Lindon y = 0.37x + 7.30 R² = 0.81 0 5 10 15 20 25 30 35 40 0 20406080 Mo d e l e d P M 2 . 5 c o n c e n t r a t i o n ( u g / m 3 ) Measured PM2.5 concentration (ug/m3) December 2013 Lindon y = 0.47x + 15.73 R² = 0.05 0 10 20 30 40 50 60 70 80 0204060Mo d e l e d P M 2 . 5 c o n c e n t r a t i o n ( u g / m 3 ) Measured PM2.5 concentration (ug/m3) February 2016 Lindon Provo, UT Maintenance Plan Section IX.A.27 31 1 (c) Photochemical Model Performance Evaluation 2 Introduction 3 To assess how accurately the photochemical model predicts observed concentrations and to demonstrate 4 that the model can reliably predict the change in pollution levels in response to changes in emissions, a 5 model performance evaluation was conducted. This model performance evaluation also provides support 6 for the model modifications and settings that were applied (ammonia injection, increase of surface 7 resistance to ammonia, zeroing-out of ozone deposition velocity, reduction of cloud-water content, snow 8 albedo enhancement, vertical diffusion modifications and paved road dust emissions adjustment) to more 9 accurately reproduce winter-time inversion episodes. A detailed explanation of these model modifications 10 is provided in the TSD. 11 Available ambient monitoring data were used for this photochemical model performance evaluation. Data 12 included 24-hr total PM2.5 and 24-hr chemically-speciated PM2.5 measurements collected at the Lindon 13 monitoring station in the Provo NAA. The evaluation was based on the December 31-January 10, 2011, 14 episode and the 2011 emissions inventory were used as input data for the model simulations. The 15 evaluation focused on days with PM2.5 concentration exceeding the NAAQS (> 35 µg/m3). Results for 16 December 31, which is a model spin-up day, are excluded from this evaluation. 17 A more detailed model performance evaluation that examines the model performance for gaseous species 18 is provided in the TSD. More details on the model performance at various sites within the Provo NAA are 19 also included in the TSD. 20 Daily PM2.5 Concentrations 21 A comparison of 24-hr modeled and observed PM2.5 during January 1-10, 2011, at the Lindon monitoring 22 station in the Provo non-attainment area showed that the model overall captures the temporal variation in 23 PM2.5 well (Figure IX.A.27.12). The gradual increase in PM2.5 concentration and its transition back to low 24 levels are generally well reproduced by the model. Moreover, the bias between measured and modeled 25 PM2.5 is overall relatively small, particularly on PM2.5 exceedance days. 26 Provo, UT Maintenance Plan Section IX.A.27 32 1 Figure IX.A.27. 12 Ten-day Time Series of Observed (black) and Modeled (red) 24-hr 2 Average PM2.5 Concentrations During January 1-10, 2011 at Lindon Monitoring Station in 3 the Provo NAA. Dashed Red Line is NAAQS for 24-hr PM2.5. 4 5 PM2.5 Chemical Speciation 6 The model performance was further evaluated for PM2.5 chemical species. Given that measurements of 7 PM2.5 chemical species were not available for a PM2.5 exceedance day during the selected modeling 8 episode, this analysis is based on a comparison of the fraction of individual PM2.5 chemical species in 9 total PM2.5 mass between 2011 model outputs and 2013 measurements. The latter correspond to filter 10 speciation data collected at Lindon during a typical winter-time inversion event in 2013. While the 2013 11 measurements cannot be directly compared to day-specific 2011 model simulations, the measurements are 12 useful to assess if the model predicts similar PM2.5 chemical composition during strong inversion 13 conditions. Although the concentration of individual PM2.5 chemical species may vary between inversion 14 events, their relative contribution to total PM2.5 mass is expected to remain the same during typical 15 inversion events. Chemical species, including nitrate (NO3), sulfate (SO4), ammonium (NH4), organic 16 carbon (OC), elemental carbon (EC), chloride (Cl), sodium (Na), crustal material (CM), and other species 17 (other mass), were considered in this analysis. The model performance evaluation for PM2.5 species on 18 non-PM2.5 exceedance days is provided in the TSD. 19 Figures IX.A.27.13 shows the percent contribution of modeled and measured chemical species to PM2.5 at 20 Lindon monitoring station on a typical 24-hr PM2.5 exceedance day. As can be seen, the chemical 21 composition of modeled PM2.5 is similar to that of measured PM2.5. Modeled NO3 accounts for about 50% 22 of PM2.5, in agreement with the contribution of measured NO3 to PM2.5 mass (about 49%). Measured and 23 modeled sulfate and ammonium also have similar fractional contributions to PM2.5 mass. The model 24 performance for OC was also good. On the other hand, the model overestimated the percent contributions 25 of EC and CM to PM2.5. This overprediction on days when the simulated atmospheric mixing was 26 particularly strong could be related to an overestimation in source emissions. A more thorough evaluation 27 is limited by the lack of speciation measurements for the selected modeling episode. 28 29 a) b) Provo, UT Maintenance Plan Section IX.A.27 33 1 Figure IX.A.27. 13 a) Measured and b) Modeled Species Contribution (in ug/m3 and %) to 2 PM2.5 at Lindon Monitoring Station in the Provo NAA during a typical 24-hr PM2.5 3 exceedance day 4 5 Summary of Model Performance 6 The model performance replicating the buildup and clear out of PM2.5 is good overall. The model captures 7 the temporal variation in PM2.5 well. The gradual increase in PM2.5 concentration and its transition back to 8 low levels are generally well reproduced by the model. The model also predicts reasonably well PM2.5 9 concentration on peak days. It also overall replicates well the composition of PM2.5 on exceedance days, 10 with good model performance for secondary NO3 and ammonium which account for over 50% of PM2.5 11 mass. 12 Several observations should be noted on the implications of these model performance findings on the 13 attainment modeling presented in the following section. First, it has been demonstrated that model 14 performance overall is good and, thus, the model can be used for air quality planning purposes. Second, 15 consistent with EPA guidance, the model is used in a relative sense to project future year values. EPA 16 suggests that this approach “should reduce some of the uncertainty attendant with using absolute model 17 predictions alone.” 18 (d) Modeled Attainment Test 19 Introduction 20 21 With acceptable performance, the model can be utilized to make future-year attainment projections. For 22 any given (future) year, an attainment projection is made by calculating a concentration termed the Future 23 Design Value (FDV). This value is calculated for each monitor included in the analysis, and then 24 NO3 49% NH4 15% SO4 2% OC 10% EC 4% Crustal Material 2% Na 0.13% Cl 1% Other Mass 17% Lindon December 12 2013 Measured PM2.5 = 45.9 ug/m3 NO3 50% NH4 15% SO4 3% OC 10% EC 8% Crustal Material 10% Na 0.25% Cl 0.49% Other mass 4% Lindon January 07 2011 Modeled PM2.5 = 45.02 ug/m3 Provo, UT Maintenance Plan Section IX.A.27 34 compared to the NAAQS (35 µg/m3). If the FDV at every monitor located within a NAA is less than the 1 NAAQS, this demonstrates attainment for that area in that future year. 2 A maintenance plan must demonstrate continued attainment of the NAAQS for a span of ten years. This 3 span is measured from the time EPA approves the plan, a date which is somewhat uncertain during plan 4 development. To be conservative, attainment projections were made for 2035. An assessment was also 5 made for 2026 as a “spot-check” against emission trends within the ten-year span. 6 PM2.5 Baseline Design Values 7 For any monitor, the FDV is greatly influenced by existing air quality at that location. This can be 8 quantified and expressed as a Baseline Design Value (BDV). The BDV is consistent with the form of the 9 24-hour PM2.5 NAAQS, which is the 98th percentile value averaged over a three-year period. 10 Quantification of the BDV for each monitor is included in the TSD, and is consistent with EPA guidance. 11 Several values were excluded when calculating the BDVs in the Provo NAA. EPA’s “Exceptional Events 12 Rule” allows states to exclude certain air quality data due to exceptional events such as wildfires or dust 13 storms. In the preamble to the 2016 amendments to the rule, EPA states that “the CAA also recognizes 14 that it may not be appropriate to use the monitoring data influenced by “exceptional” events that are 15 collected by the ambient air quality monitoring network when making certain regulatory determinations. 16 When “exceptional” events cause exceedances or violations of the NAAQS that subsequently affect 17 certain regulatory decisions, the normal planning and regulatory process established by the CAA may not 18 be appropriate.” 23 19 There were two large local wildfires during the summer of 2018 that affected the ambient monitored 20 PM2.5 values at the Spanish Fork monitor in the Provo NAA. When including the atypical data influenced 21 by wildfires, the baseline design value is just below the NAAQS at 35.4 µg/m3. Since the design value 22 complies with the NAAQS, the wildfire events are not considered “exceptional events” because they did 23 not cause exceedances or violations of the NAAQS (40 CFR 50.14). In anticipation that there would be 24 some determinations and analyses not covered by the Exceptional Events Rules that would rely on air 25 quality data that may have been influenced by atypical, extreme, or unrepresentative events, EPA 26 published further guidance on the subject24. 27 This guidance identifies the most common determinations and analyses not covered by the Exceptional 28 Events Rule and clarifies for each of them whether there is a separate existing mechanism under which 29 the exclusion, selection, or adjustment of air quality monitoring data may be appropriate. One example is 30 certain modeling analyses under EPA’s Guideline on Air Quality Models Rule, including modeling 31 analyses used for estimating base and future year design values for ozone and PM2.5 attainment 32 demonstrations. 33 According to the Guidance, these types of modeling analyses may exclude monitoring data if the data is 34 not representative to characterize base period concentrations which may impact a determinative value in a 35 23 Treatment of Data Influenced by Exceptional Events; Final Rule, 81 Fed. Reg. 68216 (Oct. 3, 2016). 24 EPA Memorandum. Additional Methods, Determinations, and Analyses to Modify Air Quality Data Beyond Exceptional Events. April 4, 2019. Provo, UT Maintenance Plan Section IX.A.27 35 projected time period. This could include data used to model future year design values for demonstrating 1 attainment. 2 In the case of the two Utah County fires, the ambient data recorded by the Spanish Fork monitor was 3 atypical. It did not characterize base period concentrations, and it would impact a determinative value in 4 the projected design value. Since this data is atypical and gives an atypical projected design value, it 5 should be excluded from the Provo NAA’s modeling and maintenance demonstration. 6 As a result, this maintenance plan modeling uses a baseline design value that excludes the atypical data at 7 the Spanish Fork monitor from the two fires. The baseline design value including the atypical data is 35.4 8 µg/m3. The baseline design value excluding the atypical data is 28.4 µg/m3. An extensive atypical event 9 write-up, including back trajectory analysis using HYSPLIT, is included in the TSDs. Table IX.A.27.8 10 details the filtered PM2.5 values that are excluded. 11 Date Value Wildfire Sources 8/7/2018 37.8 Coal Hollow 8/9/2018 50.8 Coal Hollow and other western state(s) fire(s) 8/10/2018 68.8 Coal Hollow and other western state(s) fire(s) 8/11/2018 49.6 Coal Hollow and other western state(s) fire(s) 8/13/2018 58.1 Coal Hollow and other western state(s) fire(s) 9/14/2018 71.5 Pole Creek and Bald Mountain 9/15/2018 42.6 Pole Creek and Bald Mountain 9/17/2018 74.5 Pole Creek and Bald Mountain 9/18/2018 57.7 Pole Creek and Bald Mountain 9/19/2018 76.3 Pole Creek and Bald Mountain 9/21/2018 39.3 Pole Creek and Bald Mountain Table IX.A.27. 8 Atypical Event Values Excluded from Baseline Design Value at the 12 Spanish Fork Monitor 13 14 Relative Response Factors 15 In making future-year predictions, the output from the CAMx model is not considered to be an absolute 16 answer. Rather, the model is used in a relative sense. In doing so, a comparison is made using the 17 predicted concentrations for both the year in question and a pre-selected baseline year, which for this plan 18 is 2017. This comparison results in a Relative Response Factor (RRF). 19 The UDAQ used the Software for Model Attainment Test - Community Edition (SMAT-CE) v. 1.01 20 utility from EPA25 to perform the modeled attainment test for daily PM2.5. SMAT is designed to 21 interpolate the species fractions of the PM mass from the Speciation Trends Network (STN) monitors to 22 the FRM monitors. It also calculates the RRF for grid cells near each monitor and uses these to calculate 23 25 https://www.epa.gov/scram/photochemical-modeling-tools Provo, UT Maintenance Plan Section IX.A.27 36 a future year design value for these grid cells. A grid of 3-by-3 (9) cells surrounding the monitors was 1 used as the boundary for relative response factor (RRF) calculations. 2 The State of Utah operates three Chemical Speciation Network (CSN) monitors: Hawthorne, Bountiful, 3 and Lindon. Hawthorne is located in Salt Lake County, the Bountiful monitor is in Davis to the north, and 4 the Lindon monitor is located in Utah County to the south. Of the three, Hawthorne samples one out of 5 three days, while the other two sample one in six days. 6 This mismatch in sampling frequency lead, initially, to interpolated speciation profiles that were 7 unexpectedly non-uniform across the Salt Lake Valley. To create more realistic speciation profiles, the 8 CSN data collected at the Hawthorne monitor were applied to all of the FRM sites in the SLC NAA. 9 UDAQ believes this is a reasonable assumption that is supported by recently conducted special studies. 10 Further discussion may be found in the TSD. 11 For each monitor, the FDV is calculated by multiplying the BDV by the relative response factor: FDV = 12 RRF * BDV. These FDV’s are compared to the NAAQS in order to determine whether attainment is 13 predicted at that location or not. The results for each of the monitors are shown below in Table 14 IX.A.27.9. For all projected years and monitors, no FDV exceeds the NAAQS. Therefore, continued 15 attainment is demonstrated for the Provo NAA. 16 17 Monitor Location 2016-2018 BDV 2026 FDV 2035 FDV* Lindon 31.1 29.3 29.5 Spanish Fork 28.4** 28.4 28.4 Table IX.A.27. 9 Baseline and Future Design Values (ug/m3) at Monitors in Provo NAA 18 *These values include additional emissions added to the MAG MVEB from the safety margin 19 **This value excludes data from atypical events discussed in the BDV section 20 (2) Attainment Inventory 21 The attainment inventory is discussed in EPA guidance26 as another one of the core provisions that should 22 be considered by states for inclusion in a maintenance plan. According to the guidance, the stated purpose 23 of the attainment inventory is to establish the level of emissions during the time periods associated with 24 monitoring data showing attainment. 25 In cases such as this, where a maintenance demonstration is founded on a modeling analysis that is used 26 in a relative sense, the modeled baseline inventory is used for comparison with every projection year 27 model run. For this analysis, the State compiled a baseyear inventory for the year 2017. This year falls 28 within the span of data representing current attainment of the PM2.5 NAAQS. The guidance discusses the 29 projection inventories as well, and notes that they should consider future growth, including population 30 and industry, should be consistent with the baseyear inventory, and should document data inputs and 31 assumptions. Any assumptions concerning emission rates must reflect permanent, enforceable measures. 32 Utah compiled projection inventories for use in the quantitative modeling demonstration. The years 33 selected for projection include 2026 and 2035. The emissions contained in the inventories include sources 34 26 Calcagni (n 3) Provo, UT Maintenance Plan Section IX.A.27 37 located within the modeling domain, encompassing all three PM2.5 nonattainment areas, as well as a 1 bordering region. See Figure IX.A.27.4. 2 Since this bordering region is so large, the State identified a “core area” within this domain wherein a 3 higher degree of accuracy is important. Within this core area (which includes Weber, Davis, Salt Lake, 4 Utah, Box Elder, Tooele, Cache, and Franklin, ID counties), SIP-specific inventories were prepared to 5 include seasonal adjustments and forecasting to represent each of the projection years. In the bordering 6 regions away from this core, the State used the most current National Emissions Inventory (2014) from 7 EPA for the analysis. 8 There are four general categories of emission sources included in these inventories: point sources, area 9 sources, on-road mobile sources, and non-road mobile sources. For each of these source categories, the 10 pollutants that were inventoried include: PM2.5, SO2, NOX, VOC, and NH3. The unit of measure for point 11 and area sources is the traditional tons per year. Mobile source emissions are reported in terms of tons per 12 day. The pre-processing model, SMOKE, converts all emissions to daily, weekly, and hourly values. 13 Area source emissions were projected to 2017 from the 2014 triannual inventory. Growth data from 14 appropriate data sources, including information from the Governor’s Office of Management and Budget, 15 was used to project inventories to 2026 and 2035. Point source emissions are represented as the actual 16 emissions from the 2017 triannual emissions inventory. Point sources were grown to 2026 and 2035 on a 17 case-by-case basis for the projection inventories. 18 On-road mobile source emissions were calculated for each year using MOVES2014b in conjunction with 19 the appropriate estimates for vehicle miles traveled (VMT). VMT estimates for the urban counties were 20 provided by the local metropolitan planning organizations (MPOs), including the Wasatch Front Regional 21 Council, the Mountainland Association of Governments, and the Cache Metropolitan Planning 22 Organization and are based their travel demand modeling for 2017, 2026, and 2035. Non-road mobile 23 source emission were calculated for each year using MOVES2014b. Growth data from appropriate data 24 sources was used to project to 2026 and 2035. The TSD accompanying this SIP includes the Inventory 25 Preparation Plan that details the growth factors used for each emissions source. 26 Source category emission inventories are expected to look quite different between 2017 and 2035. 27 Population is expected to steadily increase between the 18-year span. On-road mobile emissions dominate 28 the 2017 inventory; however, in 2035 area source emissions dominate the inventory. This is due to the tier 29 3 federal fuel standards and phase-in of newer cars driving on-road emission reductions. Area source 30 emissions are relatively stable from 2017 to 2026 to 2035, besides a decrease in NOx from 2017 to 2026 31 due to the phase-in of area source rules. 32 Since this SIP subsection takes the form of a maintenance plan, it must demonstrate that the area will 33 continue to attain the PM2.5 NAAQS throughout a period of ten years from the date of EPA approval. It is 34 also necessary to “spot check” this ten-year interval. Hence, projection inventories were prepared for 35 2026 and 2035. Table IX.A.27.10 summarizes these inventories. As described, it represents point, area, 36 on-road mobile, and non-road mobile sources in the modeling domain and include PM2.5, as well as the 37 precursors SO2, NOX, VOC, and NH3 as defined in 40 CFR Parts 50, 51, and 93. 38 Provo, UT Maintenance Plan Section IX.A.27 38 More detail concerning any element of the inventory can be found in the appropriate section of the TSD. 1 More detail about the general construction of the inventory can be found in the Inventory Preparation 2 Plan. 3 4 Table IX.A.27. 10 Emissions Inventory in Tons Per Average Episode Day by Source 5 Category and Year 6 7 (3) Additional Controls for Future Years 8 Since the emission limitations discussed in subsection IX.A.27.b(3) are federally enforceable and, as 9 demonstrated in IX.A.27.c(1) above, are sufficient to ensure continued attainment of the PM2.5 NAAQS, 10 there is no need to require any additional control measures to maintain the PM2.5 NAAQS. 11 (4) Mobile Source Budget for Purposes of Conformity 12 The transportation conformity provisions of section 176(c)(2)(A) of the Act requires regional 13 transportation plans and programs to show that “…emissions expected from implementation of plans and 14 programs are consistent with estimates of emissions from motor vehicles and necessary emissions 15 reductions contained in the applicable implementation plan…” EPA's transportation conformity 16 regulation (40 CFR 93, Subpart A, last amended at 77 FR 14979, March 14 2012 ) also requires that 17 motor vehicle emission budgets must be established for the last year of the maintenance plan, and may be 18 established for any years deemed appropriate (see 40 CFR 93.118((b)(2)(i)). 19 For an MPO’s Regional Transportation Plan, analysis years that are after the last year of the maintenance 20 plan (in this case 2035), a conformity determination must show that emissions are less than or equal to the 21 maintenance plan's motor vehicle emissions budget(s) for the last year of the implementation plan. 22 Emissions (tons/day)Sector PM2.5 Filterable PM2.5 Condensable PM2.5 Total NOx VOC NH3 SO2 Area Sources 1.75 0.29 2.04 5.01 13.32 6.54 0.06 Mobile Sources ––0.8315.49.070.430.09 NonRoad Sources ––0.21 3.07 1.66 0 0.01 Point Sources 0.18 0.12 0.3 1.12 0.18 0.42 0.05 Total 3.38 24.6 24.23 7.39 0.22 Area Sources 1.89 0.32 2.21 3.56 14.2 6.38 0.05 Mobile Sources ––0.42 5.79 4.58 0.36 0.05 NonRoad Sources ––0.14 2.14 1.65 0.01 0.01 Point Sources 0.19 0.12 0.31 0.97 0.17 0.44 0.06 Total 3.08 12.46 20.6 7.19 0.17 Area Sources 2.06 0.35 2.41 3.67 16.32 6.24 0.05 Mobile Sources ––1.41 5.74 6.49 0.44 0.05 NonRoad Sources ––0.13 1.84 1.8 0.01 0.01 Point Sources 0.19 0.12 0.31 0.97 0.17 0.44 0.06 Total 4.26 12.22 24.78 7.13 0.17 2017 2026 2035 Provo, UT Maintenance Plan Section IX.A.27 39 (a) Mobile Source PM2.5 Emissions Budgets 1 In this maintenance plan, Utah is establishing transportation conformity motor vehicle emission budgets 2 (MVEB) for direct PM2.5, NOX, and VOC for 2035. The MVEBs are established for tons per average 3 winter weekday for NOx, VOC, and direct PM2.5 (primary exhaust PM2.5 + brake and tire wear). 4 i. Direct PM2.5, NOx, and VOC 5 Direct (or “primary”) PM2.5 refers to PM2.5 that is not formed via atmospheric chemistry. Rather, direct 6 PM2.5 is emitted straight from a mobile or stationary source. With regard to the emission budget 7 presented herein, direct PM2.5 includes road dust, brake wear, and tire wear as well as PM2.5 from exhaust. 8 Through atmospheric chemistry, NOX and VOC emissions can substantially contribute to secondary PM2.5 9 formation. For this reason, NOX and VOC are considered PM2.5 precursors and are the only PM2.5 10 precursors emitted at a significant level by on-road mobile and therefore included in the MVEBs. 11 EPA's conformity regulation (40 CFR 93.124(a)) allows the implementation plan to quantify explicitly 12 the amount by which motor vehicle emissions could be higher while still demonstrating compliance with 13 the maintenance requirement. These additional emissions that can be allocated to the applicable MVEB 14 are considered the “safety margin.” As defined in 40 CFR 93.101, the safety margin represents the 15 amount of emissions by which the total projected emissions from all sources of a given pollutant are less 16 than the total emissions that would satisfy the applicable requirement for demonstrating maintenance. The 17 implementation plan can then allocate some or all of this "safety margin" to the applicable MVEBs for 18 transportation conformity purposes. 19 As presented in the TSD for on-road mobile sources, the estimated on-road mobile source emissions of 20 direct PM2.5, NOx, and VOC in 2035 for the Provo NAA, are listed in the first row (original MVEB) in 21 Table IX.A.27.11. These mobile source emissions were included in the maintenance demonstration in 22 Subsection IX.A.27.c which estimates a maximum PM2.5 concentration of 28.5 µg/m3 at the Lindon 23 monitor in 2035 within the Provo NAA portion of the modeling domain. These emissions numbers are 24 considered the MVEB for the maintenance plan prior to the application of any amount of safety margin. 25 The safety margin for the Provo NAA portion of the domain equates to 6.5 µg/m3 (the 2006 24-hr PM2.5 26 standard of 35.0 µg/m3 minus the initial 2035 FDV of 28.5 µg/m3). To evaluate the portion of safety 27 margin that could be allocated to the MVEBs, modeling was re-run for 2035 using the same emission 28 projections for point, area and non-road mobile sources with additional emissions attributed to the on-29 road mobile source (see 2nd row of Table IX.A.27.11 Additional Tons Per Day from Safety Margin). The 30 revised maintenance demonstration for 2035 still shows maintenance of the PM2.5 standard with a 31 maximum PM2.5 concentration of 29.5 µg/m3 at the Lindon monitor in 2035 within the Provo NAA 32 portion of the modeling domain. The final 2035 MVEB for the Provo NAA Metropolitan Planning 33 Organization, Mountainland Association of Governments, is listed in the last row of Table IX.A.27.11 34 along with the 2035 design value that includes the revised MVEB. 35 Direct PM2.5 NOX VOC Design Value @ controlling monitor Original MVEB .32 4.5 4.2 28.7 µg/m3 Provo, UT Maintenance Plan Section IX.A.27 40 Additional Tons Per Day from Safety Margin 1.18 2.0 2.8 -- Final 2035 MVEB 1.5 6.5 7.0 29.5 µg/m3 Table IX.A.27. 11 2035 Mountainland Association of Government’s Motor Vehicle 1 Emission Budget in Tons per Winter Weekday 2 3 It is important to note that the MVEBs presented in Table IX.A.27.9 are somewhat different from the on-4 road summary emissions inventory presented in Table IX.A.27.8. 5 Overall the emissions established as MVEBs are calculated using MOVES to reflect an average winter 6 weekday. The totals presented in the summary emissions inventory (Table IX.A.27.8), however, represent 7 an average-episode-day. The episode used to make this average (December 31, 2010 through January 10, 8 2011) includes seven such winter weekdays, but also includes two weekends. Emissions produced on 9 weekdays are significantly larger than those produced on both Saturdays and Sundays. Therefore, the 10 weighted average of daily emissions calculated for an episode-day will be less than that of a weekday. 11 There are also some conventions to be considered in the establishment of MVEBs. In particular, PM2.5 in 12 the summary emissions inventory totals includes direct exhaust, tire & brake wear, and fugitive dust. For 13 the MVEBs PM2.5 includes direct exhaust, tire & brake but no fugitive dust. VOC emissions in the 14 summary emissions inventory include refueling spillage and displacement vapor loss and are counted in 15 the on-road mobile category. MVEBs for VOC do not include these emissions because, in this context, 16 they are regarded as an area source. 17 40 CFR 93.118((b)(2)(i) also states “If the maintenance plan does not establish motor vehicle emissions 18 budgets for any years other than the last year of the maintenance plan, the conformity regulation requires 19 that a "demonstration of consistency with the motor vehicle emissions budget(s) must be accompanied by 20 a qualitative finding that there are not factors which would cause or contribute to a new violation or 21 exacerbate an existing violation in the years before the last year of the maintenance plan.” 22 Considering this, it is useful to compare the projected future design values in 2026 at all monitors in the 23 NAA to the on-road mobile emission inventory as well as the percent of the total inventory that the on-24 road mobile sector comprises. As can be seen in Table IX.A.27.9, the design values in the Provo NAA are 25 29.1 and 28.4 µg/m3. The Lindon monitor shows the highest value at 29.1 µg/m3, which is 5.9 µg/m3 26 below the standard. The on-road mobile source contribution to the overall inventory is shown in Table 27 IX.A.27.12. 28 Emissions tons/day PM2.5 NOX VOC 2026 emission inventory total 3.08 12.46 20.6 2026 on-road mobile inventory .42 5.79 4.58 On-road mobile % of total inventory 13.64% 46.47% 22.23% Table IX.A.27. 12 Comparison of 2026 On-Road Mobile Inventory to Total Emissions 29 Inventory 30 31 Provo, UT Maintenance Plan Section IX.A.27 41 Although the on-road mobile NOX contribution is almost half of the total NOX in the inventory, the 1 projected design values are so far below the standard, UDAQ is confident that there will not be any on-2 road mobile factors that will cause or contribute to a new violation of the NAAQS. 3 ii. Trading Ratios for Transportation Conformity 4 Per section 93.124 of the conformity regulations, for transportation conformity analyses using these 5 budgets in analysis years beyond 2035, a trading mechanism is established to allow future increases in on-6 road direct PM2.5 emissions to be offset by future decreases in plan precursor emissions from on-road 7 mobile sources at appropriate ratios established by the air quality model. Future increases in on-road 8 direct PM2.5 emissions may be offset with future decreases in NOx emissions from on-road mobile sources 9 at a NOx to PM2.5 ratio of 5.7 to 1 and/or future decreases in VOC emissions from on-road mobile sources 10 at a VOC to PM2.5 ratio of 28.6 to 1. This trading mechanism will only be used if needed for conformity 11 analyses for years after 2035. To ensure that the trading mechanism does not impact the ability to meet 12 the NOx or VOC budgets, the NOx emission reductions available to supplement the direct PM2.5 budget 13 shall only be those remaining after the 2035 NOx budget has been met, and the VOC emissions reductions 14 available to supplement the direct PM2.5 budget shall only be those remaining after the 2035 VOC budget 15 has been met. Clear documentation of the calculations used in the trading should be included in the 16 conformity analysis. The assumptions used to create the trading ratios can be found in the TSDs. 17 (5) Nonattainment Requirements Applicable Pending Plan Approval 18 CAA 175A(c) - Until such plan revision is approved and an area is redesignated as attainment, the 19 requirements of CAA Part D, Plan Requirements for Nonattainment Areas, shall remain in force and 20 effect. The Act requires the continued implementation of the nonattainment area control strategy unless 21 such measures are shown to be unnecessary for maintenance or are replaced with measures that achieve 22 equivalent reductions. Utah will continue to implement the emissions limitations and measures from both 23 PM2.5 SIPs. 24 (6) Revise in Eight Years 25 CAA 175A(b) - Eight years after redesignation, the State must submit an additional plan revision which 26 shows maintenance of the applicable NAAQS for an additional 10 years. Utah commits to submit a 27 revised maintenance plan eight years after EPA takes final action redesignating the Provo area to 28 attainment, as required by the Act. 29 (7) Verification of Continued Maintenance and Monitoring 30 Implicit in the requirements outlined above is the need for the State to determine whether the area is in 31 fact maintaining the standard it has achieved. There are two complementary ways to measure this: 1) by 32 monitoring the ambient air for PM2.5; and 2) by inventorying emissions of PM2.5 and its precursors from 33 various sources. 34 The State will continue to maintain an ambient monitoring network for PM2.5 in accordance with 40 CFR 35 Part 58 and the Utah SIP. The State anticipates that the EPA will continue to review the ambient 36 monitoring network for PM2.5 each year, and any necessary modifications to the network will be 37 implemented. 38 Provo, UT Maintenance Plan Section IX.A.27 42 Additionally, the State will track and document measured mobile source parameters (e.g., vehicle miles 1 traveled, congestion, fleet mix, etc.) and new and modified stationary source permits. If these and the 2 resulting emissions change significantly over time, the State will perform appropriate studies to 3 determine: 1) whether additional and/or re-sited monitors are necessary; and 2) whether mobile and 4 stationary source emission projections are on target. The State will also continue to collect actual 5 emissions inventory data from sources at thresholds defined in R307-150. 6 (8) Contingency Plan 7 CAA 175A(d) - Each maintenance plan shall contain contingency measures to assure that the State will 8 promptly correct any violation of the standard which occurs after the redesignation of the area to 9 attainment. Such provisions shall include a requirement that the State will implement all control 10 measures which were contained in the SIP prior to redesignation. 11 Upon redesignation, this contingency plan for the Provo NAA supersedes Subsection IX.A.22.9, 12 Contingency Measures, which is part of the moderate Provo NAA PM2.5 attainment SIP. 13 The contingency plan must also ensure that the contingency measures are adopted expeditiously once 14 triggered. The primary elements of the contingency plan are: 1) the list of potential contingency measures; 15 2) the tracking and triggering mechanisms to determine when contingency measures are needed; and 3) a 16 description of the process for recommending and implementing the contingency measures. 17 (a) List of Potential Contingency Measures 18 Section 175(d) of the CAA requires the maintenance plan to include as potential contingency measures all 19 of the PM2.5 control measures contained in the attainment SIP that were relaxed or modified prior to 20 redesignation. There were no control measures relaxed in the Provo NAA; however, below are potential 21 contingency measure that will be evaluated. If it is determined through the triggering mechanism that 22 additional emissions reductions are necessary, UDAQ will adopt and implement appropriate contingency 23 measure as expeditiously as possible. 24 1. Measures to address emissions from residential wood combustion (i.e. emissions from fireplaces 25 under the existing R307-302 rule), including re-evaluating the thresholds at which red or yellow 26 burn days are triggered. Residential wood combustion represents a large emissions inventory 27 source category at 43.6% of direct PM2.5 emissions in the 2017 county-wide inventory. 28 2. Measures to address fugitive dust from area sources. Fugitive dust represents 28.1% of direct 29 PM2.5 emissions in the 2017 county-wide inventory. 30 3. Additional measures to address other PM2.5 sources identified in the emissions inventory such as 31 on-road vehicles, non-road vehicles and engines, and industrial sources. These source categories 32 represent 43.2%, 8.3%, and 3.5%, respectively, of the overall 2017 baseyear emissions inventory. 33 In addition, UDAQ administers incentive and grant programs that reduce emissions in Utah’s NAAs. 34 The emissions reductions are not included in the quantitative maintenance demonstration; however, 35 they are expected to contribute to the mitigation of PM2.5 concentrations. Generally speaking, the 36 programs target Utah nonattainment areas. The programs include approximately $25.5 million from 37 the Volkswagen settlement and approximately $12.7 million to replace heavy-duty diesel trucks and 38 Provo, UT Maintenance Plan Section IX.A.27 43 buses that are operating under old emissions standards. Nonroad diesel upgrades will see 1 approximately $1.3 million on the Wasatch Front. Another $3.8 million of the Volkswagen funding 2 will go towards installing electric vehicle supply equipment in Utah. UDAQ is in the process of using 3 approximately $9.6 million in federal funding to implement wood stove changeout programs 4 throughout the three Utah PM2.5 NAAs. 5 (b) Tracking 6 The tracking plan for the three NAAs consists of monitoring and analyzing ambient PM2.5 concentrations. 7 In accordance with 40 CFR 58, the State will continue to operate and maintain an adequate PM2.5 8 monitoring network in SLC, Provo, and Logan NAAs. 9 (c) Triggering 10 Triggering of the contingency plan does not automatically require a revision to the SIP, nor does it mean 11 that the area will automatically be redesignated once again to nonattainment. Instead, the State will have 12 an appropriate timeframe to correct the potential violation with implementation of one or more adopted 13 contingency measures. In the event that violations continue to occur, additional contingency measures 14 will be adopted until the violations are corrected. 15 Upon notification of a potential violation of the PM2.5 NAAQS, the State will develop appropriate 16 contingency measures intended to prevent or correct a violation of the PM2.5 standard. Information about 17 historical exceedances of the standard, the meteorological conditions related to the recent exceedances, 18 and the most recent estimates of growth and emissions will be reviewed. The possibility that an 19 exceptional event occurred will also be evaluated. 20 Upon monitoring a potential violation of the PM2.5 NAAQS, including exceedances flagged as exceptional 21 events but not concurred with by EPA, the State will identify a means of corrective action within six 22 months after a potential violation. The maintenance plan contingency measures will be chosen based on a 23 consideration of cost-effectiveness, emission reduction potential, economic and social considerations, or 24 other factors that the State deems appropriate. 25 The State will require implementation of such corrective action no later than one year after the violation is 26 confirmed. Any contingency measures adopted and implemented will become part of the next revised 27 maintenance plan submitted to the EPA for approval. 28