DAQ-2025-002103

March 31, 2025

Ms. Ana Williams
Ms. Jazmine Lopez
Utah Division of Air Quality
P.O. Box 144820
Salt Lake City, Utah 84114-4820

Submitted by email to [e-mail] and [e-mail]

Subject: Utah Petroleum Association Comments on Amendment to Section R307-110-17.
Section IX, Control Measures for Area and Point Sources, Part H, Emission Limits; and
Amendments to Utah State Implementation Plan, Section IX.H.11 and Section IX.H.12:
Emission Limitations and Operating Practices

Dear Ms. Williams and Ms. Lopez:

Thank you for the opportunity for the Utah Petroleum Association (“UPA”) to submit these
comments on the Utah Division of Air Quality (“UDAQ”) proposed Amendment to Section R307-
110-17. Section IX, Control Measures for Area and Point Sources, Part H, Emission Limits; and
Amendments to Utah State Implementation Plan, Section IX.H.11 and Section IX.H.12: Emission
Limitations and Operating Practices, as approved by the Air Quality Board (“AQB”) for proposal
on February 5, 2025,1 and published as a proposed rule in the Utah Bulletin on March 1, 2025.2

UPA is a statewide oil and gas trade association established in 1958 representing companies
involved in all aspects of Utah’s oil and gas industry. UPA members range from independent
producers to midstream and service providers, to major oil and natural gas companies widely
recognized as industry leaders responsible for driving technology advancement resulting in
environmental and efficiency gains. Five member companies each operate a petroleum refinery
in the Salt Lake City Serious PM2.5 nonattainment area (“SLC NAA”), of which four are addressed
in this proposal. Additionally, UPA member companies operate oil and gas production and
midstream facilities within the Uintah Basin ozone nonattainment area (“NAA”). Thus, our
member companies have an interest in air quality and air emission controls throughout Utah.

1 See “Feb 5 2025 Packet.pdf”, at https://www.utah.gov/pmn/files/1230147.pdf, agenda item VIII
(“Proposed SIP Revision”). Minutes of meeting not published at the time of this writing.
2 Utah State Digest, March 01, 2025, Vol. 2025, No. 05, at 13.
UPA Comments on Proposed PM2.5 SIP Revisions, March 31, 2025
Page 2 of 14

The Environmental Protection Agency (“EPA”) established the 2006 24-hour PM2.5 National
Ambient Air Quality Standard (“NAAQS”) in October 2006.3 A ruling of the D.C. Circuit Court in
January 2013 established that PM2.5 implementation must follow the requirements of the Clean
Air Act (“CAA”) Title I Part D Subpart 4 instead of the general requirements in Subpart 1 that EPA
previously applied to PM2.5 nonattainment areas.4 In June 2014, EPA classified the Salt Lake City
area5 as Moderate nonattainment under the 2006 24-hour PM2.5 NAAQS.6 In May 2017, EPA
determined that the SLC NAA failed to attain by the attainment date and reclassified it to Serious
nonattainment.7 In September 2019, EPA issued a Clean Data Determination (“CDD”) for the
SLC NAA, declaring that the area had attained the NAAQS.8 Finally, in November 2020, EPA
proposed to redesignate the SLC NAA to attainment.9 EPA to-date has not finalized this proposal,
but instead, UPA understands that EPA requested changes and/or clarifications to the SIP
requirements for major source petroleum refineries within the SLC NAA. UDAQ therefore
proposed SIP revisions that are subject of this letter and responsive to EPA’s questions.

UPA thanks UDAQ for working closely with our member company petroleum refineries to resolve
EPA’s questions. We also appreciate that UDAQ worked closely with EPA to understand the
questions. This has been a major effort over the course of several years, culminating in the
proposal.

Most importantly, UPA is pleased to see the SLC NAA back on the regulatory path to
redesignation to attainment. UPA remains concerned that Utah can complete this revision
to the PM2.5 Serious State Implementation Plan (“SIP”) in time for EPA to complete its
notice and comment rulemaking for the redesignation of the SLC NAA to attainment with
a final federal rule published no later than December 31, 2025.

UPA has other concerns, summarized as follows:

• The controls already in place for petroleum refineries and those prescribed for petroleum
refineries meet Best Available Control Technology and no additional changes to controls
beyond those proposed are warranted.

3 FR 71 61144, National Ambient Air Quality Standards for Particulate Matter, Final Rule, October 17,
2006.
4 D.C. Circuit Court , Natural Resources Defense Council and Sierra Club v. Environmental Protection
Agency, No. 08-1250, decided January 4, 2013.
5 The Salt Lake City PM2.5 nonattainment area includes the full counties of Davis and Salt Lake and partial
counties of Box Elder, Tooele, and Weber. See 40 CFR 81.345 Utah—2006 24-Hour PM2.5 NAAQS
[Primary and Secondary].
6 79 FR 31566, Identification of Nonattainment Classification and Deadlines for Submission of State
Implementation Plan (SIP) Provisions for the 1997 Fine Particle (PM2.5) National Ambient Air Quality
Standard (NAAQS) and 2006 PM2.5 NAAQS, Final Rule, June 2, 2014 (“Classification and Deadline
Rule.”
7 82 FR 21711, Determinations of Attainment by the Attainment Date, Determinations of Failure To Attain
by the Attainment Date and Reclassification for Certain Nonattainment Areas for the 2006 24- Hour Fine
Particulate Matter National Ambient Air Quality Standards, Final Rule, May 10, 2017.
8 84 FR 51055, Clean Data Determination; Salt Lake City, Utah 2006 Fine Particulate Matter Standards
Nonattainment Area, Final Rule, September 27, 2019 (“CDD”).
9 85 FR 71023, Approval and Promulgation of Implementation Plans; State of Utah; Salt Lake City and
Provo, Utah PM2.5 Redesignations to Attainment and Utah State Implementation Plan Revisions,
Proposed Rule, November 6, 2020 (“Proposed Redesignation”).
UPA Comments on Proposed PM2.5 SIP Revisions, March 31, 2025
Page 3 of 14

• Petroleum refinery data needs to be annotated on each page of data, not just the cover
sheet, with a watermark to ensure that anyone accessing it will understand its context.
• UPA requests that the wording for frequency of stack tests subsequent to the initial test
revert back to the prior wording to provide appropriate flexibility.
• The SIP provides appropriate enforceable emission limitations as written and the
limitations do not need to be expressed as mass per unit of time.
• Although some may claim that the air quality in the SLC NAA is not good, the monitoring
data continue to meet the PM2.5 24-hour NAAQS and will likely continue to improve.
• Although UPA is not seeking for UDAQ to incorporate precursor demonstrations into this
SIP at this time, should additional controls be added to the SIP beyond those included in
the proposal, UPA may reevaluate its position.

Our comments address some of the comments submitted by other entities during EPA’s comment
period on the Proposed Redesignation and to UDAQ during SIP development.10

The following sections of this letter detail all of our concerns and comments about the proposed
PM2.5 SIP update for the SLC NAA.

1. We recommend completing these updates to the PM2.5 SIP for the SLC NAA
as expeditiously as practicable and supporting EPA’s timely approval and
redesignation of the NAA to attainment.

The CAA requires that a request to redesignate a NAA to attainment include a maintenance plan
that provides for maintenance of the applicable NAAQS for at least ten years after redesignation.11
The Calcagni Memo reinforces that the CAA requires the maintenance plan to provide a
demonstration of maintenance for at least ten years after EPA redesignates the area, not ten
years after the State submits the redesignation request.12

When UDAQ originally submitted the Serious PM2.5 SIP Maintenance Plan, it included a
demonstration that the area would continue to attain until ten years after the latest expected date
that EPA would redesignate the area to attainment, that is, through the year 2035.13 Thus, in
order for the ten year maintenance requirement to be fulfilled, EPA must finalize the redesignation
no later than the end of 2025. Additional steps that must be completed for the final redesignation
include UDAQ to finalize this set of revisions for the PM2.5 SIP and for EPA to evaluate the revision,
propose redesignation, accept and evaluate comments on the redesignation, and finalize the
redesignation, all by the end of 2025.

If EPA does not finalize the redesignation by the end of 2025, UDAQ will need to update the
maintenance plan including the modeling to show ongoing attainment through the end of a future
year, later than 2035, allowing additional years for UDAQ to complete the update and for EPA to

10 See comments submitted in Docket Number EPA–R08–OAR–2020–0098 at regulations.gov. Also, see
comments submitted to UDAQ as an attachment to the SIP submittal to EPA.
11 CAA §175A(a).
12 Memorandum, John Calcagni, Director of Air Quality Management Division, to EPA Directors for Air in
EPA Regions, Procedures for Processing Requests to Redesignate Areas to Attainment, September 4,
1992 (“Calcagni Memo”).
13 Utah State Implementation Plan, PM2.5 Maintenance Provisions for the Salt Lake City, UT
Nonattainment Area, Section IX.A.36 at 39, attached to EPA with Letter, Gary R. Hebert, Governor of
Utah, to Gregory Sopkin, Regional Administrator for EPA Region 8, Utah Area Designation
Recommendations for the 2006 24-Hour PM2.5 NAAQS, January 2, 2020.
UPA Comments on Proposed PM2.5 SIP Revisions, March 31, 2025
Page 4 of 14

review and approve it. More than four years have passed since EPA issued the CDD, based on
air quality data from 2016 through 2018. The road to redesignation has been long.

We understand that EPA agreed to parallel process the SIP revision and redesignation in an effort
to meet the end of 2025 date. We support parallel processing to complete timely redesignation.
We urge UDAQ to work closely with EPA to ensure that EPA can meet the deadline. We also
ask UDAQ if there is anything additional that UPA can do to support the process. Should anything
arise where UPA or its member companies can provide additional support, please advise us as
quickly as possible.

2. The controls prescribed for petroleum refineries meet the definition of Best
Available Control Technology. No additional changes beyond those
included in the proposed SIP should be made.

The EPA implementation rule for PM2.514 defines Best Available Control Measures (“BACM”) as
follows and includes Best Available Control Technology (“BACT”) within the definition of BACM:

Best available control measure -- (BACM) is any technologically and economically feasible
control measure that can be implemented in whole or in part within 4 years after the date of
reclassification of a Moderate PM2.5 nonattainment area to Serious and that generally can
achieve greater permanent and enforceable emissions reductions in direct PM2.5 emissions
and/or emissions of PM2.5 plan precursors from sources in the area than can be achieved
through the implementation of RACM on the same source(s). BACM includes best available
control technology (BACT).15

Thus, the definition includes the concepts of technological and economic feasibility.

UDAQ provides the following approach to perform a BACT analysis for the PM2.5 SIP, consistent
with EPA’s approach:

The BACT analysis for each emission unit includes the five steps in a top-down BACT
analysis. The first step identifies control options. DAQ evaluated various resources to
identify the various controls and emission rates. These include, but are not limited to,
federal regulations, Utah regulations, regulations of other states, RACT/BACT/LAER
clearinghouse (RBLC), issued permits, and emission unit vendors.

The second step in the BACT analysis eliminates the technological infeasible controls.
The remaining control options are ranked in the third step of the BACT analysis.
Combinations of various controls are also included. The fourth step of the BACT analysis
evaluates the economic feasibility of the highest ranked option. This evaluation includes
energy, environmental, and economic impacts of the control option.

The fifth step in the BACT analysis selects the “best” option. This step also includes the
necessary justification to support the DAQ’s decision. The DAQ has included the time it
will take for a source to implement the selected control if the control is not already being
implemented at the source.16

14 40 CFR Part 51 Subpart Z.
15 40 CFR 51.1000 “Best available control measure.”
16 See Control Strategies: Serious Area PM2.5 SIP located at https://deq.utah.gov/air-quality/control-
strategies-serious-area-pm2-5-sip (accessed on March 30, 2025).
UPA Comments on Proposed PM2.5 SIP Revisions, March 31, 2025
Page 5 of 14

a. Utah’s petroleum refineries are highly controlled and meet BACT.

The petroleum refineries in the SLC NAA each submitted a BACT analysis according to the above
5-step approach, then UDAQ performed a rigorous BACT determination and updated it in 2025,
concluding that the SIP had appropriate controls required to meet the stringent BACT
requirements. In this SIP update, UDAQ replaced the plant-wide caps for petroleum refinery
emissions with enforceable emission limits on equipment or processes and added test
requirements.

It is not surprising that the Serious PM2.5 SIP requires few additional controls. Utah’s petroleum
refineries are highly controlled due to decades of air permitting with associated BACT
requirements, compliance with numerous applicable EPA standards including the recent 2015
extensive revisions to petroleum refinery requirements,17 compliance with Utah rules, and
compliance with various prior SIPs including the Moderate PM2.5 SIP for the SLC NAA and the
recent Moderate ozone SIP for the Northern Wasatch Front ozone nonattainment area (“NWF”).

In fact, in establishing Reasonably Available Control Technology (“RACT”) for the Moderate PM2.5
SIP for the SLC NAA, UDAQ appropriately encouraged petroleum refineries to agree to controls
that would exceed RACT based on the possibility that the NAA would be reclassified to Serious
and require BACT. UDAQ wrote:

UDAQ always made clear to [listed sources] that they should view potential controls as
being “better than RACT” and to “focus on BACT-level controls.” UDAQ knew and
explained that potentially revisiting this issue with the possibility of replacing “just installed”
controls would be an expensive and unpopular undertaking – so better to focus on the
higher level of control from the outset.18

The nature of wintertime inversions enhances PM2.5 formation in the SLC NAA. Outside of
uncontrollable wildfire events typically in the summertime and early fall and which are potentially
subject to exclusion from the monitoring data via wildfire exceptional events, the PM2.5
nonattainment has been entirely a wintertime situation. Yet, the petroleum refineries willingly
agreed to apply PM2.5 controls year-round, outside of the wintertime PM2.5 “season.” Although
refineries agreed to year-round controls, extension of the controls beyond the PM2.5 season may
be considered as beyond BACT.

Outside of the PM2.5 SIP controls, Utah petroleum refineries voluntarily elected to make and sell
Tier 3 gasoline for the area instead of applying credits generated elsewhere, outside of Utah, as
allowed in EPA’s gasoline standards,19 thus reducing motor vehicle emissions in the SLC NAA.
The combination of on-road and non-road motor vehicle emissions comprises the largest part of
the VOC and NOx emission inventories and Tier 3 gasoline reduces VOC and NOx emissions
from onroad motor vehicles. For the NWF Moderate ozone SIP, UDAQ estimated that the
combination of Tier 3 gasoline with Tier 3 vehicles would reduce summertime onroad motor

17 80 FR 75178, Petroleum Refinery Sector Risk and Technology Review and New Source Performance
Standards.
18 UDAQ SIP submittal to EPA, Section IX, Part H, Public Comments and Responses, July 1-August 15,
2018 (“Response to Comment”) at 160/C0056. Located in the docket for the redesignation at
https://www.regulations.gov/document/EPA-R08-OAR-2020-0098-0009, attachment labeled “5. Public
Comments and Responses.”
19 40 CFR Part 1090 Subpart H.
UPA Comments on Proposed PM2.5 SIP Revisions, March 31, 2025
Page 6 of 14

vehicle emissions by 25% for VOC and 36% for NOx from 2017 to 2023 despite the population
increase.20 While the effect of Tier 3 gasoline cannot be easily separated from the effect of Tier
3 vehicles because the two were designed to work together for the best emissions performance,
EPA reports that Tier 3 gasoline lowers emissions from older motor vehicles built to standards
earlier than Tier 3, as well.21 The Tier 3 effect in the wintertime PM2.5 season would not be the
same as the summertime effect but likely still substantial.

Therefore, we conclude that the Utah petroleum refineries meet BACT requirements and go
beyond BACT with refinery agreements to year-round refinery emission controls and
commitments to Tier 3 gasoline.

b. Additional control suggestions from others do not provide
appropriate suggestions for BACT.

Non-Governmental Organizations (NGOs) submitted comments on UDAQ’s original BACT
determinations to both UDAQ22 and to EPA.23 NGOs reiterate many of the same arguments in
both letters, despite UDAQ providing painstakingly thorough responses to the comments.

• NGOs stated that each emission unit must have an emission limitation instead of the plant-
wide caps in the original SIP. They also indicated that the SIP did not adequately address
a need for more stack testing. As noted above, UDAQ resolved these items by removing
the plant-wide caps from the SIP and increasing the number of stack tests.

• In some cases, NGO comments relied on controls prescribed in a Prevention of Significant
Deterioration (“PSD”) permit for the Arizona Clean Fuels Yuma petroleum refinery in
Arizona and the Meridian Davis Refinery in North Dakota. Controls from these permits do
not fit the technological and economic feasibility requirements of BACT. The Arizona
refinery was never built so the controls have not been shown to be either economically or
technologically feasible. Construction on the Meridian Davis refinery has not completed
so these controls also have not been shown to be either economically or technologically
feasible. Thus, controls specified for these two refineries are not yet demonstrated in
practice and cannot be used as the basis for BACT for the SLC NAA PM2.5 SIP.

Furthermore, the cost to install a control on a new facility such as the Yuma or Meridian
refineries – grass roots facilities -- likely would be far less than the cost to retrofit the control
onto an existing facility such as Utah’s petroleum refineries. And, the incremental tons of
emissions reduced for the alternative control on an existing facility with some emissions

20 Utah Division of Air Quality, State Implementation Plan, 2015 Ozone NAAQS Northern Wasatch Front
Moderate Nonattainment Area, 2024, Section IX. Part D.11. Data from Tables 7 and 8, pages 26 and 27,
respectively.
21 EPA Fact Sheet, EPA Sets Tier 3 Tailpipe and Evaporative Emission and Vehicle Fuel Standards,
March 2014, located at https://www.epa.gov/regulations-emissions-vehicles-and-engines/final-rule-
control-air-pollution-motor-vehicles-tier-3 (accessed on March 31, 2025).
22 Letter, Heal Utah, Western Resource Advocates, Utah Physicians for a Healthy Environment, and the
Utah Chapter of the Sierra Club to Utah Division of Air Quality, Comments on Section IX Part H of the
Utah State Implementation Plan, August 15, 2018.
23 Letter, Heal Utah, Western Resource Advocates, and the Utah Chapter of the Sierra Club to Air and
Radiation Division, EPA Region 8, Docket ID No. EPA–R08–OAR–2020–0098, Proposed Approval and
Promulgation of Implementation Plans; State of Utah; Salt Lake City and Provo, Utah PM2.5
Redesignations to Attainment and Utah State Implementation Plan Revisions, December 7, 2020.
UPA Comments on Proposed PM2.5 SIP Revisions, March 31, 2025
Page 7 of 14

controls already in place would be much less than the tons reduced for a new facility with
no controls. Thus, the cost expressed as dollars per ton of emissions reduced on the
new facility would be far less (fewer dollars divided by more tons) than that for retrofitting
a facility that already has other control(s) in place.

• NGOs also claimed that there is no justification to eliminate a control technology from the
BACT analysis as being technologically infeasible if it cannot be designed, installed, and
in operation by December 31, 2018, as at least one Utah petroleum refiner did in their
BACT analysis. NGOs claim that controls may be considered as BACT if installed as late
as four years after the date of reclassification to Serious, which they say is May 10, 2021.
UPA does not agree with the NGOs. The attainment date for the SLC NAA is December
31, 2019, and controls not in place by the wintertime of 2019, i.e., the PM2.5 season of the
last year of the three calendar year period of data used to determine the design value for
the attainment date -- in this case the early months of 2017 through 2019 -- will have no
bearing on attainment. Accordingly, even though BACT for PM2.5 SIPs is generally
independent of attainment, the PM2.5 SIP implementation rule requires the controls to be
installed by December 31, 2018, to be operational by the beginning of the year containing
the applicable attainment date, which would be the day after December 31, 2018, in other
words to be operational by January 1, 2019:

Required timeframe for obtaining emissions reductions. For each Serious
nonattainment area, the attainment plan must provide for implementation of all
control measures needed for attainment as expeditiously as practicable. All control
measures must be implemented no later than the beginning of the year
containing the applicable attainment date, notwithstanding BACM
implementation deadline requirements in § 51.1010.24 [emphasis added]

Therefore, using a latest installation date of December 31, 2018, as one measure of the
control being technologically feasible entirely fits within the SIP requirements.

• NGOs complained that refineries did not evaluate the cost-effectiveness of Selective
Catalytic Reduction (“SCR”) in combination with Ultra Low NOx burners. However, if one
of the controls is not feasible, both controls in combination would also not be feasible.

• NGOs cast doubt on cost-effectiveness and schedule determinations for some equipment
considered in the BACT analyses. The NGOs are in no position to cast such doubt as
they do not know the peculiarities of any specific equipment in a particular petroleum
refinery that would increase costs and/or extend schedules. Such peculiarities include but
are not limited to plot space, availability of connections to tie into the process, availability
of required utilities, process unit turnaround schedule, metallurgy required for the specific
service, and others. These issues all contribute to retrofit costs typically being greater
than new facility costs.

• NGOs claim that Maximum Achievable Control Technology (“MACT”) requirements should
be imposed on petroleum refinery flaring operations. Petroleum refineries already must
comply with applicable MACT requirements including flaring requirements and, therefore,
it would not be appropriate to add this to the PM2.5 SIP. Furthermore, MACT standards
are designed to control Hazardous Air Pollutants (“HAPs”) and not the criteria pollutants

24 40 CFR 51.1011(c)(5).
UPA Comments on Proposed PM2.5 SIP Revisions, March 31, 2025
Page 8 of 14

of PM2.5 and its precursors of NOx, SO2, or volatile organic compounds (“VOC”).
Additionally, ammonia, the remaining precursor to PM2.5 formation, is not a HAP.

• NGOs state that BACT has not been applied consistently “in emissions assumptions and
cost effectiveness analyses for similar controls at the four refineries. DAQ must evaluate
and implement BACT for the refineries consistently.” This argument, however, ignores
fundamental differences in the four petroleum refinery configurations, process units,
equipment within process units, constraints such as plot space and age of existing
equipment, existing emissions controls, and other factors. BACT will not be the same for
the four refineries. NGOs claim that BACT must be evaluated consistently, but the
refineries are not consistent with each other.

This list is not exhaustive but provides some of the highlights of our concerns with comments
submitted by NGOs. UDAQ addressed these and other comments from the NGOs in the
Response to Comment.25 UPA supports UDAQ’s responses to the NGO comments.

Thus, we conclude that NGO comments for petroleum refinery BACT should not alter the BACT
determination. No additional changes to BACT beyond those included in the proposed SIP should
be made.

3. UPA requests that a watermark be placed on each page of petroleum
refinery data to explain the context.

UPA requests that a watermark be placed on each page of the petroleum refinery data supplied
and included in appendices to UDAQ’s updated BACT determinations, to explain the context. The
watermark should state:

Note: All data in this document is in raw, unprocessed form and includes periods of monitor
downtime, quality assurance, calibration, maintenance, out of control periods, potential
malfunctioning CEMs data, and exempt periods.

Although this statement has been provided on the cover page of applicable appendices, UPA
does not consider this to be adequate because cover pages may not be read, if read they may
be forgotten, and they can easily be separated from the pages of data.

4. UPA requests that all subsequent stack tests after the initial test retain the
original wording for the due dates, i.e., once every X number of years
thereafter.

Previously in Part H.12. of the SIP, the stack test dates required subsequent tests after the initial
test to be conducted once every X number of years thereafter. This timing gave petroleum
refineries some measure of flexibility. For example, if a refinery did one test in July, the following
test could be done in June, July, August, or September, and if then done in August, the tests the
year after that could still be done in June, July, August, or September. This was especially helpful
to avoid the safety hazards associated with wintertime stack testing.

Part H.12. has been changed so that subsequent stack test dates now would be required at least
once every X years from the date of the last stack test. This wording loses flexibility for petroleum

25 See Response to Comment starting at 159/C0055.
UPA Comments on Proposed PM2.5 SIP Revisions, March 31, 2025
Page 9 of 14

refineries, but more importantly results in a need to either conduct some tests on a far shorter
cycle or conduct tests in the winter months, which would pose safety hazards. For example, if a
test were conducted in April of one year, the next test would necessarily need to be scheduled for
March or February of the test year, and the following test after that would need to be scheduled
for December or January. Scheduling for an earlier month provides a necessary margin of time
to accommodate the possibility of a need to reschedule based on any number of unforeseen
circumstances such as illness of a stack tester or key refinery personnel for the test, travel
difficulties, inclement weather including snow or thunderstorms, process unit upsets or difficulties
achieving the appropriate operating requirements for the test, or any number of other situations.
Some tests might need to be conducted several months early to avoid wintertime testing.

Therefore, UPA requests that all subsequent stack tests after the initial test retain the original
wording for the due dates, i.e., once every X number of years thereafter.

If UDAQ has concerns that a facility might do one test in December and a subsequent test in
January of the following year, thus spacing the tests too closely together, then a minimum time
between tests could be specified. For example, rules for leak detection and repair (“LDAR”) for
oil and gas sources already require this as in R307-509-4(1)(d)(ii):

Semiannually after the initial monitoring survey. Consecutive semiannual monitoring surveys
shall be conducted at least four months apart and no more than seven months apart.

EPA’s National Emission Standards for Hazardous Air Pollutants (“NESHAPS”) for Major
Sources: Industrial, Commercial, and Institutional Boilers and Process Heaters rule provides
another example of boundaries on stack test timing. This regulation sets the maximum time
between tests.26

UDAQ could use these examples to retain original stack test wording, “conducted once every X
years”, and thereby retain scheduling flexibility and avoid the need for a refinery to consider
wintertime testing or to shorten the frequency by several months to avoid wintertime testing. By
adding a statement about minimum months between tests, this would resolve any concerns about
a test being conducted too soon after the prior test.

UDAQ also requests that the SIP allow petroleum refineries to consider adding a Continuous
Emissions Monitoring System (“CEMS”) in the future, as a future alternative to stack testing.

5. The proposed SIP revision provides enforceable emission limitations as
required and does not need to have emission limitations expressed in units
of mass per unit time.

An additional concern expressed by NGOs was that protection of short term ambient standards
requires limits to be written in pounds per hour. The Proposed SIP Revision includes numerical
limits and work practice limits for petroleum refinery process units and equipment, included as
general requirements for all major source petroleum refineries in the PM2.5 SIP Part H.11.g,
Petroleum Refineries, and for individual major source petroleum refineries in the PM2.5 SIP Part
H.12. SIP limits do not need to be expressed as mass per unit time and, for example, the SIP
general limits include satisfactory limits as follows:

26 40 CFR Part 63 Subpart DDDDD §63.7515.
UPA Comments on Proposed PM2.5 SIP Revisions, March 31, 2025
Page 10 of 14

• Fluid Catalytic Cracking Unit (“FCCU”) SO2 expressed as parts per million by volume, dry
(“ppmvd”)
• FCCU particulate matter (“PM”) limits expressed as 1.0 pounds PM per 1000 pounds coke
burn-off [sic]
• Leak Detection and Repair limits expressed as 500 parts per million (“ppm”)
• Uncontrolled emission limits for tank degassing emissions expressed as a percent of the
lower explosive limit (“LEL”) within the tank
• Emission limits on combustion units require operating in accordance with good
combustion practices and maintaining all combustion units following the manufacturer’s
recommendations

Although the CAA requires that SIPs (including maintenance plans) include enforceable emission
limitations,27 nothing prescribes that these limitations be expressed in units of mass (or volume)
of emissions per unit time. Therefore, all emission limitations in the SIP meet the CAA
requirement of being enforceable emission limitations. We support the emission limits as
currently stated in the proposed SIP and would not support any additional changes should NGOs
submit this comment again.

6. Air quality for PM2.5 within the SLC NAA is very good and will continue to improve.
No decisions should be made on unfounded fears of higher PM2.5 ambient
concentrations.

UDAQ indicated that all monitors attained the standards for both the 2006 24-hour and 2024
annual PM2.5 NAAQS.28

The 24-hour PM2.5 NAAQS is met when the 98th percentile 24-hour concentration, as determined
in accordance with appendix N to 40 CFR Part 50, is less than or equal to 35 µg/m3.29 From the
time period of the CDD, 2016 to 2018, through 2024, the SLC NAA design values remained within
the level of the 2006 24-hour PM2.5 NAAQS, 35 µg/m3, as shown in Figure 1.30 Moreover, 24-
hour PM2.5 design values reduced by 15.4 µg/m3 or 31% since 2006 when EPA issued the 24-
hour PM2.5 NAAQS. This dramatic reduction can only result from permanent and enforceable
emission reductions.

27 See CAA §110(a)(2)(A). See also CAA §172(c)(6) which requires that nonattainment SIPs “shall
include enforceable emission limitations, and such other control measures, means or techniques
(including economic incentives such as fees, marketable permits, and auctions of emission rights), as well
as schedules and timetables for compliance, as may be necessary or appropriate to provide for
attainment of such standard in such area by the applicable attainment date specified in this part.”
28 February 2025 Stakeholder Slides.
29 40 CFR 50.20(c).
30 Data through 2023 obtained from EPA design value spreadsheets located at https://www.epa.gov/air-
trends/air-quality-design-values. Data for 2024 obtained from UDAQ presentation, PM2.5 Update,
presented to the Industry Stakeholder Meeting held on February 5, 2025 (“February 2025 Stakeholder
Slides”.
UPA Comments on Proposed PM2.5 SIP Revisions, March 31, 2025
Page 11 of 14

Figure 1. 24-Hour PM2.5 Design Values for 3-Year Period Ending in Year Indicated, µg/m3
The increase in design values for years starting with 2021 resulted from greater smoke impacts
in 2021. UDAQ flagged five wildfire exceptional event days at the Copper View monitor for ozone
in July and August 2021 due to wildfires in California, Oregon, Washington, and Idaho. UDAQ
further noted that Utah had 80 smoke days in 2021 and the wildfire smoke conditions were
“exceptionally abnormal” compared to historical conditions.31

Figure 1 shows a small downturn for the 2024 24-hour PM2.5 design value although not as
significant as might be expected with the exceptionally high wildfire year of 2021 no longer in the
3-year average for the design value calculation. Figure 2 shows a larger downturn for the annual
design value for 2024. Although not as significant as 2021, it appears that 2024 may also have
had significant wildfire influence during the summer months of third quarter. Table 1 shows that
third quarter average PM2.5 levels in Salt Lake County dropped significantly during 2022 and 2023
compared to 2021, and then the 2024 level rose again although not as high as 2021. The number
of ozone exceedances per year, 2021 through 2024, falls and rises again with the third quarter
PM2.5 average levels.32 This analysis is not meant to be exhaustive to fully explore wildfire effects
or possibilities but does present some data towards explaining the trend in Figure 1.

Table 1. Average 3rd Quarter PM2.5 and Annual Ozone Exceedance
Comparison, Salt Lake County
Year 3rd Quarter Average
PM2.5, µg/m3
Number of Ozone
Exceedances in Year
2021 16.9 42
2022 8.7 18
2023 7.9 13
2024 11.7 26

31 Northern Wasatch Front Nonattainment Area, 2015 Ozone NAAQS, Clean Air Act 179B(b)
Demonstration; APPENDIX II: Exceptional Events and Data Modification Demonstration, prepared by
Utah Department of Environmental Quality, Utah Division of Air Quality (2024).
32 Data obtained from EPA’s Air Data Website, using the Daily Data Download and the Ozone
Exceedances Plot. Website located at https://www.epa.gov/outdoor-air-quality-data.
UPA Comments on Proposed PM2.5 SIP Revisions, March 31, 2025
Page 12 of 14

Although not relevant to this PM2.5 SIP update for the 24-hour PM2.5 NAAQS but presented here
as further evidence of declining PM2.5 concentrations that attain the NAAQS, we present Figure
2, a graph of the annual PM2.5 design values in Salt Lake County. The primary annual
PM2.5 standard is met when the annual arithmetic mean concentration, as determined in
accordance with appendix N to 40 CFR Part 50, is less than or equal to 9.0 µg/m3.33

Figure 2. Annual PM2.5 Design Values for 3-Year Period Ending in Year Indicated, µg/m3
Again, the graph shows attainment throughout recent years despite the effect of the wildfire
smoke impacts in 2021 for the design values that include data from 2021. In this case, the graph
shows a strong downward trend in 2024.

PM2.5 concentrations in ambient air in the SLC NAA will likely reduce more in the near future.
Controls for the Moderate NWF ozone SIP, which are year-round controls, serve to lower VOC
and NOx emissions which are precursors for both ozone and PM2.5. Some of these controls are
not in place yet but will soon drive emission reductions. Ongoing vehicle fleet turnover will
continue to reduce emissions and Tier 3 gasoline will yield even greater benefits as more Tier 3
vehicles replace older vehicles over time. UDAQ is also working on Serious ozone SIP controls
now, which will again reduce emissions and will further drive PM2.5 concentrations as well as
ozone concentrations lower.

We conclude that the SLC NAA attains the PM2.5 NAAQS and air quality will continue to improve.
Therefore, no decisions should be made on the basis of unfounded concerns about future
attainment.

7. Precursor demonstrations show that no additional controls would be needed in the
Serious PM2.5 SIP for major stationary sources because, at current emission
limitations, they do not have a significant effect on PM2.5 concentrations.

Both the CAA and EPA’s rules governing PM2.5 SIPs allow the use of demonstrations to show that
major stationary sources do not contribute significantly to PM2.5 levels which exceed the standard
in the area.34

33 40 CFR 50.20(b).
34 CAA §189(e) and 40 CFR Part 51 §51.1006.
UPA Comments on Proposed PM2.5 SIP Revisions, March 31, 2025
Page 13 of 14

UPA provided a precursor demonstration for NOx, sulfur oxides (SOx), VOC, and ammonia,
including photochemical modeling performed by Ramboll and submitted these as part of its
comments to UDAQ on the original Serious PM2.5 SIP.35 Ramboll developed the demonstration
and performed the modeling according to EPA draft guidance for the demonstrations,36
addressing both contribution-based and sensitivity-based analyses and employed UDAQ’s
January 2011 photochemical modeling platform supporting the development of the SLC NAA
SIP.37 Thus, the demonstration is directly relevant and parallel to UDAQ’s technical analyses.

The precursor demonstration concluded:

Estimated 2019 NOx contributions (via zero-out) to daily 24-hour PM2.5 at each monitoring
site fell well below the significance threshold of 1.5 μg/m3. Estimated 2019 SOx
contributions from major stationary sources approached the 1.5 μg/m3 threshold on a
single episode day at one site, but remained less than the threshold over all sites and
days. [Ramboll] also performed SOx sensitivity reductions of 30 to 70% for major
stationary sources, which showed PM2.5 sensitivity well below the threshold. Estimated
2019 VOC contributions from major stationary sources also resulted in PM2.5 impacts
orders of magnitude smaller than the threshold. Finally, the model also indicated no 2019
ammonia contributions from major stationary sources above the threshold.

In UDAQ’s response to comment submitted to EPA as part of its original Serious PM2.5 SIP
submittal, UDAQ stated, “There are likely some things we would do somewhat differently, but
given the conservative nature of the concentration based demonstrations, it appears that the
conclusions would probably remain much the same.”38

Although UPA is not currently advocating that UDAQ reconsider the use of precursor
demonstrations, we would reconsider our position should additional controls be considered
beyond those contained in the proposed PM2.5 SIP. The precursor demonstration shows that the
controls would provide little benefit from an industry that contributes a relatively small amount to
the overall NAA emissions inventory.

8. Conclusion and Closing

In closing, UPA only requests two changes to the proposed SIP, to retain original wording for
timing of subsequent stack tests and to apply a watermark to each page of data submitted and
included as appendices in the 2025 Addendums to the BACT determinations. In all other aspects,
we do not support any other changes except as submitted by individual refineries addressing their
own refinery such as corrections in the BACT determination documents for their refinery and are
beyond the scope of this letter.

35 Letter, Jennette King to Bryce Bird, Utah Petroleum Association Comments on Proposed Rulemaking,
Revisions to Section IX, Control Measures for Area and Point Sources, Part H Emission Limits and
Amend R307-110-17, undated but submitted August 15, 2018, first comment of Enclosure No. 1 and
Attachment A, Major Stationary Source Precursor Demonstration for NOx, SOx, VOC, and NH3 in the
Salt Lake City 24-hour PM2.5 Serious Nonattainment Area, Final.
Report, August 2018
36 “Draft PM2.5 Precursor Demonstration Guidance”, Stephen D. Page to Regional Air Division Directors
(November 17, 2016).
37 The UDAQ photochemical model is run for an exceedance PM2.5 episode that occurred during the cold
air pool event of January 1-10, 2011.
38 Response to Comment at 112/C0008, response to H-10.
UPA Comments on Proposed PM2.5 SIP Revisions, March 31, 2025
Page 14 of 14

We sincerely appreciate the dialogue with UDAQ to arrive at this proposed SIP. We stand ready
to help as appropriate to support the timely redesignation to attainment of the SLC Serious PM2.5
nonattainment area.

Sincerely,

[original approved by Rikki Hrenko-Browning]

Rikki Hrenko-Browning
President, Utah Petroleum Association

cc:
Bryce Bird - [e-mail]
Becky Close – [e-mail]
Rachel Chamberlain - [e-mail]