NAAQS Regulatory Review and Rulemaking Coalition

December 7,2021

Comments of the NAAQS Regulatory Review & Rulemaking Coalition
on EPA's Revised Draft Guidance for Ozone and Fine Particulate

Matter Permit Modeling

On September 20, 2021, EPA posted Revised Draft Guidance for Ozone and Fine
Particulate Matter Permit Modeling ("Draft Guidance" or "Guidance").1 These are the comments
of the NAAQS Regulatory Review & Rulemaking ("NR3") Coalition on the Draft Guidance. The
NR3 Coalition is an ad hoc coalition of industry groups and companies supportive of National
Ambient Air Quality Standards ("NAAQS") that provide the requisite protection of public health
and welfare and that are implemented in ways that provide that protection, consistent with the
economic health of the country. Members of the NR3 Coalition and their member companies are
committed to reducing emissions as necessary, consistent with the requirements of the Clean Air
Act ("CAA" or "Act"),2 to provide air quality protective of public health and welfare, while
continuing to facilitate economic growth in the United States. We have worked for many years
with EPA, states, and local authorities to lower concentrations of particulate matter ("PM"), ozone,
their precursors, and other common pollutants in ambient air. As a result, between 1970 and 2020,
emissions of criteria air pollutants have steadily declined, while both U.S. gross domestic product
and population have grown.3
I. Role of the Draft Guidance

Many voluntary and compulsory programs have contributed to the improvement of
impaired air quality and maintenance of existing clean air resources over the years. One such

1	https://www.epa.gov/scram/revised-draft-guidance-ozone-and-fine-particulate-matter-pemiit-modeling.

2	42 U.S.C. §§ 7401-7671q.

3	See EPA, Our Nation's Air, https://gispub.epa.gov/air/trendsreport/202l/#effects (last visited Dec. 3, 2021).

1


-------
NAAQS Regulatory Review
& Rulemaking Coalition

program is the Act's Prevention of Significant Deterioration ("PSD") program.4 The Draft
Guidance provides nonbinding recommendations for implementation of some aspects of this
program.5 It provides "recommendations on how to conduct compliance demonstrations for the

03	NAAQS and the PM2.5 NAAQS and PSD increments under the PSD program."6 In particular,
as indicated by its title, it offers technical advice on conducting air quality modeling to support
these compliance demonstrations. The requirement for such compliance demonstrations7 is just
one aspect of the PSD program. That program, which requires that new or modifying major
emitting facilities in attainment or unclassifiable areas obtain preconstruction permits,8 also
requires that these sources be "subject to the best available control technology for each pollutant
subject to regulation under" the CAA.9

The Draft Guidance must be consistent with the statutory goals of the PSD program.
Certainly, the PSD program is intended to preserve air quality in areas where NAAQS are met.
This that does not mean, however, that the program should be implemented without regard to the
burdens particular implementation measures may pose or consideration of how those burdens
relate to their benefits. One goal of the PSD program is "to insure that economic growth will occur
in a manner consistent with the preservation of existing clean air resources."10 The Draft Guidance
must be consistent with this goal. Thus, the Draft Guidance should not impose more costly or
burdensome permitting requirements than are necessary to "preserv[e] existing clean air
resources."11

4	CAA §§ 160-169B.

5	Draft Guidance at 4.

6	Id. at 7.

7	CAA 165(a)(3)

s Id. §§ 165(a), 169(1).

9	Id, § 165(c)(4).

10	Id. § 160(3).

11	Id.

2


-------
NAAQS Regulatory Review
& Rulemaking Coalition

Indeed, the costs associated with the expanded modeling requirements in the Draft
Guidance should not be allowed to exceed the benefits of that modeling. Under the Draft
Guidance, the owner of a new or modifying source that would previously have been considered
minor for PM2.5 will now face a potential requirement to conduct photochemical grid modeling of
the source if its emissions of sulfur dioxide ("SO2") or nitrogen oxides ("NOx") will exceed the
significant emission rate ("SER") for that pollutant. Those emissions would already be subject to
a requirement to use the best available control technology (BACT), however, as a result of the
exceedance of the applicable SER. The owner of that source could decide that the costs, delays,
and uncertainties associated with the newly required PM2.5 modeling render the project
unworkable. For example, the owner of a source seeking to upgrade a boiler might recognize and
accept a requirement for the use of BACT emissions (perhaps a low-NOx burner), but find
unacceptable the costs and uncertainties associated with the PM2.5 modeling required by the Draft
Guidance. It is not clear that EPA has considered whether the additional modeling that the Draft
Guidance is recommending will discourage construction of new facilities that use BACT in lieu of
the continued operation of older facilities that are not subject to a requirement for BACT and
therefore may have higher emission rates. EPA should consider this possible result before it
imposes new modeling requirements.

II. Improved Approaches and Tools for Required Modeling of PM2.5

If EPA finalizes the Draft Guidance as it currently stands, we urge the Agency to consider
improvements to its recommended modeling approaches and tools for PM2.5 to decrease their
conservatism and improve their realism. Unless EPA makes such improvements, modeling
required by the Draft Guidance will tend to overestimate air quality impacts of the source seeking

3


-------
NAAQS Regulatory Review
& Rulemaking Coalition

a permit. This, in turn, "could severely constrain the permitting process, in some instances
blocking environmentally beneficial projects."12

With regard to EPA's recommended AERMOD model, which would be used to predict the
impact of direct emissions of PM2.5, we note that EPA continues to evaluate possible improvements
to the model's treatment of modeling under low wind speed conditions. This is an issue that our
members have repeatedly highlighted in comments to and discussions with EPA.13 We urge the
Agency to improve this aspect of AERMOD as soon as possible. Additionally, Bob Paine and
Dave Heinold, both with AECOM, have identified several other potential improvements to
AERMOD and we urge EPA to improve these aspects of the model rapidly, as well. 14

Second, we recognize and appreciate that use of Modeled Emissions Rates for Precursors
("MERPs") to characterize secondary PM2.5 resulting from emissions of SO2, NOx, or both will
eliminate the need to conduct photochemical grid modeling in most cases. Predictions of PM2.5
levels using MERPs are conservative, however. Use of MERPs requires the unrealistic assumption
that the maximum direct and secondary PM2.5 concentrations attributable to the source seeking a
permit occur simultaneously and at the same site. Because direct emissions of PM2.5 are likely to
have an impact much nearer the source than are precursor emissions, simply combining the two
predictions likely means that the predicted maximum contribution of the source to ambient PM2.5
levels will be overestimated. This overestimate may result in an erroneous prediction of an

12	R. J. Paine & D.W. Heinold, Comments on Revised Guidance for Ozone and Fine Particulate Matter Permit
Modeling (submitted to EPA on Nov. 19, 2021), at 2 ("Paine & Heinold").

13	See, e.g.. Petition of the NAAQS Implementation Coalition to the Administrator of the United States
Enviromnental Protection Agency for Administrative Reconsideration of Portions of the Final Rule Entitled
"Revisions to the Guideline on Air Quality Models: Enhancements to the AERMOD Dispersion Modeling System
and Incorporation of Approaches to Address Ozone and Fine Particulate Matter (Mar. 20, 2017); Comments of the
NAAQS Implementation Coalition on the United States Enviromnental Protection Agency's Revision to the
Guideline on Air Quality Models: Enhancements to the AERMOD Dispersion Modeling System and Incorporation
of Approaches to Address Ozone and Particulate Matter: Proposed Rule (Oct. 27, 2016).

14	See Paine & Heinold at 4-5.

4


-------
NAAQS Regulatory Review
& Rulemaking Coalition

exceedance of the Significant Impact Level ("SIL"). This, in turn, may trigger a requirement for
photochemical grid modeling. Mr. Paine and Mr. Heinold recommend that EPA develop a tool to
provide more realistic MERPs-based predictions near sources to address the likely overprediction
of PM2.5 concentrations to be compared to the SIL.15 We endorse their recommendation.

Finally, if cumulative modeling is required, EPA's approach to addressing sources near the
source for which a permit is being sought is unnecessarily conservative. It requires the unrealistic
assumption that "nearby" sources operate at their allowable emissions level or federally permitted
emissions limit,16 when, in fact, data on the actual emissions of such a source is often available.
If those data are available, they should be used when modeling the source. Furthermore, in the
absence of actual emissions data, if a realistic emissions profile can be developed for a nearby
source, EPA should permit the use of such a profile. In addition, EPA should specify that
background air quality should be determined from a monitor that "is not influenced by any
modeled sources or sources local to the selected monitor" or, if that is not possible, from a monitor
when such influences are minimized, so as to avoid double-counting impacts."17

III. Interaction of the Draft Guidance with Other Aspects of EPA's PSD Regulations

Concerning PM2.5

We recognize that the Draft Guidance focuses on the ambient air quality analysis
component of PSD permitting and appreciate that it "does not change or substitute for any law,
regulation, or any other legally binding requirement."18 Nevertheless, we urge EPA to consider
whether it has implications for other aspects of PSD permitting, particularly with regard to PM2.5.

15	See Paine & Heinold at 3.

16	40 C.F.R. Pt. 51, App. W, Tbl. 8-2.

17	Paine & Heinold at 7.

18	Draft Guidance at 4.

5


-------
NAAQS Regulatory Review
& Rulemaking Coalition

Specifically, we understand that the Draft Guidance requires modeling of direct PM2.5
emissions for PSD permitting purposes even when the increase is emissions of direct PM is below
the 10 ton per year ("tpy") PM SER if SO2 or NOx emissions will exceed the applicable SER for
those emissions. The requirement to model such insignificant direct PM2.5 emissions could raise
issues concerning the rate of emissions to be modeled, particularly for modifications. The Draft
Guidance suggests in a few places that sources should model the "net emissions increase,"19 and
we agree. This approach is in line with the Guidance's scientific basis for requiring consideration
of insignificant emissions increases because to do so ensures that the source provides a full
accounting of its projected air quality impacts for the relevant NAAQS."20 If the intent truly is to
obtain a "full accounting" of a source's impact on ambient air, the analysis cannot be a one-way
street, meaning that the analysis should not just consider what would increase the impacts but must
also address what would decrease impacts. That would be the most "scientific" approach and
provide a more accurate "full accounting" of a source's ambient air impact. The goal must be a
realistic analysis, not one that is so conservative that it is meaningless.

Given EPA's acceptance of netting as a means for determining the applicability of PSD
permitting requirements, we suggest that modeling should consider both the increases resulting
from the project to be permitted and contemporaneous decreases at that unit or other units.21 The
Draft Guidance simply does not address questions such as this one that arise at the intersection of
EPA's PSD regulations and the new modeling guidance. In the absence of clear guidance from
EPA, sources and permitting authorities will be left to develop their own answers, potentially
resulting in inconsistencies between state programs.

19	Id. at 8-11.

20	Id. at 13.

21	See 85 Fed. Reg. 74890, 74894 (Nov. 24, 2020).

6


-------
NAAQS Regulatory Review
& Rulemaking Coalition

The conditions that determine the minor source baseline date for an area are another
example of a possible intersection between the Draft Guidance and other aspects of PSD
permitting. When EPA adopted the PM2.5 increments in 2010 through rulemaking, the Agency
indicated that the PM2.5 minor source baseline date would be set based on the first PSD application
for direct PM2.5 emissions after the 2011 trigger date, not on the date of the first PSD application
for emissions of NOx or SO2 after the trigger date.22 Applying PSD requirements to sources that
will not emit direct PM2.5 in significant amounts could, however, lead to questions about the minor
source baseline date. The Draft Guidance cannot change the criteria for determining the minor
source baseline date for an area. Any change to how that baseline date is determined must follow
the same rulemaking procedures the Agency followed in adopting the increments and specifying
the current procedure. Furthermore, even if such a rulemaking were conducted, it would be
administratively impractical to try to identify retroactively all potentially increment-consuming
sources over the past decade; the revised criteria for the minor source baseline date would have to
be prospective only. For clarity, EPA should acknowledge that the minor source baseline date for
an area continues to be determined based on emissions of direct PM2.5 and therefore can only be
established by a project that causes a significant net emissions increase of direct PM2.5 emissions,
notwithstanding net emissions increases of precursors resulting from the project.
IV. Conclusion

The NR3 Coalition is pleased to offer these comments on the Draft Guidance. For reasons
stated above, the Coalition urges EPA to consider whether the Draft Guidance in its present form
imposes more costly or burdensome requirements than are necessary to prevent deterioration of

22 See 75 Fed. 64864, 64868, 64886 (Oct. 20, 2010) (acknowledging that baseline areas are established based on
modeling and that modeling techniques were not available for PM2.5 precursors).

7


-------
NAAQS Regulatory Review
& Rulemaking Coalition

existing air quality. Please contact Joe Stanko (i stanko@huntonak.com) or Cindy Langworthy
(clangworthy@huntonak.com) if you have questions about them. We look forward to seeing a
final version of the guidance.

8


-------