Responses to Significant Comments on the State and Tribal Designation Recommendations for the 2015 Ozone National Ambient Air Quality Standards (NAAQS)


Responses to Significant Comments on the State and Tribal
Designation Recommendations for the 2015 Ozone National
Ambient Air Quality Standards (NAAQS)

Docket Number EPA-HQ-OAR-2017-0548
U.S. Environmental Protection Agency
April 2018

-------
Table of Contents

1.0 Introduction	4

2.0 Background	4

3.0 Responses to Significant Comments on the State and Tribal Designation Recommendations for the
2015 Ozone NAAQS	5

3.1 General Issues	5

3.1.1	Super-Regional Areas	5

3.1.2	Exceptional Events	11

3.1.3	Effective Date	11

3.1.4	NAAQS Implementation	12

3.1.5	Air Quality Data	12

3.1.6	Nonattainment Boundaries	12

3.1.7	The Term "Nearby"	12

3.1.8	Other Options and the Ozone Transport Commission	12

3.2. Area-Specific Issues	13

3.2.1 EPA Region 1	13

3.2.2.	EPA Region II	13

3.2.3.	EPA Region III	13

3.2.4.	EPA Region IV	19

3.2.5.	EPA Region V	19

3.2.6.	EPA Region VI	37

3.2.7.	EPA Region VII	42

3.2.8.	EPA Region VIII	42

3.2.9.	EPA Region IX	63

3.2.10.	EPA Region X	66

3.2.11.	Multi-State Areas	67

3.3 Comments Not Requiring a Response from EPA	81

3.3.1	NAAQS Related	81

3.3.2	Comments on Unrelated Programs	82

2

-------


List of Acronyms

CAA

Clean Air Act

CAMx

Comprehensive Air Quality Model with Extensions

CBS A

Core Based Statistical Areas

CFR

Code of Federal Regulations

CMSA

Combined Metropolitan Statistical Area

CSA

Combined Statistical Area

CSPAR

Cross-state Air Pollution Rule

DFW

Dallas-Fort Worth, Texas

EE

Exceptional Events

EGU

Electric Generating Unit

EPA

Environmental Protection Agency

FR

Federal Register

HGB

Houston-Galveston-Brazoria, Texas

HYSPLIT

Hybrid Single Particle Lagrangian Integrated Trajectory Model

IEPA

Illinois Environmental Protection Agency

I/M

(Vehicle) Inspection and Maintenance

LAER

Lowest Achievable Emission Rate

MACT

Maximum Achievable Control Technology

MDE

Maryland Department of the Environment

MPO

Metropolitan Planning Organization

MSA

Metropolitan Statistical Area

NESHAP

National Emission Standards for Hazardous Air Pollutants

NAA

Nonattainment Area

NAAQS

National Ambient Air Quality Standard

NEI

National Emissions Inventory

NFR

Notice of Final Rulemaking

NOx

Oxides of Nitrogen

NSPS

New Source Performance Standards

NSR

New Source Review

NYC

New York City

NYMA

New York Metropolitan (Nonattainment) Area

OAQPS

EPA Office of Air Quality Planning and Standards

OMB

Office of Management and Budget

OTR

Ozone Transport Region

PADEP

Pennsylvania Department of Environmental Protection

PPB

Parts Per Billion

PPM

Parts Per Million

PSD

Prevention of Significant Deterioration

QA/QC

Quality Assurance/Quality Control

RACT

Reasonably Available Control Technology

SIP

State Implementation Plan

TCEQ

Texas Commission on Environmental Quality

TPY

Tons per Year

TSD

Technical Support Document

VMT

Vehicle Miles Traveled

VOC

Volatile Organic Compounds

3

-------
1.0 Introduction

This document, together with the preamble to the final rule, and the Technical Support Documents
(TSDs) for the designations, presents the responses of the Environmental Protection Agency (EPA) to the
significant comments we received on our proposed designations. The responses presented in this
document are intended to augment the responses to comments that appear in the preamble to the final rule
and the TSD or to address comments not discussed in those documents.

2.0 Background

On October 1, 2015, the EPA promulgated revised primary and secondary ozone national ambient air
quality standards (NAAQS (80 FR 6592, October 26, 2015)). In that action, the EPA strengthened both
standards to a level of 0.070 parts per million (ppm), while retaining their indicators, averaging times, and
forms. The EPA revised the ozone standards based on an integrated assessment of an extensive body of
new scientific evidence, which substantially strengthens our knowledge regarding ozone-related health
and welfare effects, the results of exposure and risk analyses, the advice of the Clean Air Scientific
Advisory Committee and consideration of public comments.

The revised primary standard provides increased protection for children, older adults and people with
asthma or other lung diseases, and other at-risk populations against an array of adverse health effects
including lung function, increased respiratory symptoms and pulmonary inflammation and asthma
exacerbations; effects that contribute to emergency department visits or hospital admissions; and
mortality. The revised secondary standard provides protection of natural forests from adverse growth-
related effects and is expected to provide increased protection from other effects of potential public
welfare significance, including crop yield loss and visible foliar injury.

On November 6, 2017, the EPA issued final designations for the 2015 NAAQS for ozone for most areas
in the United States (U.S.). Specifically, the Agency found that most areas in the country met the
standards and designated those areas, including 2,646 counties, two tribal areas and five territories,
"attainment/unclassifiable." This represented about 85 percent of the counties in the U.S. The EPA also
designated three counties in the state of Washington as "unclassifiable", because there was not enough
data to calculate a 3-year ozone design value.

On December 22, 2017, the EPA responded to state and tribal recommendations by indicating the
anticipated area designations for the portions of the country not already designated for the 2015 ozone
standards. These responses started a 120-day period for states and tribes to provide additional
information before the EPA determines the final designations. The EPA also opened a 30-day comment
period for the public to provide input on these designations before they are finalized. Following are
summaries of significant comments received on the 2015 ozone designation recommendations and the
EPA's responses to those comments.

4

-------
3.0 Responses to Significant Comments on the State and Tribal
Designation Recommendations for the 2015 Ozone NAAQS

The following sections address the state, tribal, and public comments received by the EPA on the state
and tribal ozone designation recommendations for the 2015 Ozone NAAQS. Comment summaries and
responses are presented below. Comments and responses addressing general issues are presented first
followed by area-specific comments and responses. The EPA has provided additional detail for some
nonattainment areas in the TSD for that area. Commenters can find the TSDs in the electronic docket for
this action (www.regulations.gov. docket number EPA-HQ-OAR-2017-0548) and at the EPA's Ozone
Designations Web Page (www.epa.gov/ozone-designations').

3.1 General Issues
3.1.1 Super-Regional Areas

Comment: Several commenters noted the decision to sharply limit the extent of the nonattainment areas
despite requests from states most affected by unhealthy ozone like Connecticut, New Jersey and Delaware
is flawed. They suggested that the EPA should review and reconsider this decision to define this
nonattainment area so narrowly.

EPA Response: The EPA does not believe that creation of a larger nonattainment area to address
pollution transport is the appropriate approach. As an initial matter, section 107(d)(1) provides that areas
designated nonattainment should include any "nearby" area contributing to a violation of the NAAQS.
We believe that broad super-regional areas go beyond this by including areas that are not necessarily
"nearby" but contribute to nonattainment through long-range transport. The Clean Air Act (CAA) has
several separate provisions in the Act to address this phenomenon. Section 110(a)(2)(D) requires states to
address ozone transport that contributes to a violation of the NAAQS in another State. In addition, section
184, creates the northeast ozone transport region and also grants EPA authority to establish additional
transport regions, as appropriate. Finally, we note that the approach taken by EPA is consistent with the
approach Congress specified for Serious and above areas for the 1-hour NAAQS, where in section
107(d)(4)(A), Congress set the Consolidated Metropolitan Statistical Area (CMSA) boundaries as the
presumptive boundaries of the nonattainment area. In Catawba Co. v. EPA1, the Court upheld that
"contribute" under § 107(a)(1)(A) of the CAA does not necessarily mean "any contribution" to
nonattainment but rather a degree of contribution sufficient to deem an area nonattainment, that is,
sufficient enough to warrant designation as nonattainment. "Section 107(d) is ambiguous as to how EPA
should measure contribution and what degree of contribution is sufficient to deem an area
nonattainment... " Catawba County v. EPA, 571 F.3d 20, 39 (D.C. Cir. 2009) (Internal citation omitted
but with emphasis added). "Thus, reasonably exercising the discretion that Congress delegated to it, EPA
interpreted "contribute" to mean "sufficiently contribute," and then applied the C/MSA presumption and
nine-factor test precisely to identify those areas that meet that definition." Id.

EPA's analyses supporting boundaries for individual nonattainment areas are provided in the TSD for the
area in question.

Comment: A commenter suggested that upwind areas do not have to participate in regional air quality
planning, may attain before downwind areas and contribute insufficiently to emission reductions needed
to attain downwind.

1 Catawba County v. EPA, 571 F.3d 20, (D.C. Cir. 2009)

-------
EPA Response: As noted in the previous response, other provisions of the Act address longer-range
ozone transport and the designation process requires only that "nearby" areas that contribute to violations
of the NAAQS be included as part of the nonattainment area.

Comment: A commenter pointed out that in its proposed designations, the EPA once again refuses to
take steps to comprehensively address the causes of ozone pollution. The EPA rejects multiple states'
recommendations to establish large, multi-state nonattainment areas, noting that "other provisions of the
[Clean Air Act] address longer range transport of ozone pollution, such as sections 110(a)(2)(D), 126, and
184." EPA-HQ-OAR-2017-0548-0073 at 2; EPA-HQ-OAR-2017-0548-0106 at 20. Yet EPA has failed to
implement those provisions. It has not even provided a complete § 110(a)(2)(D) remedy for the East
Coast under the 2008 standard. It has failed to respond to various § 126 petitions and has denied multiple
states' petition to expand the Ozone Transport Region under § 184. It is unlawful and arbitrary for EPA to
play this shell game of referring people suffering from high levels of ozone pollution to other statutory
provisions, even though no statutory avenue is fully implemented.

EPA Response: The EPA understands the concerns raised by the commenter regarding long range
transport. However, as noted in other responses in this document, section 107(d)(1) of the CAA address
"nearby" areas that contribute to a violation of the NAAQS. We do not believe that these provisions that
focus on "nearby" contribution are the appropriate vehicle for addressing long range transport. The CAA
has several separate provisions in the Act to address this phenomenon.

Comment: A commenter suggested that the reasons why the term 'nearby' can be extended beyond
EPA's proposed designations; 1) EPA's modeling shows a larger impact on Connecticut's ozone from
sources in Pennsylvania than those in New Jersey. 2) Trajectories, covering up to two days' transport, in
the TSD show air passes through York, Monroe and Lehigh Counties and other higher nitrogen oxide
emitting counties in Pennsylvania on their way to peak ozone violations in Connecticut. 3) States need to
be in the same room to plan effective strategies.

EPA Response: The EPA is maintaining its established interpretation of the term "nearby" and
continuing the policy of designating as nonattainment all areas violating the ozone NAAQS and any areas
in close proximity that are contributing to a violation in another area. Under the designation provision,
only "nearby" areas that contribute to the violation must be included as part of the nonattainment area.
There are other provisions of the CAA that address longer range transport of ozone pollution, such as
sections 110(a)(2)(D), 126, and 184. The phenomenon of ozone transport must be balanced against the
need to have smaller areas that can focus on local control measures. In the absence of broad agreement
among all affected states to recommend such a large nonattainment area, we do not intend to designate a
large nonattainment area and instead intend to adhere to a common-sense interpretation of the term
"nearby." EPA considered, and rejected, similar recommendations in connection with the boundaries for
the New York Metro Area under the 2008 ozone NAAQS. At that time, EPA explained that the CAA
"does not require that all contributing areas be designated nonattainment, only the nearby areas," and that
"[Regional strategies, such as those employed in the Ozone Transport Region and EPA's NOx SIP Call
are needed to address the long-range transport component of ozone nonattainment." "Area Designations
for the 2008 Revised Ozone National Ambient Air Quality Standards" (December 4, 2008), at 4; see
Mississippi Commission on Environmental Quality v. EPA, 790 F.3d 138, 150-51 (D.C. Cir. 2015). The
D.C. Circuit upheld EPA's approach. Id. at 151-53.

Comment: A commenter said the EPA should reconsider its analyses and rely on its authority to
designate areas which best account for transport rather than looking to inconsistencies amongst the
various state submittals as justification for the status quo.

-------
EPA Response: Conducting a five-factor analysis on the given Combined Statistical Area is the starting
point basis for including or excluding any particular county within the CSA. The EPA conducts area-
specific analyses to support nonattainment area boundary recommendations and final boundary
determinations by evaluating factors such as air quality data, emission and emissions-related data,
meteorological data, geography/topography and jurisdictional boundaries. While jurisdictional boundaries
and state recommendations - including differences between the affected states' recommendations ~ are
taken into consideration, these are not the only bases for nonattainment boundary determinations.

For the reasons noted elsewhere, the EPA does not believe the creation of a larger nonattainment area to
address pollution transport is the appropriate approach.

Comment: A commenter said including the Philadelphia metropolitan area and upwind counties in the
New York Metropolitan (Nonattainment) Area (NYMA) would be more effective because otherwise the
Philadelphia area will likely be Marginal and not require any coordination of emission reductions,
whereas the NYMA area will have Moderate requirements, which will include coordination and
modeling.

EPA Response: A combined New York, Philadelphia area would result in a multistate nonattainment
area covering six states (Connecticut, Delaware, Maryland, New Jersey, New York, and Pennsylvania).
Managing a nonattainment area with six states agencies would be unwieldly at best. Furthermore,
designating this "super-regional" area, covering two CSAs and six states, would not necessarily result in
additional emission reductions. All six of the states, in their entirety, are in the Ozone Transport Region,
and therefore have planning requirements similar to a Moderate nonattainment area, including
Reasonably Available Control Technology (RACT) and New Source Review (NSR). Moreover, all of the
states, in their entirety, are subject to the regional transport requirements and are part of the Cross-State
Air Pollution Rule(CSPAR) and its Update. Further, all of the areas in the Philadelphia nonattainment
area will be subject to the basic nonattainment SIP requirements, as will all of the areas in the New York
City nonattainment area.

Comment: A commenter stated that a larger Moderate nonattainment area would have requirements for
proof that the controls will actually be useful and sufficient for attainment. Without it, downwind peak
areas will repeat the pattern of control, wait for assistance and then failure to attain, when cheaper
controls upwind would be more useful and according to EPA's own modeling, needed.

Ozone is a regional pollutant and is readily transported both short and long distances. To determine
whether a "nearby" area is contributing to a violation, EPA recommended that states conduct a technical
analysis based on a number of factors listed in the designation guidance for the 2015 ozone NAAQS,
including air quality data, emissions and emissions-related data, meteorological data, and
geography/topography and jurisdictional boundaries. In evaluating whether to modify a state's
designation recommendation, EPA also considered those factors. The justification for including or
excluding a county in the nonattainment area is provided in EPA's TSD for the area. In determining
whether an area should be designated nonattainment, EPA did not consider the costs of controls because
that is not relevant for determining whether an included area is violating the NAAQS or is a "nearby" area
that is contributing to a violation as provided under CAA section 107(d).

-------
The implementation rulemaking for the 2015 ozone NAAQS will address the control obligations for areas
designated nonattainment. As EPA considers the required elements of implementation for the 2015 ozone
NAAQS, it is our goal to propose approaches that provide flexibility and opportunity for efficiency to the
extent such approaches are consistent with the CAA and will not jeopardize expeditious attainment of the
public health and welfare goals of the CAA. In addition, we are exploring ways in which the EPA could
provide assistance to the states. Finally, to the extent the CAA does not mandate specific control
measures, states may consider economic concerns in development of their state implementation plans to
address air quality.

Comment: A commenter suggested that the analysis of the 'nearby' area for the Philadelphia metro area
is not consistent with even this proposal's usage of 'nearby' and will leave major contributing counties
out or the area will fail to attain, wasting resources. A better action would be to have a larger
nonattainment area that includes the New York City (NYC) and Philadelphia areas along with the
counties of York, Lehigh and Monroe and others as needed. This area would provide the incentive for
good planning efforts, led by the states, affording an opportunity for attainment in 2024. This is an
opportunity to use the Clean Air Act to meet the President's goal of clean air for everyone and meet the
standards in the northeastern corridor of the USA. Without the larger area, areas will implement
expensive and possibly unattainable control measures and at worst, not even provide the clean air
promised.

EPA Response: As noted, the starting point for the analysis of both areas is their CSAs. The EPA was
consistent in its evaluation of the CSAs within each TSD, having a long history of using the CSA to
define the starting point for assessing nearby contributing areas. The EPA used a similar approach with
the 1997 and 2008 ozone NAAQS. Furthermore, the Clean Air Act Amendments of 1990 set, by
operation of law, the boundaries for Serious, Severe, and Extreme nonattainment areas for the 1979 1-
hour ozone NAAQS as the metropolitan statistical area or consolidated metropolitan statistical area (as
established by the Bureau of the Census).

Ozone is a regional pollutant and is readily transported both short and long distances. To determine
whether a "nearby" area is contributing to a violation, EPA recommended that states conduct a technical
analysis based on a number of factors listed in the designation guidance for the 2015 ozone NAAQS,
including air quality data, emissions and emissions-related data, meteorological data,
geography/topography and jurisdictional boundaries. In evaluating whether to modify a state's
designation recommendation, EPA also considered those factors. The EPA did not consider economic
impacts because that is not relevant for determining whether an included area is violating the NAAQS or
is a "nearby" area that is contributing to a violation as provided under CAA section 107(d).

A combined New York, Philadelphia area would result in a multistate nonattainment area covering six
states (Connecticut, Delaware, Maryland, New Jersey, New York, and Pennsylvania). Managing a
nonattainment area with six states agencies would be unwieldly at best. Furthermore, designating this
"super-regional" area, covering two CSAs and six states, would not necessarily result in additional
emission reductions. All six of the states, in their entirety, are in the Ozone Transport Region, and
therefore have planning requirements similar to a Moderate nonattainment area, including RACT and
NSR. Moreover, all of the states, in their entirety, are subject to the regional transport requirements and
are part of the CSPAR and its Update. Further, all of the areas in the Philadelphia nonattainment area will
be subject to the basic nonattainment SIP requirements, as will all of the areas in the NYC nonattainment
area.

Comment: A commenter said the EPA needs to take the appropriate steps to address the transport of
ozone precursors from significant and nearby upwind areas. With this in mind, New Jersey revised their
nonattainment boundaries to include areas in their shared CSAs.

-------
EPA Response: The EPA evaluated the entire CSA in making its decisions for both the NYMA and the
Philadelphia nonattainment areas. The corresponding TSDs include the factor analysis for all the counties
within the CSAs. As an initial matter, section 107(d)(1) provides that areas designated nonattainment
should include any "nearby" area contributing to a violation of the NAAQS. The five-factor analysis is
conducted for each county within its corresponding CSA and a determination is made based on the factors
to include or exclude the county within the designated area.

As noted, ozone is a regional pollutant and is readily transported both short and long distances. To
determine whether a "nearby" area is contributing to a violation, EPA recommended that states conduct a
technical analysis based on a number of factors listed in the designation guidance for the 2015 ozone
NAAQS, including air quality, emissions and emissions-related data, meteorology, and
geography/topography. In evaluating whether to modify a state's designation recommendation, EPA also
considered those factors. The justification for including or excluding nearby areas in a given
nonattainment area is provided in EPA's TSD for the area.

New Jersey's revised recommendation for their nonattainment area proposed to add the four counties of
Pike, Monroe, Northampton and Lehigh, Pennsylvania, to the NYMA area and to add the four counties of
Berks, Lebanon, Lancaster and York, Pennsylvania, to the Philadelphia area. The counties of Pike,
Monroe, Northampton and Lehigh, Pennsylvania, are in the NYMA CSA (New York-Newark, NY-NJ-
CT-PA CSA) and the factor analysis was conducted for these counties in the corresponding TSD. The
basis for excluding these counties was based on the factor analysis and their overall contribution to
nonattainment within the nonattainment area. These Counties were assessed based on their air quality,
emissions and emissions-related data, meteorology, and geography/topography, and they were excluded
from the NYMA area. The counties of Lebanon, Lancaster and York, Pennsylvania, are not part of the
Philadelphia CSA (Philadelphia-Reading-Camden, PA-NJ-DE-MD CSA) and any contribution to New
Jersey from these counties can be addressed through the transport provisions in the CAA. The EPA
evaluated the factors for Berks County, Pennsylvania, and determined based on the factors that the
County should be designated separately from the Philadelphia area. However, EPA stated in its December
20, 2017 letter to Governor Tom Wolf (EPA-HQ-OAR-2017-0548-0145), that, if EPA approves
Pennsylvania's exceptional events (EE) demonstration, the 2014-2016 design value for the Berks County
monitor would move from violating to attaining the 2015 ozone NAAQS. In that case, EPA would revise
its recommendation for the Reading Area to attainment/unclassifiable. The EPA concurred on
Pennsylvania's EE demonstration for the Reading airport ozone monitor (monitor 420110011) in Berks
County on March 6, 2018. Therefore, EPA is designating Berks County as attainment/unclassifiable.
Please see corresponding TSDs for more information.

Comment: A commenter said the EPA did not specifically address New Jersey's large nonattainment
area recommendation. The determination for the size of a nonattainment area must include the areas
monitoring a violation of the standard, as well as, all areas that significantly contribute to violations of the
standard.

EPA Response: As noted, the EPA does not believe that creation of a larger nonattainment area to
address pollution transport is the appropriate approach. As an initial matter, section 107(d)(1) provides
that areas designated nonattainment should include any "nearby" area contributing to a violation of the
NAAQS. We believe that broad super-regional areas go beyond this by including areas that are not
necessarily "nearby" but contribute to nonattainment through long-range transport. The CAA has several
separate provisions in the Act to address this phenomenon. Section 110(a)(2)(D) requires states to address
ozone transport that contributes to a violation of the NAAQS in another State. In addition, section 184,
creates the northeast ozone transport region and also grants EPA authority to establish additional transport
regions, as appropriate. Finally, we note that the approach taken by EPA is consistent with the approach

-------
Congress specified for Serious and above areas for the 1-hour NAAQS, where in section 107(d)(4)(A),
Congress set the CMSA boundaries as the presumptive boundaries of the nonattainment area.

Comment: A commenter said the EPA must take steps to expand the size of New Jersey's ozone
nonattainment areas to address the nearby emissions from Pennsylvania that contribute to the poor air
quality experienced in New Jersey and the region.

EPA Response: Counties in New Jersey are contained in the NYMA or the Philadelphia nonattainment
areas. As noted above, the EPA evaluated the entire CSA in making its decisions for both the NYMA and
the Philadelphia nonattainment areas. According to the EPA's designations guidance, the starting point
for the designation of an area is the CSA. The EPA does not believe that creation of a larger
nonattainment area to address pollution transport is the appropriate approach. As an initial matter, section
107(d)(1) provides that areas designated nonattainment should include any "nearby" area contributing to a
violation of the NAAQS. We believe that broad super-regional areas go beyond this by including areas
that are not necessarily "nearby" but contribute to nonattainment through long-range transport.

Comment: A commenter suggested that it is important to add additional counties to the current NYC and
Philadelphia nonattainment areas, because the Court Ruling concerning the 2015 EPA rule implementing
the 2008 ozone standard affirmed that creditable emission reductions must come from within the ozone
nonattainment area seeking credit. A state with a smaller sized ozone nonattainment area will find it very
difficult to make any meaningful reductions when transported emissions are equal to or greater that the
contribution from the nonattainment area alone.

EPA Response: As noted, ozone is a regional pollutant and is readily transported both short and long
distances. To determine whether a "nearby" area is contributing to a violation, EPA recommended that
states conduct a technical analysis based on a number of factors listed in the designation guidance for the
2015 ozone NAAQS, including air quality, emissions and emissions-related data, meteorology, and
geography/topography. In evaluating whether to modify a state's designation recommendation, EPA also
considered those factors. The justification for including or excluding a county in the nonattainment area is
provided in EPA's TSD for the area. In determining whether an area should be designated nonattainment,
EPA did not consider economic impacts because that is not relevant for determining whether an included
area is violating the NAAQS or is a "nearby" area that is contributing to a violation as provided under
CAA section 107(d).

The implementation rulemaking for the 2015 ozone NAAQS will address the CAA's control obligations
for areas designated nonattainment. To the extent the CAA does not mandate specific control measures,
states may consider economic concerns in development of their state implementation plans to address air
quality.

Comment: One commenter contended that the addition of the Pennsylvania counties to the current
NYMA and Philadelphia nonattainment areas are appropriate based on their significant contribution to
ozone nonattainment, they are local from a geographical perspective, meteorology indicates pollution
travels from these counties to nonattainment monitors in the nonattainment area, the emissions are
significant, and the population in the counties are connected economically based on commuting data.

EPA Response: As noted, the starting point for the designation of an area is the CSA. The EPA does not
believe that creation of a larger nonattainment area to address pollution transport is the appropriate
approach. As an initial matter, section 107(d)(1) provides that areas designated nonattainment should
include any "nearby" area contributing to a violation of the NAAQS. We believe that broad super-
regional areas go beyond this by including areas that are not necessarily "nearby" but contribute to
nonattainment through long-range transport.

-------
The EPA evaluated all counties within both New Jersey CSAs and relied on factors including air quality,
emissions and emissions-related data, meteorology, and geography/topography when analyzing the
counties within the CSAs. The data included, among other things, commuting data and 2014 Total
Vehicle Miles Traveled (VMT) and the percentages of those that commuted to nonattainment counties.
Less than 20% of commuters within the Pennsylvania counties in the NYMA CSA traveled to counties
with violating monitors. In addition, the EPA evaluated the factors for Berks County, Pennsylvania, and
determined based on the factors that the County should be designated separately from the Philadelphia
area. Importantly, meteorology shows, in Figures 6a-e and 6g-o of the TSD, that violating monitors in the
Philadelphia-Wilmington-Atlantic City area are generally not impacted by Berks County relative to other
counties in the area of analysis. The HYSPLIT back trajectories for the violating monitors in the area of
analysis are predominantly from the south and southwest and Berks County is to the west or northwest of
the other counties in the area of analysis. Please see corresponding TSDs for further information.

3.1.2 Exceptional Events

Comment: A commenter suggested the impacts of wildfires that occurred in Canada in May and July
2016 are shown for the states of Connecticut, New York and New Jersey. While the EPA has recognized
these Exceptional Events (EE) in certain cases, there are instances where the same data for other monitors
is not excluded. This would ensure appropriate design values would be used for the designations,
upcoming modeling and SIP submittals such as the Good Neighbor SIPs.

EPA Response: The EPA's EE rule (40 CFR 50.14 and 51.930) requires EPA to make a concurrence or
nonconcurrence determination on EE demonstrations only for monitor-days with current regulatory
significance, which means that the demonstration would affect the outcome of a regulatory determination.
However, submitted monitor-days considered to not have regulatory significance at the time of the EPA's
demonstration review will be reconsidered for an EE determination if at some future date those monitor-
days take on regulatory significance. The EPA has taken action on all EE demonstrations with regulatory
significance submitted by states for the purpose of informing these 2015 ozone NAAQS designations.
Information on related EE actions can be found in the docket for this rulemaking.

3.1.3 Effective Date

Comment: A commenter stated that the CAA dictates that the EPA must set the effective date for the
designations to be the date of their promulgation.

EPA Response: The EPA does not agree with the comment that the CAA requires the EPA to set the
effective date at the time of promulgation. Nothing in section 107(d)(1) addresses the effective date of
designations. In a recent lawsuit claiming EPA had failed to meet its obligation to complete the
designation process for the 2015 ozone NAAQS, the court rejected an argument that the court should
order EPA to establish an "immediate" effective date for the designations. In Re Ozone Designation
Litigation (N. D. Cal. No. 17-cv-06900-HSG), the court stated: "The CAA does not set forth a specific
date by which the agency must make designations effective. The statute does not expressly equate the
Administrator's duty to promulgate with a duty to make designations effective." The EPA is establishing
an effective date of 60 days following publication, which is consistent with past practice for ozone
designations.

-------
3.1.4 NAAQS Implementation

Comment: A commenter suggested EPA should follow the letter of the law and prioritize public health at
every state of the implementation process.

EPA Response: The EPA appreciates the commenter's views regarding implementation of the NAAQS.
Under the cooperative federalism structure of the Act, States take primary responsibility for implementing
the NAAQS once the NAAQS are promulgated and areas are designated for the standard. The EPA takes
seriously its role in administering the CAA.

3.1.5 Air Quality Data

Comment: A commenter opposed any designations that do not accurately reflect the monitoring data and
that would withhold critical local air quality information from communities and states.

EPA Response: The designations reflect the most recent air quality monitoring data that was required to
be certified (2014-2016) in most instances. States are required to certify air quality data by May 1 for the
previous calendar year. Where states have submitted early certified data for 2017, the EPA is assessing
compliance with the NAAQS on 2015-2017 data. The EPA notes that some counties are designated based
on contribution to an area that is violating the standard. Therefore, some county designations are not
based solely on air quality monitoring in that county but also on a determination of whether emissions in
that county contribute to a violation at a monitor in a different, nearby county.

3.1.6 Nonattainment Boundaries

Comment: A commenter opposed designations that irrationally and illegally cut down the size of the
nonattainment areas which must include areas that contribute to unhealthy air in nearby areas.

EPA Response: The EPA evaluated each area on a case-by-case basis using the five factor approach
outlined in the ozone designations guidance (https://www.epa.gov/ozone-designations/epa-guidance-area-
designations-2015-ozone-naaqs). Consistent with the requirements of section 107(d) of the CAA, as part
of this analysis, the EPA also evaluated nearby areas that could contribute to violations in a violating area.

3.1.7 The Term "Nearby"

Comment: A commenter said the NYC TSD never defines the term 'nearby' and the Philadelphia TSD
does not even use the term. A consistent definition of 'nearby' would be to have counties that strongly
affect ozone concentrations at peak downwind receptors and are within the travel time of ozone and its
precursors during the return flow portion of high pressure systems.

EPA Response: The EPA discussed the definition of "nearby" in the TSD in section 3.0, in the
subsection entitled, "Conclusion for The New York Metro Area." The EPA relied on its established
history of using the CSA to define the starting point for "nearby" contributing areas. The EPA used a
similar approach with the 1997 and 2008 ozone NAAQS.

3.1.8 Other Options and the Ozone Transport Commission

Comment: One commenter noted that options, like section 126 petitions, section 110 SIP and
infrastructure SIPs do not account for the nearby impacts of two-day ozone and precursor travel time. The
EPA needs to cover this gap with more effective use of transport-related SIPs or by giving states the

-------
chance to work together by combining the Philadelphia and NYC ozone nonattainment areas with the
nearby high-emission counties.

EPA Response: As discussed previously, the EPA followed the historical practice of using the CSA as
the starting point for determining "nearby" contributions for the designations process. We used a similar
approach designating areas for the 1997 and 2008 ozone NAAQS. The EPA notes that Philadelphia and
New York are in separate CSAs. The options presented by the commenter are programs established
under other sections of the Clean Air Act to address long range transport.

Comment: The commenter said that the Ozone Transport Commission (OTC) has not been as effective
due to the EPA's practice of requiring unanimity of OTC states for recommendations of control measures.
This is not required by the CAA, which provides for the OTC to adopt its recommended controls to the
EPA by majority vote. This is policy change that the EPA can make to follow the law.

EPA Response: The EPA acknowledges the comment regarding the OTC but notes that it is outside the
scope of this action.

Comment: The commenter said the EPA needs to prepare additional tools to help this region and others
meet the standard, including an updated CSPAR under the 2015 standard and revisiting the flawed
October 2017 decision to limit the Ozone Transport Region.

EPA Response: The EPA acknowledges the comment addressing tools to help regions meet the standard.
However, this action designates areas under the 2015 Ozone NAAQS. Implementation of the ozone
standard is covered by other sections of the CAA and is outside of the scope of this action.

3.2. Area-Specific Issues
3.2.1 EPA Region I

Comments received regarding Connecticut counties in the NYMA are addressed below in Section 3.2.11.

3.2.2. EPA Region II

Comments received regarding areas located in EPA Region II (New York and New Jersey) are addressed
below in Section 3.2.11.

3.2.3. EPA Region III

3.2.3.1 Maryland

Comment: A commenter was unaware whether the EPA has acted on Maryland's May 26, 2017 EE
events package regarding the July 2017 Canadian wildfire. The commenter states that Maryland's
demonstration assesses design values affected by both the May and July 2016 Canadian wildfires.

EPA Response: By letters and enclosures dated May 26, 2017 and October 20, 2017, the Maryland
Department of the Environment (MDE) submitted an EE demonstration related to the May and July 2016
Canadian wildfires. MDE determined that the May and July 2016 events affected 16 and 12 monitors,
respectively, throughout Maryland. By letter dated December 26, 2017, EPA concurred on 17 monitor
days, deferred concurrence on 16 monitor days, and non-concurred on 10 monitor days based on the
information MDE provided and EPA's weight of evidence demonstrations referenced in 40 CFR

-------
50.14(a)(2) and (b)(1). The EPA deferred action on monitors that do not have regulatory significance at
this time. The EPA will retain MDE's submittal, should any deferred data become significant for any
future regulatory action. The EPA non-concurred on exceptional event claims for monitors that did not
have exceedances of the NAAQS on the requested dates, or did not have current or projected future
regulatory significance.

3.2.3.2 Pennsylvania

Comment: A commenter was not aware whether EPA has acted on Pennsylvania's November 2017 EE
package regarding the May 2016 Canadian wildfire.

EPA Response: The Pennsylvania Department of Environmental Protection (PADEP) submitted an EE
demonstration related to the May 2016 Canadian wildfire on May 31, 2017. The EPA reviewed that
demonstration and requested PADEP provide further evidence to support their request. PADEP
supplemented the demonstration and submitted a final version on February 20, 2018. By letter dated
March 6, 2018 (EPA-HQ-OAR-2017-0548-0331), the EPA concurred on 8 monitor days, deferred
concurrence on 41 monitor days, and non-concurred on 78 monitor days based on the information
PADEP provided and the EPA's weight of evidence demonstrations referenced in 40 CFR 50.14(a)(2)
and (b)(1). The EPA deferred action on monitors that do not have regulatory significance at this time.
The EPA will retain PADEP's submittal, should any deferred data become significant for any future
regulatory action. The EPA non-concurred on monitors that did not have exceedances of the NAAQS on
the requested dates, or did not have current or projected future regulatory significance.

Comment: The commenter "does not agree that any Pennsylvania counties should be included in any as
[sic] multi-county or multistate non-attainment areas for the 2015 Ozone NAAQS. This is completely
unnecessary because Pennsylvania is located entirely within the [CAA] defined Ozone Transport Region
(OTR). Simply stated, in the case of ozone, inclusion in a multi-county or multistate 'transport region'
can only be to the detriment of the individual counties that are already measuring attainment with the
2015 Ozone NAAQS."

EPA Response: The EPA assumes that the commenter meant that any Pennsylvania county with
monitors showing attainment with the ozone NAAQS should not be grouped with any county showing
nonattainment as part of a nonattainment area. However, CAA section 107(d)(l)(A)(i) requires that the
state designate as nonattainment, "any area that does not meet (or that contributes to ambient air quality in
a nearby area that does not meet) the national primary or secondary ambient air quality standard for the
pollutant." In making the final promulgations required by section 107(d)(l)(B)(i), "the Administrator
may make such modifications as the Administrator deems necessary to the designations of the areas (or
portions thereof) submitted under subparagraph (A) (including to the boundaries of such areas or portions
thereof)." See CAA section 107(d)(l)(B)(ii). The EPA is obligated to designate nonattainment areas
pursuant to CAA section 107(d)(1) regardless of an areas' location within the OTR. Furthermore,
nonattainment area designations under section 107(d)(1) addresses nearby contribution to nonattainment
problems, whereas the OTR addresses regional transport of pollutants.

The process the EPA followed for promulgating designations is described in the TSD in the docket at
EPA-HQ-OAR-2017-0548-0097. The EPA has interpreted Section 107(d)(1)(A) as recognition in the
CAA that an area could be showing attainment with a certain NAAQS, but nonetheless be contributing
pollutants to a nearby NAAQS nonattainment area (see EPA's 2016 ozone designations guidance).The
EPA has historically considered, and used in designating the 2015 ozone nonattainment areas, a weight of
evidence approach following a five -factor analysis that takes into account more than county/state borders
and monitored values in counties. The five-factor analysis is discussed in the NPR and various TSDs for
this action. See EPA-HQ-OAR-2017-0548. The presence of an attaining NAAQS monitor in an area

14

-------
does not preclude that area from affecting NAAQS compliance in a nearby nonattainment area. Many
other factors in addition must be taken into account to determine whether an area is contributing to
nonattainment in a nearby area.

Comment: A commenter suggested that some states that are parts of multi-state nonattainment areas
appear unwilling to redesignate attainment of a NAAQS in their own state (continuing nonattainment),
even though monitoring data may show attainment, to force economic burdens upon sources in
neighboring states, as shown by certain language in a December 9, 2013 letter submitting a CAA Section
176A petition to EPA.

EPA Response: The commenter did not cite a specific example of a state in a multi-state nonattainment
area refusing to seek redesignation to attainment for areas under its control in order to burden other states.
The letter cited by the commenter transmitted a large amount of information in support of the request by
some OTR states under Section 176A of the CAA to expand the OTR to include other upwind states. As
such, the letter itself is only a small portion of a much larger submission. In addition, EPA eventually
denied this request to expand the OTR. See 82 FR 51238, November 3, 20172. Area redesignations from
nonattainment to attainment and 176A petitions are outside the scope of this initial area designation action
for the 2015 ozone NAAQS.

Comment: A commenter noted that there are negative consequences to a county with monitors showing
attainment being included in a nearby nonattainment area, including:

1)	Additional, unnecessary emission reductions being required from major sources in a county
already measuring attainment to allow that multi-county or multistate area be designated as
attainment;

2)	In the case of Pennsylvania, it could eliminate the ability to average with other affected sources
under common ownership or control under the Pennsylvania RACT 2 rule; and

3)	The inability of a state to designate their county to attainment if another state with a county in a
multistate nonattainment area will not designate their included county to attainment.

EPA Response: As mentioned above, section 107(d)(l)(A)(i) requires states and EPA to include in
nonattainment areas those nearby areas contributing to nonattainment. The EPA applies a five-factor
analysis, after receiving recommendations from the state, to evaluate whether nearby counties sufficiently
contribute for the purposes of including a county as part of the nonattainment area. Control strategy
consequences associated with a nonattainment designation, compliance with the CAA's RACT provisions
for nonattainment areas, and area redesignations from nonattainment to attainment are outside the scope
of this initial area designation action for the 2015 ozone NAAQS.

Comment: One commenter said all counties in Pennsylvania, including Lebanon, Berks, York, Chester,
and Montgomery Counties, should be designated as attainment/unclassifiable for the 2015 ozone
NAAQS, with the exception of Bucks, Delaware, and Philadelphia Counties.

EPA Response: Regarding Lebanon County, the EPA has approved Pennsylvania's exceptional events
(EE) submission for the May 2016 Canadian wildfires as it pertains to Lebanon County. See the March 6,
2018 EE concurrence letter from EPA Region III Regional Administrator Servidio to Pennsylvania DEP
Secretary McDonnell. Following approval of the EE submission, the revised 2014-2016 design value
(DV) for the violating Lebanon County ozone monitor meets the attainment level for the 2015 ozone
NAAQS. By letter dated December 20, 2017 (EPA-HQ-OAR-2017-0548-0145) the EPA first proposed,

2 This final action is currently being appealed.

15

-------
based on the five factor analysis performed in the TSD, that Lebanon County and the following additional
counties be part of a new 2015 ozone nonattainment area known as the Harrisburg-York-Lebanon-
Lancaster Area: Cumberland, Dauphin, Lancaster, and York Counties. However, this letter also
anticipated that Pennsylvania's EE submission, if approved, would put Lebanon County into the
attainment/unclassifiable category, and under that circumstance alternatively proposed that the counties in
the proposed Harrisburg-York-Lebanon-Lancaster Area be designated as attainment/unclassifiable, with
the exception of York County, which would be unclassifiable.

Regarding Berks County, the EPA's approval of Pennsylvania's EE demonstration for the May 2016
Canadian Wildfires caused the 2014-2016 DV for the Reading area monitors to show attainment with the
2015 ozone standard of 70 ppm. As stated in EPA's proposed designation rulemaking, the Reading Area
(which consists solely of Berks County) will therefore be designated attainment/unclassifiable for the
2015 ozone NAAQS.

For Chester County, approval of the EE demonstration has not brought the county's DV to attainment.
The DV for Montgomery County following approval of the EE submission does not bring the monitor
into attainment, but as discussed in response to other comments, the weight of evidence of the five factors
warrants keeping Montgomery County as part of the Philadelphia nonattainment area.

Harrisburg-York-Lebanon-Lancaster. PA

Comment: The Pennsylvania Department of Environmental Projection (PADEP) requested that the
Harrisburg-York-Lebanon-Lancaster and Reading Areas should be designated as
attainment/unclassifiable if the EPA concurs on Pennsylvania's EE demonstration.

EPA Response: The EPA concurred on Pennsylvania's EE demonstration for the Berks County and
Lebanon County monitors on March 6, 2018 (EPA-HQ-OAR-2017-0548-0331). As stated in the EPA's
December 20, 2017 letter to Governor Tom Wolf (EPA-HQ-OAR-2017-0548-0145), if EPA approves the
EE demonstration, the 2014-2016 design value for those monitors would move from violating to attaining
the 2015 ozone NAAQS. In that case, EPA would revise its recommendations for the Reading Area
(Berks County) and the Harrisburg-York-Lebanon-Lancaster Area (Cumberland, Dauphin, Lebanon,
Lancaster, and York Counties) from nonattainment to attainment/unclassifiable. Therefore, the EPA is
designating Berks, Cumberland, Dauphin, Lebanon, and Lancaster Counties as attainment/unclassifiable.

The EPA's December 20, 2017 letter (EPA-HQ-OAR-2017-0548-0145) also stated that, if the EE
demonstration is approved, EPA intended to designate York County as unclassifiable because a York
County monitor has incomplete data. However, on March 19, 2018, PADEP submitted to the EPA a data
completeness analysis that includes missing days assumed to be less than the standard for the York
County Downwind monitor (monitor 421330011). The EPA approved that demonstration on April 4,
2018 (EPA-HQ-OAR-2017-0548-0332). Therefore, that incomplete monitor now has a valid design value
that meets the NAAQS. Therefore, the EPA is designating York County as attainment/unclassifiable.

Comment: A commenter supported the EPA's recommendation to expand the counties designated
nonattainment in the Harrisburg-York-Lebanon-Lancaster area to Cumberland, Dauphin, Lancaster,
Lebanon, and York Counties based on factors including emission sources of NOx and VOC, VMT,
emissions transport, population and population density.

EPA Response: In the EPA's December 20, 2017 letter to Governor Tom Wolf (EPA-HQ-OAR-2017-
0548-0145), the EPA stated its intention to designate Cumberland, Dauphin, Lebanon, Lancaster, and

16

-------
York Counties as nonattainment in the Harrisburg-York-Lebanon-Lancaster Area. However, the EPA
also stated that, if the EPA approves Pennsylvania's EE demonstration, the 2014-2016 design value for
the Lebanon County monitor would move from violating to attaining the 2015 ozone NAAQS. In that
case, the EPA would revise its recommendation for the Harrisburg-York-Lebanon-Lancaster area to
attainment/unclassifiable. The EPA concurred on Pennsylvania's EE demonstration for the Lebanon
County ozone monitor on March 6, 2018 (EPA-HQ-OAR-2017-548-0331). Therefore, the EPA is
designating Cumberland, Dauphin, Lebanon, and Lancaster Counties as attainment/unclassifiable.

The EPA's December 20, 2017 letter (EPA-HQ-OAR-2017-0548-0145) also stated that, if the EE
demonstration is approved, the EPA intended to designate York County as unclassifiable because a York
County monitor has incomplete data. However, on March 19, 2018, PADEP submitted to the EPA data
completeness analysis that includes missing days assumed to be less than the standard for the York
County Downwind monitor (monitor 421330011). The EPA approved that demonstration by letter dated
March 6, 2018 (EPA-HQ-OAR-2017-0548-0332). Therefore, that incomplete monitor now has a valid
design value that meets the NAAQS, so the EPA is designating York County as attainment/unclassifiable.

Comment: A commenter opposed the EPA's intended designation of Dauphin County, Pennsylvania as
Nonattainment. To the extent that the Commonwealth of Pennsylvania's initial recommendations dated
October 3, 2016 support our position with respect to Dauphin County, we incorporate Pennsylvania's
recommendations and data by reference. The Commonwealth recommended that Dauphin County be
designated as Attainment with the 2015 Ozone NAAQS. We also incorporate by reference Pennsylvania's
supplemental data submission to the EPA dated April 11, 2017. Finally, the commenter incorporates by
reference and fully supports the Commonwealth's EE Analysis pertaining to the dates May 24th through
May 26th of 2016 that was submitted to the EPA by Pennsylvania's Department of Environmental
Protection ("PADEP") in November of 2017. As indicated in more detail in the formal comments
uploaded to the Docket, the commenter respectfully requests that the EPA revise its designations to
exclude Dauphin County, Pennsylvania from the Harrisburg-York-Lebanon-Lancaster, Pennsylvania
Nonattainment Area and designate the county as Attainment/Unclassifiable as originally recommended by
the Commonwealth of Pennsylvania."

EPA Response: The EPA concurred on Pennsylvania's EE demonstration for the Lebanon County
monitor on March 6, 2018 (EPA-HQ-OAR-2017-0548-0331). As stated in the EPA's December 20, 2017
letter to Governor Tom Wolf (EPA-HQ-OAR-2017-0548-0145), if the EPA approves the EE
demonstration, the 2014-2016 design value for the Lebanon County monitor would move from violating
to attaining the 2015 ozone NAAQS. In that case, the EPA would revise its recommendation for the
Harrisburg-York-Lebanon-Lancaster Area, which includes Dauphin County, from nonattainment to
attainment/unclassifiable. Therefore, the EPA is designating the Harrisburg-York-Lebanon-Lancaster
Area, including Dauphin County, as attainment/unclassifiable. Because the commenter's concern about
Dauphin County being designated nonattainment is resolved by the EE approval, the EPA will not
respond to the commenter's objections to EPA's five factor analysis in the TSD.

Pittsburgh- Beaver Valley, PA

Comment: A commenter urged the EPA to designate the Pittsburgh-Beaver Valley area as
nonattainment. The commenter notes a large number of major NOx and VOC emissions sources in the
area, stating that Allegheny County has the highest total VOC emissions and the second highest NOx
emissions among counties in Pennsylvania. The commenter further notes several exceedances of the
NAAQS in 2017.

EPA Response: Section 107(d) of the Clean Air Act requires the EPA to designate an area as
nonattainment if it does not meet (or contributes to ambient air quality in a nearby area that does not

-------
meet) a NAAQS. The 2014-2016 design value for the Pittsburgh-Beaver Valley area is 70 parts per
billion (ppb), which meets the 2015 ozone NAAQS.

Furthermore, preliminary 2015-2017 data indicates that the area continues to meet the 2015 ozone
NAAQS. Two monitors in Allegheny County, the Harrison Township (monitor 420031008) and South
Fayette (monitor 420030067), recorded four and five exceedances of the 2015 ozone NAAQS,
respectively, during 2017, but the preliminary 2015-2017 design values for these monitors still meet the
2015 ozone NAAQS. Compliance with the 8-hour ozone NAAQS is not based on the number of
exceedances at any given monitor. Compliance with the NAAQS is determined by the design value,
which is calculated at each monitor as the 3-year average of the fourth highest values measured at the
monitoring site in each year. See 40 CFR 50, Appendix U. The preliminary 2015-2017 design values at
the Harrison Township monitor (monitor 420031008) and South Fayette monitor (monitor 420030067)
are 69 ppb and 70 ppb, respectively. The design value for the Pittsburgh-Beaver Valley area is meeting
the 2015 ozone NAAQS and this area is not contributing to a nearby violation of the 2015 ozone
NAAQS. Therefore, the EPA has designated the area as attainment/unclassifiable.

Comment: The commenter noted that Pennsylvania's 2016 recommendation letter includes HYSPLIT
back trajectories "that appear to show impact from the Pittsburgh-Beaver Valley area to violating
monitors at Bristol, Lebanon, Norristown, and Philadelphia (see Appendix C of the PA DEP analysis)."
The commenter goes on to state that EPA should consider the impact of the area's "significant" emissions
on intrastate nonattainment areas.

EPA Response: The EPA concurred on Pennsylvania's EE demonstration for the Lebanon County and
Norristown (Montgomery County) monitors. Therefore, these monitors are no longer violating the 2015
ozone NAAQS. The Bristol and Philadelphia monitors are in the Philadelphia-Wilmington-Atlantic City
(Philadelphia) nonattainment area, approximately 300 miles from Pittsburgh. The EPA has a long history
of using the consolidated statistical area (CSA) to define the starting point for analyzing nearby
contributing areas. The EPA used a similar approach with the 1997 and 2008 ozone NAAQS.

Furthermore, the CAA Amendments of 1990 set, by operation of law, the boundaries for Serious, Severe,
and Extreme nonattainment areas for the 1979 1-hour ozone NAAQS using the boundaries of the
metropolitan statistical area or consolidated metropolitan statistical area (as established by the Bureau of
the Census). The Pittsburgh area is not near the violating monitors cited in the comment, considering
both the plain English meaning of the term "nearby" and EPA's traditional interpretation of the term.
There are two CSAs and eight or nine counties from the eastern edge of the Pittsburgh-Beaver Valley area
and the violating Bristol and Philadelphia monitors in Bucks and Philadelphia Counties. Therefore, EPA
did not consider the Pittsburgh area's emissions in the Agency's five factor analysis to determine the
boundaries of the Philadelphia and Harrisburg-York-Lebanon-Lancaster nonattainment areas.

Reading. PA

Comment: A commenter supported the EPA's recommendation that Berks County in the Reading area
be designated as nonattainment. The commenter notes that Berks County has a violating monitor with a
2014-2016 design value above the NAAQS at 0.071 ppm and that preliminary 2017 data shows several
ozone exceedances in Berks County, "indicating continual air quality challenges for the area which could
contribute to 2015-2017 Design Values above the 2015 Ozone NAAQS."

EPA Response: In the EPA's December 20, 2017 letter to Governor Tom Wolf (EPA-HQ-OAR-2017-
0548-0145), EPA stated its intention to designate Berks County as nonattainment, as the Reading
nonattainment area. However, the EPA also stated that, if the EPA approves Pennsylvania's EE
demonstration, the 2014-2016 design value for the Berks County monitor would move from violating to

-------
attaining the 2015 ozone NAAQS. In that case, the EPA would revise its recommendation for the
Reading Area to attainment/unclassifiable. The EPA concurred on Pennsylvania's EE demonstration for
the Reading airport ozone monitor (monitor 420110011) in Berks County on March 6, 2018 (EPA-HQ-
OAR-2017-0548-0331). Therefore, the EPA is designating Berks County as attainment/unclassifiable.
Furthermore, preliminary 2015-2017 data indicates that, with the EPA's concurrence on Pennsylvania's
EE demonstration, Berks County continues to meet the 2015 ozone NAAQS.

3.2.4. EPA Region IV

Comments received regarding areas in Region IV are addressed below in section 3.2.11.

3.2.5. EPA Region V

3.2.5.1 Michigan

Comment: The State of Michigan requested that only parts of Muskegon and Allegan Counties be
designated as nonattainment. The townships that Michigan is recommending for nonattainment extend
from 8.8 miles to 17.4 miles from the shore for Muskegon County and from 15.5 to 19.2 miles from the
shore for Allegan County. Michigan based its position on the following:

• A 2009 Lake Michigan Ozone Study concluded that ozone pollution in western Michigan is
dominated by transport.

• 2016 design values for eight West Michigan monitors were compared against their distance from
shore. The closest attaining monitor (Jenison in Ottawa County) is 18.31 miles from the shore.

• 82% of the NOx and VOC point source emissions in Allegan County are emitted within the
proposed nonattainment area and 98% of the NOx and VOC point source emissions in Muskegon
County are emitted within the proposed nonattainment area.

• HYSPLIT ozone modeling results show that back trajectories on high ozone days are when air
parcels primarily travel from the Southwest and pass over the Chicago, II-Gary, IN area before
arriving at the Muskegon and Holland monitors. The NOx and VOC emissions from the west
Michigan counties are fairly minimal compared to the 274,440 tons of NOx and 206,171 tons of
VOC emitted from the Chicago CSA.

• Photochemical modeling of ozone over the Lake Michigan area demonstrate that ozone
concentrations do not attain the standard within a narrow band along the shoreline.

EPA Response: We agree with the state's recommendations above noting that the meteorological data
strongly indicates that the violating monitors in these counties are predominantly affected by the transport
of emissions over Lake Michigan. The township boundaries selected by Michigan are generally consistent
with the area along the Muskegon and Allegan shores that could be expected to exceed the ozone
standard. To the extent in-county emissions sources may also contribute to violations of the standard, we
note, as recognized by the state, a significant portion of the emissions within the two counties are also
included within the state's recommended boundary for each nonattainment area.

Comment: A commenter claimed that Ottawa County (in Western Michigan) should be designated as
nonattainment for ozone because Ottawa County emissions impact violating monitors in Muskegon and
Allegan Counties. More specifically:

• 22,558 tons of VOC and NOx from Ottawa County

-------
• JH Campbell plant (on western shore of Ottawa) emitted 5,049 and 143 tons of NOx and VOC,
respectively, in 2014. These emissions reflect control of NOx emissions at two of its three units
since 2011

• The commenter stated that J.H. Campbell can contribute ozone at levels exceeding 1 percent of
the 2015 ozone NAAQS in Allegan, Berrien and Muskegon Counties based on 2011 daily
emissions levels that J.H. Campbell continues to exceed since the NOx emissions controls were
installed on two if its three units.

• CAMx APCA OSAT modeling showed a 1.29 ppb impact in Berrien County and a 0.82 ppb
impact in Allegan County on June 13, 2011 when emissions from J.H. Campbell were 16.25 tons.
A level that J.H. Campbell exceeded on four days during the 2017 ozone season.

• CAMx APCA OSAT modeling showed a 1.07 ppb impact in Muskegon County on May 29, 2011
when emissions from J.H. Campbell were 17.62 tons.

• In 2014, J.H. Campbell accounted for about 40% of Ottawa County NOx emissions.

EPA Response: The EPA disagrees that Ottawa County should be designated as nonattainment based on
a weight of evidence analysis, which includes the following assessments:

• As reflected by HYSPLIT modeling and by analysis in the 2009 Lake Michigan Ozone Study, the
violating monitors in Western Michigan are primarily impacted by emissions from the Chicago
CSA.

• Two of the three EGUs at J.H. Campbell are controlled with SCRs. The 2011 modeling relied on
by the commenter does not reflect these controls. While the commenter states that the J.H.
Campbell plant continues to emit NOx on some days at levels similar to or above daily emissions
from the 2011 modeling, the commenter has provided no information regarding whether those
emission levels occur on days when the relevant monitors are exceeding the 2015 ozone NAAQS,
or that meteorological conditions support that emissions on those days are transported to the
violating monitors.

Thus, EPA does not agree that it should modify the State's recommendation by designating Ottawa
County as nonattainment based on this analysis.

Comment: A commenter supported EPA's intention to designate the Detroit area as a nonattainment
area.

EPA Response: The EPA appreciates the commenters' views. In agreement with the state of Michigan's
recommendation, the EPA has designated the Detroit area a nonattainment for the 2015 ozone NAAQS.

Comment: A commenter agreed that the western Michigan counties of Allegan, Muskegon and Berrien
should be designated as nonattainment. However, because these counties are primarily affected by
transported air pollution from the Chicago metropolitan area, the commenter questioned why EPA was
recommending the counties be designated as stand-alone nonattainment areas rather than as part of an IL-
IN-WI multistate area. Including those counties as part of the larger nonattainment area would allow for
appropriately informed planning for both the downwind and the upwind areas.

EPA Response: The CAA provides that EPA should designate as nonattainment areas that are violating
the standard and nearby areas contributing to air quality at violating monitors. Several other provisions in
the CAA address longer range transport of ozone and ozone precursor emissions (see, e.g., section
110(a)(2)(D)). Consistent with how Congress identified nonattainment areas for the 1-hour standard at
the time of the CAA Amendments of 1990 (section 107(d)(4)(A)(iv)), EPA has used the Core Based

-------
Statistical Area (CBSA) or Combined Statistical Area (CSA) as a starting point for its analysis in
determining appropriate boundaries for nonattainment areas. In guidance issued in February 2016, EPA
indicated it would follow this same approach for the 2015 ozone NAAQS. Consistent with the approach
EPA stated it would follow in the February 2016 Guidance, EPA used the CBSA as the area of analysis
for the western Michigan counties and addressed the Chicago CSA separately.

3.2.5.2 Ohio

Comment: The commenter contended specific days from 2016 with high ozone should be excluded from
the Ohio monitoring data due to impacts from Canadian wildfires and references EPA Exceptional Event
request packages put together by the states of Ohio, New Jersey, and Maryland and containing evidence
that Ohio air quality was impacted by these wildfires.

EPA Response: Ohio submitted an exceptional event demonstration on November 15, 2017. The EPA
is in the process of reviewing and taking action on that submission. With respect to designations for the
2015 ozone standard, Ohio did not rely on the exceptional events demonstration to support its initial
nonattainment recommendations, nor did the state submit a revised nonattainment boundary
recommendation based on the exceptional events data. As set forth in the five factor analysis in the
December TSD, EPA is agreeing with Ohio's nonattainment recommendations for the Cleveland,
Columbus, and Cincinnati areas. Because these areas continue to show monitored violations of the 2015
ozone standard even after the exceptional events data is excluded, concurrence with the states exceptional
events demonstration would not change EPA's agreement with the State's nonattainment
recommendation. Furthermore, all of the nonattainment areas in Ohio are being classified as Marginal
nonattainment areas, which is the lowest classification; therefore, any changes that lower design values
resulting from a concurrence with the exceptional events demonstration would not affect the classification
of the areas at this time.

Cleveland, OH

Comment: A commenter noted that the EPA should include Ashtabula County in the Cleveland, Ohio
nonattainment area. Ashtabula County was part of the ozone nonattainment area under the 1997 and 2008
standards and has a monitor that is near to violating the standard with both a2014-2016 and preliminary
2015-2017 design value of 0.070 ppm. The commenter contends that major sources in Ashtabula are
located near two HYSPLIT back trajectories and appear to have the potential to influence violating
monitors.

EPA Response Section 107(d)(1) of the Clean Air Act directs the EPA to designate an area as
"nonattainment" if it is violating the NAAQS or if it is contributing to a violation of the NAAQS in a
nearby area. In the Cleveland area, monitors in Cuyahoga and Geauga Counties show a violation of the
2015 ozone NAAQS based on certified 2014-2016 monitoring data. As acknowledged by the commenter,
the 2014-2016 design value forthe Ashtabula County monitor is 0.070 parts per million (ppm), which
meets the 2015 ozone NAAQS as does the preliminary design value for 2015-2017.Therefore, air quality
data does not require that Ashtabula County be included in the Cleveland nonattainment area based on a
determination that the county is violating the standard. The EPA evaluated whether emissions in
Ashtabula County were contributing to violations of the 2015 ozone NAAQS at the monitors in Cuyahoga
and Geauga Counties based on the weight-of-evidence of the five factors identified in EPA's Designation
Guidance. EPA notes that only 2 HYSPLIT back trajectories pass through Ashtabula County meaning
that this meteorological pattern was relatively infrequent on days with ozone above the standard evaluated

21

-------
in the TSD. When looking at the full five-factor analysis, EPA concludes that while Ashtabula County
has moderate emissions as compared to other counties in the area of analysis (17% and 23% of Cuyahoga
County's NOx and VOC emissions, respectively), the county ranks relatively low in population density
and VMT and has only two HYSPLIT trajectories that pass through the county on days that the violating
monitors are exceeding the NAAQS. Given the overall weight of evidence of all five factors, including
the relative rarity of trajectories passing through Ashtabula County and the greater emission levels of
other counties in the area of analysis, EPA determined not to modify the State's recommendation that
Ashtabula County not be included as part of the Cleveland nonattainment area.

Comment: The commenter agreed with EPA's intended boundary for the Cleveland, Ohio area.

EPA Response: The EPA appreciates the commenters' views. The EPA is finalizing the boundary for
the Cleveland Area as provided in the December TSD. The EPA is designating Cuyahoga, Geauga, Lake,
Lorain, Medina, Portage, and Summit Counties as the Cleveland nonattainment area for the 2015 ozone
NAAQS.

Columbus, OH

Comment: The commenter contended that EPA should include Ross County in the Columbus, Ohio
nonattainment area. The commenter argues that the P H Glatfelter paper facility in Ross County emitted
over 1,000 tons of NOx in 2016 from its two coal-fired boilers, which appear to impact the Franklin
County monitor.

EPA Response: The P H Glatfelter facility converted their No. 7 and 8 coal-fired boilers (B002 and
B003) to natural gas by January 31, 2017. This conversion is estimated to result in a 76% reduction in
NOx emissions from these units. Actual reported emissions from these units in 2016 was 1,103.3 tons
and emissions in 2017 are expected to be approximately 266 tons.

Further, the EPA does not determine contribution based on a single factor, but rather on the weight-of-
evidence of the five factors set forth in EPA's February 25, 2016 designation guidance. The commenter
has not provided a compelling reason for revising the conclusion EPA reached in the five factor analysis
contained in the TSD for the Columbus area.

Comment: The commenter agreed with EPA's intended boundary for the Columbus, Ohio area

EPA Response: The EPA appreciates the commenters' views. As documented in the TSD for this area,
the EPA is designating Delaware, Fairfield, Franklin, and Licking Counties as the Cleveland
nonattainment area for the 2015 ozone NAAQS, as provided in the December TSD.

3.2.5.3 Wisconsin

Comment: The Wisconsin Department of Natural Resources (WDNR), and Governor Walker submitted
comments asking EPA to foremost consider designating the entire state as attainment of the 2015 ozone
NAAQS as per the Governor's original September 21, 2016, recommendation, since WDNR believes
ozone levels in Wisconsin are beyond the state's control due to out-of-state emissions and meteorology. If
EPA does not designate the entire state as attainment, then WDNR urged EPA to only designate as
nonattainment narrow parcels of land near the Lake Michigan shoreline around the violating monitors

-------
(essentially dismissing any contribution analysis), since WDNR believes these monitors are not
meaningfully affected by in-state emissions.3 For these reasons, U.S. Congressman F. James
Sensenbrenner Jr. of Wisconsin (0285) would also like EPA to consider reducing the scope of EPA's
intended nonattainment designations in Wisconsin and believes pollution from beyond Wisconsin's
borders should be accurately accounted for, especially as it pertains to monitors located along Lake
Michigan.

In its April 2017, technical support document (TSD), WDNR provided a geographic estimate of areas
experiencing nonattainment air (i.e. with design values > 70 ppb) based on an estimate of a "70 ppb ozone
contour line" near the shoreline of Lake Michigan. WDNR's 70 ppb contour is based on a best-fit line
developed by plotting the design values of six of the eight violating monitors and one of the four
attainment monitors located within four miles of the Lake Michigan shoreline versus the location of each
of these seven monitors expressed as distance in miles from the shoreline of Lake Michigan. The extent of
this 70 ppb contour was described in WDNR's April 2017 TSD and again in WDNR's comment letter
(0300), which included specific modifications to the location of the 70 ppb contour in Racine County (4.2
miles inland) and in Sheboygan County (2.3 miles inland) relative to WDNR's original recommendations
regarding these counties provided in its April 2017 TSD.

As a starting point for the distance from the lakeshore going inland, WDNR suggested EPA use the U.S.
Army Corps of Engineers (USACE) Ordinary High Water Mark (OHWM), which for Lake Michigan is
581.5 feet, and cited the USACE website for the Detroit District's Regulatory Office.4 According to
WDNR the USACE OHWM is permanent (e.g., it does not change based on water level fluctuations), is
legally-defined, can be easily identified, and is already widely-used used in federal regulatory
applications.

EPA Response: Because there are valid Federal Reference ozone monitors indicating violations of the
2015 ozone NAAQS, the CAA requires that EPA designate as nonattainment those areas that do not meet
the NAAQS, regardless of the source of high ozone levels. EPA's assessment of the boundaries for
nonattainment areas associated with the violating monitors is contained in the TSD, and includes
consideration of the information supplied by Wisconsin in submissions to EPA for designations. EPA's
final nonattainment boundaries are consistent with the state's input.

Here we address the feasibility of Wisconsin's suggestion of using the OHWM to delineate a distance
inland from the shoreline rather than EPA's roadway-based approach. The federal high water mark for
Lake Michigan is currently set at 581.5 feet.5 The USACE OHWM is a jurisdictional benchmark for

3	In a September 21, 2016, letter to EPA from its Governor, Wisconsin recommended that the entire state be
designated as attainment for the 2015 ozone NAAQS, despite having violating monitors, since, in Wisconsin's
opinion, elevated ozone levels in Wisconsin are primarily due to emissions originating from other states, recent
ozone levels in Wisconsin have greatly improved, and Wisconsin has already significantly reduced ozone-causing
emissions. Later in an April 20, 2017, technical support document (TSD), WDNR submitted to EPA additional
information to support the Governor's recommendation including estimates of the geographic extent of the areas
in Wisconsin with design values above 0.070 ppm (70 ppb). Wisconsin requested that if EPA designates
nonattainment areas in Wisconsin, the EPA should ensure that the geographic scope of these areas is minimized.
Wisconsin emphasized in its April 20, 2017, submittal that these descriptions should not be construed as a
recommendation for a potential nonattainment area designation for the 2015 ozone NAAQS.

4	http://www.lre.usace.army.mil/Missions/Great-Lakes-lnformation/Links/Ordinary-High-Water-Mark-and-Low-
Water-Datum/

5	http://www.lre.usace.army.mil/Missions/Great-Lakes-lnformation/Links/Ordinary-High-Water-Mark-and-Low-
Water-Datum/

23

-------
administering its regulatory program in navigable waterways under Section 10 of the Rivers and Harbors
Act and Section 404 of the Clean Water Act. The OHWM is the line on the shore coincident with the
elevation contour that represents the approximate location of the line on the shore established by
fluctuations of water and indicated by physical characteristics such as shelving, destruction of terrestrial
vegetation, presence of litter or debris, or changes in the character of soil. Wisconsin did not provide
details on the Wisconsin OHWM and only referenced the federal OHWM via the link to the USACE
Detroit District's Regulatory Office without providing additional details or maps depicting what
Wisconsin's areas would look like if they were to be based upon the federal OHWM for Lake Michigan.
Whether using the federal OHWM or simply using a standard map of the shoreline, a distance inland to
delineate a nonattainment area is problematic, since the distance inland might bisect individual source
facilities, thus making it difficult to regulate which part of the source is in the nonattainment area and
which part of the source is outside of the nonattainment area. With roadways, it is clear as to which
portion of a facility is located in a nonattainment area if a facility has several building units one of which
may be located across a roadway from another.

Additional comments and responses regarding specific areas as well as the topic of contribution are
summarized and addressed in further detail below.

Comment: Several public health groups and an environmental law and policy group commented on
EPA's Wisconsin intended nonattainment areas. One of these commenters agreed with EPA's intent to
designate the Door, Sheboygan, and Manitowoc partial county areas as nonattainment. However, since
Door, Sheboygan and Manitowoc counties are primarily affected by transported air pollution from the
Chicago metropolitan area, this commenter questioned why EPA is recommending the counties be
designated as stand-alone nonattainment areas rather than as part of the Chicago IL-IN-WI multistate
area. Some commenters believed that Sheboygan County, Manitowoc County, and Door County,
Wisconsin, should be included as part of the Chicago IL-IN-WI nonattainment area since the data indicate
these areas receive transported emissions from the Chicago metro area. The commenter indicated, "EPA's
current proposal limits these counties' ability to ensure that they meet the standard quickly. EPA argues
that these represent areas outside of the Chicago metropolitan area, so they separate the designations.

With limited sources under their authority, these counties cannot take actions individually to control or
reduce emissions coming into these counties. Fortunately, if national measures including the 2016 Cross-
State Air Pollution Rule and the 2014 Tier 3 standards remain in effect, these counties will have federal
help to meet the standards. However, if included as part of the larger metro area planning and
implementation, those counties would have a stronger voice in ensuring improvements to protect the
health of their residents." One commenter noted that even though the Milwaukee area is between these
northern counties and the Chicago metropolitan area, back trajectory analyses show that they are affected
by transport from the Chicago area. A commenter indicated Door County is sufficiently remote that an
individual designation may be more appropriate; however, the commenter asked that EPA consider this
issue for all three counties.

EPA Response: The EPA acknowledges the receipt of these comments. While it might make sense to
include Sheboygan and possibly Manitowoc in the Milwaukee area (not the Chicago area since it is not
contiguous with Sheboygan and Manitowoc), EPA carefully considered this option and decided to
maintain its long-standing method of starting with the larger of the CSA or CBSA as the initial area of
analysis for each area with violating monitor(s).6 Sheboygan and Manitowoc are each single county areas.
The state has the discretion to make decisions regarding how best to target attainment planning for its
nonattainment areas including its single (partial) county areas that are upwind of in-state urban areas and
EPA stands ready to work cooperatively with the state in these efforts. As for contributions from out-of-

6 httpsi//www.epa.gov/ozone-designations/epa-guidance-area-designations-2015-ozone-naaqs

24

-------
state upwind areas (e.g. the Chicago area), the transport provisions of the Clean Air Act require each state
to analyze their contributions to downwind nonattainment and maintenance receptors.

Comment: A commenter emphasized that the EPA needs to act swiftly to formally designate
nonattainment areas. EPA's actions have resulted in unnecessary and prolonged delay in areas taking
steps to protect their residents and develop a successful plan to reduce emissions.

Several commenters urged the EPA to allow no further delays to complete this review and protect the
health of all Americans.

EPA Response: By this action, the EPA is finalizing the designation for all areas except one and EPA
will complete the designation process by July 17, 2018.

Comment: One commenter submitted detailed concerns about the future emissions of criteria pollutants,
hazardous air pollutants, and greenhouse gases from the proposed Foxconn facility in Racine County and
the detrimental impact these emissions would have on air quality, human health, and the environment.
The commenter cited EPA's December 22, 2017, TSD7 and stated that Racine County is already within
an EPA-designated non-attainment area for ozone. Because Foxconn will be a major source polluter of
both VOCs and NOx, and will account for about 5% of the annual local production of these contaminants
the facility will contribute significantly to local ozone production.

EPA Response: To clarify, Racine County is not currently within an EPA-designated nonattainment area
for ozone. The EPA addresses in the final nonattainment area TSD the conclusions reached for the final
boundary for the Milwaukee area, including Racine County. EPA's 5-factor analysis includes an
evaluation of emissions; however, it does not take into consideration future projected emissions in the
area of analysis. Therefore, the analysis did not consider the future emissions from the proposed Foxconn
facility. These additional emissions would be addressed by the Clean Air Act's preconstruction
permitting requirements.

Comment: Governor Walker commented that WDNR has data and modeling to show that the vast
majority of emissions leading to nonattainment come from out-of-state. WDNR (0300) commented that
LADCO source apportionment modeling results show that out-of-state emissions are responsible for the
elevated ozone concentrations observed at Wisconsin's lakeshore monitors, including those in the 5-
county Milwaukee area. For instance, only 7% of the ozone at Milwaukee's Bayside monitor originated
from Wisconsin and much larger portions came from out-of-state (e.g., 26% from Illinois and roughly
20% from "boundary conditions" also denoted as "BC" which they interpret to represent international
sources). WDNR commented that Chicago emissions dwarf those of the Wisconsin emissions, estimating
that the Chicago area emits 79-81% of the NOx and VOC emissions in the southwestern Lake Michigan
area.

Several cited the source apportionment modeling in the WDNR submittal which show the contribution
from the entire state of Wisconsin to be approximately 12%, 15%, and 7% at the Harrington, Grafton, and
Bayside monitors respectively. Commenters believed that it takes 17,349 to 25,604 tpy of precursor
emissions to make 1 ppb of ozone.8 One commenter indicated this would translate to ozone
concentrations of 0.49, 0.41, and 1.16 ppb on the high end of the range for Racine, Washington, and

7	https://www.epa.goy/sites/production/files/20	Dcuroents/wi120d tsd rewrite final.pdf

8	See WDNR "Correspondence Memorandum, Ozone Air Quality Analysis for a PSD Permit for Aarrowcast -
Shawano," Dated June 7, 2012: "...it is estimated that it takes from 17,349 tons per year to 25,604 tons per year of
total VOC and NOx reductions to result in a 1 ppb reduction in ozone concentration."

25

-------
Waukesha county emissions, respectively. This commenter indicated that "assuming this relationship to
be true, sources in Racine County would result in an increase of no more than 0.5 ppb ozone for any
monitor in Wisconsin."

EPA Response: The EPA addressed in the final nonattainment area TSD the conclusions reached for the
source apportionment modeling cited in WDNR's April 2017 submittal and February 2018 comments.
EPA notes that WDNR did not include in its April 2017 TSD, nor in its February 2018 comment
letter/TSD, a detailed, transparent description of the source apportionment modeling. A complete and
thorough description of any modeling analysis including details regarding the modeling platform,
emissions inventory, model options, post-processing methodology, and model performance evaluation
would be necessary to fully and objectively assess the modeling analysis.

With respect to the commenters assertion that it takes 17,349 to 25,604 tpy of precursor emissions to form
1 ppb of ozone, firstly the commenters appear to have misquoted the document to which they were
referring.9 Secondly, ozone formation chemistry is complex and nonlinear. It varies based upon many
factors, such as magnitude of precursor emissions, local meteorology, geographical features, etc. Any
emissions of NOx and VOC have the potential to photochemically react to form ozone. Studies show that
single sources of ozone precursor emissions can have measurable nearby downwind ozone impacts.1011
Maximum impacts typically occur within 50-100 km (31-62 miles) from the source and can occur up to
200 km (124 miles) away from the source. Actual sources10 and hypothetical single sources11 have been
analyzed for their potential downwind ozone impacts. As a high-end example, a hypothetical source in
northern Illinois emitting 500 tpy of NOx can have a maximum 8-hour downwind ozone impact of 3.88
ppb.12 As a lower end example, a hypothetical source in northwest Indiana emitting 500 tpy of NOx can
have a maximum 8-hour downwind ozone impact of 1.15 ppb.13

Comment: WDNR commented that EPA failed to acknowledge that WDNR submitted two sensitivity
modeling scenarios showing emission reductions in Wisconsin would not meaningfully impact ozone
design values along the lake and therefore local emissions have little to no impact on lakeshore ozone
concentrations. One of these scenarios involved modeling a 10% reduction in both NOx and VOC
emissions from all sectors excluding on-road and biogenic emissions from a 10-county area in southeast
Wisconsin. The other scenario involved "zero out" modeling of emissions from all sectors in Sheboygan
County (excluding biogenic emissions). The modeling results showed that eliminating Sheboygan County
emissions would not reduce the design values at the county's monitors and the emissions reductions in the
10 southeast Wisconsin lakeshore counties would not meaningfully impact ozone design values along the

9 See WDNR "Correspondence Memorandum, Ozone Air Quality Analysis for a PSD Permit for Aarrowcast -
Shawano," Dated June 7, 2012: "...it is estimated that it takes from 17,349 tons per year to 25,604 tons per year of
total VOC and NOx reductions to result in a 1 ppb reduction in ozone concentration."

10 Baker, K. R. and Kelly, J. T.: Single source impacts estimated with photochemical model source sensitivity and
apportionment approaches, Atmospheric Environment 96, 266-274, 2014.

11 Baker, K. R., Kotchenruther, R. A., and Hudman, R. C.: Estimating ozone and secondary PM2.5 impacts from
hypothetical single source emissions in the central and eastern United States, Atmospheric Pollution Research, 7,
122-133, 2016.

12 EPA's Guidance on the Development of Modeled Emission Rates for Precursors (MERPs) as a Tier 1
Demonstration Tool for Ozone and PM2.5 under the PSD Permitting Program. Draft for Public Review and
Comment. December 2, 2016.

13 EPA's Guidance on the Development of Modeled Emission Rates for Precursors (MERPs) as a Tier 1
Demonstration Tool for Ozone and PM2.5 under the PSD Permitting Program. Draft for Public Review and
Comment. December 2, 2016.

-------
lake. WDNR indicated the modeling suggests that even Wisconsin's highest-emitting counties
(Milwaukee and Waukesha) do not have a meaningful impact on lakeshore ozone concentrations, which
is further evidence that emissions from these two counties are overwhelmed by those from the upwind
regions, including the Chicago area, which emits six to seven times more NOx and VOC.

EPA Response: The EPA addresses in the final nonattainment area TSD the conclusions reached for the
sensitivity modeling cited in WDNR's April 2017 submittal and February 2018 comments. EPA also
notes that the 10% emissions cut14 model simulation cannot alone be used as a contribution analysis. Due
to the nonlinear nature of ozone chemistry this type of sensitivity analysis can be used to determine the
impacts of the specific emissions changes modeled but cannot be used to infer the overall impact that
results from total emissions from the sources in question.

Comment: WDNR commented that EPA did not include or reference WDNR's wind rose analyses,
which more accurately reflect the complex lakeshore environment than does the HYSPLIT back
trajectory model relied upon by EPA, and which confirms that ozone concentrations exceeding 70 ppb
occur when winds originate offshore. EPA inappropriately relied on HYSPLIT back trajectories to make a
connection between local emissions and locally-monitored ozone levels. WDNR believes that only the
100 m HYSPLIT back trajectories are potentially relevant when considering associations with ground-
level monitored ozone levels. Most traveled over the lake (high level trajectories represent synoptic and
not local flow). In contrast, direct measurements at these monitors found that, for virtually every single
hour with ozone concentrations above 70 ppb, the air masses came from over the lake: from 155-185
degrees for the Harrington Beach monitor and 135-175 degrees for the Grafton monitor. The wind roses
from the other lakeshore monitors showed similar results: ozone-rich air was delivered to the monitors
almost exclusively from over Lake Michigan. This comparison conclusively shows that HYSPLIT
underestimates the role of the lake in delivering ozone-rich air to this monitor and overestimates the
impact of emissions from inland portions of the counties under discussion.

Some commenters believed there is no meaningful contribution from Racine, Washington, and Waukesha
counties. Several of these commenters cited the wind roses in WDNR's April, 2017, submittal,
specifically the ones from the Harrington Beach and Grafton monitoring sites which show winds from the
south southeast during days in 2014-2016 with 1-hour ozone values greater than 70 ppb. These
commenters also cited an April 17, 2016, HYSPLIT 48-hour back trajectory from the Bayside monitor in
WDNR's submittal, showing air parcels traveling from the southeast and over the lake. Other commenters
believed that the EPA HYSPLIT back trajectories support excluding Wisconsin counties since the lowest
altitude trajectories are from over the lake. Specifically, one commenter believed meteorological data do
not support the finding that Racine County emissions contribute (e.g. low level trajectories are mainly
coming from over the lake and even on days that surface or mid to upper level air masses cross over
Racine County, these air masses have crossed over the greater Chicago area with its significantly greater
contribution of ozone precursors prior to crossing over Racine County. One commenter seemed to believe
the location of where the back trajectories originate is indicative of where the emissions are coming
from.) Another commenter thought that exceedances have only occurred on days and times when wind is
carrying ozone from over Lake Michigan (and that EPA ignored this fact in its TSD). This commenter
indicated that the Racine County population has declined and less than one quarter of Racine County
residents commute to or through Milwaukee or Ozaukee counties, so the population and activity data
indicate minimal contribution. Another commenter believed that the Milwaukee area monitors have never

14 As noted in the final nonattainment area TSD, the emissions cut to anthropogenic emissions (including on-road
sources) in the sensitivity modeling described is likely closer to a 6-8% emissions reduction.

27

-------
exceeded a standard when monitoring air coming from these counties, rather, the monitors are measuring
ozone produced from precursor emissions generated elsewhere.

EPA Response: The EPA addressed, in the final nonattainment area TSD, the conclusions reached for
both EPA's HYSPLIT trajectories and the HYSPLIT trajectories cited in WDNR's April 2017 submittal
and February 2018 comments. Wisconsin's wind roses show that the wind direction for most of the near
lakeshore monitors during the 2013-2016 hours of high ozone (>70 ppb) was predominately from the
southeast quadrant, except as follows. For the two Sheboygan sites, the predominate wind direction was
from directly south or slightly south southwest (e.g. from the direction of the Milwaukee area). For the
Milwaukee Health Center site, the wind direction was often from the south southeast, but also often from
the southwest. For the Door County (Newport) site, the wind direction was often from the
southwest/south southwest. Wisconsin seemed to use the wind roses to conclude that since the hours of
high ozone typically occur when the local wind direction at the monitors is generally (but not at all times)
coming from over the lake, that the ozone is coming from precursor emissions originating out-of-state.
Wind roses that depict wind speed and direction reported in surface observations can be used to estimate
wind speed and frequency for the immediate area of the observation, in this case the ozone monitor site,
but that representativeness diminishes with distance from the site. Extrapolating the wind pattern depicted
in a wind rose to a larger area affords a great deal of influence to the wind measured at that one site, ten
meters above the ground, and to any small-scale geographic influences that may affect wind at that site.
The HYSPLIT trajectories used in EPA's analyses were determined by the Eta Data Assimilation System
(EDAS) an archive of meteorological parameters across a nationwide grid at many vertical levels, and
incorporating surface and upper-air observations as well as wind profiler, radar, and aircraft data. Unlike
wind roses, HYSPLIT backward trajectories are just as representative of atmospheric conditions at a
distance from the trajectory starting point as they are at the starting point. HYSPLIT trajectories based
upon EDAS more accurately reflect the pertinent meteorological influences in the area under examination
than does a wind rose based upon single-point observations. HYSPLIT back trajectories at starting
heights 100, 500, and 1000 meters above ground level represent levels typically within the atmosphere's
mixed layer at the monitor, yet above the influence of local terrain. Trajectories at these three starting
heights are relevant in assessing transport of air parcels for potential contribution to ozone concentrations
at the trajectory starting point. With respect to the comments regarding the lower level (100 m)
trajectories, it is important to note that the lower level (100 m) trajectories do not exclusively occur over
the lake.

Comment: WDNR commented that despite many lines of credible evidence provided by DNR showing
that local emissions have little to no impact on areas of Wisconsin where ozone levels exceed 70 ppb,
EPA's intended nonattainment areas appear designed to include as many local sources of these emissions
as possible. WDNR comments that local precursor emissions do not meaningfully impact the ozone levels
at the violating monitors, and therefore EPA should not consider in-state emissions contributions. WDNR
also commented that in-state emissions reductions would not meaningfully impact the ozone levels at the
violating monitors. For example, WDNR indicates that NOx and VOCs from the 5-county Milwaukee
area decreased by 25% and 33%, respectively, from 2008 to 2014, however, ozone design values in the
Milwaukee area remained relatively flat during this period. With respect to Manitowoc County, WDNR
believed that EPA should not consider emissions sources located in the county since WDNR believes
emissions are low and ozone-rich air reaches the Manitowoc County monitor exclusively from over Lake
Michigan (as discussed in the wind rose section above). WDNR indicated that Manitowoc County
emissions are similar in magnitude to those of Ozaukee County and Door County, which, with respect to
Door County, EPA concluded "do not significantly contribute to ozone concentrations in the area itself or
to other areas."

28

-------
EPA Response: The EPA addressed in the final nonattainment area TSD the emissions-related
information cited in WDNR's April 2017 submittal and February 2018 comments. EPA notes that any
emissions of NOx and VOC have the potential to photochemically react to form ozone. Studies show that
single sources of ozone precursor emissions can have measurable nearby downwind ozone impacts.1516
Maximum impacts typically occur within 50-100 km (31-62 miles) from the source and can occur up to
200 km (124 miles) away from the source. Actual sources16 and hypothetical single sources17 have been
analyzed for their potential downwind ozone impacts. As a high-end example, a hypothetical source in
northern Illinois emitting 500 tpy of NOx can have a maximum 8-hour downwind ozone impact of 3.88
ppb.17 As a lower end example, a hypothetical source in northwest Indiana emitting 500 tpy of NOx can
have a maximum 8-hour downwind ozone impact of 1.15 ppb.18 Therefore, it is important to evaluate
nearby sources of potentially contributing precursor emissions. EPA also noted that the reference in the
WI TSD provided meteorology-adjusted ozone trends covering a longer time-period (2000-2015) which
do show decreasing trends in Milwaukee. Wisconsin did not provide any information on meteorology-
adjusted trends for the relatively short time-period (2008-2014) discussed in their submission or address
changes in contributing sources outside of the Milwaukee area over this time-period.

Regarding the comment about the magnitude of Door County emissions, given that Door County is the
last (northernmost) county in the series of eastern Wisconsin counties receiving transport from upwind
high-emitting urban areas like Green Bay, Milwaukee, and Chicago, EPA was able to comfortably use our
discretion to classify this area as an RTA. A similar analysis and determination is not appropriate for the
other areas that the commenter mentioned, specifically Ozaukee County, which is part of the Milwaukee
CSA or Sheboygan and Manitowoc counties which, while they are each their own area, are immediately
downwind of the Milwaukee area and each are adjacent to a metropolitan statistical area (CSA/CBSA),
which disqualifies them as potential RTAs, whereas northern Door County is not adjacent to a
CSA/CBSA.

Comment: Numerous commenters alluded to or expressed concerns about economic growth and business
operations. In particular, one commenter expressed concern about possible implications of nonattainment
such as WDNR developing SIP mandating emissions reductions from existing sources and emissions
offsets for the permitting of new sources, with specific emphasis on an electronics manufacturing district
planned for construction in Racine County. One commenter indicated that EPA's potential nonattainment
designations impose a heavy regulatory load on businesses and industry and threaten the well-being of the
state's economy. Another commenter was concerned about a negative connotation associated with and
difficulty attracting new businesses and residents to nonattainment areas.

EPA Response: Under section 107(d) of the CAA, EPA is required to designate as nonattainment an area
that is violating a new or revised national ambient air quality standard or that contributes to a nearby
violation. EPA determines an ozone violation based on certified quality assured monitoring data. Ozone is

15 Baker, K. R. and Kelly, J. T.: Single source impacts estimated with photochemical model source sensitivity and
apportionment approaches, Atmospheric Environment 96, 266-274, 2014.

16 Baker, K. R., Kotchenruther, R. A., and Hudman, R. C.: Estimating ozone and secondary PM2.5 impacts from
hypothetical single source emissions in the central and eastern United States, Atmospheric Pollution Research, 7,
122-133, 2016.

17 EPA's Guidance on the Development of Modeled Emission Rates for Precursors (MERPs) as a Tier 1
Demonstration Tool for Ozone and PM2.5 under the PSD Permitting Program. Draft for Public Review and
Comment. December 2, 2016.

18 EPA's Guidance on the Development of Modeled Emission Rates for Precursors (MERPs) as a Tier 1
Demonstration Tool for Ozone and PM2.5 under the PSD Permitting Program. Draft for Public Review and
Comment. December 2, 2016.

-------
a regional pollutant and is readily transported both short and long distances. To determine whether a
nearby area is contributing to a violation, EPA recommended that states conduct a technical analysis
based on a number of factors listed in the designation guidance for the 2015 ozone NAAQS, including air
quality data, emissions and emissions-related data, meteorological data, and geography/topography, and
jurisdictional boundaries. In evaluating whether to modify a state's designation recommendation, EPA
also considered those factors. The justification for including Wisconsin counties (whole or partial) in a
nonattainment area is provided in EPA's technical support document for each area. In determining
whether an area should be designated nonattainment, EPA did not consider economic impacts because
that is not relevant for determining whether an included area is violating the NAAQS or is a nearby area
that is contributing to a violation as provided under CAA section 107(d). In determining how best to
manage air quality in nonattainment areas in order to meet NAAQS, to the extent the CAA does not
mandate specific control measures, states may consider economic concerns in development of their state
implementation plans.

Milwaukee, WI

Comment: In addition to WDNR, several parties commented specifically on the Milwaukee intended
nonattainment area. Some of their comments are addressed in the final TSD and other comments are
addressed below. These commenters generally wanted EPA to designate Wisconsin as attainment or
reduce the size of the intended nonattainment area(s). It should be noted that one commenter represented
portions of Ozaukee, Washington, Sheboygan, Fond du Lac, and Calumet counties. Some of these
commenters did not want Waukesha, Washington, and Racine County and another commenter did not
want the City of Waukesha to be included in the nonattainment area. Some of these commenters believed
if there is a nonattainment boundary, it should be very similar to the narrow strips of land from the
shoreline inland encompassing the violating monitors per the technical analysis document submitted to
EPA by WDNR on April 7, 2017. Commenter 0266 specifically requests that EPA reconsider Racine
County's intended designation status or, in the alternative, reduce the geographic boundary of the
proposed nonattainment zone. These commenters provided the following supporting information, which
EPA addresses below along with WDNR's additional comments that were specific to the Milwaukee area:

Comment: WDNR commented that EPA inappropriately included counties with attaining monitors (e.g.
Waukesha County) as part of intended nonattainment areas. Some commenters indicated that one or more
of the following counties did not have violating monitors: Washington, Waukesha, and Racine.

EPA Response: Waukesha, Washington, and Racine counties were not included in EPA's intended
Milwaukee nonattainment area based on monitoring data. These counties, like other counties with
attaining monitors (in other nonattainment areas) across the country, were included based on an initial
contribution analysis as described in EPA's 2015 intended ozone designations for Wisconsin TSD. EPA
determines nonattainment areas based on violating monitors and nearby areas that may be contributing
based on a 5-factor "weight of evidence" analysis. EPA's TSD goes into detail on the contribution
analysis, which is why Waukesha, Washington, and Racine counties are included in the intended
nonattainment area. While the monitoring data from Racine County was not considered in EPA's analysis
due to the lack of a 2014-2016 design value, it is worth noting that the 2015 4th high is 68 and the 2016 4th
high is 76. The preliminary 2017 4th high is 80, which results in a preliminary 2017 DV of 74 ppb (with
truncation).19 Due to a dilapidated building presenting unsafe working conditions, EPA allowed WDNR
in 2013 to shut down and move the Racine County monitor to a new location approximately 5 miles north

19 The specific methodology for calculating the ozone design values, including computational formulas and data
completeness requirements, is described in 40 CFR part 50, appendix U.

-------
of its original location where a monitor had been located since 1977. WDNR was unable to get the
monitor up and running at its new location prior to the start of the 2014 ozone season causing the gap in
continuous ozone data for Racine County.

Comment: To one commenter it appeared EPA proposed to include the 5-county area because EPA
considered the 5 counties collectively to be one statistical area. Commenter pointed to EPA designations
guidance explaining that EPA methodology typically starts with a CSA or CBSA but that each area is
analyzed on a case-by-case basis and asks that EPA use its discretion to designate a smaller area for the
Milwaukee nonattainment area based on supporting info (see other comments and responses with respect
to the supporting info from this commenter).

EPA Response: As per our long-standing practice, EPA typically starts with the larger of the CSA or the
CBSA as the area of analysis for ozone nonattainment area designations.20 The Milwaukee CSA is an 8-
county area. EPA's final TSD addresses why the final nonattainment area is smaller than the CSA.

Comment: A couple of commenters noted that the Milwaukee area experiences lake breeze meteorology
just like Sheboygan and Manitowoc counties and commenter did not think that EPA addressed the lake
breeze meteorology with respect to the Milwaukee area. One commenter indicated, "It is WDNR's
opinion that because pollutants travel exclusively on coastal breezes and are detected by monitors in close
proximity to the coastline, it is unfair to use data from these monitors as justification for county-wide
non-attainment designations. This is an assessment shared by the EPA regarding the Sheboygan County
coastal air monitor. Because the EPA has already demonstrated agreement with this justification for
Sheboygan County, it should apply that same determination to Ozaukee County air monitors."

EPA Response: That the Milwaukee area experiences lake breeze meteorology is addressed in EPA's
final TSD for the nonattainment area. EPA's 5-factor contribution analysis, which included consideration
of lake breeze meteorology and local precursor emissions, started with the 8-county Milwaukee CSA.
That an area experiences lake breeze meteorology is alone not mutually exclusive with a determination
that an area may also contribute to its own ozone violations. Sheboygan and Manitowoc are separate
areas. For each area (Sheboygan and Manitowoc), the initial area of analysis was the full county. For
each, the 5-factor contribution analysis included consideration of lake breeze meteorology and local
precursor emissions. EPA notes that a reason EPA was able to justify honoring Wisconsin's request to
reduce the size of the Sheboygan County area from a full county to a partial county area is the existence
of the second ozone monitor in Sheboygan county which is attaining the standard coupled with WDNR's
lake breeze inland penetration distance analysis specific to the two Sheboygan County monitors. EPA
extended that reasoning to Manitowoc County (despite Manitowoc not having an inland attaining
monitor), since Manitowoc County is adjacent to and immediately north of Sheboygan County.

Comment: Some commenters indicated that the LADCO source apportionment modeling in the WDNR
submittal shows approximately 20% contribution from international transport and commenter indicated
that photochemical modeling from the Midwest Ozone Group shows this as well. Two commenters stated
the following: "Notwithstanding any other provision of law, any State that establishes to the satisfaction
of the Administrator that, with respect to an ozone nonattainment area in such State, such State would
have attained the national ambient air quality standard for ozone by the applicable attainment date, but for
emissions emanating from outside of the United States, shall not be subject to the provisions of section
7511(a)(2) or (5) of this title or section 751 Id of this title [of the CAA, and] since Wisconsin has shown,
clearly, that more than 20% of the ozone concentration at violating monitors is from sources outside of

20 httpsi//www.epa.gov/ozone-designations/epa-guidance-area-designations-2015-ozone-naaqs

31

-------
the United States, our state qualifies for this relief. As a result, U.S. EPA should make clear that those
provisions of the Clean Air Act will not apply to the proposed nonattainment zones in any final action on
this matter." One commenter believed the EPA should find that Wisconsin qualifies for the exemptions
set forth in Section 179b of the CAA related to areas impacted by international emissions.

EPA Response: It should be noted that the commenters seem to be interpreting "BC" or "boundary
conditions" as contributions from outside the U.S., in reality, boundary conditions are a mix of
international and natural sources as well as some recirculation from the U.S. However, as several
commenters identify, the language in section 179B(b) of the CAA may provide regulatory relief ".. .with
respect to an ozone nonattainment area..." if the affected state can establish "to the satisfaction of the
Administrator" that the subject area "would have attained the [NAAQS] but for emissions emanating
from outside of the United States...." While CAA section 179B recognizes the possibility that certain
nonattainment areas may be impacted by ozone or ozone precursor emissions from international
anthropogenic sources beyond the regulatory jurisdiction of the state, section 179B applies to designated
nonattainment areas and does not provide the authority to exclude monitoring data influenced by
international transport from regulatory determinations related to attainment/nonattainment, including area
designations for new NAAQS. Similarly, section 179B does not provide the authority to classify an area
with a lower classification than indicated by actual air quality or relax any mandatory control measures
associated with the area's classification. For designated nonattainment areas, an approved "but for"
analysis prepared under section 179B(a) as part of an attainment plan/demonstration provides relief from
attainment plan disapproval and any accompanying sanctions or Federal Implementation Plan. An
approved "but for" analysis prepared under section 179B(b)-(d), as part of an attainment determination,
provides relief from a finding of failure to attain and reclassification (e.g., relief could come in the form
of certain fee provisions (section 185) or relief from bump-ups).21

The EPA notes that while Wisconsin, and other similarly situated states, are not entitled to relief under
section 179B prior to designations forthe 2015 ozone NAAQS, they may qualify for relief under the
exceptional events provisions in section 319 of the CAA and the implementing regulations (i.e., the
Exceptional Events Rule) codified at 40 CFR 50.1, 50.14 and 51.930. The CAA recognizes that when
"exceptional" events cause exceedances or violations of the NAAQS that subsequently affect certain
regulatory decisions, including area designations, the normal planning and regulatory process established
by the CAA may not be appropriate. The Exceptional Events Rule provides a mechanism by which an air
agency can request the exclusion of event-influenced air quality data. Once EPA determines that an event
is "exceptional" under CAA section 319, it will exclude the data directly influenced by that event and
base its subsequent regulatory determination on the remaining monitoring data. Transported emissions (or
naturally-occurring ozone, in the case of stratospheric intrusions) from natural sources such as wildfires
or stratospheric ozone intrusion could qualify for data exclusion under the Exceptional Events Rule.

The EPA encourages affected air agencies to coordinate with their EPA Regional office to identify
approaches to evaluate the potential impacts of international transport and to determine the most
appropriate information and analytical methods for each area's unique situation. The EPA will also work
with states that are developing attainment plans for which section 179B is relevant, and ensure the states
have the benefit of the EPA's understanding of international transport of ozone and ozone precursors. To
assist in this effort, EPA is currently developing or has developed the following implementation tools to

21 The regulatory relief associated with attainment plans and demonstrations and provided in CAA section 179B(a)
applies to all NAAQS pollutants. Sections 179B(b)-(d) contain essentially the same regulatory relief provisions
related to attainment determinations, but the sections apply to different pollutants with 179B(b) applying to
ozone, 179B(c) applying to carbon monoxide and 179B(d) applying to particulate matter.

32

-------
help states assess the potential contributors to transported ozone: ozone/wildfire exceptional events
implementation guidance,22 stratospheric ozone intrusion exceptional events implementation guidance,23
and technical guidance on preparing approvable demonstrations under CAA section 179B.24

Comment: One commenter indicated that Racine, Waukesha, and Washington County emissions are
decreasing ozone formation at the Ozaukee County monitors and commenter 0247 is concerned that EPA
has not accounted for the "dis-benefit" of controlling NOx in these three counties.

EPA Response: No evidence was submitted to support the idea that precursor emissions from Racine,
Waukesha, and Washington County are responsible for reducing ozone concentrations detected at the
Ozaukee County monitors. Perhaps the commenters are referring to NOx scavenging? High NOx
concentrations found in downtown metropolitan areas, especially near busy streets and roads, and in
power plant plumes, can result in scavenging (sometimes referred to as titration) of ozone by reaction
with NO to form NO2 leading to localized depletion of ozone. However, as urban plumes are transported
and diluted, this NO2 can lead to photochemical production of ozone downwind of the source areas. In
addition, in areas that experience localized titration chemistry, this impact can often be reversed with
large enough reductions in local NOx emissions. Perhaps the commenters are referring to NOx-sensitivity
versus VOC-sensitivity with respect to ozone formation? After a portion of a state is designated as
nonattainment, the state is responsible for attainment planning requirements including reductions in
precursor emissions. The state has some discretion when meeting the CAA requirements with respect to
focusing control efforts on the precursor emissions that will be most effective in reducing ozone
concentrations. The state may investigate whether an area is NOx-sensitive, VOC-sensitive, or
transitional, and focus precursor emissions reductions efforts accordingly.

Comment: One commenter thought thinks specific days from 2016 with high ozone should be excluded
from the Wisconsin monitoring data due to impacts from Canadian wildfires and references EPA
Exceptional Event request packages put together by the state of New Jersey and the state of Maryland
containing evidence that Wisconsin air quality was impacted by these wildfires.

EPA Response: The CAA and EPA implementing regulations, specifically the Exceptional Events Rule
codified at 40 CFR 50.1, 50.14 and 51.930, allow for the exclusion of air quality monitoring data when
making certain regulatory determinations (including attainment/nonattainment decisions for new
NAAQS) when there are exceedances caused by certain event-related influences, including wildfires. To
exclude event-influenced data under the Exceptional Events Rule, the affected air agency must prepare a
demonstration that meets the technical criteria and the schedule and procedural requirements in the rule.
Specifically, under the Exceptional Events Rule, a demonstration must contain the following five
elements: a narrative conceptual model, evidence of a clear causal relationship between the specific event
and the monitored exceedance or violation (supported in part by a comparison to historical
concentrations), evidence that the event was not reasonably controllable or preventable, evidence that the
event was due to human activity that is unlikely to recur at a particular location or was a natural event and
documentation that the air agency conducted a public comment process and addressed comments
received, as necessary. While multiple air agencies and states affected by the same event can share the
analyses and/or narrative to support some of these required elements, other analyses and procedural steps
(e.g., the causal relationship between the event and the affected monitor and the authorizing official) will
likely be different for each air agency. Simply referencing another air agency's demonstration for a

22	Currently available at https://www.epa.gov/air-qualitv-analysis/exceptional-events-rule-and-guidance

23	under development

24	under development

33

-------
claimed shared event does not meet the requirements of the Exceptional Events Rule. Further, Wisconsin
did not submit any requests for data exclusion under the Exceptional Events Rule related to initial area
designation decisions for the 2015 ozone NAAQS.

Comment: Some commenters indicated that EPA photochemical modeling predicted that all Wisconsin
monitors will remain in or reach attainment of the 2015 ozone NAAQS by 2023 and suggest that this
supports reducing the size of the intended Milwaukee nonattainment area. Commenter 247 specifically
notes that this photochemical modeling prediction is based on "on the books" controls (without any 2015
ozone NAAQS implementation regulations) and therefore does not include additional emissions
reductions from Racine, Waukesha, or Washington counties.

EPA Response: EPA does not believe the CAA authorizes NAAQS designations based on future air
quality estimates where current valid air quality monitors indicate violations of the NAAQS. Also, it
should be noted that not all Wisconsin monitors are projected to attain the 2015 ozone NAAQS by 2023
(see EPA's March 27, 2018, transport memo).25

Comment: With respect to WDNR resources, one commenter expressed concern that a Milwaukee area
nonattainment designation would result in resources being allocated to administrative burdens at the
expense of limiting resources dedicated to scientific research designed to find solutions to ozone pollution
(e.g. participation in the Lake Michigan Ozone Study).

EPA Response: The Lake Michigan Ozone Study of 2017 involved on the order of $1.3 million in
funding from various organizations including the National Aeronautics and Space Administration
(NASA) (-46%), the National Science Foundation (NSF) and university research groups (-26%), the
National Oceanic and Atmospheric Administration (NOAA) (-9%), the Electric Power Research Institute
(EPRI) (-8%), the Environmental Protection Agency (-7%) and WDNR/LADCO (-4%). While States
can receive up to 60% of their air pollution control program funding from an EPA grant, the states
determine how to best allocate the federal and state resources available to meet federal requirements and
protect human health and the environment. Nonattainment area planning and scientific research designed
to investigate solutions to ozone pollution go hand in hand.

Comment: One commenter thought that Racine County is similar to Sheboygan, Manitowoc, and
Kenosha counties with respect to emissions, population density, VMT, etc. and should be treated
similarly.

EPA Response: Racine County is part of the Milwaukee CSA and therefore not a single county initial
area of analysis like Sheboygan and Manitowoc. Racine County is upwind of the violating monitors in the
Milwaukee area of analysis, whereas Kenosha County is the downwind-most county of the Chicago area
of analysis and contains two violating monitors. See other comments and responses above for information
on and why EPA was able to propose partial county nonattainment areas for Sheboygan and Manitowoc.

Comment: One commenter who resided in Milwaukee and was affected by ozone pollution, noted that
ozone pollution has a direct effect on the commenter's physical well-being and longevity. This
commenter was also affected by the imposition of ozone pollution regulations by being subject to motor
vehicle emissions testing. The commenter cited a news article referring to Wisconsin's request to set aside
a recent federal finding that southeast Wisconsin is violating new and tougher emissions standards for
smog and conclude the state is complying with the law. Short of that, the state is recommending federal

25 httpsi//www.epa.gov/airmarkets/march-2018-memo-and-supplemental-information-regarding-interstate-
transport-sips-2015

34

-------
officials carve out narrow strips of land of a few miles along the Lake Michigan shoreline as violating the
new standard for ozone pollution and declare the rest of the state in compliance. The commenter believed
that this request to weaken air pollution regulations in southeast Wisconsin should be denied. The
commenter believes the citizenry of Wisconsin has paid too much of a price to attract a foreign
manufacturing corporation, Foxconn, to locate in southeast Wisconsin. The commenter believed that no
matter how much Wisconsin elected officials want this corporation to do business in Wisconsin, this
corporation must be required to do business in the same manner as all other Wisconsin and U.S.
businesses, which means its operations cannot be exempted from engaging in environmentally destructive
activities. The commenter asked EPA to enforce the standards already in place and to do no more than
what EPA is tasked to do and is already doing.

EPA Response: The commenter seems to be referring to Wisconsin's recommendation that EPA
foremost designate the entire state as attainment of the 2015 ozone NAAQS or, barring that, only narrow
parcels of nonattainment land encompassing the violating monitors near the shoreline of Lake Michigan.
The commenter also seems to be referring to EPA's intent to designate some part of the Milwaukee area
as nonattainment of the 2015 ozone NAAQS based on our analysis of violating monitors and nearby
contributing areas. EPA acknowledges the receipt of this comment. EPA's final nonattainment area for
Milwaukee is addressed in the final TSD for the Wisconsin nonattainment areas. In this analysis EPA has
done no more than it is tasked with doing, which is to, in accordance with CAA section 107(d), designate
as nonattainment all areas with monitor[s] that is [are] violating the 2015 ozone NAAQS and nearby areas
that contribute to the violation [s].

Sheboygan County. WI

Comment: A couple of commenters believed the entirety of Sheboygan County should be designated
attainment. However, if EPA designates part of the county as nonattainment, one commenter provided a
list of roadways which create a narrower nonattainment boundary than EPA's intended boundary (see
docket for specific roadways). This commenter also provided a list of roadways which create a narrower
nonattainment boundary than EPA's intended boundary (see docket for specific roadways). Several
commenters believed that the Kohler Andrae monitor in Sheboygan county is improperly sited.
Specifically, one of these commenters indicated that this monitor is upwind of the majority of Sheboygan
County sources and dominated by out-of-state transport, and a commenter indicated that it is misplaced
according to a 1998 EPA document.26 Two commenters indicated the Haven monitor (as opposed to the
Kohler Andrae monitor) is properly sited downwind and measuring air from activity within the county.
One commenter believed the Haven monitor should be the only monitor used for regulatory purposes in
Sheboygan County and one commenter believed EPA should disregard the Kohler Andrae monitor. One
commenter indicated that the LADCO modeling (WDNR April submittal) shows the entire state of
Wisconsin contributes less than 10% of ozone measured at Kohler Andrae monitor. One commenter
attached the 2017 LADCO white paper on the Lake Michigan Ozone Study27 to a comment letter
indicating that the commenter believes the Kohler Andrae monitor is reading 98% ozone that has been
transported from outside Sheboygan County and that Sheboygan County NOx emissions account for 2%
of the total NOx emissions in Wisconsin.

26 U.S. Environmental Protection Agency, "GUIDELINE ON OZONE MONITORING SITE SELECTION." EPA-
454/R-9 8-002, August 1998.

27 Pierce, B., Kaleel, R., Dickens, A., Bertram T., and Stanier, C., Kenski D.: White Paper: Lake Michigan Ozone Study
2017 (LMOS 2017), http://www.ladco.org/, 2016.

-------
With respect to Sheboygan County, while WDNR recommended attainment for the entire state as per the
Governor's recommendation, if EPA designates a portion of Sheboygan County as nonattainment,

WDNR believed that the boundary should be based on the distance from shore approach and no more
than 2.3 miles inland, which is a location consistent with the best-fit line (70 ppb contour) developed by
WDNR. If EPA does not set the boundary at 2.3 miles inland, then WDNR believed it should be no more
than 2.9 miles inland, which is the location determined by comparison of design values at Sheboygan's
two ozone monitors as described in DNR's April 2017 TSD. Wisconsin did not want the boundary to be
3.2 miles inland (based on the location of the Haven monitor) nor based on roadways like EPA intends,
since the Haven monitor has a design value of 69, indicating to Wisconsin that the attainment level air
quality would be found between this monitor and the lakeshore and that that any nonattainment area
boundary should be to the east of this monitor. Wisconsin also states that Sheboygan County ozone
concentrations are heavily impacted by out-of-state transport and unfavorable meteorological and
geographic factors and are not affected by local sources of emissions. Industrial emissions comprise less
than 10% of county NOx and VOC emissions and are already well controlled. The county's largest source
of NOx emissions (Edgewater Generating Station) has significantly reduced emissions since 2011 and
forecasts even more dramatic reductions in future years.

EPA Response: The EPA addresses in the final nonattainment area TSD the conclusions reached for the
final boundary for the Sheboygan area, including our review of information provided in WDNR's April
2017 submittal and February 2018 comments.

Regarding the comments referring to the siting of the violating monitor, 40 CFR part 58, Appendix D,
provides network design criteria including site types and siting scales (Table D-l). WDNR has designated
the Sheboygan-Kohler site as a regional transport/maximum ozone concentration site and the Sheboygan-
Haven site as a population exposure site in both its EPA-approved annual network plan and in EPA's Air
Quality System. Wisconsin's annual network plan was approved by EPA on September 1, 2017 and
includes statements affirming compliance with 40 CFR part 58, Appendix E, Probe and Monitoring Path
Siting Criteria for Ambient Air Quality Monitoring for each site in Wisconsin's ambient monitoring
network, including the Sheboygan-Kohler monitoring site and the Sheboygan-Haven monitoring site. The
upwind urban region (as described in the 1988 guidance) for the Sheboygan-Kohler regional
transport/maximum ozone concentration site is not specified in the annual network plan. However, one
could surmise that the urban region could be the Milwaukee area based on the following: Ozone is a
secondary pollutant formed over some time and space from the sources of ozone precursor emissions. It
is, therefore, likely that it is measuring ozone formed predominately from upwind precursor sources in
Milwaukee which is about 60 miles south of Sheboygan, and in the Chicago area, which is farther south.
With respect to LADCO's source apportionment modeling, a 10% contribution is approximately 8 ppb of
ozone.

Manitowoc County, WI

Comment: WDNR estimated the geographic extent of nonattainment is no more than 2.9 miles inland for
Manitowoc County (other comments regarding Manitowoc were summarized and addressed above).

EPA Response: The EPA addresses in the final nonattainment area TSD the conclusions reached for the
final boundary for the Manitowoc area, including our review of information provided in WDNR's April
2017 submittal and February 2018 comments. EPA notes that a reason EPA was able to justify honoring
Wisconsin's request to reduce the size of the Sheboygan County area from a full county to a partial
county area is the existence of the second ozone monitor in Sheboygan county which is attaining the
standard coupled with WDNR's lake breeze inland penetration distance analysis specific to the two

36

-------
Sheboygan County monitors. EPA extended this reasoning to Manitowoc County (despite Manitowoc not
having an inland attaining monitor), since Manitowoc County is adjacent to and immediately north of
Sheboygan County.

Door County. WI

Comment: Both WDNR and the Door County Administrator agreed that the Door County area should be
a rural transport area, but that the nonattainment area boundary should be the Newport State Park
boundary. The Door County Administrator did not believe (based on WDNR analysis) that ozone
penetrates farther inland than the Newport State Park Boundary. Commenter indicates that EPA
HYSPLIT trajectories indicate all emission tracks causing the violation at the Door County monitor to be
outside of the Door County land base except when the emissions reach Newport State Park. Commenter
adds, "The photochemical model projections of 2017 design values along the western Lake Michigan
shoreline prepared by LADCO project the entire Door County land mass to be outside of the
nonattainment area. (See: Figure 4.1 at page 16 of DNR's Technical Support Document.)" This
commenter was concerned about Door County economy which is heavily dependent on tourism.
Commenter wants the nonattainment boundary to be the boundary of the state park. WDNR asked EPA to
exclude all the offshore islands in Door County since WDNR believes there is no recognized benefit to
designating them as nonattainment and indicates that EPA has similarly excluded islands from its
nonattainment area designations for Ventura County, California for the 2008 ozone NAAQS (insert
citation). Supporting information provided by WDNR included the following: Only two of EPA's 100 m
back trajectories passed over Door County, with the remainder passing over the lake indicating that the
elevated ozone levels measured at the ground level at the Newport monitor result from air being
transported over the lake from the south. Door County emissions are low and most of these emissions
come from sources that the state cannot control: 53% of NOx from commercial marine vessels on Lake
Michigan and 61% of VOC emissions from recreational vehicles and pleasure craft.

This commenter noted specific concerns about negative public perception of poor air quality and the local
economy which is heavily dependent on tourism, (maybe move this and lump together will all area
comments regarding economic concerns and business operations and growth)

EPA Response: The EPA addresses in the final nonattainment area TSD the conclusions reached for the
final boundary for the Door County area, including our review of information provided in WDNR's April
2017 submittal and February 2018 comments.

3.2.6. EPA Region VI

3.2.6.1 Louisiana

Comment: The State of Louisiana commented that it submitted to EPA a Notice of Potential EE for
Ozone for the dates September 13-14, 2017 and that it intends to submit an EE demonstration to EPA for
approval. The State requested that an extension of up to one year be provided by the Administrator to
make a final determination on the EE demonstration, or in the alternative, a designation of Unclassifiable
until a final determination on the EE demonstration can be made.

Response: We have reviewed the EE demonstration proposed by Louisiana and the final EE
demonstration for air quality data received on April 3, 2018. On April 13, 2018, the EPA concurred on the
EE demonstration submitted by Louisiana for the ozone monitor known as the "Dutchtown" monitor in
Ascension Parish. The documents associated with the demonstration, including the EPA's concurrence
letters to Louisiana regarding the early certified data for 2017 ozone data and the EE demonstration, are

37

-------
provided in the docket for this action. With our concurrence on the EE demonstration, the Dutchtown
monitor in Ascension Parish was found to meet the 2015 ozone NAAQS for the 2015-2017 period.
Because all ozone monitors in the Baton Rouge area are meeting the 2015 ozone standard with 2015-2017
data, we are finalizing a designation of attainment/unclassifiable for the 9 parishes in the Baton Rouge,
LA Core-Based Statistical Area.

Comment: A commenter asserted that the weight-of-evidence and the EPA's guidance require the agency
to include Pointe Coupee Parish as part of the Baton Rouge nonattainment area. The commenter presented
a number of reasons why it believes emissions in Point Coupee are contributing to a violation of the 2015
NAAQS in the Baton Rouge area, mostly relating to emissions from the Big Cajun II (BCII) power plant.

Response: As noted in the previous response, based on the most recent three years of air quality data, all
monitors in the Baton Rouge area are attaining the 2015 ozone NAAQS and the EPA is designating the
area as attainment/unclassifiable. Because there are no violations at any of the monitors in the Baton
Rouge area, the EPA does not need to perform a five-factor analysis to determine whether other counties
are contributing to such a violation. Thus, no response is needed with regard to the contribution
arguments being made by the commenter.

3.2.6.2 New Mexico
Dona Ana County. NM

Comment: A commenter asserted that El Paso County should be included in the Dona Ana
nonattainment area and provides the following statements: the EPA identifies Mexico as the largest
source of emissions into the monitor in Las Cruces, but El Paso contributes considerably more emissions
than all of Dona Ana County; El Paso has far larger point sources than does the proposed nonattainment
area; El Paso County has four times the population of all of Dona Ana County. The commenter stated that
El Paso contributes too much to the ozone violation in Las Cruces to isolate the nonattainment area to this
tiny corner of Dona Ana County. The commenter claimed that to give Las Cruces the greatest opportunity
to reduce emissions from sources within the United States and meet the standard, El Paso must be part of
the nonattainment area.

EPA Response: We disagree with the commenter's assertion that El Paso must be a part of the
nonattainment area. Our TSD for El Paso, Texas and Dona Ana County, New Mexico (EP-NM TSD)
contains a five-factor analysis that discusses this conclusion. The commenter provided no additional data
to support their position regarding contribution to the violating monitor or for the EPA to review that
would cause us to reconsider our conclusions.

Comment: Several Commenters supported the intended designation of attainment/unclassifiable for El
Paso, County, TX.

EPA Response: The EPA appreciates the commenter's views. The EPA is finalizing the boundary for the
Dona Ana County, New Mexico Area as provided in the December TSD.

Comment: Other commenters asserted that because El Paso County is in attainment with the 2015 ozone
NAAQS, the designation should be Attainment rather than Attainment/Unclassifiable.

EPA Response: We refer the commenter to our February 25, 2016 guidance memo "Area Designations
for the 2015 Ozone National Ambient Air Quality Standards," (footnote 1) that explains what we mean by
"attainment/unclassifiable":

-------
"For initial area designations for the 1997 ozone NAAQS and the 2008 ozone NAAQS, the EPA
used a designation category of "unclassifiable/attainment" for areas that were monitoring
attainment and for areas that did not have monitors but for which the EPA has reason to believe
were likely attainment and were not contributing to nearby violations. The EPA reserved the
category "unclassifiable" for areas where the EPA could not determine based on available
information whether the area was meeting or not meeting the NAAQS and the EPA had not
determined that the area contributed to a nearby violation. While states can submit
recommendations identifying areas as "attainment," the EPA expects to continue to use the
"unclassifiable/attainment" category for designations for the 2015 NAAQs.28"

3.2.6.4 Texas
Dallas-Fort Worth, TX

Comment: A commenter supports the EPA's intended designation for the Dallas-Fort Worth area, and
provides examples of emission reduction measures implemented in the area. The commenter believes we
will finalize designations based on 2015-2017 ozone monitoring data. The commenter states that the EPA
should continue to develop procedures and guidance addressing Exceptional Event and Clean Air Act
section 179(b) demonstrations.

Response: The EPA appreciates the commenter's views and supports emissions reductions measures
undertaken by TCEQ in collaboration with industry and the community; this work has and will continue
to assist the area in reducing NOx, total VOCs, and Highly Reactive VOCs, that are precursors to ozone
formation. The EPA encourages the continuation and development of these and future emissions
reductions efforts with the TCEQ.

As noted in the designations guidance, the EPA is basing these final designations on air quality
monitoring data for the 2014-2016 time period because this is the most recent certified data available for
this area.

Comment: A commenter asserts that the EPA should include in its 2015 ozone NAAQS nonattainment
designation for DFW the nearby East Texas coal-fired electric generation units ("EGUs")—Martin Lake
in Rusk County, Monticello and Welsh in Titus County, Big Brown in Freestone County, and Limestone
in Leon County. The commenter states that properly drawing the lines of the 2015 ozone nonattainment
boundaries is a critical first step in bringing all of the contributing areas to the table and achieving healthy
air quality for the DFW area. The commenter provides two sets of modeling simulations and asserts that if
these five East Texas EGUs reduce coal burning or install "end of the pipe" pollution controls to reduce
NOx, there would be meaningful reductions in ozone levels in the DFW area and the potential for other
public health co-benefits across Texas. The commenter notes that Texas has failed to attain health-based
standards for ozone pollution in the DFW area and cites the EPA's proposal to reclassify the area to
Severe under the 1997 ozone standard (see 80 FR 8274, February 17, 2015). The commenter asserts that
where nearby sources are "sufficiently" contributing to nonattainment, the EPA and the courts have
recognized that those areas may be designated as nonattainment even if they are not immediately adjacent

28 More recently, the EPA has determined to reverse the order of the terms and refer to such areas as
"Attainment/Unclassifiable" as found in the Designations Guidance and previous ozone designations.
This change in phrasing has no substantive effect but will improve public understanding and make
clearer what regulations apply to areas designated in this way, which some states have suggested is
important for the economic development of such areas.

39

-------
to the jurisdictional boundary of the violating monitor. Finally, the commenter states that between the
high levels of ozone precursors emitted by the these five EGUs and the minimal NOx controls on their
largest sources, the "emissions data" factor weighs heavily towards including these counties in the DFW
nonattainment area.

The Commenter provided a study by Haley, et al, asserting that retrofitting or closing the 5 coal plants
would result in a 5 ppb drop in ozone levels at the DFW ozone monitors.

EPA Response: The EPA disagrees with the commenter that it is necessary to designate Rusk, Titus,
Freestone, and Leon Counties as part of the DFW nonattainment area. As provided in EPA's February
2016 Designation Guidance, the EPA begins its analysis of the appropriate area to designate as
nonattainment by looking at the areas within the CSA or CBSA. EPA has determined that the CSA or
CBSA is a reasonable representation of which areas may be "nearby" for purposes of determining
contribution. The four counties mentioned by the commenter - Rusk, Titus, Freestone, and Leon Counties
- are not in the DFW CSA and thus EPA did not analyze these areas for purpose of the 120-day
notification sent to the State in December 2017. While EPA is not precluded from looking at counties
outside the CSA or CBSA, EPA does not agree that it should do so here. Three of the four counties do not
adjoin the CSA - i.e., Leon, Rusk and Titus Counties are separated from the CSA by at least one other
county that is also not part of the CSA. Thus, even if emissions from these counties do "contribute" to
exceedances of the ozone standard at violating monitors in the DFW area, EPA does not agree that they
should be considered "nearby." Freestone County is adjacent to the CSA and while there are no
geographical or topographical features limiting air pollution transport, Freestone is not adjacent to a
county with a violating monitor; thus we do not consider Freestone to be nearby. Even if Freestone were
considered nearby, however, we do not believe the five-factor analysis would support modifying the state
recommendation by including Freestone County as part of the DFW nonattainment area, as explained in
EPA's TSD for this area. Furthermore, we have recently been provided information that Luminant has
permanently retired the Big Brown Units 1 and 2 boilers (Electric Generating Units or EGUs in Freestone
County), and the associated air permit cancellation request is provided in the docket for this action. Based
on the 2014 National Emissions Inventory (NEI), the retired Big Brown units emitted approximately
5,087 tpy of NOx or 45 percent of the county-wide NOx emissions, and 76 tpy of VOC or one percent of
the county-wide VOC emissions. Thus, the NOx emissions in Freestone County are substantially less than
those in the emissions inventories and reports submitted by the commenter. Finally, Freestone County is
not part of a previously established nonattainment boundary and is not served by the metropolitan
planning organization that operates in the DFW CSA.

The modeling provided by the commenter predicted that 90 to 100 percent reductions of power plant NOx
at all five EGUs could reduce the Denton Airport ozone design value by approximately 4 ppb. There are
two issues with this modeling. First, the modeling addresses emissions from four counties combined. We
cannot determine from the modeling what the contribution might be from an individual county. This is
not consistent with EPA's county-by-county analysis for determining contribution. In addition, as already
noted in part, we have recently been provided information that Luminant has permanently retired the
Monticello Units 1, 2, and 3 boilers (in Titus County) and Big Brown Units 1 and 2 boilers, which
reduces the emissions from Titus and Freestone Counties well below the modeled levels.29 While the
modeling provided by the Commenter predicts a drop of 5 ppb through closure of the 5 referenced power
plants at the DFW ozone monitors, which would help for future emissions reductions, the retirements of
the facilities in Titus and Freestone Counties, as described above, reduces the emissions from those
counties below the modeled levels and thus, the submitted modeling does not reflect ozone contributions
that are consistent with current emissions levels

29 The associated air permit cancellation requests are provided in the docket for this action. The TCEQ's
cancellation of the permit authorizations for Monticello Units 1, 2, and 3 boilers are also in the docket.

40

-------
Comment: A commenter supports the EPA's proposed nonattainment designation for Collin, Dallas,
Denton, Ellis, Johnson, Kaufman, Parker, Rockwall, Tarrant, and Wise Counties and urges the EPA to
finalize its designation expeditiously. The commenter urges the EPA to include specific recognition of the
contribution of oil and gas emissions to the unhealthy ozone levels in the Dallas-Fort Worth area, and to
work with the State to implement robust controls on such activities. The commenter asserts that EPA's
failure to mention oil and gas emissions as a contributor to unhealthy ozone levels in the area is a major
omission. The commenter also contends that state, the EPA, and independent data clearly demonstrate
that emissions from oil and gas facilities are a large contributor of NOx and VOC emissions in the DFW
area.

The commenter states that oil and gas activities in the 10 nonattainment counties were responsible for
5,328 tpy of NOx emissions and 14,947 tpy of VOC emissions. The commenter urges the EPA to note
such contributions in its final designation. The commenter urges the EPA and the state to evaluate and
implement all available cost-effective means to reduce oil and gas emissions, including implementing the
oil and gas CTGs and establishing robust pollution control standards for existing oil and gas operations.
The commenter asserts that these measures are required regardless of the specific nonattainment area
designation (i.e., Marginal or Moderate) as in either instance the poor air quality in the DFW area
threatens human health and welfare.

EPA Response: We appreciate the technical information provided by the commenter and acknowledge
the Commenter's support of our intended designation of nonattainment for the DFW area. However, as
provided in the TSD for the DFW Area, we have determined for purposes of the final designation to
accept the State's recommendation that Rockwall not be included as part of the nonattainment area.
Assuming the oil and gas emissions provided by the Commenter are accurate, such contributions provide
approximately 10 percent of the VOC emissions and less than 4 percent of the NOx emissions in the 10-
county area proposed as nonattainment. Consistent with the CAA, the State is responsible for addressing
emission sources in each of its nonattainment areas. The State may choose to implement more stringent
controls than required by the Act in a nonattainment area, but the EPA has no authority to require controls
more stringent than those mandated by the CAA.

Houston-Galveston-Brazoria. TX

Comment: A commenter supported EPA's intended designation for the Houston-Galveston-Brazoria
(HGB) area, and provides examples of emission reduction measures implemented in the area. The
commenter assumed we will finalize designations based on 2015-2017 ozone monitoring data. The
commenter stated that EPA should continue to develop procedures and guidance addressing EE and CAA
section 179(b) demonstrations.

EPA Response: The EPA appreciates the commenter's views. Our analysis of air quality data is based on
ambient air monitoring data for the period from 2014-2016, which is the most current certified data for
this area. The State is not required to certify data from 2017 until May 1, 2018.

Comment: A commenter urged EPA to take additional measures to address emissions from neighboring
counties whose emissions contribute to ozone exceedances in the HGB area and consider expanding the
nonattainment area to neighboring counties whose emissions contribute to such exceedances. The
commenter supports designating nonattainment areas for the 2015 ozone standard from a minimum
baseline of the 2008 ozone standard. The commenter claimed that more aggressive action is required to
bring the HGB area into attainment. The commenter stated that although a regional approach to air quality

-------
has proven effective in reducing emissions, the 2015 ozone nonattainment areas should include all area
counties that contribute to elevated levels of ozone in the Houston region.

EPA Response: The nonattainment designation determinations are based on a weight of evidence
analysis that takes into consideration the five factors of air quality, emissions, meteorology, geography
and topography, and jurisdictional boundaries. Based on the five-factor analysis, EPA identifies whether
all of the counties violating the NAAQS and those nearby counties contributing to air quality at the
violating monitors are included in the nonattainment area. The final TSD in the docket for this action
describes this five-factor analysis of the counties in the HGB combined statistical area. Based on the five-
factor analysis and considering the recommendation made by the State, EPA has designated six counties
as part of the nonattainment area. The commenter does not identify any specific counties to be included in
the area nor does the commenter provide any technical analysis supporting inclusion of additional
counties. To the extent the commenter is suggesting that EPA should impose additional measures in
counties neighboring the HGB area, we note that States, not EPA, have the primary responsibility for
regulating areas in the State for purposes of attaining and maintaining the NAAQS.

San Antonio. TX

Comment: Several commenters submitted comments urging EPA to designate the multi-county San
Antonio Metropolitan Statistical Area (MSA) as nonattainment. Another commenter urged EPA to
evaluate the contribution of emissions from counties in the San Antonio-New Braunfels MSA and Eagle
Ford Shale on air quality in Bexar County. Another commenter, claiming that the 2015 ozone NAAQS is
invalid, urges EPA to designate Bexar County as attainment or unclassifiable. The commenter also noted
that the county is projected to attain the 2015 NAAQS by 2020 and claims the costs imposed by a
nonattainment designation would be burdensome. Yet another commenter urged a designation of
unclassifiable, claiming EPA's 2017 air monitoring handbook provides EPA with discretion in
interpreting monitored design values.

EPA Response: The December 22, 2017 action did not include a proposed designation for the San
Antonio area. The EPA announced its intended designations for the San Antonio area in a separate action
on March 19, 2018 (see https://www.epa.gov/ozone-designations/epa-responds-recommendations-2015-
ozone-standards-san-antonio-tx). The EPA has provided an opportunity for public comment regarding
the designation of the counties in the San Antonio area. (83 FR 13719) the EPA will consider and respond
to these comments as well as any new comments it receives during the new public comment period when
it moves forward with a final designation for the counties in the San Antonio area.

3.2.7. EPA Region VII

Comments received regarding areas located in EPA Region VII are addressed below in Section 3.2.11.

3.2.8. EPA Region VIII

3.2.8.1 Colorado

Denver Metro/North Front Range. CO

Comment:_Several commenters stated that the EPA did not evaluate or analyze the significance of the
information provided in the TSD consistent with CAA requirements and that the EPA's proposal
arbitrarily excludes the northern portions of Weld and Larimer Counties from the nonattainment area.
They contend that a proper analysis of the five factors would lead to the conclusion that ozone precursors

-------
from Northern Weld and Larimer Counties are transported into the nonattainment area and contribute to
ozone formation on high ozone concentration days.

EPA Response: The CAA requires that the EPA designate areas that fail to attain a revised NAAQS, as
well as nearby areas which contribute to violations be designated as nonattainment. The EPA utilized a
weight of evidence approach using five factors (Air Quality Data, Emissions and Emissions Related Data,
Meteorology, Geography/Topography, and Jurisdictional Boundaries) to determine nonattainment area
boundaries. Based on consideration of those factors, EPA determined not to modify the State's
recommendation that the northern portions of Weld and Larimer counties should not be included as part
of the Denver nonattainment area because emissions from those areas do not sufficiently contribute to air
quality at the violating monitors in the Denver area. The EPA's TSD describes how consideration of all
five factors led to the boundary the EPA finalized.

Comment: Several commenters alleged that the EPA's proposed exclusion of part of Weld County does
not correspond to any jurisdictional boundary. The commenters reference the EPA's ozone guidance
recommending inclusion of the entire violating or contributing county in an ozone nonattainment area.
The commenters state that the EPA did not justify excluding part of Weld County and therefore should
designate the entire county nonattainment.

EPA Response: Page 33 of the EPA TSD for Colorado (EPA-HQ-OAR-2017-0548-0069) provided with
the 120-day letter to the Governor of Colorado explained that the boundary selected for the 2015 ozone
NAAQS is identical to that used for the 1997 and 2008 ozone NAAQS nonattainment boundaries. The
EPA's ozone guidance30 provides examples of jurisdictional boundaries states may use to determine
nonattainment area boundaries, including existing nonattainment areas. The EPA utilized a weight of
evidence approach using five factors (Air Quality Data, Emissions and Emissions Related Data,
Meteorology, Geography/Topography, and Jurisdictional Boundaries) to determine nonattainment area
boundaries. Based on consideration of those factors, EPA determined not to modify the State's
recommendation these two areas should not be included as part of the Denver nonattainment area because
emissions from these areas do not contribute to air quality at the violating monitors in the Denver area.
The TSD contains a more detailed explanation of the 5-factor analysis and how the boundaries for this
area were determined.

Comment: A commenter stated that the EPA's TSD shows that the furthest north monitor with a valid
design value is exceeding the 2015 NAAQS and therefore the air quality factor is reason to include all of
Weld County in the nonattainment area.

EPA Response: The regulatory ozone monitor (08-123-0009) operated in Weld County by the state of
Colorado shows an attaining design value (0.070 ppm) for the most recent available data years (2014-
2016 data). Non-regulatory ozone monitors operated by the U. S. Forest Service north of Briggsdale,
Colorado (24 miles northeast of Greeley, and 4 miles south of the nonattainment area boundary) and at
Pawnee Buttes (42 miles northeast of Greeley and 7 miles north of the nonattainment area boundary) and
available in the EPA AQS database have never shown a design value greater than 70 ppb (2011-2016 at
Briggsdale, 2013-2016 at Pawnee Buttes). The available air quality data indicate that the northern portion
of Weld County is currently attaining the 2015 ozone NAAQS.

Comment: The commenter stated that the monitor in Weld County, 08-123-0009, shows a violation of
the NAAQS based on the 2015-2017 design value, providing further evidence that the entire county
should be designated nonattainment.

30 https://www.epa.gov/ozone-designations/epa-guidance-area-designations-2015-ozone-naaqs

-------
EPA Response: The 2017 data from the Greeley Tower ozone monitor (08-123-009) have not yet been
certified by the Colorado Department of Public Health and Environment, and so are not being used in this
designation. The EPA is required to use the most recent, certified data in making designation decisions; in
this case, that is the data from 2014-2016. Even so, the preliminary, uncertified data from the site show a
2015-2017 design value of 70 ppb and would therefore be attaining when certified.

Comment: A commenter contended that EPA's TSD ignored ozone in Boulder County and references
two studies (Reference Ex. 1 and Ex. 2 abstract) that found precursor emissions in Weld County,
including the northern portion of the county, contribute to ozone in Boulder County. The commenter also
states that the monitor located in Boulder County 08-013-0011 is violating the NAAQS based on 2013
AQS data and since Weld County emissions contribute to the violation, the entire county should be
designated nonattainment.

EPA Response: The commenter does not provide evidence of how the referenced studies support the
inclusion of the northern portion of Weld County in the nonattainment area. While the referenced study
indicates that "O&G sources amount to the second largest NOx sources with most of them attributed to
Weld County", the reference does not specifically assess what proportions of emissions come from
northern Weld County, as opposed to the portion of Weld County the EPA is including in the
nonattainment area. The EPA has addressed in the TSD and five factor analysis why the northern portion
is being excluded from the nonattainment area. The commenter cites Reference Ex. 2 abstract, which
states that "Analyses of surface ozone and wind observations from two sites, namely, South Boulder and
the Boulder Atmospheric Observatory, both near Boulder, CO, show a preponderance of elevated ozone
events associated with east-to-west airflow from regions with O&NG operations in N-ESE, and a
relatively minor contribution of transport from the Denver Metropolitan area to the SE-S". It is precisely
the southern portion of Weld county, which EPA and the state included in the recommended
nonattainment area, that is contributing to ozone formed and observed after the east to west motion
described in this abstract cited by the commenter. The cited abstract does not specifically address
transport from more than 40 miles north from the northern excluded portion of Weld County.

Comment: The commenters pointed out that Weld County has the largest emissions of any county in the
nonattainment area with NOx and VOC emissions comprising of 46% of the total emissions of the
nonattainment area. One commenter states that the northern portion of Weld County represents a large
share of the total county emissions. The commenter states that if the EPA does not include all of Weld
County in the nonattainment area, the EPA will be treating Weld County differently than similarly
situated counties.

EPA Response: While the specific proportions of emissions may not be identical, other large counties in
the western U. S. have been designated with both attainment and nonattainment portions (see, for
example, San Bernardino and Riverside counties in southern California). While Weld County may have
46% of the total emissions in the analyzed counties, the EPA is including the majority of these Weld
County emissions sources in the nonattainment area, and specifically those emissions which are nearby to
violating monitors, as required by the CAA. Moreover, the EPA utilized a weight of evidence approach
using five factors (Air Quality Data, Emissions and Emissions Related Data, Meteorology,
Geography/Topography, and Jurisdictional Boundaries) to determine nonattainment area boundaries.

Comment: A commenter provided a discussion on large point sources, identifying three large point
sources in Weld County located outside the proposed nonattainment area. The commenter contrasts this
with Douglas County which has no large point sources but is proposed as part of the nonattainment area.
The commenter points to parts of Larimer and Boulder counties that have no large point sources but are
also proposed to be part of the nonattainment area. The commenter concludes that the EPA is treating
Weld County differently than similarly situated counties.

-------
EPA Response: The EPA uses a weight of evidence approach using five factors in its TSD to determine
which areas to include or exclude in the nonattainment area. Considering the weight of evidence from the
five factors, including the proximity of the heavily populated portions of Douglas and Boulder Counties,
air quality data (Douglas County contains the violating Chatfield Reservoir monitor, while the Rocky
Flats North violating monitor lies less than 200 yards south of the Boulder County line), the relative
populations of Boulder and Douglas Counties compared to the sparsely populated northern Weld County,
the prevailing meteorology on ozone exceedance days, and topography, inclusion of Boulder and Douglas
counties is necessary. Based on consideration of those factors, EPA determined not to modify the State's
recommendation that northern Weld County should not be included as part of the Denver nonattainment
area because emissions from these areas do not contribute to air quality at the violating monitors in the
Denver area.

Comment: A commenter alleged that EPA evaluated the wrong factors in its analysis of large and small
point sources and that the CAA requires areas contributing emissions to a violation in a nearby area to be
designated nonattainment; 42 U.S.C. 7407(d)(l)(A)(i). The commenter pointed to the EPA's finding that
Wyoming contributes significantly to ozone in the current Denver Metro North Front Range
nonattainment area which therefore means Northern Weld County contributes to the Front Range
nonattainment area.

EPA Response: Under CAA section 107(d)(1), the EPA must designate an area "nonattainment" if it is
violating the NAAQS or if it is contributing to a violation of the NAAQS in a nearby area. The EPA
applied the same five factor analysis nationally in evaluating nearby contribution to nearby violations, as
is outlined in EPA's Guidance on the Area Designations for the 2015 Ozone NAAQS31. The five
factors—air quality data, emissions and emissions-related data, meteorology, geography/topography, and
jurisdictional boundaries—are evaluated on a case-by-case basis to inform the decisions on the
boundaries for nonattainment areas for initial designations after a NAAQS modification. The
nonattainment area boundary identified for the Denver Metro North Front Range area, as assessed by the
five-factors, adequately captures the nearby contributing emissions. CAA section 110(a)(2)(D)(i), often
referred to as the "good neighbor" provision of the Act, requires states, within three years of promulgation
of a new or revised NAAQS, to submit SIPs that contain adequate provisions prohibiting any source or
other type of emissions activity within the state from emitting any air pollutant in amounts which will
contribute significantly to nonattainment in, or interfere with maintenance by, any other state with respect
to any such national primary or secondary ambient air quality standard. Assessment of contribution for
the purpose of the good neighbor provisions and associated requirements for states are addressed
separately under the CAA.

Comment: Several commenters stated that the part of Weld County that the EPA intends to exclude has
higher VOC emissions than every other county in the proposed nonattainment area and higher NOx
emissions than most of the counties in the proposed nonattainment area. Also, that emissions in Northern
Weld County are significant and increasing rapidly. The commenters assert that Colorado is issuing minor
source permits without any analysis of their impacts to ozone levels. Additionally, a map of oil and gas
wells in and around the nonattainment area included in the EPA's TSD map shows that oil and gas wells
are numerous and dense in northern Weld County, relative to other areas which the EPA is recommending
for nonattainment. This contrasts with other counties in the nonattainment area which have almost none.
The commenter cites a study by Evans and Helmig that report findings "suggesting that VOCs from the
Northern Colorado Front Range region, where significant oil and gas production occurs, contribute to the
formation of ozone in the nonattainment area (Evans and Helmig, 2017)."

31 https://www.epa.gov/ozone-designations/epa-guidance-area-designations-2015-ozone-naaqs

45

-------
EPA Response: The commenter is focusing on individual issues, without taking an integrated weight of
evidence multi-factor approach as the EPA did. While emissions in northern Weld County may be greater
than emissions in counties closer to violating monitors, the weight of evidence five factor analysis
assesses emission levels, proximity to violations, population and area source emissions relative to
meteorology transporting emissions, and other factors. While oil and gas wells may appear on a map to be
relatively dense in northern Weld County, the map also shows much higher well density in the portion of
Weld County included in the nonattainment area. The EPA does not dispute the conclusion of the cited
report, that Northern Colorado Front Range oil and gas emissions contribute to ozone formation in the
nonattainment area, and for that reason have included the majority of those emission sources in the
nonattainment area, and particularly those sources which are nearby to violating monitors in directions
prevailing meteorology indicates contribute to those violations.

Comment: A commenter suggested that the low population of northern Weld County is irrelevant
because of the level of emissions coming from the northern portion of the county.

EPA Response: Population density is an indicator of the location for a number of emission factors not
related to oil and gas production, such as mobile source emissions. Population density is therefore not
irrelevant for nonattainment area boundary determinations, and is one of the five factors used nationally
in a weight of evidence approach to determine boundaries for nonattainment area designations. The lack
of population in northern Weld County indicates that few emissions, aside from those from oil and natural
gas sources, are present.

Comment: A commenter disagreed with the EPA's analysis of traffic and vehicle miles traveled and
points to emissions from vehicles used to serve the oil and gas industry.

EPA Response: Emissions from vehicles used to serve the oil and gas industry are included in the traffic
and vehicle miles traveled data that the EPA evaluated. As is shown in the TSD, the majority of VMT
emissions in the CSA are captured within the EPA's nonattainment boundary.

Comment: A commenter stated that the EPA is incorrect in its analysis, and has based its designation on
a mythology that air pollution is worst "downtown" when the worst ozone in the Front Range is to the
west of the urban core area.

EPA Response: The commenter takes the EPA's assessment that "Rapid population or VMT growth in a
county on the urban perimeter may signify increasing integration with the core urban area, and thus could
indicate that the associated area source and mobile source emissions may be appropriate to include in the
nonattainment area" (EPA TSD page 19) out of context and incorrectly suggests that the EPA is claiming
ozone is worse "downtown" than in the foothills to the west of the urban core. The EPA acknowledges
that monitors to the west of urban centers are the violating monitors, as shown it throughout the TSD (for
instance, Figure 14 shows all violating monitors west of urban centers and adjacent to the foothills). The
EPA's analysis in the TSD is consistent with the commenter's statement that ozone concentrations tend to
be highest in the foothills areas to the west of downtown Denver.

Comment: A commenter pointed to the EPA's HYSPLIT back trajectory maps to show that emissions in
northern Weld County end up at violating monitors. The commenter also alleged that Colorado's back
trajectory analysis is flawed because it uses 2013-2015 monitoring data, was not run using the ensemble
mode, and did not calculate back trajectories for any day where the maximum daily 8-hour ozone
concentration exceeded 70 ppb.

EPA Response: The EPA's HYSPLIT back trajectories show at least some trajectories originating or
passing through counties adjacent to the nonattainment area as well as the larger area of analysis in all

46

-------
directions. The presence of one or a few back trajectories in an area upwind in a given direction is not
determinative evidence that the area must be included in the nonattainment area. This is a principle
evident in the designation decisions across the country which utilize HYSPLIT. The HYSPLIT back
trajectories represent modeled simulations of general direction and timing of air flow on analyzed days,
but are not perfect representations of actual air flow. Directions and areas with many trajectories can
provide evidence of nearby emissions contributing to monitored violations on many (or most) high ozone
days, and can indicate areas which should be strongly considered for inclusion. Colorado submitted its
recommendations for ozone nonattainment boundaries in October of 2016; the state cannot be faulted for
not utilizing the incomplete 2016 dataset in a submission requested by the EPA midway through 2016. In
addition, analysis of data over a three-year period, whether 2013-2015 used by Colorado or 2014-2016
used by the EPA, would be considered equally representative of general wind patterns on high ozone
days. The fact that another year has passed, allowing the EPA to evaluate the 2014-2016 data using
HYSPLIT does not in any way invalidate the representativeness of the Colorado evaluation of wind
patterns for high ozone days in two of the same years, plus one earlier year the EPA did not evaluate.
Neither does the EPA's use of 2016 ambient ozone data forthe designation design value calculation mean
that Colorado's 2013-2015 evaluation is "flawed" for ignoring 2016. The commenter provided its own
HYSPLIT analysis of the 2013-2015 data, but used different model options. The EPA produces
HYSPLIT backward trajectories for each violating monitor on each exceedance day, using only input
meteorological data for the location of the monitor and for the exceedance day. Matrix and ensemble
applications of HYSPLIT use meteorological data or trajectory endpoint locations and times that differ
from those of the violating monitor and the exceedance day. To maintain the focus of the analysis directly
on the contribution to the exceedance at the monitor on the exceedance day, EPA does not employ options
that build such a compilation result.

Comment: A commenter stated that the EPA's HYSPLIT analysis fails to consider recirculated emissions
from the prior day and ignores the fact that pollution from further away takes longer to get to a violating
monitor than pollution from closer to the violating monitor.

EPA Response: The HYSPLIT tool uses archived, gridded meteorological data to estimate trajectories air
parcels followed in the hours before arriving at a receptor. Local topography, inaccuracies introduced by
the gridded meteorology data, and other sources of uncertainty mean that HYSPLIT trajectories are best
interpreted as general trends in source regions for a given receptor on given days, and not a precise
representation of actual air parcel travel. If the back-trajectory fails to reconsider observed or suspected
recirculation, one reason would be that the recirculation is not properly characterized in the gridded
meteorology. The EPA has added a discussion of the recirculation represented by the Denver cyclone and
vertical recirculation in the meteorology discussion under Factor 3 in the TSD.

Comment: A commenter stated that the EPA and Colorado's analysis fails to consider the Denver
Cyclone and points to a study that found that "the Denver Cyclone causes oil and gas emissions in Weld
County, which would include the area the EPA is proposing to designate attainment, to contribute to
exceedances of the ozone NAAQS."

EPA Response: The EPA has added further discussion of the recirculation represented by the Denver
Cyclone and vertical recirculation in the meteorology discussion under Factor 3 in the TSD, one of five
factors that the EPA weighed in reaching its decision about the boundaries of the Denver nonattainment
area.

Comment: A commenter disagreed with the EPA that Northern Weld County and Northern Larimer
County is outside the Denver Basin and that the Cheyenne Ridge bounds the nonattainment area airshed
to the North.

47

-------
EPA Response: Colorado included the figure below (Figure 1) to illustrate the terrain features bounding
the Denver Basin. At the right of the figure is the Cheyenne Ridge, rising from the elevation of the Cache
la Poudre and South Platte rivers to a high point near the Colorado,Wyoming border (off the right edge of
the figure). The nonattainment area boundary is shown in grey. The northern sections of Weld and
Larimer Counties fall in the elevated terrain which forms the northern boundary of the Denver Basin. The
Denver Basin is characterized by unique meteorological conditions and topographic features described in
Factor 3 and Factor 4 of the TSD, which indicate that emissions in Northern Weld and Northern Larimer
Counties are not likely to contribute to violating monitors.

The EPA has modified the TSD for the nonattainment area to more accurately characterize the elevated
terrain in northern Weld and Larimer Counties which are excluded from the nonattainment area.

Figure 1 - Topographic illustration of the Denver Basm and final nonattainment area boundary

Comment: A commenter stated that for the same reasons outlined for Weld County in comment EPA-
HQ-OAR-2017-0548-0273, the EPA has not offered a rational basis for excluding part of Larimer County
from the Front Range nonattainment area.

EPA Response: The EPA used the five factor analysis to assess the retention of the same northern
boundary for Larimer County as has been used for the nonattainment area for the 1997 and 2008 ozone
standards. Based on the weight of evidence of the five factor analysis, EPA determined not to modify the

48

-------
State's recommendation that northern Larimer County should not be included as part of the Denver
nonattainment area because emissions from these areas do not contribute to air quality at the violating
monitors in the Denver area.

Comment: Several commenters stated that Northern Larimer County includes significant sources of
ozone precursor emissions and represent a large share of the total county emissions. Larimer County VOC
emissions approach 50,000 tpy. The commenter noted that no other individual county within the NFR
non-attainment area has VOC emissions above 30,000 tpy.

EPA Response: The commenter cites the 2011, V2 NEI for a Larimer County VOC emission estimate
"approach [ing] 50,000 tpy." Table 3 of the EPA TSD shows Larimer County VOC emissions of 8,307
tons per year, based on the 2014 NEI, and omitting biogenic emissions. Table 2 shows that when biogenic
emissions are excluded for the 2014 NEI, Weld, Adams, Denver, Jefferson, and Arapahoe counties all
have greater VOC emissions that Larimer. The EPA does not consider biogenic emissions in the five
factor analysis given that no controls are available for these emission sources. The majority of Larimer
County is included in the nonattainment area, so the relative magnitude of the total county is not
necessarily relevant to whether the northern portion of Larimer County is included in the nonattainment
area. Colorado estimated that the northern portion of Larimer County includes 3,076 tpy of VOC
emissions (37% of the county total).

Comment: A commenter argued that emissions from Martin Drake, Nixon, and Rawhide Power Plants
contribute to Colorado's nonattainment problems and that the EPA must include these major sources of
NOx emissions in the nonattainment area designation because they contribute to the nonattainment
problem.

EPA Response: Martin Drake and Nixon power plants lie in the Colorado Springs CBSA, and are not
part of the Denver CSA. Martin Drake lies 51 miles south of the nearest violating monitor in the Denver
nonattainment area, and Ray Nixon is 64 miles south of that monitor. Neither of the ozone monitors in the
Colorado Springs CBSA have recorded a violation of the NAAQS. Additionally, neither has had a design
value greater than 70 ppb since the 2012-2014 monitoring year. Given these facts, as well as the relative
infrequency of HYSPLIT back trajectories moving from the Colorado Springs area to a violating monitor,
the EPA does not believe that the Colorado Springs area power plants can be considered nearby sources
contributing to violations.

The Rawhide Power Plant is in the portion of Larimer County excluded from the nonattainment area for
the reasons articulated in the TSD and elsewhere in this response to comment document. In response to
the commenter and their attached HYSPLIT model results, the EPA generated a pollution rose using
hourly ozone data from 2013-2016, along with wind speed and direction data from the adjacent
meteorological station operated at the Colorado State University Agricultural Engineering Research
Center. That ozone pollution rose shows the percent of total hours where ozone is coming from each
direction, and indicates through a color scale the directions most associated with elevated ozone levels.
The pollution rose has been incorporated into the TSD for Factor 3 (Meteorology), and shows that nearly
all ozone data with concentrations above 70 ppb comes from the south, southeast or east. Nearly no
elevated ozone comes from the direction of the Rawhide Power Plant (to the north). The EPA evaluated
HYSPLIT back trajectories seen in both the 2013-2015 data set and in the 2014-2016 that the EPA used
nationally. The EPA found that there are far fewer back trajectories originating in the northern portion of
Weld and Larimer Counties (including the Rawhide Power Plant) in the 2014-2016 data set than in the
2013-2015 data set evaluated by the commenter; indicating that the commenters model results
highlighting back trajectories passing west of the Rawhide Power Plant may have been dominated by a
single (or a few) day(s) in 2013

-------
Comment: A commenter stated that more needs to be done to protect the air we breathe citing to
recorded ozone values of 88, 86, 83 and 83 ppb at the National Renewable Energy Lab monitor in the
summer of 2017.

EPA Response: The 2017 ozone data mentioned by the commenter have not yet been certified as
complete and accurate by Colorado, and so are not being used in this ozone designation action. The
comment does not indicate the source for the 2017 data, but three of the four high data values included in
the comment do not appear in the EPA's AQS database or on the Colorado Department of Public Health
and Environment data page. Preliminary 2017 data are comparable to the 2014-2016 data shown under
Factor 1 in the EPA TSD for Colorado. On October 1, 2015, the EPA strengthened primary and
secondary ozone NAAQS and provides increased protection against an array of adverse health effects.

Comment: A commenter stated that "nonattainment areas include areas that contribute to ambient air
pollution in nearby areas that exceed that applicable air quality standard, like Northern Weld and Larimer
Counties." The commenter concluded that area designations are important for the protection of public
health and environment because of requirements associated with nonattainment designations.

EPA Response: The CAA requires that the EPA designate areas that fail to attain a revised NAAQS, as
well as nearby areas which contribute to violations be designated as nonattainment. The EPA utilized a
weight of evidence approach using five factors (Air Quality Data, Emissions and Emissions Related Data,
Meteorology, Geography/Topography, and Jurisdictional Boundaries) to determine nonattainment area
boundaries. Based on consideration of those factors, EPA determined not to modify the State's
recommendation that the northern portions of Weld and Larimer counties should not be included as part
of the Denver nonattainment area because emissions from those areas do not sufficiently contribute to air
quality at the violating monitors in the Denver area. The EPA's TSD describes how consideration of all
five factors led to the boundary the EPA finalized. On October 1, 2015, the EPA strengthened primary
and secondary ozone NAAQS and provides increased protection against an array of adverse health
effects.

Comment: A commenter asserted that including Northern Weld and Larimer Counties in the
nonattainment area will aid Colorado in developing strategies to achieve the ozone standard and cites the
State's 2014-2016 design values to support this statement. The commenter stated that the EPA should
expand the nonattainment area that is contributing to elevated ozone levels so that nonattainment controls
would be applicable to sources in those areas. The commenter concluded that this would result in
additional reductions since more emissions would be subject to controls.

EPA Response: Given the overall weight of evidence of all five factors that the EPA considered in its
TSD, including the relative rarity of trajectories from the excluded areas, the much greater emission levels
within the included areas, it is not clear that inclusions of sources in the excluded areas will aid Colorado
in attaining the NAAQS to any significant degree. If future more refined attainment planning analyses
conducted by the state of Colorado find that such controls are effective for bringing the area into
attainment, the state has CAA authority to address them.

Comment: One commenter stated that the bulk of emission in Northern Weld and Larimer Counties are
from oil and gas production activities that that these emissions are understated in the inventory. The
commenter referenced a series of studies focusing on emissions at natural gas production sites concluding
that emissions from oil and gas facilities are significantly underestimated in inventories and that as many
of these facilities as possible should be included in the nonattainment area to help Colorado reduce ozone
precursors from the industry.

50

-------
EPA Response: Oil and natural gas emissions are a significant portion of the total emissions throughout
Larimer and Weld Counties, not just in the northern portion excluded from the nonattainment area. Figure
4 in the EPA TSD shows far higher well densities in southeastern Larimer, southwestern Weld and
eastern Boulder Counties than in the northern portions of Weld and Larimer Counties excluded from the
nonattainment area. Any understatement of oil and gas emissions would apply to both the sources
included in the nonattainment area and to the sources outside the nonattainment area. The understatement
of emissions in the northern portions would not directly argue that the northern portions should have been
included, because the much greater emissions within the nonattainment area would also have been
understated.

Comment: A commenter pointed to peer-reviewed studies to support its claim that transport from oil and
gas emissions cause ozone exceedances in Colorado and that the northern portions of Weld and Larimer
Counties should be included in the nonattainment area.

EPA Response: As presented in the EPA TSD (pp. 8-11), the EPA also believes that oil and gas
emissions contribute to the ozone precursors in the NAA. The same peer reviewed article included as
supporting material in this comment concludes that the studies analysis suggests that elevated ozone
levels in the Northern Colorado Front Range "are predominantly correlated with air transport from the N-
ESE [from north to east-southeast], which are the upwind sectors where the O&NG [oil and natural gas]
operations in the Wattenberg Field area of the DJB | Denver-Julesburg Basin] are located". According to a
2011 USGS publication, the Wattenberg Field is outlined in the following figure, and is entirely contained
within the intended ozone nonattainment area.

104°W

51

-------
Figure 2 - Outline map of the Wattenberg field in the Denver Basin, Colorado, drawn to contain all wells
producing from the formations studied in this report. The undrilled area in the northern part of the field is
the city of Greeley. Reprinted from U.S. Department of Interior "Gas, Oil, and Water Production from
Wattenberg Field in the Denver Basin, Colorado", U.S. Geological Survey, Reston, Virginia, 2011. Web.
29 Mar 2018.

Comment: A commenter referenced a map produced by the Colorado Oil and Gas Conservation
Commission to demonstrate that areas of oil and gas current and future development will be significant
contributors to ozone in the nonattainment area, and therefore must be included in the nonattainment area.

EPA Response: Consistent with the CAA requirements for designations, the EPA considers existing
emissions but not potential future emissions in assessing whether an area is contributing to violations of
the NAAQS.

Comment: One commenter suggested that the jurisdictional boundaries factor supports inclusion of the
northern portions or Weld and Larimer counties as part of the nonattainment area.

EPA Response: The jurisdictional boundary factor is just one of five that the EPA considers when
determining an appropriate nonattainment boundary. In this case, evidence from the other factors supports
finalizing the proposed boundary.

3.2.8.2 Utah
Wasatch Front. UT

Comment: The State of Wyoming requested that EPA clarify that the same topographic obstacles that
prevent emissions from Summit, Juab, Wasatch, and Morgan Counties in Utah from influencing violating
monitors along the Wasatch Front, also prevent emissions from southwestern Wyoming point sources
from influencing those violating monitors.

EPA Response: In the February 25, 2016 ozone designations guidance, the EPA identifies what factors
States and Tribes should use in their boundary recommendation - as well as what the EPA considers in its
analysis to support final designations. The guidance notes the importance of analyzing whether nearby
areas contribute to a violating area, and that the EPA intends to consider information relevant to
designations associated with the counties in the Combined Statistical Area (CSA). For the Northern and
Southern Wasatch Front nonattainment areas, the area of analysis was limited to the Salt Lake City-
Provo-Orem CSA - and consequently did not include any areas in southwestern Wyoming. Thus, EPA
has not analyzed how topography may affect emissions from sources in southwestern Wyoming.

Comment: A commenter suggested that in addition to the inclusion of Salt Lake County, Davis County,
and portions of Weber and Toole Counties in the Northern Wasatch Front nonattainment area, the EPA
should include the major sources located in Morgan and Summit Counties, adjacent to counties with
violating monitors, in the Northern Wasatch Front nonattainment area. The location of major sources of
NOx in these two counties appears to coincide with many of the lower elevation back trajectories in
EPA's TSD that show impacts to the violating monitors on exceedance days, with trajectories from the
Summit County source impacting the Hawthorne, Ogden, Bountiful and North Provo monitors, and
trajectories from the Morgan County source influencing Hawthorne, Ogden, Bountiful, Harrisville and
Spanish Fork.

-------
EPA Response: While individual point sources are found in both Morgan and Summit Counties, county
level precursor emissions in these counties are lower than the emissions from all of the counties being
included in the nonattainment areas, as shown in Table 3 of the TSD. HYSPLIT trajectories as utilized in
the TSD are useful for visualizing the predominant flow characteristics during high ozone events. As
noted in the TSD Factor 3: Meteorology section, emissions originating within Davis and Salt Lake
Counties as well as the southern portion of Weber County, the northern portion of Utah County, and the
eastern portion of Tooele County, tend to be the primary influencer on violating monitors. The Wasatch
mountain range acts as a terrain block for emissions in areas to the east of the violating monitors. During
high ozone events, low surface winds inhibit transport of emissions, and mountains to the east block
ozone and ozone precursors from having a substantial effect from the Morgan and Summit County
sources. Instead, north, northwest and south, southwest components for wind directions are far more
common in the large scale. It should also be noted that the accuracy of the HYSPLIT model can be
limited with complex topography. As the commenter notes, many of the trajectories suggesting influence
from the Morgan and Summit County major sources are lower elevation trajectories, meaning that the
model is working backward from a point 100 meters above the elevation of the monitor of interest. Those
back trajectories leading to the Morgan and Summit County sources cross a mountain range with typical
elevations of 1,200 meters above the elevation of the monitors. HYSPLIT backward trajectories for the
same days and monitors, with starting heights 500m and 1000m above ground level, which are less likely
to be affected by the mountain range, show more north, northwest and south, southwest components than
do the trajectories with 100m starting heights.

Uinta Basin, UT

Comment: Several commenters expressed concern that the nonattainment designation for the Uinta Basin
might occur before a local, streamlined permitting mechanism is developed for the U&O Reservation as a
nonattainment area (the National O&NG FIP does not apply in areas of Indian country designated
nonattainment). Operators of oil and natural gas facilities wishing to locate or expand operations on the
U&O Reservation currently use the National Tribal minor NSR FIP to obtain authorizations to construct.
Upon nonattainment designation, the National O&NG FIP's construction authorization mechanism will
no longer apply. The commenters requested that the EPA revise the National O&NG FIP so that it applies
to nonattainment areas until a U&O Reservation-specific FIP is in place. The commenters also requested
that the EPA delay the effective date of a nonattainment designation for the Uinta Basin until a U&O
Reservation-specific FIP is finalized.

EPA Response: The EPA is aware that once designated nonattainment, the National O&NG FIP will no
longer apply to the Indian country portions of the Uinta Basin (i.e., the U&O Reservation). The EPA is
also aware of the importance of streamlined permitting, as afforded by the National O&NG FIP, and its
value to economic development in this airshed. We are working to avoid a gap in streamlined permitting
when the Uinta Basin is designated nonattainment for ozone, and, to this end, we are pursuing multiple
options, including a proposed revision to the National O&NG FIP. If finalized, this would allow for
continued streamlined construction authorizations on the U&O Reservation following the effective date of
a designation of nonattainment for the Uinta Basin. Additionally, the EPA is planning on proposing a
reservation-specific FIP to control new, modified, and existing minor oil and natural gas sources on the
U&O Reservation. Under the National O&NG FIP, new and modified true minor oil and natural gas
sources in the oil and natural gas production and natural gas processing segments of the oil and natural
gas sector would need to register their proposed sources with the EPA, as they currently do. In light of the

-------
court's order in In Re Ozone Designation Litigation (N.D. Cal. Nos. 4:17-cv-06900-HSG and 4:17-cv-
06936-HSG), the EPA has limited ability to further delay nonattainment designations.

Comment: Several commenters from the oil and natural gas industry, state, tribal, and local government
request that the EPA proposed a nonattainment boundary in the most targeted possible way. Suggestions
were made to either use a boundary of the exact 6,250 ft. contour line, or a boundary consisting of
"quarter-quarter" section definitions.

EPA Response: The EPA has revised the nonattainment boundary to be defined as the 6,250-ft contour
line. Generally, all land lying below an external contiguous perimeter defined at 6,250 feet in Uintah and
Duchesne Counties will be in the nonattainment area, and lands outside that perimeter above 6,250 feet
will be attainment/unclassifiable. See the TSD for more detailed descriptions of this boundary.

Comment: A commenter contended that based on research cited from Dr. Seth Lyman, the EPA should
postpone nonattainment designations for additional analysis and focus on areas responsible for generating
background ozone levels. The commenter cited an attached report stating that analysis of historical
weather indicates that high ozone would occur in only about Vz of years, high ozone has only been seen in
2/3 of years since 2012, that not all monitors see high ozone in all high ozone winters, that winter time
background ozone from sources outside the Uinta Basin are 40 to 55 ppb, which is 60 to 80% of the 70
ppb standard, and that more work is needed before computer models can be reliably used to develop
mitigation strategies for winter ozone.

EPA Response: The Agency does not give special consideration to background ozone levels as part of
the designation process. The level of the NAAQS and the procedures for determining whether an area is
meeting the NAAQS were established at the time the 2015 ozone NAAQS was promulgated. See 40 CFR
50.19 and Part 50, App. U. Moreover, EPA is not at liberty to postpone designations. The CAA requires
EPA to designate areas within 2 years of promulgation of a NAAQS and allows for a one-year extension
where there is "insufficient information to promulgate the designations." CAA section 107(d)(l)(B)(i).
On December 4, 2017, a coalition of environmental and health organizations filed suit against the
Administrator claiming that EPA failed to meet its mandatory obligation to designate all areas of the
United States for the 2015 ozone NAAQS by October 1, 2017. American Lung Association, et al v. Pruitt
(N.D. Cal. No. 4:17-cv-06900). A coalition of fifteen states also filed a similar suit on December 5, 2017.
State of California v. Pruitt (N.D. Cal. No. 4:17-cv-06936). In a March 12, 2018, order, the court granted
the motions in part and ordered the EPA "to promulgate final designations for all areas of the country
except for the eight undesignated counties composing the San Antonio area no later than April 30, 2018."
See In Re Ozone Designation Litigation (N.D. Cal. Nos. 4:17-cv-06900-HSG and 4:17-cv-06936-HSG)
Although not relevant for purposes of the designations, EPA notes that sources outside the Uinta Basin
have little to no impact on air quality within the Basin. The meteorological conditions (temperature
inversions) associated with high wintertime ozone trap local emissions in the valley and isolate the valley
air from impacts originating from regional or longer-range transport that occurs in upper layers of the
atmosphere. The result is that these wintertime exceedances are the direct result of local emissions rather
than other background sources of ozone. The comments on the frequency of high ozone winters is correct,
but the area violates because those conditions occur frequently enough to cause three-year average ozone
levels to be above the standard.

Comment: Several commenters from local government object to the proposed 6,250 ft. elevation and
believed that a 6,000 ft. elevation boundary is more scientifically supported. One commenter stated that

-------
"We understand that exceedances from the Whiterocks monitor are likely due to the close proximity and
density of wells currently running within a few miles of that monitor" (which is the only violating
regulatory monitor above 6,000 feet), and that the EPA should not have raised the elevation above 6,000
feet, as recommended by Utah, unless other monitors over 6,000 feet also showed violations.

EPA Response: The EPA disagrees that it was not appropriate to raise the elevation boundary above
6,000 feet. As the commenter recognizes, the Whiterocks monitor, which is violating the standard is
located above 6,000 feet. To the extent the commenter is suggesting that only the emission sources in
close proximity to the Whiterocks monitor are contributing to ozone exceedances in the area, we disagree.
Ozone is formed photochemically over time in the atmosphere from precursor (NOx and VOC) emissions
and thus tends to be more regional than local in nature. That formation over time allows for movement of
precursors before ozone reaches a high level causing an ozone exceedance at a particular monitor, even
under relatively low wind speed conditions under winter temperature inversions. During daylight hours in
high ozone episodes, Whiterocks winds are generally from the southwest to southeast, rather than
completely random. This allows precursors from sources within the Basin but at significant distances to
the south of the monitor to move northward while ozone is forming photochemically. In addition, on days
when the Whiterocks monitor exceeds the ozone standard, several other ozone monitors across the basin
at significant distance generally also exceed the standard. For example, during the severe ozone episodes
in early 2013, the Whiterocks monitor recorded ozone exceedances ranging from 72 to 89 ppb on 13 days
between January 22 and February 20. On six of the 13 days in that period, all the other monitors within
the nonattainment area also recorded exceedances; and on no other exceedance day did fewer than 5 of 7
other monitors in the nonattainment area record a coincident exceedance with the Whiterocks monito.

This kind of simultaneous widespread elevated ozone over a monitored area extending sixty miles east to
west and 42 miles north to south is much more indicative of widespread regional ozone formation, and
not of local "hot spot" ozone formation around emission sources in a small area.

Comment: A commenter suggested that the proposed designations misrepresent data from the Rabbit
Mountain monitoring station. The EPA only considered the regulatory data collected at the site in 2012
and 2013, and did not consider the additional non-regulatory data collected at the site from October 2013
until the end of the 2016-2017 winter (non-regulatory data was provided to the EPA with the comment).
The commenter suggests that the monitor has not recorded any exceedances of the 2015 ozone NAAQS
since 2013 and the additional data should be considered in determining the extent of the nonattainment
boundary.

EPA Response: The EPA used all the valid regulatory data available in the EPA AQS database relevant
to the Uinta Basin. That includes five values above the level of the standard recorded at Rabbit Mountain
in 2012, and twelve values above the standard recorded in 2013, the only years the Rabbit Mountain
monitor collected regulatory data. The EPA was not provided additional non-regulatory data collected at
the Rabbit Mountain monitoring station after regulatory monitoring ceased in 2013 until the public
comment period. While non-regulatory monitoring has not recorded ozone exceedances at the site since
2013, quality assurance and quality control (QA/QC) data for the monitor has not been provided to the
EPA; the EPA cannot therefore assess the data quality of this post-2013 data. In addition, during the 2017
ozone season, data was not recorded at the site during the only period in which monitors at other locations
in the basin recorded ozone exceedances that winter; the commenter's suggestion that the site has not
exceeded the standard since 2013 is therefore based on an incomplete data set which misses at least one
critical period. None of the ozone years since 2013, when non-regulatory data was collected and provided
by the commenter has been as severe at other monitors in the Uinta Basin as was 2013. Even with the

55

-------
subsequent data from Rabbit Mountain, located at an elevation of 6,165 feet, the EPA's conclusion that
ozone at levels above the standard can be recorded at the location during the most severe ozone seasons is
not incorrect, and that conclusion strongly supports the use of a 6,250 ft. elevation boundary across the
entire Uinta Basin, including that portion of the basin which includes the Rabbit Mountain monitor.

Please see page 34 of the Utah TSD for a discussion of how this, and other non-regulatory monitors, were
used in EPA's evaluation of the Uinta Basin.

Comment: A commenter suggested that the proposed designations overestimate the emissions in the
vicinity of the Rabbit Mountain monitoring station, since the EPA focused on emissions from oil and gas
wells, population centers, and traffic and vehicle miles traveled, none of which occur in proximity to the
location of the Rabbit Mountain ozone monitor.

EPA Response: The EPA's statutory obligation is to designate all areas violating the ozone standard, as
well as nearby areas contributing to those violations as nonattainment. Oil and gas operations can be
found within four miles of the Rabbit Mountain monitor. The EPA used typical mixing height during
winter ozone episodes to define the upper elevation of the nonattainment area boundary, and the sources
near the Rabbit Mountain monitor indicate it is an appropriate boundary in the southeastern portion of the
basin near Rabbit Mountain as it is elsewhere in the basin.

Comment: A commenter suggested that the proposed designations incorrectly consider meteorology and
topography in the vicinity of the Rabbit Mountain monitoring station. The commenter noted that ozone at
the Rabbit Mountain monitoring station more closely tracks the Rangely, Colorado ozone monitor (700
feet lower and outside the eastern boundary of the nonattainment area) than it tracks the Whiterocks
violating monitor (51 feet higher than Rabbit Mountain and 60 miles northwest).

EPA Response: The EPA used available meteorology and ozone measurement data to determine that an
elevation defined boundary for the nonattainment area was an appropriate boundary that included both the
area violating the standard and nearby areas contributing to the violations.

Comment: A commenter suggested that the area around the Rabbit Mountain monitoring station should
be excluded based on the few emissions sources and lack of recent recorded violations at the monitor.

EPA Response: We disagree. The EPA used available meteorology and ozone measurement data to
determine that an elevation defined boundary for the nonattainment area was an appropriate boundary that
included both the area violating the standard and nearby areas contributing to the violations. As
previously noted, the Rabbit Mountain monitor is a non-regulatory monitor and we cannot draw any
conclusions from that data about whether the area is currently attaining the standard. We note that
historical monitoring data at the site from a monitor that met EPA's regulatory monitoring requirements
indicate that elevated ozone may occur at the location during severe ozone seasons.

Comment: A commenter requests that the area around their phosphate mining operations be excluded
from the nonattainment boundary. The commenter noted that their mining operations are primarily at an
elevation above 6,000 ft. and are therefore not expected to have contributed to ozone violations.
Additionally, the phosphate mine is not a large point source of NOx or VOC emissions, nor are there
significant oil and natural gas development activities in the area. Finally, the commenter contends that the
meteorological conditions that result in ozone violations, the low-emitting ozone precursors from the

-------
mine are not expected to transport downward and contribute to ozone violations. Consequently, the
commenter proposes that the EPA exclude certain townships from the final nonattainment area.

EPA Response: The commenter provides no data to support the statement that ozone precursors emitted
from mine operations are not expected to transport to monitors at lower elevations and thus contribute to
ozone violations at those monitors. The statute does not require and it is not practical for EPA to consider
carving out small portions of a county for inclusion in a nonattainment area based on claims by a single
source that its emissions are not causing or contributing to the violation. We note, however, that the
modification of the boundary in the final TSD to be based on elevation will result in much of this
commenter's mining operation not being included within the nonattainment area.

Comment: Several commenters from the oil and natural gas industry, state, and local government object
to the "> 10 percent" methodology used to determine which townships are included in the nonattainment
boundary. The commenters note that using 10 percent of land area as a basis for determining inclusion in
the nonattainment is arbitrary.

EPA Response: The EPA disagrees that the "greater than 10 percent" methodology was arbitrary. The
EPA used that methodology at the time of the 120-day letters because it had been recommended by the
State of Utah. Based on further dialogue with the State and based on other comments received, for the
final designation the EPA is setting a contiguous 6,250 ft. contour boundary.

Comment: Several environmental groups sent comments in support of the Uinta Basin proposed
nonattainment boundary.

EPA Response: The EPA acknowledges and appreciates the supportive comments. However, based on
further dialogue with the State and consideration of additional comments, the EPA is setting the boundary
as a contiguous 6,250 ft. contour boundary rather than including all townships with greater than 10
percent of land at lower than 6,250 ft.

Comment: Several commenters request that the EPA designate the Uinta Basin at the Moderate
classification either through using 2015-2017 monitoring data (versus 2014-2016), or the EPA's CAA
discretion to designate at a higher classification. One commenter states that the EPA should classify the
Uinta Basin as a Moderate nonattainment area because existing emissions controls will not allow the area
to attain the NAAQS. Further, the commenter contends that the suspension, revision, and/or rescission of
federal oil and gas controls will eliminate the primary emissions control regulations applicable in much of
the Basin and that energy development in the area is expected to increase.

EPA Response: The EPA's final rule published on March 9, 2018, (83 CFR 10376) established the
design value thresholds for area classifications. Based on that rule, the Uinta Basin is being classified as
Marginal. Under section 181(a)(4) of the CAA, an area may be reclassified if the design value for the area
is within 5 percent of the threshold for that other classification. The EPA has not historically reclassified
areas under this provision on its own initiative and is not planning to do so here. If a state or tribe submits
such a request, the EPA will consider whether such reclassification is supported. Regarding the comment
that the area is unlikely to attain the 2015 ozone NAAQS by the Marginal area attainment date without
imposing further controls, the EPA notes that nothing in the Act precludes a state from adopting controls
if they are needed to bring the area into attainment by the applicable attainment date. Regardless of an
area's classification, the Act places the primary responsibility for attaining and maintaining the NAAQS

-------
on the State. Thus, the commenter's concerns about federal regulation is misplaced. If the area remains
classified as Marginal and fails to attain the standard by the Marginal area attainment date, it will be
reclassified at that time to Moderate and the state will be obligated to demonstrate attainment in the area
by the Moderate area attainment date.

Comment: A commenter indicates that they are supportive of designating the Uinta Basin nonattainment;
however, there should be no elevation cap to the nonattainment boundary. The commenter argues that
"almost certainly" atmospheric mixing at elevations higher than 6,250 ft. contribute to nonattainment at
the Whiterocks monitor on at least some days. Further, the commenter claims that the 8% of emissions
above the elevation boundary do influence ozone production in the Basin. The Clean Air Act requires the
EPA to finalize a nonattainment area boundary that includes all sources which contribute to
nonattainment in the Uinta Basin, and by the agency's own admission the 6,250-ft elevation cap does not
meet this requirement, and that it is arbitrary for the EPA to reject Utah's recommended 6,000 ft. upper
boundary that excludes 12% of oil and gas emissions only to choose instead a boundary that excludes 8%
of oil and gas emissions. The EPA shows that their 6,250 ft. boundary does not include all emissions
contributing to nearby violations.

EPA Response: The commenter has not submitted any additional evidence or data to support their claim
that sources above an elevation of 6,250 ft. contribute meaningfully to air quality at the violating
monitors. In light of the State and Tribe's recommendations and the location of the violating monitor; as
well as the majority of sources, the EPA believes that using an elevation of 6,250 ft. is a reasonable
estimation of the area violating the standard and the area with sources contributing to violations of the
standard.

During the Uinta Basin winter ozone studies of 2011, 2012, 2013 and 2014, vertical temperature and
ozone profiles were primarily collected at the Ouray National Wildlife Refuge site (4,692 feet in
elevation), the Fantasy Canyon site (at 4,823 feet) and the Horsepool monitoring site (at 5,148 feet).
Average and extreme ozone measurements aloft are shown below for ozone balloon soundings from
Ouray Wildlife Refuge for several ozone exceedance days in 2013. The standard deviation of ozone is not
over 70 ppb at elevations more than 375 meters above the surface, and mean peak ozone over 70 ppb does
not occur above 325 meters above the surface; these elevations above ground level correspond to
elevations above sea level of 5,922 feet and 5,758 feet, respectively.

-------
450
400
350
300

iD

< 250

v>

i_

QJ

% 200

E

QD

"a>

150

100

50



2013 Ouray
Wildlife Refuge
Ozone Profiles





Morning

(blue)

0 10 20 30 40 50 60 70 SO 90 100 110 120 130 140

Ozone (ppbv)

Figure 1. Summary plot of the 2013 average ozone mixing ratio and standard deviations measured
at all sites during morning (between sunrise and local noon, in blue) and afternoon (noon to sunset,
in red). Note the large range of ozone concentrations in 2013 and the large photochemical
production of ozone in the afternoons.32

At the Horsepool site, above ground level ozone was measured using the NOAA Tunable Optical Profiler
for Aerosol and oZone (TOPAZ) lidar. The TOPAZ lidar found that peak ozone concentrations were
often constrained to the lower 300 meters of the atmosphere (below an elevation of 6,134 feet). Thus,
available measurements of vertical distribution of ozone on high ozone days, as well as the elevation of
surface monitors in the Uinta Basin showing ozone violations support an upper boundary of 6,250 feet.

Comment: A commenter stated that that the EPA has not identified the height of the inversion layer
throughout the Uinta Basm on the worst ozone days. The EPA considered data about inversion layer

32 Final Report, 2013 Uinta Basin Winter Ozone Study, March 2014, ENVIRON (ed.), Section 8, Tethered
Ozonesonde and Surface Ozone Measurements in the Uinta Basin, Winter 2013, p. 8-46; available in the docket for
this action.

59

-------
heights gathered with an ozonesonde from Ouray National Wildlife Refuge, but did not conduct similar
analysis of inversion layer elevation throughout the rest of the Uinta Basin. The EPA's choice of a 6,250-
ft elevation cap thus lacks a rational basis and is not supported by adequate evidence. Rather than set an
arbitrary cap that excludes at least 8% of ozone precursor emissions, the EPA should simply designate the
entire Uinta Basin as a nonattainment area without an elevation cap.

EPA Response: The burden was on the State and the EPA to determine how a boundary line should be
drawn. Portions of the Uinta Basin have never recorded ozone exceedances or violations, and large
portions lack any appreciable ozone precursor emission sources. The EPA used the available data on
mixing height during winter temperature inversions and surface ozone measurements to conclude that an
elevation of 6,250 feet would encompass mixing height on most winter ozone days, as well as
encompassing all violating ozone monitors and those emissions under the mixing height which contribute
to ozone violations. No additional data on mixing height indicating that a higher boundary would be
appropriate is available. The 6,250 ft. elevation was used throughout the basin to define the portion of the
basin contributing to and subject to ozone formation.

Comment: A commenter suggested that the EPA should consider potential future emissions in the Uinta
Basin. Specifically, the commenter identified the Enefit proposed facility known as "South Project"
which the commenter claims has the potential to emit 1,243 tons of NOx and 244 tons of VOC once
complete. They also noted that these sources of future emissions should be considered by the EPA in
determining a nonattainment boundary.

EPA Response: The statutory language governing designations is not written to consider future
violations or future contributions. Rather, the statute focuses on the present through phrases such as "does
not meet" and "contributes." In designating areas, the EPA considers the most recent three-years of air
quality data and also considers the most recent information on emissions and meteorology. Future
projections are not considered for purposes of determining a current violation of the NAAQS or current
contribution to a violation.

Comment: A commenter noted that the contribution to Colorado's elevated ozone levels further supports
the EPA's intended nonattainment designation for the Uinta Basin - specifically, that emissions from
Utah impact Class I visibility areas in Colorado. The commenter stated that the EPA should acknowledge
this and incorporate findings about Utah's contribution to western Colorado's ozone problems into its
final nonattainment designation decision.

EPA Response: The EPA appreciates the commenter's views, but visibility issues are addressed in
separate provisions of the CAA and are outside the scope of this action.

Comment: A commenter stated that the EPA correctly expanded the recommended nonattainment
boundary to ensure that it contains all violating monitors and emissions sources contributing to
nonattainment. The commenter supports the EPA's proposed boundary of 6,250 ft.

EPA Response: The EPA appreciates the commenter's support but notes that it has modified the
boundary for the final designation. The EPA still relies on 6,250 feet for setting the boundary, but as
described above and in the TSD, is using a different methodology for identifying the precise border of the
area.

-------
Comment: A commenter alleged that emissions from oil and gas facilities are a major contributor to
ozone pollution and cites to several sources supporting this statement in relation to air quality in the Uinta
Basin.

EPA Response: The EPA appreciates the commenter's view and has recognized that oil and gas facilities
are the predominant source of emissions in the Uinta Basin.

Comment: A commenter suggested that the EPA should pursue strategies along with the nonattainment
designation to improve air quality in the Uinta Basin.

Response: Once areas are designated for a NAAQS, States - not the EPA - take primary responsibility
for developing plans and choosing the sources to regulate in order to attain and maintain the NAAQS. For
areas under Tribal jurisdiction, the EPA will work with the Ute Tribe to achieve meaningful emissions
reductions to help bring the Uinta Basin into compliance with the 2015 ozone standard as expeditiously as
possible.

Comment: A commenter noted that the Uinta Basin air quality continues to deteriorate, citing a
(preliminary) 2015-2017 design value at the Ouray monitor of 89 ppb, up from the 2014-2016 design
value of 80 ppb. They argue that this is evidence that a Moderate classification rather than a Marginal
classification is needed in the area.

EPA Response: The designations reflect the most recent air quality monitoring data that was required to
be certified (2014-2016), or data from 2015-2017 where a state or tribe has chosen to early certify such
data. Utah and the Ute Indian Tribe of the Uintah and Ouray Reservation have not yet certified data for
2017; and at this time we are not certain what the design value will be for the area using 2015-2017 data.
We note, however, that it is not unusual for design values to fluctuate and that a higher design value in
one three-year period is not necessarily indicative of air quality "deteriorating," but instead could be a
function of meteorology. We respond earlier to requests that EPA immediately to reclassify the area to
Moderate.

Comment: A commenter requested clarification to the TSD discussion on Pages 37-38 whether
townships are included in the analysis for oil and natural gas well inclusion or if the percentages are based
on elevation only.

EPA Response: For purposes of the preliminary 120-day TSD, the EPA intended to establish the
boundary based on townships with land area greater than 10 percent below 6,250 ft. Thus, information
provided in that TSD regarding the scope of oil and gas source emissions located in the nonattainment
area was based on oil and gas sources located in townships with greater than 10 percent of land area
below 6,250 ft. As noted, for the final designations, EPA is basing the nonattainment boundary on the
contiguous 6,25 0-ft. elevation contour.

Comment: A commenter recommended a 6,000 to 6,250 topographic line for establishing the
nonattainment area boundary or that townships should only be included in the nonattainment boundary if
the majority of the land area in the township is below the 6,000 or 6,250 ft. elevation.

EPA Response: The EPA has established the nonattainment area boundary at the contiguous 6,250-ft
elevation contour.

-------
Comment: A commenter alleged that the map on p. 38 of the EPA's TSD is misleading and that the
majority of land area identified as "Tribal" land is actually Ashley National Forest land. The commenter
recommends Figure 17 of the TSD be corrected to show forest service land.

EPA Response: The map referred to by the commenter originated in the recommendation TSD submitted
by the State. The EPA does not suggest or maintain that the map shows areas of tribal or state jurisdiction
for air quality management. Rather, the map from Utah shows what is commonly referred to as the
exterior boundary of the Uintah and Ouray Indian Reservation.

Comment: Several commenters requested that the EPA hold a public hearing in the Uinta Basin before
official nonattainment area designations.

EPA Response: While not required specifically to do so under CAA section 107(d), the EPA invited
public input on our responses to states and tribes regarding these areas during the 30-day comment period.
For full consideration, input from the public was required to be submitted to the docket on February 5,
2018. The EPA believes the public comment period provides sufficient opportunity for members of the
public to comment on this action, which is national in scope. EPA has reviewed and attempted to respond
to all relevant comment received.

Comment: A commenter suggested that the EPA should explain how it has taken into consideration the
complex emission sources, meteorology, and topography of Uinta Basin.

EPA Response: The EPA addressed emissions, meteorology and topography in the December TSD as
part of the five factor analysis. The commenter does not indicate how that analysis is deficient. The EPA
has included the five-factor analysis as part of the TSD for the final designations.

Comment: A commenter noted that historical ozone measurements show no correlation between ozone
readings and oil and gas production, and that there is little reason to think that even shutting down all oil
and gas production in the Uinta Basin completely would cause improved ozone readings in subsequent
years.

EPA Response: We disagree with the suggestion that oil and gas production in the Uinta Basin is not
related to elevated wintertime ozone levels. Winter ozone in the Uinta Basin occurs only during specific
winter meteorological conditions. To form winter ozone, snow cover, persistent cold pool temperature
inversions and sunlight must be present. While snow cover, persistent cold pool temperature inversions
and sunlight occur in many areas, winter ozone has only been observed in two areas - the Uinta Basin and
the Upper Green River Basin in Wyoming. In both of these areas significant oil and gas emissions of
ozone precursors are trapped by the winter meteorology. Poor historical correlation between emissions
and ozone levels is a result of the deterministic nature of the variability in interannual meteorology.

Comment: A commenter indicates that nonattainment designations and subsequent restrictions have a
chilling effect on oil and gas production revenues, and place the Uinta Basin in a competitive
disadvantage to other regional, national, and international oil and gas opportunities.

EPA Response: As an initial matter, we note that economic considerations are not relevant for
determining whether an area violates the NAAQS or contributes to a violation in a nearby area.

Nationally, existing oil and gas production areas do not appear to have been significantly negatively

62

-------
impacted by being included in designated nonattainment areas (e.g., Denver-Julesburg and Wattenburg
fields in the Denver nonattainment area, Barnett Shale in the Dallas-Fort Worth nonattainment area, and
the Jonah and Pinedale fields in the Upper Green River Basin nonattainment area). Some (such as the
Upper Green River Basin) have successfully attained the ozone NAAQS after initial nonattainment
designation without significant curtailment of production and development. The others remain
competitive and are seeing development while operating in current nonattainment areas.

Comment: A commenter cites evidence compiled by Dr. Seth Lyman that indicates the elevation limit
could be 6,000 feet or lower, and to raise that level to 6,250 feet is arbitrary and capricious.

EPA Response: As previously noted, a boundary of 6,000 feet is not supported because areas above that
elevation are violating the 2015 ozone NAAQS. We note that the commenter did not submit, nor directly
cite the evidence they are relying on.

Comment: A commenter alleges that the criteria used by the EPA would capture 93% of Uinta Basin oil
and gas wells within the nonattainment area and represent a substantial economic development impact to
the region. The commenter claims that "this treatment of elevation levels is an example of government
abusing its discretion in a way that negatively impacts property rights."

EPA Response: The EPA increased the elevation level recommended by the state (6,000 feet) by 250 feet
primarily to capture the violating Whiterocks monitor; but also to include areas with sources contributing
to violations of the 2015 ozone NAAQS. The EPA disagrees that the criteria it uses to determine which
areas to designate as nonattainment and the appropriate boundary for a nonattainment area is an "abuse of
discretion." These criteria directly relate to whether an area is violating the NAAQS or is contributing to
air quality in an area violating the NAAQS and these are the statutory basis for designating an area as
nonattainment. See CAA section 107(d)(l)(A)(i).

Comment: A commenter suggested that the EPA clearly distinguish areas of jurisdictional control for air
quality requirements per the Clean Air Act.

EPA Response: In the final TSD and Part 81 tables, the EPA has clarified that the Uinta Basin
nonattainment area includes land under both State and Tribal jurisdiction - and the State of Utah and Ute
Tribe are responsible for 2015 ozone NAAQS implementation requirements in their respective
jurisdictions.

3.2.9. EPA Region IX
3.2.9.1 Arizona
Yuma, AZ

Comment: One commenter proposed that the EPA expand the Yuma nonattainment area to include the
area of high vehicle miles traveled (VMT) directly south of the exceeding monitor to capture more on-
road sources of pollution in Yuma County. The commenter asserts that based on back trajectory modeling
presented in the EPA's TSD, it appears that the area to the south impacts the monitor in Yuma County on
exceeding days.

EPA Response: Based on the analysis presented in our TSD, the EPA does not agree with the commenter
that the EPA should modify the State's recommended area to include the area south of the Yuma monitor
as part of the nonattainment area. The boundary for the Yuma nonattainment area includes the violating

-------
monitor and the most significant stationary sources of ozone precursor emissions in the county. As
discussed in the EPA's TSD, HYSPLIT data shows that approximately 98 percent of back trajectories
originate or flow through Mexico before reaching the Yuma monitor, indicating that exceedances at the
Yuma monitor are heavily impacted by emissions sources in Mexico. As noted in the TSD for this area,
the EPA's pollution transport modeling indicates that anthropogenic sources in all of Arizona contribute
approximately 6% to the projected 2017 design value at the Yuma monitor. Emissions from the nearby
Mexican municipalities of Mexicali and San Luis Rio Colorado are on the order of five times larger than
emissions in Yuma County, and emissions in upwind Imperial County in California, which is also being
designated nonattainment, are also considerably larger than Yuma County emissions. Emissions sources
south of the city of Yuma down to the Mexican border comprise less than two percent of the area-wide
total of ozone precursor emissions when emissions from the upwind areas in Mexico and Imperial County
are considered. For these reasons, the EPA determined not to modify the boundary recommended by the
state for the Yuma area.

3.2.9.2 California
Sacramento, CA

Comment: A commenter contended that the EPA should expand the proposed Sacramento Metro
nonattainment area to include nearby Amador County, Calaveras County, the western portion of Nevada
County, and all of Sutter County. The commenter stated that emissions sources and trajectories provide
convincing evidence that ozone levels in these counties are driven primarily by emissions from the
Sacramento metropolitan area and that following traditional nonattainment area boundaries ignores that
fact. The commenter also asserted that there is a disconnect in the EPA's decision because all of Sutter
County is part of the Sacramento-Roseville Combined Statistical Area but only the southern portion of
Sutter County is included in the Sacramento Metro nonattainment area and the elevated Sutter Buttes are
made their own isolated nonattainment area despite having no population or sources within the boundary.

EPA Response: The State requested that the EPA designate Amador, Calaveras, and Nevada counties as
separate nonattainment areas from the Sacramento Metro area, consistent with how the areas were
designated for both the 1997 and 2008 ozone NAAQS. These three counties are located in a separate air
basin than the Sacramento Valley. As discussed in the California TSD, California has historically been
divided into fifteen distinct air basins that were determined by grouping together areas with similar
geographical and meteorological features. Amador, Calaveras, and Nevada counties are in the Mountain
Counties Air Basin, to the east and southeast of the Sacramento Valley Air Basin. The Mountain Counties
Air Basin consists of a group of counties that are similar in air quality, contain more pronounced
topography, and are more rural in character as compared to the flatter and more populous metropolitan
Sacramento Valley air basin to the west. Thus, geographical and topographical factors associated with
California's air basins support separating Amador, Calaveras, and Nevada counties from the Sacramento
Metro area. Placer County and El Dorado County are also located within the Mountain Counties Air
Basin but are appropriately included in the Sacramento Metro area, and distinguished from the other
counties in the Mountain Counties Air Basin, due to the extent of urbanized land use development
radiating from the City of Sacramento along the Interstate 80 and U.S. Route 50 corridors passing through
the two counties.33 These two counties are more closely integrated with the urban core of the Sacramento
Area. In addition, both air quality and transportation planning are carried out by different agencies in
Amador, Calaveras, and Nevada counties than in the Sacramento Metro area. For instance, all the

33 See population density, population growth, traffic, and commuting information presented in the Factor 2
discussions in the Sacramento Metro, Amador County, Calaveras County, and Nevada County (Western part)
sections of the EPA's TSD for California.

64

-------
counties in the Sacramento Metro area belong to metropolitan planning organizations (MPOs) that
coordinate transportation planning and funding on a multi-county basis,34 whereas Amador, Calaveras,
and Nevada counties each have their own transportation planning agencies, consistent with the more rural
nature of each of those counties.

The EPA believes that it is appropriate to also designate Sutter Buttes as a separate nonattainment area
from the Sacramento Metro area. As discussed in the Technical Analysis for the Sutter County (Sutter
Buttes) section of the California TSD, the violating monitor is atop a mountain at approximately 2,115
feet. The Sutter Buttes monitoring site was established at the special Sutter Buttes location specifically to
assess transport aloft, rather than as an indicator of conditions in Sutter County. Its unique characteristics
distinguish it from the other locations included in the Sacramento Metro, and warrants its status as a
separate nonattainment area. Sutter County itself has no major pollution sources, and total county
emissions are small compared to the more urbanized areas included in the Sacramento Metro area. There
is no evidence of elevated ozone there aside from the peak of Sutter Buttes. A second monitor in Sutter
County located on the valley floor in Yuba City is closer to higher population census tracts, small point
sources, and areas of higher VMT and is attaining the 2015 NAAQS. Thus, the evidence suggests that the
mountaintop area is predominately affected by transport from both the Sacramento Metro and San
Francisco Bay areas, which EPA is also designating nonattainment. The Sutter Buttes boundary for the
2015 ozone standard is consistent with the nonattainment boundary for the 1997 ozone standard and the
area has established, operational air planning mechanisms in place resulting from the previous
designation.

Based on our analysis, the EPA also does not intend to modify the State's recommendation to designate
part of Sutter County as attainment. As recommended by the State, the EPA is including a southern
portion of Sutter County in the Sacramento Metro nonattainment area, consistent with the boundaries for
the 1997 and 2008 ozone standards. The northern part of the county is being designated attainment,
except for the area around the violating Sutter Buttes monitor, which is affected by wind and pollution
conditions aloft rather than emissions from within Sutter County. As discussed above, the second Sutter
County monitor located on the valley floor is attaining the 2015 NAAQS. Thus, because a monitor
indicates that this portion of Sutter County is attaining and the evidence does not demonstrate a
contribution to concentrations at a violating monitor, we do not intend to modify the State's
recommendation.

3.2.9.3 Nevada
Las Vegas. NV

Comment: One commenter contended that EPA should expand the Las Vegas Nonattainment Area to
include major sources of NOx to the east of the Las Vegas Valley in Hydrographic Area 215. The
commenter asserted that the two sources identified in the EPA's TSD as Nevada Cogeneration Associates
#2 (near the northern end of Lake Mead) and PABCO Building (just east of the nonattainment area
boundary) have the potential to influence monitors in the Las Vegas Valley, despite topographic and
meteorological obstacles, based on lower elevation back trajectories presented in the EPA's TSD.

34 The Sacramento Area Council of Governments (SACOG) is the MPO for El Dorado, Placer, Sacramento, Yolo and
Sutter counties, and the Metropolitan Transportation Commission (MTC) is the MPO for Solano County.

-------
EPA Response: First, the EPA would like to clarify the locations of the point sources discussed by the
commenter. The commenter identifies Nevada Cogeneration Associates #2 as being near the northern end
of Lake Mead; however, Nevada Cogeneration Associates #2 and PABCO Building Products are located
just east of the Las Vegas Valley, within less than one mile of one another. These sources are difficult to
distinguish on the maps in the EPA's TSD due to their proximity. The source near Lake Mead is J. R.
Simplot Company (Simplot Silica Products). The EPA's TSD mistakenly identified this source as being
in Hydrographic Area 215, but it is actually located in Hydrographic Area 220, near the border with
Hydrographic Area 215. Given the error and to ensure we fully address the comment; the EPA is
addressing all three of these sources in our response.

As discussed in the EPA's TSD, Nevada Cogeneration Associates #2 and PABCO Building Products are
separated from the Las Vegas Valley by topographical barriers, including Sunrise Mountain, Frenchman
Mountain, and several smaller, generally north-south trending ridges. The data indicate that winds during
exceedance days are predominately from the southwest, originating in southern California and passing
over locations in the southern half of Clark County. Of the 108 exceedances at violating monitors
(occurring on 49 different calendar days), seven exceedances have HYSPLIT 24-hour back trajectories
passing through the general area identified by the commenter. These trajectories tend to spend less time
near the Nevada Cogeneration Associates #2 and PABCO Building Products source area and more time in
the more heavily populated, high-VMT areas of the Las Vegas Valley. Furthermore, these point sources
account for just 0.4% of total county NOx and VOC emissions,35 the area is sparsely populated, and
vehicle miles traveled (VMT) levels are low relative to the Las Vegas Valley.

As also discussed in the EPA's TSD and above, the predominant wind flow is not consistent with a
contribution to violations from J.R. Simplot Company, the source near the northern tip of Lake Mead.
While there are two exceedance days when back trajectories originate near J.R. Simplot Company, its
emissions occur 24 hours away in transport time and more than 80 miles away in transport distance, and
are likely dispersed by the time they reach monitors in the Las Vegas Valley. Furthermore, wind does not
typically flow directly from the area towards the Las Vegas Valley - trajectories originating near J.R.
Simplot Company typically also travel through the more heavily populated, high-VMT areas of the Las
Vegas Valley (from the south to the north) before reaching the violating monitors. The area near J.R.
Simplot Company is sparsely populated and is characterized by low VMT relative to the Las Vegas
Valley, and J.R. Simplot Company, which is the only point source in Hydrographic Area 220, accounts
for just 0.1% of total county NOx and VOC emissions.

For these reasons, EPA does not believe that an evaluation of the five factors supports modification of the
State's recommendation to include the areas near these sources as part of the designated nonattainment
area.

3.2.10. EPA Region X

The EPA received no comments from either the public or states and tribes pertaining to the areas in
Region X.

35 Emissions information are based on 2014 National Emissions Inventory (NEI) data.

-------
3.2.11. Multi-State Areas

3.2.11.1 New York-Northern New Jersey-Long Island, NY-NJ-CT

Comment: A commenter noted that none of the counties in Pennsylvania are included in the NYC area,
eliminating Pennsylvania's impact on anticipated Moderate nonattainment area planning for the area.

EPA Response: Conducting a five-factor analysis on the given CSA is the basis for including or
excluding any particular county within the CSA. The EPA conducts area-specific analyses to support
nonattainment area boundary recommendations and final boundary determinations by evaluating factors
such as air quality data, emission and emissions-related data, meteorological data, geography/topography
and jurisdictional boundaries.

As the NYMA TSD explains, EPA has excluded the counties of Dutchess, Orange, Putnam and Ulster in
New York; and Carbon, Lehigh, Monroe, Northampton and Pike in Pennsylvania from the New York
Metro nonattainment area because of the area-specific factors analysis EPA concluded that the excluded
counties' contributions were not sufficient to include them in the nonattainment area. We explain why we
excluded other counties in Pennsylvania that are outside the CSA in other responses to comments.

While Pennsylvania is not included in the NYMA nonattainment area, it is already part of the Ozone
Transport Region and therefore has planning requirements similar to a Moderate nonattainment area,
including RACT and New Source Review (NSR). Moreover, Pennsylvania in its entirety, is subject to the
regional transport requirements and is part of the CSPAR and its Update.

Comment: A commenter suggested that Pennsylvania should be included in an expanded nonattainment
area because EPA's CSAPR Update modeling showed it contributed more to nonattainment monitors in
southwest Connecticut than in-state sources by as much as a factor of two and contributed as much, or
more, than New Jersey.

EPA Response: As noted, the starting point for analyzing the designation of an area is the CSA. The EPA
does not believe that creation of a larger nonattainment area to address pollution transport is the
appropriate approach. As an initial matter, section 107(d)(1) provides that areas designated nonattainment
should include any "nearby" area contributing to a violation of the NAAQS. We believe that broad super-
regional areas go beyond this by including areas that are not necessarily "nearby" but contribute to
nonattainment through long-range transport. EPA used a similar approach with the 1997 and 2008 ozone
NAAQS. Furthermore, the Clean Air Act Amendments of 1990 set, by operation of law, the boundaries
for Serious, Severe, and Extreme nonattainment areas for the 1979 1-hour ozone NAAQS as the
metropolitan statistical area or consolidated metropolitan statistical area (as established by the Bureau of
the Census).

Comment: A commenter noted the EPA did not evaluate the emissions from the entire New Jersey CSAs.

EPA Response: The EPA evaluated the entire New York-Newark, NY-NJ-CT-PA CSA and the entire
Philadelphia-Reading-Camden, PA-NJ-DE-MD CSA in its corresponding TSDs. These two CSAs include
the entire State of New Jersey.

Comment: A commenter contended that the EPA's TSD did not address New Jersey's technical data or
evaluate the emissions of all of the counties in the CBSA/CSA, or provide a basis for why the technical
data provided by New Jersey were not adequate for including these counties in the recommended ozone
nonattainment area.

-------
EPA Response: The EPA evaluated the entire New York-Newark, NY-NJ-CT-PA CSA and the entire
Philadelphia-Reading-Camden, PA-NJ-DE-MD CSA in its corresponding TSDs and made note of New
Jersey's recommendation. The State made a recommendation for a super-regional area including parts of
several nonattainment areas. The EPA reviewed and fully evaluated the State's recommendation, in
addition to other states' recommendations for the designation process. In the EPA's factor analysis, the
analytical starting point for the designation of an area is the CSA. The EPA does not believe that creation
of a larger nonattainment area to address pollution transport is the appropriate approach. As an initial
matter, section 107(d)(1) provides that areas designated nonattainment should include any "nearby" area
contributing to a violation of the NAAQS. We believe that broad super-regional areas go beyond this by
including areas that are not necessarily "nearby" but contribute to nonattainment through long-range
transport.

Comment: A commenter noted that NJDEP's November 2016 designation recommendation is consistent
with EPA's guidance, because it considered the entire CSA, which includes the
Allentown/Bethlehem/Easton and East Stroudsburg, PA Metropolitan Statistical Areas. Adding these
"nearby" counties to New Jersey's northern ozone nonattainment area would add the Commonwealth of
Pennsylvania to the ozone planning process for the northern New Jersey nonattainment area to address
their significant portion of the ozone nonattainment.

EPA Response: The EPA evaluated the entire New York-Newark, NY-NJ-CT-PA CSA in its
corresponding TSD for the NYMA. Consistent with EPA's Designation Guidance, the analysis is based
on the counties within the CSA, not on the overall contribution of the Commonwealth of Pennsylvania.
The EPA evaluated the counties of Pike, Monroe, Northampton and Lehigh, Pennsylvania based on air
quality data, emission and emissions-related data, meteorological data, geography/topography and
jurisdictional boundaries, and concluded that the excluded counties' contribution was not sufficient to
include them in the nonattainment area.

Comment: A commenter noted the EPA's five-factor analysis must be the criteria for determining the
extent of nonattainment, independent of state recommendations.

EPA Response: The EPA agrees that EPA has an independent role in the designations process, and that
the Administrator must designate area as he deems appropriate and consistent with the CAA provisions
governing area designations. EPA fully considers state recommendations and supporting information
within the framework of a five-factor analysis.

Comment: A commenter pointed out that the NYC TSD reviews the five factors for areas outside the
NYC CSA, but does not distinguish between the differing effects of emissions of VOCs and NOx. Using
a 10,000 per year per county (or lower) threshold for NOx alone would help address the lack of
Pennsylvania counties in the NYC nonattainment area. Using VOC and NOx at the same threshold,
despite their differing effects, does not have a technical basis, since even in nearby locations NOx control
is more effective than VOC control. The NYC metro nonattainment area needs to be expanded to counties
in Pennsylvania with higher NOx emissions

EPA Response: The New York-Newark, NY-NJ-CT-PA CSA, with the additional county of Middlesex
in Connecticut, was the area of analysis for the NYMA TSD. The counties in Table 1, with the exception
of Middlesex County in Connecticut, are part of this CSA. Mercer and Ocean (NJ) counties in the CSA
were excluded from the area of analysis because they were analyzed as part of the current Philadelphia-
Wilmington-Atlantic City, PA-NJ-MD-DE nonattainment area and discussed in separate TSDs.

-------
Within the "nearby" area the EPA reviewed county-wide emissions of nitrogen oxides (NOx) and VOCs
in the area of analysis, and the EPA also reviewed emissions from large point sources and mapped the
sources out during its analysis. The EPA looked at both the NOx and VOC emissions from each of the
counties along with the analysis of the remaining factors to conclude the basis for inclusion or exclusion
of the counties in the designation area.

With respect to the excluded Pennsylvania counties in the NYMA CSA, we disagree that distinguishing
between NOx and VOC emissions would have resulted in including any of those counties in the New
York nonattainment area. To briefly reiterate the EPA's analysis in the TSD, only one of the Pennsylvania
counties, Northampton, has more than 10,000 total NOx tons per year (tpy), and other relevant factors
point against including all the counties, including Northampton, as part of the nonattainment area.
Specifically:

• The emissions and emissions-related data. Carbon, Lehigh, Monroe and Pike counties have less
than 10,000 total NOx tons per year (tpy), while only Northampton County has more than 10,000
- it has 12,944 total NOx tpy. All five counties have less than 10,000 total VOC tpy.

• Population density and degree of urbanization. The population in these counties ranges from
55,000 to 360,000 people.

• Traffic and VMT. Carbon and Pike counties were the lowest counties for 2014 total VMT within
the CSA, while none of the five counties had more than 20% of commuters commuting to
counties with violating monitors.

• Meteorology. Based on the HYSPLIT (HYbrid Single-Particle Lagrangian Integrated Trajectory)
data and the overall Ozone Concentration/Wind Direction Frequency Plots for a Coastal Monitor
(Westport), the EPA concluded the typical trajectories for exceedance days was from the
southwest towards the northeast, minimizing the potential impacts of the five Pennsylvania
counties within the CSA (which are mostly western).

• Geography/topography. These considerations did not directly impact the conclusions.

• Jurisdictional boundaries. The EPA considered existing jurisdictional boundaries and considered
the pre-existing nonattainment boundaries for the New York Metro area. As noted, EPA's
intended boundary for the nonattainment area is the same as EPA's previously established
nonattainment boundaries associated with the 1997 and 2008 ozone NAAQS.

Based on the EPA's analysis of all the factors, the EPA excluded from the NYMA nonattainment area the
five Pennsylvania counties in the CSA.

Comment: A commenter noted the New Jersey's TSD demonstrate that the VOC and NOx
emissions from particular Pennsylvania counties are much greater than the emissions from the counties
within New Jersey that are already a part of the ozone nonattainment areas.

EPA Response: As noted, the EPA used a full five-factor analysis to determine if a county was excluded
or included in its designated areas within each CSA. Using a single factor to determine applicability
would not be valid. See corresponding TSDs.

Comment: A commenter pointed out that Lehigh, Northampton, Monroe, and Pike counties in
Pennsylvania have over 37,000 daily commuters into New Jersey. It appeared to the commenter that the
EPA did not evaluate this data since there is no response to New Jersey's analysis within EPA's TSDs.

EPA Response: The EPA evaluated Lehigh, Northampton, Monroe, and Pike counties in Pennsylvania,
as they are part of the NYMA CSA. As noted, a full five-factor analysis was performed for the five
Pennsylvania counties that are in the NYMA CSA, including evaluating the extent of commuting from
each of those counties to counties with violating monitors. See NYMA TSD, Table 5. Traffic and

-------
Commuting Patterns. After considering that information, as well as the other factors, the EPA excluded
these counties from the designated area.

3.2.11.2. Philadelphia-Wilmington-Atlantic City, PA-NJ-MD-DE

Comment: A commenter stated that including New Castle County, Delaware, in the Philadelphia Non-
Attainment Area (NAA) will not produce additional emission reductions or reduced impacts on other
counties because Delaware is already implementing all Marginal or Moderate nonattainment area
requirements. The commenter states that EPA includes New Castle County in the proposed 2015
Philadelphia NAA rationalizing that it is "nearby" whereas EPA arbitrarily excludes upwind areas which
are causing the area's problem because they are not "nearby." Delaware's impact on the Philadelphia NAA
is dwarfed by the impact of pollution which is transported into the region. Delaware is being harmed by
being subject to the nonattainment area provisions of the CAA while upwind areas that are the primary
cause of the problem are not subject to such provisions. As a result, cost effective emissions reduction
opportunities outside the Philadelphia NAA area will not be realized, and the uncontrolled or under-
controlled sources that are contributing to the problem will remain so.

EPA Response: CAA section 107(d)(1) directs the EPA to designate as nonattainment any area that does
not meet (or that contributes to ambient air quality in nearby areas that does not meet) the NAAQS. New
Castle County, Delaware, has a monitor violating the 2015 ozone NAAQS based on the most recent three
years of air quality monitoring data; for that reason, New Castle County must be designated
nonattainment under CAA section 107(d)(1). The EPA does not arbitrarily exclude areas that are not
"nearby" from the nonattainment area. Rather, the EPA is acting in accordance with the EPA's guidance
when determining that New Castle County should be included in the Philadelphia Nonattainment Area,
discussed in the TSD for the Philadelphia area (EPA-HQ-OAR-2017-0548-0083). The EPA determined
that emissions from sources in "nearby" areas, such as New Castle County, and neighboring area in
Pennsylvania, New Jersey, and Maryland are contributing to local air quality.

Outside of the designations process, the CAA requires that states prohibit certain emissions from in-state
sources if such emissions impact the air quality of downwind states. See CAA section 110(a)(2)(D)(i).
The EPA does not dispute that certain named upwind states in the petition might impact air quality in one
or more downwind states that are measuring violations of the 2008 ozone NAAQS. However, the EPA
believes that states and the EPA can effectively address the upwind states' impacts on downwind ozone
air quality through the good neighbor provision. The EPA has already taken steps to address interstate
transport with respect to the 2008 ozone NAAQS through the promulgation of the CSAPR Update, which
reduces emissions starting with the 2017 ozone season. The EPA used the authority of CAA sections
110(a)(2)(D)(i)(I) and 110(c) to tailor a remedy focused on the precursor pollutant most likely to improve
ozone levels (currently NOx) in downwind states and those sources that can most cost-effectively reduce
emissions within a limited timeframe (i.e., EGUs). 82 FR 51245.

Germaine to this discussion, the EPA further considered Delaware's nonattainment with the 2008 ozone
NAAQS (in New Castle County) and that cost-effective controls required by the OTR requirements
should be applied, in its response to the petition filed by some OTR states (including Delaware) to expand
the size of the OTR under section 176A of the Act. See Dec. 9, 2013 Petition Seeking Expansion of OTR
at http://www.regulations.gov, Docket No EPA-HQ-OAR-2016-0596. The EPA's full response to the
petition is set forth at 82 FR 51238 (Nov. 3, 2017). See also Proposed Response to Section 176A Petition,
82 FR 6509 (Jan. 19, 2017).

Comment: The State of Delaware commented that because the EPA has chosen not to align the
nonattainment areas with the scope of the problem (i.e., designating the smallest areas possible as

-------
nonattainment), the state requests that New Castle County be designated as a stand-alone nonattainment
area, rather than be included in the Philadelphia NAA.

EPA Response: As explained in the TSD for the Philadelphia Area (pp. 2-5), the starting point used by
the EPA in evaluating appropriate boundaries for a nonattainment area is the CSA or CBSA in which a
violating monitor is located. This is consistent with the approach Congress specified for Serious and
above areas for the 1-hour ozone NAAQS for designations immediately following passage of the Clean
Air Act Amendments of 1990 (see section 107(d)(4)(A)). This is the same approach the EPA used for
designations for the 1997 and 2008 ozone NAAQS and the approach the EPA said it would follow in
setting nonattainment area boundaries for the 2015 ozone NAAQS. See EPA's Designations Guidance for
the 2015 NAAQS, https://www.epa.gov/ozone-designations/epa-guidance-area-designations-2015-ozone-
naaqs. The EPA then performs a five-factor analysis to determine whether any of the counties in the CSA
are likely to contribute to nonattainment in another county. As noted in the TSD, HYSPLIT back
trajectories indicate that emissions from New Castle County impact most violating monitors in the
Philadelphia CSA. There is a high density of trajectories through New Castle County to monitor
240150003 in Cecil County, Maryland; monitor 340070002 in Camden County, New Jersey, monitor
340150002 in Gloucester County, New Jersey; monitor 420450002 in Delaware County, Pennsylvania;
and monitors 421010024 and 421010048 in Philadelphia County, Pennsylvania. Similarly, emissions
from nearby counties in neighboring Pennsylvania, New Jersey, and Maryland impact violations in New
Castle County. Based on this and other information, the EPA made a determination that New Castle
County is appropriately included with neighboring counties in the Philadelphia area. TSD, pp. 58-72.

Comment: A commenter noted that Delaware's impact on the Philadelphia NAA is dwarfed by the
impact of pollution transported into the region.

EPA Response: The standard in section 107(d)(1) for designating an area as nonattainment is "any area
that does not meet (or that contributes to ambient air quality in a nearby area that does not meet)" the
standard. The question of the degree of contribution necessary to include a nearby area in a
nonattainment area was addressed by the Court in Catawba County v. EPA, 571 F. 3d. 20 (D.C. Cir.
2009). In that case, the Court found that section 107(d) was ambiguous regarding how EPA is to measure
contribution and "what degree of contribution is sufficient to deem an area nonattainment..." Id. at 39
(emphasis added). The court determined that EPA reasonably applies its multi-factor analysis to each
specific case in evaluating whether counties in the metropolitan area "sufficiently contribute" for
purposes of including a county as part of the nonattainment area. Id. The issue of whether regional
pollution sources that are not "nearby" also contribute or contribute larger amounts, while an important
part of the context of understanding the nature of the ozone problem in each area, is not central to our
analysis and conclusion that New Castle County is interconnected to the greater Philadelphia area, and
has emissions that contribute to nonattainment in other nearby counties in the Philadelphia area, and that
emissions from the nearby counties in Maryland, New Jersey, and Pennsylvania, all surrounding New
Castle County, contribute to nonattainment in New Castle County.

Comment: A commenter noted that the CAA requirement for RACT in the OTR has helped reduce
emissions in the OTR via reasonably available and economically feasible controls, and there is no reason
why all major upwind sources should not have RACT level controls.

EPA Response: The EPA appreciates the commenter's views; however, the issue of imposing RACT on
all upwind sources goes beyond the scope of designations which, as previously noted, address designation
of areas violating the standard and areas contributing to air quality as a nearby violating monitor. As
noted in the previous response, the EPA has separately addressed a petition from Delaware requesting that
EPA expand the OTR and thus subject additional areas to the RACT requirement that applies to areas in
the OTR.

-------
Comment: A commenter pointed out that the EPA's likely classification of the Philadelphia Area and
other upwind areas as "Marginal" nonattainment will likely not result in new or upgraded controls in
those areas, so upwind ozone and transport of precursors from areas upwind of Delaware will not be
sufficiently mitigated.

EPA Response: As stated in EPA's response to the section 176A petition, EPA believes that the
measures adopted under Section 110(a)(2)(D)(i)(l) of the Act, such as CSAPR and the CSAPR update,
and other emission reduction requirements for on-road and non-road mobile sources adopted under other
provisions of the Act, will effectively reduce the emission of NOx and VOCs that contribute to ozone
formation from upwind areas that are not part of the designated nonattainment area. See 82 FR 51244.

Comment: The PADEP recommended that Montgomery County should be removed from the
Philadelphia area because the monitor is attaining and EPA has reduced the size of nonattainment areas
boundaries to be smaller than the core based statistical area (CBSA)/CSA. Pennsylvania cites the
following examples of areas where EPA's intended boundaries for the 2015 ozone NAAQS contain fewer
counties than the same areas for the 2008 ozone NAAQS.

Nonattainment Area

Ohio Counties for the 2008

Proposed Ohio Counties for the



NAAQS

2015 NAAQS

Cincinnati, OH-KY-IN

Butler

Butler



Clermont

Clermont



Clinton





Hamilton

Hamilton



Warren

Warren

Cleveland-Akron-Lorain, OH

Ashtabula





Cuyahoga

Cuyahoga



Geauga

Geauga



Lake

Lake



Lorain

Lorain



Medina

Medina



Portage

Portage



Summit

Summit

Columbus, OH

Delaware

Delaware



Fairfield

Fairfield



Franklin

Franklin



Knox





Licking

Licking



Madison



EPA Response: The EPA agrees that the boundaries for a nonattainment area for the 2015 ozone
NAAQS need not be identical to the boundaries for the 2008 ozone NAAQS and that for several areas the
EPA has agreed with state recommendations to designate a different, typically smaller, area as
nonattainment for the 2015 ozone NAAQS. Each decision on whether an area is included as part of the
nonattainment area is based on a five-factor analysis considering all of the counties within the area of
analysis. As an initial matter, we note that Pennsylvania's recommendation to exclude Montgomery
County relies solely on the fact that Montgomery County is not violating the 2015 ozone NAAQS based
on the most recent three years of air quality data.

Although Pennsylvania did not provide support to show that Montgomery County is not contributing to
air quality at other monitors in the Philadelphia area, the EPA believe that it's five factor analyses for

72

-------
Montgomery County distinguishes that County from Clinton County in the Cincinnati area, Ashtabula
County in the Cleveland-Akron-Lorain (Cleveland) area, and Knox and Madison Counties in the
Columbus area.

As stated in the Cincinnati TSD (EPA-HQ-OAR-2017-0548-0095), Clinton County has relatively low
NOx and VOC emissions, population, population density, and VMT, and less dense HYSPLIT back
trajectories than the counties included in the nonattainment area. In the Cincinnati area of analysis,
Clinton County is ranked 12th for NOx emissions and 9th for VOC emissions. Clinton County is 10th in
population and 9th in population density. Clinton County has low VMT, accounting for less than 3
percent of CSA VMT. The 2014-2016 HYSPLIT back trajectories show that transport winds blew
predominantly from the west southwest, southwest, southeast and east directions during times when the
violating monitors in the Cincinnati area measured exceedances of the 2015 ozone NAAQS. Clinton
County is in the northeastern portion of the area of analysis. Clinton County has a less dense pattern of
back trajectories. These combined factors informed the EPA's decision to exclude Clinton County from
the Cincinnati area.

As stated in the Cleveland TSD (EPA-HQ-OAR-2017-0548-0094): "While Ashtabula County has
moderate emissions as compared to other counties in the area of analysis (17% and 23% of Cuyahoga
County's NOx and VOC emissions, respectively), the county ranks relatively low in population density
and VMT and has only two HYSPLIT trajectories that pass through the county on days that the violating
monitors are exceeding the NAAQS." The 2014-2016 HYSPLIT back trajectories show that transport
winds blew predominantly from the west, southwest, and south during times when the violating monitors
in the Cleveland area measured exceedances of the 2015 ozone NAAQS. Together, these figures show a
dense pattern of HYSPLIT back trajectories across counties in the western, southwestern, and southern
portions of the area of analysis. Ashtabula is in the northwestern portion of the Cleveland area of analysis
and only two back trajectories cross it, which is the fewest within the area of analysis. These factors
support EPA's decision to exclude Ashtabula County from the Cleveland area.

As stated in the Columbus TSD (EPA-HQ-OAR-2017-0548-0094), Knox County has very few HYSPLIT
back trajectories passing over it, and "does not stand out sufficiently with respect to any of the other
factors to support inclusion in the nonattainment area." In the Columbus area of analysis, Knox County
ranks 15th in NOx emissions and 11th in VOC emissions, 8th in population and population density, and 15th
on VMT. Similarly, although Madison County has a moderately dense pattern of HYSPLIT back
trajectories, those trajectories pass almost completely through Franklin County, which has much higher
emissions, before reaching the violating monitor. Madison County ranks 10th in NOx and 13th in VOC
emissions, 12th in population and population density, 8th in VMT, and 7th in number of workers
commuting to or within a county with a violating monitor. These combined factors informed EPA's
decision to exclude Knox and Madison Counties from the Columbus area.

By contrast, as stated in the Philadelphia TSD (EPA-HQ-OAR-2017-0548-0097), Montgomery County
ranks second in both NOx and VOC emissions in the Philadelphia area of analysis. Montgomery County
ranks second in population, fourth in population density, and first in VMT. Numerous HYSPLIT back
trajectories pass through Montgomery County on their way to the various violating monitors in the
Philadelphia area, with a high density of trajectories through Montgomery County to the violating
monitors in Bucks County, PA, Mercer County, NJ, and Philadelphia County, PA (monitor 421010024).
This means that the dominant winds on days when the monitors in Bucks County, PA, Mercer County,
NJ, and Philadelphia County, PA go through Montgomery County. Therefore, Montgomery County is
upwind of and therefore contributing to, those violating monitors. Considering these factors, EPA
reasonably concludes that emissions from Montgomery County contribute to air quality at violating
monitors in the Philadelphia area

-------
Comment: The PADEP stated that while EPA's TSD shows that 90.1 percent of commuters who live in
Montgomery County commute to counties with violating monitors, after EPA concurs on Pennsylvania's
EE demonstration, that figure will be 42.2 percent.

EPA Response: The EPA notes that 42 percent of commuters from Montgomery County commuting to
counties with violating monitors is not a small percentage. In addition, the percentage of commuters into
counties with violating monitors is only one sub-factor in the EPA's five factor analysis. As described
above, EPA's analysis as set out in the Philadelphia area TSD (EPA-HQ-OAR-2017-0548-0097) shows
that other factors support a conclusion that emissions from Montgomery County contribute to nearby
violating monitors in the Philadelphia area, and therefore Montgomery County should be included in the
Philadelphia nonattainment area.

Comment: A commenter noted that the Philadelphia TSD does not evaluate outside counties' potential
influences on its nonattaining monitors.

EPA Response: The Philadelphia TSD evaluates the Philadelphia-Reading-Camden, PA-NJ-DE-MD
CSA in addition to Mercer and Ocean Counties in New Jersey. Consistent with the approach Congress
adopted in 1990 for areas classified as Serious or higher for the 1-hour ozone NAAQS, EPA
recommended to states that they use (and EPA indicated in guidance36 that it would use) Core Based
Statistical Areas (CBSAs) or Combined Statistical Areas (CSAs) as a starting point for considering the
geographic boundaries for ozone nonattainment area for the 2015 ozone NAAQS. This boundary could be
adjusted to either include additional areas or exclude areas based on EPA's analysis five factors, to
evaluate nearby contribution to the violating monitors.37 EPA believes that factor analysis is a logical way
of assessing nearby contribution and thus determining whether specific areas should be included as part of
the designated nonattainment area.

The current Philadelphia TSD evaluates the nearby areas that impact the nonattainment monitors. Going
beyond the nearby areas and addressing ozone transport is beyond the scope of this designation because
other provisions of the Act address longer-range ozone transport and the designation provisions require
only that "nearby" areas that contribute to violations of the NAAQS be included as part of the
nonattainment area.

Comment: A commenter stated that the Pennsylvania counties of Berks, Lebanon, Lancaster and York
should be included in New Jersey's southern multi-state ozone nonattainment area. These counties have
large amounts of VOC and NOx emissions that are transported on high ozone days to New Jersey. These
four counties should be added to the southern multi-state ozone nonattainment area.

EPA Response: As noted, the starting point for analyzing the designation of an area is the CSA. The EPA
does not believe that creation of a larger nonattainment area to address pollution transport is the
appropriate approach. As an initial matter, section 107(d)(1) provides that areas designated nonattainment
should include any "nearby" area contributing to a violation of the NAAQS. We believe that broad super-
regional areas go beyond this by including areas that are not necessarily "nearby" but contribute to
nonattainment through long-range transport.

36	EPA's February 25, 2016 guidance memorandum, "Area Designations for the 2015 Ozone National Ambient Air
Quality Standards

37	EPA has used area-specific analyses to support nonattainment area boundary recommendations and final
boundary determinations by evaluating factors such as air quality data, emission and emissions-related data,
meteorology, geography/topography and jurisdictional boundaries.

74

-------
The EPA used a similar approach with the 1997 and 2008 ozone NAAQS. Furthermore, the Clean Air Act
Amendments of 1990 set, by operation of law, the boundaries for Serious, Severe, and Extreme
nonattainment areas for the 1979 1-hour ozone NAAQS as the metropolitan statistical area or
consolidated metropolitan statistical area (as established by the Bureau of the Census).

The counties of Lebanon, Lancaster and York, Pennsylvania, are not part of the Philadelphia CSA
(Philadelphia-Reading-Camden, PA-NJ-DE-MD CSA). The EPA evaluated the factors for Berks County,
Pennsylvania, and determined based on the factors that the County should be designated as separate from
the Philadelphia area. However, EPA stated in its December 20, 2017 letter to Governor Tom Wolf
(EPA-HQ-OAR-2017-0548-0145), that, if EPA approves Pennsylvania's exceptional events (EE)
demonstration, the 2014-2016 design value for the Berks County monitor would move from violating to
attaining the 2015 ozone NAAQS. In that case, EPA would revise its recommendation for the Reading
Area to attainment/unclassifiable. The EPA concurred on Pennsylvania's EE demonstration for the
Reading airport ozone monitor (monitor 420110011) in Berks County on March 6, 2018. Therefore, EPA
is designating Berks County as attainment/unclassifiable. Please see corresponding TSDs for more
information.

PA's analysis of all the factors, the EPA excluded the five Pennsylvania counties in the CSA.

Comment: A commenter suggested that the Philadelphia area should be expanded to include counties
westward to the high-emitting counties in Pennsylvania.

The only county excluded from the nonattainment area that is within the Philadelphia-Reading-Camden,
PA-NJ-DE-MD CSA is Berks County. The EPA evaluated the factors for Berks County, Pennsylvania,
and determined based on the factors that the County should be designated nonattainment as a single-
county area separate from the Philadelphia area. Importantly, meteorology shows, in Figures 6a-e and 6g-
o in the TSD, that violating monitors in the Philadelphia-Wilmington-Atlantic City area are generally not
impacted by Berks County relative to other counties in the area of analysis. The HYSPLIT back
trajectories for the violating monitors in the area of analysis are predominantly from the south and
southwest and Berks County is to the west or northwest of the other counties in the area of analysis.

Please see corresponding TSDs for further information.

3.2.11.3. Washington, DC-MD- VA

Comment: A commenter asked for clarification and confirmation that Montgomery County, Maryland
(MD) is included in the EPA's proposed Washington, DC-MD-VA nonattainment area for the 2015 ozone
NAAQS. The TSD, which the EPA provided for the proposed Washington, DC-MD-VA nonattainment
area, stated that: 1). The State of Maryland had requested that Montgomery County be included in the
Washington, DC-MD-VA nonattainment area, and 2). according to the Agency's own analysis in its TSD,
Montgomery County should be included in the designation recommendation. However, the EPA
neglected to verify Montgomery County as included in the proposed nonattainment area in table 1 of the
TSD and in its initial 120-day letter to Maryland. The commenter asks that the EPA confirm that
Montgomery County, MD is included in the proposed Washington, DC-MD-VA nonattainment area for
the 2015 ozone NAAQS.

-------
EPA Response: The EPA erred in its TSD for the proposed Washington, DC-MD-VA nonattainment
area for the 2015 ozone NAAQS, by neglecting to include Montgomery County, MD in table 1, which
lists the counties that the Agency proposed to include in the Washington, DC-MD-VA nonattainment area
for the 2015 ozone NAAQS. In a December 20, 2017 120-day letter to the State of Maryland, the EPA
also neglected to include Montgomery County, MD in the list of counties that the State had recommended
to be included in the Washington, DC-MD-VA nonattainment area as well as the counties that the EPA
intended to designate as part of the area. The EPA became aware of its error shortly after publication of
the TSD in the Federal Register on January 5, 2018. By letter dated February 9, 2018, the EPA notified
Maryland's Governor Hogan of this mistake and confirmed that the EPA supports the State's
recommendation by proposing that Montgomery County, MD be designated as part of the Washington,
DC-MD-VA nonattainment area. A copy of the letter was published in the Federal Register on February
23, 2018 and placed in the docket for this matter.

3.2.11.4	Louisville, KY-IL

Comment: A commenter agreed with and incorporates by reference comments submitted to this docket
from the Rural Electric Cooperative Association, the Kentucky Division for Air Quality and the Midwest
Ozone Group.

EPA Response: The EPA notes that it did not receive comments regarding the EPA's intended
nonattainment boundaries from the Kentucky Division of Air Quality or the Rural Electric Cooperative
Association. To the extent the commenter is referring to any Kentucky Division of Air Quality 2016
ozone designation recommendations for the 2015 ozone NAAQS, Kentucky recommended a designation
of attainment/unclassifiable for the Louisville Area and did not include an analysis for the Louisville
Area. Their recommendation was based on attaining ambient air quality data from 2013, 2014 and 2015.
The EPA must use the most recent three-years of quality assured, certified ambient air quality data and
therefore, the EPA is using 2014, 2015 and 2016 data which shows a violation of the 2015 ozone
NAAQS. Based on this data, the EPA conducted a five factor analysis that is found in this docket in the
Louisville, Kentucky TSD. As a result of this analysis, the EPA is designating the county with a violating
monitor, Jefferson County, Kentucky, and four other counties (two more in Kentucky and two in Indiana)
as the Louisville nonattainment area. The EPA did receive comments from the Midwest Ozone Group and
addresses those comments in separate summaries and responses in this response to comment document.

Comment: A commenter asserted that the Louisville air quality designation for the 2015 Ozone NAAQS
impacts the East Kentucky Power Cooperative's (EKPC's) Bluegrass Station located in Oldham County,
Kentucky, and that the EPA should re-evaluate the EPA's analysis of the Louisville KY-IN area as it
pertains to Oldham County, Kentucky, and should designate Oldham County, Kentucky as
unclassifiable/attainment based on the following:

•	FACTOR 1 - the Oldham County monitor is in attainment with the 2015 Ozone NAAQS and
therefore, the EPA focused its analysis on determining whether Oldham County is an area that
contributes to the NAAQS violation in Jefferson County, the only surrounding county with a
violating monitor.

•	FACTOR 2 - The commenter provided the following additional information and asserts that it
was not provided in the EPA's TSD but that it demonstrates that Oldham County should be
designated attainment:

o The Kentucky Transportation Cabinet data showing 1-71 traffic counts suggests that

".. .mobile source emissions are a large contributor to air quality in Oldham County, rather
than the other sources of emissions within the county..." They suggest that the EPA perform
additional analysis of mobile source emissions, e.g., the benefits of reformulated gasoline and
"nitrous oxide (NOx)" controls on compression ignition engines should be better understood

76

-------
and quantified. The commenter concludes from this that "... EPA should not impose
restrictions on stationary sources that are not a primary contributor to air quality concerns in
the Louisville area." The commenter states that a large number of commuters into Jefferson
County come from Jefferson County and that "... reformulated gasoline demonstrates the path
to NAAQS compliance, rather than more stringent regulation of point sources in Oldham
County."

o The commenter described the Bluegrass Station as a minor Title V/Prevention of Significant
Deterioration (PSD) source that operated zero hours in 2014, 484 hours in 2015 and 355
hours in 2016, with normal load peak in the winter and the commenter contends, therefore,
the Bluegrass Station is not a contributor to Jefferson County ozone air quality. The
commenter contended that "Imposition of stricter controls on minor stationary sources like
Bluegrass will not affect the air quality in Jefferson County during the summer ozone season"
and concluded that the EPA should perform further study on mobile source emissions prior to
designating the Louisville area,
o The commenter contended that the 2010-2015 population growth in Bullitt, Oldham, Shelby
and Clark documented in the EPA TSD does not "favor" designating these counties with
"small population growth" as nonattainment.

•	FACTOR 3 - The commenter recommended that the EPA further examine the meteorological
data. The commenter contended that Jefferson County is likely contributing to the air quality in
Oldham County, that Oldham County is downwind from Jefferson County, that "the typical wind
direction for the majority of the time in the Louisville area comes from the south/southwest."
The commenter pointed to a March 2009 boundary recommendation from the Kentucky Energy
and Environmental Cabinet for the 2008 ozone NAAQS designation process and 2006-2008
HYSPLIT modeling as evidence.

•	FACTOR 4 - The commenter agreed with the EPA that there are no physical features impacting
air pollution transport in the region.

•	FACTOR 5 - The commenter stated that the EPA has the flexibility to exclude Oldham County
from the nonattainment boundary because the EPA has excluded Oldham County from the
Louisville ozone nonattainment boundary in the past for the 1979 ozone NAAQS, and designated
the entire Louisville area attainment for the 2008 ozone NAAQS.

•	The commenter stated that costs of NOx controls on minor sources must be considered in
addition to the five Factor analysis. The commenter provided cost information for installation of
NOx controls on combustion turbines and states that it is "nonsensical" to expend these costs
when considering weight of evidence that Jefferson County's ozone air quality concerns do not
come from Oldham County point sources.

EPA Response: We agree that the Oldham County monitor is attaining and the Jefferson County monitor
is violating the 2015 Ozone NAAQS for the 2014-2016 time period. The EPA must designate as
nonattainment not only those areas violating the NAAQS, but also those areas contributing to a nearby
violation of the NAAQS. In our TSD, we have considered all of the technical information available to the
EPA and the EPA's TSD provides our full analysis supporting the inclusion of Oldham County as part of
the designated nonattainment area.38

The EPA agrees with the commenter that mobile source emissions are a large contributor to air quality in
Oldham County, Kentucky and in the Louisville Area. The commenter suggests that because mobile
source emissions are a large component of the emissions profile for the area, "EPA should not impose
restrictions on stationary sources." We note that in considering whether a county is contributing to a
violation in a nearby area, the EPA evaluates all emissions and emissions related data from the National
Emissions Inventory including emissions from on-road mobile sources, point sources, non-point (i.e.,

38 The Commonwealth of Kentucky did not provide ozone designation recommendations for the Louisville KY-IN area.

77

-------
area) sources, non-road mobile sources, and fires. If, full consideration of all five factors supports
inclusion of a county as a contributing county, then the EPA includes the county as part of the designated
nonattainment area. Contrary to the suggestion of the commenter, the EPA does directly impose controls
on stationary sources based on their inclusion in a nonattainment area. States have primary responsibility
to determine appropriate emissions controls during the NAAQS implementation process. We further note
that the EPA does not consider regulatory controls that may apply to an area in determining whether that
area is either violating the NAAQS or is a nearby area contributing to a violation of the NAAQS. The
EPA disagrees that further studies on mobile source emissions must be performed prior to designating the
Louisville area; the EPA is required to designate areas within two years of promulgation of a new or
revised NAAQS with a possible one-year extension where there is insufficient information to promulgate
the designation. The commenter has not supported that current information is insufficient to promulgate a
designation for the area.

The EPA agrees that a large number of commuters come from Jefferson County. However, as stated in
the EPA's TSD, the EPA has determined that commuting patterns from Oldham and Bullitt Counties,
taken with other factors, such as population growth and population density in conjunction with
meteorology, indicate emissions from Oldham and Bullitt Counties contribute to air quality at the
Jefferson County violating monitor. The commenter has not provided compelling evidence that under a
five-factor analysis, emissions in these counties do not contribute to air quality at the violating monitor.

In their comment, the commenter states that Jefferson County sources are likely contributing to air quality
in Oldham County and references meteorology analyses from the designations for the 2008 Ozone
NAAQS to support this claim. The EPA agrees that the 2006-2008 meteorology analyses referenced by
EKPC show that air quality in Oldham County was likely impacted by sources in Jefferson County during
the 2006-2008 timeframe. However, this information is not relevant for the EPA's current assessment of
the potential for emissions sources in Oldham County to contribute to violations measured in Jefferson
County during the 2014-2016 timeframe. Since the monitor in Oldham County is not currently violating
the 2015 Ozone NAAQS, the EPA has not assessed the potential for sources in Jefferson County to
contribute to air quality in Oldham County for this designation action. Meteorology can vary from year to
year. The EPA has evaluated the designation for the 2015 ozone NAAQS on the most recent HYSPLIT
data rather than data from 10 years ago that was relevant for designations for the previous ozone NAAQS.

The EPA also notes that we did not receive additional information from Kentucky in response to our
intended designations released on December 20, 2017. The comments from EKPC refer to meteorology
information collected by the Commonwealth of Kentucky which appears to contradict the EPA's
conclusion about Oldham County contributing violations in Jefferson County. Neither EKPC nor the
Commonwealth have provided any additional meteorology information for the 2014-2016 timeframe.

We note that while we do consider the nonattainment boundaries of an area that was designated
nonattainment for the 1979 1-hour ozone NAAQS, 1997 8-hour ozone NAAQS or, in this instance,
attainment area for the 2008 ozone NAAQS, we consider the jurisdictional primarily in determining how
to draw the boundary of the area once we determine the areas that are violating the 2015 ozone standard
and the nearby areas that are contributing to air quality at the violating monitors. The fact that a particular
area was included or excluded from the nonattainment area for a previous ozone NAAQS is not
controlling for purposes of determining a violation or contribution for a different NAAQS. As a point of
clarification and correction, the EPA notes that portions of Oldham and Bullitt Counties were included in
the nonattainment area for Louisville for the 1979 1-hour ozone NAAQS, and the entire counties of
Oldham and Bullitt were included in the nonattainment boundary for the 1997 8-hour ozone NAAQS.

-------
Finally, we note that the designation provision in section 107(d) of the CAA does not provide for
consideration of regulatory burden or compliance costs in determining whether an area is violating the
NAAQS or contributing to a nearby violation of the NAAQS.

Comment: Several commenters supported the EPA's intended boundary for Louisville, KY-IN area.
One commenter also supported lowering the ozone standard to at or below 60 ppb.

EPA Response: The EPA recognizes the commenter's views. Revision of the ozone standard is beyond
the scope of this action.

3.2.11.5. Chicago, IL-IN- WI

Comment: Several commenters expressed support for EPA's intended boundary for the Chicago, IL-IN-
WI area.

EPA Response: The EPA appreciates the commenters' views. Based upon further analysis, EPA is
modifying the boundaries for the area as described in the final TSD.

Comment: The State of Indiana and other commenters contended that Lake and Porter Counties in
Indiana should be designated as attainment/unclassifiable for the 2015 ozone NAAQS. One commenter
noted that these counties have not recorded a violation of the NAAQS. This commenter further contended
that these counties should not be designated nonattainment on the basis of any contribution to violating
monitors within the Chicago area. Another commenter endorsed the comment from the previous
commenter.

EPA Response: The EPA agrees with the commenters that there are no monitors in Lake or Porter
County that show a violation of the 2015 ozone NAAQS. However, areas must be designated
nonattainment not only if they are violating the 2015 ozone NAAQS, but also if they contribute air
quality at a monitor in a nearby area that is violating the 2015 NAAQS. As provided in the final TSD, the
EPA is designating Porter County and a portion of Lake County in Indiana as attainment/unclassifiable
for the 2015 ozone NAAQS. The EPA is designating the remaining portion of Lake County in Indiana as
nonattainment for the 2015 ozone NAAQS.

Comment: The mayor of the City of Kenosha requested that a smaller portion of Kenosha County, WI be
designated as nonattainment, and recommends that Hwy. 31 (Green Bay Road) rather than 1-94 be used as
the boundary for the portion of Kenosha County included in the Chicago nonattainment area.

EPA Response: The EPA disagrees with the suggestion that Hwy 31 be used as the boundary for the
portion of Kenosha County included in the Chicago nonattainment area; this boundary would capture
only one of the two violating monitors in Kenosha County, WI. As provided in the final TSD, however,
EPA is modifying the boundary from what was provided in the 120-day letter to the State; EPA is
designating the area inclusive and east of 88th Ave as nonattainment for the 2015 ozone NAAQS. The
EPA is designating the area west of 88th Ave as attainment/unclassifiable.

3.2.11.6 Cincinnati, OH-KY-IN

Comment: The commenters expressed support for the counties and partial counties included in the
EPA's intended nonattainment designation for the Cincinnati area. Some commenters particularly
expressed support of EPA's intended nonattainment designation for Lawrenceburg Township in Dearborn
County, Indiana as part of the Cincinnati area. One commenter also contended that the EPA should

-------
extend the nonattainment area further into Dearborn County to include sources in Aurora, Indiana, citing
a high density of HYSPLIT back trajectories for a Hamilton County monitor as showing potential impacts
on exceedance days.

EPA Response: In response to the EPA's intended designations, The Indiana Department of
Environmental Management (IDEM) provided a summary of photochemical modeling39 analyzing ozone
precursor emissions from Dearborn County. For this analysis, IDEM performed several photochemical
modeling simulations using with the Comprehensive Air Quality Model with Extensions (CAMx)40
version 6.3 for the period of June 15 - August 4 using EPA's 2011 modeling platform with 2023el
emissions41. The CAMx simulations included one case which used emissions from EPA's 2023el
emissions platform and several cases where a portion of 2023 emissions in Dearborn County were
removed ("zeroed-out"). IDEM then compared the model-predicted ozone concentrations from each zero-
out scenario back to model-predicted ozone concentrations from the base 2023el scenario with
unperturbed 2023 emissions. In one scenario, IDEM zeroed-out point source VOC emissions in Dearborn
County; in a second scenario IDEM zeroed out both VOC and NOx point source emissions in Dearborn
County. The model showed changes to maximum daily 8-hour average ozone concentrations in the
Cincinnati area of up to 0.1 ppb when both NOx and VOC point source emissions were zeroed out and of
up to 0.001 ppb for when only VOC point source emissions were zeroed out. These impacts are minor.

Ideally IDEM would have modeled 2011 or a recent year rather than a projected 2023 emissions case
since designations are based on an assessment of current conditions rather than future conditions. In
addition, due to an expansion at the largest VOC point source in Dearborn County which was not
accounted for in the 2023 emissions projection, the 2023 VOC emissions attributed to that source were
substantially underestimated. Further, it would have presented a more complete case if Indiana had
zeroed-out all anthropogenic emissions in Dearborn County rather than just point sources. Nevertheless,
given the very small impacts on ozone predicted by the zero-out simulations, modifications to the
simulation reflecting these changes would likely have no more than a negligible impact on the minor
impacts demonstrated by the modeling submitted by the state. In addition, the negligible impact from
VOC reductions alone suggests that the Cincinnati area is NOx-limited and that underestimates of 2023
Dearborn County VOC emissions are unlikely to impact model-predicted ozone levels in the Cincinnati
area. Therefore, the EPA is not including Dearborn County, Indiana in the Cincinnati, OH-KY
nonattainment area. The EPA is designating Butler, Clermont, Hamilton and Warren Counties in Ohio
and the parts of Boone, Campbell and Kenton Counties in Kentucky identified in Kentucky's
recommendation as the Cleveland nonattainment area for the 2015 ozone NAAQS.

3.2.11.7	Cincinnati, OH-KY-IN and Louisville, KY-IN

Comment: Several commenters expressed support for EPA's intended boundary for the Louisville, KY-
IN and Cincinnati, OH-KY areas, but also contend that EPA should designate Jefferson County in Indiana
as nonattainment for the 2015 ozone NAAQS. These commenters expressed concern about the
accumulation of pollution in the Ohio River Valley from large coal-burning power plants and other
sources that contribute to ozone formation, especially the Clifty Creek power plant in Jefferson County,

39	See February 16, 2018 letter from Bruno L.Pigott, Commissioner, Indiana Department of Environmental
Management to Cathy Stepp, Regional Administrator, USEPA Region 5 (docket document ID number EPA-HQ-
OAR-2017-0548-0292).

40	Model documentation and download available at http://www.camx.com/

41	These emissions are documented as part of EPA's 2011 Version 6.3 Air Emissions Modeling Platform available at
https://www.epa.gov/air-emissions-modeling/2011-version-63-platform

80

-------
IN. The commenters state that Jefferson County, IN contributes to nonattainment in both Louisville and
Cincinnati.

EPA Response: Section 107(d)(1) of the Clean Air Act directs EPA to designate an area as
"nonattainment" if it is violating the NAAQS or if it is contributing to a violation of the NAAQS in a
nearby area. As described in EPA's February 25, 2016 designation guidance for the 2015 ozone NAAQS
(ozone designation guidance), after identifying each monitor indicating a violation of the ozone NAAQS,
EPA analyzed those nearby areas with emissions potentially contributing to the violating area. The ozone
designation guidance provided that using the Core Based Statistical Area (CBSA) or Combined Statistical
Area (CSA) as a starting point for the contribution analysis is a reasonable approach to ensure that the
nearby areas most likely to contribute to a violating area are evaluated. Jefferson County, Indiana is not
part of nor adjacent to the Cincinnati-Wilmington-Maysville, OH-KY-IN CSA, which was the area of
analysis for the Cincinnati area Rather, Jefferson County, IN is part of the Louisville/Jefferson County-
Elizabethtown-Madison, KY-IN CSA. Therefore, EPA considered Jefferson County for its potential to
contribute to the violation of the standard in the Louisville area. The EPA's ozone designation guidance
identifies a five-factor analysis which is used to determine the areas that should be included as part of a
designated nonattainment area for the 2015 ozone NAAQS. These factors are: air quality data, emissions
and emissions-related data, meteorology, geography/topography, and jurisdictional boundaries. Based on
consideration of all five factors as described in the TSD for the Louisville area, EPA is not including
Jefferson County, IN in the nonattainment area. Jefferson County, IN has low population, low population
density, and low population growth, as well as low VMT and low commuter links to counties within the
Louisville CSA that are violating the 2015 ozone NAAQS. Significantly, the meteorological data did not
suggest that emissions from Jefferson County, IN are transported to the violating monitors on days on
which the 2015 ozone NAAQS is exceeded.

3.2.11.7 St. Louis, MO-IL

Comment: A commenter urges the EPA to finalize its intended designations for the St. Louis MO-IL
nonattainment area that includes Franklin, Jefferson, St. Charles, and St. Louis counties and the City of
St. Louis. The commenter disagrees with Missouri Department of Natural Resources '(MDNR) proposed
exclusion of Franklin and Jefferson counties as nonattainment. The commenter also believes it would be
irrational for MDNR to rely upon 2015-2017 monitoring data while also using 2014 NEI data.

EPA Response: The EPA's final designation for the St. Louis MO-IL nonattainment area is based on the
latest available monitoring data, which is now 2015-2017 data early-certified by both Missouri and
Illinois. EPA's new analysis of this information is included in the final TSD. In that analysis EPA
concludes that Jefferson is not included in the nonattainment area, and a portion of Franklin County
(Boles Township) is included in the nonattainment area.

The 2017 NEI is not yet available and the 2014vl NEI is the most recent emissions data compiled by the
EPA. Thus, there is not a requirement for the state to use an inventory that is not yet available.

3.3 Comments Not Requiring a Response from EPA
3.3.1 NAAQS Related

Comment: Some commenters congratulated EPA for setting high air quality standards. One commenter
noted that poor air quality leads to breathing problems, higher medical costs and also has a negative effect
attracting people, businesses and tourism.

-------
EPA Response: The EPA appreciates the commenters' views and agrees that poor air quality is a public
health concern. The impact of air quality on medical costs or local economy is beyond the scope of this
action.

Comment: Some Commenters submitted comments in support of revisiting or delaying the 2015
NAAQS.

EPA Response: The EPA appreciates the comments. Revising or delaying the 2015 ozone NAAQS is
outside the scope of this action.

Comment: One commenter supported protective NAAQS and specifically mentions the Louisville,
Kentucky area.

EPA Response: The EPA appreciates the commenter's views and agrees standards protective of public
health are important.

Comment: One commenter supported proposal to designate Louisville nonattainment for 2015 ozone
NAAQS. The commenter also supports lowering ozone standard to at or below 60 ppb.

EPA Response: The EPA appreciates the commenter's views. Revision of the ozone standard is beyond
the scope of this action.

Comment: Several commenters stated that ground-level ozone poses a serious threat to public health and
environment and that ozone precursors should be reduced to improve regional air quality for the benefit of
community health and the environment.

EPA Response: The EPA appreciates the comments and agrees that high levels of ground-level ozone
can pose a threat to public health and the environment. On October 1, 2015, the EPA Administrator
signed a notice of final rulemaking that revised the primary and secondary ozone NAAQS (80 FR 65292;
October 26, 2015). The EPA lowered the revised primary and secondary ozone NAAQS from .075 parts
per million (ppm) to 0.070 ppm. Designating nonattainment areas is an important step toward identifying
areas where additional reductions in ozone precursor emissions can improve air quality.

Comment: Several commenters supported revoking the 2008 NAAQS.

EPA Response: Revoking the 2008 ozone NAAQS is beyond the scope of this action.

3.3.2 Comments on Unrelated Programs

Comment: The EPA received several comments addressing the listing of endangered species or other
aspects of the NEPA program.

EPA Response: These comments do not address the ozone designations which are the subject of this
action. They are beyond the scope of the action and do not require a response.

Comment: The EPA received a comment addressing the Regulatory Flexibility Act.

EPA Response: This comment does not address ozone designations and does not require a response.

-------