^tDSR%
I A \
PH«I«^
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
-------
-------
EPA-454/R-19-003
April 2019
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
U.S. Environmental Protection Agency
Office of Air Quality Planning and Standards
Air Quality Assessment Division
Air Quality Modeling Group
Research Triangle Park, NC
-------
Contents
EXECUTIVE SUMMARY 5
1. Background 9
2. 03 and Secondary PM2.5 Formation in the Atmosphere 12
3. Framework for Developing MERPs as a Tier 1 Demonstration Tool 16
3.1. Definition of MERPs as a Tier 1 Demonstration Tool 17
3.2. Development of MERPs through Photochemical Modeling 18
3.2.1. EPA Single Source Photochemical Modeling for 03 and Secondary PM2.5 19
3.2.1.1. EPA Modeled Impacts: Annual and Daily PM2.5 23
3.2.1.2. EPA Modeled Impacts: 8-hour Ozone 26
3.2.1.3. EPA Illustrative MERPs: Annual and Daily PM2.5 29
3.2.1.4. EPA Illustrative MERPs: 8-hour Ozone 34
3.2.2. Use of Other Photochemical Modeling to Develop MERPs for 03 and Secondary PM2.5 36
3.2.2.1. Developing Area Specific MERPs 38
4. Application of the MERPs to Individual Permit Applications 40
4.1. Illustrative MERP Tier 1 Demonstrations for Example PSD Permit Scenarios 44
4.1.1. Source Impact Analysis: 03 and PM2.5 NAAQS 44
4.1.2. Source Impact Analysis: Class 1 PSD Increment for PM2.5 50
4.1.3. Cumulative Impact Analysis: 03 and PM2.5 NAAQS 54
5. References 61
Appendix A. Hypothetical Sources Included in the EPA's Modeling Assessment 64
4
-------
EXECUTIVE SUMMARY
EPA finalized revisions to the Guideline on Air Quality Models (the "Guideline," published as
Appendix W to 40 CFR part 51) that recommend a two-tiered approach for addressing single-
source impacts on ozone (O3) and secondary particulate matter less than 2.5 microns in
diameter (PM2.5) (U.S. Environmental Protection Agency, 2017a). The first tier (or Tier 1)
involves use of appropriate and technically credible relationships between emissions and
ambient impacts developed from existing modeling studies deemed sufficient for evaluating a
project source's impacts. The second tier (or Tier 2) involves more sophisticated case-specific
application of chemical transport modeling (e.g., with an Eulerian grid or Lagrangian model).
As EPA introduced in the preamble to the 2015 proposed revisions to the Guideline, Modeled
Emission Rates for Precursors (MERPs) can be viewed as a type of Tier 1 demonstration tool
under the Prevention of Significant Deterioration (PSD) permitting program that provides a
simple way to relate maximum downwind impacts with a critical air quality threshold (e.g., a
significant impact level or SIL) (U.S. Environmental Protection Agency, 2018). The purpose of
this document is to provide a framework for permitting authorities and permit applicants on
how air quality modeling can be used to develop relationships between precursors and
maximum downwind impacts for the purposes of developing a technically credible Tier 1
demonstration tool.
A conceptual understanding of an area's emission sources and which precursor emissions limit
the formation of secondary pollutants such as O3 and PM2.5 is useful for interpreting modeled
and monitored impacts due to changes in emissions to that area. O3 formation is a complicated,
nonlinear process that depends on meteorological conditions in addition to volatile organic
compounds (VOC) and nitrogen oxides (NOx) concentrations (Seinfeld and Pandis, 2012). Warm
temperatures, clear skies (abundant levels of solar radiation), and stagnant air masses (low
wind speeds) increase O3 formation potential (Seinfeld and Pandis, 2012). In the case of PM2.5,
or fine PM, total mass is often categorized into two groups: primary (i.e., emitted directly as
PM2.5from sources) and secondary (i.e., PIVh.sformed in the atmosphere by precursor
emissions from sources). PM2.5 organic carbon is directly emitted from primary sources and also
formed secondarily in the atmosphere by reactions involving VOCs. PM2.5 sulfate, nitrate, and
ammonium are predominantly the result of chemical reactions of the oxidized products of sulfur dioxide
(S02) and NOx emissions and direct ammonia (NH3) emissions (Seinfeld and Pandis, 2012).
A Tier 1 demonstration tool, as described in the Guideline, consists of technically credible air
quality modeling that relates precursor emissions and secondary pollutant impacts from
specific or hypothetical sources (U.S. Environmental Protection Agency, 2017a). Existing
credible air quality modeling generally may include single source modeling based on an
approved State Implementation Plan (SIP) demonstration, a more recent submitted but not yet
approved SIP demonstration, or modeling not used to support a SIP demonstration but
considered representative of the current air quality in the area and of sufficient quality that is
comparable to a model platform supporting a SIP demonstration.
5
-------
Figure ES-1 illustrates the framework for MERPs as a Tier 1 demonstration tool. This framework
is the organizing flow of this guidance and sequences from the concept of a MERP, how MERPs
can be developed from either existing EPA modeling or other credible sources, and then how
that information can be credibly used for a source impact analysis and, if necessary, a
cumulative impact analysis.
Figure ES-1. Framework for MERPs as a Tier 1 demonstration tool.
Define concept of MERPs (Section 3.1}
i
Development of MERPs with photochemical modeling
(Section 3.2)
Use of MERPS for individual permits (Section 4)
Properly supported MERPs provide a straightforward way to relate modeled downwind impacts
with an air quality threshold that is used to determine if such an impact causes or contributes
to a violation of the appropriate National Ambient Air Quality Standard (NAAQS). To derive a
MERP value for the purposes of a PSD compliance demonstration, the model predicted
relationship between precursor emissions from hypothetical sources and their modeled
downwind impacts can be combined with the appropriate SIL value using the following
equation:
- _ ___ _ . . Modeled emission rate from hypothetical source
Eq 1. MERP = appropriate SIL value X
Modeled air quality impact from hypothetical source
MERPs can be derived using any air quality threshold of concern ("critical air quality threshold")
and are not necessarily dependent on SILs. In practice, MERPs are intended to be used with SILs
as analytical tools for PSD air quality analyses. For PM2.5, the modeled air quality impact of an
increase in precursor emissions from the hypothetical source is expressed in units of |-ig/m3. For
O3, the modeled air quality impact is expressed in ppb.
As stated in the preamble to the 2017 final revisions to the Guideline (U.S. Environmental
Protection Agency, 2017a), the EPA believes that use of photochemical models for the purpose
of developing MERPs is scientifically appropriate and practical to implement. In this guidance
6
-------
document, EPA presents existing and new photochemical modeling of hypothetical single
source impacts on downwind O3 and secondary PM2.5. This modeling was configured, applied,
and post-processed consistent with EPA single source modeling guidance (U.S. Environmental
Protection Agency, 2016a). The locations of hypothetical sources included here are shown in
Figure ES-2. The single source impacts detailed in this section are collected from various past
and more recent photochemical grid model-based assessments. More than 100 locations were
modeled with hypothetical source emissions and are presented here.
Figure ES-2. Hypothetical sources modeled for downwind secondary air quality impacts
included in this assessment.
O
O
o
LO
~~
o
o
o
LO
o
o
~ New work presented here
O Baker et al, 2016
v Kelly etal, 2015
A IWAQM3-NFI Report
O
-2000
-1000
0
1000
2000
The relationships shown here for these hypothetical sources are not intended to provide an
exhaustive representation of all combinations of source type, chemical, and physical source
environments but rather to provide insightful information about secondary pollutant impacts
from hypothetical single sources in different parts of the U.S. Based on these annual
photochemical model simulations, the maximum impacts for daily PM2.5, annual PM2.5 and daily
maximum 8-hr average O3 are provided for each modeled source described in Appendix Table
A-l in an Excel spreadsheet on EPA's Support Center for Regulatory Atmospheric Modeling
(SCRAM) website. It is expected that the information in the Excel spreadsheet will be updated
over time as newer modeling is done consistent with EPA's single source modeling guidance
(U.S. Environmental Protection Agency, 2016a).
7
-------
Based on these photochemical modeling data, EPA recommends that the permit applicant in
consultation with the appropriate reviewing authority follow a three-step process:
1) Identify a representative hypothetical source (or group of sources for an area) from
EPA's modeling results (as described in Section 3.2.1).
¦S If a representative hypothetical source is not available, then consider whether
any of these derived MERP values available for the geographic location of the
project source may be appropriate to use. Alternatively, one can consider
conducting photochemical modeling (as described in Section 3.2.2) to derive a
source- or area-specific value.
2) Acquire the source characteristics and associated modeling results for the hypothetical
source(s).
3) Apply the source characteristics and photochemical modeling results from Step 2 above
with the appropriate SIL to the MERP equation for comparison with the project emission
rate.
Section 4 provides details on the use of MERPs for PSD compliance demonstrations for: 1)
source impact analysis, 2) PM2.5 increment analysis, and 3) cumulative impact analysis. It also
provides illustrative examples that show how existing EPA hypothetical source modeling can be
used to support a Tier 1 demonstration.
For PM2.5, based on EPA modeling presented here and recommended PM2.5 SILs, the illustrative
MERPs for NOx as a precursor to daily PM2.5 range from 1,073 tons per year (tpy) to over
100,000 tpy, while the illustrative MERPs for sulfur dioxide (SO2) as a precursor to daily PM2.5
range from 188 tpy to over 27,000 tpy. The illustrative MERPs for NOx as a precursor to annual
PM2.5 range from 3,182 tpy to over 700,000 tpy, while the illustrative MERPs for SO2 to annual
PM2.5 range from 859 tpy to over 100,000 tpy. For this assessment, the illustrative MERPs are
generally lower for SO2 than NOx reflecting that SO2 tends to form PM2.5 more efficiently than
NOx.
For O3, based on EPA modeling presented here and recommended O3 SIL, the illustrative MERPs
for NOx as a precursor to daily maximum 8-hr O3 range from 125 tpy to over 5,000 tpy, while
the illustrative MERPs for VOC as a precursor to daily maximum 8-hr O3 range from 1,049 tpy to
over 140,000 tpy. For this assessment, illustrative MERPs for NOx tend to be lower than VOC
which suggests most areas included in this assessment are more often NOx limited rather than
VOC limited in terms of O3 formation.
8
-------
1. Bad
EPA finalized revisions to the Guideline on Air Quality Models (the "Guideline," published as
Appendix W to 40 CFR part 51) that recommend a two-tiered approach for addressing single-
source impacts on ozone (O3) and secondary particulate matter less than 2.5 microns in
diameter (PM2.5) (U.S. Environmental Protection Agency, 2017a). The first tier (or Tier 1)
involves use of appropriate and technically credible relationships between emissions and
ambient impacts developed from existing modeling studies deemed sufficient for evaluating a
project source's impacts. The second tier (or Tier 2) involves more sophisticated case-specific
application of chemical transport modeling (e.g., with an Eulerian grid or Lagrangian model).
This guidance document is intended to provide a detailed framework that applicants may
choose to apply, in consultation with the appropriate permitting authority, to estimate single-
source impacts on secondary pollutants under the first-tier approach put forth in the Guideline
(i.e., Sections 5.3.2.b and 5.4.2.b).
For Tier 1 assessments, EPA generally expects that applicants would use existing empirical
relationships between precursors and secondary impacts based on modeling systems (e.g.,
chemical transport models) appropriate for this purpose. The use of existing credible technical
information that appropriately characterizes the emissions to air quality relationships will need
to be determined on a case-by-case basis. Existing credible air quality modeling would generally
include single source modeling based on an approved State Implementation Plan (SIP)
demonstration, a more recent submitted but not yet approved SIP demonstration, or modeling
not used to support a SIP demonstration but considered representative of the current air
quality in the area and of sufficient quality that is comparable to a model platform supporting a
SIP demonstration. The applicant should describe how the existing modeling reflects the
formation of O3 or PM2.5 in that geographic area. Information that could be used to describe the
comparability of two different geographic areas include average and peak temperatures,
humidity, terrain, rural or urban nature of the area, nearby local and regional sources of
pollutants and their emissions (e.g., other industry, mobile, biogenic), and ambient
concentrations of relevant pollutants where available.
As EPA introduced in the preamble to the 2015 proposed revisions to the Guideline, Modeled
Emission Rates for Precursors (MERPs) can be viewed as a type of Tier 1 demonstration tool
under the Prevention of Significant Deterioration (PSD) permitting program that provides a
simple way to relate maximum downwind impacts with a critical air quality threshold (e.g., a
significant impact level or SIL) (U.S. Environmental Protection Agency, 2018). EPA had initially
planned to establish generally applicable MERPs through a future rulemaking. However, after
further consideration, EPA believes it is preferable for permit applicants and permitting
authorities to consider site-specific conditions when deriving MERPs and to allow for the
development and application of locally and regionally appropriate values in the permitting
process. Thus, instead of deriving generally-applicable MERP values, the EPA is providing this
guidance document for consideration and use by permitting authorities and permit applicants
on a permit specific basis.
9
-------
This guidance is relevant for the PSD program and focuses on assessing the ambient impacts of
precursors of PM2.5 and O3 for purposes of that program. The MERP framework may be used to
describe an emission rate of an individual precursor that is expected to result in a change in the
level of ambient O3 or PM2.5, as applicable, that would be less than a specific air quality
threshold for O3 or PM2.5 that a permitting authority adopts and chooses to use in determining
whether a projected impact causes or contributes to a violation of the NAAQS for O3 or PM2.5,
such as the SILs recommended by EPA. In the context of the PSD program, precursors to O3
include volatile organic compounds (VOC) and nitrogen oxides (NOx) and precursors to PM2.5
generally include sulfur dioxide (SO2) and NOx. MERPs relate emissions of a specific precursor of
O3 or PM2.5 to ambient impacts of O3 or PM2.5 and do not provide a single demonstration for all
NAAQS pollutants.
If approved by the permitting authority as a PM2.5 Tier 1 demonstration tool for a PSD source in
a PM2.5 attainment or unclassifiable area, a finding that projected increases in the PM2.5
precursor emissions of NOx and/or SO2 from a project are below the respective MERPs may be
part of a sufficient demonstration that the project will not cause or contribute to violation of
the applicable NAAQS (hereafter "demonstration of compliance" or "compliance
demonstration"). Similarly, for the O3 NAAQS, an appropriate Tier 1 demonstration may include
a finding that the projected increases in O3 precursor emissions of NOx and/or VOC are below
the respective MERPs.
For situations where project sources are required to assess multiple precursors of PM2.5 or of
O3, EPA recommends that the impacts of multiple precursors should be estimated in a
combined manner for comparison to the appropriate SIL such that the sum of precursor
impacts would be lower than the SIL in a demonstration of compliance. Examples of combining
precursor impacts are provided in Section 4 of this document. Further, where project sources
are required to assess both primary PM2.5 and precursors of secondary PM2.5, EPA recommends
that applicants combine the primary and secondary impacts to determine total PM2.5 impacts as
part of the PSD compliance demonstration. An example of combining primary and secondary
impacts is provided in Section 4 of this document.
The purpose of this document is to provide a framework for using air quality modeling to
develop relationships between precursors and maximum downwind impacts for the purposes
of developing and using MERPs as a Tier 1 demonstration tool. We provide hypothetical single
source impacts on O3 and secondary PM2.5 to illustrate how this framework can be
implemented by permit applicants. The relationships presented here in some cases may
provide relevant technical information to assist or inform an applicant in providing a first-tier
demonstration for their specific permit situation and as a template for stakeholders and/or
state or local agencies to develop information relevant to a specific area or source type. Based
on the EPA modeling conducted to inform these illustrative MERPs provided here, such values
will vary across the nation reflecting different sensitivities of an area's air quality level to
changes in levels of precursor emissions thereby providing an appropriate technical basis for
evaluating the impacts of these precursors to PM2.5 and O3 formation because they reflect the
10
-------
regional or local atmospheric conditions for particular situations.
This document is not a final agency action and does not reflect a final determination by the EPA
that any particular proposed source with emissions below an illustrative MERP value developed
by EPA (or a MERP developed by another party using methods recommended by EPA) will not
cause or contribute to a violation of an O3 or PM2.5 NAAQS or PM2.5 PSD increments. A
determination that a proposed source does not cause or contribute to a violation can only be
made by a permitting authority on a permit-specific basis after consideration of the permit
record. The illustrative MERP values identified by the EPA have no practical effect unless and
until permitting authorities decide to use those values in particular permitting actions. This
guidance document does not require the use, nor does it require acceptance of the use, of this
framework or any result using this framework by a permit applicant or a permitting authority.
Permit applicants and permitting authorities retain the discretion to use other methods to
complete a first-tier assessment under Sections 5.3.2.b and 5.4.2.b of the Guideline and to
require additional information from a permit applicant to make the required air quality impact
demonstration. This guidance document does not create any binding requirements on EPA,
permitting authorities, permit applicants, or the public.
Subsequent sections of this document include information about O3 and secondary PM2.5
formation in the atmosphere, a conceptual description of MERPs, information about developing
MERPs using photochemical modeling, using MERPs for individual permit demonstrations, and
several illustrative examples of using MERPs to support hypothetical permit applications.
11
-------
. Vi 1r" wan! *\\ IMMi i, T ;niifi «iikii iii ilir -'nifit: 4«llk-ik
A conceptual understanding of an area's emissions sources and which precursor emissions limit
the formation of secondary pollutants such as O3 and PM2.5 is useful for interpreting modeled
and ambient impacts due to changes in emissions in that area. The formation regime favoring a
particular precursor may vary seasonally, day to day, and by hour of the day. It is important to
understand how the atmosphere will respond to changes in emissions to make informed
decisions about how changes in emissions from a source might impact ambient pollutant levels.
Typically, reductions in emissions of primary pollutants or precursors of secondary pollutants
result in some level of reduction in ambient pollutant concentrations.
Secondary PM2.5 and O3 are closely related to each other in that they share common sources of
emissions and are formed in the atmosphere from chemical reactions with similar precursors
(U.S. Environmental Protection Agency, 2017a). Air pollutants formed through chemical
reactions in the atmosphere are referred to as secondary pollutants. For example, ground-level
O3 is predominantly a secondary pollutant formed through photochemical reactions driven by
emissions of NOx and VOCs in the presence of sunlight. O3 formation is a complicated nonlinear
process that depends on meteorological conditions in addition to VOC and NOx concentrations
(Seinfeld and Pandis, 2012). Warm temperatures, clear skies (abundant levels of solar
radiation), and stagnant air masses (low wind speeds) increase 03 formation potential (Seinfeld
and Pandis, 2012).
O3 Formation
O3 formation may be limited by either NOx or VOC emissions depending on the meteorological
conditions and the relative mix of these pollutants. When O3 concentrations increase (decrease)
because of increases (decreases) in NOx emissions, the O3 formation regime is termed "NOx
limited." Alternatively, the O3 formation regime is termed "VOC limited" when ambient ozone
concentrations are very sensitive to changes in ambient VOC. The VOC-limited regime is
sometimes referred to as "radical-limited" or "oxidant-limited" because reactions involving
VOCs produce peroxy radicals that can lead to O3 formation by converting nitric oxide (NO) to
nitrogen dioxide (NO2) in the presence of sunlight. In a NOx-limited regime, ozone decreases
with decreasing NOx and has very little response to changes in VOC. The NOx-limited formation
regime is more common in rural areas of the U.S. where high levels of biogenic VOC exist and
relatively few man-made, or anthropogenic, NOx emissions occur. O3 decreases with decreasing
VOC in a VOC-limited formation regime. The O3 formation regime for some urban areas in the
U.S. is locally VOC-limited during daytime hours due to large NOx emissions from mobile and
industrial sources and relatively smaller amount of biogenic and anthropogenic VOC emissions.
Additional information on O3 formation regimes based on modeling (U.S. Environmental
Protection Agency, 2017b) and satellites (Chang et al., 2016; Duncan et al., 2010; Jin et al.,
2017) are available elsewhere. An example is shown in Figure 2-1.
12
-------
Figure 2-1. The ratio of the change in monthly peak daily maximum 8-hr (MDA8) O3 from the
50% reduction in NOx to the change in monthly peak MDA8 O3 from a 50% reduction in VOC.
Note: Ratios greater than one (shown in purple) indicate that ozone was reduced more effectively by similar percentage
reductions in NOx emissions than reductions in VOC emissions. Ratios less than one (shown in green) indicate that ozone was
reduced more effectively by similar percentage reductions in VOC emissions than reductions in NOx emissions.
Source: https://www.epa.gov/sites/production/files/2017-05/documents/national_modeling.advance.may_2017.pdf
(Max MDA8 03 change: NOx)/(Max MDA8 03 change: VOC)
July 2 OIL
188 "
5.000
4.000
3.000 ¦
2.000 "
1.000
0.750
0.500
0.250
0.000
1
PIVb.s Formation
In the case of PM2.5, or fine PM, total mass is often categorized into two groups: primary (i.e.,
emitted directly as PIVh.sfrom sources) and secondary (i.e., PM2.5formed in the atmosphere by
precursor emissions from sources). The ratio of primary to secondary PM2.5 varies by location
and season. In the U.S., PM2.5 is dominated by a variety of chemical components: sulfate,
nitrate, ammonium, organic carbon (OC), elemental carbon (EC), crustal elements, sea-spray
constituents, and oxidized metals. PM2.5 EC, crustal elements, and sea spray are directly
emitted into the atmosphere from primary sources. PM2.5 OC is directly emitted from primary
sources but is also formed secondarily in the atmosphere by reactions involving VOCs. PM2.5
sulfate, nitrate, and ammonium are predominantly the result of chemical reactions of the
oxidized products of SO2 and NOx emissions and direct NH3 emissions (Seinfeld and Pandis,
2012). Figure 2-2 shows the average composition by season (spring, summer, fall and winter)
for PM2.5 data collected during 2013-15. In the eastern United States, sulfate is high in the
spring (March-May) and summer (July-September). Nitrate is most evident in the Midwest and
western cities and highest during the winter. Organic mass (OM) is a large component
throughout the year.
13
-------
Figure 2-2. Average composition by season for PM2.5 data collected during 2013-15.
Note: Quarter 1 (top left), quarter 2 (top right), quarter 3 (bottom left), and quarter 4 (bottom right).
V
/
L__
\{ I
-/ V
Ml-, C
_j
\
nitrate
• 6 pg rrv3
0 12 |jg m
IS (jg m
• 6 pg rrr3
0 12 |jg m
18 |jg rrr
sulfate
sulfate
I sea-salt
crustal
, sea-salt
nitrate
¦crustal
V--'
EC
OC
EC
OC
• 6 pg m 3
0 12 pg m
IS |jg m
sulfate
nitrate
EC
• 6 (jg m 3
^ 12 [ig m
18 |jg m
Sulfur dioxide emissions are oxidized in the atmosphere and form sulfuric acid, which has a very
low vapor pressure and tends to exist in the particulate phase. Particulate sulfuric acid reacts
with NH3 to form ammonium bisulfate and ammonium sulfate. Aqueous phase reactions are
also an important pathway for particulate sulfate formation. SO2 dissolves into cloud and fog
droplets and is oxidized to sulfate via reaction pathways involving hydrogen peroxide, O3, and
other oxidants. Since sulfate is essentially non-volatile under atmospheric conditions, sulfate
formed in clouds persists as particulate sulfate after the cloud evaporates. Sulfur dioxide
emission reductions lead to reductions in particulate sulfate. The process is not completely
linear, especially when aqueous phase production is significant, and so changes in SO2
emissions may not result in the same proportion of change in PM2.5 sulfate concentration.
Emissions of NOx are chemically transformed to nitric acid (HNO3) through gas-phase and
heterogeneous reactions. Nitric acid may condense onto particles to form particulate nitrate
depending on the conditions. Condensation of HNO3 onto particles is favored by low
temperature, high relative humidity, and relatively less acidic conditions associated with high
levels of NH3 and particulate cations. HNO3 formation may be oxidant or NOx-limited, and PM2.5
ammonium nitrate formation may be limited by the availability of either nitric acid or NH3 or by
14
-------
meteorological conditions. When PM2.5 ammonium nitrate is limited by the availability of NH3,
the formation regime is termed "ammonia-limited," and the formation regime is termed "nitric
acid-limited" when the opposite situation exists (Stockwell et al., 2000). In general, a decrease
in NOx emissions will result in a decrease in PM2.5 nitrate concentration (Pun et al., 2007). Since
PM2.5 ammonium nitrate formation is preferred under low temperature and high relative
humidity conditions and in the presence of NH3, ammonium nitrate concentrations tend to be
greater during colder months and in areas with significant NH3 emissions. NOx emission
changes during warm temperatures may result in less change in ambient PM2.5 compared to
cold months due to HNO3 staying in the gas rather than particle phase due to higher
temperatures. Additionally, NOx emission changes in places with very little or no ambient
ammonia may result in little change in ambient PM2.5 ammonium nitrate.
15
-------
••• Fi iiihnvil icI II vwlkpiii ¦ h'llFJNI , « r« IVI I iMriniMrnK.h
A Tier 1 demonstration tool as described in the Guideline consists of technically credible air
quality modeling done to relate precursor emissions and peak secondary pollutant impacts
from specific or hypothetical sources (U.S. Environmental Protection Agency, 2017a). With
appropriate supporting information, permit applicants may use existing appropriate air quality
modeling as part of an assessment of air quality impacts from a proposed new or modified
source under the PSD permitting program. Permit applicants should provide a narrative
explanation describing how project source emissions relate to the information provided as part
of their Tier 1 demonstration. It should be made clear how the chemical and physical
environments modeled as part of an existing set of information included in their Tier 1
demonstration are relevant to the geographic area of the project and key receptors.
As detailed below, this framework for developing MERPs focuses on use of photochemical
modeling to relate the modeled air quality impacts and a critical air quality threshold (e.g.,
appropriate SIL value) to estimate a MERP for comparison with the project source emissions.
However, a similar screening approach would be to adjust the modeled air quality impacts
based on the relationship between the modeled and project source emissions to then compare
the resulting air quality impact with the appropriate SIL.
Existing credible air quality modeling generally may include single source modeling based on an
approved SIP demonstration, a more recent submitted but not approved SIP demonstration, or
modeling not used to support a SIP demonstration but considered representative of the current
air quality in the area and of sufficient quality that is comparable to a model platform
supporting a SIP demonstration. The specifications for single source demonstration model
platforms (e.g., horizontal grid spacing, vertical resolution, non-project source emission
treatment, etc.) are detailed in the 2016 EPA guidance document "Guidance on the use of
models for assessing the impacts of emissions from single sources on the secondarily formed
pollutants O3 and PM2.5" (U.S. Environmental Protection Agency, 2016a).
Figure 3-1 illustrates the EPA's framework for MERPs as a Tier 1 demonstration tool. This
framework is intended to show how the elements and concepts described in this document
relate to each other and where more information is provided in this document about each step
of the process. This flow diagram shows how MERPs can be developed from either existing EPA
modeling or another source of data and how that information can be credibly used for a source
impact analysis and, if necessary, a cumulative impact analysis. In this framework, the source
impact analysis for the PM2.5 NAAQS may also satisfy Class II PSD increment since the
recommended EPA SILs are the same.
16
-------
Figure 3-1. EPA's framework for MERPs as a Tier 1 Demonstration Tool.
Define concept of MERPs (Section 3.1)
I ~
Development of MERPs with photochemical modeling
(Section 3.2}
Permit applicant
EPA modeling of provided or use of
hypothetical sources other appropriate
(Section 3.2.1) existing modeling
(section 3.2.2)
i 1
Use of MERPS for individual permits (Section 4)
r i
Source Impact Analysis for
NAAQS (Section 4.1.1)
*same as Class II PSD increment demonstration
| If project impacts > SIL
Cumulative Impact Analysis
for NAAQS (Section 4.1.3)
3.1, Definition of MERPs as a Tier 1 Demonstration Tool
Properly-supported MERPs provide a simple way to relate modeled downwind impacts with an
air quality threshold that is used to determine if such an impact causes or contributes to a
violation of the appropriate NAAQS. In the discussion that follows and in reported results in
computing MERP values, we use the EPA's recommended SIL values for O3 and PM2.5 as the
relevant air quality threshold (U.S. Environmental Protection Agency, 2018). Consistent with
EPA's SILs guidance, to the extent a permitting authority elects to use a SIL to help quantify a
level of impact that does not cause or contribute to a violation of the O3 and/or PM2.5 NAAQS or
PM2.5 PSD increment(s), such values will need to be justified on a case-by-case basis. To derive a
MERP value for the purposes of a PSD compliance demonstration, the model predicted
relationship between precursor emissions from hypothetical sources and their downwind
modeled impacts can be combined with the appropriate SIL value using the following equation:
Source Impact Analysis for
Class I PSD increment
(Section 4.1.2)
17
-------
- - . Modeled emission rate (tpy) from hypothetical source
Eq. 1 MERP = appropriate SIL value X
Modeled air quality impact from hypothetical source
For PM2.5, the modeled air quality impact of an increase in precursor emissions from the
hypothetical source is expressed in units of |-ig/m3. For O3, the modeled air quality impact is
expressed in ppb. As discussed in Section 4, these modeled impacts would reflect the maximum
downwind impacts for PM2.5 and O3. The SIL value is expressed as a concentration for PM2.5 (in
l-ig/m3) and mixing ratio for O3 (in ppb). Consistent with the air quality model application used
here to predict a change in pollutant concentration, MERPs are expressed as an annual
emissions rate (in this case as tons per year).
I»e 4"i 1 Ik lie i it Mil M kliis. 11 h I! 11 v I- -clno 1 ifii' «l Mi-- Idiii,'
As stated in the preamble to the 2017 revisions to the Guideline (U.S. Environmental Protection
Agency, 2017a), the EPA believes that use of photochemical models for estimating single source
secondary pollutant impacts is scientifically appropriate and practical to implement. Publicly
available and fully documented Eulerian photochemical grid models such as the Comprehensive
Air Quality Model with Extensions (CAMx) (Ramboll ENVIRON, 2016) and the Community
Multiscale Air Quality (CMAQ) (Byun and Schere, 2006) model treat emissions, chemical
transformation, transport, and deposition using time and space variant meteorology. These
modeling systems simulate primarily emitted species and secondarily formed pollutants such as
O3 and PM2.5 (Chen et al., 2014; Civerolo et al., 2010; Russell, 2008; Tesche et al., 2006). Even
though single source emissions are injected into a grid volume, photochemical transport
models have been shown to adequately capture single source impacts when compared with
downwind in-plume measurements (Baker and Kelly, 2014; Baker and Woody, 2017; Zhou et al.,
2012). Where set up appropriately for the purposes of assessing the air quality impact of single
sources to ambient levels of primary and secondarily formed pollutants, photochemical grid
models could be used with a variety of approaches to estimate these impacts. These
approaches generally fall into the categories of source sensitivity (how air quality changes due
to changes in emissions) and source apportionment (what air quality impacts are related to
certain emissions).
The simplest source sensitivity approach, commonly referred to as a brute-force change to
emissions, would be to simulate two sets of conditions, one with all emission sources and a
subsequent simulation with all emission sources and the post-construction characteristics of
the new source or modification being the only difference from the original baseline simulation
(Cohan and Napelenok, 2011). The difference between these model simulations provides an
estimate of the air quality change related to the change in emissions from the project source. In
addition to the brute force approach, some photochemical models have been "instrumented"
with techniques that allow tracking of air quality impacts from the emissions of a particular
sector or source. One sensitivity approach is the decoupled direct method (DDM), which tracks
the sensitivity of an emission source through all chemical and physical processes in the
modeling system (Dunker et al., 2002). Sensitivity coefficients relating source emissions to air
quality are estimated during the model simulation and output at the resolution of the host
18
-------
model. Unlike the brute force approach, a second simulation is not necessary when using DDM,
although additional resources are required as part of the initial baseline simulation when DDM
is applied.
Some photochemical models have been instrumented with source apportionment capabilities
which tracks emissions from specific sources through chemical transformation, transport, and
deposition processes to estimate source-specific impacts to predicted air quality at downwind
receptors (Kwok et al., 2015; Kwok et al., 2013). Source apportionment has been used to
differentiate the air quality impact from single sources on model predicted O3 and PM2.5 (Baker
and Foley, 2011; Baker and Kelly, 2014; Baker and Woody, 2017). DDM has also been used to
estimate O3 and PM2.5 impacts from specific sources (Baker and Kelly, 2014; Bergin et al., 2008;
Kelly et al., 2015) as well as the simpler brute-force sensitivity approach (Baker and Kelly, 2014;
Bergin et al., 2008; Kelly et al., 2015; Zhou et al., 2012). Limited comparison of single source
impacts between models (Baker et al., 2013) and approaches to differentiate single source
impacts (Baker and Kelly, 2014; Kelly et al., 2015) show generally similar downwind spatial
gradients and impacts.
Near-source in-plume aircraft based measurement field studies provide an opportunity to
evaluate model estimates of (near-source) downwind transport and chemical impacts from
single stationary point sources (ENVIRON, 2012b). Photochemical grid model source
apportionment and source sensitivity simulation of single-source downwind impacts compare
well against field study primary and secondary ambient in-plume measurements (Baker and
Kelly, 2014; Baker and Woody, 2017; ENVIRON, 2012b). This work indicates photochemical grid
models using source apportionment or source sensitivity approaches provide meaningful
estimates of single source impacts.
! II II ' lii.'k' iHIKt I! Ilk-r. oiliHInl' 4|| IK'iefii-llllli - icil 1 I, »|| 1 StH 1 Ilk1! !l
II
This section presents a summary of EPA photochemical modeling of hypothetical single source
impacts on downwind O3 and secondary PM2.5. The locations of hypothetical sources modeled
are shown in Figure 3-2. A total of 113 locations were modeled. The single source impacts
detailed in this section were collected from various past and recent photochemical grid model-
based assessments. The resulting relationships were based on photochemical modeling studies
that estimated single source impacts in California (Kelly et al., 2015), the Detroit and Atlanta
urban areas (U.S. Environmental Protection Agency, 2016b), and at rural and suburban
locations in the central and eastern United States (Baker et al., 2016). Additional photochemical
modeling was conducted by EPA consistent with the approach described in Baker et al., 2016
for hypothetical sources in the western, central, and eastern U.S. to provide broader geographic
coverage across the nation.
19
-------
Figure 3-2. Location of hypothetical sources modeled for downwind secondary air quality
impacts included in EPA's assessment.
O
o
in
8 J
O
o
o
If)
o
o
o _
un
o
o
o
~ New work presented here
O Baker et al, 2016
v Kelly etal, 2015
A IWAQM3-NFI Report
-2000
-1000
0
1000
2000
Atlanta and Detroit both include a single hypothetical source modeled at 4 km horizontal grid
resolution for an entire year. The California sources were also modeled at 4 km but only include
a sub-set of an entire year meaning the maximum impact from those hypothetical sources may
not be realized as part of that study design. The western, central, and eastern U.S. sources were
modeled at 12 km horizontal grid resolution for the entire year of 2011. It is possible that the
maximum impacts from each of these hypothetical sources may not have been realized using a
single year of meteorology and that another year with more conducive meteorology for
secondary formation of O3 and/or PM2.5 might be more appropriate and result in greater
downwind impact. As shown, we define the following source types throughout the continental
U.S. that reflect different release heights and multiple emissions rates:
• Source release type "L" refers to sources modeled with surface level emissions releases:
stack height of 10 m, stack diameter of 5 m, exit temperature of 311 K, exit velocity of 27
m/s, and flow rate of 537 m3/s.
• Source release type "H" refers to sources modeled with tall stack emissions releases: stack
height of 90 m, stack diameter of 5 m, exit temperature of 311 K, exit velocity of 27 m/s,
and flow rate of 537 m3/s.
Hypothetical sources for this assessment include impacts based on multiple emission rates and
emitted with a near-surface release or tall stack. Information about each hypothetical source
modeled is provided in Appendix A.
20
-------
The relationships shown here for these hypothetical sources are not intended to provide an
exhaustive representation of all combinations of source type, chemical, and physical source
environments but rather to provide insightful information about secondary pollutant impacts
from single sources in different parts of the U.S. The maximum impacts for daily PM2.5, annual
PM2.5 and daily maximum 8-hr average O3 are shown in the following sub-sections for the
hypothetical sources modeled for an entire year and do not include sources modeled for an
episode.
Tables showing the maximum impacts for sources modeled with annual simulations are
provided in an Excel spreadsheet on EPA's SCRAM website. Impacts for each source include the
maximum daily PM2.5 impacts, maximum annual PM2.5 impacts, and maximum daily 8-hr O3
impacts over annual simulations. Emissions are shown in tpy and release height in meters. VOC
speciation used for these assessments is shown in Table 3-1. More information about these
hypothetical sources and how the model output was processed to generate maximum impacts
are described in more detail in (Baker et al., 2016).
Table 3-1. Assumed VOC speciation for hypothetical sources presented
nere.
Carbon bond specie
Fraction
Carbon bond specie
Fraction
ALD2
0.0152
MEOH
0.0054
ALDX
0.0155
NVOL
0.0008
ETH
0.0324
OLE
0.1143
ETHA
0.0094
PAR
0.4057
ETOH
0.0090
TERP
0.0170
FORM
0.0757
TOL
0.1148
IOLE
0.0088
UNR
0.1080
ISOP
0.0007
XYL
0.0674
Additional information has been provided for each source to facilitate qualitative comparison
between hypothetical sources with project sources. The additional information includes the
terrain within 50 km of the source and maximum grid cell percent urban landcover within 50
km of the source to provide some additional information about nearby orography and whether
the source is in proximity to population centers. This additional information is illustrated in
Figure 3-3.
The spreadsheet also includes the climate zone where the source is located as shown in Figure
3-4. These regional classifications are used to aggregate impacts in summarizing modeling
results in subsequent sections.
21
-------
Figure 3-3. Maximum terrain height (top) and fractional urban coverage (bottom) within 50
km of each of the hypothetical sources modeled.
o
o
0 o
o*
o
o
o
o
o
J 0
n ° ° n
° 0 0 0 %
O I
. y
• o
o
0
• o
O o
o o
o w
O o
o o
•o
oo
o
o
qJ
o
- 2000
- 1500
0
o,
o
° o
7 I?
OO""
o
/ o
o
o
o
o
o
o
/ °
/ °
0 0
o
o o
-§±9
o
o
r ch
O o
Oo
o
o° ° O °o
o
o
° O To W°:
oo •
o o
o o
> jMkJ
/«™
o
o o
o o
B
. _
I
100
80
- 60
- 40
20
0
22
-------
Figure 3-4. NOAA climate zone map with number of hypothetical source locations
modeled in each climate zone.
Source: https://www.ncdc.noaa.gov/monitoring-references/maps/us-climate-regions.php
Climate Zone
Sources
Northeast
10
Southeast
9
Ohio Vally
19
Upper Midwest
12
Rockies/Plains
14
South
17
Southwest
15
West
6
Northwest
3
3.2.1.1. EPA Modeled Impacts: Annual and Daily PM2.5
The maximum daily average PM2 5 sulfate ion from SO? emissions and maximum daily average
PM2.5 nitrate ion from NOx emissions are shown in Figure 3-5 by emission rate and area.
Downwind maximum PM2.5 impacts generally increase as rates of precursor emissions increase.
However, differences in chemical (e.g. NOx/VOC ratio, NH3 concentrations) and physical (e.g.,
terrain and meteorology) regimes among these hypothetical sources result in differences in
downwind impacts even for similar types of sources. Differences in maximum impacts can also
be seen between the different areas and studies. One such example is described in Section
3.2.1.3 of this document.
23
-------
Figure 3-5. Maximum daily average PM2.5 nitrate ion impacts from NOx emissions and
PM2.5 sulfate ion impacts from SO2 emissions.
Note: These impacts are from multiple modeling studies estimating downwind impact from hypothetical sources. The
distribution shown for each climate zone represents multiple emission rates.
24-hr PM2.5 - NOX precursor
All Sources
cn
3.
Q
<3
c
Q
O
T
-r
° o
O 8
T
-r
1-
T
NE SE OV UM RP S SW W
MOM Climate Zone
01
=),
s
c
O
O
Emission Rate (tpy)
24-hr PM2.S - SO2 precursor
All Sources
e
61
c
S
2
c
s
c
o
a
1—1 1 1—1 1 1—r
NE SE OV UM RP S SW W
a>
c
.2
c
s
c
o
O
8
The distance from the source of maximum daily and annual average secondary PM2.5 impact is
shown in Figure 3-6. Peak impacts tend to be in close proximity to the source. For NOx
precursor, the peak 24-hour PM2.5 impacts are typically within 20 to 50 kilometers, while peak
annual average PM2.5 impacts are typically within 20 kilometers of the source. For SO2
precursor, the peak 24-hour PM2.5 impacts are shown to be mostly within 10 to 40 kilometers,
while peak annual average PM2.5 impacts are largely within 20 kilometers. These peak impacts
become less common as distance from the source increases. Figure 3-7 shows maximum annual
average impacts from SO2 emissions on modeled PM2.5 sulfate ion and NOx emissions on
modeled PM2.5 nitrate ion. Downwind impacts tend to increase as emissions of precursors
increase. Also, impacts vary from area to area.
24
-------
Figure 3-6. Maximum daily and annual average secondary PM2.5 nitrate ion impacts from NOx
emissions and PM2.5 sulfate ion impacts from SO2 emissions shown by distance from the
source.
24-hr PM2.5 - NOX precursor 24-hr PM2.5 - S02 precursor
m
o
o
o
V § w
o | T
tr1!? 1 8 1 ° 0 '
0
i y m 8 n
1,,.. 1.
1111111111111111111
6 20 40 60 80 110 160 210
Distance from the source (km)
E
c
o
c
0
o
c
o
o
I ~! ? T
e D
6 20 40 60 100 120
Distance from the source (km)
Annual PM2.5 - NOX precursor
Annual PM2.5 - S02 precursor
O
O
O
O
6 20 40 70 90 110 150
Distance from the source (km)
E
C
o
c
0
O
c
o
O
o
OJ
o
o
o
LO
o
d
o
o
o
6 10 20 30 40 50
Distance from the source (km)
The tendency for secondary PM2.5 to be larger near the source is important when considering
how to use impact estimates to inform different types of permit demonstrations. For NAAQS
demonstrations, peak impacts tend to be near the source. Class I impacts are likely to be
further downwind of the project source, so a near-source impact estimate would typically not
be as relevant.
25
-------
Figure 3-7. Maximum annual average secondary PM2.5 nitrate ion impacts from NOx
emissions and PM2.5 sulfate ion impacts from SO2 emissions.
Note: These impacts are from multiple modeling studies estimating downwind impact from hypothetical sources. The
distribution shown for each climate zone represents multiple emission rates.
Annual PM2.5 - NOX precursor
All Sources
0T
E
3.
c
.Q
2
E
8
c
o
U
S
o
8
8 +
1 r
—
1-
-r
—
0
1-
T
NE SE OV UM RP S 5W W
NOAA Climate Zone
0T
E
j±
c
.S
2
E
S
c
o
O
s
<6
500 1000 3000
Emission Rate (tpy)
Annual PM2.5 -S02 precursor
All Sources
—1—1—1—1—1—1—1—i—
NE SE OV UM RP S SW W
01
3.
c
Q
-------
physical (e.g., terrain and meteorology) regimes among these hypothetical sources result in
differences in downwind impacts even for similar types of sources.
Figure 3-8. Maximum 8-hr ozone impacts from NOx emissions and from VOC emissions.
Note: These impacts are from multiple modeling studies estimating downwind impact from hypothetical sources. The
distribution shown for each climate zone represents multiple emission rates.
8-hr Ozone - NOX precursor All Sources
T
CM
-i 1 1 1 1 1 1 r
NE SE OV UM RP S SW W
NOAA Climate Zone
a
DC
x
-
500 1000 3000
Emission Rate (tpy)
8-hr Ozone - VOC precursor
ft "
a
o
0
—I 1 1 1 1 1 1 r
NE SE OV UM RP S SW W
All Sources
JO
ft
oc
£
X
"1
CM
500
1000
3000
Each of the hypothetical source impacts modeled as part of EPA's assessment used a typical
industrial assumption for speciation of VOC emissions (see Table 3-1 for VOC speciation profile).
To better understand the influence of VOC speciation, as a sensitivity analysis, EPA modeled a
set of hypothetical sources with near-surface releases in the western and eastern U.S. with an
alternative VOC emissions speciation that assumed 100% of the VOC emissions were emitted as
formaldehyde to provide a more reactive profile than typically used. Figure 3-9 shows a
comparison of the downwind maximum daily 8-hr average O3 impacts using the typical VOC
profile compared with impacts where these same sources are modeled with formaldehyde-only
VOC emissions. For both sets of emissions scenarios, a total of 500 tpy of VOC was emitted, the
only difference being the VOC speciation. The formaldehyde only simulations for these sources
generally resulted in higher downwind O3 impacts than the simulations of hypothetical sources
27
-------
with VOC speciation shown in Table 3-1. The increases in impacts are typically between 1.5 and
2 times higher (Figure 3-9).
Since VOC reactivity can be important, some areas may want to develop separate VOC to O3
relationships using typical VOC profiles and VOC profiles that may be more reflective of certain
types of sources that exist in that area or are anticipated to operate in that area in the future.
Figure 3-9. Maximum 8-hr ozone impacts from 500 tpy of near-surface VOC emissions using a
typical industrial VOC speciation profile and assuming all VOC emissions are formaldehyde.
Note: these impacts are for the eastern and western U.S. hypothetical sources presented here and do not include
information from any other studies.
VOC & FORM to period peak max 8-hr ozone
o
o
Aq
c\j
d
o Eastern US
A Western US
o
o
0.0
0.2
0.4
0.6
0.8
VOC Emissions Impact (ppb)
The distance from the source of the maximum daily 8-hr average O3 impacts are shown in
Figure 3-10. Like maximum daily PM2.5 impacts, maximum daily 8-hr average O3 impacts tend to
be in close proximity to the source and are less frequent as distance from the source increases.
This is particularly notable where distance from the source exceeds 50 km.
28
-------
Figure 3-10. Maximum 8-hr ozone impacts from NOx emissions and from VOC emissions by
distance from the source.
Note: These impacts are from multiple modeling studies estimating downwind impact from hypothetical sources.
8-hr Ozone - NOX precursor
8-hr Ozone -
- VOC precursor
CM _
0
0
LO _
8
_Q
O _
0 8
.Q
CM
o
o
LL
Q.
/-1
00 -
I
LL
Q.
O
CM
o | o
CO
DC
O)
c
CD -
"sf -
0
01— -
001
u
"CO
cc
U)
c
1.0 1.5
I I
o
g *
§ 1 8
; ! o
X
CM -
O -
B 9 B n
A j_ A 5 i . i .
"x
LO
o
o _
o
« 2 .
O
1 1 1 1 1 1 1 1
6 10 20 30 50 60 80 110
i i i i i i
6 20 40
I I I I I I I
60 80 130 150
Distance from source (km)
Distance from source (km)
3.2.1.3. EPA Illustrative MERPs: Annual and Daily PM2.5
The hypothetical single source modeling presented here was used to develop illustrative MERPs
based on equation 1 and the EPA recommended SIL. Based on the EPA's photochemical
modeling results across all hypothetical sources presented above and detailed in Appendix A of
this document, Figure 3-11 shows NOx to annual maximum daily average PM2.5 nitrate ion and
SO2 to annual maximum daily average PM2.5 sulfate ion MERPs that illustrate the range of
potential values for these sources and time period. Neither PM2.5 sulfate nor PM2.5 nitrate was
assumed to be neutralized by ammonium. For this illustrative example, consistent with EPA's
SILs guidance (U.S. Environmental Protection Agency, 2018), the EPA recommended 24-hour
PM2.5 NAAQS SILs value of 1.2 |-ig/m3 was used to estimate daily average PM2.5 MERPs.
The illustrative MERPs for NOx to daily PM2.5 range from 1,073 tpy to over 100,000 tpy, while
the illustrative MERPs for SO2 to daily PM2.5 range from 188 tpy to over 27,000 tpy for the
hypothetical sources modeled and presented here based on the selected air quality threshold.
The variation from source to source is related to different chemical and meteorological
environments around the source that range in terms of conduciveness toward secondary PM2.5
formation.
Similarly, based on EPA's photochemical modeling results of hypothetical sources, Figure 3-12
shows NOx to maximum annual average PM2.5 nitrate ion and SO2 to maximum annual average
PM2.5 sulfate ion MERPs to illustrate the range of potential values for these sources and this
time period. Neither PM2.5 sulfate nor PM2.5 nitrate were assumed to be neutralized by
ammonium.
29
-------
Figure 3-11. NOx and SO2 daily average PM2.5 MERPs estimated from single source
hypothetical emissions impacts on PM2.5 nitrate ion and PM2.5 sulfate ion respectively.
Note: Daily PM2.5 MERPs derived here based on EPA recommended 24-hour PM2.5 NAAQS SIL value of 1.2 ng/m3 and neither
PM2.5 sulfate nor nitrate is assumed to be neutralized by ammonia.
24-hr PM2.5 MERPs - MOX precursor
All Soirees
3
S
£ 8
§
s
s
B
UM
NOAA Climate Zone
24-hr PM2.5 MERPS -SQ2 precursor
All Soirees
S -
S -
~~r-
NE
SE
*
-I-
~
1
•
1)
LJ
Is
' * * *
1
OV
1
UM
II
HP
4 *
iS=
—I 1 1"
s sw w
>1
O
For this illustrative example, consistent with EPA's SILs guidance, the EPA recommended annual
PM2.5 NAAQS SILs value of 0.2 |ag/m3 was used to estimate annual average PM2.5 MERPs. The
illustrative MERPs for NOx to annual PM2.5 range from 3,182 tpy to over 700,000 tpy, while the
illustrative MERPs for SO2 to annual PM2.5 range from 859 tpy to over 100,000 tpy for the
hypothetical sources presented here based on the selected air quality threshold. The variation
from source to source is related to different chemical and meteorological environments around
the source that range in terms of conduciveness toward secondary PM2.5 formation.
30
-------
Figure 3-12. NOx and SO2 annual average PM2.5 MERPS shown by geographic region.
Note: Annual PM2.5 MERPs derived here based on EPA recommended annual PM2.5 NAAQS SIL value of 0.2 pg/m3 and
neither PM2.5 sulfate nor nitrate is assumed to be neutralized by ammonia.
Annual PM2.5 MERPs - NOX precursor
All Sources
o
9
3
A
«
*
*
1
<
1
f
1
NC
1
sc
1
OV
1
UM
—1—
RP
—I—
sw
—r~
w
MOAA Climate Zone
Annual PM25 MERPs - SO2 piecursor All Sources
NC SC OV UW HP s SW W
As shown, the illustrative MERPs are generally lower for SO2 than NOx meaning that SO2 tends
to form PM2.5 more efficiently than NOx. This is consistent with the conceptual model of
secondary PM2.5 formation in many parts of the United States reflecting that the PM2.5 sulfate
ion has a lower vapor pressure than PM2.5 nitrate ion and tends to stay in the particulate phase
in a greater range of meteorological conditions.
The distribution of illustrative MERPs for both SO2 and NOx to daily PM2.5 are shown to vary
between regions of the United States. This is expected since the chemical (e.g., oxidants,
neutralizing agents) and physical (e.g., terrain) environments vary regionally in the United
States. Figure 3-13 shows the lowest MERP at each hypothetical source location for daily (left
panels) and annual (right panels) PM2.5 from SO2 (top panels) and NOx (bottom panels)
emissions. These plots show broad regional patterns in PM2.5 formation potential which are
generally related to regions with conducive meteorology, available neutralizing agents, and
other emission sources competing for these neutralizing agents.
31
-------
Figure 3-13. Lowest MERP value at each hypothetical source location for daily (left panels)
and annual (right panels) PM2.5 from SO2 (top panels) and NOx (bottom panels) emissions.
2000
1000
6000
4000
15000
10000
Daily PM25 from S02 emissions
Daily PM25 from NOX emissions
Annual PM25 from S02 emissions
Annual PM25 from NOX emissions
Figure 3-13 also shows that sometimes there are notable differences in PM2.5 formation
potential for sources in close proximity. Again, these differences are related to differences in
local to regional mix of pollution, terrain, and meteorology. This also shows that spatial
interpolation between these hypothetical sources would not always provide a realistic
representation of model response to the introduction of new precursor emissions.
One interesting example of sources in close proximity with different PM2.5 formation potential
for sulfate and nitrate are the two hypothetical sources in western North Dakota. These sources
are in fairly close proximity but are situated by very different types of emissions sources (e.g.,
large complex of industrial sources, animal operations). Figure 3-14 shows the location of these
sources relative to modeled monthly average ammonia concentration and annual NO2
emissions from the oil and gas sector.
32
-------
Figure 3-14. Monthly average ammonia concentrations estimated by CAMx for July 2011 and
annual total NO2 emissions from the oil and gas sector based on the 2011 National Emission
Inventory.
Figure 3-14 shows that the northern source is in very close proximity to a very large ammonia
source which provides a readily available neutralizing agent for PM2.5 formation when weather
conditions are favorable. However, when winds are out of the north the southern source is in
closer proximity to ammonia emissions located to the south in South Dakota. Further, the
northern source is closer to the Bakken shale which is an area of high emissions that can
provide oxidants for secondary chemical production and compete for neutralizing agents like
ammonia.
N02 emissions from oil&gas sector
-600 -400 -200 0
Ammonia
-600 -400 -200
Therefore, depending on meteorology, these sources will often have different potential for
PM2.5 production given their proximity to other industrial emissions sources and ammonia
emissions sources. Figure 3-15 shows illustrative MERPs estimated for modeled sources for the
daily and annual average forms of the PM2.5 NAAQS.
33
-------
Figure 3-15. Illustrative PM2.5 MERPs for NOx (left panel) and SO2 (right panel) estimated
from single source hypothetical emissions impacts on PM2.5 nitrate ion and PM2.5 sulfate
ion respectively. Note: Daily average PM2.5 MERPs are directly compared with annual average PM2.5 MERPs.
NOX 10 PM2.5 MERPs
S02 10 PM2.5 MERPs
s
:j--::o
2e+05 4©+05 be+05 8e+05
! §
8 -
o
o o (
3 <6
o o
00
O O O
0
OcP
o Elevated release
o Surface release
1 1 1 1—
Oe+OQ 2e+04 4e+04 Ge+04 0e+O4 1e+€5
Annual PM2.5 MERP (fpy)
Annual PM2.5 MERP (tpy)
3.2.1.4. EPA Illustrative MERPs: 8-hour Ozone
The hypothetical single source modeling presented here was used to develop illustrative MERPs
based on equation 1 and the EPA recommended SIL. Figure 3-16 shows illustrative MERPs for
NOx and VOC to daily maximum 8-hr average O3 to illustrate the variability between
regions/studies for the hypothetical sources included in this assessment. The modeled impacts
reflect the highest annual 8-hr O3 impacts from various hypothetical sources presented in this
assessment (Baker et al., 2016; Kelly et al., 2015; U.S. Environmental Protection Agency,
2016b). The hypothetical source impacts presented here were not intended to capture O3
formation associated with winter time cold pool events and are not appropriate for situations
where peak impacts would be expected during these meteorological conditions.
Based on EPA's SILs guidance (U.S. Environmental Protection Agency, 2018), the recommended
8-hour O3 NAAQS SIL of 1.0 ppb was used for this illustrative example. The illustrative VOC
MERPs are based on single source VOC impacts on downwind daily maximum 8-hr O3, while the
illustrative NOx MERPs are based on single source NOx impacts on downwind daily maximum 8-
hr O3. The illustrative MERPs for NOx to daily maximum 8-hr O3 range from 125 tpy to over
5,000 tpy, while the illustrative MERPs for VOC to daily maximum 8-hr O3 range from 1,049 tpy
to over 140,000 tpy for the hypothetical sources presented here.
For this assessment, illustrative MERPs for NOx tend to be lower than VOC which suggests most
areas included in this assessment are often more NOx limited rather than VOC limited in terms
of O3 formation regime. This finding is consistent with the information provided in Section 2.
The distribution of illustrative MERPs for both NOx and VOC are shown to vary between areas
34
-------
modeled as part of this assessment. Similar to PM2.5, this is expected since the chemical (e.g.,
oxidants) and physical (e.g., terrain) environments vary regionally in the United States. The
area-to-area availability of oxidants will determine whether O3 production is NOx or VOC
limited which will be an important factor in how much an emissions source of NOx or VOC will
impact O3 production.
Figure 3-16. NOx (top panels) and VOC (bottom panels) MERPs estimated from single source
hypothetical emissions impacts on daily maximum 8-hr O3.
Note: 8-hr 03 MERPs derived here based on EPA recommended 8-hour 03 NAAQS SIL value of 1.0 ppb
0-hr Ofcone MERPs - NOX precursor
All Sources
8 H
8
E 2
NOAA Climate Zone
0-hr Ofcone MERPs - VOC precursor
All Sources
8 -
8 -
The lowest MERP value for each of the hypothetical source locations is shown for NOx (top) and
VOC (bottom) in Figure 3-17. This shows that even within geographic areas there are
sometimes notable differences in O3 production potential for these precursors. Some broader
patterns do emerge such as VOC emissions having less potential for O3 formation in areas rich
in regional VOC such as the southeast and intermountain west. Differences are also sometimes
seen for sources located in fairly close proximity, which is related to local scale differences in
emissions and meteorology. Figure 3-3 provides additional information about each of the
hypothetical sources to help interpret conceptual differences in O3 formation that may be
related to terrain or proximity to urban areas.
35
-------
Figure 3-17. Lowest MERP value for each hypothetical source location for O3 from NOx (top
panel) and VOC (bottom panel) emissions.
5000
4000
3000
- 2000
- 1000
03 from NOX emissions
03 from VOC emissions
3.2.2. Use of Other Photochemical Modeling to Develop MERPs for O3
and Secondary PM2.5
Given the spatial variability in illustrative MERPs for each precursor for PM2.5 and O3,
stakeholders choosing to develop their own Tier 1 demonstration tool will need to conduct air
quality modeling. Therefore, the air quality modeling should be consistent with the type of
modeling system, model inputs, model application and estimation approach for O3 and
secondary PM25 recommended in the Guideline and the "Guidance on the use of models for
assessing the impacts from single sources on secondarily formed pollutants ozone and PM2.5"
(U.S. Environmental Protection Agency, 2016a). The chosen modeling system should be applied
with a design scope similar to that shown in this document where multiple hypothetical single
sources with varying emission rates and stack release parameters are simulated for a period
that includes meteorology conducive to the formation of O3 and/or secondary PM2.5. A
36
-------
modeling protocol should be developed and shared with the EPA Regional office that details the
planned approach for developing MERPs based on photochemical modeling to ensure a sound
technical basis for development of a suitable Tier 1 demonstration tool.
There is no minimum number of hypothetical sources to include in developing a MERPs Tier 1
demonstration tool, but the benefit of including more hypothetical sources is that more
information is available for future sources to use in predicting secondary pollutant impacts from
their post-construction emissions. Permitting authorities or permit applicants should examine
existing recent (e.g., last 5 to 10 years) permit applications in that area to determine what types
of emission rates and stack characteristics (e.g., surface and elevated release) should be
reflected in the hypothetical project sources included in the model simulations. These model
simulations should include a credible representation of current or post-construction conditions
around the project source and key receptors.
Existing regulatory modeling platforms can be used to minimize resource burden. The most
recently submitted regulatory demonstration (e.g., O3 or PM2.5 attainment demonstration,
Regional Haze SIP demonstration) modeling platform considered appropriate for the purposes
of permit related single source secondary impact demonstrations by the reviewing authority
could provide a platform for development of a MERPs Tier 1 demonstration tool. This could
include the last approved SIP demonstration, a more recent submitted but not yet approved SIP
demonstration, or modeling not used to support a SIP demonstration but considered
representative of the current air quality in the area and of sufficient quality that is comparable
to a model platform supporting a SIP demonstration.
Where multiple appropriate modeling platforms are available for a particular area, the platform
that is considered to be the most reflective of the current atmosphere in a particular area
should be used for the demonstration to account for growth in an area and the changing mix of
sources. For instance, if an area has a SIP modeling platform with a baseline year of 2011 and
projected future year of 2018 and the current year is 2018, then the projected future year may
better represent air quality in that area.
For areas that do not have an existing regulatory demonstration modeling platform, a new
modeling platform that represents the current air quality and conforms to the specifications
outlined for attainment demonstration modeling could be acceptable. The specifications for
permit related demonstration model platforms (e.g., horizontal grid spacing, vertical resolution,
non-project source emission treatment) are detailed in the "Guidance on the use of models for
assessing the impacts from single sources on secondarily formed pollutants ozone and PM2.5"
(U.S. Environmental Protection Agency, 2016a).
These platforms should be assessed for reasonableness with respect to predictive capability
compared to ambient data to ensure that single sources are modeled in a realistic chemical and
physical environment.
37
-------
\ I II v1 ,1- || •in i" • ik'! i-eciiiiK hill IRC ¦
Photochemical modeling conducted for an area by a source, a governmental agency, or some
other entity that is deemed sufficient may be adequate for air agencies to conduct permit
related demonstrations and also or alternatively leading to the development of area-specific
MERPs.
8-hr Ozone: The general framework for such developmental efforts for O3 should include the
following steps:
1) Define the geographic area(s)
2) Conduct a series of source sensitivity simulations with appropriate air quality models to
develop a collection of modeled O3 impacts associated with emissions of O3 precursors
(i.e., VOC and NOx) from typical industrial point sources within the area of interest.
3) Extract the highest daily 8-hr average modeled impact related to each hypothetical
source anywhere in the domain from each model simulation (U.S. Environmental
Protection Agency, 2016a).
4) Calculate the MERP estimate(s) using Equation 1.
5) Conduct quality assurance of the resulting MERP estimate(s) and evaluate the
interpretation and appropriateness given the nature of O3 precursor emissions sources
and chemical formation in the area of interest. This evaluation will likely require
emissions inventory data, observed ambient data for O3 and precursors, a comparison
of baseline total model predictions against ambient data, and qualitative comparison to
MERPs estimated here and elsewhere.
Daily PIVb.s: The general framework for such developmental efforts for daily PM2.5 should
include the following steps:
1) Define the geographic area(s)
2) Conduct a series of source sensitivity simulations with appropriate air quality models to
develop a collection of modeled PM2.5 impacts associated with emissions of PM2.5
precursors (i.e., SO2 and NOx) from typical industrial point sources within the area of
interest.
3) Extract the highest daily 24-hr average modeled impact related to each hypothetical
source anywhere in the domain from each model simulation (U.S. Environmental
Protection Agency, 2016a).
4) Calculate the MERP estimate(s) using Equation 1.
6) Conduct quality assurance of the resulting MERP estimate(s) and evaluate the
interpretation and appropriateness given the nature of PM2.5 precursor emissions
sources and chemical formation in the area of interest. This evaluation will likely require
emissions inventory data, observed ambient data for PM2.5 and precursors, a
comparison of baseline total model predictions against ambient data, and qualitative
comparison to MERPs estimated here and elsewhere.
38
-------
Annual PM?.s: The general framework for such developmental efforts for annual PM2.5 should
include the following steps:
1) Define the geographic area(s)
2) Conduct a series of source sensitivity simulations with appropriate air quality models to
develop a collection of modeled PM2.5 impacts associated with emissions of PM2.5
precursors (i.e., SO2 and NOx) from typical industrial point sources within the area of
interest.
3) Extract the highest annual average modeled impact related to each hypothetical source
anywhere in the domain from each model simulation (U.S. Environmental Protection
Agency, 2016a).
4) Calculate the MERP estimate(s) using the Equation 1.
7) Conduct quality assurance of the resulting MERP estimate(s) and evaluate the
interpretation and appropriateness given the nature of PM2.5 precursor emissions
sources and chemical formation in the area of interest. This evaluation will likely require
emissions inventory data, observed ambient data for PM2.5 and precursors, a
comparison of baseline total model predictions against ambient data, and qualitative
comparison to MERPs estimated here and elsewhere.
If there are questions about what steps are appropriate in each instance or how to apply the
steps described above, air agencies should contact their Regional office modeling contact for
further technical consultation.
39
-------
4. Application of the MERPs to Individual Permit Applications
The Guideline recommends a two-tiered approach for addressing single-source impacts on O3
or secondary PM2.5 (U.S. Environmental Protection Agency, 2017a) with the first tier involving
use of appropriate and technically credible relationships between emissions and ambient
impacts developed from existing modeling studies deemed sufficient for evaluating a project
source's impacts. Consistent with the recommendations in EPA's Guideline, the appropriate tier
for a given application should be selected in consultation with the appropriate reviewing
authority (paragraph 3.0(b)) and after reviewing EPA guidance. This section describes how
applicants might choose, in consultation with the appropriate permitting authority, to use
MERPs in estimating single-source impacts on secondary pollutants under the first-tier
approach (i.e., sections 5.3.2.b and 5.4.2.b of the Guideline).
The use of MERPs as a Tier 1 demonstration tool can be based on either (1) EPA photochemical
modeling with the source-specific value for a representative hypothetical source (as described
in Section 3.2.1) or (2) the source- or area-specific value derived from a more similar
hypothetical source modeled by a permit applicant or permitting authority (as described in
Section 3.2.2). In some situations, the most conservative (lowest) MERP value across a
region/area could be considered representative. The relevant geographic area could range from
a county or airshed to a state or multi-state region. The selection of this geographic area may
be determined in consultation with the appropriate reviewing authority and technical
justification should be provided in the modeling protocol and/or permit-related
documentation.
EPA recommends that the permit applicant follow a three-step process as shown in Figure 4-1.
1) Identify a representative hypothetical source (or group of sources for an area) from EPA's
modeling as detailed in Appendix Table A-l or the Excel spreadsheet available on SCRAM. If
a representative hypothetical source is not available, then consider whether an EPA derived
MERP value available for the broader geographic area of the project source may be
adequately representative and thus appropriate to use (see Table 4-1). Alternatively, one
can consider conducting photochemical modeling (as described in Section 3.2.2) to derive
appropriate information to derive a source- or area-specific value.
The permit applicant should provide the appropriate permitting authority with a technically
credible justification that the source characteristics (e.g., stack height, emissions rate) of the
specific project source described in a permit application and the chemical and physical
environment (e.g., meteorology, background pollutant concentrations, and regional/local
emissions) near that project source are adequately represented by the selected
hypothetical source(s).
2) Acquire the source characteristics and associated modeling results for the hypothetical
source(s). If using EPA modeling, then access these data from the on-line spreadsheet on
40
-------
EPA's SCRAM website. If using other modeling, then access these data from the relevant
input and output files.
3) Apply the source characteristics and photochemical modeling results from Step 2 to the
MERP equation with the appropriate SIL value to assess the project source impacts.
Section 4.1 provides several example PSD permit application scenarios that illustrate how to
use source characteristics and photochemical modeling results to derive a MERP Tier 1
demonstration tool. In general, for situations where the project source emits only one
precursor for O3 or secondary PM2.5 (and no primary PM2.5 emissions), the project source
emissions for that precursor can be compared directly to the appropriate MERP value for
that precursor to determine if the applicable SIL is exceeded or not. For situations where
project sources are required to assess multiple precursors, EPA recommends that the
project source impacts on O3 or secondary PM2.5 reflect the sum of air quality changes
resulting from each of those precursors for comparison to the EPA recommended SIL.
Further, where project sources are required to assess both primary PM2.5 and precursors of
secondary PM2.5, EPA recommends that applicants combine the primary and secondary
impacts to determine total PM2.5 impacts as part of the PSD compliance demonstration. In
such cases, the project source impacts associated with their direct PM2.5 emissions should
be assessed through dispersion modeling.
At the start of this process, EPA recommends that the permit applicant consult with the
appropriate reviewing authority in developing a modeling protocol (per Section 9 of the
Guideline) and that both parties confirm, at that time, the appropriateness of using these
modeling results for the permitting situation. As part of the protocol, the permit applicant
should include a narrative that provides a technical justification that the existing information or
planned photochemical modeling is appropriate for the project source(s).
Derived from EPA modeling results, Table 4-1 summarizes the distribution of illustrative MERPs
values across climate zones showing the lowest, highest and median values. Consistent with
Step 1 outlined above, the most conservative (lowest) illustrative MERP value may, in some
cases, be considered adequately representative to characterize the responsiveness of ozone or
secondary PM2.sto precursors emitted in a region or area and then be considered for the Tier 1
demonstration in an individual permit application. Climate zones are only used here to
summarize the MERPs values for the reader. EPA recommends that the permit applicant
consult with the appropriate reviewing authority to determine the relevant geographic area
and/or hypothetical source from which to select a representative MERP value.
41
-------
Figure 4-1. EPA recommended multi-step process for use of MERPs in PSD compliance
demonstrations.
Step 1
Identify representative hypothetical source(s)
From EPA modeling (Table A-l)
From other modeling
If representative source not available
1
Consider whether EPA-
derived MERP values
available for region of
the project source
If no existing modeling has a
representative source,
consider conducting
photochemical modeling to
derive appropriate source or
area specific MERP value
Acquire source characteristics and associated source impact
modeling results
From EPA modeling (online
spreadsheet)
From other photochemical
modeling
I
Apply the source characteristics and photochemical modeling results
from Step 2 to the MERP equation with the appropriate SIL value to
assess the project source impacts
42
-------
Table 4-1. Lowest, median, and highest illustrative MERP values (tons per year) by precursor,
pollutant and climate zone.
Note: illustrative MERP values are derived based on EPA modeling and EPA recommended SILs from EPA's final SILs guidance
(U.S. Environmental Protection Agency, 2018).
8-hr03 from NOx
8-hr 03 from VOC
Climate Zone
Lowest
Median
Highest
Lowest
Median
Highest
Northeast
209
495
5,773
2,068
3,887
15,616
Southeast
170
272
659
1,936
7,896
42,964
Ohio Valley
126
340
1,346
1,159
3,802
13,595
Upper Midwest
125
362
4,775
1,560
2,153
30,857
Rockies/Plains
184
400
3,860
1,067
2,425
12,788
South
190
417
1,075
2,307
4,759
30,381
Southwest
204
422
1,179
1,097
10,030
144,744
West
218
429
936
1,094
1,681
17,086
Northwest
199
373
4,031
1,049
2,399
15,929
Daily PM2.5from NOx
Daily PM2.5from S02
Climate Zone
Lowest
Median
Highest
Lowest
Median
Highest
Northeast
2,218
15,080
34,307
623
3,955
8,994
Southeast
1,943
8,233
23,043
367
2,475
5,685
Ohio Valley
2,570
10,119
32,257
348
3,070
16,463
Upper Midwest
2,963
10,043
29,547
454
2,482
6,096
Rockies/Plains
1,740
9,389
31,263
251
2,587
19,208
South
1,881
8,079
24,521
274
1,511
10,112
Southwest
6,514
26,322
101,456
1,508
8,730
27,219
West
1,073
8,570
34,279
188
2,236
24,596
Northwest
3,003
11,943
20,716
1,203
3,319
8,418
Annual
PM2.5from NOx
Annual PM2.5from S02
Climate Zone
Lowest
Median
Highest
Lowest
Median
Highest
Northeast
10,142
47,396
137,596
4,014
21,353
41,231
Southeast
5,679
45,076
137,516
859
14,447
25,433
Ohio Valley
7,625
31,931
150,868
3,098
23,420
58,355
Upper Midwest
10,011
33,497
139,184
2,522
17,997
45,113
Rockies/Plains
9,220
39,819
203,546
2,263
16,939
106,147
South
7,453
41,577
110,478
1,781
11,890
58,612
Southwest
11,960
128,564
779,117
10,884
38,937
105,417
West
3,182
29,779
103,000
2,331
11,977
66,773
Northwest
7,942
21,928
71,569
11,276
15,507
18,263
43
-------
! I III'! .'rati, h'llfjr IF IlkT l IVliKJli Hi'iKh' ik I ( «ln|'kl -I» I! rll lult
S
In this section, several example PSD permit application scenarios are presented to illustrate
how modeled emissions and secondary pollutant impacts from EPA's modeling of hypothetical
sources (described in Section 3.2.1) could be used to derive a MERP Tier 1 demonstration tool
(as described in Section 3.1) for a given location. Some of these examples demonstrate how to
account for multiple precursor impacts on secondary PM2.5 formation. One scenario (i.e.,
scenario D) reflects a situation where a project source emits both primary PM2.5 and precursors
to secondary PM2.5. In those situations, applicants should consult the appropriate sections of
the Guideline (U.S. Environmental Protection Agency, 2017a) and related permit modeling
guidance for information about estimating primary PM2.5 impacts. As illustrated in these
examples, representative MERPs for each precursor may be developed based on either the
most conservative (lowest) value across a region/area or the source-specific value derived from
a more similar hypothetical source modeled by a permit applicant, permitting authority, or EPA.
For multiple areas, Table 4.1 shows an example of the most conservative (i.e., lowest)
illustrative MERP for each precursor and NAAQS across all sources and studies. These
illustrative values in Table 4.1 are based on the EPA modeling of hypothetical sources described
in Section 3.2.1. For reference at the individual source level, the maximum predicted downwind
impacts for each of the hypothetical sources modeled with annual simulations are provided in
the Excel spreadsheet available on EPA's SCRAM website.
! 1 1 1mii e iiii|" ii'i ''mii »lh --I'Himl FIMi ¦«, 111
The following section provides examples of developing a suitable Tier 1 demonstration tool for
each precursor and secondary pollutant as part of a PSD source impact analysis for the O3 and
PM2.5 NAAQS. Where only a single precursor of O3 or PM2.5, and no direct PM2.5, is emitted by
the project source, then the MERP for that precursor may be directly applied. For situations
where project sources are required to assess multiple precursors of PM2.5 or of O3, EPA
recommends that the impacts of multiple precursors should be estimated in a combined
manner for comparison to the appropriate SIL such that the sum of precursor impacts would be
lower than the SIL in a demonstration of compliance. Further, where project sources are
required to assess both primary PM2.5 and precursors of secondary PM2.5, EPA recommends
that applicants combine the primary and secondary impacts to determine total PM2.5 impacts as
part of the PSD compliance demonstration. In such cases, the project source impacts associated
with their direct PM2.5 emissions should be assessed through dispersion modeling.
In this assessment, the maximum downwind impact from each source is chosen over the length
of the model simulation period and matched with the annual emission rate. The maximum
impact is selected since a single year of meteorology (or less in some instances) is used to
generate these relationships. Additional or alternative meteorological patterns may result in
44
-------
different impacts in some areas. The following illustrative examples are intended to show how
MERP values may be used in specific PSD permit air quality demonstrations.
Scenario A: Single precursor assessment for PM2 5 and additive O3 impacts
In this scenario, a PSD permit applicant with a proposed increase in emissions of 0 tpy of
primary PM2.5, 130 tpy of VOC, 72 tpy of NOx, and 0 tpy of SO2 located in the upper midwest
region.
O3 analysis: The project source is not located in an area with unusual circumstances regarding
complex terrain, proximity to very large sources of either NOx or VOC, or meteorology. Thus,
the climate zone may be defined as the relevant geographic area such that the lowest MERPs
from Table 4-1 for the upper midwest region could be considered representative and chosen
for comparison with the project emissions rather than selecting a particular hypothetical source
from this same climate zone. In practice, EPA recommends that the permit applicant consult
with the appropriate reviewing authority to determine the relevant hypothetical source and
geographic area from which to select representative MERP values.
The NOx emissions of 72 tpy and VOC emissions of 130 tpy from the project source are well
below the lowest (most conservative) MERP values for NOx as an O3 precursor (i.e., 125 tpy)
and VOC as an O3 precursor (i.e., 1,560 tpy), respectively, of all sources modeled by EPA in the
upper midwest region, as shown in Table 4-1. In this case, air quality impacts for each O3
precursor from this source would be expected to be below the EPA recommended 8-hour O3
SIL.
However, for this example, EPA recommends that the NOx and VOC precursor impacts on 8-hr
daily maximum O3 be considered together to determine if the project source's air quality
impact would exceed the O3 SIL. In such a case, the project source's emissions increase can be
expressed as a percent of the MERP for each precursor and then the percentages can be
summed. A value less than 100% indicates that the EPA recommended 8-hour O3 SIL will not be
exceeded when considering the combined impacts of these precursors on 8-hr daily maximum
03.
Example calculation for additive precursor impacts on 8-hr daily maximum O3:
(72 tpy NOx from source/125 tpy NOx 8-hr daily maximum O3 MERP) + (130 tpy VOC
from source/1,560 tpy VOC 8-hr daily maximum O3 MERP) = .58 + .08 = .66 * 100 = 66%
A value less than 100% indicates that the O3 SIL would not be exceeded when considering the
combined impacts of these precursors. Thus, the project level O3 impacts associated with both
NOx and VOC precursor emissions from this source would be expected to be below the EPA
recommended 8-hour O3 SIL.
PM?.s analysis: The project source is not located in an area with unusual circumstances
45
-------
regarding complex terrain, proximity to very large sources of pollutants that impact
atmospheric chemistry (i.e., NOx, SO2, NH3) or meteorology. Thus, similar to the O3 analysis
above, the climate zone may be defined as the relevant geographic area such that the lowest
MERPs from Table 4-1 for the upper midwest region could be considered adequately
representative and chosen for comparison with the project emissions rather than selecting a
particular hypothetical source from this same region. EPA recommends that the permit
applicant consult with the appropriate reviewing authority to determine the relevant
hypothetical source and geographic area from which to select representative MERP values.
The project source emits no direct PM2.5 nor SO2 so the demonstration focuses only on the NOx
emissions increase of 72 tpy, which is well below the lowest (most conservative) MERP value in
the upper midwest region for NOx as a precursor for the daily and annual PM2.5 NAAQS shown
in Table 4-1, i.e., 2,963 tpy and 10,011 tpy respectively. In this case, air quality impacts of PM2.5
from this source are expected to be below the EPA recommended 24-hour and annual PM2.5
SILs.
Scenario B: Single precursor assessment for O3 impacts and additive secondary PM2.5 impacts
In this scenario, a facility with a proposed increase in emissions of 0 tpy of primary PM2.5, 0 tpy
of VOC, 220 tpy of NOx, and 75 tpy of SO2 located in the southeast region.
O3 analysis: The project source is not located in an area with unusual circumstances regarding
complex terrain, proximity to very large sources of either NOx or VOC, or meteorology. The
project source does not emit VOC so the demonstration focuses only on the NOx emission
increase of 220 tpy, which is greater than the lowest (most conservative) NOx MERP for 8-hr O3
in the southeast region (i.e., 170 tpy). Thus, for this example, even though the project source's
surrounding environment does not raise an obvious regional feature that would influence
downwind O3 impacts, it is likely more appropriate to use a specific hypothetical source in the
same region or other appropriate geographic area for comparison.
A comparable hypothetical source is identified to be representative of this source (e.g.,
southeast region source located in Tallapoosa County, Alabama with elevated emissions
release). Here, equation 1 is used with the modeled emissions rates and air quality impact
information from this hypothetical source. Since multiple hypothetical sources were modeled at
this location with an elevated release, the source with the lowest MERP was selected for
comparison with the project source, i.e.,
MERP for selected representative hypothetical source (tpy) = 1.0 ppb * (500 tpy /1.528
ppb) = 327 tpy
In this case, based on EPA modeling results for a representative hypothetical source, the project
source emissions are less than the calculated NOx to 8-hr O3 MERP such that air quality impacts
of O3 from this source would be expected to be less than the EPA recommended 8-hour O3 SIL.
46
-------
PIVb.s analysis: The project source is not located in an area with unusual circumstances
regarding complex terrain, proximity to very large sources of pollutants that impact
atmospheric chemistry (i.e., NOx, SO2, NH3) or meteorology. Thus, the climate zone may be
defined as the relevant geographic area such that the lowest MERPs from Table 4-1 for the
southeast region could be considered adequately representative and chosen for comparison
with the project emissions rather than selecting a particular hypothetical source from this same
region. In practice, EPA recommends that the permit applicant consult with the appropriate
reviewing authority to determine the relevant hypothetical source and geographic area from
which to select representative MERP values.
For this example, both the NOx emissions of 220 tpy and SO2 emissions of 75 tpy are well below
the lowest (most conservative) daily PM2.5 MERP values of any source modeled in the
southeastern region, i.e., 1,943 tpy for NOx and 367 tpy for SO2 respectively. These emission
rates are also well below the annual PM2.5 MERP values of any source modeled in the
southeastern region (see Table 4-1).
However, for this example, EPA recommends that the NOx and SO2 precursor impacts to both
daily and annual average PM2.5 are considered together to determine if the project source's air
quality impact on PM2.5 would exceed the PM2.5 SILs. In this case, the project source's emissions
increase can be expressed as a percent of the MERP for each precursor and then the
percentages can be summed. A value less than 100% indicates that the EPA recommended daily
or annual PM2.5 SIL would not be exceeded when considering the combined impacts of these
precursors on daily or annual PM2.5.
Example calculation for additive secondary impacts on daily PM2.5:
(220 tpy NOx from source/1,943 tpy NOx daily PM2.5 MERP) + (75 tpy SO2 from
source/367 tpy S02 daily PM2.5 MERP) = .11 + .20 = .31 * 100 = 31%
Example calculation for additive secondary impacts on annual PM2.5:
(220 tpy NOx from source/5,679 tpy NOx annual PM2.5 MERP) + (75 tpy SO2 from
source/859 tpy S02 annual PM2.5 MERP) = .04 + .09 = .13 * 100 = 13%
A value less than 100% indicates that the PM2.5 SIL would not be exceeded when considering
the combined impacts of these precursors on daily or annual PM2.5. Thus, in this case, the air
quality impacts of PM2.5 from precursor emissions of NOx and SO2 from this source would be
expected to be less than the EPA recommended daily and annual PM2.5 SILs.
Scenario C: Single precursor assessment for O3 and additive PM2.5 impacts
In this scenario, a facility with a proposed increase in emissions of 0 tpy of primary PM2.5, 0 tpy
of VOC, 920 tpy of NOx, and 259 tpy of SO2 located in the Rockies region.
47
-------
Os analysis: The project source is not located in an area with unusual circumstances regarding
complex terrain, proximity to very large sources of either NOx or VOC, or meteorology. The
project source does not emit VOC so the demonstration focuses only on the NOx emission
increase of 920 tpy, which is greater than the lowest (most conservative) NOx MERP for 8-hr O3
in the Rockies region (i.e., 184 tpy). Thus, for this example, even though the project source's
surrounding environment does not raise an obvious regional feature that would influence
downwind O3 impacts, it is likely more appropriate to use a hypothetical source for comparison.
A comparable hypothetical source is identified to be representative of this source (e.g., Rockies
region in Iron County, Utah with elevated release). Here, equation 1 is used with the modeled
emissions rates and air quality impact information from the selected comparable source. Since
multiple hypothetical sources were modeled at this location with an elevated release, the
source with the most similar emission rate was selected for comparison with the project
source, i.e.,
MERP for selected representative hypothetical source (tpy) = 1.0 ppb * (1000 tpy / 1.314
ppb) = 761 tpy
In this case, based on EPA modeling results for a representative hypothetical source, the project
source emissions are greater than the calculated NOx to 8-hr O3 MERP such that air quality
impacts of O3 from this source are expected to exceed the EPA recommended 8-hour O3 SIL.
Given that the NOx emissions from this project source are expected to have air quality impacts
that exceed the O3 SIL, a cumulative impact analysis would be the next step in this scenario.
More information for this type of demonstration is provided in Section 4.1.3.
PM?.s analysis: The project source is not located in an area with unusual circumstances
regarding complex terrain, proximity to very large sources of pollutants that impact
atmospheric chemistry (i.e., NOx, SO2, NH3) or meteorology. The NOx emissions of 920 are
below the lowest (most conservative) daily and annual PM2.5 MERP value of any source
modeled in the Rockies region (i.e., 1.740 tpy and 9,220 tpy respectively), while the SO2
emissions of 259 tpy are slightly higher than the lowest daily PM2.5 MERP value of any source
modeled in the Rockies region (i.e., 251 tpy for daily and 2,263 tpy for annual). Thus, for this
example, even though the project source's surrounding environment does not raise an obvious
regional feature that would influence downwind secondary PM2.5 impacts, it is likely more
appropriate to use a hypothetical source for comparison.
A hypothetical representative source is identified to be representative of this source (e.g.,
Rockies region in Iron County, Utah) and has a 1,000 tpy elevated release NOx MERP for daily
PM2.5 of 25,754 tpy and SO2 MERP for daily PM2.5 of 7,515 tpy, which are both much larger than
the increase in emissions of the project source such that the source's impact on daily PM2.5
would be expected to be less than the EPA recommended daily PM2.5 SIL. The same
hypothetical source has a NOx MERP for annual PM2.5 of 166,670 tpy and SO2 MERP for annual
PM2.5 of 37,997 tpy, which are both much larger than the increase in emissions of the project
source such that the source's impact on annual PM2.5 would be expected to be less than the
48
-------
EPA recommended annual PM2.5 SIL. However, for this example, EPA recommends that the NOx
and SO2 precursor contributions to both daily and annual average PM2.5 are considered
together to determine if the project source's air quality impact of PM2.5 would exceed the PM2.5
SILs. In this case, the project source's emissions increase can be expressed as a percent of the
MERP for each precursor and then the percentages can be summed.
Example calculation for additive secondary impacts on daily PM2.5:
(920 tpy NOx from source/25,754 tpy NOx daily PM2.5 MERP) + (259 tpy SO2 from
source/7,515 tpy S02 daily PM2.5 MERP) = .036 + .034 = .07 * 100 = 7%
Example calculation for additive secondary impacts on annual PM2.5:
(920 tpy NOx from source/166,670 tpy NOx annual PM2.5 MERP) + (259 tpy SO2 from
source/37,997 tpy S02 annual PM2.5 MERP) = .006 + .007 = .013 * 100 = 1.3%
A value less than 100% indicates that the PM2.5 SIL would not be exceeded when considering
the combined impacts of these precursors on daily or annual PM2.5. Thus, in this case, the air
quality impacts of PM2.5 from precursor emissions of NOx and SO2 from this source would be
expected to be less than both the EPA recommended daily and annual PM2.5 SILs.
Scenario D: NOx and SO2 precursor assessment for additive secondary PM2.5 impacts along
with direct PM2.5
In this scenario, a facility with a proposed increase in emissions of 250 tpy of primary PM2.5, 0
tpy of VOC, 220 tpy of NOx, and 75 tpy of SO2 located in the southeast region. This scenario is
like Scenario B above, except that EPA recommends that in assessing PM2.5 the primary PM2.5
emissions be accounted for along with the secondary impacts of PM2.5 precursor emissions as
part of the Tier 1 demonstration.
O3 analysis: See scenario B above.
PM?.s analysis: Same as Scenario B as to PM2.5 precursors. The combined impacts of the
proposed increases in PM2.5 precursor emissions of NOx and SO2 would not exceed the EPA
recommended daily or annual PM2.5 SILs.
However, for this example, EPA recommends that the primary PM2.5 impacts be added to the
secondary impacts for a full account of total PM2.5 impacts in comparison to the daily and
annual PM2.5 SILs. The primary PM2.5 impacts should be estimated using AERMOD or an
approved alternative model as outlined in the Guideline (U.S. Environmental Protection Agency,
2017a) and consistent with EPA guidance for combining primary and secondary impacts of
PM2.5 for permit program assessments.
In this scenario, a representative secondary PM2.5 impact for this source is added to the
49
-------
appropriately estimated primary PM2.5 impacts. The highest ambient impact at any receptor for
primary PM2.5 should be divided by the daily or annual PM2.5 SIL values to estimate the primary
impact calculated as a percentage of the SIL value and then added to the previously calculated
secondary impacts.
For the daily PM2.5 NAAQS, a peak primary PM2.5 impact from AERMOD in this scenario is
estimated to be 0.41 |-ig/m3. Compared with a 1.2 |-ig/m3 SIL for daily PM2.5 means that the
primary impact is 34% of the SIL. When this primary impact is summed with the secondary
impacts of 31% the total is 65% which is below 100% suggesting this source impact is below the
EPA recommended daily PM2.5 SIL.
For the annual PM2.5 NAAQS, annual average primary PM2.5 impact from AERMOD is estimated
to be 0.11 |-ig/m3 for the scenario above. Compared with a 0.2 |-ig/m3 SIL for annual PM2.5
means that the primary impact is 55% of the SIL. When this primary impact is summed with the
secondary impacts of 13% the total is 68% which is below 100% suggesting this source impact is
below the EPA recommended annual PM2.5 SIL.
Accounting for spatial correlation of primary and secondary impacts: As a variant on this
scenario, for the daily PM2.5 NAAQS, if the peak primary PM2.5 impact from AERMOD is
estimated to be 0.90 |-ig/m3 for the above scenario, then the percent primary contribution to
the SIL would be 75%. When summed with the secondary contribution of 31%, the total source
impact exceeds 100% and, therefore, is greater than the EPA recommended daily PM2.5 SIL. In
this case, the spatial nature of the primary and secondary PM2.5 impacts of the project source
may be resolved in a more detailed manner to gain a better estimate of the project source
impact for comparison to the PM2.5 SILs. Primary impacts tend to be higher in closer proximity
of the source, whereas secondary impacts can be higher further downwind (beyond the
property fence line). For example, the primary and secondary PM2.5 impacts could be resolved
at varying distances from the source (e.g., within 5-10 km, between 10 and 25 km, and between
25 and 50 km) and then combined at each distance range for a comparison with the EPA
recommended PM2.5 SILs. If the more spatially resolved assessment still finds combined
percentages above 100%, then a cumulative impact analysis would be the next step for this
demonstration. More information for this type of demonstration is provided in Section 4.1.3.
1 f J"IU6 1II11111 h i Hii «lh 'II 1 I < - f I" If" III in Ik liirhi k II I M2.5
This section provides information for single source permit demonstrations for PSD increment of
PM2.5at Class I areas. According to 40 CFR 51.166(c)(1) and 52.21(c), an allowable PSD
increment based on an annual average may not be exceeded, and the allowable PSD increment
for any other time period may be exceeded once per year at any one location. Currently there is
no PSD increment for O3 so no PSD increment demonstration for O3 is necessary. The PM2.5 PSD
increment SIL values recommended by EPA for Class II and III areas are the same as the
recommended PM2.5 NAAQS SIL values so no separate PSD increment demonstration is needed
for Class II and III areas.
50
-------
The hypothetical model results provided in this document represent peak impacts for
secondary PM2.5, which are typically within 50 km from the source (see section 3.2.1). These
impacts may not be applicable for PSD increment demonstrations at Class I area receptors that
may be far downwind (beyond 50 km) of the project source. As stated in the Guideline,
AERMOD is the preferred dispersion model for estimating primary PM2.5 impacts from single
sources for distances up to 50 km. Currently, there is no preferred modeling system for
estimating long range transport impacts (i.e., beyond 50 km). The Guideline establishes a
screening approach for such assessments (U.S. Environmental Protection Agency, 2017a).
The screening approach for the primary PM2.5 component of a PSD Class I area demonstration
beyond 50 km could include AERMOD estimates at or about 50 km from the project source
(Section 4.2.c.i of the Guideline) or a second level assessment based on modeling primary
PM2.5 that does not include plume-depleting processes to ensure a conservative estimate
(Section 4.2.c.ii of the Guideline). The Guideline suggests a Lagrangian or comparable modeling
system would be appropriate for a second level assessment. Photochemical grid models have
been shown to demonstrate similar skill to Lagrangian models for long range pollutant
transport when compared to measurements made from multiple mesoscale field experiments
(ENVIRON, 2012a; U.S. Environmental Protection Agency, 2016c). EPA modeled a subset of the
hypothetical sources shown in Figure 3-2 with tracking of primary PM2.5 contribution (N=36)
using the CAMx model applied without chemistry. A table of maximum daily average and
maximum annual average primary PM2.5 impacts by emission rate are shown in Table 4-2. This
table is intended to provide illustrative information about peak downwind primary PM2.5
impacts at distances beyond 50 km and where agreed to by the appropriate reviewing authority
may provide relevant information to support Tier 1 PSD Class I increment demonstrations.
Table 4-2. Maximum daily average and maximum annual average primary PM2.5 impacts at
100, 200, and 300 km from modeled hypothetical source.
Highest Daily Average Highest Daily Average Highest Annual Average Highest Annual Average
Emission Distance from Concentration (ng/m3) -Concentration (ng/m3) - Concentration (ng/m3) - Concentration (ng/m3) -
Rate (tpy)
source (km)
tall stack
surface release
tall stack
surface release
100
300
0.0117
0.0123
0.0008
0.0009
100
200
0.0223
0.0212
0.0016
0.0015
100
100
0.0537
0.0445
0.0070
0.0049
150
300
0.0180
0.0184
0.0012
0.0013
150
200
0.0328
0.0311
0.0024
0.0022
150
100
0.0807
0.0632
0.0102
0.0073
500
300
0.0610
0.0625
0.0044
0.0045
500
200
0.1167
0.1095
0.0087
0.0078
500
100
0.2717
0.2536
0.0379
0.0238
1000
300
0.1186
0.1217
0.0087
0.0089
1000
200
0.2300
0.2161
0.0175
0.0157
1000
100
0.5445
0.5009
0.0731
0.0477
Single source impacts on secondary PM2.5 tend to decrease as distance from the source
increases (Baker et al., 2016), which means peak source impacts presented in previous sections
51
-------
to inform a PM2.5 NAAQS air quality assessment may not provide relevant information for the
spatial scales involved between project sources and Class I areas. Given that project source
impacts will be lower at greater distances (see also Figure 3.6), the illustrative MERPs listed in
Section 4 would not usually be relevant (unless the source and Class I area were in close
proximity), so applicants should follow the screening approach described in this section for a
Tier 1 demonstration of compliance with the Class I PSD increment for PM2.5.
The hypothetical source impact information generated as part of the illustrative examples
shown here or other credible existing single source modeling could provide information
relevant for Class I SIL screening demonstrations. Rather than using the peak impact, the
entirety of modeled information available for a specific project source (if available) or
hypothetical source (such as but not limited to the sources modeled as part of this document)
could be used to provide an estimate of secondary PM2.5 impacts at distances further
downwind.
Consistent with the long-range transport (LRT) screening approach in the Guideline, the initial
screening step would be to select one or more of the hypothetical sources modeled as part of
the illustrative assessment provided in this document that are found to be similar to the project
source. Then, modeled maximum secondary PM2.5 impacts at or greater than 50 km would be
used in combination with primary PM2.5 impacts estimated with AERMOD at 50 km downwind
of the source for comparison to the EPA recommended PM2.5 Class I SIL value. Information
about using AERMOD to support a LRT demonstration for primary pollutants is provided
elsewhere (U.S. Environmental Protection Agency, 2016d).
If the results of the initial screening step show an exceedance of the PM2.5 Class I SIL value, a
second more refined screening step would involve selecting the highest modeled secondary
PM2.5 impact at or less than the downwind distance of the Class I area relative to the project
source. That value would be combined with primary PM2.5 impacts estimated with AERMOD at
50 km downwind and compared with the EPA recommended PM2.5 Class I SIL. Another option
for this screening step would also involve selecting the highest modeled secondary PM2.5 impact
at or near the downwind distance of the Class I area relative to the project source but include
an estimate of primary PM2.5 impacts estimated with a chemical transport model (e.g.,
Lagrangian or photochemical model) at or less than the downwind distance of the Class I area
relative to the project source.
An illustrative example of this type of a screening demonstration for Class I PM2.5 increment
would be a 3,000 tpy NOx project source that emits near the surface in the northeast U.S. This
project source does not emit SO2 so secondary formation of PM2.5 sulfate ion does not need to
be considered in addition to PM2.5 nitrate formation from the NOx emissions. The nearest Class I
area is ~300 km downwind of the project source. Multiple hypothetical sources (3 for this
particular example) with ground-level emission release characteristics near the project source
were examined for annual and 24-hr average PM2.5 nitrate impacts at or greater than 50 km and
at or near 300 km downwind of the source in any direction. Figure 4-2 shows the peak
hypothetical source impacts from 500 tpy of emissions at ~50 km downwind on PM2.5 nitrate for
52
-------
daily PM2.5 is 0.032 |ig/m3 and annual PM2.5 is 0.002 |ig/m3. As shown, at approximately 310 km
from the project source, the peak hypothetical source impacts on PM25 nitrate for daily PM2.5
would be 0.01 |ig/m3 and 0.0003 ng/m3 for annual PM2.5 (see Figure 4-2).
Figure 4-2. Modeled peak daily average (top) and annual average (bottom) PM2.5 nitrate ion
impacts from a hypothetical 500 tpy surface level source of NOx emissions by distance
downwind of the source.
NITR-24h
rh
E
u.
0.06
0.05
0.04
0.03
0.02
0.01
0
11IIIII
0 20 40 60 80 100 120 140 160 180 200 220 240 260 280 300 320 340
Distance (km)
NITR-Ann.
0.007
0.006
c
c
1*
0.005
3
++
!^l
0.004
na
CL
E
0.003
15
3
0.002
E
<
0.001
0
I I I I I i i ¦ i ¦ i
¦ j I • i i IIi¦
0 20 40 60 80 100 120 140 160 180 200 220 240 260 280 300 320 340
Distance (km)
The hypothetical source NOx emission rate is 500 tpy and the project source emission rate is
3,000 tpy. Impacts from the 500 tpy hypothetical sources are linearly scaled (increased in this
example) to be better representative of the project source emission rate. For example, the daily
PM2.5 nitrate impacts at 50 km downwind would be adjusted to 0.192 |ig/m3: 0.032 |ig/m3 *
53
-------
3000 tpy/500 tpy = 0.192 |-ig/m3. The annual PM2.5 nitrate impacts at 300 km downwind would
be adjusted to 0.0018 |ag/m3: 0.0003 |-ig/m3 * 3000 tpy/500 tpy = 0.0018 |-ig/m3.
As part of the initial screening step, the project source impact of 0.192 |-ig/m3 for daily PM2.5 at
50 km downwind is added to its primary impact estimated with AERMOD at 50 km for
comparison with the EPA recommended 24-hr PM2.5Class I area SIL of 0.27 |-ig/m3. Assuming
the primary impacts are below 0.078 |-ig/m3, the project source could include this screening
demonstration in its PSD application. Otherwise, the project source would move on to the
second step with more refined screening demonstration based on 0.01 |-ig/m3 impacts per 500
tpy NOx at 300 km distance downwind, i.e., 0.01 |-ig/m3 * 3000 tpy/500 tpy = 0.06 |-ig/m3 of
PM2.5 nitrate.
This estimate of secondary contribution at the distance of the Class I area from the project
source would then be added to the primary impacts modeled with AERMOD at 50 km and be
compared with the EPA recommended PM2.5 Class I SIL. If the sum of the more refined
secondary contribution paired with the primary PM2.5 contribution exceeds the SIL, the next
step in the screening demonstration would utilize an estimate of primary PM2.5 using a chemical
transport model (e.g., Lagrangian or photochemical model) that can be paired with the
secondary impact at 300 km downwind (as shown above). In situations where the screening
demonstration does not show downwind impacts of PM2.5 at Class I areas below the SIL, then a
more refined approach to estimate the impacts from their project source based on methods
suggested for Tier 2 demonstrations may be considered prior to conducting a cumulative
impact analysis.
! I MiiiiinhG k II11111 >''i i '"Ml «lv II. i. .'II ! M 1 .'5 NAAQS
As detailed in Section 9 of the Guideline, for situations where the project source is not able to
demonstrate compliance through the source impact analysis, a cumulative impact analysis can
be conducted that accounts for the impacts from the project source, impacts from nearby
sources (as appropriate), and monitored background levels. The cumulative impacts are then
compared to the NAAQS to determine whether the project source could cause or contribute to
a NAAQS exceedance.
The following section provides examples of developing a suitable Tier 1 demonstration tool for
each precursor and secondary pollutant for the purposes of a cumulative impact analysis.
Where only a single precursor of O3 or PM2.5 necessitates a demonstration, then a direct
application of this approach would be appropriate. For situations where project sources are
required to assess multiple precursors of PM2.5 or of O3, EPA recommends that the impacts of
multiple precursors should be estimated in a combined manner for comparison to the
appropriate SIL such that the sum of precursor impacts would be lower than the SIL in a
demonstration of compliance. Further, where project sources are required to assess both
primary PM2.5 and precursors of secondary PM2.5, EPA recommends that applicants combine
the primary and secondary impacts to determine total PM2.5 impacts as part of the PSD
54
-------
compliance demonstration. In such cases, the project source impacts associated with their
direct PM2.5 emissions should be assessed through dispersion modeling. The examples below
include each of these situations.
The Tier 1 demonstration approach detailed in Section 3 of this document can be modified for
use in a cumulative impact assessment. Here, existing relevant single source modeled impacts
can be estimated and then added to the appropriate background contribution for comparison
to the NAAQS. The MERP equation (Eq. 1) can be rearranged such that instead of calculating a
modeled emission rate based on a critical air quality threshold such as a SIL value, a project
specific impact would be estimated. Equation 2 shows how a project source impact would be
the product of the relevant hypothetical source air quality impact relative to emissions scaled
either upwards or downwards to the emission rate of the project.
_ _ _ . . _ . Modeled air quality impact from hypothetical source
Eq. 2 Project Impact = Project emission rate X
Modeled emission rate from hypothetical source
For simplicity in these examples, nearby and background levels are represented by the design
value from a representative monitor. In this situation, the cumulative assessment would include
the sum of equation 2 and that monitored design value.
Eq. 3 Projected Design Value with Project = Project Impact (Eq. 2) + Monitored Design Value
If equation 3 results in an air quality level less that the NAAQS, then there is no NAAQS violation
for which the source could cause or contribute to. However, if equation 3 results in an air
quality level greater than the NAAQS, then the permit applicant should consult with the
reviewing authority to determine the next step in the demonstrating project source impact at
the location of the NAAQS violation. This may necessitate more refined modeling to reconcile
project source impacts and monitored design values to complete the second phase of the
cumulative impact analysis.
The following illustrative examples are intended to show how existing modeling information
may be used in specific permit demonstrations.
Scenario A: Single precursor assessment for O3 and additive secondary PM2.5 impacts
In this scenario, a facility with a proposed increase in emissions of 0 tpy of primary PM2.5, 0 tpy
of VOC, 600 tpy of NOx, and 3,100 tpy of SO2 located in the southeast region.
O3 source impact analysis: The project source is not located in an area with unusual
circumstances regarding complex terrain, proximity to very large sources of either NOx or VOC,
or meteorology. However, the NOx emissions of 600 tpy are larger than the lowest (most
conservative) NOx MERP for 8-hr O3 in the southeast region (i.e., 170 tpy). Thus, even though
the project source's surrounding environment does not raise an obvious regional feature that
would influence downwind O3 impacts, it is likely more appropriate to use a hypothetical source
55
-------
in the same region or other appropriate geographic area for comparison. In practice, EPA
recommends that the permit applicant consult with the appropriate reviewing authority to
determine the relevant hypothetical source and geographic area from which to select
representative MERP values.
A comparable hypothetical source is identified to be representative of the project (e.g.,
southeast region source located in Tallapoosa County, Alabama with elevated emissions
release). Since multiple hypothetical sources were modeled at this location with an elevated
release, the source with the lowest MERP was selected for comparison with the project source.
The project source does not emit VOC so a MERP approach addressing only NOx emission is
sufficient in this example. For this example, equation 2 was used to estimate air quality impacts
using the hypothetical source information rather than equation 1 because this form of the Tier
1 demonstration approach more clearly fits into the subsequent cumulative assessment.
Project source impact (ppb) = 600 tpy * (1.528 ppb / 500 tpy) = 1.83 ppb
In this case, based on EPA modeling results for a representative hypothetical source, air quality
impacts of O3 from this project source would be expected to exceed the EPA recommended 8-
hour O3 SIL.
O3 cumulative impact analysis: For the cumulative impact analysis, the impact estimated with
equation 2 in the source impact analysis was used with an estimate of nearby source impacts
and background O3, which was a nearby monitor design value. The representative monitor near
the project source has a design value of 65 ppb.
Projected Design Value with Project Source (ppb) = 1.83 ppb + 65 ppb = 66.83 ppb
When the source impact is combined with the nearby monitor design value using equation 3,
the projected value is below the level of the O3 NAAQS of 70 ppb.
PM?.s source impact analysis: The project source is not located in an area with unusual
circumstances regarding complex terrain, proximity to very large sources of pollutants that
impact atmospheric chemistry (i.e., NOx, SO2, NH3) or meteorology. Both the NOx and SO2
emissions are below the lowest (most conservative) daily and annual PM2.5 MERP values of any
source modeled in the southeast region. The SO2 emissions are not very far below the most
conservative MERP relating SO2 emissions to daily PM2.5 impacts. Thus, for simplicity in this
example, even though the project source's surrounding environment does not raise an obvious
regional feature that would influence downwind secondary PM2.5 impacts, it is likely more
appropriate to use a specific hypothetical source in the same region or other appropriate
geographic area for comparison. In practice, EPA recommends that the permit applicant consult
with the appropriate reviewing authority to determine the relevant hypothetical source and
geographic area from which to select representative MERP values.
A comparable hypothetical source is identified to be representative of this project (e.g.,
56
-------
southeast region source located in Tallapoosa County, Alabama with elevated emissions
release) and has a source derived NOx MERP for 24-hr PM2.5 of 12,686 tpy and SO2 MERP for 24-
hr PM2.5 of 2,593 tpy. This hypothetical source has a derived NOx MERP for annual PM2.5 of
116,399 tpy and SO2 MERP for annual PM2.5 of 21,106 tpy.
For this example, EPA recommends that the NOx and SO2 precursor impacts on both daily and
annual average PM2.5 are considered together to determine if the project source's air quality
impact of PM2.5 would exceed the PM2.5 SILs. In this case, the project source's emissions
increase can be expressed as a percent of the MERP for each precursor and then the
percentages can be summed. A value less than 100% indicates that the EPA recommended
PM2.5 SILs would not be exceeded when considering the combined impacts of these precursors
on daily and annual PM2.5.
Example calculation based on equation 1 for additive precursor impacts on daily PM2.5:
(600 tpy NOx from source/12,686 tpy NOx daily PM2.5 MERP) + (3,100 tpy SO2 from
source/2,593 tpy S02 daily PM2.5 MERP) = .05 + 1.20 = 1.21 * 100 = 121%
Example calculation based on equation 1 for additive precursor impacts on annual PM2.5:
(600 tpy NOx from source/116,399 tpy NOx annual PM2.5 MERP) + (3,100 tpy SO2 from
source/21,106 tpy S02 annual PM2.5 MERP) = .005 + .147 = .15 * 100 = 15%
A value less than 100% indicates that the EPA recommended PM2.5 SIL would not be exceeded
when considering the combined impacts of these precursors on daily or annual PM2.5. Thus, in
this case, the air quality impacts of PM2.5 from precursor emissions of NOx and SO2 from this
source would be expected to be above the daily PM2.5 SIL and less than the annual PM2.5 SIL.
PM?.s cumulative impact analysis: For the cumulative impact analysis on daily PM2.5 impacts,
equation 2 is used with the modeled emissions rates and air quality impact information from
this representative hypothetical source with an elevated release. Since multiple hypothetical
sources were modeled at this location with an elevated release the source with the lowest
MERP was selected for comparison with the project source.
Source nitrate impact (|ag/m3) = 600 tpy * (0.047 |-ig/m3 / 500 tpy) = 0.056 |-ig/m3
Source sulfate impact (|ag/m3) = 3,100 tpy * (0.891 |-ig/m3 / 3,000 tpy) = 0.921 |-ig/m3
A representative monitor near the project source has a 24-hour PM2.5 design value of 14 |-ig/m3.
Projected Design Value with Project Source (|ag/m3) = 0.056 |-ig/m3 + 0.921 |-ig/m3 + 14
l-ig/m3 = 14.98 |-ig/m3
When the source impact is combined with the nearby monitor design value using equation 3,
the projected value is below the level of the daily PM2.5 NAAQS of 35 |-ig/m3.
57
-------
Scenario B: Additive demonstration for O3 and secondary PM2.5 with primary PM2.5 impacts
In this scenario, a facility with a proposed increase in emissions of 500 tpy of primary PM2.5, 62
tpy of VOC, 920 tpy of NOx, and 259 tpy of SO2 located in the western region.
O3 source impact analysis: The project source is not located in an area with unusual
circumstances regarding complex terrain, proximity to very large sources of either NOx or VOC,
or meteorology. However, the NOx emissions of 920 tpy are larger than the lowest (most
conservative) NOx MERP for 8-hr O3 in the western region of the U.S. Thus, even though the
project source's surrounding environment does not raise an obvious regional feature that
would influence downwind O3 impacts, it is likely more appropriate to use a specific
hypothetical source in the same region or other appropriate geographic area for comparison. In
practice, EPA recommends that the permit applicant consult with the appropriate reviewing
authority to determine the relevant hypothetical source and geographic area from which to
select representative MERP values.
A comparable hypothetical source is identified to be representative of this source (e.g., western
(Rockies) region in Iron County, Utah with elevated release). Here, equation 1 is used with the
modeled emissions rates and air quality impact information from the selected comparable
source. Since multiple hypothetical sources were modeled at this location with an elevated
release the source with the MERP with the most similar emission rate was selected for
comparison with the project source, i.e.,
1. NOx MERP for selected representative hypothetical source (tpy) = 1.0 ppb *
(1000 tpy / 1.314 ppb) = 761 tpy
2. VOC MERP for selected representative hypothetical source (tpy) = 1.0 ppb * (500
tpy / 0.0407 ppb) = 12,275 tpy
3. Combining impacts from both NOx and VOC: (920/761 + 62/12,275) * 100 =
121%
In this case, based on modeling results for a representative hypothetical source, the project
source emissions are greater than the calculated 8-hr O3 MERP such that air quality impacts of
O3 from this source are expected to exceed the EPA recommended 8-hour O3 SIL.
O3 cumulative impact analysis: For the cumulative impact analysis, equation 2 is used with the
modeled emissions rates and air quality impact information from this representative
hypothetical source with an elevated release. Since multiple hypothetical sources were
modeled at this location with an elevated release the source with the most similar emission
rate was selected for comparison with the project source.
Source impact from NOx (ppb) = 920 tpy * (1.314 ppb / 1000 tpy) = 1.208 ppb
Source impact from VOC (ppb) = 62 tpy * (0.0407 ppb / 500 tpy) = 0.005 ppb
58
-------
A representative monitor near the project source has a design value of 62 ppb.
Projected Design Value with Project Source (ppb) = 1.213 ppb + 62 ppb = 63.213 ppb
When the source impact is combined with the nearby monitor design value using equation 3,
the projected value is below the level of the O3 NAAQS.
PIVb.s source impact analysis: The project source is not located in an area with unusual
circumstances regarding complex terrain, proximity to very large sources of pollutants that
impact atmospheric chemistry (i.e., NOx, SO2, NH3) or meteorology. However, the NOx
emissions of 920 are marginally below the lowest (most conservative) daily and annual PM2.5
MERP value of any source modeled in the continental U.S., while the SO2 emissions of 259 tpy
are slightly higher than the lowest daily PM2.5 MERP value of any source modeled in the
western U.S. region.
Thus, for simplicity in this example, even though the project source's surrounding environment
does not raise an obvious regional feature that would influence downwind secondary PM2.5
impacts, it is likely more appropriate to use a hypothetical source in the same region or other
appropriate geographic area for comparison. In practice, EPA recommends that the permit
applicant consult with the appropriate reviewing authority to determine the relevant
hypothetical source and geographic area from which to select representative MERP values.
A hypothetical source is identified to be representative of this source (e.g., western (Rockies)
region in Iron County, Utah). Since multiple hypothetical sources were modeled at this location
with an elevated release the source with the lowest MERP was selected for comparison with
the project source. The 1,000 tpy MERP was chosen for NOx and the 500 tpy MERP for SO2
impacts. Both reflect elevated emissions release.
For this example, EPA recommends that the NOx and SO2 precursor contributions to both daily
and annual average PM2.5 are considered together to determine if the project source's air
quality impact of PM2.5 would exceed the EPA recommended PM2.5 SILs. In this case, the project
source's emissions increase can be expressed as a percent of the MERP for each precursor and
then the percentages can be summed.
Example calculation for additive precursor impacts on daily PM2.5:
(920 tpy NOx from source/25,754 tpy NOx daily PM2.5 MERP) + (259 tpy SO2 from
source/6,386 tpy S02 daily PM2.5 MERP) = 0.04 + 0.04 = 0.08 * 100 = 8%
Example calculation for additive precursor impacts on annual PM2.5:
(920 tpy NOx from source/166,670 tpy NOx daily PM2.5 MERP) + (259 tpy SO2 from
source/33,561 tpy S02 daily PM2.5 MERP) = 0.0055+ 0.0077 = 0.013 * 100 = 1.3%
59
-------
The emissions rates for both NOx and SO2 are much lower than the daily and annual PM2.5
MERP based on the modeling results for a representative hypothetical source. However, for
purposes of illustration in this hypothetical example, an assumption is made that primary PM2.5
modeling with AERMOD (daily impact assumed to be 1.8 |ag/m3and annual impact assumed to
be 0.02 |-ig/m3) showed an exceedance of the EPA recommended daily (but not annual) PM2.5
SIL so that a cumulative impact analysis example is presented below for the daily form of the
NAAQS. Note that no AERMOD simulations were done to relate primary PM2.5 emissions and
downwind impacts; the levels of impact used here are purely to support this illustrative
example. When considering primary and secondary impacts for the annual form of the NAAQS,
the source's impact would be expected to be less than the EPA recommended PM2.5 SIL.
PIVb.s cumulative impact analysis: For the cumulative impact analysis, equation 2 is used with
the modeled emissions rates and air quality impact information from this representative
hypothetical source with an elevated release.
Source nitrate impact (|ag/m3) = 920 tpy * (0.047 |-ig/m3 / 1000 tpy) = 0.043 |-ig/m3
Source sulfate impact (|ag/m3) = 259 tpy * (0.094 |-ig/m3 / 500 tpy) = 0.049 |-ig/m3
A representative monitor near the project source has a daily PM2.5 design value of 11 |-ig/m3. A
hypothetical downwind primary PM2.5 impact from other analysis for this source was
determined to be 1.8 |-ig/m3, which is included in the CIA together with the secondary impact
analysis.
Projected Design Value with Project Source (|ag/m3) = 0.043 |-ig/m3 + 0.049 |-ig/m3 + 11
l-ig/m3 + 1.8 |-ig/m3 = 12.89 |-ig/m3
When the project source primary impact (from AERMOD) and secondary impacts (from MERP
equation) are combined with the nearby monitor design value using equation 3, the projected
value is below the level of the daily PM2.5 NAAQS.
60
-------
5. References
Baker, K.R., Foley, K.M., 2011. A nonlinear regression model estimating single source
concentrations of primary and secondarily formed PM2.5. Atmospheric Environment 45, 3758-
3767.
Baker, K.R., Kelly, J.T., 2014. Single source impacts estimated with photochemical model source
sensitivity and apportionment approaches. Atmospheric Environment 96, 266-274.
Baker, K.R., Kelly, J.T., Fox, T., 2013. Estimating second pollutant impacts from single sources
(control #27). http://aqmodels.awma.ore/conference-proceedings/.
Baker, K.R., Kotchenruther, R.A., Hudman, R.C., 2016. Estimating ozone and secondary PM 2.5
impacts from hypothetical single source emissions in the central and eastern United States.
Atmospheric Pollution Research 7,122-133.
Baker, K.R., Woody, M.C., 2017. Assessing Model Characterization of Single Source Secondary
Pollutant Impacts Using 2013 SENEX Field Study Measurements. Environmental Science &
Technology 51, 3833-3842.
Bergin, M.S., Russell, A.G., Odman, M.T., Cohan, D.S., Chameldes, W.L., 2008. Single-Source
Impact Analysis Using Three-Dimensional Air Quality Models. Journal of the Air & Waste
Management Association 58, 1351-1359.
Byun, D., Schere, K.L., 2006. Review of the governing equations, computational algorithms, and
other components of the models-3 Community Multiscale Air Quality (CMAQ) modeling system.
Applied Mechanics Reviews 59, 51-77.
Chang, C.-Y., Faust, E., Hou, X., Lee, P., Kim, H.C., Hedquist, B.C., Liao, K.-J., 2016. Investigating
ambient ozone formation regimes in neighboring cities of shale plays in the Northeast United
States using photochemical modeling and satellite retrievals. Atmospheric Environment 142,
152-170.
Chen, J., Lu, J., Avise, J.C., DaMassa, J.A., Kleeman, M.J., Kaduwela, A.P., 2014. Seasonal
modeling of PM 2.5 in California's San Joaquin Valley. Atmospheric Environment 92, 182-190.
Civerolo, K., Hogrefe, C., Zalewsky, E., Hao, W., Sistla, G., Lynn, B., Rosenzweig, C., Kinney, P.L.,
2010. Evaluation of an 18-year CMAQ simulation: Seasonal variations and long-term temporal
changes in sulfate and nitrate. Atmospheric environment 44, 3745-3752.
Cohan, D.S., Napelenok, S.L., 2011. Air quality response modeling for decision support.
Atmosphere 2, 407-425.
Duncan, B.N., Yoshida, Y., Olson, J.R., Sillman, S., Martin, R.V., Lamsal, L., Hu, Y., Pickering, K.E.,
Retscher, C., Allen, D.J., 2010. Application of OMI observations to a space-based indicator of
61
-------
NOx and VOC controls on surface ozone formation. Atmospheric Environment 44, 2213-2223.
Dunker, A.M., Yarwood, G., Ortmann, J.P., Wilson, G.M., 2002. The decoupled direct method for
sensitivity analysis in a three-dimensional air quality model - Implementation, accuracy, and
efficiency. Environmental Science & Technology 36, 2965-2976.
ENVIRON, 2012a. Documentation of the Evaluation of CALPUFF and Other Long Range
Transport Models using Tracer Field Experiment Data, EPA Contract No: EP-D-07-102. February
2012. 06-20443M4.
ENVIRON, 2012b. Evaluation of chemical dispersion models using atmospheric plume
measurements from field experiments, EPA Contract No: EP-D-07-102. September 2012. 06-
20443M6.
Jin, X., Fiore, A.M., Murray, L.T., Valin, L.C., Lamsal, L.N., Duncan, B., Folkert Boersma, K., De
Smedt, I., Abad, G.G., Chance, K., 2017. Evaluating a Space-Based Indicator of Surface Ozone-
NOx-VOC Sensitivity Over Midlatitude Source Regions and Application to Decadal Trends.
Journal of Geophysical Research: Atmospheres 122.
Kelly, J.T., Baker, K.R., Napelenok, S.L., Roselle, S.J., 2015. Examining single-source secondary
impacts estimated from brute-force, decoupled direct method, and advanced plume treatment
approaches. Atmospheric Environment 111, 10-19.
Kwok, R., Baker, K., Napelenok, S., Tonnesen, G., 2015. Photochemical grid model
implementation of VOC, NO x, and O 3 source apportionment. Geoscientific Model
Development 8, 99-114.
Kwok, R., Napelenok, S., Baker, K., 2013. Implementation and evaluation of PM2.5 source
contribution analysis in a photochemical model. Atmospheric Environment 80, 398-407.
Pun, B.K., Seigneur, C., Bailey, E.M., Gautney, L.L., Douglas, S.G., Haney, J.L., Kumar, N., 2007.
Response of atmospheric particulate matter to changes in precursor emissions: A comparison
of three air quality models. Environmental science & technology 42, 831-837.
Ramboll ENVIRON, 2016. User's Guide Comprehensive Air Quality Model with Extensions
version 6, www.camx.com. ENVIRON International Corporation, Novato.
Russell, A.G., 2008. EPA Supersites program-related emissions-based particulate matter
modeling: initial applications and advances. Journal of the Air & Waste Management
Association 58, 289-302.
Seinfeld, J.H., Pandis, S.N., 2012. Atmospheric chemistry and physics: from air pollution to
climate change. John Wiley & Sons.
Stockwell, W.R., Watson, J.G., Robinson, N.F., Steiner, W., Sylte, W.W., 2000. The ammonium
nitrate particle equivalent of NO x emissions for wintertime conditions in Central California's
62
-------
San Joaquin Valley. Atmospheric Environment 34, 4711-4717.
Tesche, T., Morris, R., Tonnesen, G., McNally, D., Boylan, J., Brewer, P., 2006. CMAQ/CAMx
annual 2002 performance evaluation over the eastern US. Atmospheric Environment 40, 4906-
4919.
U.S. Environmental Protection Agency, 2016a. Guidance on the use of models for assessing the
impacts from single sources on secondarily formed pollutants ozone and PM2.5. EPA 454/R-16-
005. https://www3.epa.gov/ttn/scram/appendix w/2016/EPA~454 R-16-005.pdf.
U.S. Environmental Protection Agency, 2016b. Interagency Workgroup on Air Quality Modeling
(IWAQM) Phase 3 Summary Report: Near-Field Single Source Secondary Impacts. EPA-454/R-
16-003.
U.S. Environmental Protection Agency, 2016c. Interagency Workgroup on Air Quality Modeling
Phase 3 Summary Report: Long-range Transport and Air Quality Related Values (AQRVs). EPA-
454/R-16-002. June 2016.
U.S. Environmental Protection Agency, 2016d. Technical Support Document (TSD) for AERMOD-
Based Assessments of Long-Range Transport Impacts for Primary Pollutants.
https://www3.epa.gov/ttn/scram/appendix w/2016/AppW LRT TSD.pdf.
U.S. Environmental Protection Agency, 2017a. Revisions to the Guideline on Air Quality Models:
Enhancements to the AERMOD Dispersion Modeling System and Incorporation of Approaches
to Address Ozone and Fine Particulate Matter. 40 CFR Part 51. Federal Register. Vol. 82, No. 10,
January 17, 2017.
U.S. Environmental Protection Agency, 2017b. Supplemental Information for Ozone Advance
Areas Based On Pre-Existing National Modeling Analyses.
https://www.ep3.gov/sites/production/files/2017~
05/documents/national modeling.advance.may 2017.pdf.
U.S. Environmental Protection Agency, 2018. Guidance on Significant Impact Levels for Ozone
and Fine Particles in the Prevention of Significant Deterioration Permitting Program.
https://www.epa.gov/sites/production/files/2Q18-
04/documents/sils policy guidance document final signs t , u!.
Zhou, W., Cohan, D.S., Pinder, R.W., Neuman, J.A., Holloway, J.S., Peischl, J., Ryerson, T.B.,
Nowak, J.B., Flocke, F., Zheng, W.G., 2012. Observation and modeling of the evolution of Texas
power plant plumes. Atmospheric Chemistry and Physics 12, 455-468.
63
-------
Appendix A. Hypothetical Sources Included in the EPA's Modeling
Assessment
Table A-l. Complete list of EPA modeled hypothetical sources presented in this document. A list
of emission rates and stack height combinations modeled for each domain are provided in
Table A-2. The "Max Nearby Urban (%)" column provides the highest percentage urban
landcover in any grid cell near (within 50 km) the source. Source locations are shown in Figures
A-l, A-2, A-3, and A-4.
Max Max
Nearby Nearby
Terrain Urban
FIPS
State
County
Domain
Source
Latitude
Longitude
(m)
(%)
1001
Alabama
Autauga
12EUS2
4
32.522
-86.550
179
25
1123
Alabama
Tallapoosa
12EUS3
19
32.848
-85.809
306
10
4005
Arizona
Coconino
12US2
36
35.428
-111.270
2483
7.4
4007
Arizona
Gila
12WUS1
14
33.469
-110.789
1592
4.3
4012
Arizona
La Paz
12WUS1
17
33.400
-113.408
757
0.9
5119
Arkansas
Pulaski
12EUS2
13
34.724
-92.275
235
32.2
6029
California
Kern
12WUS1
26
35.356
-119.508
1195
49.1
6037
California
Los Angeles
12WUS1
21
34.696
-118.414
1528
39.9
6047
California
Merced
12WUS1
25
37.274
-120.708
547
14.6
6063
California
Plumas
12WUS1
24
39.920
-121.263
1773
17.5
6107
California
Tulare
12WUS1
20
36.324
-119.404
566
18.1
8011
Colorado
Bent
12WUS1
4
37.685
-102.994
1698
1.4
8069
Colorado
Larimer
12WUS1
8
40.841
-105.826
3288
0.5
8093
Colorado
Park
12US2
31
38.919
-105.990
3535
2.2
8109
Colorado
Saguache
12WUS1
9
37.965
-106.234
3374
2.7
8109
Colorado
Saguache
12WUS1
9
37.965
-106.234
3374
2.7
8123
Colorado
Weld
12WUS1
3
40.621
-104.037
1609
6.2
12005
Florida
Bay
12EUS2
5
30.269
-85.700
55
9.8
17021
Illinois
Christian
12US2
16
39.509
-89.092
209
11.6
17145
Illinois
Perry
12EUS2
7
38.078
-89.547
194
6.8
17155
Illinois
Putnam
12EUS2
6
41.200
-89.446
243
16.4
17177
Illinois
Stephenson
12US2
15
42.455
-89.606
296
14.4
18011
Indiana
Boone
12US2
11
40.009
-86.574
290
47.3
18037
Indiana
Dubois
12EUS2
2
38.255
-86.724
224
4.4
18053
Indiana
Grant
12EUS3
17
40.623
-85.589
285
10.3
18127
Indiana
Porter
12EUS2
1
41.380
-87.185
235
52.3
19027
Iowa
Carroll
12US2
20
42.092
-94.693
435
3.9
19095
Iowa
Iowa
12EUS2
11
41.674
-92.060
295
17.3
20091
Kansas
Johnson
12EUS2
17
38.746
-94.949
325
38.8
20109
Kansas
Logan
12US2
26
38.909
-101.173
1121
1.6
20155
Kansas
Reno
12EUS2
22
38.121
-97.899
542
12.7
64
-------
21009
Kentucky
Barren
12EUS3
18
36.828
-85.830
269
4.5
21187
Kentucky
Owen
12US2
33
38.536
-84.707
279
7.4
22001
Louisiana
Acadia
12EUS2
15
30.241
-92.616
16
6.5
22061
Louisiana
Lincoln
12EUS2
14
32.476
-92.711
97
5.8
22071
Louisiana
Orleans
12EUS2
10
30.092
-89.879
10
50.4
23003
Maine
Aroostook
12EUS3
1
46.772
-67.850
365
4.6
23031
Maine
York
12EUS3
2
43.367
-70.580
237
13.3
25011
Massachusetts
Franklin
12EUS3
4
42.582
-72.459
583
21.6
25021
Massachusetts
Norfolk
12EUS3
3
42.139
-71.234
224
60
26099
Michigan
Macomb
12EUS3
11
42.822
-82.872
317
63.9
26103
Michigan
Marquette
12EUS3
15
46.570
-87.395
518
4
26117
Michigan
Montcalm
12EUS3
16
43.319
-85.368
309
42.8
26129
Michigan
Ogemaw
12US2
5
44.164
-84.069
382
4.4
26159
Michigan
Van Buren
12US2
10
42.410
-86.027
273
25.3
27037
Minnesota
Dakota
12US2
19
44.785
-93.311
339
52.4
27137
Minnesota
St Louis
12US2
13
47.913
-92.331
485
2.8
27159
Minnesota
Wadena
12US2
18
46.401
-95.086
464
2.2
28129
Mississippi
Smith
12EUS2
9
32.177
-89.345
142
2.3
29029
Missouri
Camden
12EUS2
12
38.014
-93.006
378
6.2
29155
Missouri
Pemiscot
12US2
17
36.223
-89.851
104
5.1
29177
Missouri
Ray
12US2
21
39.504
-94.135
305
39
30013
Montana
Cascade
12US2
28
47.367
-111.447
1803
18.1
30075
Montana
Powder River
12WUS1
7
45.299
-105.895
1238
0.6
30083
Montana
Richland
12WUS1
6
47.367
-104.447
862
2.3
30111
Montana
Yellowstone
12WUS1
11
45.786
-108.207
1641
22.2
31001
Nebraska
Adams
12EUS2
21
40.673
-98.327
655
18.2
31055
Nebraska
Douglas
12EUS2
16
41.364
-96.155
424
43.3
31101
Nebraska
Keith
12US2
25
41.247
-102.006
1197
2.1
32001
Nevada
Churchill
12WUS1
19
39.941
-118.748
1599
9.2
34041
New Jersey
Warren
12US2
2
41.017
-75.000
577
31.2
35031
New Mexico
Mc Kinley
12US2
32
35.368
-107.382
2577
3.6
35035
New Mexico
Otero
12WUS1
10
32.757
-105.767
2618
4.4
36005
New York
Bronx
12EUS3
5
40.819
-73.909
273
75.4
36019
New York
Clinton
12US2
1
44.477
-73.836
889
3.2
36051
New York
Livingston
12EUS3
7
42.877
-77.603
532
34
37009
North Carolina
Ashe
12EUS3
13
36.301
-81.374
1168
6.9
37109
North Carolina
Lincoln
12US2
8
35.439
-81.154
457
32.1
37127
North Carolina
Nash
12US2
4
35.922
-78.187
123
22.1
38057
North Dakota
Mercer
12WUS1
1
47.287
-101.879
719
1.8
38059
North Dakota
Morton
12WUS1
2
46.861
-101.925
799
1
39103
Ohio
Medina
12US2
6
41.238
-81.813
344
51.7
39157
Ohio
Tuscarawas
12EUS3
12
40.541
-81.396
356
26.9
40017
Oklahoma
Canadian
12EUS2
23
35.463
-97.913
473
43.1
65
-------
40101
Oklahoma
Muskogee
12EUS2
18
35.751
-95.507
236
30.4
40127
Oklahoma
Pushmataha
12US2
22
34.390
-95.567
294
2.5
40149
Oklahoma
Washita
12US2
27
35.311
-99.187
662
4.4
41049
Oregon
Morrow
12WUS1
18
45.790
-119.475
894
8.2
42001
Pennsylvania
Adams
12EUS3
8
40.009
-77.111
364
26.9
42029
Pennsylvania
Chester
12US2
3
39.940
-75.822
188
32.2
45005
South Carolina
Allendale
12EUS3
14
32.973
-81.407
84
2.2
45051
South Carolina
Horry
12EUS3
10
34.083
-79.187
33
7.1
46055
South Dakota
Haakon
12US2
23
44.287
-101.879
842
1.4
46097
South Dakota
Miner
12US2
24
43.861
-97.425
535
5.4
47001
Tennessee
Anderson
12US2
12
36.079
-84.149
611
25.4
47055
Tennessee
Giles
12EUS2
3
35.291
-86.897
286
8.4
47157
Tennessee
Shelby
12EUS2
8
35.124
-90.002
117
42.4
48187
Texas
Guadalupe
12EUS2
25
29.553
-97.991
349
43.8
48201
Texas
Harris
12EUS2
20
29.592
-95.418
41
64.7
48213
Texas
Henderson
12EUS2
19
32.314
-95.556
155
27.6
48367
Texas
Parker
12EUS2
24
32.610
-97.736
384
35.7
48445
Texas
Terry
12WUS1
5
33.369
-102.146
1112
31.9
49013
Utah
Duchesne
12WUS1
12
40.407
-110.618
3395
0.9
49015
Utah
Emery
12US2
35
38.804
-110.630
2090
0.6
49021
Utah
Iron
12WUS1
16
37.608
-113.092
2870
5.5
49037
Utah
San Juan
12WUS1
13
37.905
-109.899
2450
0.2
49049
Utah
Utah
12WUS1
15
40.110
-111.936
2235
21.7
51053
Virginia
Dinwiddie
12EUS3
9
36.919
-77.707
133
9
53039
Washington
Klickitat
12WUS1
23
45.938
-121.191
1699
4.9
53057
Washington
Skagit
12WUS1
22
48.466
-122.559
497
9.6
54017
West Virginia
Doddridge
12US2
7
39.299
-80.633
454
10.4
55107
Wisconsin
Rusk
12US2
14
45.596
-90.768
482
2.3
55115
Wisconsin
Shawano
12US2
9
44.733
-88.263
309
32.2
56001
Wyoming
Albany
12US2
30
41.829
-105.857
2898
0.3
56005
Wyoming
Campbell
12US2
29
44.299
-105.895
1532
8.1
56023
Wyoming
Lincoln
12US2
34
41.905
-110.326
2585
1.3
66
-------
Table A-2. A list of emission rates and stack release height combinations modeled for each
domain. A complete list of hypothetical sources in each domain are provided in Table A-l.
Figures showing the location of specific sources by domain are provided in Figures A1-A4.
NAAQS & Precursors Modeled
# hypothetical
sources
Emission
Geographic
within the
Release
Rate
Daily
Annual
Region
region
Type
(tpy)
8-hr 03
PM2.5
PM2.5
12EUS3
18
H
3000
NOX, VOC
NOX, S02
NOX, S02
(eastern US)
18
H
1000
NOX, VOC
NOX, S02
NOX, S02
18
H
500
NOX, VOC
NOX, S02
NOX, S02
18
L
500
NOX, VOC
NOX, S02
NOX, S02
12EUS2
25
H
3000
NOX, VOC
NOX, S02
NOX, S02
(central US)
25
H
1000
NOX, VOC
NOX, S02
NOX, S02
25
L
1000
VOC
NOX, S02
NOX, S02
25
H
500
NOX
NOX, S02
NOX, S02
25
L
500
NOX, VOC
NOX, S02
NOX, S02
12WUS1
26
H
3000
NOX, VOC
NOX, S02
NOX, S02
(western US)
26
H
1000
NOX, VOC
NOX, S02
NOX, S02
26
H
500
NOX, VOC
NOX, S02
NOX, S02
26
L
500
NOX, VOC
NOX, S02
NOX, S02
12US2
36
H
1000
NOX
NOX, S02
NOX, S02
(contiguous US)
36
H
500
NOX
NOX, S02
NOX, S02
36
L
500
NOX, VOC
NOX, S02
NOX, S02
67
-------
Figure A-l. Hypothetical source locations for the eastern U.S. (12EUS3) domain.
Model Domain and Hypothetical Sources
20OQ
-------
Figure A-2. Hypothetical source locations for the central U.S. (12EUS2) domain.
Model Domain and Hypothetical Sources
O
O
C
.Q
H
lT
5
To
E=
Of
Q_
O
-1000
69
-------
Figure A-3. Hypothetical source locations for the western U.S. (12WUS1) domain.
Model Domain and Hypothetical Sources
13 01JD
¦2000
-------
re A-4. Hypothetical source locations for the contiguous U.S. (12US2) domain.
80
40
230
!0O
.160
2 50
40
120
!2n
220
-2000
-1000
0
1000
2000
71
-------
United States Office of Air Quality Planning and Standards Publication No. EPA-454/R-19-003
Environmental Protection Air Quality Assessment Division April 2019
Agency Research Triangle Park, NC
------- |