Technical Support Document (TSD)
for the Final Revised CSAPR Update for the 2008 Ozone NAAQS

Docket ID No. EPA-HQ-OAR-2020-0272

Ozone Transport Policy Analysis
Final Rule TSD

U.S. Environmental Protection Agency
Office of Air and Radiation
March 2021

1


-------
The analysis presented in this document supports the EPA's final Revised Cross-State Air
Pollution Rule Update for the 2008 Ozone NAAQS (Revised CSAPR Update). This TSD
includes analysis to help quantify upwind state emissions that significantly contribute to
nonattainment or interfere with maintenance of the 2008 ozone NAAQS in downwind states and
quantification of emission budgets (i.e., limits on emissions) and the resulting effects on air
quality. The analysis is described in Sections VI and VII of the preamble to the rule. This TSD
also broadly describes how the EPA used historical data and the Integrated Planning Model
(IPM) to inform air quality modeling, budget setting, and policy analysis aspects of this rule.
Finally, this TSD includes an assessment on the effects of ozone concentrations on forest health.
This TSD is organized as follows:

A.	Background on EPA's Analysis to Quantify Emissions that Significantly Contribute to
Nonattainment or Interfere with Maintenance of the 2008 Ozone NAAQS

B.	Using Engineering Analytics and Integrated Planning Model (IPM) to Assess Air Quality
Modeling, EGU NOx Mitigation Strategies, and Policy Impacts

C.	Calculating Budgets from Historical Data and IPM Analysis

1.	Calculating 2021-2025 engineering baseline for NOx (from adjusted historical data)

2.	Estimating impacts of combustion and post combustion controls on state emission budgets

3.	Estimating emission reduction potential from generation shifting

4.	Variability limits and RIA scenarios

D.	Analysis of Air Quality Responses to Emission Changes Using an Ozone Air Quality
Assessment Tool (AQAT)

1.	Introduction: development of the ozone AQAT

2.	Details on the construction of the ozone AQAT

3.	Description of analytical results

4.	Comparison between the air quality assessment tool estimates

E.	Observations on Cost and Air Quality Factors for 2024

F.	Assessment of the Effects of Ozone on Forest Health

2


-------
A. Background on EPA's Analysis to Quantify Emissions that Significantly Contribute to
Nonattainment or Interfere with Maintenance of the 2008 Ozone NAAQS

In the preamble, EPA describes the 4-Step Good Neighbor Framework that it is applying
to identify upwind states' emissions that significantly contribute to nonattainment or interfere
with maintenance with respect to the 2008 ozone National Ambient Air Quality Standard
(NAAQS) in other states and to implement appropriate emission reductions. This framework was
used in the original CSAPR rulemaking to address interstate transport with respect to the 1997
ozone NAAQS and the 1997 and 2006 PM2.5 NAAQS and was also used in the 2016 CSAPR
Update to address interstate transport with respect to the 2008 ozone NAAQS.

The first step of the Good Neighbor Framework uses air quality analysis to identify
nonattainment and maintenance receptors for the 2008 ozone NAAQS. The second step of the
framework uses further air quality analysis to identify upwind states whose ozone pollution
contributions to these monitoring sites meet or exceed a specified threshold and therefore merit
further analysis. See section V of the preamble for details on applying these steps with respect to
interstate emissions transport for the 2008 ozone NAAQS.

The third step in the Good Neighbor Framework quantifies upwind state emissions that
significantly contribute to nonattainment or interfere with maintenance of the 2008 ozone
NAAQS at the downwind receptors, and identifies the electricity generating unit (EGU) NOx
emission budgets and/or non-EGU emissions reduction for each state that represent the reduction
of these emissions levels. See section VI of the preamble with respect to interstate emissions
transport for the 2008 ozone NAAQS. Finally, the fourth step of the Good Neighbor Framework
implements the emission budgets in each state through the CSAPR NOx ozone season allowance
trading program or other enforceable mechanism. See section VII of the preamble for details on
implementation for this rule.

This TSD primarily addresses step three of the Good Neighbor Framework related to
EGU emissions as well as to the effects on air quality of both EGU and non-EGU emissions
reductions. In order to establish EGU NOx emissions budgets for each linked upwind state, EPA
first identifies various possible uniform levels of NOx control stringency based on available
EGU NOx control strategies and represented by cost thresholds.1 The EGU emission reductions
pertaining to each level of control stringency are derived using historical data, engineering
analyses, and EPA's integrated planning model (IPM) for the power sector as described in
sections B and C of this TSD. The adjusted historical data and the model data are combined in
order to quantify a series of potential EGU NOx emission budgets for each linked upwind state at
each level of uniform NOx control stringency. A similar assessment for one scenario was done
for non-EGUs. Next, EPA uses the ozone Air Quality Assessment Tool (AQAT) to estimate the
air quality impacts of the upwind state emissions reductions on downwind ozone pollution levels
for each of the assessed cost threshold levels. Specifically, EPA looks at the magnitude of air
quality improvement at each receptor at each level of control, it also examines whether receptors
change status (shifting from either nonattainment to maintenance, or from maintenance to
attainment), and looks at the individual contributions of each state to each of its receptors. See
section D in this TSD for discussion of the development and use of the ozone AQAT.

Finally, the EPA uses this air quality information within the multi-factor test, along with
NOx reduction potential, cost, and other considerations to select a particular level of uniform

1 See the EGU NOx Mitigation Strategies Final Rule TSD

3


-------
NOx control stringency that addresses each state's significant contribution to nonattainment and
interference with maintenance (see Section VI.D of the preamble for additional information).

In this TSD, EPA assesses the EGU NOx mitigation potential for all states in the
contiguous U.S. EPA assessed the air quality impacts for all monitors in the contiguous U.S.
from emission reductions that met the criteria for developing air quality contribution estimates.
In applying the multi-factor test for purposes of identifying the appropriate level of control, the
EPA evaluated NOx reductions and air quality improvements at the four receptors from the two
home states and the 12 upwind states that were linked to downwind receptors in step two of the
4-Step Good Neighbor Framework. The 12 upwind linked states are listed in Table A-l below.

Table A-l. Upwind States Evaluated in the Multi-factor Test

Ozone Season NOx

Illinois

New Jersey

Indiana

New York

Kentucky

Ohio

Louisiana

Pennsylvania

Maryland

Virginia

Michigan

West Virginia

B.	Using the Engineering Analytics Tool and the Integrated Planning Model (IPM) to
Assess Air Quality Modeling, EGU NOx Mitigation Strategies, and Policy Impacts

Similar to the final CSAPR Update, EPA relied on adjusted historical data (engineering
analytics) and its power sector modeling platform using IPM as part of the process to quantify
significant contribution at step three within the 4-Step Good Neighbor Framework. Historical
data were adjusted through the engineering analytics tool and used along with IPM to analyze the
ozone season NOx emission reductions available from EGUs at various uniform levels of NOx
control stringency, represented by cost per ton, in each upwind state. Finally, IPM was used to
evaluate compliance with the rule and the rule's regulatory control alternatives (i.e., compliance
with the emission budgets, with a more stringent alternative, and with a less stringent
alternative). EPA also used its engineering analytics tool and IPM projections to perform air
quality assessment and sensitivity analysis for steps 1 and 2.

The engineering analytics tool uses the latest historical representative emissions and
operating data reported under 40 CFR part 75 by covered units (which were 2019 ozone-season
data at the time of this analysis).2 It is a tool that builds estimates of future unit-level and state-
level emissions based on exogenous changes to historical heat input and emissions data
reflecting fleet changes known to occur subsequent to the last year of available data. See Section

C.	Calculating Budgets from Historical Data and IPM Analysis for a detailed description of the
engineering analytics tool.

IPM is a multiregional, dynamic, deterministic linear programming model of the U.S.
electric power sector that EPA uses to analyze cost and emissions impacts of environmental

2 As explained in preamble section VII.B, EPA did not use 2020 data as a representative historical year due to the
global COVID-19 pandemic.

4


-------
policies.3 All IPM cases for this rule included representation of the Title IV SO2 cap and trade
program; the NOx SIP Call; the CSAPR and CSAPR Update regional cap and trade programs;
consent decrees and settlements; and state and federal rules as listed in the IPM documentation
referenced above.

Application of the 4-Step Good Neighbor Framework requires robust data collection,
IPM modeling and engineering analytics is time consuming. Rather than freezing all IPM and
engineering analytic data sets at the outset of EPA's analysis for the rule, the EPA allowed for
ongoing improvement of the relied-upon EGU data. As a result, each step of EPA's analysis for
the final rule is informed by the best available data at the time the analysis was conducted.

In the power sector modeling done for this rule, the EPA needed to quantify emissions for
three different analytic purposes. The first purpose was to provide future base case EGU
emissions for input to air quality modeling to identify nonattainment and maintenance receptors
and quantify interstate contribution to inform steps 1 and 2 of the 4-Step Good Neighbor
Framework. This base case incorporated the most important fleet changes and retrofits identified
through comments up to Fall of 2019 using the National Electricity Energy Data System
(NEEDS) EGU inventory, January 8, 2020 version. The version of the NEEDS file reflects EGU
fleet updates through November 2019.4

The second purpose was to construct an illustrative base case and control case to study
the potential cost and reduction potential of different uniform technology scenarios. This set of
cases is referred to as the "Illustrative Cases." These illustrative cases are primarily cost
threshold runs that EPA performed where the agency would first adjust the base case to reflect
the relevant control technologies being considered and would then perform a sensitivity where a
dollar per ton price constraint (e.g., $1,600 per ton) was applied to that adjusted base case to
estimate the additional reductions to be expected from generation shifting at a dollar per ton level
commensurate with the technology operating cost.

The third purpose was to estimate system impacts of the final rule and confirm the impact
of implementing the state emissions budgets in a region-wide trading program. This set of cases
is referred to as the "Final Policy Cases." For the Final Policy Cases, the EPA applied the state
emission budgets and corresponding state and regional caps to the same base case used in the
illustrative cases. EPA also performed a "less stringent" and "more stringent" control scenario
policy case using lower and higher state emission budgets respectively. The "Final Cases" were
used to inform the cost and benefits of this rulemaking, as described in the Regulatory Impact
Analysis, or RIA, for this rule.

Table B-l below summarizes the various IPM runs conducted and Appendix C provides
further details on each of these scenarios.

3	See "Documentation for EPA's Power Sector Modeling Platform v6 using January 2020 Reference Case".
Available at https://www.epa.gov/airmarkets/epas-power-sector-modeling-platform-v6-using-ipm-january-2020-
reference-case.

4	https://www.epa.gov/airmarkets/national-electric-energy-data-system-needs-v6

5


-------
Table B-l. Summary of Sets of Scenarios.



Air Quality Modeling
Base Case

Illustrative Cases

Final Policy Cases

Scenarios Run

Base Case (IPM)

Base Case Sensitivity

(Engineering

Analytics)

Base Case (IPM)

Base Case (Engineering
Analytics)

Uniform Control/Cost
Threshold (IPM and
Engineering Analytics)

Policy Cases (IPM)

What Analysis Each
Set of Runs Informs

Base Case air quality
modeling to identify
nonattainment and
maintenance receptors
and estimate upwind
contributions (steps 1
and 2)

Development of a set of
state emission budgets
for each cost threshold
(step 3)

RIA analysis to
gauge system impacts
when budgets are
implemented through
a trading program
(step 4)

EGU Updates
Captured in Each Set
of Runs

Updates as of
November 2019 for
IPM scenarios, and as
of December 2020 for
Engineering Analytic
Sensitivity

Updates as of lune 30,
2020 for IPM scenarios
and December 2020 for
Engineering Analytic
scenarios

Updates as of
December 2020

For the "Illustrative Case" IPM runs, the EPA modeled the emissions that would occur
within each state in a Base Case. The EPA then designed a series of IPM runs that imposed
increasing cost thresholds representing uniform levels of NOx controls and tabulated those
projected emissions for each state at each cost level. The EPA has referred to these runs as "Cost
Threshold Runs" and these tabulations, when combined with adjusted historical data, as "cost
curves."5 The cost curves report the remaining emissions at each cost threshold after the state
has made emission reductions that are available up to the particular cost threshold analyzed.

In each Cost Threshold run, the EPA applied the applicable ozone-season cost level to all
fossil-fuel-fired EGUs with a capacity greater than 25 MW in all states, though only the
estimates for the four receptors, the two "home states" for those receptors, and the 12 linked

5 These projected state level emissions and heat input for each "cost threshold" run are presented in several formats.
The IPM analysis outputs available in the docket contain a "state emissions" file for each analysis. The file contains
two worksheets. The first is titled "all units" and shows aggregate emissions for all units in the state. The second is
titled "all fossil > 25MW" and shows emissions for a subset of these units that have a capacity greater than 25 MW.
The 2021 emissions and heat input in the "all fossil > 25 MW" worksheet is used to derive the generation shifting
component of the state emission budgets for each upwind state at the cost thresholds.

6


-------
states affect the results in Step 3. As described in the EGU NOx Mitigation Strategies Final Rule
TSD, because of the time required to build advanced pollution controls, the model was prevented
from building any new post-combustion controls, such as SCR or SNCR, before 2025, in
response to the cost thresholds.6 Similarly, the model was not enabled to build incremental new
units in that time frame. In response to the ozone-season NOx cost, the modeling allows turning
on idled existing SCR and SNCR, optimization of existing SCR, shifting generation to lower-
NOx emitting EGUs, and adding or upgrading NOx combustion controls (such as state-of-the-art
low NOx burners (LNB)) in 2021/2022. In this TSD, we refer to state-of-the-art combustion
controls, or SOA CC, generally, as combustion controls, or LNB).

In these scenarios, EPA imposed cost thresholds of $500, $1,600, $1,800, $3,900, $5,800,
$9,600 per ton of ozone season NOx. See Preamble Section VI for a discussion of how the cost
thresholds were determined. Table B-2 below summarizes the reduction measures that are
broadly available at various cost thresholds.

Table B-2. Reduction strategies available to EGUs at each cost threshold.

Cost Threshold ($ per
ton Ozone-Season NOx)

Reduction Options

$500

-Generation shifting

$1,600

-Above option; and

-Retrofitting state-of-the-art combustion controls;
-Optimizing idled SCRs (to 0.08 lb/MMBtu);
-Optimizing operating SCRs (to 0.08 lb/MMBtu);

$1,800

-Above options; and

-Optimizing operating SNCRs7 ($3,900 for optimizing
idled SNCR)

$5,800

-Above options; and

-Installing SNCR on certain coal units lacking post-
combustion retrofit

$9,600

-Same as above options; and

-Installing SCR on certain coal units lacking SCR post-
combustion controls (rather than SNCR).

For both Engineering analytics and IPM:

•	At $500/ton:

o Engineering Analytics - no change.

o IPM - cost of $500 per ton applied to base case for EGUs > 25 MW.

•	At $l,600/ton:

6	IPM results do include certain newly built post-combustion NOx control retrofits in base case modeling, cost
curve runs, and remedy runs. These pre-2020 retrofits do not reflect any controls installed in response to the rule,
but instead represent those that are already announced and/or under construction and expected to be online by 2021,
or controls that were projected to be built in the base case in response to existing consent decree or state rule
requirements.

7	As explained in the preamble section VLB, EPA notes that this technology becomes widely available at $1,800 per
ton. For purposes of assessing generation shifting available at this technology level's commensurate cost, EPA
relies on its $1,600 per ton IPM analysis.

7


-------
o Engineering Analytics - If 2019 adjusted baseline rate was greater than 0.08
lb/MMBtu for SCR controlled units, that rate and corresponding emissions were
adjusted down to 0.08 lb/MMBtu starting in 2021; for units with LNB upgrade
potential and an adjusted historical rate greater than 0.199 lb/MMBtu, their rates
were adjusted downwards to 0.199 lb/MMBtu starting in 2022.
o IPM - cost of $1,600 per ton applied for EGUs > 25 MW; units with existing
SCRs have their emission rates lowered to the lower of their mode 4 NOx rate in
NEEDS and the "widely achievable" optimized emissions rate of 0.08
lbs/MMBtu.8

•	At $l,800/ton:

o Engineering Analytics - Same as $l,600/ton; additionally, units with SNCRs
were given their mode 2 NOx rates if they were not already operating at that level
or better in 2019.
o IPM - Same as $l,600/ton;

•	At $5,800/ton:

o Engineering Analytics - Same as $l,800/ton; additionally, coal units greater than
100 MW and lacking a post-combustion control were given a 25% reduction to
adjusted historical baseline emissions starting in 2024 to reflect SNCR
installation.9

o IPM - Same as $l,800/ton; additionally, coal units greater than 100 MW and
lacking a post combustion control were given a 25% reduction from their mode 2
rate reflecting SNCR installation, starting in model run year 2025.10 Cost of
$5,800 per ton applied for EGUs > 25 MW. Additionally, units with idled SNCRs
were identified as units equipped with SNCR and mode 2 NOx rates in NEEDS
greater than 0.30 lbs/MMBtu. These units were given NOx rates 25% lower to
reflect SNCR operation.

•	At $9,600/ton:

o Engineering Analytics - Same as $l,800/ton; additionally, coal units greater than
100 MW and lacking a SCR were given an emission rate equal to the greater of a
reduction of 90% or 0.07 lb/MMBtu reflecting SCR installation starting in 2024.
o IPM - Same as $3,900/ton; additionally, coal units greater than 100 MW and
lacking SCR were assigned a mode four emission rate of 0.07 lb/MMBtu
reflecting SCR installation starting in model run year 2025. Cost of $9,600 per ton
applied for EGUs > 25 MW.

8	The mode 4 NOx rate, as described in Chapter 3 of the Documentation for EPA Base Case v.6 Using Integrated
Planning Model, represents post-combustion controls operating and state-of-the-art combustion controls, where
applicable. For units determined to be operating their SCR, the rate is typically equal to the unit's rate reported in
previous year ETS data. For units not operating their SCRs, the mode 4 rate is calculated as described in Attachment
3-1 of the Documentation for EPA Base Case v.6 Using Integrated Planning Model available at
https://www.epa.gov/airmarkets/ipm-v6-power-system-operation-assumptions-attachment-3-l-nox-rate-
development-epa.

9	As described in preamble section VI. C, EPA does not believe these controls to be available on a regional scale
until 2025. However, the EPA shows their impact from 2024 onwards in its engineering analysis. For its IPM
analysis, there is no model run year for 2024, so 2025 is the first year for which they can be assumed.

10	EPA's Power Sector Model v.6 using IPM does not have a 2024 model run year.

8


-------
As described in preamble section VI.B, the EPA limited its assessment of generation
shifting to reflect shifting only to other EGUs within the same state as a proxy for generation
shifting that could occur during the near-term implementation timeframe of the rule. EPA did
this by limiting state generation in each Cost Threshold run to not go below the level in its
respective Base Case. EPA also limited the potential for any new build in response to the price
signal in the near term as it was interested in capturing generation shifting among the existing
fleet.

Section C.l-3 of this TSD describes how state emissions budgets were calculated using a
combination of historical data and data from the IPM cost threshold cases. Once these budgets
were calculated, EPA used the budgets for covered states to conduct IPM Final Policy Cases to
investigate the impact of compliance with the budgets calculated from the $500, $1,800, $9,600
per ton cases. These cases reflect a less stringent scenario, the final policy scenario, and a more
stringent scenario. Specifically, the budgets informed by the Illustrative $1,800 per ton Cost
Threshold case were used for the final policy scenario, and the budgets informed by the
Illustrative $500 and $9,600 per ton cost threshold cases informed the less and more stringent
scenarios. These scenarios were used to inform the RIA.

To model these scenarios in IPM, EPA used the calculated state emissions budget and
assurance levels (121% of the state emission budget) to set state and regional ozone-season NOx
emissions limits. Additionally, EPA assumed a starting bank of allowances equal to 21% of the
sum of the 12 states' budgets. States could individually emit up to their assurance levels in each
run year, and collectively could not have emissions exceeding the sum of their regional budget
and banked allowances in each run year. In the final policy scenario and the more stringent
scenario, units with existing operating SCRs were assumed to operate them at the lower of their
mode 4 NOx rate in NEEDS and the "widely achievable" emissions rate of 0.08 lb/MMBtu, as
EPA determined this was a cost-effective mitigation strategy. Additionally, for these same two
scenarios, coal units with identified combustion control upgrade potential were assumed to
upgrade to state-of-the-art combustion controls. In all scenarios, the model provided the units the
option to retrofit with post-combustion controls. While the EPA conservatively limited
generation shifting in developing the state emission budgets, through use of state-level
generation constraints, the EPA believes that generation shifting may occur broadly among states
as a compliance mechanism and so removed that constraint for the IPM Final Policy Cases
reflecting program implementation.

9


-------
C. Calculating Budgets from Historical Data and IPM Analysis

As described in Section VII.B of the preamble, similar to CSAPR Update, the EPA
determined it was appropriate to calculate state emission budgets by combining historical
emissions and heat input data with projections from IPM to derive state emission budgets.

Section VII.B notes there are three primary steps in this process: 1) EPA determines a future year
baseline using historical data, 2) EPA adjusts that baseline to reflect the combustion and post-
combustion control mitigation measures deemed cost-effective at a given cost threshold, and 3)
EPA factors in emission reduction potential from generation shifting at a cost threshold
commensurate with that mitigation technology's control operation cost. Similar to CSAPR
Update, in this final rule the EPA calculated state budgets with the following formula:

2021 State OS NOx Budget =

2021 State OS Baseline Heat Input *[2021 State OS NOx Emissions Rate —

(2021 IPM Base Case OS NOx Emissions Rate — 2021 IPM Cost Threshold OS NOx Emissions
Rate)]11

The first two variables in the equation are derived from historical data and are the primary
determinants of states' emissions budgets. They are described in sections C.l and C.2 below.
The last two variables are identified through IPM analysis and described in section C.3 below.12
In section C.4, EPA discusses variability limits and RIA scenarios.

1. Calculating 2021-2025 Engineering Baseline for NOx (from adjusted historical data)

The underlying data and calculations described below can be found in the workbook titled
(Appendix A - Final Rule State Emission Budget Calculations and Engineering Analytics). They
are also available in the docket and on the EPA website.

EPA starts with 2019 reported, seasonal, historical NOx emissions and heat input data for each
unit.13 This reflects the latest representative owner/operator reported data available at the time of
EPA analysis.14 The NOx emissions data for units that report data to EPA under the Acid Rain
Program (ARP), the Cross-State Air Pollution Rule (CSAPR) and CSAPR Update Rule are
aggregated to the summer/ozone season period (May-September). Because the unit-level NOx
emissions for the summer/ozone-season period are relevant to determining ozone-season
emissions budgets, those files are shown in the "unit 2021" through "unit 2025" sheets in the
"Appendix A: Final Rule State Emission Budget Calculations and Engineering Analytics" file
accompanying this document.15 In that file, unit-level details such as facility name, unit ID, unit
type, capacity, etc. are shown in columns A through H of the "unit 2021" through "unit 2025"

11	The year in the formula changes for each year of budget calculation.

12	Given the proximity of the first implementation year to the analytics for this rulemaking and its promulgation,
EPA determined the use of this budget setting approach provided the most precision and expediency for this
rulemaking.

13	"Seasonal" refers to the ozone-season program months of May through September.

14	Preamble section VII.B addresses EPA's consideration of 2020 reported data as representative data.

15	The EPA notes that historical unit-level ozone season EGU NOx emission rates are publicly available and quality
assured data. The data are monitored using continuous emissions monitors (CEMs) or other monitoring approaches
available to qualifying units under 40 CFR part 75 and are reported to the EPA directly by power sector sources.

10


-------
worksheets. Reported historical data for these units such as historical emissions, heat input,
generation, etc. are shown in columns I through L. The 2019 historical emissions value is in
column I. The assumed future year baseline emissions estimate (e.g., 2021-2025) is shown in
column U, and reflects either the same emissions level as that observed in 2019, or a
modification of that value based on changes expected to the operational or pollution control
status of that unit.16 These modifications are made due to:

a. Retirements - Emissions from units with upcoming confirmed retirement dates prior
to that designated year are adjusted to zero. Retirement dates are identified through a
combination of sources including EIA Form 860, utility-announced retirements,
stakeholder and commenter feedback provided to EPA, and the National Electricity
Energy Data System (NEEDS) December 2020 file. The impact of retirements on

ami fipi /amp i fi cil^rvn7«	\ i\ TU a ra+i n m rt lim+n nra -fl nrtrrA/^	~\.T ^



2019

Future Year (e.g., 2021)

Unit x

10,000 MMBtu x .2 lb/MMBtu = 1 ton

0 MMBtu x .2 lb/MMBtu = 0 ton

b. Coal to Gas Conversion - Emissions from coal units with scheduled conversions to
natural gas fuel use by the designated future year are adjusted to reflect reduced
emission rates associated with natural gas. To reflect a given unit's conversion to gas,
that unit's future emission rates for NOx are assumed to be half of its 2019 coal-fired
emission rates while utilization levels are assumed to remain the same.18 Therefore,
the future year estimated emissions for these converting units are expected to be half
of 2019 levels for NOx. Units expected to convert to gas are flagged using EIA Form
860, NEEDS June 2020, and stakeholder feedback. The impact of coal to gas
conversion for the future year is shown in column Q, flagged in column R. The

example below pertains to NOx emission estimates.



2019

Future Year (e.g., 2021)

Unit x

10,000 MMBtu x .2 lb/MMBtu = 1 ton

10,000 MMBtu x .1 lb/MMBtu = .5 ton

c. Retrofits - Emissions from units with scheduled SCR or SNCR retrofits are adjusted
to reflect the emission rates expected with new SCR installation (0.075 lb/MMBtu of
NOx) and new SNCR (-25% representative decrease in emission rate) and are
assumed to operate at the same 2019 utilization levels.19 These emission rates were
multiplied by the affected unit's 2019 heat input to estimate the future year emission
level. The impact of post-combustion control retrofits on future year emissions
assumptions is shown in column S, flagged in column T.

For SNCR:

2019

Future Year (e.g., 2021)

16	Based on data and changes known at time of analysis.

17	EPA updated its inventory of units flagged as retiring in column N based on commenter input on the proposed
rule and the latest data from EIA 860 and the PJM retirement tracker.

18	This is consistent with NOx rate change used in IPM. See "Documentation for EPA Base Case v.5.13 Using the
Integrated Planning Model.", table 5-21.

19	Ibid.

11


-------
Unit x | 10,000 MMBtu x .2 Ib/MMBtu = 1 ton | 10,000 MMBtu x .15 Ib/MMBtu = .75 ton

For SCR:



2019

Future Year (e.g., 2021)

Unit x

10,000 MMBtu x .2 Ib/MMBtu = 1 ton

10,000 MMBtu x .075 Ib/MMBtu = .375 ton

d.	Other - EPA also made several unit-specific adjustments to 2019 emission levels to
reflect forthcoming emission or emission rate requirements specified in consent
decrees, BART requirements, and/or other revised permit limits. The impacts for
future year emission assumptions are shown in column U, flagged in column V.20

e.	New Units - Emissions for new units are identified in the "New units" worksheet.
They reflect under-construction and/or permitted units greater than 25 MW that are
expected to be in commercial operation by the designated future year. These assumed
emission values for new units are reflected in column F and the online years are in
column H. To obtain these emissions, EPA identified all new fossil-fired EGUs
coming online after 2019 according to EIA Form 860 and in NEEDSv.6 December
2020. EPA then identified the heat rate and capacity values for these units using EIA
Form 860, NEEDSv.6 December 2020 and stakeholder-provided data. Next, EPA
identified the 2019 average seasonal capacity factor for similar units that came online
between 2015-2019. EPA used these seasonal capacity factors (e.g., 65% for
NGCC), the unit's capacity, the unit's heat rate, and the unit's estimated NOx rate to
estimate future year emissions (capacity x capacity factor x number of hours in ozone
season x heat rate x NOx emission rate = NOx emissions)2122	



2019

Future Year (e.g., 2021)

Unit x

0 MMBtu x .0 Ib/MMBtu = 0 ton

100 MW * .65 *(153x24) *8000 Btu/KWh *.01
Ib/MMBtu = 9 tons

After completing these steps, EPA has unit-level and state-level future year baselines that
originate from the most recently reported representative data (2019) and incorporate known EGU
fleet changes.23 The state-level file reflects a summation of the unit-level values and provides the

211 EPA checked its inventory of units impacted by consent decrees based on input provided by commenters at
proposal. No units were determined to be impacted as described in the Allowance Allocation under the Revised
CSAPR Update Final Rule TSD.

21	Based on comment, EPA also incorporated new NGCC units that had received their regulatory approvals for
construction according to EIA 860m (October 2020), had not reported starting construction by that time, but that
were reporting planned commercial operation dates prior to the start of the 2023 or 2024 ozone season. Some of
these units appeared to have begun construction post October 2020. Moreover, regardless of whether these new units
come online as scheduled, EPA views their anticipated heat input, generation, and emissions as reflective of
expected fleet behavior from total NGCC operation in response to fleet turnover and retirement of higher emitting
units, and therefore they are included in the baseline.

22	Emission rate data is informed by the NEEDS data and historical data for like units coming online in the last five
years. See "2019 and 2020 new NGCC Data" worksheet in the "EGU Power Sector 2019 and 2020 data" file in the
docket.

23	As explained in preamble section VII.B, EPA did not consider recently available 2020 data totals as representative
for future years due to the unique global Covid-19 pandemic impacting that year.

12


-------
state-level heat input value used as the first variable in the emissions budget formula below. It
also provides the starting value for the second variable (i.e., showing the future year baseline
emission rate) before any mitigation technologies beyond the baseline are incorporated.24

^	—7fl?1 Stutr OS NO Rml rt ~					

^2021 State OS Baseline Heat7np).it ^021 State OS NOx Emissions Ra^ —

(z.uz.1 IPM Bas(TCase OS NO^bmissions tiate — 20ziTPM C0st Tlireshold OS NOx Emissions
Rate)]

2. Estimating impacts of combustion and post combustion controls on state emission budgets

Next, EPA evaluates the impact of the different combustion and post-combustion controls and
establishes the impact on the state OS NOx Emission rate to complete the second variable in the
equation above. Similar to the methodology above, EPA continued to adjust the historical data to
reflect a future year with specific uniform control assumptions. However, these adjustments
were to capture changes incremental to the baseline reflecting different uniform control
measures. EPA applied these adjustments for analytical purposes to all states, but only the 12
linked states' adjustments are relevant for emission budgets finalized in this rule. Each of these
adjustments is shown incrementally for the relevant mitigation technology in the "unit 2021"
through "unit 2025" worksheets.

a. SCR optimization - Emissions from units with existing SCRs, but that operated at an
emission rate greater than 0.08 lb/MMBtu in 2019, were adjusted downwards to reflect
expected emissions when the SCR is operated to achieve a 0.08 lb/MMBtu emission rate.
The 0.08 lb/MMBtu emission rate was identified as the emission rate that reflected the
fleet-average optimization assumption for SCR controlled units that were not currently
optimizing their controls. The optimized emission rate is multiplied by baseline heat
input levels to arrive at the future year emissions estimate. The impact on future year
emission assumptions is shown in column W and flagged in column X of the "unit 2021"
through "unit 2025" worksheets. EPA notes this assumption only applies to ozone-season
NOx as that is the season in which the Revised CSAPR Update Rule would likely
incentivize such operation. In the final rule, EPA also incorporated a flag in column X,
based on commenter input, for units with SCRs and a shared stack. For these units, based
on commenter provided data, EPA did not assume potential emission reductions
attributable to SCR optimization as explained in preamble section VI.B.	



2019

Future Year (e.g., 2021)

Unit x

10,000 MMBtu x .2 lb/MMBtu = 1 ton

10,000 MMBtu x .08 lb/MMBtu = .4 ton

b. State-of-the- art combustion controls - Emissions from units that were operating in 2019
without state-of-the-art combustion controls were adjusted downwards to reflect assumed

24 While less relevant to emission budgets setting, EPA also created a future year baseline for 1) NOx and SO2
emission from EGUs not currently covered under existing EPA programs that require emissions monitoring and
reporting under 40 CFR part 75, and for other pollutants for all grid connected EGUs (e.g., PM2.5, P.M10. CO).
These data points were used in some of the air quality analysis and in some of the system impacts estimates for the
RIA. The EPA also evaluates whether the assumed aggregate heat input changes given retirements and new builds
are consistent with trends observed historically in the fleet and with new planned units identified in EIA Form 860.
This evaluation is in the appendix to this document.

13


-------
installation of, or upgrade to, these controls and their expected emission rate impact.
EPA assumed a future year emission rate of 0.199 for units expected to install/upgrade
combustion controls. These emission rates were multiplied by each unit's future year
baseline heat input to estimate its future emission level. Details of EPA's assessment of
state-of-the-art NOx combustion controls and corresponding emission rates are provided
in the EGU NOx Mitigation Strategies Final Rule TSD. The impact of state-of-the-art
combustion controls on future year emission assumptions is shown in column Y and
flagged in column Z of the "unit 2021" through "unit 2025" worksheets. EPA also
incorporated a flag in column Z, based on commenter input, for units with a shared stack.
For these units, based on commenter provided data, EPA did not assume potential
emission reductions attributable to state-of-the-art combustion controls as explained in
preamble section VI.B. Note, these assumptions apply to both winter and ozone season
emissions adjustments as the controls operate continuously once installed.



2019

Future Year (e.g., 2021)

Unit x

10,000 MMBtu x .4 Ib/MMBtu = 2 ton

10,000 MMBtu x ,199lb/MMBtu = ~1 ton

c. SNCR optimization - Emissions from units with existing SNCRs, but that operated at an
emission rate greater than the SNCR optimization rate, were adjusted downwards to
reflect expected emissions when the SNCR is optimized. This emission rate was
identified specific to each unit based on historical data and is described in the EGU NOx
Mitigation Strategy Final Rule TSD. The optimized emission rate is multiplied by future
year baseline heat input levels to arrive at the future year emissions estimate. For the
units affected by this adjustment, the impact on future year emission assumptions is
shown in column AA and flagged in column AB of the "unit 2021" through "unit 2025"
worksheets. Note, this assumption only applies to ozone-season NOx as that is the season
in which the Revised CSAPR Update Rule would likely incentivize such installation and



2019

Future Year (e.g., 2021)

Unit x

10,000 MMBtu x .2 Ib/MMBtu = 1 ton

10,000 MMBtu x .15 Ib/MMBtu = .75 ton

d. SNCR retrofit- Emissions from coal units greater than 100 MW without post-combustion
controls were adjusted downwards to reflect expected emissions if an SNCR were to be
retrofitted on the unit. The emission rate was identified as 75% of the unit's baseline
emission rate level (i.e., reflecting a 25% reduction from the technology). The adjusted
emission rate is multiplied by future year baseline heat input levels to arrive at the future
year emissions estimate for that technology. For the units affected by this adjustment, the
impact on future year emission assumptions is shown in column AC and flagged in
column AD of the "unit 2021" through "unit 2025" worksheets. Note, this assumption
only applies to ozone-season NOx as that is the season in which the Revised CSAPR
Update Rule would likely incentivize such installation and operation.



2019

Future Year (e.g., 2021)

Unit x

10,000 MMBtu x .2 Ib/MMBtu = 1 ton

10,000 MMBtu x .15 Ib/MMBtu = .75 ton

14


-------
e. SCR retrofit- Emissions from coal units greater than 100 MW without SCR controls were
adjusted downwards to reflect expected emissions if an SCR were to be retrofitted on the
unit. The emission rate was identified as 10% of the unit's baseline emission rate or 0.07
lb/MMBtu (i.e., a 90% reduction with an emission rate floor of 0.07 lb/MMBtu).25 The
adjusted emission rate is multiplied by future year baseline heat input levels to arrive at
the future year emissions estimate for that technology. For the units affected by this
adjustment, the impact on future year emission assumptions is shown in column AE and
flagged in column AF of the "unit 2021" through "unit 2025" worksheets. Note, this
assumption only applies to ozone-season NOx as that is the period in which the Revised



2019

Future Year (e.g., 2021)

Unit x

10,000 MMBtu x .2 lb/MMBtu = 1 ton

10,000 MMBtu x .07lb/MMBtu = .35 ton

These adjustments for each uniform control technology resulted in adjusted OS NOx emissions,
heat input, and emission rates at the unit-level. When summed up to the state level, these
changes resulted in the State OS NOx Emission Rate listed second in the formula below. EPA
notes, this emission rate for any given uniform control level times the baseline heat input would
provide the state emissions budget without generation shifting. These pre-generation shifting
emission budget levels at the state-level are visible in the worksheets titled "State 2021" through
"State 2025" in the Appendix A: Final Rale State Emission Budget Calculations and Engineering
Analytics workbook accompanying this document.26

State 2021 OS NOx Budget =				

2021 State OS Baseline Heat Input^j2021 State OS NOx Emissions ~Rate}~

(2021 IPM Base Case OSNO^ hi ill!, s iuiis Rule 2U21 iFKlCost Threshold OS NOx Emissions
Rate)]

3. Estimating Emission Reduction Potential from Generation Shifting

The last two variables in the equation relate to emission reductions from generation shifting.
Here, as in the CSAPR Update, EPA uses the Integrated Planning Model (IPM) to capture the
change in emission rate in a state's fossil-fuel fired power fleet when. While holding everything
else equal, EPA applies a given dollar per ton marginal cost constraint. EPA relies on IPM for
this analysis as generation shifting occurs on a cost continuum and is a function of least-cost
dispatch under different constraints. To derive this value, EPA first prepares an adjusted base
case that reflects all the combustion or post-combustion mitigation measures discussed above for

25	This is a conservative estimate based on the floor rates for new SCRs in the 1PM documentation, ranging from
0.05 to 0.07 lbs/mmBtu, depending on coal type. See "Documentation for EPA Base Case v.5.13 Using the
Integrated Planning Model," table 5-5.

26	EPA makes these illustrative unit-level details described in C.l and C.2 available, before aggregating those
values to use at the state and regional level. The illustrative unit-level values are meant to be a tool to inform a state-
level estimate, not a prediction of how each unit will operate in the future. Although anchored in historical data,
EPA recognizes at the unit-level some units will overperfonn and some units will underperfonn the unit-level
illustrative values. It is an exercise in projecting reasonable state-level and region-level totals, not an exercise that
purports to predict the future of millions of operational variables at the unit-level. This is discussed further in the
Budgets section of the Response to Comment Document.

15


-------
a given cost threshold. These adjusted base cases are specific to the uniform mitigation scenario.
For instance, for the $1,600 per ton scenario EPA adjusts its base case to reflect the optimization
of SCRs and combustion control upgrades by adjusting the emission rates to the levels discussed
above for relevant units not already achieving that level. EPA then executes an IPM run with
these new exogenous assumptions and observes the state-level emission rate for fossil-fuel fired
units greater than 25 MW. This is the third variable in the emissions budget formula.

Next, EPA performs a sensitivity for these adjusted base case runs where it applies the same set
of assumptions in variable three, but layers on a commensurate marginal cost price signal (e.g.,
$1,600 per ton). In addition to the mitigation measures assumed, the entire fossil-fuel fired EGU
fleet greater than 25 MW in the state is subjected to a cost-per-ton price associated with that
technology. The model solves for least-cost dispatch given this additional marginal cost
constraints for seasonal ozone emissions. EPA observes the state-level emission rate for fossil-
fuel fired units greater than 25 MW. This data point becomes the fourth variable in the state-
emissions budget formula. The difference between the third and fourth variables reflects the
change in emission rate due solely to generation shifting at a given dollar per ton level.

State 2021 OS NOx Budget =

2021 State OS Baseline Heat Input *[2021 State OS NOx Emissions Rate —	

—t-z!U/Il II'm Base Case OS NOx Emissions Rate — 2021 IPM Cost Threshold us Lniijwons
	Rate)!27			

This difference in the state-level emission rate between the two IPM cases is shown in columns
B through F in the worksheet titled "Generation Shifting". These values are in the Appendix A:
Final Rale State Emission Budget Calculations and Engineering Analytics workbook
accompanying this document.

Once EPA calculated the change in emissions rate between the IPM adjusted base case
and each cost threshold case, the EPA then subtracted this change in emissions rate from the
state OS NOx emission rate without generation shifting (the second variable in the formula).

This yielded state-level, historically-anchored, emission rates reflecting NOx reduction potential
for a given uniform control measure.

Finally, the EPA multiplied these rates by each state's adjusted heat input (historical heat
input adjusted for retirements and new builds identified in variable one of the formula) to yield
emission budgets for each cost threshold. The state budgets for the different cost thresholds are
displayed in Table C-l through C-5. EPA notes that budgets are calculated for all states for the
purpose of AQAT analysis, as explained in section D of this TSD, even if the state is not covered
by the Revised CSAPR Update Rule.

In addition to being shown below, the state-level emission budgets are calculated in the
far right-hand side columns of each "State" worksheet for each mitigation technology scenario
available that year. These budgets reflect an application of the formula described above to the
data in the spreadsheet. These state-emission budgets reflect the inclusion of generation shifting.

27 The year in the formula changes for each year of budget calculation.

16


-------
The difference between these final state-emission budgets shown in the far right columns that
include "generation shifting" it the column title and the immediately preceding columns with the
same column title but without "generation shifting" reflect the additional reduction due to
generation shifting at a given cost threshold.

Generation shifting accounts for a small amount of emissions reductions relative to the
combustion and post combustion control mitigation measures (see Table C-10). In its cost
threshold modeling, EPA limited generation levels in a state to its base case level so that states
would not achieve emission reductions by importing more generation from out-of-state EGUs
and reducing in-state generation (e.g., emissions leakage). This assumption ensures that the
generation shifting-based reductions are, and can be achieved, within a state. EPA also only
assumes generation shifting from the projected baseline fleet, it does not incorporate generation
shifting from any assumed incremental new build capacity that could be incentivized by a price
level. Finally, in EPA's budget setting process it only includes generation shifting at dollar per
ton levels that encourage the optimization of existing or newly installed controls considered at
that cost level. Capturing reduction potential from generation shifting in the state's emission
budgets is meant to preserve the incentive to implement EPA's identified control strategy.
Factoring generation shifting into the state emissions budgets helps promote an allowance price
commensurate with these levels. In this rule, generation shifting is intended to be a mitigation
measure supportive to those combustion and post-combustion control measures, not incremental
to it. Therefore, EPA designed its IPM analysis and utilized the results for emission budget
purposes in a manner that did not allow for, or include, emission reductions from projected
model new builds or retirements that occurred in response to a dollar per ton price signal.28
Instead, EPA examined generation shifting that was expected to occur among the baseline fleet
at cost threshold levels commensurate with post-combustion control operation (e.g., $1,600 per
ton) at fossil fuel-fired units greater than 25 MW for 2021.29

28	EPA also relied on the modeled emission rate change in the IPM 2021 results for each year of the budget
calculation to avoid capturing generation shifting attributable to model-projected new builds in later years that are
not yet under construction.

29	As explained in preamble Section VLB. and VI.C, EPA does not believe regional post-combustion control
installation (represented by higher cost thresholds of $5,800/ton and $9,600/ton) is possible prior to 2025, and thus
not relevant for consideration in this action as there are no nonattainment or maintenance receptors in 2025 after
reductions available at $l,800/ton are implemented. However, for illustrative purposes, EPA assessed reductions at
these levels as well. For the higher cost thresholds of $5,800/ton and $9,600/ton pertaining to the later years of
analysis (2024 and 2025), EPA used the generation shifting emission rate delta consistent with the cost of operating
any idled existing post-combustion control (e.g., $3,900/ton) to ensure that all controls (existing and new) would
have an incentive to operate if installed. EPA also performed a feasibility check on its generation shifting
assumptions to assess whether such generation shifting would still be likely once those assumed controls were
installed. If the state's assumed emission rate reductions from generation shifting were greater than 10% of the IPM
baseline, and its adjusted historical baseline for that year was less than 90% of the IPM baseline, then no additional
reductions were assumed from generation shifting at higher cost thresholds of $5,800 and $9,600 in EPA's 2024
analysis. While this last assessment was done for all states, only Utah and Arizona (states which are not covered in
this rulemaking) were affected.

17


-------
Table C-l. 2021 Emissions for States at Different Uniform Control Scenarios (Reflected by dollar
		per ton)	



OS NOx (tons)



2021
Baseline

$500/ton

0.08 SCR

Optimization

($1,600/ton)

0.08 SCR
Optimization +
LNB ($1,600/ton)

0.08 SCR
Optimization +
LNB + SNCR
Optimization
($1,800/ton)

Alabama

7,786

7,785

7,786

7,693

7,693

Arizona

5,389

5,100

4,616

4,616

4,616

Arkansas

8,731

8,655

8,708

8,708

8,708

California

1,112

1,111

1,062

1,062

1,061

Colorado

7,484

7,487

7,471

7,471

7,449

Connecticut

344

316

307

307

307

Delaware

223

223

206

206

205

Florida

15,286

15,276

13,869

13,869

13,788

Fort Mojave

53

53

53

53

53

Georgia

7,833

7,833

7,808

7,808

7,808

Idaho

204

204

204

204

204

Illinois

9,368

9,348

9,198

9,198

9,102

Indiana

15,856

15,677

13,085

13,085

13,051

Iowa

8,567

8,447

7,714

7,659

7,659

Kansas

6,057

6,053

5,384

5,384

5,338

Kentucky

15,588

15,606

15,307

14,057

14,051

Louisiana

15,476

15,430

15,389

15,389

14,818

Maine

67

67

67

67

67

Maryland

1,501

1,501

1,499

1,499

1,499

Massachusetts

336

333

333

333

331

Michigan

13,898

13,126

12,732

12,614

12,610

Minnesota

5,969

5,842

5,448

5,448

5,448

Mississippi

8,070

8,067

8,065

7,739

7,739

Missouri

12,439

12,379

11,352

11,352

11,276

Montana

3,553

3,553

3,553

3,553

3,553

Navajo

1,319

1,319

1,319

1,319

1,319

Nebraska

8,078

8,013

8,037

7,530

7,530

Nevada

2,434

1,833

1,456

1,456

1,456

New Hampshire

386

386

299

299

299

New Jersey

1,346

1,346

1,253

1,253

1,253

New Mexico

4,656

4,624

4,502

4,502

4,488

New York

3,469

3,463

3,416

3,416

3,416

North Carolina

15,911

15,814

11,227

11,227

11,083

North Dakota

11,885

11,829

11,774

11,774

11,135

Ohio

15,829

15,487

9,690

9,690

9,690

Oklahoma

8,964

8,878

8,717

8,717

8,717

Oregon

350

350

350

350

350

Pennsylvania

11,896

11,807

8,379

8,379

8,379

Rhode Island

233

233

233

233

233

South Carolina

4,979

4,979

3,839

3,839

3,839

South Dakota

591

583

581

581

581

Tennessee

4,547

4,549

4,367

4,367

4,367

Texas

44,767

43,841

42,349

42,349

42,308

Utah

6,729

4,862

4,837

4,837

4,837

Ute

2,144

2,144

2,144

2,144

2,144

Vermont

51

51

51

51

51

Virginia

4,664

4,661

4,614

4,345

4,284

Washington

1,609

1,609

1,609

1,609

1,603

18


-------
West Virginia

15,165

15,017

13,686

13,205

12,884

Wisconsin

5,251

5,120

4,952

4,952

4,945

Wyoming

11,480

11,480

11,366

10,623

10,623

12 Linked States Total

124,057

122,469

108,248

106,130

105,037

"Note - For 2021 EPA shows $1,600 ton with and without LNB upgrade; given it is not requiring budgets reflecting LNB controls until

2022.

Table C-2. 2022 Emissions for States at Different Uniform Control Scenarios (Reflected by dollar
		per ton)	



OS NOx (tons)



2022 Baseline

$500/ton

0.08 SCR Optimization
+ LNB ($1,600/ton)

0.08 SCR Optimization
+ LNB + SNCR
Optimization
($1,800/ton)

Alabama

7,786

7,785

7,693

7,693

Arizona

5,389

5,100

4,616

4,616

Arkansas

8,731

8,655

8,708

8,708

California

1,104

1,103

1,055

1,053

Colorado

7,484

7,487

7,471

7,449

Connecticut

341

313

304

304

Delaware

220

220

203

202

Florida

14,976

14,966

13,641

13,560

Fort Mojave

53

53

53

53

Georgia

7,833

7,833

7,808

7,808

Idaho

204

204

204

204

Illinois

9,368

9,348

9,198

9,102

Indiana

15,383

15,206

12,615

12,582

Iowa

8,567

8,447

7,659

7,659

Kansas

6,057

6,053

5,384

5,338

Kentucky

15,588

15,606

14,057

14,051

Louisiana

15,476

15,430

15,389

14,818

Maine

67

67

67

67

Maryland

1,267

1,267

1,266

1,266

Massachusetts

336

333

333

331

Michigan

13,459

12,688

12,295

12,290

Minnesota

5,888

5,761

5,369

5,369

Mississippi

8,070

8,067

7,739

7,739

Missouri

12,439

12,379

11,352

11,276

Montana

3,249

3,249

3,249

3,249

Navajo

1,319

1,319

1,319

1,319

Nebraska

8,078

8,013

7,530

7,530

Nevada

1,500

931

575

575

New Hampshire

386

386

299

299

New Jersey

1,346

1,346

1,253

1,253

New Mexico

4,656

4,624

4,502

4,488

New York

3,469

3,463

3,416

3,416

North Carolina

15,326

15,231

10,658

10,514

North Dakota

11,885

11,829

11,774

11,135

Ohio

15,927

15,569

9,773

9,773

Oklahoma

8,964

8,878

8,717

8,717

Oregon

350

350

350

350

Pennsylvania

11,896

11,806

8,373

8,373

Rhode Island

233

233

233

233

South Carolina

4,979

4,979

3,839

3,839

South Dakota

591

583

581

581

19


-------
Tennessee

4,547

4,549

4,367

4,367



Texas

44,773

43,835

42,326

42,285



Utah

6,729

4,862

4,837

4,837



Ute

2,144

2,144

2,144

2,144



Vermont

51

51

51

51



Virginia

4,274

4,270

3,957

3,897



Washington

1,609

1,609

1,609

1,603



West Virginia

15,165

15,017

13,205

12,884



Wisconsin

4,992

4,864

4,700

4,693



Wyoming

10,918

10,918

10,061

10,061



12 Linked States Total

122,619

121,016

104,797

103,705



Table C-3. 2023 Emissions for States at Different Uniform Control Scenarios (Reflected by dollar

per ton).





OS NOx (tons)



2023 Baseline

$500/ton

0.08 SCR Optimization
+ LNB ($1,600/ton)

0.08 SCR Optimization
+ LNB + SNCR
Optimization
($1,800/ton)

Alabama

7,786

7,785

7,693

7,693

Arizona

5,389

5,100

4,616

4,616

Arkansas

8,731

8,655

8,708

8,708

California

1,104

1,103

1,055

1,053

Colorado

6,663

6,666

6,650

6,629

Connecticut

341

313

304

304

Delaware

220

220

203

202

Florida

14,496

14,486

13,162

13,080

Fort Mojave

53

53

53

53

Georgia

7,154

7,154

7,129

7,129

Idaho

204

204

204

204

Illinois

8,413

8,393

8,275

8,179

Indiana

15,357

15,179

12,587

12,553

Iowa

7,647

7,531

6,753

6,753

Kansas

6,057

6,053

5,384

5,338

Kentucky

15,588

15,606

14,057

14,051

Louisiana

15,476

15,430

15,389

14,818

Maine

67

67

67

67

Maryland

1,267

1,267

1,266

1,266

Massachusetts

320

317

317

315

Michigan

11,182

10,386

9,980

9,975

Minnesota

4,655

4,545

4,198

4,198

Mississippi

8,070

8,067

7,739

7,739

Missouri

12,160

12,101

11,074

10,997

Montana

3,249

3,249

3,249

3,249

Navajo

1,319

1,319

1,319

1,319

Nebraska

8,078

8,013

7,530

7,530

Nevada

1,386

828

479

479

New Hampshire

386

386

299

299

New Jersey

1,346

1,346

1,253

1,253

New Mexico

1,693

1,673

1,594

1,594

New York

3,474

3,468

3,421

3,421

North Carolina

15,326

15,231

10,658

10,514

North Dakota

9,166

9,128

9,091

8,452

20


-------
Ohio

15,927

15,569

9,773

9,773

Oklahoma

8,964

8,878

8,717

8,717

Oregon

350

350

350

350

Pennsylvania

11,896

11,806

8,373

8,373

Rhode Island

233

233

233

233

South Carolina

4,979

4,979

3,839

3,839

South Dakota

591

583

581

581

Tennessee

4,547

4,549

4,367

4,367

Texas

44,582

43,646

42,138

42,097

Utah

6,729

4,862

4,837

4,837

Ute

2,144

2,144

2,144

2,144

Vermont

51

51

51

51

Virginia

4,361

4,357

4,041

3,980

Washington

1,609

1,609

1,609

1,603

West Virginia

15,165

15,017

13,205

12,884

Wisconsin

4,857

4,734

4,576

4,569

Wyoming

10,337

10,337

9,480

9,480

12 Linked States Total

119,453

117,824

101,620

100,526

Table C-4. 2024 Emissions for States at Different Uniform Control Scenarios (Reflected by dollar



OS NOx (tons)











0.08 SCR

0.08 SCR







0.08 SCR

0.08 SCR
Optimization +

Optimization +
LNB + SNCR

Optimization +
LNB + SNCR



2024



Optimization
+ LNB

LNB + SNCR
Optimization

Optimization +
SNCR Retrofit

Optimization +
SCR Retrofit



Baseline

$500/ton

($1,600/ton)

($1,800/ton)

($5,800/ton)

($9,600/ton)

Alabama

7,786

7,785

7,693

7,693

7,694

7,515

Arizona

5,389

5,100

4,616

4,616

4,816

3,916

Arkansas

8,731

8,655

8,708

8,708

6,642

4,661

California

1,104

1,103

1,055

1,053

1,053

1,053

Colorado

5,950

5,953

5,938

5,916

5,085

3,826

Connecticut

341

313

304

304

302

302

Delaware

220

220

203

202

194

194

Florida

14,505

14,495

13,170

13,089

12,446

11,752

Fort Mojave

53

53

53

53

53

53

Georgia

7,154

7,154

7,129

7,129

7,097

7,097

Idaho

204

204

204

204

204

204

Illinois

8,292

8,272

8,154

8,059

7,239

6,891

Indiana

12,232

12,083

9,585

9,564

8,923

8,430

Iowa

7,647

7,531

6,753

6,753

5,201

2,817

Kansas

6,057

6,053

5,384

5,338

4,815

3,658

Kentucky

15,588

15,606

14,057

14,051

12,322

9,775

Louisiana

15,476

15,430

15,389

14,818

14,378

12,622

Maine

67

67

67

67

67

67

Maryland

1,350

1,350

1,348

1,348

1,168

1,168

Massachusetts

320

317

317

315

307

307

Michigan

10,968

10,188

9,791

9,786

8,670

7,344

Minnesota

4,655

4,545

4,198

4,198

3,034

2,581

Mississippi

8,070

8,067

7,739

7,739

7,081

6,436

Missouri

12,160

12,101

11,074

10,997

10,744

9,301

Montana

3,249

3,249

3,249

3,249

2,445

1,544

Navajo

1,319

1,319

1,319

1,319

1,319

1,319

Nebraska

7,347

7,283

7,299

7,299

5,966

4,246

21


-------
Nevada

1,386

828

479

479

332

100

New Hampshire

386

386

299

299

299

299

New Jersey

1,346

1,346

1,253

1,253

1,257

1,257

New Mexico

1,693

1,673

1,594

1,594

1,187

1,187

New York

3,456

3,450

3,403

3,403

3,297

3,297

North Carolina

15,326

15,231

10,658

10,514

7,981

4,691

North Dakota

9,166

9,128

9,091

8,452

7,237

2,250

Ohio

15,927

15,569

9,773

9,773

9,644

9,222

Oklahoma

8,964

8,878

8,717

8,717

7,820

7,251

Oregon

350

350

350

350

350

350

Pennsylvania

11,896

11,806

8,373

8,373

7,921

7,851

Rhode Island

233

233

233

233

233

233

South Carolina

4,903

4,903

3,769

3,769

3,762

3,762

South Dakota

591

583

581

581

583

583

Tennessee

4,547

4,549

4,367

4,367

4,280

4,280

Texas

43,265

42,338

40,845

40,804

35,342

29,460

Utah

6,729

4,862

4,837

4,837

5,094

3,174

Ute

2,144

2,144

2,144

2,144

1,608

573

Vermont

51

51

51

51

51

51

Virginia

4,025

4,021

3,707

3,663

3,618

3,184

Washington

1,609

1,609

1,609

1,603

1,603

761

West Virginia

15,165

15,017

13,205

12,884

12,837

10,568

Wisconsin

4,161

4,043

3,893

3,886

3,862

3,862

Wyoming

10,337

10,337

9,480

9,480

6,997

3,972

12 Linked States Total

115,722

114,138

98,038

96,975

91,274

81,609

Table C-5. 2025 Emissions for States at Different Uniform Control Scenarios (Reflected by dollar

per ton).



OS NOx (tons)



2025
Baseline

$500/ton

0.08 SCR
Optimization +
LNB

($1,600/ton)

0.08 SCR
Optimization +
LNB + SNCR
Optimization
($1,800/ton)

0.08 SCR
Optimization +
LNB + SNCR
Optimization +
SNCR Retrofit
($5,800/ton)

0.08 SCR
Optimization +
LNB + SNCR
Optimization +
SCR Retrofit
($9,600/ton)

Alabama

7,786

7,785

7,693

7,693

7,694

7,515

Arizona

5,389

5,100

4,616

4,616

4,816

3,916

Arkansas

8,731

8,655

8,708

8,708

6,642

4,661

California

1,104

1,103

1,055

1,053

1,053

1,053

Colorado

5,950

5,953

5,938

5,916

5,085

3,826

Connecticut

341

313

304

304

302

302

Delaware

220

220

203

202

194

194

Florida

13,938

13,929

12,604

12,523

11,882

11,188

Fort Mojave

53

53

53

53

53

53

Georgia

7,154

7,154

7,129

7,129

7,097

7,097

Idaho

204

204

204

204

204

204

Illinois

8,281

8,261

8,143

8,047

7,228

6,880

Indiana

12,232

12,083

9,585

9,564

8,923

8,430

Iowa

7,647

7,531

6,753

6,753

5,201

2,817

Kansas

6,057

6,053

5,384

5,338

4,815

3,658

Kentucky

14,551

14,567

13,352

13,345

11,796

9,529

Louisiana

15,476

15,430

15,389

14,818

14,378

12,622

Maine

67

67

67

67

67

67

Maryland

1,350

1,350

1,348

1,348

1,168

1,168

Massachusetts

320

317

317

315

307

307

Michigan

11,009

10,211

9,804

9,800

8,678

7,352

22


-------
Minnesota

4,655

4,545

4,198

4,198

3,034

2,581

Mississippi

8,070

8,067

7,739

7,739

7,081

6,436

Missouri

12,147

12,088

11,061

10,985

10,732

9,288

Montana

3,249

3,249

3,249

3,249

2,445

1,544

Navajo

1,319

1,319

1,319

1,319

1,319

1,319

Nebraska

7,347

7,283

7,299

7,299

5,966

4,246

Nevada

1,386

828

479

479

332

100

New Hampshire

386

386

299

299

299

299

New Jersey

1,346

1,346

1,253

1,253

1,257

1,257

New Mexico

1,693

1,673

1,594

1,594

1,187

1,187

New York

3,456

3,450

3,403

3,403

3,297

3,297

North Carolina

14,281

14,188

9,632

9,538

7,055

4,483

North Dakota

9,166

9,128

9,091

8,452

7,237

2,250

Ohio

15,927

15,569

9,773

9,773

9,644

9,222

Oklahoma

8,964

8,878

8,717

8,717

7,820

7,251

Oregon

350

350

350

350

350

350

Pennsylvania

11,896

11,806

8,373

8,373

7,921

7,851

Rhode Island

233

233

233

233

233

233

South Carolina

4,903

4,903

3,769

3,769

3,762

3,762

South Dakota

591

583

581

581

583

583

Tennessee

3,953

3,954

3,907

3,907

3,826

3,826

Texas

43,125

42,199

40,708

40,667

35,207

29,325

Utah

6,729

4,862

4,837

4,837

5,094

3,174

Ute

2,144

2,144

2,144

2,144

1,608

573

Vermont

51

51

51

51

51

51

Virginia

4,162

4,158

3,839

3,795

3,745

3,312

Washington

1,609

1,609

1,609

1,603

1,603

761

West Virginia

15,165

15,017

13,205

12,884

12,837

10,568

Wisconsin

3,769

3,660

3,519

3,512

3,490

3,490

Wyoming

10,337

10,337

9,480

9,480

6,997

3,972

12 Linked States Total

114,850

113,248

97,467

96,403

90,872

81,488

As noted in Section VI of the Preamble, EPA identified $1,800 per ton as the point for
determining significant contribution from EGUs under the Step 3 multifactor test. Section VII
explains that EPA applied this cost threshold to each year through 2024 to arrive at a budget
estimate for that year. Those state-level emissions budgets for the 12 states along with the
corresponding percent reduction relative to 2019 and the state's baseline emissions for that year
are shown below in Tables C-6 through C-10.30

Table C-6 - OS NOx, 2021 Emissions Budget, and % Reduction



2016 OS
NOx

2019 OS
NOx

Baseline
2021 OS
NOx

2021
Budget

%

Reduction
from 2019

% Reduction
from 2021
Baseline

Illinois

14,553

11,877

9,368

9,102

23%

3%

Indiana

34,636

16,594

15,856

13,051

21%

18%

Kentucky

25,403

19,117

15,588

15,300

20%

2%

Louisiana

19,615

15,365

15,476

14,818

4%

4%

Maryland

4,471

1,662

1,501

1,499

10%

0%

311A table providing state emission budgets and associated variability limits for these 12 linked states is provided in
Appendix G

23


-------
Michigan

17,632

14,055

13,898

12,727

9%

8%

New Jersey

2,463

1,346

1,346

1,253

7%

7%

New York

6,534

3,225

3,469

3,416

-6%

2%

Ohio

24,205

16,390

15,829

9,690

41%

39%

Pennsylvania

31,896

12,093

11,896

8,379

31%

30%

Virginia

9,833

4,668

4,664

4,516

3%

3%

West Virginia

21,178

15,615

15,165

13,334

15%

12%

Total

212,418

132,006

124,057

107,085

19%

13.7%

Table C-7. OS NOx, 2022 Emissions Budget, and % Reduction



2016

2019

Baseline



%

% Reduction



OS

OS

2022 OS

2022

Reduction

from 2022



NOx

NOx

NOx

Budget

from 2019

Baseline

Illinois

14,553

11,877

9,368

9,102

23%

3%

Indiana

34,636

16,594

15,383

12,582

24%

18%

Kentucky

25,403

19,117

15,588

14,051

27%

10%

Louisiana

19,615

15,365

15,476

14,818

4%

4%

Maryland

4,471

1,662

1,267

1,266

24%

0%

Michigan

17,632

14,055

13,459

12,290

13%

9%

New Jersey

2,463

1,346

1,346

1,253

7%

7%

New York

6,534

3,225

3,469

3,416

-6%

2%

Ohio

24,205

16,390

15,927

9,773

40%

39%

Pennsylvania

31,896

12,093

11,896

8,373

31%

30%

Virginia

9,833

4,668

4,274

3,897

17%

9%

West Virginia

21,178

15,615

15,165

12,884

17%

15%

Total

212,418

132,006

122,619

103,705

21%

15.4%

Table C-8. OS NOx, 2023 Emissions Budget, and % Reduction



2016

2019

Baseline



%

% Reduction



OS

OS

2023 OS

2023

Reduction

from 2023



NOx

NOx

NOx

Budget

from 2019

Baseline

Illinois

14,553

11,877

8,413

8,179

31%

3%

Indiana

34,636

16,594

15,357

12,553

24%

18%

Kentucky

25,403

19,117

15,588

14,051

27%

10%

Louisiana

19,615

15,365

15,476

14,818

4%

4%

Maryland

4,471

1,662

1,267

1,266

24%

0%

Michigan

17,632

14,055

11,182

9,975

29%

11%

New Jersey

2,463

1,346

1,346

1,253

7%

7%

New York

6,534

3,225

3,474

3,421

-6%

2%

Ohio

24,205

16,390

15,927

9,773

40%

39%

Pennsylvania

31,896

12,093

11,896

8,373

31%

30%

24


-------
Virginia

9,833

4,668

4,361

3,980

15%

9%

West Virginia

21,178

15,615

15,165

12,884

17%

15%

Total

212,418

132,006

119,453

100,526

24%

15.8%

Table C-9. OS NOx, 2024 Onward: Emissions Budget, anc



2016

2019

Baseline



%

% Reduction



OS

OS

2024 OS

2024

Reduction

from 2024



NOx

NOx

NOx

Budget

from 2019

Baseline

Illinois

14,553

11,877

8,292

8,059

32%

3%

Indiana

34,636

16,594

12,232

9,564

42%

22%

Kentucky

25,403

19,117

15,588

14,051

27%

10%

Louisiana

19,615

15,365

15,476

14,818

4%

4%

Maryland

4,471

1,662

1,350

1,348

19%

0%

Michigan

17,632

14,055

10,968

9,786

30%

11%

New Jersey

2,463

1,346

1,346

1,253

7%

7%

New York

6,534

3,225

3,456

3,403

-6%

2%

Ohio

24,205

16,390

15,927

9,773

40%

39%

Pennsylvania

31,896

12,093

11,896

8,373

31%

30%

Virginia

9,833

4,668

4,025

3,663

22%

9%

West Virginia

21,178

15,615

15,165

12,884

17%

15%

Total

212,418

132,006

115,722

96,975

27%

16.2%

% Reduction

Table C-10. Emission Reduction Attributable to Generation Shifting (for 12 linked states)



Baseline
OS NOx

Budget
Without
Gen
Shifting

Budget
With
Gen.
Shifting

% Reduction from
Generation Shifting as a
Percentage of Baseline

2021

124,057

109,578

107,085

2%

2022

122,619

106,211

103,705

2%

2023

119,453

103,077

100,526

2%

2024

115,722

99,446

96,975

2%

4. Variability Limits and RIA Scenarios

Once EPA determined state-emission budgets, EPA calculated the variability limits and
assurance levels for each state based on the calculated emission budgets. Each state's variability
limit is 21% of its budget, and its assurance level is the sum of its budget and variability limit (or
121% of its budget). The variability limits and assurance levels are further described and shown
in section VII of the preamble for the Revised CSAPR Update and shown in Table Appendix G-
1.

As explained in the preamble, the EPA is finalizing EGU NOx ozone season emission
budgets reflecting the uniform cost threshold of $1,800 per ton to eliminate significant
contribution to nonattainment and interference with maintenance.

25


-------
For the RIA analysis, EPA used the budgets informed by the $1,800 per ton cost
threshold scenario. Additionally, the RIA includes analysis of the less stringent policy option,
using the budgets from a $500 per ton cost threshold case, and a more stringent policy
alternative, using 2025 budgets from the $9,600 per ton cost threshold case reflecting SCR
retrofits at coal units greater than 100 MW lacking such controls.

The IPM runs performed for this analysis are listed in Appendix C of this TSD. Table
Appendix C-l lists the name of each IPM run next to a description of the run. The output files of
these model runs can be found in the rulemaking docket. Detailed budget calculations for all
cost per ton cases can be found in Appendix A - Final Rule State Emission Budget Calculations
and Engineering Analytics.

26


-------
D. Analysis of Air Quality Responses to Emission Changes Using an Ozone Air Quality
Assessment Tool (AQAT)

EPA has defined each linked upwind state's significant contribution to nonattainment and
interference with maintenance of downwind air quality using a multi-factor test (described in the
preamble at section VI in step three of the 4-Step Good Neighbor Framework) which is based on
cost, emissions, and air quality factors. A key quantitative input for determining the amount of
each state's emission reduction obligation is the predicted downwind ambient air quality impacts
of upwind EGU emission reductions under the budgets at various levels of NOx emission control
and under a scenario of potential upwind non-EGU emissions reductions. See section C of this
TSD for information regarding EGUs and see preamble section VI for information about non-
EGUs. The emission reductions from the various cost thresholds can potentially result in air
quality improvements such that individual receptors drop below the level of the NAAQS based
on the cumulative air quality improvement from the upwind states, or potentially decrease each
upwind state's contributions such that they possibly drop below the 1% linkage threshold (used
in step two of the Good Neighbor Framework to identify the states for further analysis).

Air quality modeling would be the optimal way to examine these questions at each cost
threshold level from EGUs and non-EGUs. However, due to time and resource limitations EPA
was unable to use photochemical air quality modeling for all but a few emissions scenarios.
Therefore, in order to estimate the air quality impacts for the various levels of emission
reductions and to ensure that each step of its analysis is informed by the evolving emissions data,
EPA used a simplified air quality assessment tool (AQAT).31 The simplified tool allows the
Agency to analyze many more NOx emission budget levels than would otherwise be possible.
EPA recognizes that AQAT is not the equivalent of photochemical air quality
modeling. However, AQAT is directly informed by air quality modeling data. Further, AQAT
has evolved through iterative development under the original CSAPR and the CSAPR
Update. One such evolution is its calibration of the change in air quality based on air quality
modeling of a particular emission reduction scenario. As done at proposal, EPA examined
various cost threshold scenarios for the year 2024 using two different calibration factors as a
mechanism to estimate the range of results.

The inputs and outputs of the tool can be found in the "Ozone AQAT Final.xlsx" excel
workbook.

The remainder of section D of this document will:

•	Present an introduction and overview of the ozone AQAT;

•	Describe the construction of the ozone AQAT; and

•	Provide the results of the NOx emission cost threshold analyses.

1. Introduction: Development of the ozone AQAT

The ozone AQAT was developed for use in the rule's step three air quality analysis as
part of the multi-factor test. Specifically, the AQAT was designed to evaluate air quality

31 EPA used CAMx to model several base cases (i.e., one of 2016, one of 2023, and one of 2028). The EPA
calculated air quality contributions for each state for both the 2023 and 2028 cases. EPA did not explicitly model
2021.

27


-------
changes in response to emissions changes in order to quantify necessary emission reductions
under the good neighbor provision and to evaluate potential budgets for over-control as to either
the 1% threshold or the downwind receptor status. EPA described and used a similar tool in the
original CSAPR to evaluate good neighbor obligations with respect to the fine particulate matter
(PM2.5) NAAQS and in the CSAPR Update to evaluate good neighbor obligations with respect to
ozone. For the CSAPR Update, EPA refined both the construction and application of the
assessment tool for use in estimating changes in ozone concentrations in response to changes in
NOx emissions. We followed the methodology developed in the CSAPR Update rulemaking
where we calibrate the response of a pollutant using two CAMx simulations at different emission
levels.32'33 The construction of the AQAT for the final rule is essentially the same as that for the
proposed rule.

A critical factor in the assessment tool is the establishment of a relationship between
ozone season NOx emission reductions and reductions in ozone. Within AQAT, on a receptor-
by-receptor basis, we assume that the reduction of a ton of emissions of NOx from the upwind
state results in a particular level of improvement in air quality downwind.34 For the purposes of
developing and using an assessment tool to compare the air quality impacts of NOx emission
reductions under various emission cost threshold emission levels, we determine the relationship
between changes in emissions and changes in ozone contributions on a receptor-by-receptor
basis. Specifically, EPA assumed that, within the range of total NOx emissions being considered
(as defined by the cost threshold emission levels or changes from year-to-year), a change in
ozone season NOx emissions leads to a proportional change in downwind ozone contributions.35
This proportional relationship was then modified using calibration factors created using the 2023
base case contribution air quality modeling and either the 2016 base case (for cases between
2016 and 2023) or the 2028 base case (for cases from 2023 to 2028) to account for the majority
of the nonlinearity between emissions and ozone concentrations. For example, for a particular
receptor in 2022, we could assume that a 20% decrease in the upwind state's emissions leads to a

32	In CSAPR, we estimated changes in sulfate using changes in SO2 emissions.

33	In this rule, as was the case for the CSAPR Update, we used CAMx to calibrate the assessment tool's predicted
change in ozone concentrations to changes in NOx emissions. This calibration is receptor-specific and is based on
the changes in NOx emissions and resulting ozone concentrations between the 2023 base case and either the 2016
base case or the 2028 base case. One of these two calibration points (either 2016 or 2028) was used to create site-
specific calibration factors so that the response of ozone concentrations to upwind NOx emission changes would
more closely align with ozone estimates from CAMx. For time periods before 2023, we used the 2016 calibration
point, for 2023 and later, we used the 2028 calibration point.

34	This downwind air quality improvement is assumed to be indifferent to the source sector or the location of the
particular emission source within the state where the ton was reduced. For example, reducing one ton of NOx
emissions from the power sector is assumed to have the same downwind ozone reduction as reducing one ton of
NOx emissions from the mobile source sector.

35The relationship between NOx emissions and ozone concentrations is known to be non-linear when examined over
large ranges of NOx emissions (e.g., J.H. Seinfeld and S.N. Pandis, Atmospheric Chemistry and Physics From Air
Pollution to Climate Change, 2nd Edition, John Wiley and Sons, 2006, Hoboken, NJ, pp 236-237). However, for
smaller ranges of NOx and VOC emissions, while meteorological conditions are held constant, the relationship may
be reasonably linear. The nonlinearities are evident over tens of ppb of ozone changes with tens of percent changes
in the overall emission inventories. For most states examined here, under the various control scenarios, most
changes in the emission inventory are on the order of a few percent and most air quality changes are on the order of
a fraction of a ppb. In this assessment tool, we are assuming a linear relationship between NOx emissions and ozone
concentrations calibrated between two CAMx simulations. A significant portion of the nonlinearity is accounted for
by using the calibration factor and having the air quality estimates occur at levels of emissions between the 2023
base case and the other base case used in the calibration (which were both modeled in CAMx).

28


-------
20% decrease in its downwind ozone contribution in the "uncalibrated" ozone AQAT, while
following the application of the calibration factor (based on the change to 2016) it may only
decrease by 10% in "calibrated" AQAT (where the calibration factor is 0.5). Typically, the
calibration factors were substantially less than one, often on the order of 0.3 (thus, a 10%
decrease in emissions would result in a 3% decrease in ozone concentration). The creation of the
calibration factors is described in detail in section D.2.c (1) of this TSD.

In summary, because the tool is only being used over a range for which a calibration
factor has been developed, and because other options such as using CAMx to model all scenarios
is cost and time-prohibitive, EPA used ozone AQAT to estimate the downwind ozone reductions
due to upwind NOx emission reductions for the air quality input to the multi-factor test for this
rule. Other options, such as directly scaling the results (i.e., an "uncalibrated ozone AQAT")
will likely greatly overestimate the air quality impacts of emission reductions.36

Section D.2, below, is a technical explanation of the construction of the ozone AQAT.
Readers who prefer to access the results of the analysis using the ozone AQAT are directed to
section D.3.

2. Details on the construction of the ozone AQAT
(a) Overview of the ozone AQAT

This section describes the step-by-step development process for the ozone AQAT. All
the input and output data can be found in the Excel worksheets described in Appendix B. In the
ozone AQAT, EPA links state-by-state NOx emission reductions (derived from the
photochemical model, the non-EGU assessment and/or the IPM EGU modeling combined with
the EGU engineering assessment) with 2023 CAMx modeled ozone contributions in order to
predict ozone concentrations at different levels of emission levels at monitoring sites. The
reduction in ozone contributions for each year at each cost threshold level and the resulting air
quality improvement at monitoring sites with projected nonattainment and/or maintenance
problems were then considered in a multi-factor test for identifying the level of emissions
reductions that define significant contribution to nonattainment and interference with
maintenance.

In applying AQAT to analyze air quality improvements at a given receptor for the cost
threshold scenarios, emissions were reduced in only those upwind states that were "linked" to
that receptor in step 2 of the Good Neighbor Framework (i.e., those states that contributed an air
quality impact at or above 1 percent of the NAAQS). Emissions were also reduced in the state
that contained that receptor (regardless of the level of that state's contribution) at a level of
control stringency consistent with the budget level applied in upwind states.

Specifically, the key estimates from the ozone AQAT for each receptor are:

36 Directly scaling the results is the equivalent of using a second calibration point with an assumption that zero
emissions results in zero contribution. While clearly this is a reasonable assumption for that emission level, using
this in the calibration process assumes that the emission and air quality relationship is linear throughout the entire
ozone regime (e.g., from 0 ppb all the way up to 80 ppb or so). Clearly, there is important non-linearity over this
range.

29


-------
•	The ozone contribution as a function of emissions at each cost threshold level, for
each upwind state that is contributing above the 1 percent air quality threshold and the
state containing the receptor.

•	The ozone contribution under base case NOx emissions in the various years, for each
upwind state that is not above the 1 percent air quality threshold for that receptor.

•	The non-anthropogenic (i.e., background, boundary, biogenic, and wildfire) ozone
concentrations. These are assumed to be constant and equal to the contributions from
the 2023 base case source apportionment modeling.

The results of the ozone AQAT analysis for each emission cost threshold level for EGUs and
non-EGUs can be found in section D.3 of this document.

(b)	Data used to construct the ozone AQAT for this rule

Several air quality modeling and emissions inventory sources were used to construct the
calibrated ozone AQAT for this rule. Using the calibration factors, EPA modulated the 2023
CAMx ozone season contributions for each upwind state to each downwind receptor. These
modulations were enough to adjust the concentrations to represent a different year (e.g., 2021).
In all cases, the starting point was the 2023 base case CAMx run with contributions. For each
scenario, EPA multiplied each state's percent change in emissions relative to the 2023 base case
ozone season NOx emission inventories from all source sectors used in the source apportionment
CAMx air quality modeling (this includes all anthropogenic sources and excludes biogenic
sources and wildfires) by the receptor-specific calibration factor and the state's base case
contribution. Note that the 2023 scenario in CAMx used IPM emission estimates while the 2016
base case used EGU continuous emissions monitoring system (CEMS) data. The base case
emission inventories for the 2023 base case and the 2016 and 2028 base cases are discussed in
the Air Quality Modeling TSD. An additional emission inventory (i.e., for 2021) was also
developed. The EGU emissions for this inventory were replaced with emission estimates used
throughout Step 3. This emission inventory is described in the Emission Inventory TSD. The
ozone season NOx EGU and non-EGU emissions for each emission scenario including the base
case as modeled in AQAT are described in section C of this TSD.

As described in the Air Quality Modeling TSD, the air quality contributions and
emissions were modeled for all states in the contiguous United States and the District of
Columbia. Thus, in the ozone AQAT, any emission differences between the 2023 air quality
modeling base case and the scenario would result in changes in air quality contributions and
ozone concentrations at the downwind monitors.

(c)	Detailed outline of the process for constructing and utilizing the ozone AQAT

The ozone AQAT was created and used in a multi-step process. First, a calibrated ozone
AQAT was created using the contributions and emission inventory from the 2023 base case air
quality modeling as well as the 2016 base case (for all scenarios with years greater than or equal
to 2023, the calibrated AQAT used the design values and emission inventory from the 2028 base
case). The construction of this was identical to that from the proposal. For each emissions cost
threshold scenario evaluated, for each state, EPA identified the percent change in anthropogenic

30


-------
NOx emissions relative to the 2023 base case and multiplied this by the receptor-specific
calibration factor as well as by the state- and receptor-specific contribution. This resulted in a
state- and receptor-specific "change in contribution" relative to the 2023 base case. Each state's
change in contribution value was then added to (or subtracted from) its 2023 base case
contribution and the results summed for all states for each receptor. To this total of each state's
contribution to each receptor, the receptor-specific base case contributions from the other source-
categories were added (modified if necessary in the same way if necessary to adjust to a different
year), resulting in an estimated design value for each receptor.37 The calibrated ozone AQAT
was used to project the ozone concentrations for each NOx emission budget level on a receptor-
by-receptor basis for every monitor throughout the domain.

In order to facilitate understanding of the calibration process, EPA describes below a
demonstrative example: monitor number 090019003 in Fairfield County, Connecticut, with a
2023 base case projected ozone average design value of 76.9 parts per billion (ppb) and
maximum design value of 77.2 ppb.

(1) Create the calibration factors

The process for creating the calibration factors remains unchanged from the method used
in the CSAPR Update. This section repeats the process and data from the proposal. To create
the calibration factors, EPA used emissions, contributions, and design values from the 2023
CAMx run that used IPM for emissions, and the emissions and design values from either the
2016 or 2028 CAMx base cases. All changes in emissions and air quality are relative to the
proposed 2023 CAMx base case.

First, EPA used ozone season state-level 2023 base case total NOx emissions from all
source sectors. This emissions data is divided into multiple source sectors for the purposes of air
quality modeling: airports, beis, cmv_clc2_12, cmv_c3_12, nonpt, nonroad, np oilgas, onroad,
pt oilgas, ptagfire, ptegu, ptfire, ptnonipm, rail, rwc (see the Preparation of Emissions
Inventories for 2016vl North American Emissions Modeling Platform TSD for additional details
on the emissions inventories used in the CAMx air quality modeling). The anthropogenic state-
level total NOx emissions used in the air quality contribution modeling are the sum of emissions
from all these source sectors except (beis, ptagfire, and ptfire). Next, EPA summed the ozone
season total anthropogenic NOx emissions across all relevant source sectors for both the 2016
and 2028 base cases. EPA calculated the ratio of the emissions for each of these two base cases
to the total emissions for the 2023 base case for each state modeled in CAMx. More information
on the emissions inventories can be found in the preamble to the proposed rule. The total
emissions data and resulting ratios can be found in Table D-l and in the ozone AQAT worksheet
"calib_emiss".

For each monitor, the "uncalibrated" change in concentration was found by multiplying
each state's 2023 base case ozone air quality contribution by the difference in the state's ratio of
emissions. The difference in the ratio of emissions was calculated as the difference in total
ozone season anthropogenic NOx emissions between the either the 2016 base case (or the 2028
base case) and the 2023 base case scenario divided by the 2023 base case emission. Thus, when
the 2016 or 2028 base case had smaller emissions than the 2023 base case, the net result was a
negative number. Each state's fractional change in emissions was multiplied by its 2023 base

37 Details on procedures for calculating average and maximum design values can be found in the Air Quality
Modeling TSD.

31


-------
case contribution to get a state- specific change in contribution (Table D-l). For each monitor,
the change in concentrations was summed across all states. The result was the total
"uncalibrated" change in concentration.

32


-------
Table D-l. The 2023,2016, and 2028 Base Cases Total Anthropogenic NOx Emissions with 2023 Ozone
Contributions (ppb) and Uncalibrated Contributions for 2016 and 2023 for the Westport Monitor Number
090019003 in Fairfield County, Connecticut. 					









Fraction of 2016

Fraction of 2028



Uncalibrated

Uncalibrated







2023 Base

2016 Base

2028 Base

Base Case

Base Case

2023 Base

Change in

Change in

Uncalibrated

Uncalibrated

State

Case NOx

Case NOx

Case NOx

Emissions to 2023

Emissions to

Case Ozone

AQ

AQ

Contribution

Contribution



Emissions

Emissions

Emissions

Base Case
Emissions

2023 Base Case
Emissions

Contributions

Contribution
for 2016

Contribution
for 2028

2016

2028

AL

67,839

101,168

60,574

0.491

-0.107

0.113

0.056

-0.012

0.169

0.101

AZ

45,043

70,225

37,041

0.559

-0.178

0.016

0.009

-0.003

0.025

0.013

AR

46,552

68,756

40,093

0.477

-0.139

0.176

0.084

-0.024

0.259

0.151

CA

145,157

212,134

133,619

0.461

-0.079

0.035

0.016

-0.003

0.051

0.032

CO

61,473

86,684

55,852

0.410

-0.091

0.065

0.027

-0.006

0.091

0.059

CT

12,724

18,874

11,227

0.483

-0.118

2.682

1.296

-0.316

3.979

2.367

DE

6,985

10,193

6,274

0.459

-0.102

0.425

0.195

-0.043

0.620

0.382

DC

1,610

2,338

1,348

0.453

-0.163

0.041

0.019

-0.007

0.060

0.035

FL

114,045

186,866

99,721

0.639

-0.126

0.075

0.048

-0.009

0.123

0.065

GA

73,702

115,451

65,044

0.566

-0.117

0.165

0.093

-0.019

0.258

0.146

ID

19,924

29,416

16,389

0.476

-0.177

0.028

0.013

-0.005

0.041

0.023

IL

103,625

143,831

93,027

0.388

-0.102

0.798

0.310

-0.082

1.107

0.716

IN

82,323

129,702

73,428

0.576

-0.108

1.239

0.713

-0.134

1.952

1.105

IA

48,818

67,053

41,876

0.374

-0.142

0.169

0.063

-0.024

0.232

0.145

KS

69,568

91,022

61,361

0.308

-0.118

0.129

0.040

-0.015

0.168

0.114

KY

51,946

88,409

45,953

0.702

-0.115

0.854

0.599

-0.098

1.453

0.755

LA

105,245

138,804

98,692

0.319

-0.062

0.266

0.085

-0.017

0.351

0.250

ME

13,132

19,133

11,619

0.457

-0.115

0.007

0.003

-0.001

0.010

0.006

MD

27,785

43,974

24,694

0.583

-0.111

1.181

0.688

-0.131

1.869

1.050

MA

33,006

45,621

29,957

0.382

-0.092

0.079

0.030

-0.007

0.110

0.072

MI

87,686

118,418

81,141

0.350

-0.075

1.678

0.588

-0.125

2.266

1.553

MN

63,293

89,970

53,343

0.421

-0.157

0.188

0.079

-0.030

0.267

0.159

MS

33,963

56,364

30,491

0.660

-0.102

0.101

0.067

-0.010

0.168

0.091

MO

74,595

116,616

63,863

0.563

-0.144

0.356

0.200

-0.051

0.556

0.305

MT

28,901

40,605

24,884

0.405

-0.139

0.075

0.030

-0.010

0.105

0.064

NE

43,475

59,121

37,771

0.360

-0.131

0.076

0.028

-0.010

0.104

0.066

NV

19,070

30,601

15,844

0.605

-0.169

0.012

0.007

-0.002

0.020

0.010

NH

7,763

11,753

6,806

0.514

-0.123

0.018

0.009

-0.002

0.028

0.016

NJ

37,738

58,456

33,624

0.549

-0.109

8.446

4.637

-0.921

13.083

7.525

NM

53,165

70,102

47,356

0.319

-0.109

0.038

0.012

-0.004

0.051

0.034

NY

77,766

106,248

69,004

0.366

-0.113

14.141

5.179

-1.593

19.320

12.547

NC

76,448

101,378

64,646

0.326

-0.154

0.548

0.179

-0.085

0.727

0.464

ND

46,471

58,914

41,774

0.268

-0.101

0.080

0.021

-0.008

0.101

0.071

OH

98,579

141,543

87,637

0.436

-0.111

2.506

1.092

-0.278

3.598

2.228

OK

101,105

120,286

92,941

0.190

-0.081

0.195

0.037

-0.016

0.233

0.180

OR

31,443

46,235

26,228

0.470

-0.166

0.028

0.013

-0.005

0.041

0.023

PA

112,449

160,648

100,275

0.429

-0.108

6.723

2.882

-0.728

9.604

5.995

RI

4,742

6,994

4,106

0.475

-0.134

0.010

0.005

-0.001

0.015

0.009

SC

46,385

67,218

40,781

0.449

-0.121

0.177

0.079

-0.021

0.256

0.155

SD

13,753

21,784

10,688

0.584

-0.223

0.047

0.028

-0.011

0.075

0.037

TN

57,191

89,762

50,022

0.569

-0.125

0.318

0.181

-0.040

0.499

0.278

TX

314,342

418,972

288,147

0.333

-0.083

0.582

0.194

-0.049

0.776

0.534

UT

35,868

53,661

31,932

0.496

-0.110

0.033

0.016

-0.004

0.049

0.029

VT

4,047

6,078

3,340

0.502

-0.175

0.014

0.007

-0.002

0.021

0.011

VA

57,856

93,453

51,398

0.615

-0.112

1.273

0.783

-0.142

2.057

1.131

WA

58,962

89,605

49,093

0.520

-0.167

0.055

0.028

-0.009

0.083

0.046

wv

53,854

61,692

52,047

0.146

-0.034

1.459

0.212

-0.049

1.671

1.410

WI

48,283

75,140

41,731

0.556

-0.136

0.229

0.127

-0.031

0.357

0.198

WY

41,098

53,908

36,966

0.312

-0.101

0.078

0.024

-0.008

0.103

0.070

TRIBAL

5,858

17,759

5,875

2.032

0.003

0.004

0.009

0.000

0.013

0.004

CAN MEX







0

0

2.52895

0

0

2.52895

2.52895

OFFSHORE







0

0

0.67413

0

0

0.67413

0.67413

FIRE







0

0

0.34482

0

0

0.34482

0.34482

ICBC







0

0

20.63261

0

0

20.63261

20.63261

BIOG







0

0

4.68736

0

0

4.68736

4.68736

33


-------
Next, the estimate of the monitor specific ozone responses under the 2016 or 2028 base
cases was used to calibrate the ozone AQAT to CAMx and to derive the calibration factors. One
factor was created using the 2016 base and is applied to all scenarios from 2016 through 2022,
the other factor was created using the 2028 base and is applied to all scenarios from 2023 to
2028. First, the changes in ozone predicted by the ozone AQAT and CAMx for the average
design values were calculated for each monitor for the 2016 or 2028 base case relative to the
2023 base case concentrations. The change in ozone predicted by CAMx was then divided by
the change in ozone predicted by the uncalibrated AQAT, resulting in a monitor-specific
calibration factor (see Table D-2 for an example calculation using the 2016 and 2028 base cases).
The calculation of these monitor-specific calibration factors provided EPA with the ability to
align the ozone response predicted by the ozone AQAT to the ozone response predicted by
CAMx.

The ozone AQAT and CAMx concentration differences can be found in the "Ozone
AQAT Final.xlsx" excel workbook in columns BK and BL, respectively, on worksheets
"calib_2016" and "calib_2028". The resulting calibration factors can be found in column BM of
the aforementioned excel worksheets. The calibration factor, multiplied by the fractional change
in emissions (relative to the 2023) base and multiplied by the 2023 base air quality contribution,
results in the fractional change in air quality contribution for any alternative scenario.

Table D-2. Design Values in the 2023 Base Case and the 2016 and 2028 Base Cases and
Estimated Change in Design Value Relative to the 2023 Base Case from CAMx and
Uncalibrated AQAT for the Westport Monitor Number 090019003 in Fairfield County,
Connecticut.



2023 Base
Case
Concentration
(PPb)

2016 Base
Case
Concentration
(PPb)

2028 Base
Case
Concentration
(PPb)

Change in
Concentration
from 2016 to
2023
(PPb)

Change in
Concentration
from 2028 to
2023
(PPb)

CAMx

76.90

82.70

74.30

5.80

-2.60

Uncalibrated AQAT

76.90

98.04

71.70

21.14

-5.20

Calibration Factor -
Change in
Concentration from
CAMx Divided by
Change in

Concentration from the
Uncalibrated AQAT







0.2743

0.4998

(2) Create a calibrated version of the ozone AQAT for emission budget analysis for the proposed
rule

Next, as was done at proposal, EPA used 2023 base case emissions and 2023 base case
air quality ozone contributions from the air quality modeling along with either the 2016-based or
2028-based calibration factors to create a "calibrated" AQAT specific to the year being assessed.

34


-------
EPA examined the changes in the 2023 air quality contributions from changes in emissions
relative to the 2023 base case emissions (while using the calibration factor). This calibrated
AQAT was then used to estimate the change in predicted ozone due to NOx emission reductions
under each emission cost threshold level evaluated for EGUs and non-EGUs in each year.

First, as described in section VI of the preamble and above in section C of this TSD for
EGUs, EPA identified various cost threshold levels of emissions based on projected changes in
emissions rates and adjusted historical data. For each state, for each year, the total anthropogenic
NOx emissions (excluding the EGU emissions) are presented in Table D-3. These "straight line"
emissions inventories were created by linearly interpolating the emissions for all anthropogenic
source sectors except EGUs between 2023 and 2016 (or between 2023 and 2028). An additional
set of sensitivity analyses were done for 2021 and 2022 using the 2021 emission inventory
directly and by linearly interpolating the emissions for all source sectors (except EGUs) between
the 2021 and 2023 inventories.

The EGU point inventory is composed of emissions from units that report emissions to
EPA's Clean Air Markets Division (CAMD) under 40 CFR Part 75 (most emissions from these
sources are measured by CEMS) and units that are typically included in EPA's power sector
modeling using the Integrated Planning Model (IPM) but that do not report to CAMD and
typically lack CEMS (i.e., the nonCEM units). Within the air quality modeling platform,
different approaches are taken depending on whether an emissions inventory for EGUs is created
using an IPM-based emission estimates or an engineering analysis based platform. The nonCEM
components for various air quality model platforms are shown in Table D-4. For each year,
based on the available air quality modeling runs available (in 2016, 2023, and 2028) an
engineering-based nonCEM point EGU component was created (Table D-5). For the years from
2016 through 2023, this was a straight line linear interpolation of the 2016 and 2023 nonCEM
component from the engineering based air quality runs. For years 2023 through 2028, we used
2023 nonCEM values held constant for all years. The component of the EGU point inventory
from CAMD reporting units (labelled "CEMs") was developed using engineering analysis (see
section C for details). For each year, we show EGU emissions for units with CEMs as a
function of cost threshold level (see Tables C-l through C-5 for the years 2021 through 2025,
respectively). These levels include:

•	Engineering Baseline,

•	$500/ton,

•	Optimize SCR,

•	Optimize SCR + State-of-the-Art Combustion Controls (referred to as Low NOx
Burners, or LNB),

•	Optimize SNCR+ SCR ,

•	Optimize SNCR+ SCR + LNB ,

•	New SNCR + Optimize SNCR+ SCR + LNB

•	New SCR + Optimize SNCR+ SCR + LNB.

In the construction of AQAT, for each scenario, we assembled an emission inventory
from all anthropogenic sources for each state. In other words, we combine the year-specific
anthropogenic emissions from Table D-3, with the relevant EGU nonCEM component from D-5,
and one of the EGU CEM estimates from Tables C-l through C-5.

Finally, these emission totals are compared to the 2023 case that was modeled with
CAMx. For each emission cost threshold level in each analysis year, EPA calculated the ratio of

35


-------
the emission differences from the scenario and the 2023 air quality modeling base case to the
total NOx emissions for the 2023 air quality modeling base case used in the air quality modeling
for each state (see Tables D-6 through D-10). Scenarios that are not viable, for technical or
policy reasons, have been grayed out in these tables.

For each year, we also created a complete "straight line" emissions inventory including a
linear interpolation of the EGU inventory from the air quality modeling between 2023 and 2016
and between 2023 and 2028. For the sensitivity analysis, the interpolation was between 2021
and 2023. The emission differences and air quality estimates for the two sensitivity scenarios
can be found in Appendix H. In Table D-l 1, we examined the emission reduction for non-EGUs
in tranche 1 for glass and cement controls below $2,000 per ton, and then estimated the ratio of
the emission difference relative to the 2023 air quality modeling base case.

For each cost threshold level analyzed, on a receptor-by-receptor basis, the emissions
change for each upwind state is associated with one of two emission levels (either the
engineering base case emission level for that year or the particular cost threshold level)
depending on whether the upwind state is "linked" to that receptor or if the receptor is located
within the state. States that are contributing above the air quality threshold (i.e., greater than or
equal to 1 percent of the NAAQS) to the monitor, as well as the state containing the monitor,
make NOx emission reductions available at the particular cost threshold level for that year. The
emissions for all other states are adjusted to the engineering base case level for that year.

For the $l,800/ton control case at various years, all states that were linked to any receptor
in 2021 were simultaneously adjusted to the emission levels in the control case, regardless of
whether (or not) the state was "linked" to a particular receptor. This scenario examines the
emission results when budgets have been applied to the geography. For each monitor, the
predicted change in contribution of ozone from each state is calculated by multiplying the state-
specific 2023 base case ozone contributions from the air quality modeling by the calibration
factor as well as by the ratio of the change in emissions (Tables D-l and D-6 through D-10, for
either the emission cost threshold level or the engineering base case emission level depending on
whether the state is linked).38 This calibrated change in ozone is then added to the ozone
contribution from the 2023 base case air quality modeling. The result is the state and receptor
specific "calibrated" total ozone contribution after implementation of the emission at a particular
cost threshold level.

For each monitor, these state-level "calibrated" contributions are then summed to
estimate total ozone contribution from the states to a particular receptor in the CAMx modeling
domain. Finally, "other" modeled ozone contributions ("CAN MEX", "OFFSHORE",
"USCANMEX FIRE", "ICBC", and "BIOG") are added from the 2023 base case air quality
modeling to the state contributions to account for other sources of ozone affecting the modeling
domain. The total ozone from all the states and "other" contributions equals the average design
values estimated in the assessment tool. The maximum design values were estimated by
multiplying the estimated average design values by the ratio of the modeled 2023 base case
maximum to average design values.

Generally, as the emission cost threshold stringency increased, the estimated average and
maximum design values at each receptor decreased. In the assessment tool, the estimated value
of the average design value was used to estimate whether the location will be out of attainment,

38 The change in concentration can be positive or negative, depending on whether the state's total anthropogenic
ozone season NOx emissions for the scenario are larger or smaller than the 2023 air quality modeling base case
emission level.

36


-------
while the estimated maximum design value was used to estimate whether the location will have
problems maintaining the NAAQS. The area was noted as having a nonattainment or
maintenance issue if either estimated air quality level was greater than or equal to 76 ppb.

37


-------
Table D-3. Ozone Season Anthropogenic NOx Emissions (Tons) without EGUs for Each

State.

State

2021

2022

2023

2024

2025

2026

2021
Sensitivity

2022
Sensitivity

Alabama

69,987

66,163

62,338

60,603

58,868

57,134

66,126

64,232

Arizona

45,793

42,719

39,644

38,220

36,796

35,371

42,094

40,869

Arkansas

42,259

39,586

36,914

35,793

34,672

33,551

40,275

38,594

California

161,697

152,246

142,795

140,823

138,852

136,880

155,951

149,373

Colorado

59,562

56,774

53,986

52,934

51,881

50,829

55,717

54,852

Connecticut

12,579

11,683

10,787

10,496

10,205

9,914

11,598

11,193

Delaware

7,443

7,053

6,664

6,514

6,365

6,216

7,046

6,855

District of Columbia

1,817

1,713

1,609

1,556

1,504

1,452

1,678

1,643

Florida

113,717

105,687

97,657

94,900

92,143

89,385

101,801

99,729

Georgia

77,170

71,638

66,107

64,178

62,248

60,319

71,469

68,788

Idaho

22,332

21,039

19,745

19,039

18,333

17,627

21,057

20,401

Illinois

103,427

98,260

93,094

90,938

88,782

86,626

97,378

95,236

Indiana

73,016

68,763

64,511

62,670

60,829

58,988

69,443

66,977

Iowa

42,556

39,903

37,250

35,915

34,580

33,246

40,668

38,959

Kansas

67,361

64,151

60,941

59,345

57,749

56,153

64,927

62,934

Kentucky

49,241

46,459

43,676

42,486

41,296

40,106

46,190

44,933

Louisiana

101,112

98,267

95,423

94,142

92,862

91,582

99,116

97,269

Maine

13,285

12,535

11,786

11,486

11,186

10,886

12,544

12,165

Maryland

28,862

26,913

24,963

24,334

23,705

23,076

25,561

25,262

Massachusetts

33,806

32,138

30,469

29,869

29,269

28,669

31,101

30,785

Michigan

81,760

77,760

73,761

72,229

70,696

69,163

77,274

75,518

Minnesota

62,574

58,958

55,342

53,789

52,236

50,683

61,433

58,387

Mississippi

35,723

33,389

31,055

30,299

29,543

28,787

33,414

32,234

Missouri

68,437

63,937

59,436

57,302

55,169

53,035

64,699

62,068

Montana

27,412

26,140

24,868

24,079

23,290

22,501

27,238

26,053

Nebraska

38,582

36,322

34,063

32,847

31,631

30,414

37,204

35,634

Nevada

21,229

19,836

18,443

17,822

17,202

16,581

19,156

18,799

New Hampshire

8,429

7,948

7,466

7,280

7,093

6,906

7,803

7,634

New Jersey

41,044

38,117

35,189

34,314

33,439

32,564

37,016

36,102

New Mexico

52,452

50,678

48,905

47,794

46,684

45,573

51,454

50,179

New York

78,610

74,563

70,517

68,874

67,231

65,587

71,570

71,043

North Carolina

68,043

63,994

59,944

58,207

56,469

54,732

63,635

61,790

North Dakota

37,522

36,370

35,218

34,332

33,446

32,561

38,415

36,816

Ohio

90,701

85,504

80,307

78,080

75,853

73,626

86,304

83,306

Oklahoma

96,329

94,061

91,794

89,825

87,856

85,887

95,756

93,775

Oregon

34,601

32,676

30,751

29,716

28,681

27,646

32,399

31,575

Pennsylvania

106,545

102,733

98,920

96,913

94,906

92,899

98,613

98,767

Rhode Island

5,095

4,739

4,384

4,258

4,133

4,007

4,661

4,522

South Carolina

45,792

42,877

39,963

38,799

37,636

36,472

42,507

41,235

South Dakota

15,556

14,414

13,273

12,653

12,032

11,412

14,992

14,132

Tennessee

61,367

57,590

53,813

52,363

50,913

49,462

56,856

55,335

Texas

297,010

283,927

270,845

265,662

260,480

255,297

287,305

279,075

Utah

33,095

31,333

29,572

28,803

28,034

27,266

30,909

30,240

Vermont

4,583

4,311

4,038

3,897

3,756

3,614

4,332

4,185

Virginia

61,278

57,313

53,347

51,893

50,439

48,985

55,980

54,664

Washington

65,990

62,147

58,305

56,338

54,371

52,404

62,042

60,173

West Virginia

37,555

37,047

36,540

36,056

35,573

35,089

37,819

37,179

Wisconsin

50,430

47,071

43,713

42,447

41,182

39,917

47,763

45,738

Wyoming

34,845

34,165

33,486

33,011

32,536

32,061

34,546

34,016

Tribal Data

2,742

2,743

2,744

2,754

2,764

2,773

2,723

2,734

38


-------
Table D-4. EGU Point Source NOx Emissions (Tons) from Units without CEMs from AQ

Modeling Inventory.

State

2016 Eng EGU

2023 IPM

EGU
nonCEMs

2023 Eng EGU

2028 IPM EGU

nonCEMs

nonCEMs

nonCEMs

Alabama

482

473

200

450

Arizona

367

1,012

377

1,117

Arkansas

141

526

87

528

California

1,972

1,674

1,968

444

Colorado

334

604

277

998

Connecticut

1,272

1,759

1,362

1,788

Delaware

80

131

80

142

District of Columbia

0

0

0

0

Florida

6,189

8,376

6,466

8,569

Georgia

1,580

838

1,640

856

Idaho

528

164

413

161

Illinois

55

1,070

49

1,128

Indiana

611

1,165

722

1,115

Iowa

635

797

618

880

Kansas

109

1,001

93

653

Kentucky

1

366

1

561

Louisiana

3,885

1,943

3,908

1,964

Maine

1,972

1,280

1,908

1,275

Maryland

901

1,930

924

1,983

Massachusetts

2,363

2,044

2,349

2,046

Michigan

1,367

3,825

1,367

3,939

Minnesota

1,740

1,556

1,502

1,522

Mississippi

1,726

959

1,341

952

Missouri

471

331

456

349

Montana

933

3

933

3

Nebraska

665

750

664

748

Nevada

155

268

155

253

New Hampshire

374

215

374

206

New Jersey

1,083

1,844

1,022

1,955

New Mexico

98

72

98

88

New York

1,996

5,068

2,094

5,142

North Carolina

740

1,559

862

1,827

North Dakota

156

116

14

121

Ohio

722

1,881

981

1,961

Oklahoma

1

753

277

695

Oregon

712

515

712

515

Pennsylvania

2,187

5,945

2,543

5,573

Rhode Island

35

313

35

308

South Carolina

604

647

698

834

South Dakota

30

23

30

24

Tennessee

7

510

116

508

Texas

1,996

5,101

2,026

4,623

Utah

561

109

48

91

Vermont

61

9

0

9

Virginia

2,995

2,772

2,996

2,962

Washington

1,536

565

1,503

550

West Virginia

1

0

1

0

Wisconsin

61

612

92

617

Wyoming

11

0

0

0

Tribal Data

50

455

71

3,080

39


-------
Table D-5. EGU Point Source NOx Emissions (Tons) from Units without CEMs Adjusted

by Year.

State

2021

2022

2023

2024

2025

2026

2021
Sensitivity

2022
Sensitivity

Alabama

280

240

200

200

200

200

280

240

Arizona

374

376

377

377

377

377

374

376

Arkansas

102

95

87

87

87

87

102

95

California

1,969

1,969

1,968

1,968

1,968

1,968

1,969

1,969

Colorado

294

286

277

277

277

277

294

286

Connecticut

1,337

1,349

1,362

1,362

1,362

1,362

1,337

1,349

Delaware

80

80

80

80

80

80

80

80

District of Columbia

0

0

0

0

0

0

0

0

Florida

6,387

6,426

6,466

6,466

6,466

6,466

6,387

6,426

Georgia

1,623

1,631

1,640

1,640

1,640

1,640

1,623

1,631

Idaho

446

429

413

413

413

413

446

429

Illinois

50

50

49

49

49

49

50

50

Indiana

690

706

722

722

722

722

690

706

Iowa

623

621

618

618

618

618

623

621

Kansas

98

95

93

93

93

93

98

95

Kentucky

1

1

1

1

1

1

1

1

Louisiana

3,902

3,905

3,908

3,908

3,908

3,908

3,902

3,905

Maine

1,926

1,917

1,908

1,908

1,908

1,908

1,926

1,917

Maryland

918

921

924

924

924

924

918

921

Massachusetts

2,353

2,351

2,349

2,349

2,349

2,349

2,353

2,351

Michigan

1,367

1,367

1,367

1,367

1,367

1,367

1,367

1,367

Minnesota

1,570

1,536

1,502

1,502

1,502

1,502

1,570

1,536

Mississippi

1,451

1,396

1,341

1,341

1,341

1,341

1,451

1,396

Missouri

460

458

456

456

456

456

460

458

Montana

933

933

933

933

933

933

933

933

Nebraska

664

664

664

664

664

664

664

664

Nevada

155

155

155

155

155

155

155

155

New Hampshire

374

374

374

374

374

374

374

374

New Jersey

1,039

1,031

1,022

1,022

1,022

1,022

1,039

1,031

New Mexico

98

98

98

98

98

98

98

98

New York

2,066

2,080

2,094

2,094

2,094

2,094

2,066

2,080

North Carolina

827

845

862

862

862

862

827

845

North Dakota

54

34

14

14

14

14

54

34

Ohio

907

944

981

981

981

981

907

944

Oklahoma

198

237

277

277

277

277

198

237

Oregon

712

712

712

712

712

712

712

712

Pennsylvania

2,441

2,492

2,543

2,543

2,543

2,543

2,441

2,492

Rhode Island

35

35

35

35

35

35

35

35

South Carolina

671

684

698

698

698

698

671

684

South Dakota

30

30

30

30

30

30

30

30

Tennessee

85

100

116

116

116

116

85

100

Texas

2,017

2,021

2,026

2,026

2,026

2,026

2,017

2,021

Utah

195

122

48

48

48

48

195

122

Vermont

18

9

0

0

0

0

18

9

Virginia

2,996

2,996

2,996

2,996

2,996

2,996

2,996

2,996

Washington

1,513

1,508

1,503

1,503

1,503

1,503

1,513

1,508

West Virginia

1

1

1

1

1

1

1

1

Wisconsin

83

88

92

92

92

92

83

88

Wyoming

3

2

0

0

0

0

3

2

Tribal Data

65

68

71

71

71

71

65

68

40


-------
Table D-6. 2021 Fractional Difference in Emissions Relative to 2023 Air Quality Modeling
Base Case for Each State.

State

Eng

Baseline

$500/ton

Optimize
SCR



Optimize

SNCR+
SCR







Straight
Line

Interpolatio
n

Alabama

0.151

0.151

0.151



0.151







0.140

Arizona

0.145

0.138

0.127



0.127







0.160

Arkansas

0.098

0.096

0.097



0.097







0.136

California

0.135

0.135

0.135



0.135







0.132

Colorado

0.095

0.095

0.095



0.095







0.117

Connecticut

0.121

0.119

0.118



0.118







0.138

Delaware

0.109

0.109

0.107



0.106







0.131

District of
Columbia

0.129

0.129

0.129



0.129







0.129

Florida

0.187

0.187

0.175



0.174







0.182

Georgia

0.175

0.175

0.175



0.175







0.162

Idaho

0.153

0.153

0.153



0.153







0.136

Illinois

0.089

0.089

0.087



0.086







0.111

Indiana

0.088

0.086

0.054



0.054







0.164

Iowa

0.060

0.058

0.043



0.043







0.107

Kansas

0.057

0.057

0.047



0.046







0.088

Kentucky

0.248

0.248

0.243



0.242







0.201

Louisiana

0.145

0.144

0.144



0.139







0.091

Maine

0.163

0.163

0.163



0.163







0.131

Maryland

0.126

0.126

0.126



0.126







0.166

Massachusetts

0.106

0.106

0.106



0.106







0.109

Michigan

0.107

0.098

0.093



0.093







0.100

Minnesota

0.108

0.106

0.100



0.100







0.120

Mississippi

0.332

0.332

0.332



0.332







0.188

Missouri

0.090

0.090

0.076



0.075







0.161

Montana

0.104

0.104

0.104



0.104







0.116

Nebraska

0.089

0.087

0.088



0.088







0.103

Nevada

0.249

0.217

0.198



0.198







0.173

New Hampshire

0.184

0.184

0.173



0.173







0.147

New Jersey

0.151

0.151

0.148



0.148







0.157

New Mexico

0.076

0.075

0.073



0.073







0.091

New York

0.082

0.082

0.081



0.081







0.105

North Carolina

0.109

0.108

0.048



0.046







0.093

North Dakota

0.064

0.063

0.062



0.048







0.077

Ohio

0.090

0.086

0.028



0.028







0.125

Oklahoma

0.043

0.043

0.041



0.041







0.054

Oregon

0.134

0.134

0.134



0.134







0.134

Pennsylvania

0.075

0.074

0.044



0.044







0.122

Rhode Island

0.131

0.131

0.131



0.131







0.136

South Carolina

0.109

0.109

0.084



0.084







0.128

South Dakota

0.176

0.176

0.176



0.176







0.167

Tennessee

0.154

0.154

0.151



0.151







0.163

Texas

0.094

0.091

0.086



0.086







0.095

Utah

0.116

0.064

0.063



0.063







0.142

Vermont

0.150

0.150

0.150



0.150







0.143

Virginia

0.192

0.191

0.191



0.189







0.176

Washington

0.172

0.172

0.172



0.172







0.148

West Virginia

-0.021

-0.024

-0.049



-0.055







0.042

Wisconsin

0.155

0.152

0.149



0.149







0.159

Wyoming

0.127

0.127

0.124



0.124







0.089

Tribal Data

0.079

0.079

0.079



0.079







0.580

Note: Scenarios that are not viable, for technical r	icy reasons, have been grayec 	

41


-------
Table D-7. 2022 Fractional Difference in Emissions Relative to 2023 Air Quality Modeling
Base Case for Each State.

State

Eng

Baseline

$500/ton

Optimize

SCR

Optimize

SCR + LNB

Optimize

SNCR+
SCR

Optimize

SNCR+
SCR+LNB

New SNCR
+ Optimize

SNCR+
SCR + LNB

New SCR +
Optimize

SNCR+
SCR+LNB

Straight
Line

Interpolatio
n

Alabama

0.094

0.094



0.092



0.092





0.070

Arizona

0.076

0.070



0.059



0.059





0.080

Arkansas

0.040

0.038



0.039



0.039





0.068

California

0.070

0.070



0.070



0.070





0.066

Colorado

0.050

0.050



0.050



0.049





0.059

Connecticut

0.051

0.049



0.048



0.048





0.069

Delaware

0.053

0.053



0.050



0.050





0.066

District of
Columbia

0.064

0.064

0.064

0.064

0.064

0.064

0.064

0.064

0.065

Florida

0.114

0.114



0.103



0.102





0.091

Georgia

0.100

0.100



0.100



0.100





0.081

Idaho

0.088

0.088



0.088



0.088





0.068

Illinois

0.039

0.039



0.037



0.037





0.055

Indiana

0.031

0.029



-0.003



-0.003





0.082

Iowa

0.006

0.003



-0.013



-0.013





0.053

Kansas

0.011

0.011



0.001



0.000





0.044

Kentucky

0.194

0.195



0.165



0.165





0.100

Louisiana

0.118

0.117



0.117



0.112





0.046

Maine

0.106

0.106



0.106



0.106





0.065

Maryland

0.047

0.047



0.047



0.047





0.083

Massachusetts

0.055

0.055



0.055



0.055





0.055

Michigan

0.056

0.047



0.043



0.043





0.050

Minnesota

0.049

0.047



0.041



0.041





0.060

Mississippi

0.262

0.262



0.252



0.252





0.094

Missouri

0.030

0.029



0.015



0.014





0.080

Montana

0.049

0.049



0.049



0.049





0.058

Nebraska

0.037

0.035



0.024



0.024





0.051

Nevada

0.127

0.097



0.078



0.078





0.086

New Hampshire

0.122

0.122



0.111



0.111





0.073

New Jersey

0.073

0.073



0.071



0.071





0.078

New Mexico

0.043

0.042



0.040



0.039





0.046

New York

0.030

0.030



0.029



0.029





0.052

North Carolina

0.049

0.047



-0.012



-0.014





0.047

North Dakota

0.039

0.038



0.037



0.023





0.038

Ohio

0.039

0.035



-0.024



-0.024





0.062

Oklahoma

0.021

0.020



0.019



0.019





0.027

Oregon

0.073

0.073



0.073



0.073





0.067

Pennsylvania

0.042

0.041



0.010



0.010





0.061

Rhode Island

0.056

0.056



0.056



0.056





0.068

South Carolina

0.046

0.046



0.022



0.022





0.064

South Dakota

0.093

0.093



0.093



0.093





0.083

Tennessee

0.088

0.088



0.085



0.085





0.081

Texas

0.052

0.049



0.044



0.044





0.048

Utah

0.065

0.013



0.012



0.012





0.071

Vermont

0.080

0.080



0.080



0.080





0.072

Virginia

0.116

0.116



0.111



0.110





0.088

Washington

0.107

0.107



0.107



0.107





0.074

West Virginia

-0.030

-0.033



-0.067



-0.073





0.021

Wisconsin

0.080

0.077



0.074



0.074





0.079

Wyoming

0.097

0.097



0.076



0.076





0.045

Tribal Data

0.080

0.080



0.080



0.080





0.290

42


-------
Table D-8. 2023 Fractional Difference in Emissions Relative to 2023 Air Quality Modeling
Base Case for Each State.

State

Eng

Baseline

$500/ton

Optimize

SCR

Optimize

SCR+LNB

Optimize

SNCR+
SCR

Optimize

SNCR+
SCR+LNB

New SNCR
+ Optimize

SNCR+
SCR+LNB

New SCR +
Optimize

SNCR+
SCR + LNB

Straight
Line

Interpolatio
n

Alabama

0.037

0.037



0.035



0.035





0.000

Arizona

0.008

0.002



-0.009



-0.009





0.000

Arkansas

-0.018

-0.019



-0.018



-0.018





0.000

California

0.005

0.005



0.005



0.005





0.000

Colorado

-0.009

-0.009



-0.009



-0.009





0.000

Connecticut

-0.018

-0.021



-0.021



-0.021





0.000

Delaware

-0.003

-0.003



-0.005



-0.006





0.000

District of
Columbia

-0.001

-0.001

-0.001

-0.001

-0.001

-0.001

-0.001

-0.001

0.000

Florida

0.040

0.040



0.028



0.028





0.000

Georgia

0.016

0.016



0.016



0.016





0.000

Idaho

0.022

0.022



0.022



0.022





0.000

Illinois

-0.020

-0.020



-0.021



-0.022





0.000

Indiana

-0.021

-0.023



-0.055



-0.055





0.000

Iowa

-0.068

-0.070



-0.086



-0.086





0.000

Kansas

-0.036

-0.036



-0.045



-0.046





0.000

Kentucky

0.141

0.141



0.111



0.111





0.000

Louisiana

0.091

0.090



0.090



0.085





0.000

Maine

0.048

0.048



0.048



0.048





0.000

Maryland

-0.023

-0.023



-0.023



-0.023





0.000

Massachusetts

0.004

0.004



0.004



0.004





0.000

Michigan

-0.016

-0.025



-0.029



-0.029





0.000

Minnesota

-0.028

-0.030



-0.036



-0.036





0.000

Mississippi

0.191

0.191



0.182



0.182





0.000

Missouri

-0.034

-0.035



-0.049



-0.050





0.000

Montana

0.005

0.005



0.005



0.005





0.000

Nebraska

-0.015

-0.017



-0.028



-0.028





0.000

Nevada

0.048

0.019



0.000



0.000





0.000

New Hampshire

0.060

0.060



0.049



0.049





0.000

New Jersey

-0.005

-0.005



-0.007



-0.007





0.000

New Mexico

-0.046

-0.047



-0.048



-0.048





0.000

New York

-0.022

-0.022



-0.022



-0.022





0.000

North Carolina

-0.004

-0.005



-0.065



-0.067





0.000

North Dakota

-0.045

-0.045



-0.046



-0.060





0.000

Ohio

-0.014

-0.017



-0.076



-0.076





0.000

Oklahoma

-0.001

-0.002



-0.003



-0.003





0.000

Oregon

0.012

0.012



0.012



0.012





0.000

Pennsylvania

0.008

0.007



-0.023



-0.023





0.000

Rhode Island

-0.019

-0.019



-0.019



-0.019





0.000

South Carolina

-0.016

-0.016



-0.041



-0.041





0.000

South Dakota

0.010

0.010



0.010



0.010





0.000

Tennessee

0.022

0.022



0.019



0.019





0.000

Texas

0.010

0.007



0.002



0.002





0.000

Utah

0.013

-0.039



-0.039



-0.039





0.000

Vermont

0.011

0.010



0.010



0.010





0.000

Virginia

0.049

0.049



0.044



0.043





0.000

Washington

0.042

0.042



0.042



0.042





0.000

West Virginia

-0.040

-0.043



-0.076



-0.082





0.000

Wisconsin

0.008

0.005



0.002



0.002





0.000

Wyoming

0.066

0.066



0.045



0.045





0.000

Tribal Data

0.081

0.081



0.081



0.081





0.000

43


-------
Table D-9. 2024 Fractional Difference in Emissions Relative to 2023 Air Quality Modeling
Base Case for Each State.

State

Eng

Baseline

$500/ton

Optimize

SCR

Optimize

SCR + LNB

Optimize

SNCR+
SCR

Optimize

SNCR+
SCR + LNB

New SNCR
+ Optimize

SNCR+
SCR+LNB

New SCR +
Optimize

SNCR+
SCR + LNB

Straight
Line

Interpolatio
n

Alabama

0.011

0.011



0.010



0.010





-0.021

Arizona

-0.023

-0.030



-0.041



-0.041





-0.036

Arkansas

-0.042

-0.043



-0.042



-0.042





-0.028

California

-0.009

-0.009



-0.009



-0.009





-0.016

Colorado

-0.038

-0.038



-0.038



-0.038





-0.018

Connecticut

-0.041

-0.043



-0.044



-0.044





-0.024

Delaware

-0.024

-0.024



-0.027



-0.027





-0.020

District of
Columbia

-0.033

-0.033

-0.033

-0.033

-0.033

-0.033

-0.033

-0.033

-0.033

Florida

0.016

0.016



0.004



0.004





-0.025

Georgia

-0.010

-0.010



-0.010



-0.010





-0.023

Idaho

-0.013

-0.013



-0.013



-0.013





-0.035

Illinois

-0.042

-0.042



-0.043



-0.044





-0.020

Indiana

-0.081

-0.083



-0.114



-0.114





-0.022

Iowa

-0.095

-0.097



-0.113



-0.113





-0.028

Kansas

-0.059

-0.059



-0.068



-0.069





-0.024

Kentucky

0.118

0.118



0.089



0.088





-0.023

Louisiana

0.079

0.078



0.078



0.072





-0.012

Maine

0.025

0.025



0.025



0.025





-0.023

Maryland

-0.042

-0.042



-0.042



-0.042





-0.022

Massachusetts

-0.014

-0.014



-0.014



-0.014





-0.018

Michigan

-0.036

-0.044



-0.049



-0.049





-0.015

Minnesota

-0.053

-0.055



-0.060



-0.060





-0.031

Mississippi

0.169

0.169



0.159



0.159





-0.020

Missouri

-0.063

-0.063



-0.077



-0.078





-0.029

Montana

-0.022

-0.022



-0.022



-0.022





-0.028

Nebraska

-0.060

-0.062



-0.061



-0.061





-0.026

Nevada

0.015

-0.014



-0.032



-0.032





-0.034

New Hampshire

0.036

0.036



0.025



0.025





-0.025

New Jersey

-0.028

-0.028



-0.030



-0.030





-0.022

New Mexico

-0.067

-0.068



-0.069



-0.069





-0.022

New York

-0.043

-0.043



-0.044



-0.044





-0.023

North Carolina

-0.027

-0.028



-0.088



-0.090





-0.031

North Dakota

-0.064

-0.064



-0.065



-0.079





-0.020

Ohio

-0.036

-0.040



-0.099



-0.099





-0.022

Oklahoma

-0.020

-0.021



-0.023



-0.023





-0.016

Oregon

-0.021

-0.021



-0.021



-0.021





-0.033

Pennsylvania

-0.010

-0.011



-0.041



-0.041





-0.022

Rhode Island

-0.045

-0.045



-0.045



-0.045





-0.027

South Carolina

-0.043

-0.043



-0.067



-0.067





-0.024

South Dakota

-0.035

-0.035



-0.036



-0.036





-0.045

Tennessee

-0.003

-0.003



-0.006



-0.006





-0.025

Texas

-0.011

-0.014



-0.018



-0.019





-0.017

Utah

-0.008

-0.060



-0.061



-0.061





-0.022

Vermont

-0.024

-0.024



-0.024



-0.024





-0.035

Virginia

0.018

0.018



0.013



0.012





-0.022

Washington

0.008

0.008



0.008



0.008





-0.033

West Virginia

-0.049

-0.052



-0.085



-0.091





-0.007

Wisconsin

-0.033

-0.035



-0.038



-0.038





-0.027

Wyoming

0.055

0.055



0.034



0.034





-0.020

Tribal Data

0.082

0.082



0.082



0.082





0.001

44


-------
Table D-10. 2025 Fractional Difference in Emissions Relative to 2023 Air Quality

Modeling Base Case for

Cach State.

State

Eng

Baseline

$500/ton

Optimize

SCR

Optimize

SCR+LNB

Optimize

SNCR+
SCR

Optimize

SNCR+
SCR + LNB

New SNCR
+ Optimize

SNCR+
SCR + LNB

New SCR +
Optimize

SNCR+
SCR+LNB

Straight
Line

Interpolatio
n

Alabama

-0.015

-0.015



-0.016



-0.016

-0.016

-0.019

-0.043

Arizona

-0.055

-0.062



-0.072



-0.072

-0.068

-0.088

-0.071

Arkansas

-0.066

-0.067



-0.066



-0.066

-0.111

-0.153

-0.055

California

-0.022

-0.022



-0.023



-0.023

-0.023

-0.023

-0.032

Colorado

-0.055

-0.055



-0.055



-0.055

-0.069

-0.089

-0.037

Connecticut

-0.064

-0.066



-0.067



-0.067

-0.067

-0.067

-0.047

Delaware

-0.046

-0.046



-0.048



-0.048

-0.049

-0.049

-0.041

District of
Columbia

-0.066

-0.066

-0.066

-0.066

-0.066

-0.066

-0.066

-0.066

-0.065

Florida

-0.013

-0.013



-0.025



-0.026

-0.031

-0.037

-0.050

Georgia

-0.036

-0.036



-0.036



-0.036

-0.037

-0.037

-0.047

Idaho

-0.049

-0.049



-0.049



-0.049

-0.049

-0.049

-0.071

Illinois

-0.063

-0.063



-0.064



-0.065

-0.073

-0.076

-0.041

Indiana

-0.104

-0.106



-0.136



-0.136

-0.144

-0.150

-0.043

Iowa

-0.122

-0.125



-0.141



-0.141

-0.172

-0.221

-0.057

Kansas

-0.081

-0.082



-0.091



-0.092

-0.099

-0.116

-0.047

Kentucky

0.075

0.075



0.052



0.052

0.022

-0.022

-0.046

Louisiana

0.067

0.066



0.066



0.060

0.056

0.039

-0.025

Maine

0.002

0.002



0.002



0.002

0.002

0.002

-0.046

Maryland

-0.065

-0.065



-0.065



-0.065

-0.072

-0.072

-0.044

Massachusetts

-0.032

-0.032



-0.032



-0.033

-0.033

-0.033

-0.037

Michigan

-0.053

-0.062



-0.066



-0.066

-0.079

-0.094

-0.030

Minnesota

-0.077

-0.079



-0.085



-0.085

-0.103

-0.110

-0.063

Mississippi

0.147

0.147



0.137



0.137

0.118

0.099

-0.041

Missouri

-0.091

-0.092



-0.106



-0.107

-0.110

-0.130

-0.058

Montana

-0.049

-0.049



-0.049



-0.049

-0.077

-0.108

-0.056

Nebraska

-0.088

-0.090



-0.089



-0.089

-0.120

-0.160

-0.052

Nevada

-0.017

-0.046



-0.065



-0.065

-0.072

-0.085

-0.068

New Hampshire

0.012

0.012



0.000



0.000

0.000

0.000

-0.049

New Jersey

-0.051

-0.051



-0.054



-0.054

-0.054

-0.054

-0.044

New Mexico

-0.088

-0.089



-0.090



-0.090

-0.098

-0.098

-0.044

New York

-0.064

-0.064



-0.065



-0.065

-0.066

-0.066

-0.045

North Carolina

-0.063

-0.064



-0.124



-0.125

-0.158

-0.191

-0.062

North Dakota

-0.083

-0.084



-0.084



-0.098

-0.124

-0.232

-0.040

Ohio

-0.059

-0.063



-0.121



-0.121

-0.123

-0.127

-0.044

Oklahoma

-0.040

-0.040



-0.042



-0.042

-0.051

-0.057

-0.032

Oregon

-0.054

-0.054



-0.054



-0.054

-0.054

-0.054

-0.066

Pennsylvania

-0.028

-0.028



-0.059



-0.059

-0.063

-0.064

-0.043

Rhode Island

-0.072

-0.072



-0.072



-0.072

-0.072

-0.072

-0.054

South Carolina

-0.068

-0.068



-0.092



-0.092

-0.092

-0.092

-0.048

South Dakota

-0.080

-0.081



-0.081



-0.081

-0.081

-0.081

-0.089

Tennessee

-0.039

-0.039



-0.039



-0.039

-0.041

-0.041

-0.050

Texas

-0.028

-0.031



-0.035



-0.036

-0.053

-0.072

-0.033

Utah

-0.029

-0.081



-0.082



-0.082

-0.075

-0.129

-0.044

Vermont

-0.059

-0.059



-0.059



-0.059

-0.059

-0.059

-0.070

Virginia

-0.004

-0.005



-0.010



-0.011

-0.012

-0.019

-0.045

Washington

-0.025

-0.025



-0.025



-0.025

-0.025

-0.039

-0.067

West Virginia

-0.058

-0.061



-0.094



-0.100

-0.101

-0.143

-0.013

Wisconsin

-0.067

-0.069



-0.072



-0.072

-0.073

-0.073

-0.054

Wyoming

0.043

0.043



0.022



0.022

-0.038

-0.112

-0.040

Tribal Data

0.084

0.084



0.084



0.084

-0.007

-0.184

0.001

45


-------
Table D-ll. 2023 Ozone Season Anthropogenic NOx Emissions Reductions (Tons) for non-
EGUs and Fractional Difference in Emissions for the non-EGU Scenario Relative to the
2023 Air Quality Modeling Base Case for Each State.

State

Non-EGU
glass and
cement,
refined
analysis,
others
unchanged,

below
$2,000/ton
(Tons)

Fractional
Difference

EGU
$l,800/ton
+non-EGU
tranche 1
glass &
cement
analyzed

Alabama

-

0.035

Arizona

-

-0.009

Arkansas

-

-0.018

California

-

0.005

Colorado

-

-0.009

Connecticut

-

-0.021

Delaware

-

-0.006

District of
Columbia

-

-0.001

Florida

-

0.028

Georgia

-

0.016

Idaho

-

0.022

Illinois

464

-0.027

Indiana

666

-0.063

Iowa

-

-0.086

Kansas

-

-0.046

Kentucky

-

0.111

Louisiana

-

0.085

Maine

-

0.048

Maryland

-

-0.023

Massachusetts

-

0.004

Michigan

-

-0.029

Minnesota

-

-0.036

Mississippi

-

0.182

Missouri

-

-0.050

Montana

-

0.005

Nebraska

-

-0.028

Nevada

-

0.000

New Hampshire

-

0.049

New Jersey

-

-0.007

New Mexico

-

-0.048

New York

238

-0.025

North Carolina

-

-0.067

North Dakota

-

-0.060

Ohio

-

-0.076

Oklahoma

-

-0.003

Oregon

-

0.012

Pennsylvania

-

-0.023

Rhode Island

-

-0.019

South Carolina

-

-0.041

South Dakota

-

0.010

Tennessee

-

0.019

Texas

-

0.002

Utah

-

-0.039

Vermont

-

0.010

Virginia

138

0.040

Washington

-

0.042

West Virginia

-

-0.082

Wisconsin

-

0.002

Wyoming

-

0.045

Tribal Data

-

0.081

46


-------
47


-------
3. Description of the analytic results.

For each year 2021-2025, EPA used the ozone AQAT to estimate improvements in
downwind air quality at base case levels, at $1,800 per ton emission budget levels, and at higher
dollar per ton emission budget levels. At each cost threshold level, using AQAT, EPA examined
the average and maximum design values for each of the receptors. EPA evaluated the degree of
change in ppb and whether it decreased average or maximum values to below 76 ppb (at which
point their nonattainment and maintenance issues, respectively, would be considered resolved).
EPA also examined each state's air quality contributions at each emission budget level, assessing
whether a state maintained at least one linkage (i.e., greater than or equal to 1% (.75 ppb) to a
receptor that was estimated to remain in nonattainment and/or maintenance. EPA examined the
engineering base case, $l,600/ton, $l,800/ton, $5,800/ton, and $9,600/ton. EPA also created
"straight line" estimates comparable to those used at Steps 1 and 2. The preamble explains at
section VI.D how EPA considered the results of the air quality analyses described in this TSD to
determine the appropriate emission levels for eliminating significant contribution to
nonattainment and interference with maintenance.

For each year, the average and maximum design values (in ppb) estimated using the
assessment tool for each identified receptor for each cost threshold level have been rounded to
hundredths of a ppb and can be found in Tables D-12 through D-21. There are four monitors,
three in Connecticut and one in Texas. Scenarios that are not viable, for technical or policy
reasons, have been grayed out in these tables.

In 2021, we observe that the Stratford monitor 090013007 in Fairfield County,
Connecticut, switches to maintenance at the $l,600/ton level (where SCRs are optimized). In
other words, its average design value drops below 76 ppb (Table D-12), while its maximum
design value stays above 76 ppb (Table D-13). The Madison monitor 090099002 in New Haven
County, Connecticut, has both its average and maximum design values below 76 at all cost
levels, including in the Engineering Base. It was estimated to have a maintenance issue in the
2021 Base Case interpolated from the air quality modeling and used at Steps 1 and 2 (with its
maximum design value higher than 76 ppb).

In 2024, there is only one receptor remaining (the Westport monitor 090019003 in
Fairfield County, Connecticut). This receptor switches from nonattainment to maintenance at
$l,600/ton (Tables D-18 and D-19).

EPA also assessed changes in air quality for the non-EGU scenarios for 2023, 2024, and
2025. In these cases, we included EGU emission reductions at the $l,800/ton cost threshold
level. The results are shown in Table D-22.

In the assessment of air quality using the calibrated assessment tool, we are able to
estimate the change in the air quality contributions of each upwind state to each receptor (see the
description of the state and receptor-specific contributions in section D.2.c.(2)) in order to
determine whether any state's contribution is below the 1 percent threshold used in step 2 of the
4-Step Good Neighbor Framework to identify "linked" upwind states. For this over-control
assessment, we compared each state's adjusted ozone concentration against the 1% air quality
threshold at each of the cost threshold levels up to $9,600/ton at each remaining receptor, using
AQAT. To see static air quality contributions and design value estimates for the four receptors

48


-------
of interest for each of the years for each of the cost levels, see the individual worksheets labeled:
$9600; $5800; $3900; $1600 w CC & non-EGU; $1600 w CC; $1600 wo CC; $500;
straightline_base; and eng_base. For interactive worksheets, refer to the "scenario_202X"
worksheets after setting the desired scenario in the "summaryDVs" worksheet. Then, adjust
cells J27 and J28 to match the desired scenario of interest. The numbering for the various
scenarios is shown in Table D-23. For a cost threshold run, cell J27 would be a value of 1
through 8, while cell J28 should be fixed with a value of 1. For all linked states, in the cost
threshold analysis, we did not see any instances where a state's contributions dropped below 1%
of the NAAQS for all its linkages to remaining downwind receptors. That is, if a state was
linked to a receptor in 2021 in the base case, and that receptor remained either nonattainment or
maintenance in other years or at other cost thresholds, the state remained linked with a
contribution greater than or equal to 1% of the NAAQS. This is not a surprising result because,
for a linkage to be resolved by emission reductions of just a few percent, the original base
contribution would need to be within a few percent of the threshold. As a hypothetical example,
if the state is making a 6% emission reduction in its overall anthropogenic ozone season NOx
emissions, and the calibration factor was 0.5, its original base case maximum contribution to a
remaining unresolved nonattainment and/or maintenance receptor would need to be just under
1.03% of the NAAQS or 0.77 ppb, to drop below the 0.75 ppb linkage threshold.

Lastly, once the EGU budgets for the rule were established (based on the results of the
multi-factor test), it was possible to estimate air quality concentrations in the "control scenario"
at each downwind receptor for each year using the ozone AQAT (Table D-24). Here, we apply a
scenario where all states (regardless of whether they are linked to a particular receptor or to a
different receptor in the geography) have the same cost threshold applied as do the "linked"
states. We observe very little effect of this on air quality at the receptor and in no case are the
changes large enough to shift the status of a receptor from either nonattainment to maintenance
or from maintenance to attainment. This is not surprising because the contributions to each
receptor from these non-linked states are already below the 1% threshold.

49


-------
Table D-12. 2021 Average Ozone DVs (ppb) for NOx Emissions Cost Threshold Levels
($/ton) Assessed Using the Ozone AQAT for All Receptors.	









Assessment Tool Average Ozone Design Values (ppb).

Monitor
Identification
Number

State

County

CAMx
2023 Base
Case (ppb)

Straight
line

Engineering
Baseline

$500/to
n

Optimi

ze

SCR

(

J

Optimi

ze
SNCR+
SCR

Optimi

ze
SNCR+
SCR +
LNH

e

New SCR

+

Optimize
SNCR+
SCR +
LNIJ

90013007

CT

Fairfield

74.3

76.50

76.13

76.11

75.95

75,93

75.94

75,93

75,89

75,83

90019003

CT

Fairfield

76.9

78.56

78.27

78.26

78.13

78,12

78.13

78,12

78,08

78,03

90099002

CT

New Haven

71.7

73.98

73.59

73.57

73.40

73.38

73.39

73.37

73.32

73.25

482010024

TX

Harris

74.0

75.51

75.62

75.58

75.51

75,51

75.50

75,50

75,25

74.95

Note: Scenarios that are not viable, for technical or policy reasons, have been grayed out

Table D-13. 2021 Maximum Ozone DVs (ppb) for NOx Emissions Cost Threshold Levels
($/ton) Assessed Using the Ozone AQAT for All Receptors.	

Monitor
Identification
Number

State

County

CAMx
2023 Base
Case (ppb)

Assessment Tool Maximum Ozone Design Values (ppb).

Straigh
t line

Engineering
Baseline

$500/t
on

Optimi

ze
SCR



Optimi

ze
SNCR
+ SCR







90013007

CT

Fairfield

75.2

77.43

77.05

77.03

76.87



76.86







90019003

CT

Fairfield

77.2

78.86

78.58

78.57

78.44



78.43







90099002

CT

New Haven

73.8

76.15

75.74

75.72

75.54



75.54







482010024

TX

Harris

75.6

77.15

77.25

77.21

77.15



77.13







Table D-14. 2022 Average Ozone DVs (ppb) for NOx Emissions Cost Threshold Levels
($/ton) Assessed Using the Ozone AQAT for All Receptors.	

Monitor
Identification
Number

State

County

CAMx
2023 Base
Case (ppb)

Assessment Tool Average Ozone Design Values (ppb).

Straight
line

Engineering
Baseline

$500/t
on



Optimi

ze SCR
+ LNB



Optimi

ze
SNCR
+ SCR
+ LNB





90013007

CT

Fairfield

74.3

75.40

75.16

75.14



74.97



74.96





90019003

CT

Fairfield

76.9

77.73

77.55

77.54



77.39



77.39





90099002

CT

New Haven

71.7

72.84

72.58

72.56



72.37



72.37





482010024

TX

Harris

74.0

74.76

74.98

74.94



74.88



74.87





Table D-15. 2022 Maximum Ozone DVs (ppb) for NOx Emissions Cost Threshold Levels
($/ton) Assessed Using the Ozone AQAT for All Receptors.	

Monitor
Identification
Number

State

County

CAMx
2023 Base
Case (ppb)

Assessment Tool Maximum Ozone Design Values (ppb).

Straight
line

Engineering
Baseline

$500/to
n



Optimi

ze SCR
+ LNB



Optimi

ze
SNCR
+ SCR
+ LNB





90013007

CT

Fairfield

75.2

76.31

76.07

76.05



75.87



75.87





90019003

CT

Fairfield

77.2

78.03

77.85

77.84



77.70



77.69





90099002

CT

New Haven

73.8

74.98

74.71

74.69



74.49



74.49





482010024

TX

Harris

75.6

76.37

76.60

76.56



76.50



76.48





50


-------
Table D-16. 2023 Average Ozone DVs (ppb) for NOx Emissions Cost Threshold Levels
($/ton) Assessed Using the Ozone AQAT for All Receptors.	









Assessment Tool Average Ozone Design Values (ppb).

Monitor
Identification
Number

State

County

CAMx
2023 Base
Case (ppb)

Straight
line

Engineering
Baseline

$500/to
n

Optimi

ze

SCR

Optimi

ze SCR
+ LNB

Optimi

ze
SNCR+
SCR

Optimi

ze
SNCR+
SCR +
LNB

New
SNCR +
Optimiz
e

SNCR+
SCR +
LNB

New SCR

+

Optimize

SNCR+
SCR +
LNB

90013007

CT

Fairfield

74.3

74.30

74.13

74.11

73.88

73.86

73.87

73.86

73.80

73.72

90019003

CT

Fairfield

76.9

76.90

76.74

76.72



76.46



76.46





90099002

CT

New Haven

71.7

71.70

71.51

71.48



71.21



71.21





482010024

TX

Harris

74.0

74.00

74.55

74.49



74.38



74.36





Table D-17. 2023 Maximum Ozone DVs (ppb) for NOx Emissions Cost Threshold Levels
($/ton) Assessed Using the Ozone AQAT for All Receptors.	









Assessment Tool Maximum Ozone Design Values (ppb).

Monitor
Identification
Number





CAMx
2023 Base
Case (ppb)



Engineering
Baseline

$500/t
on

Optimi

ze

Optimi

ze SCR

Optimi

ze

Optimi

ze

New
SNCR +

New SCR

+

State

County

Straigh
t line





SCR

+ LNB

SNCR
+ SCR

SNCR
+ SCR
+ LNB

Optimiz

e

SNCR+
SCR +
LNB

Optimize

SNCR+
SCR +
LNB

90013007

CT

Fairfield

75.2

75.20

75.03

75.01

74.78

74.76

74.77

74.75

74.69

74.61

90019003

CT

Fairfield

77.2

77.20

77.04

77.02



76.76



76.76





90099002

CT

New Haven

73.8

73.80

73.60

73.58



73.30



73.29





482010024

TX

Harris

75.6

75.60

76.17

76.10



75.99



75.97





Table D-18. 2024 Average Ozone DVs (ppb) for NOx Emissions Cost Threshold Levels
($/ton) Assessed Using the Ozone AQAT for All Receptors.	

Monitor
Identification
Number

State

County

CAMx
2023 Base
Case (ppb)

Assessment Tool Average Ozone Design Values (ppb).

Straight
line

Engineering
Baseline

$500/t
on

Optimi

ze
SCR

Optimi

ze SCR
+ LNB

Optimi

ze
SNCR
+ SCR

Optimi

ze
SNCR
+ SCR
+ LNB

New
SNCR +
Optimiz
e

SNCR+
SCR +
LNB

New SCR

+

Optimize

SNCR+
SCR +
LNB

90013007

CT

Fairfield

74.3

73.76

73.58

73.56

73.33

73.31

73.32

73.30

73.25

73.17

90019003

CT

Fairfield

76.9

76.38

76.20

76.18



75.92



75.92





90099002

CT

New Haven

71.7

71.14

70.93

70.90



70.63



70.63





482010024

TX

Harris

74.0

73.58

74.05

73.98



73.88



73.86





Table D-19. 2024 Maximum Ozone DVs (ppb) for NOx Emissions Cost Threshold Levels
($/ton) Assessed Using the Ozone AQAT for All Receptors.	









Assessment Tool Maximum Ozone Design Values (ppb).











Engineering

$500/t

Optim

Optimi

Optimi

Optimi

New

New SCR

Monitor
Identification
Number





CAMx
2023 Base
Case (ppb)



Baseline

on

lze

ze SCR

ze

ze

SNCR +

+

State

County

Straigh
t line





SCR

+ LNB

SNCR
+ SCR

SNCR
+ SCR
+ LNB

Optimiz

e

SNCR+
SCR +
LNB

Optimize

SNCR+
SCR +
LNB

90013007

CT

Fairfield

75.2

74.65

74.47

74.45

74.22

74.20

74.21

74.19

74.14

74.06

90019003

CT

Fairfield

77.2

76.68

76.50

76.48



76.22



76.22





90099002

CT

New Haven

73.8

73.22

73.01

72.98



72.70



72.70





482010024

TX

Harris

75.6

75.17

75.65

75.58



75.48



75.46





51


-------
Table D-20. 2025 Average Ozone DVs (ppb) for NOx Emissions Cost Threshold Levels
($/ton) Assessed Using the Ozone AQAT for All Receptors.	









Assessment Tool Average Ozone Design Values (ppb).

Monitor
Identification
Number

State

County

CAMx
2023 Base
Case (ppb)

Straight
line

Engineering
Baseline

$500/t
on

' 	 		

Optimi

ze SCR
+ LNB

Optimi

Optimi

ze
SNCR
+ SCR
+ LNB

New
SNCR +
Optimiz
e

SNCR+
SCR +
LNB

New SCR
+ Optimize

SNCR+
SCR +
LNB

90013007

CT

Fairfield

74.3

73.22

73.04

73.02

72.79

72.77

72.79

72.77

72.72

72.64

90019003

CT

Fairfield

76.9

75.86

75.68

75.66

75,42

75.41

75,42

75.40

75.34

75.26

90099002

CT

New Flaven

71.7

70.58

70.37

70.35

70,10

70.08

70,09

70.07

70.00

69.91

482010024

TX

Flarris

74.0

73.16

73.62

73.55

73.45

73.45

73.43

73.43

73.04

72.59

Table D-21. 2025 Maximum Ozone DVs (ppb) for NOx Emissions Cost Threshold Levels
($/ton) Assessed Using the Ozone AQAT for All Receptors.	









Assessment Tool Maximum Ozone Design Values (ppb).

Monitor
Identification
Number

State

County

CAMx
2023 Base
Case (ppb)

Straigh
t line

Engineerin
g Baseline

$500/t
on

' 	 		

Optimi

ze SCR
+ LNB

Optimi
ze

Optimi

ze
SNCR
+ SCR
+ LNB

New
SNCR +
Optimiz
e

SNCR+
SCR +
LNB

New SCR

+

Optimize

SNCR+
SCR +
LNB

90013007

CT

Fairfield

75.2

74.11

73.93

73.91

73.67

73.66

73.67

73.65

73.60

73.52

90019003

CT

Fairfield

77.2

76.16

75.98

75.96

75,72

75.70

75,71

75.70

75.64

75.56

90099002

CT

New Flaven

73.8

72.65

72.43

72.41

72.15

72.13

72.14

72.12

72.05

71.96

482010024

TX

Flarris

75.6

74.74

75.21

75.14

75,04

75.04

75,02

75.02

74.62

74.16

Table D-22. Average and Maximum Ozone DVs (ppb) for non-EGU NOx Emissions Level*

for Each Year Assessed.

Monitor
Identification
Number

State

County

CAMx
2023 Base
Case Avg
DV (ppb)

Average Design Value (ppb)

Maximum Design Value (ppb)

2023

2024

2025

2023

2024

2025

90013007

CT

Fairfield

74.3

73.83

73.28

72.74

74.72

74.16

73.62

90019003

CT

Fairfield

76.9

76.43

75.89

75.37

76.73

76.19

75.67

90099002

CT

New Flaven

71.7

71.17

70.60

70.04

73.26

72.66

72.09

482010024

TX

Flarris

74.0

74.36

73.86

73.43

75.97

75.46

75.02

*non-EGU AQAT air quality estimates include EGU emission reductions at the $l,600/ton level.

Table D-23. Description of the Various Scenarios Modeled in AQAT.

Scenario

Cost
Threshold
Level

Short
Description

Description

1

$0

Eng base

Baseline 202x OS NOx + engineering nonCEMs

2

$500

$500

Baseline 202x OS NOx + engineering nonCEMs + $500/ton Generation Shifting

3

$1,600

$1,600 w/o
LNB

Baseline 202x OS NOx + engineering nonCEMs + 0.08 SCR Cap + $l,600/ton Generation Shifting

4

$1,600

$1,600 w
LNB

Baseline 202x OS NOx + engineering nonCEMs + 0.08 SCR Cap + LNB + $l,600/ton Generation
Shifting

52


-------
5

$1,800

$1,800 w/o
LNB

Baseline 202x OS NOx + engineering nonCEMs + 0.08 SCR Cap + SNCR Optimize + $l,600/ton
Generation Shifting

6

$1,800

$1,800 w
LNB

Baseline 202x OS NOx + engineering nonCEMs + 0.08 SCR Cap + LNB + SNCR Optimize +
$l,600/ton Generation Shifting

7

$5,800

$5,800

Baseline 202x OS NOx + engineering nonCEMs + 0.08 SCR Cap + LNB + SNCR Optimize +
SNCR Retrofit + $5,800/ton Generation Shifting

8

$9,600

$9,600

Baseline 202x OS NOx + engineering nonCEMs + 0.08 SCR Cap + LNB + SNCR Optimize +
SCR Retrofit + $9,600/ton Generation Shifting

9

NA

Straightline
base

202X Straight line emissions interpolation (an appoximation of that used for Steps 1 and 2).

10

NA



Analyzed... number of tons of non-EGU glass and cement, refined analysis.

11

$1,800 up
to $2,000

$1,800 w
LNB &
non-EGU

Baseline 202x OS NOx + engineering nonCEMs + 0.08 SCR Cap + LNB + SNCR Optimize +
$l,600/ton Generation Shifting + non-EGU glass and cement, refined analysis, others unchanged,
below $2,000 per ton.

Table D-24. Average and Maximum Ozone DVs (ppb) for the $1,800 Per Ton "Control
Scenario" for each Year Assessed.

Monitor
Identification
Number

State

County

CAMx
2023
Base
Case
(PPb)

Average Design Value (ppb)

Maximum Design Value (ppb)

2021

2022

2023

2024

2025

2021

2022

2023

2024

2025

90013007

CT

Fairfield

74.3

75.94

74.96

73.85

73.30

72.77

76.86

75.87

74.75

74.19

73.65

90019003

CT

Fairfield

76.9

78.13

77.39

76.45

75.92

75.40

78.43

77.69

76.75

76.21

75.69

90099002

CT

New Haven

71.7

73.39

72.37

71.20

70.62

70.07

75.54

74.48

73.29

72.69

72.12

482010024

TX

Harris

74.0

75.50

74.86

74.36

73.86

73.43

77.13

76.48

75.97

75.45

75.02

4. Comparison between the air quality assessment tool estimates

As described earlier, AQAT was calibrated using CAMx data from either 2016 or 2028.
Thus, it was possible to evaluate the estimates from the tool for a comparable scenario. The
average design values from AQAT for 2024 for the various scenarios are shown using the 2016-
based calibration factor (Table D-25), as well as the differences between the values in Tables D-
18 and D-25. The differences are shown in Table D-26. The AQAT values and the differences
in the table have been rounded to a hundredth of a ppb. For this set of scenarios, the differences
are moderate, with a maximum value of 0.51 ppb.

There can be a small offset between the estimates (based on the impacts of the two
different calibration factors). Within a set of estimates, the differences are likely to be
comparable. That is, comparing two different scenarios in Table D-18 with the same two
scenarios in Table D-25, produces similar changes in air quality. For example, the difference
between the engineering base and the $1,800 per ton level where SCR and SNCR are optimized
and combustion controls are installed results in a difference of 0.29 ppb when the 2028
calibration factor is applied and 0.19 ppb when the 2016 calibration factor is applied. The results
of this demonstrate that, considering the time and resource constraints faced by the EPA, the
AQAT provides reasonable estimates of air quality concentrations for each receptor, can provide
reasonable inputs for the multi-factor assessment, and can serve as a method to test for linkages
dropping below the threshold.

53


-------
Table D-25. 2024 Average Ozone DVs (ppb) for NOx Emissions Cost Threshold Levels
($/ton) Assessed Using the Ozone AQAT for All Receptors Using the Calibration Factor
from the 2016 Modeling.

Monitor
Identification
Number

State

County

CAMx
2023 Base
Case (ppb)

Assessment Tool Average Ozone Design Values (ppb).

Straight
line

Engineering
Baseline

$500/t
on

Optimi

ze
SCR

Optimi

ze SCR
+ LNB

Optimi

ze
SNCR
+ SCR

Optimi

ze
SNCR
+ SCR
+ LNB

New
SNCR +
Optimiz
e

SNCR+
SCR +
LNB

New SCR

+

Optimize

SNCR+
SCR +
LNB

90013007

CT

Fairfield

74.3

73.76

73.79

73.78

73.61

73.60

73.61

73.60

73.56

73.50

90019003

CT

Fairfield

76.9

76.38

76.52

76.50

76.38

76.36

76.37

76.36

76.33

76.28

90099002

CT

New Haven

71.7

71.14

71.15

71.14

70.96

70.95

70.96

70.94

70.89

70.82

482010024

TX

Harris

74.0

73.58

74.03

73.99

73.93

73.93

73.91

73.91

73.68

73.40

Table D-26. 2024 Difference in Average Ozone DVs (ppb) for NOx Emissions Cost
Threshold Levels ($/ton) Assessed Using the Ozone AQAT for All Receptors for the
Estimates with the Two Different Calibration Factors.

Monitor
Identification
Number

State

County

CAMx
2023 Base
Case (ppb)

Assessment Tool Average Ozone Design Values (ppb).

Straight
line

Engineering
Baseline

$500/t
on

Optimi

ze

SCR

Optimi

ze SCR
+ LNB

Optimi

ze
SNCR
+ SCR

Optimi

ze
SNCR
+ SCR
+ LNB

New
SNCR +
Optimiz
e

SNCR+
SCR +
LNB

New SCR

+

Optimize

SNCR+
SCR +
LNB

90013007

CT

Fairfield

74.3

0.16

0.21

0.22

0.28

0.29

0.29

0.29

0.31

0.33

90019003

CT

Fairfield

76.9

0.23

0.31

0.32

0.43

0.44

0.43

0.44

0.47

0.51

90099002

CT

New Haven

71.7

0.16

0.23

0.23

0.30

0.31

0.31

0.31

0.33

0.36

482010024

TX

Harris

74.0

0.16

-0.02

0.01

0.05

0.05

0.06

0.06

0.21

0.38

E. Observations on Cost and Air Quality Factors in 2024

Section VI of the preamble discusses the cost and air quality factors in the multifactor test
and reaches the conclusions about the requisite level of emissions control for each year. The
higher mitigation technology scenarios associated with post-combustion control installation were
not considered in the 2021 multi-factor test as they pertain to technologies not possible to install
at a regional scale until 2025. However, for illustrative purposes, EPA examined their reduction
potential and air quality impact of these controls starting in 2024. As described in sections C and
D of this TSD, EPA quantified emissions from upwind states at various levels of
uniform NOx control stringency, each represented by uniform control technology and
corresponding NOx reduction and then evaluated the potential air quality consequences of these
potential reductions.

EPA combines costs, EGU NOx reductions, corresponding improvements in
downwind ozone concentrations, and other considerations for different control levels in its multi-
factor test. EPA examines whether any receptor shifts from nonattainment to maintenance or
from maintenance to attainment. In 2024, the last receptor (Westport) in Fairfield Connecticut
shifts from nonattainment to maintenance at $1,600 per ton. This receptor is minimally above the

54


-------
75.9 threshold in 2024 and is fully resolved by 2025. No additional changes are observed at
higher cost threshold levels in 2024 or 2025. EPA analysis of these more stringent scenarios in
2024 also results in a "knee-in-the-curve" graph (see preamble section VI for details about this
figure for 2021). Figure E-l below illustrates the air quality improvement for the mitigation
technologies up to $9,600 per ton for EGUs for 2024. In Figure E-l, the 2024 "knee" is also at a
point where emission budgets reflect a control stringency with an estimated marginal cost of
$1,600 and $1,800 per ton. The more stringent emission budget levels (e.g., emission budgets
reflecting mitigation technologies that cost $5,800 per ton or greater) yield fewer additional
emission reductions and fewer air quality improvements relative to the increase in control costs.
These control measures also involve significant capital investment and the installation of new
hardware at the EGU. For the reasons described in section VI of the preamble, the $1,800 per ton
cost threshold is a reasonable stopping level for 2021 and 2022. Although EPA evaluated the
potential reductions from post combustion controls, that technology did not qualify as an option
for future years as EPA explains those controls are not possible on a regional scale until 2025,
and EPA expects no remaining air quality problems for the 2008 ozone NAAQS standard in that
year at the $1,800 per ton level.

Figure E-l. EGU Ozone Season NOX Reduction Potential in 11 Linked States and
Corresponding Total Reductions in Downwind Ozone Concentration at the Westport
Fairfield Connecticut Receptor for each Cost Threshold Level Evaluated in 2024.





0.45







c





0.35

o





Cl



0.3

£





>-

-Q

0.25



CL



"ro
13

Q.

0.2

a





t_



0.15

<



O)



0.1

CtD



ro





S
>



0.05

<









0

c

o

$2,000

$4,000 $6,000
Cost per Ton

$8,000

30,000
25,000
20,000
15,000
10,000
5,000
0

$10,000

c

o

3

0)
OS
X

o

c

o

to
ra


00



o

55


-------
F. Assessment of the Effects of Ozone on Forest Health39

Air pollution can impact the environment and affect ecological systems, leading to
changes in the ecological community and influencing the diversity, health, and vigor of
individual plant species. When ozone is present in the environment, it enters the plant through
the stomata and can interfere with carbon gain (photosynthesis) and allocation of carbon within
the plant, making fewer carbohydrates available for plant growth, reproduction, and/or yield
(2020 PA, section 4.3.1 and 2013 ISA, p. 1-15).40,41 Ozone can impact a variety of commercial
and ecologically important species throughout the United States. These include forest tree and
herbaceous species as well as crops. Such effects at the plant scale can also be linked to an array
of effects at larger spatial scales and higher levels of biological organization, causing impacts to
ecosystem productivity, water cycling, ecosystem community composition and alteration of
below-ground biogeochemical cycles (2020 PA, section 4.3.1 and 2013 ISA, p. 1-15)..42 With
the data sets available to the Agency, here, we focus on selected forest tree species.

Assessing the impact of ozone on forests in the United States involves understanding the
risk to tree species from ozone concentrations in ambient air and accounting for the prevalence
of those species within the forest. Across several reviews of the ozone NAAQS and based on
longstanding body of scientific evidence, EPA has evaluated concentration-response functions
which relate ozone exposure to growth-related effects in order to consider the risk of ozone-
related growth impacts on forest trees (2020 PA, section 4.3.3, 2013 ISA and 2020 ISA). For
this purpose, EPA has focused on cumulative, concentration-weighted indices of exposure, such
as the W126-based cumulative exposure index (2020 PA, section 4.3.3.1.1, 2020 ISA, section
ES.3). Measured ozone concentrations in ambient air of the United States are used to calculate
the W126-based index as the annual maximum 3-month sum of daytime hourly weighted ozone
concentrations, averaged over 3 consecutive years. The sensitivity of different trees species
varies about the growth impacts of ozone exposure. Based on well-studied datasets relating
W126 index to reduced growth, exposure response functions have been developed for 11 tree
species (2020 PA, section 4.3.3.1.2 and Figure 4-3 and 2013 ISA, section 9.6). For these
species, the impact from ozone exposure has been determined by exposing seedlings to different
levels of ozone concentrations over one or more seasons (which have been summarized in terms
of W126 index) and measuring reductions in growth (which are then summarized as "relative
biomass loss"). The magnitude of ozone impact on a forest community will depend on the

39	Analysis of the environmental effects of ozone is not within scope of the Revised CSAPR Update rule. See Legal
section of the Response to Comments document for further information.

40	U.S. EPA (2020). Policy Assessment for the Review of the Ozone National Ambient Air Quality Standards. U.S.
Environmental Protection Agency. Office of Air Quality Planning and Standards, Health and Environmental
Impacts Division, Research Triangle Park, NC. EPA-452/R-20-001.

Available https://www.epa.gov/sites/production/files/2020-05/documents/o3-final_pa-05-29-20compressed.pdf

41	U.S. EPA (2020). Integrated Science Assessment for Ozone and Related Photochemical Oxidants. U.S.
Environmental Protection Agency. Washington, DC. Office of Research 3A-35 and Development. EPA/600/R-
20/012. Available at: https://www.epa.gov/isa/integrated-science-assessment-isa-ozone-and-related-photochemical-
oxidants.

42	U.S. EPA (2013). Integrated Science Assessment of Ozone and Related Photochemical Oxidants (Final Report).
Office of Research and Development, National Center for Environmental Assessment. Research Triangle Park, NC.
U.S. EPA. EPA-600/R-10-076F. February 2013. Available

at: https://nepis. epa.gov/Exe/ZyP URL. cgi ?Dockey=Pl 00KETF. txt.

56


-------
prevalence of different tree species of relatively more versus less sensitivity to ozone and the
abundance in the community.

The most common tree species in the eastern United States, where the benefits from this
rule will be most pronounced, are black cherry (Prunus serotina), yellow or tulip-poplar
(Liriodendron tulipifera), sugar maple (Acer saccharum), eastern white pine (Pinus strobus),
Virginia Pine {Pinus virginiana), red maple (Acer rubrum), and quaking aspen (Populus
trenuloides). Since 2008, EPA has assessed the impact of ozone on these tree species within the
eastern United States for the period from 2000 to 2018 as part of the Clean Air Market Division
(CAMD) annual power sector programs progress report.43 Over this time period ozone
concentrations have improved substantially because of various emission reduction programs,
such as NBP, CAIR, CSAPR, CSAPR Update, and other local and mobile source reductions such
as Tier2 and Tier3 rules. Past EPA assessments have shown that the improvements in ozone are
evident both for the regulatory metric, 3-year average of 4th highest 8-hr daily maximum ozone
concentration, and for the W126 metric. 44 In forests where certain sensitive species dominate
the forest community, the estimates of relative biomass loss from ozone have decreased
substantially. However, for the period from 2017-2019, the eastern United States still has areas
with up to 11.5% estimated relative biomass loss for the seven tree species - black cherry,
yellow poplar, sugar maple, eastern white pine, Virginia pine, red maple, and quaking aspen
(Figure F-l)45.

Ozone levels are expected to continue to decrease through 2024 based on model
projection of the impacts on ozone concentrations resulting from baseline "on the books" control
programs as well as by emission reductions under this rule. As ozone declines, estimates of
relative biomass loss of these trees' species will also decline as they have from 2000 to 2019,
indicating increased protection of forest ecosystems and resources. Under this rule, ozone
concentrations are expected to decline faster than without the rule (e.g., under the base case).
While EPA does not have the tools to quantify the expected level of improvement at this time,
based on the previous relationships between ozone design values and W126 determined as part
of the review of the 2020 ozone NAAQS (2020 PA, section 4D.3.2.3 and Table 4D-12), W126
values are expected to improve as design values decrease. As described in the preamble, the rule
is expected to decrease design values by 0.17 ppb, on average, in 2021. The reductions from this

43	See the annual progress report at https://www3.epa.gov/airmarkets/progress/reports/index.html

44	U.S. EPA (2020). Policy Assessment for the Review of the Ozone National Ambient Air Quality Standards. U.S.
Environmental Protection Agency. Office of Air Quality Planning and Standards, Health and Environmental
Impacts Division, Research Triangle Park, NC. EPA-452/R-20-001.

Available https://www.epa.gov/sites/production/files/2020-05/documents/o3-final_pa-05-29-20compressed.pdf

45	To estimate the biomass loss for forest ecosystems across the eastern United States, the
biomass loss for each of the seven tree species was calculated using the three-month, 12-hour
W126 exposure metric at each location, along with each tree's individual C-R functions. The
W126 exposure metric was calculated using monitored ozone data from CASTNET and AQS
sites, and a three-year average was used to minimize the effect of variations in meteorological
and soil moisture conditions. The biomass loss estimate for each species was then multiplied by
its prevalence in the forest community using the U.S. Department of Agriculture (USDA) Forest
Service IV index of tree abundance calculated from Forest Inventory and Analysis (FIA)
measurements.

57


-------
rule are likely to provide further protection to natural forest ecosystems by reducing the potential
for ozone-related impacts.



i ¦

jl/:

lI. jM



I



Biomass (% Loss)

I I > 1%

1 to 3%

3 to 6%

6 to 9%

>9% Max = 11.6%

Figure F-l: Estimated Black Cherry, Yellow Poplar, Sugar Maple, Eastern White Pine, Virginia Pine, Red
Maple, and Quaking Aspen Biomass Loss due to Ozone Exposure for 2016-2018. See the annual progress
report at https://www3.epa.gov/airmarkets/progress/reports/index.html

58


-------
Appendix A: State Emission Budget Calculations and Engineering
Analytics

See Excel workbook titled "Final Rule State Emission Budget Calculations and Engineering
Analytics" on EPA's website and in the docket for this rulemaking

59


-------
Appendix B: Description of Excel Spreadsheet Data Files Used in

the AQAT

60


-------
EPA placed the Ozone AQAT Final.xlsx Excel workbook file in the Revised CSAPR
Update docket that contains all the emission and CAMx air quality modeling inputs and resulting
air quality estimates from the AQAT. The following bullets describe the contents of various
worksheets within the AQAT workbook:

State-level emissions

•	"2021 EGU" through "2026 EGU" contain EGU emissions measurements and estimates
for each state. Various columns contain the 2016 and 2019 OS measured emissions,
CSAPR Update Budgets, and then emissions for the engineering base along with each of
the cost thresholds.

•	"2016fhl" contains state and source-sector specific ozone-season NOx emission totals
for the 2016 base case modeled in CAMx.

•	"2023fhl" contains state and source-sector specific ozone-season NOx emission totals for
the 2023 base case with EGU estimates from IPM modeled in CAMx.

•	"2023fhl_eng" contains state and source-sector specific ozone-season NOx emission
totals for the 2023 base case with EGU estimates from engineering analysis that could be
used in CAMx.

•	"2028fhl" contains state and source-sector specific ozone-season NOx emission totals for
the 2028 base case with EGU estimates from IPM modeled in CAMx.

•	"AQMEGUemiss" has a breakdown of the point EGU emission inventory used in the
air quality modeling, for the units with CEMs and those that don't (nonCEMs).

•	"calib emiss" has the total anthropogenic emissions by state for each of the base cases
modeled in CAMx. This worksheet also contains the fraction change for each of these
scenarios relative to the 2023fhl base case modeled in CAMx.

•	"2021_emiss_total", "2022_emiss_total", "2023_emiss_total", "2024_emiss_total",
"2025_emiss_total", "2026_emiss_total" each of these worksheets reconstructs total
anthropogenic emissions for the year, with various EGU emission inventories for
different cost threshold (including the engineering base case). The total anthropogenic
emissions can be found for each state in columns BW through CE. These totals are then
compared to the 2023fhl emission level (column CE) to make a fractional change in
emissions in columns CF through CN. Non-EGU emissions change and fractional
change (inclusive of EGU changes at $l,800/ton) are found in columns CO and CP,
respectively

Air quality modeling design values and contributions from CAMx

•	"2023_contribs" contains average and maximum design values as well as state by state
contributions for the 2023fhl base case modeled in CAMx.

•	"2028fhl DVs" contains average and maximum design values for each receptor in 2028
with EGU estimates from IPM.

Calibration factor creation and assessment

•	"calib_2016" includes the calculation of the calibration factor based on the 2023
contributions, and percent change of 2016 emissions relative to 2023 emissions. The
calibration factor can be found in column BM.

61


-------
•	"calib_2028" includes the calculation of the calibration factor based on the 2023
contributions, and percent change of 2028 emissions relative to 2023 emissions. The
calibration factor can be found in column BM.

•	"calib comp" includes a summary of the three calibration factors (one based on 2016and
one based on 2028).

Air quality estimates

•	"Summary DVs" contains the average and maximum design value estimates (rounded to
two decimal places) for receptors that were nonattainment or maintenance in the 2021
base case interpolation modeling. Values for each year (2021 through 2015), for each
cost threshold are shown. Grey or black filled cells are not considered viable scenarios
for technical or policy reasons. Each scenario has the cost threshold shown for that run
the linked and unlinked states. Adjustment to cells J27 and J28will result in interactive
adjustment for the other worksheets and will adjust the average design values in column J
and maximum design values in column X.

•	"scenario_2021" through "scenario_2025" contains the average and maximum design
value estimates (as well as the individual state's air quality contributions) for a particular
scenario identified in cells G2 and G3. The fractional emission changes for each of the
linked and unlinked states are shown in rows 2 and 3.

•	"scenario_2021_sens" and "scenario_2022_sens" contains the average and maximum
design value estimates (as well as the individual state's air quality contributions) for a
particular EGU scenario identified in cells G2 and G3 where the remainder of the
emission inventory was created by interpolating between a projected 2021 inventory and
the 2023 inventory. The fractional emission changes for each of the linked and unlinked
states are shown in rows 2 and 3. The fractional emission changes for these scenarios are
calculated in "2021_emiss_total_sens" and "2021_emiss_total_sens".

•	"straightline_2021" through "straightline_2026" contains the average and maximum
design value estimates (as well as the individual state's air quality contributions) for the
emissions scenario that is a linear interpolation of the emissions between the 2016 base
case and the 2023fhl base case (or between the 2023fhl base case and 2028fhl base
case).

•	"control_2021" through "contol_2025" contains the average and maximum design value
estimates (as well as the individual state's air quality contributions) for a particular
scenario identified in cells G2 and G3. States that are "linked" to any receptor in the
geography are assigned the values in row 2 while nonlinked states are assigned the values
in row 3. Note that, as the "home" States, Texas and Connecticut are both assigned the
"linked" State level of reductions.

•	"scenario_2024 alt calibration" contains the average and maximum design value
estimates (as well as the individual state's air quality contributions) for a particular
scenario identified in cells G2 and G3. The fractional emission changes for each of the
linked and unlinked states are shown in rows 2 and 3. This uses the calibration factor
based on the 2016 air quality modeling, rather than the calibration factor based on the
2028 air quality modeling.

•	The individual cost level worksheets labeled: "New_SCR"; "New_SNCR";

"SCR SNCR LNB"; "SCR_SNCR"; "SCRLNB"; "SCR"; "$500"; "straightline_base";

62


-------
and "eng base" contain static air quality contributions and design value estimates for the
four receptors of interest for each of the years.

63


-------
Appendix C: IPM Runs Used in Transport Rule Significant
Contribution Analysis

64


-------
Table C-l lists IPM runs used in analysis for this rule. The IPM runs can be found in the
docket for this rulemaking under the IPM file name listed in square brackets in the table below.

Table Appendix C-l. IPM Runs Used in Transport Rule Significant Contribution Analysis

Run Name
[IPM File Namel

Description

Air Quality Modeling Base Case
EPA617BC75L

Model run used for the air quality modeling base case at steps 1
and 2, which includes the national Title IV S02 cap-and-trade
program; NOx SIP Call; the Cross-State Air Pollution trading
programs, and settlements and state rules. It also includes key
fleet updates regarding new units, retired units, and control
retrofits that were known by Fall of 2019.

Illustrative Base Case

EPA617_CURR_ 1 g

Model run used as the base case for the Illustrative Analysis of
cost threshold analyses. Based on the air quality modeling base
case, but with projected retirements and retrofits in 2021 limited.

Illustrative Base Case with optimization
technology + LNB upgrade

EPA617_CURR_5d

Imposes state-level generation constraints starting in 2021 for
fossil-fuel fired units greater than 25 MW that is equal to
Illustrative Base Case levels. Also assumes optimization of
existing post-combustion controls and upgrade of combustion
controls if mode 3
-------
Illustrative More Stringent Policy Case
EPA617CURR13

Same as Illustrative Base Case, but with ozone season emissions
budgets with variability limits applied for the 12 states reflecting
$1600 per OS NOx ton through 2023 and $9600 per OS NOx ton
for 2025 model run year, along with a regional cap equal to the
sum of the 12 states' budgets for each year.

Illustrative LessStringent Policy Case
EPA617CURR12

Same as Illustrative Base Case, but with ozone season emissions
budgets with variability limits applied for the 12 states reflecting
$500 per OS NOx ton starting in 2021, along with a regional cap
equal to the sum of the 12 states' budgets for each year.

Proposed Policy Scenario

EPA617CURR14

Same as Illustrative Base Case, but with ozone season emissions
budgets with variability limits applied for the 12 states reflecting
$1600 per OS NOx ton starting in 2021, along with a regional
cap equal to the sum of the 12 states' budgets for each year.

Final Illustrative More Stringent Policy
Case

EPA617_CURR_20e

Same as Illustrative Base Case, but with ozone season emissions
budgets with variability limits applied for the 12 states reflecting
optimization of existing controls and combustion control upgrade
through 2023, and SCR retrofit in model run year 2025 and
beyond, along with a regional cap equal to the sum of the 12
states' budgets for each year.

Final Illustrative LessStringent Policy
Case

EPA617CURR2 lb

Same as Illustrative Base Case, but with ozone season emissions
budgets with variability limits applied for the 12 states reflecting
$500 per OS NOx ton starting in 2021, along with a regional cap
equal to the sum of the 12 states' budgets for each year.

Final Policy Scenario

EPA617_CURR_ 19d

Same as Illustrative Base Case, but with ozone season emissions
budgets with variability limits applied for the 12 states reflecting
$1600 and $1800 per OS NOx ton starting in 2021and 2022
respectively, along with a regional cap equal to the sum of the 12
states' budgets for each year.

Final Policy Sensitivity
EPA617 CURR_22

Same as Final Policy Scenario used for RIA, but with state
emission budgets reflecting $1800 per ton stating in 2021

Base Case with Additional Nuclear
Retirements - Sensitivity
EPA617 CURR_18b

Same as Illustrative Base Case, but with potential Dresden/Byron
nuclear retirements included.

Final Policy Scenario Sensitivity - with
additional nuclear retirements
EPA617 CURR_23

Same as Final Policy Scenario assumptions, but applied to the
base case with additional nuclear retirements to reflect potential
Dresden/Byron nuclear retirements.

66


-------
Appendix D: Ozone-Season NOx Emissions Limits for IPM Modeling

Table Appendix D-l. State and Regional Caps for IPM Final Case Analysis for the Final Rule



Policy Case



Less Stringent Case ($500/ton)



More Stringent Case ($9,600/ton)



Assurance Level (121% of
Budget)



Assurance Level (121% of
Budget)



Assurance Level (121% of Budget)



2021*46

2023

2025



2021

2023

2025



2021

2023

2025

Illinois

11,129

9,896

9,751



11,311

10,155

10,009



11,129

9,896

8,338

Indiana

15,833

15,189

11,573



18,969

18,366

14,621



15,833

15,189

10,201

Kentucky

18,521

17,001

17,001



18,883

18,883

18,883



18,521

17,001

11,827

Louisiana

18,621

17,929

17,929



18,670

18,670

18,670



18,621

17,929

15,272

Maryland

1,814

1,532

1,631



1,816

1,533

1,633



1,814

1,532

1,414

Michigan

15,406

12,070

11,841



15,882

12,567

12,327



15,406

12,070

8,887

New Jersey

1,517

1,517

1,517



1,629

1,629

1,629



1,517

1,517

1,521

New York

4,133

4,139

4,117



4,190

4,196

4,174



4,133

4,139

3,989

Ohio

11,725

11,825

11,825



18,740

18,839

18,839



11,725

11,825

11,158

Pennsylvania

10,139

10,131

10,131



14,287

14,285

14,285



10,139

10,131

9,499

Virginia

5,583

4,816

4,432



5,640

5,272

4,866



5,583

4,816

3,853

West Virginia

16,560

15,589

15,589



18,170

18,170

18,170



16,560

15,589

12,787

























Region Cap (Budget
Total)

108,248

100,525

96,974



122,468

117,822

114,138



108,248

100,525

81,609

46 As explained in the RIA, the modeled budgets represented here in the 2021 policy and more stringent scenario are approximately 1% higher than the final rule
budgets due to the rulemaking schedule limitations for beginning the RIA analysis.

67


-------
Appendix E: Generation Shifting Analysis

Table Appendix E-l. Tons of EGU NOx Reduction Potential from Shifting Generation Compared to Adjusted Historical
Baseline Emissions.



2021

2022

2023

2024



Baseline
(tons)

Reductions
from

generation
shiftingat
$1600 per
ton (tons)

Reductions
from

generation
shifting (%)

Baseline
(tons)

Reductions
from

generation
shiftingat
$1600 per
ton (tons)

Reductions
from

generation
shifting (%)

Baseline
(tons)

Reductions
from

generation
shiftingat
$1600 per
ton (tons)

Reductions
from

generation
shifting (%)

Baseline
(tons)

Reductions
from

generation
shifting at
$1600 per
ton (tons)

Reductions
from

generation
shifting (%)

Illinois

9,368

52

1%

9,368

52

1%

8,413

53

1%

8,292

52

1%

Indiana

15,856

317

2%

15,383

313

2%

15,357

316

2%

12,232

265

2%

Kentucky

15,588

-11

0%

15,588

-12

0%

15,588

-12

0%

15,588

-12

0%

Louisiana

15,476

87

1%

15,476

87

1%

15,476

87

1%

15,476

87

1%

Maryland

1,501

2

0%

1,267

1

0%

1,267

1

0%

1,350

2

0%

Michigan

13,898

1,167

8%

13,459

1,164

9%

11,182

1,203

11%

10,968

1,177

11%

New Jersey

1,346

-1

0%

1,346

-1

0%

1,346

-1

0%

1,346

-1

0%

New York

3,469

53

2%

3,469

53

2%

3,474

53

2%

3,456

53

2%

Ohio

15,829

315

2%

15,927

330

2%

15,927

330

2%

15,927

330

2%

Pennsylvania

11,896

361

3%

11,896

367

3%

11,896

367

3%

11,896

367

3%

Virginia

4,664

48

1%

4,274

47

1%

4,361

51

1%

4,025

48

1%

West Virginia

15,165

105

1%

15,165

104

1%

15,165

104

1%

15,165

104

1%

Total

124,057

2,493

2%

122,619

2,506

2%

119,453

2,551

2%

115,722

2,471

2%

68


-------
Appendix F: Feasibility Assessment for Engineering Analytics
Baseline

Similar to CSAPR Update Final Action, EPA analyzed and confirmed that the assumed fleet
operations in its baseline emissions and budget estimates were compatible with future load
requirements by verifying that new units in addition to the existing fleet would provide enough
generation, assuming technology-specific capacity factors, to replace the retiring generation
expected to occur in years 2021 through 2024. EPA assessed generation adequacy specific to the
12 states covered under this action. EPA uses these observations to determine whether any
assumed replacement generation from the existing fleet is necessary to offset the announced
retirements and continue to satisfy electricity load. Additionally, EPA looked at whether the
combination of new units (both fossil and non-fossil) provide sufficient new generation to
replace retiring generation. In this case, EPA found that the new unit generation from fossil and
renewable generation would exceed the generation from retiring units in all three scenarios
examined, indicating that no further replacement generation from existing units is needed.
Moreover, EPA found the change in generation from the covered fossil units to be within the
observed historical trend. EPA updated its analysis below at final rule taking into account the
latest announcements and commenter data on new units and retirements. EPA's conclusion was
further supported by its updated analysis and data.

•	EPA first identified the collective baseline heat input and generation from the 12 states
covered in this action and compared it to historical trends for these same states (Scenario
1). This illustrated that the assumed heat input and generation from fleet turnover was
well within with recent historical trends (see tables Appendix F-l,Appendix F-2, and
Appendix F-3 below).

•	EPA then compared the collective baseline heat input and generation from the 12 states
covered in this action to a scenario where fossil generation remains at 2019 levels instead
of continuing to decline (Scenario 2).

•	Finally, EPA identified the 2020 Energy Information Administration's Annual Energy
Outlook (EIA AEO) annual growth projections from 2019 through 2024 total electricity
demand levels (0.8%) from its reference case, and estimated an upperbound future year
scenario where covered fossil generation grew at levels matching this fleet-wide total
growth rate (Scenario 3).47

•	EPA's assessment illustrates the amount of generation in its baseline, factoring in
retirements and new fossil units, is more than sufficient to accommodate all three
scenarios.48 For instance, generation from covered fossil sources in these 12 states has
dropped at an average rate greater than 2% per year between 2016 and 2019 (410 TWh to

47	Department of Energy, Annual Energy Outlook 2020. Available at

https://www.eia.gov/outlooks/aeo/data/browser/#/?id=8-AE02020&cases=ref2020&sourcekey=0

48	Based on historical trends, modeling, and company statements, EPA expects levels similar to scenario 1 and
scenario 2 to be most likely.

69


-------
384 Twh). However, EPA's assumed baseline generation from covered fossil sources for
the 12 states reflects a rate of decline less than 2% per year. See Table Appendix F-2.

•	EPA then identified new RE capacity under construction, testing, or in site prep by 2021.
For years beyond 2021, EPA also identified new RE capacity that was planned but with
regulatory approvals pending for years 2022 and beyond (as this capacity is unlikely to
have yet started construction).49

•	EPA calculated and added the RE generation values to the fossil baseline to estimate
future year generation in the state (see Table Appendix F-3). EPA used a capacity factor
of 42.7% for wind, 21.6% for solar, and 65% forNGCC.

•	Using these technology-specific capacity factors based on past performance and IPM
documentation, EPA anticipated over 20 TWh from new generation already under
construction or being planned with regulatory approval received. This combined with the
baseline generation from existing units exceeds the expected generation load for the 12
states under all three scenarios.50

•	Not only is the future baseline generation level assumed in EPA's engineering analysis
well within the recent historical fossil generation trend (See Table Appendix F-2) on its
own (which illustrates no need for replacement generation), but it is also exceeds an
upper bound analysis for future covered fossil generation that assumes 0.8% growth from
the existing fossil fleet (scenario 3). Moreover, the potential new generation from RE
(over 7 TWh) when added to the baseline fossil generation values further increases the
amount by which baseline generation exceeds the historical fossil generation for the 12
states with assumed annual growth of 0.8%. This indicates that available capacity and
generation assumed would serve load requirements in this upper bound scenario.

Not included in the tables below nor in EPA's baseline, but listed in the latest EIA 860m
is even more planned NGCC combined cycle for years 2023 and 2024 that is pending
regulatory approval. Assuming some of this (low emitting generation) becomes available
in the outer years, that constitutes additional generation that exceeds EPA's upperbound
generation levels below - further bolstering the observation that no replacement
generation from existing units needs to be assumed to fill generation from retiring units.

Table Appendix F-l: Heat Input Change Due to Fleet Turnover (Historical and Future)



Reported Heat Input from Covered Fossil
Units (TBtu)



Assumed Heat Input from Covered Fossil
(TBtu)



2016

2017

2018

2019



2021

2022

2023

2024

Illinois

383.4

333.2

379.3

311.8



267

267

270

266

Indiana

415.6

379.1

432.3

356.5



357

352

356

297

49	Department of Energy, EIA Form 860, Generator Form 3-1. 2019 Early Release. Available at
https://www.eia.gov/electricity/data/eia860/

50	While EPA notes the baseline generation exceeds the covered fossil load in all three scenarios in Table F-3, EPA
anticipates scenarios 1 and 2 being more representative of likely covered fossil load based on historical trends,
future modeling, and utility resource plans.

70


-------
Kentucky

360.2

319.1

351.3

313.8



287

287

287

287

Louisiana

331.8

302

312.2

317.4



344

345

345

345

Maryland

108.7

76.9

95.7

83



82

79

79

94

Michigan

331.5

317

344.4

316.1



315

315

325

318

New Jersey

178.7

145.1

150.8

144.9



145

145

145

145

New York

269.7

199.7

228.6

195.6



227

227

228

228

Ohio

429

401

392.2

391.2



376

394

394

394

Pennsylvania

515.8

473.1

460.1

485.2



517

526

526

526

Virginia

259.9

228.2

241

237.9



249

246

261

252

West Virginia

323.1

324

303.6

287.9



285

285

285

285

Total

3,907

3,498

3,692

3,441



3,450

3,468

3,501

3,437

Table Appendix F-2: Generation Change Due to Fleet Turnover (Historical and Future)



Reported Generation from Covered
Fossil Units (TWh)



Assumed Generation from Covered
Fossil (TWh)

State

2016

2017

2018

2019



2021

2022

2023

2024

Illinois

38.6

33.9

38.7

32.7



28.3

28.3

29.6

29.2

Indiana

42.7

39.4

45.8

38.8



38.8

38.4

39.0

33.8

Kentucky

37.1

33.8

37.2

33.6



31.0

31.0

31.0

31.0

Louisiana

36

33.1

34.6

36.1



39.8

39.9

39.9

39.9

Maryland

11

7.9

10.4

9.5



9.4

9.1

9.1

11.5

Michigan

31.8

30.8

34

31.7



31.6

31.7

35.2

34.5

New Jersey

20.5

17.2

18.2

18



18.0

18.0

18.0

18.0

New York

30

22.5

25.6

22.5



26.4

26.4

26.5

26.5

Ohio

47.9

45.1

45.5

45.8



44.4

46.9

46.9

46.9

Pennsylvania

53.6

49.7

49.8

56.8



62.1

63.6

63.6

63.6

Virginia

27.8

25.5

27.1

28.9



30.8

30.8

33.3

32.4

West Virginia

33.9

33.8

31.8

29.9



29.9

29.9

29.9

29.9

Total

410.9

372.7

398.7

384.3



390.5

394.0

402.0

397.2

Table Appendix F-3: Assumed Baseline OS Generation and Expected New Build Generation from Covered

Fossil Units (TWh)



2019

2020

2021

2022

203

2024

Scenario 1 - Generation Levels (with continued pace of 2%
decline)

384.2

376.5

369.0

361.6

354.4

347.3

Scenario 2 - Generation Levels (no change from 2019)

384.2

384.2

384.2

384.2

384.2

384.2

Scenario 3 - Generation Levels (.8% growth from covered
fossil)

384.2

387.3

390.4

393.5

396.6

399.8















Assumed Baseline Fossil Generation with Reported Fossil
Retirement and Reported New Build





390.5

394.0

402.0

397.2

71


-------
New Build (Non-Fossil)





4.6

7.2

7.2

7.2

Total Baseline Generation





395.1

401.2

409.2

404.4

72


-------
Appendix G: State Emission Budgets and Variability Limits

Table Appendix G-l: State Emission Budgets and Variability Limits (tons)



2021

2022

2023

2024

State

Emission
Budgets

Variability
Limit

Emission
Budgets

Variability
Limit

Emission
Budgets

Variability
Limit

Emission
Budgets

Variability
Limit

Illinois

9,102

1,911

9,102

1,911

8,179

1,718

8,059

1,692

Indiana

13,051

2,741

12,582

2,642

12,553

2,636

9,564

2,008

Kentucky

15,300

3,213

14,051

2,951

14,051

2,951

14,051

2,951

Louisiana

14,818

3,112

14,818

3,112

14,818

3,112

14,818

3,112

Maryland

1,499

315

1,266

266

1,266

266

1,348

283

Michigan

12,727

2,673

12,290

2,581

9,975

2,095

9,786

2,055

New Jersey

1,253

263

1,253

263

1,253

263

1,253

263

New York

3,416

717

3,416

717

3,421

718

3,403

715

Ohio

9,690

2,035

9,773

2,052

9,773

2,052

9,773

2,052

Pennsylvania

8,379

1,760

8,373

1,758

8,373

1,758

8,373

1,758

Virginia

4,516

948

3,897

818

3,980

836

3,663

769

West Virginia

13,334

2,800

12,884

2,706

12,884

2,706

12,884

2,706

73


-------
Appendix H: AQAT Estimates for the 2021 and 2022 Sensitivity
Cases

EPA performed a series of sensitivity scenarios using an inventory based on 2021 and
one (for 2022) interpolated between 2021 and 2023. The fractional change in emissions are
shown below. Scenarios that are not viable, for technical or policy reasons, have been grayed
out in these tables.

Table H-l. 2021 Fractional Difference in Emissions Relative to 2023 Air Quality Modeling
Base Case for Each State.

State

Eng

Baseline

$500/
ton

Optimize

SCR

Optimize

SCR +
LNB

Optimize

SNCR+
SCR

Optimize

SNCR+
SCR +
LNB

New SNCR +
Optimize
SNCR+ SCR +
LNB

New SCR +
Optimize
SNCR+ SCR
+ LNB

Straight
Line

Interpolati
on

Alabama

0.094

0.094

0.094



0.094







0.140

Arizona

0.062

0.056

0.045



0.045







0.160

Arkansas

0.055

0.053

0.054



0.054







0.136

California

0.096

0.096

0.095



0.095







0.132

Colorado

0.033

0.033

0.033



0.032







0.117

Connecticut

0.044

0.041

0.041



0.041







0.138

Delaware

0.052

0.052

0.050



0.050







0.131

District of Columbia

0.042

0.042

0.042



0.042







0.129

Florida

0.083

0.083

0.070



0.070







0.182

Georgia

0.098

0.098

0.098



0.098







0.162

Idaho

0.089

0.089

0.089



0.089







0.136

Illinois

0.031

0.030

0.029



0.028







0.111

Indiana

0.045

0.042

0.011



0.010







0.164

Iowa

0.021

0.019

0.004



0.004







0.107

Kansas

0.022

0.022

0.012



0.011







0.088

Kentucky

0.189

0.190

0.184



0.184







0.201

Louisiana

0.126

0.125

0.125



0.120







0.091

Maine

0.107

0.107

0.107



0.107







0.131

Maryland

0.007

0.007

0.007



0.007







0.166

Massachusetts

0.024

0.024

0.024



0.024







0.109

Michigan

0.055

0.047

0.042



0.042







0.100

Minnesota

0.090

0.088

0.081



0.081







0.120

Mississippi

0.264

0.264

0.264



0.264







0.188

Missouri

0.040

0.039

0.026



0.025







0.161

Montana

0.098

0.098

0.098



0.098







0.116

Nebraska

0.057

0.055

0.056



0.056







0.103

Nevada

0.140

0.109

0.089



0.089







0.173

New Hampshire

0.103

0.103

0.092



0.092







0.147

New Jersey

0.044

0.044

0.042



0.042







0.157

New Mexico

0.057

0.057

0.054



0.054







0.091

New York

-0.009

-0.009

-0.009



-0.009







0.105

North Carolina

-0.065

-0.065

-0.065



-0.065







-0.052

North Dakota

0.051

0.050

-0.010



-0.012







0.093

Ohio

0.084

0.082

0.081



0.067







0.077

Oklahoma

0.024

0.024

0.024



0.024







0.034

Oregon

0.045

0.042

-0.017



-0.017







0.125

Pennsylvania

0.038

0.037

0.035



0.035







0.054

Rhode Island

0.064

0.064

0.064



0.064







0.134

South Carolina

0.004

0.004

-0.027



-0.027







0.122

South Dakota

0.039

0.040

0.040



0.040







0.136

Tennessee

0.038

0.038

0.014



0.014







0.128

Texas

0.135

0.135

0.135



0.135







0.167

Utah

0.075

0.075

0.072



0.072







0.163

Vermont

0.063

0.060

0.055



0.055







0.095

Virginia

0.076

0.076

0.076



0.076







0.580

Washington

0.055

0.003

0.002



0.002







0.142

West Virginia

0.088

0.087

0.087



0.087







0.143

74


-------
Wisconsin

0.100

0.100

0.099



0.097







0.176

Wyoming

0.105

0.105

0.105



0.105







0.148

Tribal Data

-0.016

-0.019

-0.044



-0.050







0.042

75


-------
Table H-2. 2022 Fractional Difference in

Emissions Relative to 2023 Air Quality Modeling Base Case for Each State

State

Eng

Baseline

$500/ton

Optimize

SCR

Optimize

SCR + LNB

Optimize

SNCR+
SCR

Optimize

SNCR+
SCR+LNB

New SNCR
+ Optimize

SNCR+
SCR + LNB

New SCR +
Optimize

SNCR+
SCR+LNB

Straight
Line

Interpolatio
n

Alabama

0.065

0.065



0.064



0.064





0.070

Arizona

0.035

0.029



0.018



0.018





0.080

Arkansas

0.019

0.017



0.018



0.018





0.068

California

0.050

0.050



0.050



0.050





0.066

Colorado

0.019

0.019



0.018



0.018





0.059

Connecticut

0.013

0.010



0.010



0.010





0.069

Delaware

0.024

0.024



0.022



0.022





0.066

District of
Columbia

0.021

0.021

0.021

0.021

0.021

0.021

0.021

0.021

0.065

Florida

0.062

0.062



0.050



0.050





0.091

Georgia

0.062

0.062



0.061



0.061





0.081

Idaho

0.056

0.056



0.056



0.056





0.068

Illinois

0.010

0.010



0.008



0.007





0.055

Indiana

0.009

0.007



-0.025



-0.025





0.082

Iowa

-0.014

-0.016



-0.032



-0.032





0.053

Kansas

-0.007

-0.007



-0.017



-0.017





0.044

Kentucky

0.165

0.165



0.136



0.136





0.100

Louisiana

0.108

0.108



0.108



0.102





0.046

Maine

0.077

0.077



0.077



0.077





0.065

Maryland

-0.012

-0.012



-0.012



-0.012





0.083

Massachusetts

0.014

0.014



0.014



0.014





0.055

Michigan

0.030

0.022



0.017



0.017





0.050

Minnesota

0.040

0.038



0.032



0.032





0.060

Mississippi

0.228

0.228



0.218



0.218





0.094

Missouri

0.005

0.004



-0.010



-0.011





0.080

Montana

0.046

0.046



0.046



0.046





0.058

Nebraska

0.021

0.019



0.008



0.008





0.051

Nevada

0.073

0.043



0.024



0.024





0.086

New Hampshire

0.081

0.081



0.070



0.070





0.073

New Jersey

0.020

0.020



0.017



0.017





0.078

New Mexico

0.033

0.033



0.030



0.030





0.046

New York

-0.015

-0.015



-0.016



-0.016





0.052

North Carolina

-0.032

-0.032



-0.032



-0.032





-0.026

North Dakota

0.020

0.019



-0.041



-0.043





0.047

Ohio

0.049

0.048



0.046



0.033





0.038

Oklahoma

0.012

0.012



0.012



0.012





0.017

Oregon

0.016

0.013



-0.046



-0.046





0.062

Pennsylvania

0.019

0.018



0.016



0.016





0.027

Rhode Island

0.038

0.038



0.038



0.038





0.067

South Carolina

0.006

0.005



-0.025



-0.025





0.061

South Dakota

0.010

0.010



0.010



0.010





0.068

76


-------
Tennessee

0.011

0.011



-0.014



-0.014





0.064

Texas

0.073

0.072



0.072



0.072





0.083

Utah

0.049

0.049



0.046



0.046





0.081

Vermont

0.037

0.034



0.029



0.029





0.048

Virginia

0.079

0.078



0.078



0.078





0.290

Washington

0.034

-0.018



-0.019



-0.019





0.071

West Virginia

0.049

0.049



0.049



0.049





0.072

Wisconsin

0.070

0.070



0.065



0.064





0.088

Wyoming

0.073

0.073



0.073



0.073





0.074

Tribal Data

-0.028

-0.031



-0.064



-0.070





0.021

Table H-3. 2021 Average Ozone DVs (ppb) for NOx Emissions Cost Threshold Levels ($/ton) Assessed Using the Ozone AQAT

Monitor
Identification
Number

State

County

CAMx
2023 Base
Case (ppb)

Assessment Tool Average Ozone Design Values (ppb).

Straight
line

Engineering
Baseline

$500/t
on

Optimi

ze

SCR

Optimi

ze SCR
+ LNB

Optimi

ze
SNCR
+ SCR

Optimi

ze
SNCR
+ SCR
+ LNB

New
SNCR +
Optimiz
e

SNCR+
SCR +
LNB

New SCR

+

Optimize

SNCR+
SCR +
LNB

90013007

CT

Fairfield

74.3

76.50

74.77

74.76

74.59

74.58

74.59

74.58

74.54

74.48

90019003

CT

Fairfield

76.9

78.56

77.26

77.25

77.12



77.11







90099002

CT

New Flaven

71.7

73.98

72.21

72.19

72.02



72.01







482010024

TX

Flarris

74.0

75.51

75.14

75.10

75.03



75.02







Table H-4. 2021 Maximum Ozone DVs (ppb) for NOx Emissions Cost Threshold Levels ($/ton) Assessed Using the Ozone
AQAT for All Receptors.		









Assessment Tool Maximum Ozone Design Values (ppb).

Monitor
Identification
Number





CAMx
2023 Base
Case (ppb)



Engineering
Baseline

$500/t
on

Optim

ize

Optimi

ze SCR

Optimi

ze

Optimi

ze

New
SNCR +

New SCR
+ Optimize

State

County

Straigh
t line





SCR

+ LNB

SNCR
+ SCR

SNCR
+ SCR
+ LNB

Optimiz

e

SNCR+
SCR +
LNB

SNCR+
SCR +
LNB

90013007

CT

Fairfield

75.2

77.43

75.68

75.66

75.50

75.48

75.49

75.48

75.44

75.38

90019003

CT

Fairfield

77.2

78.86

77.56

77.55

77.42



77.41







90099002

CT

New Flaven

73.8

76.15

74.32

74.31

74.13



74.12







482010024

TX

Flarris

75.6

77.15

76.76

76.72

76.66



76.65







Table H-5. 2022 Average Ozone DVs (ppb) for NOx Emissions Cost Threshold Levels ($/ton) Assessed Using the Ozone AQAT
for All Receptors.

77


-------
Monitor
Identification
Number

State

County

CAMx
2023 Base
Case (ppb)

Assessment Tool Average Ozone Design Values (ppb).

Straight
line

Engineering
Baseline

$500/t
on



Optimi

ze SCR
+ LNB



Optimi

ze
SNCR
+ SCR
+ LNB





90013007

CT

Fairfield

74.3

75.40

74.48

74.46



74.29



74.28





90019003

CT

Fairfield

76.9

77.73

77.04

77.03



76.89



76.88





90099002

CT

New Haven

71.7

72.84

71.89

71.87



71.68



71.68





482010024

TX

Harris

74.0

74.76

74.74

74.70



74.64



74.63





Table H-6. 2022 Maximum Ozone DVs (ppb) for NOx Emissions Cost Threshold Levels ($/ton) Assessed Using the Ozone

AQAT for All Receptors.

Monitor
Identification
Number

State

County

CAMx
2023 Base
Case (ppb)

Assessment Tool Maximum Ozone Design Values (ppb).

Straigh
t line

Engineering
Baseline

$500/t
on



Optimi

ze SCR
+ LNB



Optimi

ze
SNCR
+ SCR
+ LNB





90013007

CT

Fairfield

75.2

75.38

75.37

75.20



75.20



75.14





90019003

CT

Fairfield

77.2

77.34

77.33

77.20



77.19



77.15





90099002

CT

New Haven

73.8

74.00

73.98

73.80



73.79



73.73





482010024

TX

Harris

75.6

76.36

76.32

76.25



76.24



75.98





78


-------