[Federal Register Volume 85, Number 36 (Monday, February 24, 2020)]
[Proposed Rules]
[Pages 10350-10357]
From the Federal Register Online via the Government Publishing Office [www.gpo.gov]
[FR Doc No: 2020-03507]


=======================================================================
-----------------------------------------------------------------------

ENVIRONMENTAL PROTECTION AGENCY

40 CFR Part 52

[EPA-R05-OAR-2015-0700; FRL-10005-64-Region 5]


Air Plan Approval; Indiana; Attainment Plan for Sulfur Dioxide in 
Southwest Indiana

AGENCY: Environmental Protection Agency (EPA).

ACTION: Proposed rule.

-----------------------------------------------------------------------

SUMMARY: The Environmental Protection Agency (EPA) is reproposing to 
approve under the Clean Air Act an element of the State Implementation 
Plan (SIP) revision for attaining the 1-hour sulfur dioxide 
(SO2) primary national ambient air quality standard (NAAQS) 
for the Southwest Indiana nonattainment area (including parts of 
Daviess and Pike Counties), based on revised limits for the 
Indianapolis Power and Light's Petersburg facility (IP&L-Petersburg) 
that Indiana submitted on September 18, 2019. Indiana's revised limits 
are based on the same dispersion modeling and the same 1-hour average 
emission rates that EPA proposed to conclude would result in 
attainment. However, the revised limits reflect revised calculations of 
the degree of adjustment needed for the 30-day average limits to be 
comparably stringent to 1-hour limits at the modeled emission rates. 
EPA is soliciting additional comments that may arise from these 
revisions.

DATES: Comments must be received on or before March 25, 2020.

ADDRESSES: Submit your comments, identified by Docket ID No. EPA-R05-
OAR-2015-0700 at http://www.regulations.gov, or via email to 
[email protected]. For comments submitted at Regulations.gov, follow 
the

[[Page 10351]]

online instructions for submitting comments. Once submitted, comments 
cannot be edited or removed from Regulations.gov. For either manner of 
submission, EPA may publish any comment received to its public docket. 
Do not submit electronically any information you consider to be 
Confidential Business Information (CBI) or other information whose 
disclosure is restricted by statute. Multimedia submissions (audio, 
video, etc.) must be accompanied by a written comment. The written 
comment is considered the official comment and should include 
discussion of all points you wish to make. EPA will generally not 
consider comments or comment contents located outside of the primary 
submission (i.e., on the web, cloud, or other file sharing system). For 
additional submission methods, please contact the person identified in 
the FOR FURTHER INFORMATION CONTACT section. For the full EPA public 
comment policy, information about CBI or multimedia submissions, and 
general guidance on making effective comments, please visit http://www2.epa.gov/dockets/commenting-epa-dockets.

FOR FURTHER INFORMATION CONTACT: John Summerhays, Environmental 
Scientist, Attainment Planning and Maintenance Section, Air Programs 
Branch (AR-18J), Environmental Protection Agency, Region 5, 77 West 
Jackson Boulevard, Chicago, Illinois 60604, (312) 886-6067, 
[email protected].

SUPPLEMENTARY INFORMATION: Organization of this document. The following 
outline is provided to aid in locating information in this preamble.

Table of Contents

I. History of Nonattainment Planning for SO2 in Southwest 
Indiana
II. Indiana's Revisions to Limits for IP&L-Petersburg
III. EPA Guidance Regarding Data Handling for Calculating Longer 
Term Average SO2 Emission Limits
IV. EPA's Evaluation of the IP&L-Petersburg Limit Revisions
V. EPA's Proposed Action
VI. Incorporation by Reference
VII. Statutory and Executive Order Reviews

I. History of Nonattainment Planning for SO2 in Southwest 
Indiana

    In 2013, in implementing its 2010 1-hour primary SO2 
NAAQS of 75 parts per billion (ppb), EPA designated a first set of 29 
areas of the country as nonattainment for this NAAQS, including the 
Southwest Indiana area (defined to include portions of Daviess and Pike 
Counties). See 78 FR 47191 (August 5, 2013), codified at 40 CFR part 
81, subpart C. In response to the resulting Clean Air Act requirements 
to adopt and submit to EPA a SIP demonstrating attainment of the NAAQS, 
Indiana submitted nonattainment plans for this and for three other 
areas on October 2, 2015. Indiana then submitted supplemental material 
pertinent in part to Southwest Indiana on November 15, 2017.
    On August 15, 2018, EPA published a proposed rule that proposed to 
approve the SO2 nonattainment plans for the Southwest 
Indiana, Indianapolis, and Terre Haute areas. (See 83 FR 40487.) EPA 
received no comments addressing the Indianapolis and Terre Haute areas, 
and EPA published a final rule regarding these two areas on March 22, 
2019 (84 FR 10692). EPA also published separate actions regarding the 
SO2 attainment plan for Morgan County, including a proposed 
rule published on July 9, 2019 (at 84 FR 32672) and a final rule 
published on September 23, 2019 (at 84 FR 49659). This rule therefore 
does not address these three areas that were addressed in Indiana's 
2015 submittal, and only addresses SO2 in Southwest Indiana.
    Indiana's plan for SO2 in Southwest Indiana addresses a 
number of Clean Air Act requirements that SO2 nonattainment 
plans must meet in order to be approved by EPA, including requirements 
for emission limits sufficient to provide for attainment, a modeling 
demonstration that these limits in fact provide for attainment, and 
requirements for an emissions inventory, reasonably available control, 
reasonable further progress (RFP), and contingency measures.\1\ 
Emission limits were submitted in October 2015 for both IP&L-Petersburg 
and the Frank E. Ratts facility. EPA's August 15, 2018 proposed action 
on this rule addressed these requirements.
---------------------------------------------------------------------------

    \1\ As discussed in the notice of proposed rulemaking, 83 FR 
40487, 40497 (Aug. 15, 2018), EPA already approved Indiana's 
nonattainment new source review rules on October 7, 1994, with 
subsequent amendments approved July 8, 2011. See 59 FR 51108 and 76 
FR 40242. EPA explained that these rules provide for appropriate new 
source review for SO2 sources undergoing construction or 
major modification without the need for modification of the approved 
rules. Therefore, EPA concluded that Indiana had satisfied 
nonattainment new source review requirements previously for the 
Southwest Indiana area and did not need to include any provisions to 
address these requirements in its 2015 submittal.
---------------------------------------------------------------------------

    In response to the August 15, 2018 proposed rule, EPA received 
comments on Indiana's 30-day average limits for IP&L-Petersburg, which 
prompted Indiana to reevaluate these limits.\2\ This reevaluation led 
Indiana to adopt a revised set of limits for IP&L-Petersburg, 
incorporated in Commissioner's Order Number 2019-2, which Indiana 
submitted to EPA on September 18, 2019. As explained below, these 
revised limits are based on the same modeling that was used to derive 
the limits in Indiana's October 2015 submittal and, thus, reflect the 
same critical emission values that Indiana identified in its October 
2015 submittal. However, as explained further below, Indiana 
reevaluated the adjustment factor that it used to determine the 30-day 
average emission limits for IP&L-Petersburg, which resulted in 
calculation of a revised adjustment factor and therefore revised 
emission limits. These limits were incorporated in Commissioner's Order 
Number 2019-2, which was issued on July 31, 2019 and became effective 
on August 18, 2019.
---------------------------------------------------------------------------

    \2\ EPA received no comments regarding the 1-hour limits for the 
Frank E. Ratts facility.
---------------------------------------------------------------------------

    Indiana's October 2, 2015 submittal addresses reasonably available 
control measures (RACM) and RFP by means of its adopted limits, so the 
limit revisions implicitly affect these elements of the plan. However, 
Indiana's recent submittal did not otherwise revise its plan with 
respect to these elements, and EPA continues to believe that Indiana 
has met these requirements. The primary focus of this proposed action 
is to evaluate whether these revised limits, in conjunction with other 
limits that Indiana submitted previously, provide for attainment of the 
SO2 NAAQS in Southwest Indiana and continue to support EPA's 
proposed conclusions regarding Indiana's satisfaction of the RACM and 
RFP elements.

II. Indiana's Revisions to Limits for IP&L-Petersburg

    Indiana's October 2015 submittal included two sets of limits for 
IP&L-Petersburg, including one set using 1-hour average emission limits 
and one set using 30-day average limits, with provisions for IP&L to 
select which limits would apply. IP&L has requested that the 30-day 
average limits apply, and IP&L's involvement in pursuing modified 30-
day average limits suggests that IP&L envisions continuing to be 
subject to 30-day average limits. Nevertheless, Indiana requested that 
EPA approve both the 1-hour limits in 326 IAC 7-4-15 and the 30-day 
average limits in the commissioner's order, and EPA is reproposing 
action accordingly.
    Historically, EPA required states to establish short-term emission 
limits at the level that modeling shows provides for NAAQS attainment, 
a level known as the critical emission value, with averaging times of 
limits expected to match the averaging time of the relevant NAAQS. EPA 
guidance for SO2 nonattainment plans under the 2010 1-hour 
NAAQS states that limits with

[[Page 10352]]

averaging times up to 30 days may, in appropriate circumstances, 
provide a suitable basis for plans to ensure attainment of that NAAQS. 
However, EPA recommends that, to serve this purpose, any such limit 
should be designed to have comparable stringency to a 1-hour average 
limit at the critical emission value. Appendix C of EPA's guidance 
provides a recommended procedure for determining adjustment factors 
which may be multiplied by the value of a candidate 1-hour limit to 
estimate a longer term averaged limit that is presumptively comparably 
stringent. This procedure uses a pertinent hourly emissions data set to 
determine the 99th percentile among 1-hour average emission values, to 
determine the 99th percentile among longer term averaged values, and to 
calculate the ratio between these two 99th percentile values in order 
to determine an adjustment factor to be applied in determining the 
longer term average limit. This adjustment factor represents an 
estimate of the change in stringency from applying the limit on a 
longer term average basis rather than on a 1-hour basis, so that the 
adjusted longer term limit is estimated to be comparably stringent to a 
1-hour limit at the critical emission value. The guidance document 
(including appendix C) provides extensive guidance on the data sets and 
the calculation procedures that EPA advises be used in these 
determinations.\3\
---------------------------------------------------------------------------

    \3\ This guidance, issued on April 23, 2014, entitled, 
``Guidance for 1-Hour SO2 Nonattainment Area SIP 
Submissions,'' is available at https://www.epa.gov/sites/production/files/2016-06/documents/20140423guidance_nonattainment_sip.pdf. This 
guidance is discussed at length in the August 15, 2018 notice of 
proposed rulemaking identified above.
---------------------------------------------------------------------------

    Indiana used this general approach to determine the 30-day average 
limits adopted for purposes of its 2015 submittal. Based on historical 
emissions data from a stack that vents controlled emissions from Unit 2 
of IP&L-Petersburg, Indiana calculated an adjustment factor of 80 
percent, leading Indiana to establish 30-day average limits at a level 
that was 80 percent of 1-hour emission rates that were reflected in its 
attainment demonstration modeling.
    EPA's proposed rulemaking on Indiana's 2015 submittal elicited 
public comments that, among other issues, addressed the suitability of 
elements of the derivation of this adjustment factor. In response, 
Indiana recalculated the adjustment factor to be applied in determining 
the 30-day average limits for IP&L-Petersburg, and submitted these 
revised calculations and the resulting adopted limits on September 18, 
2019. Although this recalculation used the same data set as the 
original submittal, namely the 2006 to 2010 emissions from the main 
stack at IP&L-Petersburg Unit 2, Indiana used an edited data set 
reflecting removal of a number of inappropriate zero entries (for hours 
with no operation and, thus, no valid pound per million British Thermal 
Unit (lb/MMBTU) value) and removal of selected hours with questionable 
data. The revised calculations are provided in a spreadsheet that is 
available in the docket for this action, along with spreadsheets 
showing related EPA calculations described below.
    Indiana's recalculated adjustment factor was 68 percent. That is, 
Indiana's revised evaluation determined that the 30-day average limits 
for IP&L-Petersburg should be 68 percent (reduced from 80 percent) of 
the 1-hour average emissions limit indicated by the attainment 
demonstration modeling. Indiana conducted no additional modeling, and 
instead relied on the same critical emission values as were described 
in its 2015 submittal. The revised limits are shown in Table 1, along 
with the original limits. This table also shows the emission rates 
(identified as critical emission rates, expressed in lbs/MMBTU) that 
correspond (at maximum heat input) to the modeled critical emission 
values.

                                   Table 1--Revised Limits for IP&L-Petersburg
----------------------------------------------------------------------------------------------------------------
                                                              Critical
                          Unit                             emission rate      Revised limit      Original limit
                                                            (lbs/MMBTU)        (lbs/MMBTU)        (lbs/MMBTU)
----------------------------------------------------------------------------------------------------------------
Unit 1.................................................               0.15               0.10               0.12
Unit 2.................................................               0.15               0.10               0.12
Unit 3.................................................               0.37               0.25               0.29
Unit 4.................................................               0.35               0.24               0.28
----------------------------------------------------------------------------------------------------------------

    Indiana provided additional rationale for its selection of data for 
performing these calculations. IP&L reports data for two emission 
streams at Petersburg Unit 2, identified as main stack emissions and 
bypass stack emissions. Indiana explained that the main stack vents 
emissions that have been controlled by the unit's flue gas 
desulfurization equipment, whereas the bypass stack vents emissions 
that bypass such control. Therefore, Indiana explained, data on 
emissions from the main stack \4\ provide the best representation of 
the prospective variability of emissions that are controlled well 
enough to meet the limits necessary to provide for attainment. Indiana 
explained further that while its limits govern all emissions from each 
unit (which is to say the sum of emissions from the main stack and from 
the bypass stack at Unit 2, and similarly for the pair of stacks at 
Unit 1), compliance with the adopted limits will require nearly 
eliminating emissions that bypass the control system.
---------------------------------------------------------------------------

    \4\ Indiana refers to this stack as the ``FGD stack,'' i.e., the 
stack venting emissions controlled by the flue gas desulfurization 
system.
---------------------------------------------------------------------------

    In addition to the availability of a data set of controlled 
emissions at Unit 2, Indiana also provided an additional rationale for 
using data from Unit 2 to assess variability expected upon compliance 
with SIP limits. Unit 1 has a similar setup as Unit 2, with separate 
vents for controlled versus uncontrolled emissions. However, Indiana 
explained that historic data from Unit 1 included a high fraction of 
times when emissions exited through the bypass stack, so that the 
resulting data set is both less robust and less predictive of effective 
control equipment operation. Units 3 and 4 do not have separate vents 
for controlled versus uncontrolled emissions, so data from these units 
do not properly represent the variability of controlled emissions. 
Units 3 and 4 only have single stacks, venting a combination of 
controlled and uncontrolled emissions, so the historic data from these 
units show variability that is dominated by variability in control 
level, thus providing a poor data set for projecting the variability of 
controlled emissions.
    Indiana's October 2015 submittal included both mass limits (in 
pounds per hour (lbs/hr)) and emission rate limits (lbs/MMBTU) for 
IP&L-

[[Page 10353]]

Petersburg, and applied the same adjustment factor to the modeled 1-
hour values for these respective variables. EPA guidance provides for 
separate calculations of adjustment factors for these separate limits, 
which would reflect the different impact on stringency that can result 
from expressing a mass limit versus an emission rate limit as a 30-day 
average limit. Accordingly, Indiana reconsidered this feature of its 
October 2015 submittal, with the result that the replacement 30-day 
average limits for IP&L-Petersburg only include emission rate limits 
(in lbs/MMBTU), based on a view that limits on emission rates alone 
suffice, even at maximum heat inputs, to assure that the area will 
attain the standard.
    An important aspect of any longer term average emission limit is 
the set of data handling procedures to be used in determining 
compliance. Indiana's commissioner's order makes no direct statement 
regarding data handling procedures. However, the order states that the 
``requirements of this Order are in addition to any less stringent 
requirements applicable to [IP&L] pursuant to 326 IAC 7-4-15,'' 
implying that the state intends that compliance with the 30-day average 
limits in the order is to be evaluated using the same procedures as 
those for the 30-day average limits in the rule. Paragraph (d) of 326 
IAC 7-4-15, which Indiana requests be incorporated into the SIP, states 
that ``Compliance with [the 30-day average limits in the rule] shall be 
determined by calculating the thirty (30) boiler operating day rolling 
arithmetic average emission rate at the end of each boiler operating 
day using all of the quality assured hourly average continuous emission 
monitoring system data for the previous thirty (30) boiler operating 
days.''
    Indiana's submittal also includes a copy of the letter which 
transmitted the commissioner's order to IP&L. This letter describes the 
order as applying the data handling procedures of 326 IAC 7-4-15(d), 
and notes further that the ``methodology is documented in IPL-
Petersburg's [compliance] assessment protocol, which follows 
methodologies recommended in U.S. EPA's Mercury and Air Toxics Standard 
(MATS) rule guidance and the U.S. EPA memorandum `Guidance for 1-Hour 
SO2 Nonattainment Area SIP Submissions'.''

III. EPA Guidance Regarding Data Handling for Calculating Longer Term 
Average SO2 Emission Limits

    EPA's guidance on 1-hour SO2 nonattainment plans, issued 
in April 2014, provides numerous detailed recommendations regarding 
longer term average SO2 emission limits, including several 
recommendations regarding data handling procedures.\5\ The guidance 
states that the rule promulgating MATS provides a good prototype for 
procedures for data handling. The guidance recommends the MATS approach 
of only averaging data obtained during operating hours, so that the 
compliance assessment focuses on how well emissions are controlled and 
is not influenced by the fraction of time that the facility operates. 
The guidance recommends that emission limits averaged over multiple 
days be addressed by averaging emissions over the pertinent number of 
operating days, as is done in MATS, which improves robustness of the 
compliance determination by helping assure that the compliance 
determination reflects an adequate set of data. The guidance recommends 
determining compliance with limits on emission factors (e.g., limits on 
pounds of emissions per megawatt-hour) by dividing total mass over the 
30 operating days by the total electrical output during that period. 
(The analogous approach for a limit expressed in pounds per MMBTU is to 
divide total pounds of emissions over the averaging period by total 
heat input in MMBTU during the period.) The guidance explains that this 
approach effectively weights each hour's data point according to the 
hour's emissions (more precisely, according to the hour's electrical 
output or heat input), and thus better indicates the average rate of 
emissions than, for example, computing an average of hourly average 
emission rates.
---------------------------------------------------------------------------

    \5\ Guidance is cited in footnote 3 above. See especially page 
32.
---------------------------------------------------------------------------

    Unfortunately, in this last respect, EPA's SO2 
nonattainment planning guidance misrepresents the data handling 
procedures in MATS. In fact, MATS, consistent with common practice, 
determines compliance by averaging the pertinent hourly values, either 
in pounds per megawatt or in pounds per MMBTU (reflecting the units of 
the applicable limit). See 40 CFR 63.10021. On the other hand, while 
EPA promulgated MATS as a national emission standard for hazardous air 
pollutants (NESHAP) under Clean Air Act section 112, EPA also 
simultaneously promulgated revisions to new source performance 
standards (NSPS) under Clean Air Act section 111 with limits in which, 
for facilities constructed, modified, or reconstructed after May 3, 
2011, ``compliance . . . is determined by dividing the sum of the 
SO2 . . . emissions for the 30 successive boiler operating 
days by the sum of the [energy output] for the 30 successive boiler 
operating days.'' See 40 CFR 60.48Da(d), promulgated on February 16, 
2012, 77 FR 9304, 9454. Thus, while the substance of EPA's 
recommendations was clear, the guidance was incorrect in its 
description of the data handling procedures of MATS, and the guidance 
should have cited the revisions to the NSPS for sources that began 
construction, modification or reconstruction after May 3, 2011 as a 
template for relevant data handling provisions, rather than the 
procedures of MATS. The following section reviews Indiana's revised 
submittal in light of this clarified guidance.

IV. EPA's Evaluation of the IP&L-Petersburg Limit Revisions

    EPA conducted multiple analyses of the expected variability of 
emissions at IP&L-Petersburg upon compliance with Indiana's limits. 
These analyses inform EPA's judgment as to whether Indiana's revised 
limits can be expected to be comparably stringent to 1-hour limits at 
the critical emission values.
    The first analysis used the data provided by Indiana but used a 
different data handling procedure. Indiana's rule (326 IAC 7-4-15) 
specifies that compliance with the 30-day average limits in the rule 
shall be evaluated by determining the ``30 boiler operating day rolling 
arithmetic average emission rate at the end of each boiler operating 
day using all of the quality assured hourly average continuous emission 
monitoring system data.'' This indicates that if, for example, a 30-
operating day period has 700 operating hours with valid data, the 
compliance determination for that period would be based on the average 
of those 700 hourly values.
    The variability analysis provided by Indiana deviates from this 
procedure by first calculating daily average emission rates and then 
calculating averages of 30 operating days of daily averages. This 
approach gives more weight to days with fewer operating hours than the 
approach in 326 IAC 7-4-15. To evaluate the significance of this 
difference, EPA calculated a set of 30-day average emission rates based 
on the arithmetic average of all hourly emission rates. EPA's guidance 
is to use the same data handling approach in the assessment of 
variability as is provided in the state's compliance determination 
procedures, in order best to determine the degree to which use of a 
long term average limit affects stringency of the limit with those 
compliance procedures. Nevertheless, EPA's analysis found that

[[Page 10354]]

the use of averaging procedures consistent with Indiana's compliance 
determination procedures only modestly affected the resulting 
adjustment factor; compared to IP&L's adjustment factor of 68 percent, 
use of Indiana's compliance determination procedures using the same 
data set yielded an adjustment factor of 68.2 percent.
    IP&L explained that the data it used in its analysis are for the 
``FGD stack,'' which corresponds to the monitoring site identified in 
data reported to EPA as ``MS2S.'' However, the data reported to EPA for 
these emissions differ from the emissions used by IP&L for slightly 
over the first three years, most of the data reported to EPA appear to 
reflect approximately an 11 percent bias adjustment that is not 
reflected in the data used by Indiana. Therefore, EPA conducted an 
additional analysis of data reported to EPA for the MS2S monitoring 
site. Despite the difference in magnitudes of the emissions in these 
two data sets, the variability of emissions is similar, with EPA 
suggesting an adjustment factor of 65.0 percent, modestly lower than 
the 68.0 percent estimated by Indiana.
    EPA also examined data reported to EPA for the main stack at Unit 1 
for the same period examined by Indiana (2006 to 2010). EPA concurs 
with Indiana that this is a less robust data set that appears less 
representative of future controlled operations at this plant. The 
adjustment factor calculated from data for this stack (62.2 percent) is 
somewhat lower than the 68.0 percent adjustment factor that IP&L 
calculated from Unit 2 main stack data, which may reflect what appears 
to be comparatively unstable operation of control at Unit 1. Therefore, 
these Unit 1 data are consistent with Indiana's view that the historic 
data from the main stack at Unit 2 are the best predictors of 
variability from the four units at IP&L-Petersburg upon compliance with 
the limits.
    EPA's general objective is to evaluate the degree of variability, 
in particular the impact of variability on the stringency of an 
emission limit expressed in this case as a 30-day average limit rather 
than as a 1-hour limit. EPA seeks for this evaluation to be predictive 
of the degree of variability that can be expected once the source is 
complying with the control requirements of the SIP. The rules Indiana 
submitted in October 2015 required compliance with the limits by 
January 2017. Although Indiana's September 18, 2019 submittal imposes 
slightly more stringent limits than its October 2, 2015 submittal, the 
control measure in either case is the existing flue gas desulfurization 
equipment, and EPA anticipates that the slight increase in control 
efficiency needed to meet the new limits will not materially increase 
the variability in emissions upon compliance with these limits. 
Therefore, the data that are available for 2\1/2\ years starting 
January 2017 provide a valuable indication of the likely degree of 
emissions variability that can be expected to apply into the future 
with compliance with the newer limits.
    For these reasons, EPA analyzed the emissions data from January 
2017 to June 2019 for each of the four units at IP&L-Petersburg. In 
this analysis, for Units 1 and 2, in both cases EPA used the sum of 
emissions from the main stack and from the bypass stack, reflecting the 
fact that Indiana's limits govern total emissions from each unit. In 
order to apply the same data handling procedures as are used to 
determine compliance with the limits, EPA considered only days in which 
the unit operated, EPA computed 30-operating-day averages ending at the 
end of each operating day, and EPA computed the average emission rate 
as an arithmetic average among the valid operating hour emission rate 
data. Substitution data (conservative emission estimates derived 
according to trading program requirements in cases where information 
needed for a precise emission calculation was missing) appeared to be 
rare and unlikely to affect results significantly, and so EPA's 
analyses used a complete data set that reflected no deletion of any 
substitution data.
    EPA summarizes the results of these analyses in Table 2. Two 
spreadsheets that are included in the docket, including one for 2006 to 
2010 data and one for 2017 to 2019 data, show the data and the 
calculations used in these analyses.

             Table 2--Adjustment Factors for IP&L-Petersburg
------------------------------------------------------------------------
             Analysis                    Resulting adjustment factor
------------------------------------------------------------------------
IP&L analysis, using Unit 2 Main    68.0 percent.
 Stack data (2006-2010).
Using IP&L data with Indiana        68.2 percent.
 compliance statistics.
Using EPA data (Unit 2 main stack,  65.0 percent.
 2006-2010).
Using 2017--mid-2019 data.........  Unit 1: 73.0 percent.
                                    Unit 2: 57.6 percent.
                                    Unit 3: 68.6 percent.
                                    Unit 4: 70.4 percent.
                                    Average: 67.4 percent.
------------------------------------------------------------------------

    As noted above, Indiana used data from the stack at Unit 2 that 
vents controlled emissions to determine an adjustment factor to apply 
in determining 30-day average limits. Indiana has confirmed that these 
limits govern the total of all emissions from the respective units; in 
particular the limits for Units 1 and 2 govern the sum of emissions 
from the main stack plus the emissions from the bypass stack for each 
of these two units. The determination of an adjustment factor from just 
the main stack data reflects a premise that the historic data for the 
controlled emission stack is most indicative of the prospective 
variability of all emissions once the control requirements of the SIP 
are met. This premise in turn reflects an expectation that 
implementation of the control strategy will result in (uncontrolled) 
bypass stack emissions being minimal.
    EPA used the available 2017 to 2019 data to test these premises. 
For 2006 to 2010, according to data reported to EPA, bypass stack 
emissions for the 5 years accounted for 89 percent of the total Unit 1 
emissions and 30 percent of the total Unit 2 emissions. In contrast, 
for 2017 to mid-2019, bypass stack emissions accounted for only 3 
percent of emissions from Unit 1 and 0.2 percent of emissions from Unit 
2.
    In any case, the adjustment factors shown in Table 2 above for 2017 
to mid-2019 are based on statistics for total emissions for each unit, 
which for Units 1 and 2 reflect the sum of emissions from the main 
stack plus emissions from the bypass stack. Thus, the results in Table 
2 for recent emissions represent

[[Page 10355]]

the strongest evidence that the 2006 to 2010 data for the main stack at 
Unit 2 provides a suitable projection of the degree of variability in 
total emissions upon implementation of the SIP limits.
    Since the methods recommended in appendix C of the guidance rely on 
99th percentile values, the guidance recommends assuring that these 
assessments be based on a robust data set. For this reason, the 
guidance recommends using a data set with three to five years of data. 
The post-control data being used here represent only 2\1/2\ years. 
Therefore, EPA averaged the adjustment factors for the four units 
(shown in Table 2) in order to improve the robustness of the analysis.
    As shown in Table 2 above, the post-control data for the four units 
at IP&L-Petersburg support an average adjustment factor of 67.4 
percent, very close to the 68.0 percent adjustment factor applied by 
Indiana.\6\ The similarity of these percentages support several 
findings. First, the 2006 to 2010 data for the stack known as MS2S, the 
stack that vents controlled emissions from Unit 2, provide a good 
representation of the variability of emissions to be expected upon 
implementation of the limits in Indiana's plan. Most plants do not have 
separate vents for controlled versus uncontrolled emissions, but the 
availability here of separate data for controlled versus uncontrolled 
emissions results in the availability of a good representation of the 
variability of emissions to be expected when the plan requires virtual 
elimination of uncontrolled emissions. Second, the similarity of 
percentages further supports Indiana's assertion that the controlled 
emissions from Unit 2 provide a better forecast of emissions 
variability for controlled emissions of all four units than would be 
obtained from the controlled emissions from Unit 1. Finally, this 
similarity supports a finding that the use of 2006 to 2010 data for the 
controlled emission stack for Unit 2 provides a good basis for 
estimating the degree of adjustment for determining 30-day average 
limits at IP&L-Petersburg that are comparably stringent to the 1-hour 
limits that would otherwise apply.
---------------------------------------------------------------------------

    \6\ The slightly lower adjustment factor suggests the 
possibility that the limits Indiana adopted correspond to (are 
comparably stringent to 1-hour limits at) slightly higher emission 
rates than Indiana modeled. However, Indiana's attainment 
demonstration (with a design value of 189.68 micrograms per cubic 
meter) provides a sufficient attainment margin so that these 
differences in adjustment factors would not alter the conclusion 
that Indiana's limits provide for attainment.
---------------------------------------------------------------------------

    As noted in Section II, Indiana's rule provides for computing 30-
day average emission rates as an arithmetic average of the hourly lbs/
MMBTU values during operating hours. Notwithstanding the potential for 
confusion regarding EPA's guidance on this point (as discussed above), 
this approach differs from the recommendation in EPA's guidance to 
compute 30-day average emission rates as the ratio between the 30-day 
total emissions divided by the 30-day total heat input.
    Therefore, EPA conducted additional evaluation, using the 2017 to 
mid-2019 data from the four units at IP&L-Petersburg, to compare the 
results of these two data handling approaches. This evaluation focused 
on 99th percentile values of the 30-day average emission rates 
calculated using these two approaches, in order to focus on periods 
when compliance is most challenging. Table 3 shows the results of this 
evaluation.

           Table 3--Effect of Data Handling Approach on 99th Percentile 30-Day Average Emission Rates
----------------------------------------------------------------------------------------------------------------
                                                             Arithmetic      Total emissions/
                          Unit                             average  (lbs/    total heat input      Ratio  (%)
                                                               MMBTU)          (lbs/MMBTU)
----------------------------------------------------------------------------------------------------------------
1......................................................              0.097              0.088                110
2......................................................              0.117              0.121                 97
3......................................................              0.214              0.219                 98
4......................................................              0.214              0.220                 97
Average................................................  .................  .................                100
----------------------------------------------------------------------------------------------------------------

    These results suggest several conclusions. First, the results of 
these approaches, at least at times of most concern (i.e., times with 
relatively high emissions), tend to be quite similar. Second, neither 
approach is necessarily more conservative than the other. Third, the 
variation in results across the four units lends some support to the 
view that the arithmetic average approach gives slightly less stable 
results, but the results are sufficiently similar that either approach 
is a suitable approach for evaluating compliance.
    While Indiana's submittal (in the State's letter to the company 
dated September 18, 2019) describes the commissioner's order as 
applying the compliance methodology ``recommended'' in MATS, the 
applicable compliance provisions (in 326 IAC 7-4-15(d)) provide for 
averaging ``all of the quality assured hourly average . . . data,'' 
which would include data collected during startup and shutdown of the 
units. Thus, Indiana's submittal does not raise questions as to whether 
it is permissible to exclude data during startup and shutdown in an 
attainment plan.
    As noted above, EPA guidance recommends calculating adjustment 
factors using data obtained according to the procedures used in 
determining compliance. Since compliance with IP&L's 30-day average 
limits is evaluated on the basis of an arithmetic average of operating 
hour emission rates, the appropriate adjustment factors here are 
calculated on that basis. For reasons discussed above, EPA believes 
that Indiana has adopted limits that reflect suitable adjustments, such 
that these limits are comparably stringent to the 1-hour limits that 
Indiana's modeling has demonstrated would provide for attainment.
    The August 2018 proposed rule observed that this facility, upon 
complying with its 30-day average limits, can be expected to have only 
a limited frequency and magnitude of hours with emissions exceeding the 
critical emission value. Since the changes in Indiana's plan for IP&L-
Petersburg retain the same critical emission value but establish lower 
30-day average emission limits, these changes can be expected to reduce 
the frequency and magnitude of occasions when emissions exceed the 
critical emission value.
    Nevertheless, more pertinent data are now available to address this 
question. EPA previously examined this question based on 2006 to 2010 
data from the main stack at Unit 2, but EPA now has data for 2017 to 
mid-2019 for all four

[[Page 10356]]

units, for a period when IP&L was required to meet limits similar to 
the final limits. For this period, Units 1, 3, and 4 are complying with 
the revised emission limits and are exceeding the critical emission 
values (i.e., the modeled mass emissions in lbs/hour) for 0.9 percent, 
0.1 percent, and 0.4 percent of the hours, respectively. Unit 2 is 
exceeding its revised limit 17 percent of the time, while exceeding the 
critical emission value 3 percent of the time.\7\ This suggests that 
the necessary improvements in scrubber efficiency at Unit 2 would 
likely yield a percentage of hours with emissions above the critical 
emission value that is similar to the percentages found for the three 
units that are already complying with limits.
---------------------------------------------------------------------------

    \7\ The exceedances of the new Unit 2 limit, while somewhat 
frequent, are modest in magnitude; during this 2\1/2\-year period, 
Unit 2 met the prior limit for all but one 30-day average period, 
and a majority among the 30-day periods with averages above 0.10 
lbs/MMBTU had average emission rates below 0.11 lbs/MMBTU.
---------------------------------------------------------------------------

    EPA proposed previously that Indiana's modeling provides an 
appropriate estimation of the critical emission values that will 
provide for attainment, and Indiana has made no changes that warrant 
EPA revisiting that finding. Instead, Indiana has changed only its 
calculation of an adjustment factor and, by applying the resulting 
revised adjustment factor, determined and adopted a revised set of 30-
day average limits that EPA now judges to be comparably stringent to 1-
hour limits at the critical emission values. Accordingly, in this 
proposed rule, EPA is soliciting comments on the revised adjustment 
factor calculations, the resulting revisions in Indiana's plan, and 
EPA's evaluation of these revisions. EPA is not soliciting additional 
comments on Indiana's plan and EPA's evaluation of that plan other than 
with respect to those elements of Indiana's plan and EPA's evaluation 
that have changed since EPA's prior proposed rulemaking.
    Indiana's September 18, 2019 submittal requests that EPA approve 
326 IAC 7-4-15, including the 1-hour limits for IP&L-Petersburg, except 
for the 30-day average limits for IP&L-Petersburg in that rule, and 
approve the commissioner's order, which includes substitute 30-day 
average limits. In seeking approval of both the rule and the 
commissioner's order, Indiana seeks to allow IP&L to switch between 1-
hour limits in 326 IAC 7-4-15(a) and the 30-day average limits in the 
commissioner's order. Indiana clarified that the 30-day average limits 
in the commissioner's order are to be viewed as substitutes for the 30-
day average limits in 326 IAC 7-4-15(c), and that references to the 
limits in subsection (c) in 326 IAC 7-4-15 should be understood as 
references to the limits in the order. See email from Mark Derf to John 
Summerhays dated November 19, 2019. Indiana further clarified that 326 
IAC 7-4-15(e) thus provides terms under which IP&L may choose to switch 
between being required to comply with the 30-day average limits in the 
commissioner's order and being required to comply with the 1-hour 
limits in 326 IAC 7-4-15(a). EPA is proposing action in accordance with 
this interpretation.

V. EPA's Proposed Action

    EPA is proposing to conclude that, based on revised adjustment 
factor calculations, the revised emission limits that Indiana has 
adopted for IP&L-Petersburg are a suitable element of an approvable 
plan for attaining the 2010 1-hour SO2 NAAQS for Southwest 
Indiana. This action is a supplement to a prior proposed rule, 
published August 15, 2018, at 83 FR 40487, which addressed the full 
range of requirements that the SO2 nonattainment plan for 
Southwest Indiana must meet.
    EPA is not soliciting additional comments on the other elements of 
Indiana's plan for Southwest Indiana, aside from any ramifications of 
Indiana's revised emission limits for IP&L-Petersburg. In response to 
comments received, Indiana has only revised its calculation of the 
degree of adjustment needed for 30-day average limits at IP&L-
Petersburg to be comparably stringent to the 1-hour limits that would 
otherwise be necessary, and has adopted the limits that this revised 
calculation indicated to be warranted. Accordingly, EPA is only 
soliciting comments on the revisions that Indiana made and EPA's 
evaluation of these revisions. EPA acknowledges receipt of other 
comments on Indiana's plan and EPA's August 2018 proposed action, 
including comments on the general acceptability of 30-day average 
limits. EPA plans to address those comments as part of final rulemaking 
on Indiana's plan for SO2 in Southwest Indiana.
    EPA's August 2018 proposed action specifies particular Indiana 
rules that EPA proposed to incorporate by reference into the Indiana 
SIP. Two of these rules (Title 326 Indiana Administrative Code Rules 7-
1.1-3 and 7-2-1 (326 IAC 7-1.1-3 and 7-2-1)) provide compliance 
deadlines, reporting requirements and compliance determination 
procedures not just for sources in Southwest Indiana but also for 
sources in the Indianapolis, Terre Haute, and Morgan County areas. EPA 
has already approved these rules as part of its action on the 
Indianapolis and Terre Haute area plans, as published on March 22, 2019 
at 84 FR 10692, and so no further action on these rules is necessary. 
EPA also proposed to approve limitations for Pike County, in 326 IAC 7-
4-15, which includes limitations for IP&L-Petersburg and for the Frank 
E. Ratts power plant. EPA continues to intend to approve most of this 
rule, specifically paragraphs a, b, d, and e, incorporating the limits 
for the Frank E. Ratts plant, the 1-hour limits for IP&L-Petersburg, 
and associated compliance provisions into the SIP. The only portion of 
326 IAC 7-4-15 that EPA is proposing not to take action on is paragraph 
c, the paragraph with the prior 30-day average limits for IP&L-
Petersburg; as requested by Indiana, EPA is instead proposing to 
approve the commissioner's order that Indiana submitted September 18, 
2019, which EPA considers to provide substitute 30-day average limits 
for the 30-day average limits in 326 IAC 7-4-15(c).

VI. Incorporation by Reference

    In this rule, EPA is proposing to include in a final EPA rule 
regulatory text that includes incorporation by reference. In accordance 
with requirements of 1 CFR 51.5, EPA is proposing to incorporate by 
reference Commissioner's Order Number 2019-2, effective August 18, 
2019, and 326 IAC 7-4-15 Pike County sulfur dioxide emission 
limitations (except for paragraph (c)), effective October 30, 2015. EPA 
has made, and will continue to make, these documents generally 
available through www.regulations.gov, and at the EPA Region 5 Office. 
(Please contact the person identified in the FOR FURTHER INFORMATION 
CONTACT section of this preamble for more information.)

VII. Statutory and Executive Order Reviews

    Under the Clean Air Act, the Administrator is required to approve a 
SIP submission that complies with the provisions of the Clean Air Act 
and applicable Federal regulations. 42 U.S.C. 7410(k); 40 CFR 52.02(a). 
Thus, in reviewing SIP submissions, EPA's role is to approve state 
choices, provided that they meet the criteria of the Clean Air Act. 
Accordingly, this proposed action merely approves state law as meeting 
Federal requirements and does not impose additional requirements beyond 
those imposed by state law. For that reason, this proposed action:
     Is not a ``significant regulatory action'' subject to 
review by the Office

[[Page 10357]]

of Management and Budget under Executive Order 12866 58 FR 51735, 
October 4, 1993) and 13563 (76 FR 3821, January 21, 2011);
     Does not impose an information collection burden under the 
provisions of the Paperwork Reduction Act (44 U.S.C. 3501 et seq.);
     Is certified as not having a significant economic impact 
on a substantial number of small entities under the Regulatory 
Flexibility Act (5 U.S.C. 601 et seq.);
     Does not contain any unfunded mandate or significantly or 
uniquely affect small governments, as described in the Unfunded 
Mandates Reform Act of 1995 (Pub. L. 104-4);
     Does not have federalism implications as specified in 
Executive Order 13132 (64 FR 43255, August 10, 1999);
     Is not an economically significant regulatory action based 
on health or safety risks subject to Executive Order 13045 (62 FR 
19885, April 23, 1997);
     Is not a significant regulatory action subject to 
Executive Order 13211 (66 FR 28355, May 22, 2001);
     Is not subject to requirements of Section 12(d) of the 
National Technology Transfer and Advancement Act of 1995 (15 U.S.C. 272 
note) because application of those requirements would be inconsistent 
with the Clean Air Act; and
     Does not provide EPA with the discretionary authority to 
address, as appropriate, disproportionate human health or environmental 
effects, using practicable and legally permissible methods, under 
Executive Order 12898 (59 FR 7629, February 16, 1994).
    In addition, the SIP is not approved to apply on any Indian 
reservation land or in any other area where EPA or an Indian tribe has 
demonstrated that a tribe has jurisdiction. In those areas of Indian 
country, the rule does not have tribal implications and will not impose 
substantial direct costs on tribal governments or preempt tribal law as 
specified by Executive Order 13175 (65 FR 67249, November 9, 2000).

List of Subjects in 40 CFR Part 52

    Environmental protection, Air pollution control, Incorporation by 
Reference, Intergovernmental relations, Reporting and recordkeeping 
requirements, Sulfur oxides.

    Dated: February 10, 2020.
Kurt A. Thiede,
Regional Administrator, Region 5.
[FR Doc. 2020-03507 Filed 2-21-20; 8:45 am]
 BILLING CODE 6560-50-P