SNPR Response to Comments Document

December 2011

Determination of Significant Contribution to Nonattainment of Ozone
NAAQS

General 

Commenter generally questions the validity of EPA’s conclusion
regarding the six States determined to significantly contribute to
nonattainment in downwind States and covered in this SNPR, especially
based on the Allegan, MI receptor.

Comment:  Why were States that did not significantly impact a
non-compliance monitor included in this rulemaking?  The goal of the
rulemaking is to ensure that all monitor locations are in attainment by
2014.  Including States that only showed a significant impact on a
maintenance monitor is excessive and does nothing to accomplish the
goal.  [Oklahoma Municipal Power Authority (OMPA)
(EPA-HQ-OAR-2009-0491-4573), p. 2]

RESPONSE:  The comment is outside the scope of the SNPR because it
addresses issues that were within the scope of the request for comments
in the proposed Transport Rule but not in the SNPR.  While EPA is not
reconsidering or reopening the issue, we note that the issue was
previously addressed in the final Transport Rule, section V.C.2, 76 FR
48227, August 8, 2011 and that the North Carolina v. EPA opinion
requires EPA to give independent meaning to the interfere with
maintenance prong of section 110(a)(2)(D)(i)(I).  Further, EPA notes
that the commenter incorrectly describes the objective of the Transport
Rule.

 

Comment:  PSO and AEP respectfully submit that the data included in the
proposal does not support EPA's conclusion that emissions from EGU
sources in Oklahoma significantly contribute to the projected
maintenance issues at the Allegan County, Michigan monitor, or that
further reductions in NOX emissions from these sources will address the
modeled maintenance issue.  The monitor in question has already attained
the 1997 ozone standard, and EPA's modeling erroneously predicts a
maintenance issue at this location.  Moreover, even if the modeled
values were accurate, they are the result of a well understood
phenomenon resulting from rush-hour mobile source emissions in the
Chicago area and no additional reductions in Oklahoma will aid in
maintaining the standard. [Public Service Company of Oklahoma
(EPA-HQ-OAR-2009-0491-4600), p. 2-3.]

RESPONSE:  EPA’s methodology for identifying nonattainment and
maintenance receptors is based on using modeling to project measured air
quality for specific monitors, not on the designation status of an area
as explained in section III.D of the preamble to the final supplemental.
 EPA believes this approach is appropriate for reasons explained in
section V.C.2 of the preamble to the final Transport Rule.  76 FR 48230,
August 8, 2011.  

The commenter claims that ozone at the Allegan County monitor is the
result of rush-hour mobile source emissions in Chicago and no additional
reductions in Oklahoma will aid in maintaining the standard.  However,
the commenter provides not analyses or references to substantiate this
claim.  EPA’s CAMx source apportionment modeling shows that 10 states
(including Oklahoma) contribute above the 1% of the ozone NAAQS
threshold to Allegan County.  EPA’s modeling does show that the
largest ozone contribution is from Illinois.  But that does not change
the fact that 9 other states interfere with maintenance in Allegan
County and it does not alleviate the need for those states, including
Oklahoma, to eliminate their interference with maintenance.  Further,
the North Carolina v. EPA opinion requires EPA to give independent
meaning to the interfere with maintenance prong of section
110(a)(2)(D)(i)(I).

Comment:  EPA's decision to use the CSAPR program as the basis for
further NOX reductions in Oklahoma is not sound for all of the reasons
presented in previous comments submitted on the proposed Transport Rule
by AEP, the Utility Air Regulatory Group (UARG).  The Edison Electric
Institute (EEI), the Midwest Ozone Group (MOG) and ignores the
substantial additional reductions already required to be made under the
Regional Haze SIP developed by Oklahoma, and proposed to be approved by
EP A in a notice earlier this year.  The reductions required under the
Regional Haze SIP will more than adequately address any contribution of
NOX emissions by Oklahoma sources to the Allegan County, MI monitor. 
[Public Service Company of Oklahoma (EPA-HQ-OAR-2009-0491-4600), p.
2-3.]

RESPONSE:  EPA’s methodology for identifying nonattainment and
maintenance receptors is based on using modeling to project measured air
quality for specific monitors, not on the designation status of an area
as explained in section III.D of the preamble to the final supplemental.
 EPA believes this approach is appropriate for reasons explained in
section V.C.2 of the preamble to the final Transport Rule, 76 FR 48230,
August 8, 2011.  Additional responses on the MOG modeling data are also
contained in section V.C.2.  Responses to CAIR-related baseline comments
are in section V.B. of the preamble to the final Transport Rule, 76 FR
48223, August 8, 2011.  EPA disagrees with the comment that the Regional
Haze SIP will more than adequately address any contribution of NOX
emissions by Oklahoma sources to the Allegan County, Michigan monitor. 
EPA agrees that the regional haze SIP requirements will require
reductions in NOX emissions from several EGUs in Oklahoma, and EPA
expects to issue a final approval of those regional haze NOX
requirements by December 31, 2011.  Those required reductions, however,
have compliance dates five years after EPA’s SIP approval — that is,
late 2016.  This time period is much later than 2012.  Accordingly,
those requirements would not have affected EPA’s 2012 base case
inventory for Oklahoma, nor would they have affected EPA’s
determination of whether Oklahoma’s contribution was sufficient for
inclusion under the Transport Rule.    

Comment:  EPA's application of its CSAPR modeling methodology to the
state of Oklahoma requires the entire state (along with 3 other states
in this supplemental notice of proposed rulemaking (SNPR) (Iowa,
Wisconsin and Kansas)) to significantly reduce, at significant total
cost, EGU emissions of NOX during the ozone season to address a modeled
maintenance issue.  The area in question is a single county (Allegan
County) in Michigan.  The county is located in western Michigan in an
area directly East across Lake Michigan from the Chicago/Milwaukee metro
areas.  It is important to note that the monitors in the area show the
county to be in attainment.  The modeling data developed by the Midwest
Ozone Group (MOG) and its Industrial Modeling Coalition, used in
conjunction with ambient air quality data collected by USEPA, show that
not only are transport criteria met by the existing CAIR program, the
requirements of which are still in effect and under which significant
emissions reductions have been achieved and continue, but full
compliance with the NAAQS targeted by this rulemaking are satisfied for
this county. [Public Service Company of Oklahoma
(EPA-HQ-OAR-2009-0491-4600), p. 2-3.]

 

RESPONSE:  EPA’s methodology for identifying nonattainment and
maintenance receptors is based on using modeling to project measured air
quality for specific monitors, not on the designation status of an area
as explained in section III.D of the preamble to the final supplemental.
 EPA believes this approach is appropriate for reasons explained in
section V.C.2 of the preamble to the final Transport Rule, 76 FR 48230,
August 8, 2011.  Additional responses on the MOG modeling data are
contained in section V.C.2 of the preamble to the final Transport Rule,
76 FR 48230, August 8, 2011.  The comment suggesting that the transport
criteria are met by the existing CAIR program is outside the scope of
issues we requested comment on in this SNPR, ignores the fact that this
rule is to replace, not supplement the CAIR, and was responded to in the
record for the final Transport Rule.  Responses to CAIR-related baseline
comments are in section V.B. of the preamble to the final Transport
Rule, 76 FR 48223, August 8, 2011.  Responses to comments on current air
quality in Allegan County are in section III.D of the final supplemental
rule preamble.

Comment:  The reductions required in the six states are significant. 
The proposed 2012 Oklahoma budget 21,835 tons is 56% lower than the
total of the highest emissions for each unit during 2006 through 2010,
and 42% lower than the total 2010 emissions for the applicable units in
the state.  For PSO plants operating in Oklahoma, the allocations would
require these facilities to reduce emissions by 65% from their maximum
annual during 2006 through 2010, and 52% below their 2010 emissions.  Of
these units, only two are coal combustion units with the remaining being
natural gas.  PSO and AEP have developed plans and a schedule to install
additional NOX controls on many of these units.  However, those controls
are scheduled to be phased in over several years.  Based on our
experience of installing controls for NOX emissions over the past
several years, there would not be enough time to complete installation
of all of these controls by the 2012 ozone season, even if expedited
delivery and installation were feasible. [Public Service Company of
Oklahoma (EPA-HQ-OAR-2009-0491-4600), p. 2-3]

RESPONSE:  EPA made a number of adjustments between proposal and final
that affect the size of the Oklahoma budget.  See the preamble to this
final rule, section III.B.iv, for an explanation of how the final budget
was calculated.  In calculating the 2012 budget, EPA did not assume NOX
combustion control installations or upgrades in 2012.  Sources will be
able to phase in those controls as the above comment suggests.

Comment:  PSO does not believe that the timelines and stringent budgets
within this Proposed Supplemental Transport Rule are justified by the
minimal maintenance impact predicted by the model and the total costs of
imposed reductions, especially considering the number of states/units
affected only because of this one receptor. [Public Service Company of
Oklahoma (EPA-HQ-OAR-2009-0491-4600), p. 2-3].  Given that the model
predicts that this single receptor still attains the ozone air quality
standard but is marginally approaching levels that suggest additional
reductions will assure maintenance of the standard, and that the
integrity of the modeling results is uncertain as discussed below, it is
excessive to require such significant reductions at such a significant
cost. The cost to achieve this projected 0.1 ppb reduction is not
justified. In addition, the short compliance timeframe imposes greater
costs to expedite construction schedules than accounted for in EPA's
estimate. Oklahoma should not be included in the CSAPR program. Public
Service Company of Oklahoma (EPA-HQ-OAR-2009-0491-4600), p. 2-3]

RESPONSE:  See preamble section III.D of this final rule for discussion
of EPA’s justification for inclusion of the Allegan site as a
receptor, including discussion of how recent air quality monitoring
reinforces EPA’s position.  As noted in this preamble discussion, a
very large proportion of ozone at the Allegan receptor is caused by
upwind state contributions.  Accordingly, it is especially important for
such a location to ensure that upwind states are meeting the
requirements of section 110(a)(2)(D)(i).

EPA finds that Oklahoma clearly contributes to this ozone problem
receptor.  Oklahoma contributes more than 2 ppb, which is more than
twice the threshold.  

EPA finds the costs imposed by the rule to be modest in comparison to
the types of controls routinely in place elsewhere.  For example, for
Oklahoma EPA predicts, at most, additional low NOX burner (LNB)
installations on certain facilities.  Such technologies, and more
stringent controls such as selective catalytic reduction (SCR), have
routinely been in place for states in the areas of the eastern United
States which contribute to downwind state ozone nonattainment.  One
illustration of the degree of SCR installation may be seen by comparing
2008 and 2009 annual NOX emissions.  Because CAIR essentially required
SCR installations to operate year round, which were not required to
operate year-round in 2008, nearly one million tons of NOX was reduced
between 2008 and 2009.  EPA finds the costs of LNB controls that may
result from the rule’s inclusion of Oklahoma to be very reasonable.

EPA has considered the commenter concerns with respect to the compliance
schedule.   In the final rule, EPA is setting the Oklahoma 2012 ozone
season NOX budget at a level that reflects emission reductions
achievable through actions (such as some changes in generation unit
dispatch) that do not include additional LNB installations or
significant changes in dispatch of certain oil/gas units.  EPA is
setting the Oklahoma ozone season NOX budget for 2013 and beyond at the
level that was proposed in the SNPR, i.e., to reflect NOX levels
achievable with additional LNB installations that can be completed
before the 2013 ozone season without necessitating the shutdown of units
during the summer peak demand period in 2012.   

Comment:  USEPA has failed to completely examine its own record relating
to the Allegan County ozone monitor (260030005).  In examining the
emission inventory from the proposed rule to the new modeling supporting
the final rule, the emission inventories offer several alternative
perspectives on why this monitor has suddenly changed from a monitor
that was not of interest to a monitor that is now classified as
Maintenance.  These alternatives clearly show that this monitor should
not be used to include additional states in the CSAPR program. [Public
Service Company of Oklahoma (EPA-HQ-OAR-2009-0491-4600), p. 4-6.]

RESPONSE:  EPA disagrees with the comment.  EPA has carefully examined
the information regarding the Allegan County ozone monitor and explained
the relevance of this information in   Section III.D of the preamble to
the final supplemental rulemaking.

Comment:  First, the states that are shown to contribute to the monitor
based on the 2012 source apportionment modeling should be considered.
These states can be grouped roughly as follows: 

Lake Michigan States Michigan, Indiana, Illinois, Wisconsin

Intermediate Range States -Iowa, Kansas, Missouri

Long Range States -Arkansas, Oklahoma, Texas

When the NOX inventories in the Lake Michigan and Intermediate Range
States are examined based on the information in the Technical Support
Documents for both the Proposed and Final Modeling, the total NOX
inventories for these states have changed very little (2,683,513 Tons
vs. 2,684,457 Tons).  The state level changes are shown in a table (see
Table 1 of the comment letter) for these states.

This state level analysis shows that while significant decreases in the
EGU portion of the inventory occur in all of the Lake Michigan and
intermediate range states, the total NOX inventory decreases by less, or
increases.  The contributions of other source categories from these
nearer states therefore play a far more significant role in any
projected maintenance for the Allegan County monitor. 

Past studies that USEPA modeling staff is aware of (as at least two of
them were participants in various of these studies), determined that two
of the critical states for driving ozone levels in western Michigan are
Illinois and Wisconsin.  For those two states, the total NO" inventory
increases while the EGU contribution decreases by some 34,728 tons
between the cases. This clearly shows that other source categories are
growing the inventory by some 97,699 tons.  Many of these other source
categories also generate organic emissions concurrently with the NOX
emissions, which then reacts in the atmosphere over Lake Michigan to
become ozone.  This ozone then blows on shore in Western Michigan giving
the elevated readings observed. 

Taking this one step further, if the four Lake Michigan States are
considered, the numbers become an increase in total NOX of 16,316, but a
decrease in EGU NOX of 150,850 tons, giving a total increase in NO" for
all non-EGU sources of 167,166 tons.  The inventory changes in just
these four key states impacting the Allegan County monitoring site
suggest that the VOC/NOX ratio in the Lake Michigan area has been
significantly shifted toward VOC's between the Proposal and Final
Modeling.  Since these changes have occurred in areas known to
contribute to elevated ozone levels in Western Michigan, this strongly
suggests that local, non-EGU sources are the cause of the increase
modeled at the Allegan County monitor.  With the reduced emission
inventories of the EGU Sector in relation to the total NOX inventory,
addressing the maintenance issues at the Allegan County monitor is being
wrongly placed on the shoulders of EGUs and USEPA must perform more
detailed source apportionment modeling that is able to identify
responsible source categories as well as states.  In addition, this
modeling must be extended to the 2014 Remedy Case, at a minimum, to
properly demonstrate that the remedy USEPA is attempting to impose will
actually achieve the predicted reductions. [Public Service Company of
Oklahoma (EPA-HQ-OAR-2009-0491-4600), p. 4-6.]

RESPONSE:  The comment is outside the scope of the SNPR to the extent it
is suggesting EPA should make changes to the methodology used to
quantify a state’s significant contribution.  EPA addressed numerous
comments regarding this methodology in the final Transport Rule and did
not reopen this methodology for comment in the SNPR. .  

EPA agrees that there were numerous emissions updates, including updates
to both the EGU and non-EGU components of the inventory between the
proposal and the final rule.  Some components of the emissions inventory
increased and some decreased.  In some cases EGU emissions went down
between proposal and final.  EPA agrees that, in general, mobile source
emissions went up, due to the availability of updates to the mobile
source emissions model (MOVES).  EPA also agrees that changes to NOX and
VOC emissions in the inventories used for air quality modeling could
change the VOC/NOX ratio and thereby change projected ozone
concentrations and contributions.  In this regard, EPA reran CAMx for
the final rule using the updated emissions inventories in order to
explicitly account for any chemistry affects from the net changes in
emissions.  As a result of the updated modeling, the Allegan receptor
went from a level that was slightly below the NAAQS in the proposal
modeling to a level that was slightly above the NAAQS (for maintenance)
in the final rule modeling.  Therefore, the Allegan county monitor
became a maintenance receptor for the final rule analysis.  This
receptor was treated in the same way as every other potential
nonattainment and/or maintenance receptor.  EPA believes that the
Allegan county receptor is a valid maintenance receptor and that it is
not identified as a receptor due to any errors in the modeling or the
calculation of future design values (see also responses to comments on
current air quality in Allegan County are in section III.D. of the final
supplemental rule preamble).

 

The commenter says that EPA’s approach to addressing the maintenance
issues at the Allegan County monitor is being wrongly placed on the
shoulders of EGUs and EPA must perform more detailed source
apportionment modeling that is able to identify responsible source
categories as well as states.  The comment questions EPA’s approach
that considers the collective contribution from all anthropogenic source
categories of NOX (i.e., including both EGU and non-EGU emissions) from
each upwind state along with cost in determining the amount of
significant contribution.  The commenter suggests an approach that is
based on air quality contributions from individual sectors.  This
comment represents a broad challenge to EPA’s approach of considering
cost and collective contribution.   The identification of EGU emission
reductions by EPA in the remedy analysis was based on a determination
that, at the cost thresholds selected under the final Transport Rule,
the power sector offers the most cost-effective emission reductions to
reduce a state’s significant contribution to nonattainment or
interference with maintenance.  Information on EPA’s decision to
include cost as part of the analysis is provided in the final Transport
Rule preamble.

Comment:  USEPA redesignated Allegan County to attainment in 2009.  In
2010, the monitor had a three year average value of 74 ppb and for the
period including through 7/21/2011 a three year average of 77 ppb.  As
discussed in previous comments on the Transport Rule, the Midwest Ozone
Group (MOG) supplied new modeling based on the use of a 2008 base year. 
That modeling does not indicate that the Allegan County monitor would be
out of attainment with the 84 ppb standard in any of the cases.  The MOG
modeling shows this monitor at roughly 81 ppb in the base year, 76 ppb
in 2014, and 73 ppb in 2018 with CAIR level reductions (which did not
include Oklahoma) and other rules in place.  The actual monitor
three-year average values are much more consistent with the 2008 values
modeled by MOG and much less consistent with the values being generated
by USEPA using the 2005 base year.  We therefore continue to believe
that USEPA has incorrectly used 2005 for the base year of its analysis
and should re-perform its modeling using a base year of 2008.  [Public
Service Company of Oklahoma (EPA-HQ-OAR-2009-0491-4600), p. 6.]

RESPONSE:   The comment is outside the scope of the SNPR because issues
concerning the base case were within the scope of the request for
comments in the proposed Transport Rule but not in the SNPR.  While EPA
is not reconsidering or reopening the issue, we note that the issue was
addressed in the final Transport Rule, section V.B, 76 FR 48223-48224. 
Additional information is available in section III.D of the preamble to
the final supplemental rulemaking. 

Comment:   In 2008 the U.S. Court of Appeals for the D.C. Circuit found
the Clean Air Interstate Rule (CAIR) to be “fundamentally flawed,”
initially vacating and remanding it to EPA See North Carolina v. EPA,
531 F.3d 896 (D.C. Cir. 2008).  However some months later in December,
the D.C. Circuit Court of Appeals remanded the CAIR without vacatur,
North Carolina v. EPA, 550 F.3d 1176 (D.C. Cir. 2008), instead, leaving
CAIR in place until EPA cured the flaws in CAIR.  Therefore CAIR is a
viable rule and utilities must currently comply with the stringent
nitrogen oxide (NOX) and sulfur dioxide (SO2) limits prescribed by the
rule.

During the last half of the previous decade utilities have installed and
continue to install equipment to control NOX to meet the requirements of
CAIR.  In Ameren’s case we have installed add-on controls including
Selective Catalytic Reduction (SCR) and combustion controls including
low NOX burners and over-fire air systems on Ameren's fleet of coal
plants.  Ameren has also been a pioneer in enhancing the use of
combustion controls in addition to low NOX burners, such as over fire
air and combustion optimization systems to achieve cost effective
reduction in NOX emissions.  Installation of these controls began in the
1990s to enable Ameren to comply with the Acid Rain Program, state
implementation plans and the NOX Budget Trading Programs and continued
into the subsequent years to achieve regional reductions for the CAIR
program.

EPA has assumed as part of its regulatory analysis that the CAIR program
is not in place and that affected sources are not complying with the
rules.  This assumption is in error. By not including compliance with
CAIR in its assumptions, the EPA analyses have over stated the impact of
these sources prior to the imposition of the proposed inclusion of these
six states into the ozone season control period of the CSAPR.  This
assumption completely invalidates EPA’s analyses and the results
produced.  EPA should redo the analyses assuming that CAIR is fully in
place and the affected sources are complying with the regulations.  EPA
should view CAIR as the valid control program that it is and allow
sources to smoothly transition to the future rules to be promulgated in
a more reasonable time frame, just as was done between the NOX SIP Call
trading program and the CAIR.  [Ameren Corporation
(EPA-HQ-OAR-2009-0491-4596), p. 2.]

RESPONSE:  The comment is outside the scope of the SNPR because it
addresses issues that were within the scope of the request for comments
in the proposed Transport Rule but not in the SNPR.  The comment also
ignores the fact that the Transport Rule replaces, and does not simply
supplement, the CAIR.  While EPA is not reconsidering or reopening the
issue, we note that the issue was previously addressed in the final
Transport Rule, sections V.B and V.C, 76 FR 48223-48236, August 8, 2011.
    

Commenter supports generally EPA’s supplementary proposal.

Comment:  The Sierra Club commends the EPA for recognizing the important
role of interstate air pollution in the non-attainment and interference
with maintenance of National Ambient Air Quality Standards
(“NAAQS”), and its efforts to address the issue in the proposed
rule. The Sierra Club is the oldest and largest grassroots environmental
group, with over 1.3 million members. Sierra Club members live, work,
attend school, travel and recreate in areas adversely by power plant
emissions. Our members enjoy and are entitled to the benefits of natural
resources that are adversely affected by interstate air pollution,
including air, water and soil; forests and cropland; parks, wilderness
areas and other green space; and flora and fauna. Our membership and
their families include sensitive populations such as asthmatics, the
elderly and children who are at elevated risk for deleterious health
effects posed by power plant emissions.

Sierra Club supports EPA’s proposal to include six states in the
cross-state rule’s ozone season NOX program: Iowa, Kansas, Michigan,
Missouri, Oklahoma and Wisconsin. Using the methodologies and modeling
previously available for public comment, EPA identified two additional
counties that are having trouble maintaining the 1997 ozone standards:
Allegan County, MI and Harford County, MD.  EPA found that Iowa, Kansas,
Missouri, Oklahoma and Wisconsin significantly contribute to Allegan
County’s maintenance problem and Michigan significantly contributes to
the maintenance problem in Harford County.  EPA’s proposal includes
NOX emissions budgets for all the covered units in the six states.  The
NOX emissions at issue, and the ozone it forms, causes significant harm
to public health and the environment.

As EPA explains in the proposed rule, there has been ample time and
stakeholder process for public comment on all aspects of modeling and
methodology.  EPA made all the relevant background data available for
public comment previously in connection with the proposed transport
rule.  Reopening public comment on methodology and modeling would serve
only to delay emissions reductions that are long overdue.  The six
affected states had incentive to review that proposal closely because it
included all six states in at least one of the rule’s trading
programs.

All states that are included in the proposed rule significantly
contribute to ozone maintenance in other states. Although Sierra Club
continues to maintain that EPA’s threshold should be lower, EPA used 1
percent of the NAAQS as the threshold for determining which states to
include in this rule.  For the 1997 8-hour ozone NAAQS of 0.08 ppm, the
contribution threshold is 0.0008 ppm (0.8 ppb).  All states that
contribute .8 ppb or above to a nonattainment or maintenance receptor
are included in the rule for ozone season. Most of the six states
contribute much more.  For Allegan County, MI, Oklahoma contributes over
three times that threshold at 2.8 ppb. Missouri contributes even more at
4.8, Wisconsin 2.2 ppb, Kansas 1 ppb and Iowa .9 ppb.  The numbers show
why it is vital that every contributing state do its fair share of
reductions. If only the few largest state contributors were responsible
for reductions, the air quality standards could not be achieved.  The
sum of all the many small contributors forms a substantial portion of
the total. For Harford County, only 42% of the total contribution is
captured if a 5% threshold is used (4 ppb). Using a 1% threshold
captures 81% of the total contribution to exceedances.  For Allegan
County, only 72% of the total is captured with a 5% threshold, while 92%
is captured with a 1% threshold.  [Sierra Club
(EPA-HQ-OAR-2009-0491-4587), p. 1-2.]

RESPONSE:  EPA concurs with the comments that  issues concerning
modeling and methodology should not be reopened, because adequate
opportunity for public comment was already provided on these matters,
and the comments supporting the inclusion of  Iowa, Michigan, Missouri,
Oklahoma and Wisconsin in the Transport Rule ozone season NOX program. 
As explained in the preamble to this final rule, EPA is not finalizing a
FIP for Kansas at this time for the reasons provided in section III.C.

Comment:   In brief, we support EPA’s conclusion that these six states
significantly contribute to nonattainment and interference with
maintenance of the 1997 ozone NAAQS, as well as EPA’s application of
the CSAPR remedial program to power plants in those states.  Given
EPA’s analysis and methodology employed in CSAPR, and the emission
inventories for those and other states, we do not think EPA could come
to a different conclusion, either as a policy or legal matter.

 

Specifically, we find no error in EPA’s application of CSAPR
methodologies with respect to the significant contribution of Iowa,
Kansas, Michigan, Missouri, Oklahoma, or Wisconsin to nonattainment and
interference with maintenance of the 1997 ozone NAAQS in downwind
states.  Furthermore, we support the Agency’s decision not to reopen
issues related to its overall methodology employed in CSAPR.  Clearly,
EPA’s decision is final with respect to the contribution of 20 other
eastern states to ozone nonattainment and maintenance problems in
downwind states, and to the establishment of emissions budgets and
related FIP provisions for power plants in those states.  As a matter of
law and policy, EPA is required to apply the same methodologies to the
six states addressed by the SNPR.  Thus, we support EPA’s ozone season
budgets and allocation of allowances to covered power plants in those
states.  [Clean Air Task Force, et al. (EPA-HQ-OAR-2009-0491-4588), p.
2.]

RESPONSE:   EPA concurs with the comments supporting the inclusion of 
Iowa, Michigan, Missouri, Oklahoma and Wisconsin in the Transport Rule
ozone season NOX program and the comments that  issues concerning
modeling and methodology should not be reopened, because adequate
opportunity for public comment was already provided on these matters. 
As explained in the preamble to this final rule, EPA is not finalizing a
FIP for Kansas at this time for the reasons provided in section III.C.

Iowa

 Commenter supports EPA’s decision to include Iowa in the ozone season
NOX program.  

Comment:  Based on the ozone levels in Iowa this year, as monitored by
the statewide SLAMS network and reported by the State Hygienic Lab and
in light of EPA’s upcoming change in the ozone NAAQS, Iowa’s
inclusion in the supplement is entirely warranted.  [Iowa Environmental
Council (EPA-HQ-OAR-2009-0491-4572), p. 1.]

RESPONSE:  EPA concurs with the comment supporting the inclusion of Iowa
in the Transport Rule ozone season NOX program.  Iowa is included in the
Transport Rule ozone season NOX program because of the contributions
from NOX emissions in Iowa to the downwind maintenance receptor in
Allegan County, MI.  Reductions in ozone season NOX emissions in Iowa
from implementation of the Transport Rule will help reduce ozone
concentrations within Iowa in addition to reducing ozone in downwind
states.

 EPA’s approach improperly includes Iowa.

Comment:  The final Transport Rule elaborates extensively on EPA’s
justification for not utilizing an updated emissions baseline to reflect
reductions that have been made under the provisions of CAIR, attempting
to draw a distinction regarding the enforceability or perceived lack
thereof based on CAIR’s legal status because “unlike most regulatory
requirements, the emission limitations contained in CAIR are only
temporary.”  EPA’s distinction is disingenuous.  Certainly, EPA
would not have hesitated to utilize its enforcement authority against a
facility that did not hold a sufficient number of emission allowances
under CAIR.  Controls were installed and are operated and emission
limits imposed under state-issued permits that are federally
enforceable, yet, EPA has not made it clear what, if any, controls are
reflected in its modeling.  Further, EPA takes the position that any
costs associated with emission reductions already implemented are
considered “sunk costs” and are not included in projected costs to
comply with the Transport Rule’s requirements.  Neither the emission
reductions achieved to date nor the costs with achieving those emissions
reductions are invisible.  [MidAmerican Energy Company
(EPA-HQ-OAR-2009-0491-4597), p. 3.]

RESPONSE:  The comment is outside the scope of the SNPR because issues
concerning the legal status of CAIR were within the scope of the request
for comments in the proposed Transport Rule but not in the SNPR.  While
EPA is not reconsidering or reopening the issue, we note that the issue
was previously addressed in the final Transport Rule, sections V.B and
V.C, 76 FR 48223-48236, August 8, 2011.

Comment:  Emissions of NOX in Iowa have declined substantially by virtue
of the requirements imposed under CAIR; those emission reductions have a
cost associated with them.  As EPA recognized in the final Transport
Rule, “ultimately we believe the electric power industry will pass
along most of the costs of the rule to consumers, so that the costs of
the rule will largely fall on the consumers of electricity.”  It is
critical that in making its determination to impose another layer of
regulation on sources (and, most importantly, electricity consumers) in
Iowa for allegedly making a less than 1 part per billion impact on an
ozone monitor in Michigan that is in attainment of the ozone standard,
EPA recognize both the emissions reduction efforts and cost impacts
imposed on the citizens of Iowa.  [MidAmerican Energy Company
(EPA-HQ-OAR-2009-0491-4597), p. 3.]

RESPONSE:    EPA used the best practices available to estimate the
future costs and benefits of this action and found that the emission
reductions implemented under this rulemaking, combined with the
reductions of the final Cross-State Rule, provide annual net benefits
starting in 2014 valued between $120 to $280 billion (2007$, 3% discount
rate).  The quantified PM2.5 and ozone health benefits within Iowa alone
are valued between $790 million and $1.9 billion (2007$, 3% discount
rate).  To put the benefits in Iowa in context, EPA estimates that the
total social costs of the entire rule, across all states, is $810
million (2007$, 3% discount rate).  EPA also chose to finalize
interstate trading programs as the remedy for the emission reductions to
provide for the most cost-effective implementation of any emission
controls.  See section VII in the preamble to the final Transport Rule,
76 FR 48271, August 8, 2011.

Comment:  The Allegan County, Michigan ozone monitor has been in
operation, collecting data, since 1992.  On May 12, 2010, prior to the
issuance of the proposed Transport Rule on July 6, 2010, the Michigan
Department of Natural Resources & Environment (“DNRE”) submitted a
request to EPA to redesignate Allegan County to attainment of the 0.08
parts per million 8-hour ozone National Ambient Air Quality Standard, to
change the legal status of the areas from nonattainment to attainment,
and to approve the maintenance plan as a revision to the Michigan State
Implementation Plan.  The DNRE’s request included a maintenance plan
consistent with Clean Air Act Section 175A which demonstrated an
attainment year inventory for 2008, an interim year inventory for 2018,
and a projected maintenance inventory for 2021 under which the 2021
inventories for VOC and NOX remained below attainment year 2008
emissions levels.  The DNRE noted that “continuing reductions in ozone
precursor emissions will be realized from fleet turnover, Maximum
Achievable Control Technology (“MACT”) standards for hazardous air
pollutants and federal diesel emissions programs.”  In addition to
Michigan’s regulations requiring less volatile consumer and commercial
products, the DNRE noted that:

“Due to the legal uncertainty of the Clean Air Interstate Rule (CAIR)
future year NOX reductions, a conservative 2018 and 2021 inventory was
prepared without the NOX reductions assumed from CAIR for the two
subject sources in this county.”(emphasis supplied) 

EPA approved Michigan’s request to redesignate the Allegan County,
Michigan nonattainment area to attainment for the 1997 8-hour ozone
standard on September 24, 2010 (See 75 Fed. Register 58315), prior to
issuance of the final Transport Rule.  EPA’s redesignation and
approval of Michigan’s maintenance plan, assuming no underlying
reductions from sources in Allegan County, or from upwind states under
the Transport Rule is contrary to EPA’s assertion that NOX emissions
from sources in Iowa interfere with maintenance of the ozone standard in
Allegan County, Michigan.  MEC believes EPA has incorrectly included
Iowa in the Transport Rule’s ozone season NOX reduction requirements.

 

In addition, EPA should reject arguments that Iowa and other states
should be included in the NOX ozone season reduction requirements
because lower ozone standards may be promulgated in the future.  First,
0.08 parts per million is the existing 8-hour ozone National Ambient Air
Quality Standard, and that is the standard which must be met today. 
Second, if the standard is ultimately lowered, there is no certainty
that Allegan County, Michigan wouldn’t still be in compliance, as
ozone levels in the area continue to trend downward.  [MidAmerican
Energy Company (EPA-HQ-OAR-2009-0491-4597), p. 3-4.]

RESPONSE:  See section III.D of the preamble to the final supplemental;
and the preamble (76 FR 48230) and Transport Rule Primary Response to
Comments document in the docket to this rulemaking
[EPA-HQ-OAR-2009-0491-4513].  This issue was discussed in the preamble
to the NOX SIP Call (see 63 FR 57375, October 27, 1998, footnote 25),
and the U.S. Court of Appeals for the D.C. Circuit’s decision in
Michigan v. EPA, 213 F.3d 663 (2000), further supports the position that
determinations of significant contribution or interference with
maintenance under CAA section 110(a)(2)(D)(i)(I) should not be based on
an area’s attainment designation.  Also, under the Clean Air Act
statutory structure, section 110(a)(2)(D)(i)(I) SIPs are due 3 years
from promulgation of a NAAQS, and the statute allows designations to be
done, in some cases, as late as 3 years after promulgation of a new
NAAQS.  This statutory structure suggests that designations are not
relevant to the analysis of what is required to satisfy section
110(a)(2)(D)(i)(I).  Finally, this rule addresses significant
contribution with respect to the 1997 ozone standard only, and the
requirements of the rule are necessary to address emissions that
significantly contribute or interfere with maintenance of that standard.
 The commenter, however, misstates the level of the  existing 8-hour
ozone National Ambient Air Quality Standard, which is 0.075 ppm. 

Kansas

Commenter(s) dispute EPA’s findings that Kansas significantly
interferes with the Allegan, MI receptor.

Comment:  Commenters first note the inadequate opportunity to comment on
this issue, and assert that EPA is committing reversible error in not
providing adequate opportunity to comment.  No prior notice was given
that Kansas interfered with a receptor in Allegan County, MI;
previously, Kansas was linked for interference with maintenance only
with a receptor in Dallas, TX. See Proposed Air Transport Rule, 76 FR
45210, 45269 (Aug. 2, 2010). No explanation was given in the SNPR for
the change or for how the model shifted from a receptor south of Kansas
to one north of Kansas in the interim.  As Allegan MI was newly
identified in the SPNR, commenter, who is with other Kansas utilities
engaging Trinity Consultants, Inc. to review the EPA data and modeling,
had no prior notice that this would be the area of concern, nor had
Trinity had a prior opportunity to consider "whether there are errors in
the Agency's application of the Transport Rule methodologies with
respect to Kansas's significant contribution to nonattainment and
interference of the 1997 ozone NAAQS" in Allegan County, MI. 76 FR at
40665/2. [Westar Energy (EPA-HQ-OAR-2009-0491-4583), p. 1-3; Kansas City
Board of Public Utilities (EPA-HQ-OAR-2009-0491-4585), p. 1-3.]

RESPONSE:  The comment is outside the scope of the SNPR to the extent it
seeks to challenge the methodology selected for quantifying significant
contribution and interference with maintenance.  EPA provided ample
opportunity to comment on this methodology in response to the  proposed
Transport Rule.  While EPA is not reconsidering or reopening the issue
in the SNPR, we note that our methodologies were documented in the
preambles for the proposed rule and final rule and the proposal and
final rule Air Quality Modeling TSDs.  There was ample opportunity to
comment on these methods based on information provided in the proposed
Transport Rule, three subsequent NODAs, and the additional opportunity
to comment on the conclusion in the proposed SNPR.  Further, EPA
disagrees that the opportunity to comment on the linkage between Kansas
and Allegan County was inadequate.  The purpose of the SNPR was to
provide notice of, and an opportunity to comment on, the linkage to
Allegan (and Harford County, MD) that was identified in finalizing the
Transport Rule.  See the preamble, section III.C., of the final
supplemental rule for the status of Kansas.   

Comment:  In an August 19, 2011 email denying requests for extension of
time to file comments in this matter, EPA stated: "The supplemental
proposal and all of the key supporting documentation and model results
were posted on our website on July 7, 2011 and the SNPR was formally
published in the Federal Register on July 20, 2011." This statement is
not true. Trinity determined after initially reviewing the information
posted on EPA's website that only emissions data were available, and not
the actual CAMx model input and output files. Those files contain key
supporting documentation for the proposed results. After contacting EPA
Staff to obtain the files, Trinity sent on August 2 via overnight mail
four 2-TB hard drives for EPA to upload the previously-unavailable
files. This process was finally completed on August 11, when Trinity
received the uploaded files. August 11 thus is the date on which key
supporting documentation was actually made available to Trinity. Once
the files were received, Trinity began the weeks-long process of
transferring the data to Trinity's computers, performing necessary model
runs using the newly available data, and processing results. Until those
steps are completed, Trinity will not be in a position to determine
whether EPA made any errors with respect to the claimed interference by
Kansas in Allegan, MI. [Westar Energy (EPA-HQ-OAR-2009-0491-4583), p.
1-3; Kansas City Board of Public Utilities (EPA-HQ-OAR-2009-0491-4585),
p. 1-3.]

RESPONSE:  Air quality modeling input and output files developed by EPA
for the final Transport Rule were available in the docket on the date of
signature of the final rule (see docket item EPA-HQ-OAR-2009-0491-4228,
"Final Transport Rule Air Quality Modeling Data Drives").   This docket
item provides instructions on how to obtain electronic copies of the
final Transport Rule modeling input and output files.  The material that
was available for instant access online was sufficient to understand the
application of the modeling in order to provide meaningful comment.  The
modeling code is available as a public resource in concert with the
substantive material presented for immediate consumption and review
online.  Additionally, all of this material was available in the docket
office as of the day of signature and prior to publication in the
Federal Register.  Also, a commenter submitted a follow-up analysis
pertaining to certain modeling input files by Trinity Consultants that
came in after the close of the 60-day comment period.  These comments
are responded to in section 6 (B), below.

Comment:  As of the filing of the comments, Trinity has not had enough
time to complete that process, which is critical to determining whether
the SNPR modeling erred, and how such errors affected the final results.
 This precludes commenter from filing meaningful comments because of the
limited nature of the topics (whether errors in the modeling occurred)
allowed to be addressed at this time.  A complete analysis of the actual
CAMx model input and output files is needed to address matters of
central relevance to the SNPR, and to show errors that would increase
the likelihood that the current SNPR must be significantly changed. 
EPA's denial of an extension prevents the filing of meaningful analysis
and comments that could invalidate any rule that results from the SPNR,
and thus constitutes reversible error. [Westar Energy
(EPA-HQ-OAR-2009-0491-4583), p. 1-3; Kansas City Board of Public
Utilities (EPA-HQ-OAR-2009-0491-4585), p. 1-3.]

RESPONSE:  Our methodologies were documented in the preambles for the
proposed rule and final rule and the proposal and final rule Air Quality
Modeling TSDs.  There was ample opportunity to comment on these methods
based on information provided in the proposed rule.   

Comment:   Trinity's initial assessment of the SNPR modeling suggests
several areas that could uncover errors in the final results after full
examination and analysis. In this regard, the projected maximum ozone
concentration at the Allegan receptor was 85.1 ppb, or 0.1 ppb over the
threshold of 85 ppb for a maintenance site. Previously, EPA projected
that the concentration at the Allegan receptor was 84.7 ppb. The
increased projected concentration between the Proposed Rule and the SNPR
would seem to suggest that Kansas EGU emission projections increased
since the prior projections were made. Yet, that is not the case;
instead, Kansas EGU projected emissions dropped and other Kansas sources
were projected to have higher emissions. Certainly, this calls into
question whether Kansas EGU emissions should be considered to be the
cause of interference at the Allegan MI receptor. [Westar Energy
(EPA-HQ-OAR-2009-0491-4583), p. 1-3; Kansas City Board of Public
Utilities (EPA-HQ-OAR-2009-0491-4585), p. 1-3.]

RESPONSE:  While EPA is not reconsidering or reopening the issue, we
note that in both the Transport Rule proposal and final rule, CAMx
photochemical modeling for the 2012 “no CAIR” base case was used to
determine nonattainment and maintenance receptors.  The final rule
modeling was updated based on comments from the proposal.  There were
numerous emissions updates, including updates to both the EGU and
non-EGU components of the inventory.  The commenter is correct that in
some cases EGU emissions went down between proposal and final.  In
general mobile source emissions went up, due to the availability of
updates to the mobile source emissions model (MOVES).  EPA ran CAMx for
the final rule and followed the proposed rule methodology for
determining nonattainment and maintenance receptors and contributions
from upwind states.  The Allegan receptor went from a level that was
slightly below the NAAQS in the proposal modeling to a level that was
slightly above the NAAQS (for maintenance) in the final rule modeling. 
Specifically, in the final rule modeling, ozone at the Allegan receptor
was projected to be 85.1 ppb which is 0.2 ppb above the NAAQS (projected
maximum design values of 85 ppb and higher are indicative of a
maintenance problem).  Therefore, the Allegan county monitor became a
maintenance receptor for the final rule analysis.  This receptor was
treated in the same way as every other potential nonattainment and/or
maintenance receptor.  EPA believes that the Allegan county receptor is
a valid maintenance receptor and is not identified as a receptor due to
any errors in the modeling or the calculation of future design values. 
The modeling system was updated and the receptor was found to have a
maintenance problem in 2012.  The methodology for the Transport Rule
identifies the most cost-effective emission reductions available from
upwind states, which are not necessarily from whichever sector may be
the largest absolute contributor of emissions at any specific receptor. 
All upwind state NOX emissions (including EGU, non-EGU point sources,
and mobile sources) are tracked collectively to determine state-level
downwind ozone contributions.  The contribution modeling was used to
identify states with an ozone contribution that was at least 1% of the
NAAQS.  The identification of EGU emission reductions in the remedy
analysis was based on a determination that, at the cost thresholds
selected under the final Transport Rule, the power sector offers the
most cost-effective emission reductions to reduce a state’s
significant contribution to nonattainment or interference with
maintenance.    

Comment:  Further, the 0.1 ppb differential between the maintenance
threshold and the projected concentration at Allegan heightens the
possibility that any error could dramatically change the results. One
problem with EPA's approach involves its reliance on linear modeling to
represent nonlinear photochemistry for ozone and secondary PM formation.
Linear modeling does not account for limiting factors that act
differently in various emissions scenarios. A related concern involves
the use of a latter version (5.3 versus 5.01) of the CAMx modeling
program for SNPR purposes. Different chemical parameter values are found
in the two versions, with a possible key difference related to changed
chemical reactivity values. What effect changes in these values, which
are used to estimate the projected rates of ozone and PM species
formation, had in the final results were not discussed in the SNPR.
Because there simply has not been enough time since receipt of the
previously unavailable files, Trinity has not yet been able to determine
what effects these changes had on the results or whether they adequately
represent the complex photochemistry involved in quantifying pollutant
concentrations.

It appears also that the SNPR modeling also is biased in favor of
attributing emissions to anthropogenic, rather than to biogenic,
sources. This bias does not reasonably represent the source
contributions at receptors, and fails to account for the limiting factor
among the photochemical reactions when attributing source contributions
to ozone formation. The SNPR modeling does not take into consideration
the ozone formation conditions present near large water bodies, such as
Lake Michigan where the Allegan MI receptor is located, which differ
from those at other areas, and thus unless factored into the model, may
lead to erroneous results. These preliminary assessments of some
possible areas of concern in the modeling suggest areas that could
produce erroneous results. Because an extension was not granted,
commenter does not have a full analysis from Trinity that would
determine whether these and possibly other problems have led to
erroneous results. Commenter reserves its right to submit supplemental
comments into the docket that more fully explain the potential errors
raised by any such concerns after Trinity has had adequate time to
conduct a full analysis of the model using the recently obtained data
files. [Westar Energy (EPA-HQ-OAR-2009-0491-4583), p. 3-4; Kansas City
Board of Public Utilities (EPA-HQ-OAR-2009-0491-4585), p. 3-4.]

RESPONSE:  These comments question our use of CAMx as the modeling tool
for the Transport Rule and our choice of the source apportionment
technique for ozone.  Our methodologies were documented in the preambles
for the proposed rule and final rule and the proposal and final rule Air
Quality Modeling TSDs.  There was ample opportunity to comment on these
methods based on information provided in the proposed rule.  The focus
of the CSAPR transport analysis for ozone was on the contribution to
ozone at downwind receptors resulting from emissions of NOX from
anthropogenic sources in upwind states irrespective of whether the
anthropogenic NOX formed ozone through reactions with anthropogenic VOC
or biogenic VOC.  For this purpose, the APCA ozone source apportionment
technique was used for the proposed and final rule modeling because this
technique allocates ozone formed from anthropogenic NOX reacting with
either anthropogenic or biogenic VOC to the sources of anthropogenic
NOX.  Ozone formed from reactions of biogenic VOC and biogenic NOX are
allocated not includes as part of the state contributions in the CSAPR
modeling.  The commenter claims that only the portion of the ozone
formed from the reaction of anthropogenic NOX with biogenic VOC under
“NOX limited” conditions should be allocated to the anthropogenic
NOX and that ozone formed under “VOC limited” conditions should be
allocated to biogenic sources.  Specifically, the commenter suggests
that we should have used an alternative technique (OSAT) which splits
the ozone from anthropogenic NOX plus biogenic VOC to the biogenic
sources and anthropogenic sources based on assumptions as to whether the
ozone is formed under NOX-limited or VOC-limited regimes.  EPA disagrees
with this comment because even when the amount of ozone is limited about
the amount of biogenic VOC, the ozone is still formed by reactions with
anthropogenic NOX.  We did not get any comments on the proposed rule
that said that we should have used OSAT.  In addition, the commenter
contends that EPA’s approach relied on a linear model.  This is not
correct since the chemical mechanism in CAMx includes non-linear
chemistry associated with the formation of ozone and secondary PM2.5
species concentrations.  The rationale for changes in the chemical
parameter file in CAMx v5.3 versus v5.0 are documented in the CAMx v5.3
User’s Guide (docket item EPA-HQ-OAR-2009-0491-4380).  This document
clearly states the users should not “mix and match” a version of
CAMx with a different version of the chemical parameter file, as
suggested by the commenter.  Rather, the chemical parameter files
specified in the user’s guide for CAMx v5.3, as applied by EPA, should
be used when running CAMx v5.3.  The commenter contends that the SNPR
modeling does not take into consideration the ozone formation conditions
present near large water bodies, such as Lake Michigan where the
Allegan, MI receptor is located.  However, the commenter provides no
information to support this contention.  In fact, EPA’s CAMx air
quality modeling and the MM-5 meteorological modeling used to provide
weather inputs to CAMx do include Lake Michigan (and other large bodies
of water) in terms of the thermodynamic effects and differential
deposition properties of these bodies of water on pollutant
concentrations.  EPA compared the CAMx ozone predictions for Allegan
County to monitored data and found that the model predictions closely
replicate observations, adding confidence to EPA’s model predictions
for this location (see Appendix A of EPA-HQ-OAR-2009-0491-4140, "Air
Quality Modeling Final Rule Technical Support Document").

Michigan

No comments received on this issue.

Missouri

EPA’s supplementary proposal does not require full elimination of
Missouri’s significant contribution.

Comment:   Finally, Sierra Club notes that the cross state rule does not
eliminate each of these states’ significant contribution to downwind
ozone nonattainment and/or interference with maintenance.  The Act
requires EPA to promulgate FIPs for states that contain inadequate
provisions “prohibiting…any source or other type of emissions
activity within the State from emitting any air pollutant in amounts
which will contribute significantly to nonattainment in, or interfere
with maintenance by, any other State with respect to any such national
primary or secondary ambient air quality standard…”.  Nonetheless,
EPA confesses in the Preamble to proposed rule that it has not satisfied
this obligation for all of the states covered by the final rule’s
ozone program, including, Missouri.  EPA expressly explains that
“[t]he one state addressed in the supplemental proposal for which
EPA’s analysis identifies reductions that are necessary but may not be
sufficient to satisfy section 110(a)(2)(D)(i)(I) with respect to the
1997 ozone NAAQS is Missouri.” For this reason, the cross state rule
and proposed supplement will not fully satisfy EPA’s
110(a)(2)(D)(i)(I) FIP duty. EPA must require additional reductions in
ozone forming emissions that are sufficient to eliminate Missouri’s
significant contribution to ozone nonattainment and/or interference with
maintenance in the States to which it is linked.  [Sierra Club
(EPA-HQ-OAR-2009-0491-4587), p. 3.]

RESPONSE:  EPA has noted that the Transport Rule FIP may not fully
address Missouri’s obligations under section 110(a)(2)(D)(i)(I) with
regard to the 1997 ozone NAAQS.  The final rule states that further
analysis is needed to determine if these FIPs do fully satisfy section
110(a)(2)(D)(i)(I) or if additional reductions are needed.  The FIP for
Missouri makes measureable progress toward satisfying the requirements
of section 110(a)(2)(D)(i)(I) with respect to the 1997 ozone NAAQS.  To
the extent that significant contribution to nonattainment and/or
interference with maintenance is not entirely eliminated, EPA will
address these instances in a future action.  For more information, see
sections IV.C (76 FR 48218), VI.A (76 FR 48246), and VI.D (76 FR 48263,
August 8, 2011) within the preamble of the final Transport Rule, and see
section IV of the Transport Rule Primary Response to Comments document
[EPA-HQ-OAR-2009-0491-4513].

 (B) EPA is using databases that are out of date and do not represent
current air quality.

Comment:  EPA Should Use a More Recent Base Year Than 2005.  EPA has
used inventory and meteorological data that is not representative of
current conditions.  Inventories and meteorology are available for 2008.
Currently the Lake Michigan Air Directors Consortium (LADCO) is
performing modeling with both 2007 and 2008 emissions and meteorological
data.  In addition the Midwest Ozone Group has performed modeling (see
comments submitted by the Midwest Ozone Group, October 1, 2010 to the
original proposed rule (75 Fed. Reg. 45210-45465 (August 2, 2010)) that
shows a totally different picture than that portrayed by EPA’s
modeling. EPA’s reliance on non-current data produces results that are
not reflective of today's environment and should not be relied upon as
the basis of such a far reaching rule. EPA should redo its analysis with
more recent and relevant data. In the case of Missouri's contribution to
the maintenance and nonattainment monitors the July 11, 2011 Federal
Register notice sites 3 monitors; Harris County, TX, Brazoria County, TX
and Allegan County, MI.

Harris and Brazoria, TX Counties - Both of these counties are located in
the Houston metropolitan area; the most populous city in Texas
surrounded by many large VOC and NOX sources from refineries, mobile
sources, power stations and one of the largest ports in the US.
Missouri's contribution to these monitors is 1 ppb ozone or less or
approximately 0.2 ppb ozone over the 1% arbitrary threshold (0.8 ppb)
EPA has chosen. EPA in its current analysis is using 2005 base case
emissions that are significantly higher than those of later years.
Commenter provides table developed by the Midwest Ozone Group using EPA
compiled emissions data compares Missouri statewide NOX emissions from
1999 to 2009. Using the information presented in that table, if EPA
would have chosen 2008 as its base year the NOX inventory for Missouri
would have dropped by 129,580 tons. This would lead to a similarly large
drop in EPA 2012 and 2014 projected emissions. Clearly this large of a
reduction in NOX emissions would significantly lower Missouri's impact
on these Texas monitors to the point of no significance.

Allegan County, MI monitor – This county according to EPAs designation
in the modeling performed is considered a maintenance monitor since its
maximum 2012 ozone design value is 85.1 ppb ozone; 0.2 ppb over the
threshold for an attaining county. According to EPAs modeling Missouri
significantly contributes to this monitor. However this significant
contribution is based on the projected 2005 emissions inventory.
Commenter provides table developed by the Midwest Ozone Group using EPA
compiled emissions data compares Michigan statewide NOX emissions from
1999 to 2009. Using the information presented in that table, if EPA
would have chosen 2008 as its base year the NOX inventory for Michigan
would have dropped by 95,775 tons. This would lead to a similarly large
drop in EPA 2012 and 2014 projected emissions. Clearly this large
reduction in Michigan NOX emissions (as well as those in Missouri) would
have lowered the Allegan County monitor to an ozone level much below the
significance threshold of 84.9 ppb ozone.

Clearly if EPA would have chosen a more recent and relevant base
emissions case that was reflective of current NOX emissions Missouri's
impact on these 3 monitors would be insignificant. EPA needs to revisit
this analysis using more current data sets that reflect actual air
quality.  [Ameren Corporation (EPA-HQ-OAR-2009-0491-4596), p. 4-7.]

RESPONSE:  The comment about the base case is outside the scope of the
SNPR because it addresses issues that were within the scope of the
request for comments in the proposed Transport Rule but not in the SNPR.
 While EPA is not reconsidering or reopening the issue, we note that EPA
took comment on the rationale for selecting 2005 as the base year and
the use of monitoring data for the period 2003 through 2007 as the
starting point for projecting ozone and PM2.5 concentrations in 2012 and
2014 in the Transport Rule proposal.  Also see the Transport Rule
Primary Response to Comments document in the docket to this rulemaking
[EPA-HQ-OAR-2009-0491-4513] on the use of 2005 as the base year and the
use of 2003 through 2007 monitoring data.

Comment:   Current 2007-2009 8-hour ozone design values show major
inconsistencies with the results of EPA’s analysis for the CSAPR.
Commenter provides a table that shows a comparison of 2003-2009 8-hour
ozone design values and EPA’s 2014 remedy case for selected monitors
in Kansas and Missouri.  As is evident from the table all of these
monitors have current design values that are already lower than what EPA
predicts for its 2014 remedy case.  EPA by overestimating the impact of
states on downwind nonattainment and maintenance areas will tend to
increase the level of control perceived to be required to meet the
desired goals.  This will lead to the installation controls that in
reality may not be needed.  EPA’s base case selection and control
assumptions are fatally flawed and need be re-analyzed using a more
recent and relevant information.  [Ameren Corporation
(EPA-HQ-OAR-2009-0491-4596), p. 7-8.]

RESPONSE:  This comment is outside the scope of the SNPR because it
addresses issues that were within the scope of the request for comments
in the proposed Transport Rule but not in the SNPR.  While EPA is not
reconsidering or reopening the issue, we note that EPA took comment on
the rationale for selecting 2005 as the base year and the use of
monitoring data for the period 2003 through 2007 as the starting point
for projecting ozone and PM2.5 concentrations in 2012 and 2014 in the
Transport Rule proposal.  Also see the Transport Rule Primary Response
to Comments document in the docket to this rulemaking
(EPA-HQ-OAR-2009-0491-4513) on the use of 2005 as the base year and the
use of 2003 through 2007 monitoring data.  

In response to the assertion that EPA overestimates the impact of states
on downwind nonattainment and maintenance and thus increases the level
of control perceived to be required to meet the desired goals, EPA
refutes this notion on two grounds.  First, EPA does not overstate the
emission impacts from individual sources.  It is very clearly described
in the preamble section V.B of the program.  Because the rule is
substitution, not incremental, for CAIR,  it cannot assume CAIR in its
base case and any assertions that its base case overstates emissions
because of its failure to reflect CAIR-driven reductions ignores this
important component of the Court’s decision in North Carolina.  On the
second notion that it overstates the control required in the state, EPA
also disagrees.  The first controls that would be operated in a state
would be those already in place (and in some cases already operating). 
Therefore, in many states the difference between the base case and the
policy case emissions is not the installation of additional controls,
but simply the operation of existing controls.  The commenter’s
generation portfolio is primarily located in Missouri and Illinois.  EPA
does not anticipate any new controls in these states in response to
Transport Rule.  In fact, for many of the programs in these states, the
state is already emitting at levels less than their 2012 budget level,
suggesting no additional reductions/controls are required from current
levels.  Where this is not the case, EPA modeling may be capturing
additional reductions that are on the way because of state rules (as is
the case with Illinois and its multi-pollutant rule requirements in
2012).  Finally, any additional reductions that are driven by the
Transport Rule (through fuel switching or other means) are reductions
that are incremental to current levels and are necessary to eliminate
that state’s significant contribution.

Oklahoma

Commenter generally opposes inclusion of Oklahoma.

Comment:  The inclusion of the new states in the ozone transport rule is
another example of why EPA is getting beat up by the politicians and
people who are in the know.  As an environmental professional for 35
years I’ve always said EPA is its own worst enemy. You lose the trust
of the people, and it’s a matter of time. In this case, the EPA is
working against the State of OK because OK is causing a violation in one
county in MI. Great model!  [EST Inc. (EPA-HQ-OAR-2009-0491-4570), p. 1]

Several individuals in Oklahoma similarly stated that they question the
logic that Oklahoma's emissions could, in fact, have an impact on one
single county in Michigan, and that this implausible notion is being
used as the basis for including their state in these proposed rules.
These commenters recommend that EPA should stop and consider the impact
the proposed rule would have on PSC of Oklahoma and its generation
fleet, and not include Oklahoma in the

Transport Rule. [Individual email comments (EPA-HQ-OAR-2009-0491-4608
through 4618), p. 1]

RESPONSE:  As explained in the preamble to the final supplemental rule,
EPA is obligated under its Clean Air Act authority to protect human
health and welfare.  Using the notice and comment process, EPA is
finalizing the supplemental rule based on our conclusions and taking
into account public comments.  The rule will help reduce ozone season
NOX emissions that expose hundreds of million Americans to harmful
health impacts from ozone pollution.  EPA’s methodology for
identifying nonattainment and maintenance receptors is based on modeled
projections of measured air quality at specific monitors, not on the
designation status of an area.  EPA believes this approach is
appropriate for the reasons explained in section V.C.2 of the preamble
to the final Transport Rule.  76 FR 48230, August 8, 2011.  Further, the
North Carolina v. EPA opinion requires EPA to give independent meaning
to the interfere with maintenance prong of section 110(a)(2)(D)(i)(I).

 

 (B) EPA lacks a reasonable basis for concluding that Oklahoma
significantly interferes with maintenance of the 1997 Ozone NAAQS in
Michigan.

Comment:  In the 2010 Proposed Transport Rule, EPA linked sources in
Oklahoma with significant impacts in a nonattainment area in Tarrant
County, Texas. The emission reductions required of Oklahoma in the 2010
Proposed Rule were not significant and did not require costly controls
or the shut-down of units. In the Supplemental Proposal, EPA now
abandons its claim of significant impacts in Texas and for the first
time asserts that Oklahoma interferes with maintenance of the 1997 ozone
NAAQS based exclusively on a newly-identified ozone maintenance receptor
in Allegan County, Michigan. This new claim is not only based on a new
wind direction, but also would require a 40% reduction from actual
emissions in approximately six months time. The maintenance area in
question is roughly 700 miles from most of the affected EGUs in
Oklahoma, and the area does not have a nonattainment problem.

Some of the commenters note that EPA's determination that it is
necessary to regulate Oklahoma under CSAPR is arbitrary and capricious.
EPA is proposing to include Oklahoma in the ozone-season NOX program
based solely on its projected significant contribution to or
interference with maintenance of the ozone NAAQS in Allegan County,
Michigan. Allegan County currently is in attainment. EPA is proposing to
include Oklahoma in the ozone-season NOX program not because emissions
from Oklahoma are causing any nonattainment problems, but because EPA's
models predict that ambient ozone concentration in Allegan County will
increase to 85.1 parts per billion ("ppb"), just 0.1 ppb above the 1997
ozone NAAQS. See 62 Fed. Reg. 38,856 (July 18, 1997).

EPA's decision to subject Oklahoma to regulation under CSAPR solely
based on predicted increase in ozone of just 0.1 ppb is not supportable.
The CAMx model is highly variable and will produce significantly
different results based on relatively small changes in the underlying
emissions inventory and other modeling assumptions. The emissions
inventory data relied on by EPA contain a significant number of errors
and, thus, raise concerns regarding the validity of EPA's modeling
results. Additionally, the model relied on by EPA has a significant
degree of uncertainty when used to predict future emissions. According
to EPA's technical support documents, the predicted increase in ozone in
Allegan County during the ozone season has degree of uncertainty of -3.0
to +3.6 percent of the normalized mean bias. The predicted exceedance of
0.1 ppb is well within the range of potential uncertainty and is not
sufficient justification for subjecting Oklahoma EGUs to extremely
burdensome and costly regulatory requirements.  [Oklahoma Gas & Electric
Company (OGE) (EPA-HQ-OAR-2009-0491-4590), p. 5; Oklahoma Utility Group
(EPA-HQ-OAR-2009-0491-4591), p. 3-4; Public Service Company of Oklahoma
(EPA-HQ-OAR-2009-0491-4600), p. 6-7; Western Farmers Electric
Cooperative (EPA-HQ-OAR-2009-0491-4589), p. 6-7; Grand River Dam
Authority ("GRDA") (EPA-HQ-OAR-2009-0491-4586), p. 3-4.]

RESPONSE:  In response to comments on the proposed Transport Rule, EPA
made numerous updates to the emission inventories used for the final
rule modeling.  As a result of all of the updates made to the modeling
platform, the maximum 2012 ozone design value at the Allegan receptor
went from 84.7 ppb (i.e., a level that was slightly below the NAAQS) in
the proposal modeling to 85.1 ppb (i.e., a level that was slightly above
the NAAQS) in the final rule modeling.  This relatively small increase
in ozone concentration between the proposal modeling and final rule
modeling of 0.4 ppb is not “highly variable,” as is contended by the
commenter.  The Allegan county monitor was treated in the same way as
every other nonattainment and/or maintenance receptor.  EPA believes
that the Allegan county receptor is a valid maintenance receptor and is
not identified as a receptor due to any errors in the modeling or the
calculation of future design values.   As described in the Transport
Rule Response to Comments document, EPA believes that it is
inappropriate to use model performance results, such as normalized mean
bias, as a quantitative indicator of model uncertainty in part because
the modeling results are used in a relative sense to project measured
concentrations to 2012 and 2014.  In other words, EPA is not basing the
Transport Rule purely on air quality modeling output, but on an
application of that modeling output to real observed air quality data at
the relevant sites.  A 2002 report by the National Research Council
(Estimating the Public Health Benefits of Proposed Air Pollution
Regulations) states that the accuracy of modeling is increased when the
model predictions for a recent base period and a future time period are
used in a relative sense as scaling (or response) factors to
“calibrate” observed pollutant levels from the recent base period.
 The report states that, “This approach may help reduce the bias
introduced by modeling errors and, therefore, may be more accurate than
using model results directly (absolute values) to estimate future
pollutant levels.” Thus, it is not appropriate to simply use base year
model performance results as a quantitative measure or gauge of the
uncertainty in the air quality modeling process.

 Comment:  Even assuming that EPA can include Oklahoma in the CSAPR
based solely on modeled impacts, commenters are concerned about several
uncertainties in EPA's modeling that need to be addressed in order to
reasonably evaluate whether the Allegan County monitor represents a
maintenance issue for the ozone NAAQS. A report on these uncertainties
was prepared by Trinity Consultants and is included with various comment
letters as Attachment A.

First, EPA's modeling overstates mobile source emissions by failing to
account for new corporate average fuel economy standards and greenhouse
gas emission standards for mobile sources. The latest version of the
Motor Vehicle Emission Simulator was not used in EPA's modeling and
typically predicts less VOC emissions compared to the version used by
EPA for CSAPR. Second, EPA used different chemical parameters in the
CAMx for the Supplemental Proposal than for the 2010 Proposed Rule. This
change in chemistry is not explained. Third, the location of the Allegan
County monitor near the banks of Lake Michigan creates cause for concern
that EPA's model may not adequately address certain unique aspects of
near-water chemistry or atmospheric conditions.

Other of the Oklahoma utility commenters referenced the following key
findings from the Trinity Consultants report:

Comparing the CSAPR to the proposed Transport Rule, NOX and VOC
emissions were noticeably less. Considering that Allegan County was not
identified as a non-attainment or maintenance concern in the proposed
Transport Rule, EGUs do not appear to be the primary source category
impacting Allegan County. Therefore, EGU s should not be required to
make aggressive and disproportionate NOX reductions of 60% when other
sources are having a more significant impact. 

Mobile source emissions increased from those predicted in the proposed
Transport Rule. EPA did not apply the most recent available version of
MOVES for predicting the mobile emission impact. The newer version would
likely project lower emissions from mobile sources which may show
Allegan County to not be above the maintenance threshold.

EPA changed the CAMx version used between the proposed Transport Rule
and the CSAPR. The two versions of CAMx have different chemical
parameters and represent a potentially significant change in methodology
results. These changes may be enough to cause Allegan County to have
been reclassified and a change in methodology this significant should
not be made without the opportunity for public comment or adequate
explanation of the impact of the differences. 

EPA may have overestimated the anthropogenic contribution of NOX and VOC
emissions. As a result, the costly controls imposed by CSAPR may not
produce the reduced impact to Allegan County predicted. EPA should use
the Ozone Source Apportionment Technology (OSAT) for CAMx and reevaluate
the impacts to properly account for the limiting photochemical factors. 

The Allegan County location has unique characteristics that may cause
the modeling not to be representative of actual impacts. The effects of
its proximity to Lake Michigan suggests that the impacts are more likely
to be from more local sources that from such distances as Oklahoma
sources.

Since the Allegan County receptor exceeds EPA's threshold for being a
maintenance receptor by just 0.1 ppb, failure to account for these
uncertainties in the modeling could significantly impact EPA's
conclusion in the Supplemental Proposal. EPA should address these
uncertainties before including Oklahoma in the CSAPR based on modeled
impacts in Allegan County, Michigan. One commenter adds that they
believe that the model has other errors that tend to overstate source
emissions and the corresponding modeled impacts in Allegan County.
Commenter reserves the right to supplement these comments if additional
errors are identified. [Oklahoma Gas & Electric Company (OGE)
(EPA-HQ-OAR-2009-0491-4590), p. 5; Oklahoma Utility Group
(EPA-HQ-OAR-2009-0491-4591), p. 3-4; Public Service Company of Oklahoma
(EPA-HQ-OAR-2009-0491-4600), p. 6-7; Western Farmers Electric
Cooperative (EPA-HQ-OAR-2009-0491-4589), p. 6-7; Grand River Dam
Authority ("GRDA") (EPA-HQ-OAR-2009-0491-4586), p. 3-4.]

In comments submitted in mid-November well after the close of the public
comment period, some of these commenters provided further comment on
this issue and attached a follow-up study from Trinity Consultants.  The
commenters note that EPA, in its Technical Support Document (“TSD”)
for the Proposal, stated that onroad vehicle emissions were “primarily
based on the publicly released 12/21/2009 version of the Motor Vehicle
Emissions Simulator (MOVES2010).” However, Trinity Consultants was
unable to replicate EPA’s mobile source emissions using the publicly
available information for MOVES2010. Upon further correspondence with
EPA after the close of the comment period, Trinity Consultants learned
that different datasets need to be used in lieu of the publicly
available database for MOVES2010 in order to duplicate EPA’s mobile
source emissions in CSAPR. Trinity further identified that EPA adjusted
the MOVES2010 inventory output after processing by converting the model
output into emissions per vehicle mile traveled (“VMT”) and then
applied data from a separate VMT database. This is different from the
approach explained in the TSD. The actual VMT information used by EPA to
project Allegan County, Michigan as a maintenance interference location
apparently has not been made available for public review and comment.

These commenters state that Oklahoma is included in the Proposal based
solely on a link to a projected future maintenance problem in Allegan
County.  If EPA had used the publicly available default options for
vehicle emissions in MOVES2010 as described in its TSD, Trinity’s
analysis shows that the vehicle emissions contributed to Allegan County
from upwind states would be roughly five percent less than what EPA
actually modeled. Furthermore, because EPA predicts in the Proposal that
future ozone concentrations in Allegan County will be just slightly
above the ozone standard (by just 0.1 ppb), this difference in vehicle
emissions could affect whether or not Allegan County has a projected
maintenance problem under CSAPR.

These commenters continue by stating that in other rulemaking actions
unrelated to CSAPR, EPA has designated Allegan County as an attainment
area for ozone based on actual, real-world monitoring data.  EPA also
approved a revision to the Michigan State Implementation Plan
(“SIP”) that will maintain compliance with the 1997 8-hour ozone
standard through 2021 in Allegan County. The SIP does not require
emission reductions from out-of-state sources to maintain attainment. It
is only in CSAPR modeling that EPA projects a future maintenance problem
for Allegan County, and even there only by a very slight margin. If EPA
were to utilize the public information as stated in the TSD for the
vehicle emission modeling, we believe that the CSAPR modeling for
Allegan County would show future attainment with the ozone standard
consistent with actual monitoring data and the Michigan maintenance SIP
for that area.

The commenters conclude by arguing that EPA’s authority to impose
emission reduction requirements based on modeling is limited in
important respects. A complete analytic defense of a model’s
application is required in each case, and EPA must consider real-world
data to ensure that modeled predictions are reasonable and accurate.
Where real-world data contradict a model, EPA must address with
reasonable analysis the discrepancies between its model and the actual
data.  Based on Trinity’s analysis, the commenters argue that it is
not appropriate for EPA to issue a final rule without addressing the
discrepancies noted and explaining why they deviated from the approach
set forth in the TSD.  EPA should also reconsider other aspects of the
modeling, such as using maximum design values to identify maintenance
sites, even though EPA guidance seems to call for the use of average
design values in attainment modeling. [AES Shady Point et al.
(EPA-HQ-OAR-2009-0491-4739), p. 1-3 and attachment]

RESPONSE:   These comments include questions about our use of CAMx as
the modeling tool for the Transport Rule and our choice of the source
apportionment technique for ozone.  Our air quality modeling
methodologies were documented in the preambles for the proposed rule and
final rule and the proposal and final rule Air Quality Modeling TSDs. 
There was ample opportunity to comment on these methods based on
information provided in the proposed rule.  The focus of the CSAPR
transport analysis for ozone was on the contribution to ozone at
downwind receptors resulting from emissions of NOX from anthropogenic
sources in upwind states irrespective of whether the anthropogenic NOX
formed ozone through reactions with anthropogenic VOC or biogenic VOC. 
For this purpose, the APCA ozone source apportionment technique was used
for the proposed and final rule modeling because this technique
allocates ozone formed from anthropogenic NOX reacting with either
anthropogenic or biogenic VOC to the sources of anthropogenic NOX. 
Ozone formed from reactions of biogenic VOC and biogenic NOX are
allocated not includes as part of the state contributions in the CSAPR
modeling.  The commenter claims that only the portion of the ozone
formed from the reaction of anthropogenic NOX with biogenic VOC under
“NOX limited” conditions should be allocated to the anthropogenic
NOX and that ozone formed under “VOC limited” conditions should be
allocated to biogenic sources.  Specifically, the commenter suggests
that we should have used an alternative technique (OSAT) which splits
the ozone from anthropogenic NOX plus biogenic VOC to the biogenic
sources and anthropogenic sources based on assumptions as to whether the
ozone is formed under NOX-limited or VOC-limited regimes.  EPA disagrees
with this comment because even when the amount of ozone is limited about
the amount of biogenic VOC, the ozone is still formed by reactions with
anthropogenic NOX.  We did not get any comments on the proposed rule
that said that we should have used OSAT.  The rationale for changes in
the chemical parameter file in CAMx v5.3 versus v5.0 are documented in
the CAMx v5.3 User’s Guide (docket item EPA-HQ-OAR-2009-0491-4380). 
This document clearly states the users should not “mix and match” a
version of CAMx with a different version of the chemical parameter file,
as suggested by the commenter.  Rather, the chemical parameter files
specified in the user’s guide for CAMx v5.3, as applied by EPA, should
be used when running CAMx v5.3.  The commenter contends that the SNPR
modeling does not take into consideration the ozone formation conditions
present near large water bodies, such as Lake Michigan where the
Allegan, MI receptor is located.  However, the commenter provides no
information to support this contention.  In fact, EPA’s CAMx air
quality modeling and the MM-5 meteorological modeling used to provide
weather inputs to CAMx do include Lake Michigan (and other large bodies
of water) in terms of the thermodynamic effects and differential
deposition properties of these bodies of water on pollutant
concentrations.  EPA compared the CAMx ozone predictions for Allegan
County to monitored data and found that the model predictions closely
replicate observations, adding confidence to EPA’s model predictions
for this location (see Appendix A of EPA-HQ-OAR-2009-0491-4140, "Air
Quality Modeling Final Rule Technical Support Document").  The response
to the commenter’s claims about the mobile emissions used in EPA’s
air quality modeling are provided below.

Comment:  In comments submitted by AES Shady Point above, they claim
that EPA did not apply the most recent available version of MOVES for
predicting the mobile emission impact and that the newer version would
likely project lower emissions from mobile sources which may show
Allegan County to not be above the maintenance threshold.  On November
11, after the close of the public comment period, further comment was
provided on the modeled mobile source emissions along with a study from
Trinity Consultants.  In the November 11 submittal, the commenters note
that in the Emissions Inventory Technical Support Document (“TSD”)
for the final rule, EPA stated that onroad vehicle emissions were
“primarily based on the publicly released December 21, 2009 version of
the Motor Vehicle Emissions Simulator (MOVES2010).” However, the
commenters claim that Trinity Consultants was unable to replicate
EPA’s mobile source emissions using the publicly available information
for MOVES2010.  Upon further correspondence with EPA after the close of
the comment period, Trinity Consultants learned that different datasets
need to be used in lieu of the publicly available database for MOVES2010
in order to duplicate EPA’s mobile source emissions in CSAPR.  Trinity
further identified that EPA adjusted the MOVES2010 inventory output
after processing by converting the model output into emissions per
vehicle mile traveled (“VMT”) and then applied data from a separate
VMT database.  This is different from the approach explained in the TSD.
 The actual VMT information used by EPA to project Allegan County,
Michigan as a maintenance interference location apparently has not been
made available for public review and comment.

These commenters state that Oklahoma is included in the proposal based
solely on a link to a projected future maintenance problem in Allegan
County.  If EPA had used the publicly available default options for
vehicle emissions in MOVES2010 as described in its TSD, Trinity’s
analysis shows that the vehicle emissions contributed to Allegan County
from upwind states would be roughly five percent less than what EPA
actually modeled.  Furthermore, because EPA predicts in the proposal
that future ozone concentrations in Allegan County will be just slightly
above the ozone standard (by just 0.1 ppb), this difference in vehicle
emissions could affect whether or not Allegan County has a projected
maintenance problem under CSAPR.

If EPA were to utilize the public information as stated in the TSD for
the vehicle emission modeling, the commenters believe that the CSAPR
modeling for Allegan County would show future attainment with the ozone
standard consistent with actual monitoring data and the Michigan
maintenance SIP for that area.

Based on Trinity’s analysis, the commenters argue that it is not
appropriate for EPA to issue a final rule without addressing the
discrepancies noted and explaining why they deviated from the approach
set forth in the TSD.  [AES Shady Point et al.
(EPA-HQ-OAR-2009-0491-4739), p. 1-3 and attachment]

RESPONSE:  The calculation of onroad mobile emissions involves the use
of the MOVES model code along with several input data sets.   The inputs
include a “database” with information for calculating emissions
factors, vehicle miles travelled (VMT) data, and meteorological data. 
For the final Transport Rule, EPA used the MOVES2010 code that was
released to the public on 12/21/2009 (i.e., December 2009 release).  In
addition to the MOVES code, this release also included several default
data sets that could be used as inputs for running MOVES.  In general,
the defaults are provided for users who do not have more appropriate,
representative data for their particular application.  There is no
requirement to use the default data sets when running the MOVES model. 
As described in the Emissions Inventory Technical Support Document (TSD)
and summarized below, EPA did not use all of the default data sets that
came with the December 2009 release for creating onroad mobile emissions
for the final Transport Rule.  The commenter misinterpreted the TSD and
incorrectly assumed that EPA used all the default data sets that were
included with the December 2009 release of MOVES.  The TSD says that we
primarily used MOVES 2010 to compute onroad emissions, but clearly
states that we did not use all of the default data provided with the
December release.  Instead, we strove to use the best available data at
the time of the modeling as input to MOVES.  

For the final Transport Rule modeling, EPA did not use the default
“database” that came with the December release.  Rather, we used the
“database” created and released publicly in May 2010.  As noted by
the commenter, the December 2009 database contains errors that were
subsequently corrected in the May 2010 version, and therefore EPA did
not use the December 2009 version for the final Transport Rule Modeling.
 

For 2005, the TSD states that, “The onroad emissions were primarily
based on the publicly released 12/21/2009 version of the Motor Vehicle
Emissions Simulator (MOVES2010) (http://www.epa.gov/otaq/models/moves/).
MOVES was run with a state/month aggregation using county-average fuels
for each state, state/month-average temperatures, and national default
vehicle age distributions. 2005 Vehicle Miles Travelled (VMT),
consistent with the 2005v2 NEI, were used.”  For 2012, the TSD states
“The onroad emissions were primarily based on the 2010 version of the
Motor Vehicle Emissions Simulator (MOVES2010) – the same version as
was used for 2005... Vehicle Miles Travelled (VMT) were projected using
growth rates from the Department of Energy’s AEO2009.”  The TSD
states that we used state-month specific 2005 meteorology, not the
default climatology that was included with the December 2009 release.
The TSD also states that we used 2005 VMT from the 2005v2 NEI for our
2005 modeling, and that VMT based on DOE AEO2009 was used for our future
year modeling.  These are clearly not the default datasets that were
included in the December 2009 release.  Thus, EPA disagrees with the
commenter’s assertion that the TSD does not provide adequate
documentation on the meteorological data, VMT, and VMT growth rates EPA
used for the final Transport Rule.  With a base year of 1999, the
default VMT is known to not properly reflect different growth rates in
different states – even with a year as close to 1999 as 2005.  The
2005 and 2012 state-specific VMT used by EPA provide a more appropriate
representation of VMT for these years versus using the default
1999-based VMT data provided with MOVES.  Since mobile source emissions
of NOX and VOC are sensitivity to temperature, the use by EPA of 2005
meteorology, rather than climatological data, provides for greater
consistency with the 2005 meteorology used to drive the air quality
model.  Thus, we believe that the onroad emissions developed for the
final Transport Rule using MOVES 2010 with the May 2010 database and
other inputs developed by EPA, as described in the TSD, provide a
credible projection of onroad emissions for use in the Transport Rule
air quality modeling.  

The commenter also ran the MOVES model using the “database” and
other default inputs that came with the December release and found that
using these default data produced onroad mobile NOX and VOC emissions
that were lower than those generated by EPA.  EPA disagrees with the
commenter that onroad mobile emissions using the December 2009 defaults,
which contain known errors, should be used for the Transport Rule
modeling.

The onroad mobile emissions EPA created and used for the final Transport
Rule were provided as state totals in the TSD and as county totals in
files in the docket.  Thus, the commenter had ample opportunity to
review the actual emissions data used by EPA.  The commenter requested
and received from EPA the version of the “database” and other inputs
that EPA used for calculating onroad mobile source emissions for the
final rule.  With these inputs, the commenter was able to successfully
replicate EPA’s onroad mobile emissions.  The commenter did not
provide comments saying that EPA had made any errors in the application
of these data.

Comment:  In addition, the analysis provided by Trinity Consultants
suggests that EPA's model tends to overstate the contribution of
Oklahoma sources to ozone formation in Allegan County.  Commenters are
concerned that the drastic emissions reduction required for Oklahoma in
the Supplemental Proposal goes beyond what is required to address
Oklahoma's impact. EPA also should consider that NOX emissions predicted
for all source groups in Oklahoma, other than mobile sources, were lower
in the modeling for the Supplemental Proposal than in the modeling for
the 2010 Proposed Transport Rule.  EGUs showed a decrease of 20,493 tons
per year of NOX, while NOX emissions from mobile sources increased by
21,344 tons per year.  Oklahoma showed an overall increase in VOC
emissions in the modeling for the Supplemental Proposal compared to the
2010 Proposed Rule, but this VOC emissions increase is attributable to
an increase in modeled emissions from mobile and non-point sources
rather than EGUs.  All told, these changes indicate that EGUs are not
the likely cause of the newly identified maintenance issue in EPA's
modeling for Allegan County, Michigan.  EGUs in Oklahoma should not be
held responsible for remedying an increase in emissions from other types
of sources. [Oklahoma Gas & Electric Company (OGE)
(EPA-HQ-OAR-2009-0491-4590), p. 6; Oklahoma Utility Group
(EPA-HQ-OAR-2009-0491-4591), p. 4.]

RESPONSE:  The comment concerning overstating the contribution of
Oklahoma sources to ozone in Allegan County is outside the scope of the
SNPR because it addresses modeling and inventory issues that were within
the scope of the request for comments in the proposed Transport Rule but
not in the SNPR.  While EPA is not reconsidering or reopening the issue,
we note that in both the Transport Rule proposal and final rule, CAMx
photochemical modeling for the 2012 “no CAIR” base case was used to
determine nonattainment and maintenance receptors.  The final rule
modeling was updated based on comments from the proposal.  There were
numerous emission updates, including updates to both the EGU and non-EGU
components of the inventory.  

Comment:  Furthermore, commenters question EPA's authority to include
Oklahoma in the Supplemental Proposal based solely on modeled impacts in
Allegan County, which is in attainment with the applicable ozone NAAQS. 
As noted, EPA is predicting possible impacts in Allegan County from
Oklahoma sources based on its use of the CAMx.  EPA's use of this
computer-modeling approach is overly conservative and tends to
exaggerate the impact of source emissions on air quality.  EPA's
modeling, at most, establishes a tenuous link to a predicted problem in
Michigan, but does not establish that Oklahoma sources actually
"contribute significantly to nonattainment, or interfere with
maintenance of NAAQS," as contemplated by the CAA.  EPA oversteps the
bounds of its authority by proposing to impose stringent requirements on
Oklahoma sources to address an issue that does not exist in the real
world. [Oklahoma Gas & Electric Company (OGE)
(EPA-HQ-OAR-2009-0491-4590), p. 6; Oklahoma Utility Group
(EPA-HQ-OAR-2009-0491-4591), p. 4.]

RESPONSE:  The comment is outside the scope of the SNPR because it
addresses EPA’s authority and modeling issues that were within the
scope of the request for comments in the proposed Transport Rule but not
in the SNPR.  While EPA is not reconsidering or reopening these issues,
we refer the commenter to section IV.A of the final Transport Rule for a
discussion of EPA’s authority, as well as section III.D of the
preamble to the final supplemental rule for a discussion of the Allegan
County, MI receptor.  Additionally, in response to comments on the
proposed rule, EPA performed a trajectory analysis to examine the upwind
transport patterns on days with measured exceedances at the Transport
Rule nonattainment and maintenance sites (see docket item
EPA-HQ-OAR-2009-0491-4360).  EPA’s trajectory analysis included
backward trajectories from Allegan County for multiple times on
exceedance days at Allegan not only for 2005, but also for other years
during the period 2003 through 2007.  This analysis shows multiple days
when air crosses portions of Oklahoma within three days of passing over
Allegan County on days when an exceedance was measured at the Allegan
County monitor in 2005 and on several days in the other years examined. 
EPA believes that the trajectory analysis corroborates the air quality
contribution modeling which links emissions in Oklahoma to the ozone
maintenance problem in Allegan County.

Comment: AES Shady Point believes there are significant technical flaws
in the modeling and allocation process.  In general terms, the need for
significant reductions in Oklahoma for the maintenance of attainment in
Allegan County, Michigan, which is over 900 miles from the AES Shady
Point facility, does not make logical sense because Allegan County is
directly across Lake Michigan from the Chicago/Milwaukee airshed. The
emissions from that airshed (and Gary, IN and St. Louis, MO) would seem
to dwarf any contribution from several hundred miles further away in
Oklahoma. Modeling by Midwest Ozone Group does not indicate that Allegan
County is expected to have attainment maintenance issues.  The fact that
Allegan County, Michigan already is in attainment with the NAAQS
supports the exclusion of Oklahoma in the proposed rule.  There is not
an immediate non-attainment issue that needs to be addressed.  It is
appropriate to allow an orderly process and period to achieve compliance
with a rule that is intended solely for maintenance of current
attainment.  [AES Shady Point (EPA-HQ-OAR-2009-0491-4595), p. 5.]

RESPONSE:  EPA’s methodology for identifying nonattainment and
maintenance receptors is based on modeled projections of measured air
quality at specific monitors, not on the designation status of an area. 
EPA believes this approach is appropriate for the reasons explained in
section V.C.2 of the preamble to the final CSAPR.  76 FR 48230, August
8, 2011.  EPA does not believe it would be appropriate to rely on the
designation status of an area.  The statute does not require EPA to do
so.   In fact, the statutory requirement that section 110(a)(2)(D)(i)(I)
SIPs be submitted within 3 years of promulgation or revision of a NAAQS
would make it difficult to do so, suggesting that Congress did not
intend section 110(a)(2)(D)(i)(I) SIPs to be linked in any way to
designation status.  Further, even areas that have never been in
nonattainment or that have been re-designated to attainment (including
those where the majority of pollution comes from out of state) continue
to be at risk for falling into nonattainment due to the impact of
emissions from upwind states.  We therefore disagree with the commenter
and continue to believe that Allegan was appropriately determined to be
a maintenance receptor.  See Section III.D of the preamble to the final
supplemental.

 

Furthermore, the Transport Rule CAMx source apportionment modeling shows
that 10 states (including Oklahoma) contribute above the 1% of the ozone
NAAQS threshold to Allegan County.  As would be expected, the largest
ozone contribution is from Illinois.  But that does not change the fact
that 9 other states interfere with maintenance in Allegan County and it
does not alleviate the need for those states to eliminate their
interference with maintenance.  Further, the North Carolina v. EPA
opinion requires EPA to give independent meaning to the interfere with
maintenance prong of section 110(a)(2)(D)(i)(I).

 

Comment:   In addition to the Trinity Consultant's evaluation of the
modeling methodology, AEP/PSO found a specific apparent error in the IPM
TR Base Case Final results. PSO operates a combined cycle at its
Northeastern Power Station. Units 1A and 1B are combustion turbines
equipped with heat recovery steam generators that provide steam to the
original Unit 1 turbine. The Unit 1 boiler has not operated since about
1999. Although the NEEDS v4.1 database does not include the Unit 1
boiler as an emission source, EPA's IPM file shows Unit 1 as emitting
0.0789 Mton of NOX in the 2012 Base Case. This raises the question
whether EPA has double counted Units 1A and 1B's emissions and whether
EPA has accurately quantified emissions in the IPM projections. [Public
Service Company of Oklahoma (EPA-HQ-OAR-2009-0491-4600), p. 7 plus
attachment.]

RESPONSE:  EPA did not double count units 1A and 1B emissions in its
base case modeling.  Double counting would suggest a number that was
higher than historic levels.  However, EPA’s base case modeling shows
total ozone-season emissions for the Northeastern combined cycle
facility of approximately 102 tons, where as the 2010 emission levels
were approximately 145 tons.  EPA uses the three units as reported in
EIA 860 by the source owner.  The units are modeled as two combustion
turbines and one steam turbine, as supported by the commenter.  The
model accounts for unit level emissions by taking the facility level
emissions and distributing them among the three units. The steam turbine
is dependent on the combustion turbines for operation, therefore,
emissions from the facility are spread across all three units
proportional to their generation.  A failure to do so would suggest a
unit that could generate electricity without any associated fuel use
related to that generation.  In reality, these emissions (the 79 tons
referenced by commenter) would still occur, but would be emitted from
the other two units at the facility.  However, IPM attributes the
emissions to the steam turbine to appropriately capture the economics. 
If this was not the case, and the model could generate electricity from
a steam turbine without any corresponding emissions, then it would
dispatch steam turbines as economic before the combustion turbines – a
sequence that is not possible.  In short, the facility level emissions
are correctly captured at the Northeaster unit noted above and there is
no double counting as speculated by the commenter.   

Comment:  AEP PSO did an analysis of air parcel back trajectories for
days in 2005 when ozone exceeded 80 ppb at the Allegan monitoring site. 
Their analysis identified two “elevated ozone days where the winds
were originating far to the south of the Lake Michigan region showing a
reasonable possibility of Oklahoma emissions participating in ozone
formation that impacts the Allegan County region”.  The commenter
claims that this “calls into question the accuracy of the
meteorological fields being used by USEPA as they relate to the
transport from elevated sources in the Oklahoma region and the potential
benefits of NOX emissions reductions from elevated sources in Oklahoma
in reducing ozone at the Allegan County monitor in a meaningful
fashion”.  This work is included in an attachment to this letter
titled "USEPA's Analysis of the Allegan County Ozone Monitor Is
Incomplete and Does Not Demonstrate a Connection to the CSAPR Remedy
Proposed in the SNPR".  [Public Service Company of Oklahoma
(EPA-HQ-OAR-2009-0491-4600), p. 7 plus attachment.]

RESPONSE:  In response to comments on the proposed rule, EPA performed a
trajectory analysis to examine the upwind transport patterns on days
with measured exceedances at the Transport Rule nonattainment and
maintenance sites (see docket item EPA-HQ-OAR-2009-0491-4360).  EPA’s
trajectory analysis was more extensive than the analysis performed by
AEP PSO in that EPA’s analysis included backward trajectories from
Allegan County for multiple times on exceedance days at Allegan not only
for 2005 but also for other years during the period 2003 through 2007. 
In general, backward trajectory analyses are intended to show the
upstream pathway followed by air that arrives over a particular location
at a particular time.  In EPA’s trajectory analysis we examined the
airflow pathways within 3 days of when air crossed Allegan County on
each day that an exceedance was measured at the Allegan County monitor. 
 EPA’s trajectory analysis shows that air crossed over portions of
Oklahoma within three days before passing over Allegan County on several
of the days when an exceedance was measured at the Allegan County
monitor.  EPA believes that the trajectory analysis corroborates the air
quality contribution modeling which links emissions in Oklahoma to the
ozone maintenance problem in Allegan County.

Comment:  Under the EPA's Supplemental Proposal, the State of Oklahoma
is being required to significantly reduce emissions of NOX during the
ozone season for the maintenance of a single county in Michigan (Allegan
County).  The county in question is hundreds of miles away from
Oklahoma, and modeling data shows that Allegan County, Michigan, is not
expected to have attainment maintenance issues and, in fact, is already
in attainment with the NAAQS targeted by the proposed rulemaking.  It
does not make sense for the EPA to require significant reductions in
Oklahoma for attainment in Allegan County, Michigan, in an area directly
across Lake Michigan from the Chicago/Milwaukee airshed (and also the
Gary, Indiana, and St Louis, Missouri, airsheds), where these areas
would be far greater contributors to nonattainment in Allegan County,
Michigan, than any contribution from Oklahoma.  In addition, the fact
that Allegan County, Michigan, is already in attainment, as shown by
modeling, supports the exclusion of Oklahoma in the Supplemental
Proposal rulemaking.  [Oklahoma Attorney General
(EPA-HQ-OAR-2009-0491-4592), p. 8.]

 

RESPONSE:  EPA’s methodology for identifying nonattainment and
maintenance receptors is based on modeled projections of measured air
quality at specific monitors, not on the designation status of an area. 
EPA believes this approach is appropriate for the reasons explained in
section V.C.2 of the preamble to the final CSAPR.  76 FR 48230, August
8, 2011.  EPA does not believe it would be appropriate to rely on the
designation status of an area.  The statute does not require EPA to do
so.   In fact, the statutory requirement that 110(a)(2)(D)(i)(I) SIPs be
submitted within 3 years of promulgation or revision of a NAAQS would
make it difficult to do so, suggesting that Congress did not intend
110(a)(2)(D)(i)(I) SIPs to be linked in any way to designation status. 
Further, even areas that have never been in nonattainment or that have
been re-designated to attainment (including those where the majority of
pollution comes from out of state) continue to be at risk for falling
into nonattainment due to the impact of emissions from upwind states. 
We therefore disagree with the commenter and continue to believe that
Allegan was appropriately determined to be a maintenance receptor.  See
Section III.D of the preamble to the final supplemental.

 

Furthermore, the Transport Rule CAMx source apportionment modeling shows
that 10 states (including Oklahoma) contribute above the 1% of the ozone
NAAQS threshold to Allegan County.  As would be expected, the largest
ozone contribution is from Illinois.  But that does not change the fact
that 9 other states interfere with maintenance in Allegan County and it
does not alleviate the need for those states to eliminate their
interference with maintenance.  Further, the North Carolina v. EPA
opinion requires EPA to give independent meaning to the interfere with
maintenance prong of section 110(a)(2)(D)(i)(I).

 

Comment:  The DEQ does not believe that the conclusion that Oklahoma
significantly impacts the air quality in Allegan County, Michigan is
accurate.  The modeling conducted by EPA alleged that the NOX sources in
the state of Oklahoma interfere with maintenance of the ozone standard
for Allegan County, Michigan.  Allegan Co. was designated attainment of
the ozone standard on July 20, 2010, and is now a maintenance area. The
design values for Allegan Co. for the years 2005, 2006, 2007, 2008,
2009, and 2010 are 89, 88, 93, 86, 81, and 74 ppb respectively.  It is
apparent that the emission reductions obtained from the CAIR sources
have helped to lower recent design values to well below the standard,
and that Allegan County’s maintenance of the standard is not
threatened (even the modeled value of 85.1 in the CSAPR was just over
the standard).  If EPA were to update their base case to the year 2008
or later, Allegan Co. would model attainment with no threat to their
maintenance status in a normal weather year. The monitor in Allegan
County is the only monitor projected to be above the standard that EPA
identified Oklahoma as impacting; therefore, if Allegan County is in
attainment of the standard, then Oklahoma can and should be completely
removed from the rule. It is DEQ’s belief that the conclusion that
Oklahoma emissions impact Michigan is false.  [Oklahoma Department of
Environmental Quality (EPA-HQ-OAR-2009-0491-4593), p. 1-2.]

RESPONSE:  See section III.D of the preamble for the final supplemental
regarding Allegan County.  The comments on CAIR, modeling, and the base
case are outside the scope of the SNPR because they address issues that
were within the scope of the request for comments in the proposed
Transport Rule, but not in the SNPR.  While EPA is not reconsidering or
reopening these issues, we note that these issues were previously
addressed in sections V.C.2 (76 FR 48230, August 8, 2011) and V.B (76 FR
48223, August 8, 2011) of the preamble to the final Transport Rule.  

Comment:  It is important to note that Oklahoma is currently being
included for its alleged impact on one county in the entire United
States. Commonsense suggests that carbon emissions in Oklahoma are not
dictating the air quality of Allegan County, Michigan, while somehow
having little or no impact on states with which we share a border.
Without additional and conclusive scientific data, I reject the EPA’s
conclusions in this instance, which I believe call into question the
overall validity of its study and modeling data.  I strongly advocate
that further scientific and empirical modeling is necessary to truly
determine if Oklahoma is impacting any area of the country in relation
to air quality.  I am asking today that the EPA revisit this important
issue to ensure that it is using proven science and techniques to make
an informed decision on this issue.  [Governor, State of Oklahoma
(EPA-HQ-OAR-2009-0491-4594), p. 1.]

RESPONSE:  See section III.D of the preamble to the final supplemental. 
Regarding the comment on carbon emissions, the Transport Rule does not
regulate carbon emissions.  Regarding comments on the validity of
EPA’s modeling, the comment is outside the scope of the SNPR because
it addresses issues that were within the scope of the request for
comments in the proposed Transport Rule but not in the SNPR.  While EPA
is not reconsidering or reopening the issue, we note that our
methodologies were documented in the preambles for the proposed rule and
final rule and the proposal and final rule Air Quality Modeling
Technical Support Documents.  See EPA-HQ-OAR-2009-0491-4372, "Air
Quality Modeling Final Rule Technical Support Document".  There was
ample opportunity to comment on these methods based on information
provided in the proposed rule.

Comment:  Since its promulgation in 2005, CAIR has achieved significant
emissions reductions. For the purpose of determining the base case
(i.e., the analytical baseline emissions scenario), EPA arbitrarily
assumed that CAIR is not in effect. EPA claims that it is required to do
so because if it "were to consider all reductions associated with CAIR
in the 'base case,' the baseline emissions would not adequately reflect
the true 2012 baseline in each state." 76 Fed. Reg. at 48,223.  EPA
incorrectly assumes that pre-CAIR emission levels represent the
emissions that "would occur in each state in 2012 if the Transport Rule
did not require any reductions in that state."  Id. EPA's reasoning is
arbitrary because it lacks a "rational relationship to the real world."
See, e.g., Appalachian Power Co. v. EPA, 249 F.3d 1032, 1053 (D.C. Cir.
2001) (citing Chemical Mfrs. Ass'n v. EPA, 28 F.3d 1259, 1265 (D.C. Cir.
1994» ("While courts routinely defer to agency modeling of complex
phenomena, model assumptions must have a 'rational relationship' to the
real world."); Columbia Falls Aluminum Co. v. EPA, 139 F.3d 914, 923
(D.C. Cir. 1998) ("An agency's use of a model is arbitrary if that model
bears no rational relationship to the reality it purports to
represent.") (internal quotations omitted).

In light of the numerous regulatory developments since 2005 and the
billions of dollars invested in installing controls at EGUs, it is
unreasonable for EPA to assume that all of the EGUs currently subject to
CAIR would instantly stop operating these controls on January 1, 2012. 
The D.C. Circuit in North Carolina remanded without vacatur the federal
CAIR program; it did not in any way invalidate state regulations
implementing CAIR.  Since 2005, EPA has set significantly more stringent
NAAQS for Ozone, SO2, NOX, and PM.  Many states have become dependent on
the emissions reductions achieved by their CAIR programs to attain or
maintain these newly promulgated, and more stringent, NAAQS within their
own borders.  EPA cannot assume that these states will forgo these
reductions simply because the federal mandate to implement CAIR expires.

It is far more rational for EPA to acknowledge the emissions reductions
achieved by CAIR.  EPA cannot simply assume that pre-CAIR emission
levels are an appropriate base case without first determining whether
states will continue to require CAIR reductions.  The U.S. Court of
Appeals remanded CAIR without vacatur and has consistently declined to
set a deadline for EPA to replace CAIR. See North Carolina v. EPA, 550
F.3d 1176 (D.C. Cir. 2008) (remanding without vacating CAIR); Order,
North Carolina v. EPA, No. 09-1316 (D.C. Cir. Feb 2, 2010) (denying
request for writ of mandamus ordering EPA to issue proposed rule).
Nothing requires EPA to discount emissions reductions achieved by CAIR.

Failing to account for CAIR undermines the ability of EPA's model to
accurately determine whether states significantly contribute to
nonattainment or maintenance of the NAAQS. Michigan is a CAIR state and
Allegan County currently is in attainment of the 1997 ozone NAAQS.  In
2010, the Holland monitor, located in Allegan, Michigan, had a
three-year average of 74 ppb.  If EPA used this actual data, which
reflects CAIR reductions, its modeling would show that Allegan, Michigan
would remain in attainment. See Docket No. EPA-HQ2009-0491-2809
(Comments Submitted by the Midwest Ozone Group). Modeling submitted by
the Midwest Ozone Group in response to the 2010 Proposal shows that if
EPA considered CAIR, and other rules currently in place, ambient air
quality at the Holland monitor would be approximately 81 ppb in the base
year, 9 76 ppb in 2014, and 73 ppb in 2018. See id.  There simply is no
rational basis for EPA's determination that Oklahoma will significantly
contribute to nonattainment or interference with the 1997 ozone NAAQS in
Allegan County, Michigan. [Western Farmers Electric Cooperative
(EPA-HQ-OAR-2009-0491-4589), p. 6-7; Grand River Dam Authority ("GRDA")
(EPA-HQ-OAR-2009-0491-4586), p. 4-5.]

RESPONSE:  The comment is outside the scope of the SNPR because it
addresses issues that were within the scope of the request for comments
in the proposed Transport Rule but not in the SNPR.  While EPA is not
reconsidering or reopening the issue, we note that EPA took comment on
the rationale for selecting 2005 as the base year and the use of
monitoring data for the period 2003 through 2007 as the starting point
for projecting ozone and PM2.5 concentrations in 2012 and 2014 in the
Transport Rule proposal.  See section V.B of the final Transport Rule
preamble.  Also see section IV.A of the Transport Rule Primary Response
to Comments document in the docket to this rulemaking
[EPA-HQ-OAR-2009-0491-4513].

	7.  Wisconsin

No comments received on this issue.

II.  FIPs (use of ozone season NOX program in Transport Rule as FIP)   

General

(A) EPA lacks authority to promulgate the proposed FIP.

Comment:  For the reasons discussed in detail in its comments on the
Proposed Transport Rule, UARG opposes EPA’s imposition of FIPs
establishing ozone-season NOX emission reduction requirements in these
states without first giving the states adequate time and opportunity to
develop and submit state implementation plans (“SIPs”) implementing
section 110(a)(2)(D)(i) of the Clean Air Act (“CAA”).  See Docket ID
No. EPA-HQ-OAR-2009-0491-2756.1, at 19-26 (Oct. 1, 2010).  For example,
UARG notes that, even assuming for the sake of argument that, as EPA
asserts, CAA section 110(a)(2)(D) “requires all states, within 3 years
of promulgation of a new or revised NAAQS, to submit SIPs that prohibit
certain emissions of air pollutants because of the impact they would
have on air quality in other states,” 76 Fed. Reg. at 40664 -- an
assertion that ignores that the 3-year SIP submission deadline appears
only in CAA section 110(a)(1), which explicitly applies to SIPs that
“provide[] for implementation, maintenance, and enforcement” of
NAAQS “within,” not outside, the state that submits the SIP -- that
is no basis for EPA’s imposition of FIPs in this rule.  As UARG notes
in its comments, any schedule for submission of section 110(a)(2)(D)(i)
SIPs for the 1997 NAAQS should be restarted “upon EPA’s promulgation
of a valid final rule” replacing the currently effective Clean Air
Interstate Rule.5 Docket ID No. EPA-HQ-OAR-2009-0491-2756.1, at 26. 
[Utility Air Regulatory Group (EPA-HQ-OAR-2009-0491-4598), p. 2-3.]

RESPONSE:  The comment is outside the scope of this action because it
questions the legality of using a FIP, not whether the Transport Rule
FIPs are the appropriate method for addressing the emissions.  While EPA
is not reconsidering or reopening the issue, we note that the legality
question was previously addressed in section IV.A of the preamble to the
final Transport Rule. 

Iowa

No comments received on this issue. 

Kansas

No comments received on this issue.

Michigan

Commenter disagrees with use of the FIP and the associated timeframe.

Comment:  Within the proposed rules, the U.S. Environmental Protection
Agency (EPA) again indicated their decision to use the final Transport
Rule program's federal implementation plans (FIPs) to address the
emissions in these six states.  Once again, and as also stated in our
original comments to the August 2, 2010, Federal Register, "Federal
Implementation Plans to Reduce Interstate Transport of Fine

Particulate Matter and Ozone," and comments to the January 7, 2011
Federal Register, "Notice of Data Availability for Federal
Implementation Plans to Reduce Interstate Transport of Fine Particulate
Matter and Ozone," one of the MDEQ's most critical concerns is "the
implementation timing for the proposed Interstate Transport Rules, where
Phase One begins in 2012 and Phase Two in 2014.  The MDEQ continues to
believe the shortened timeline for implementation of the Transport Rule
has adverse impacts on the states and sources affected." The MDEQ
requests the implementation of the FIP be delayed a minimum of 18
months, until states can finalize an approvable state implementation
plan (SIP) to effectively replace the FIP requirements.  [Michigan
Department of Environmental Quality (EPA-HQ-OAR-2009-0491-4606), p. 1.]

RESPONSE:  The comment is outside the scope of the SNPR because it
addresses issues that were within the scope of the request for comments
in the proposed Transport Rule, but not in the SNPR.  While EPA is not
reconsidering or reopening these issues we note that the issues were
previously addressed in sections VII.C and section X of the preamble to
the final Transport Rule, as well as in the Transport Rule Primary
Response to Comments document in the docket to this rulemaking
[EPA-HQ-OAR-2009-0491-4513].  Additionally, questions regarding the
legality of using a FIP were addressed in the preamble.  

Missouri

No comments received on this issue. 

Oklahoma

(A) EPA lacks authority to promulgate the proposed FIP for Oklahoma.

Comment:  The proposed FIP for Oklahoma is premature and inconsistent
with the structure of CAA.  In the Supplemental Proposal, EPA claims
that it has authority to include Oklahoma in the CSAPR by virtue of the
Agency's general finding on April 25, 2005 that all 50 states, including
Oklahoma, failed to submit State Implementation Plans ("SIPs") to
address the requirements of CAA Section 110(a)(2)(D)(i) with respect to
the 1997 ozone NAAQS.  In a Technical Support Document ("TSD") for the
Supplemental Proposal, EPA says that it "has not, subsequent to that
date, received or approved a SIP revision to correct the deficiency" for
Oklahoma.  Commenters disagree with EPA's characterization of the status
of Oklahoma's SIP.  First, EPA's claim misrepresents the status of
Oklahoma's SIP and ignores subsequent actions taken by Oklahoma to
implement the good neighbor requirements. It also ignores the reality
that EPA only in this proposal gave Oklahoma any notice about the
putative contribution to Allegan County's prospective and/or projected
nonattainment.  It would have been impossible for Oklahoma to have taken
steps previously to ensure that it is meeting its good neighbor
obligations because EPA had never identified the potential problem which
Oklahoma must address.

Also, contrary to EPA's assertion in the TSD, Oklahoma submitted an
interstate transport SIP on May 1, 2007 to address the requirements of
CAA Section 110(a)(2)(D)(i).  In the 2007 SIP, the State informed EPA as
follows:

“As required by Section 110(a)(2)(D)(i) of the Clean Air Act, the
State of Oklahoma submits to the USEPA that it has met the requirements
for the 8-hour ozone and PM2.5 National Ambient Air Quality Standards
("NAAQS") and does not significantly contribute to nonattainment or
interfere with maintenance of the NAAQS in another State.”

On December 5, 2007, the State supplemented its interstate transport SIP
with additional assessments and also provided EPA with the following
certification regarding the adequacy of its SIP under CAA Section
110(a)(1) and (2):

“We have evaluated our existing SIP for implementation of the 1997
8-hour ozone NAAQS to determine if it includes all the requirements in
Section 110(a)(1) and (2) of the CAA and is consistent with the guidance
provided to the Regional Air Division Directors from William T. Harnett,
Director, Air Quality Policy Division on October 2,2007.  We conclude
and certify that our SIP does meet these requirements.”

The guidance referred to in this certification is an EPA memorandum
titled "Guidance on SIP Elements Required Under Sections 110(a)(1) and
(2) for the 1997 8-hour Ozone and PM2.5 National Ambient Air Quality
Standards." The memorandum lists the basic elements that EPA thinks
should be part of a SIP and also says, "To the extent that existing SIPs
for ozone and particulate matter already meet these requirements, States
need only certify that fact to [EPA]."  Consistent with EPA's guidance,
Oklahoma certified in 2007 that the State's Interstate Transport SIP met
all of the requirements of CAA Section 110(a)(l) and (2), which includes
the interstate transport requirements related to NAAQS under CAA Section
110(a)(2)(D)(i)(I).  Since then, EPA has not approved or disapproved of
the State's conclusion in 2007 that its SIP meets the requirements of
CAA Section 110(a)(2)(D)(i)(I) related to the 1997 8-hour ozone NAAQS. 
EPA has approved certain portions of the Oklahoma SIP relating to
Prevention of Significant Deterioration ("PSD") requirements and has
also issued a proposal to partially approve and partially disapprove
other specific portions related to visibility.

EPA is now proposing to move directly to imposing a Federal
Implementation Plan ("FIP"), eliminating the State's opportunity to
receive notice of, and correct, a perceived deficiency in the

2007 SIP submittal as it relates to the requirements of CAA section
110(a)(2)(D)(i)(I). The EPA does not have authority under the CAA to
proceed with implementing a FIP prior to approving or disapproving the
State's 1997 SIP as it relates to the requirements of CAA section
110(a)(2)(D)(i)(I).  The CAA requires the EPA to act on a SIP revision
within 12 months of having received a completed submission (such SIP
submissions are deemed complete after 6 months unless EPA makes a
finding otherwise) see CAA section 110 (k)(l)(B), (k)(2). When a State
submits a SIP, the CAA allows EPA to issue a FIP within 2 years after
the Administrator disapproves the SIP in whole or part. Also, the CAA
gives the EPA the authority to issue a FIP within two years after
finding that a State failed to submit a SIP, or submitted a SIP that was
found to be incomplete. However, the authority to issue a FIP expires
two years after EPA issues a finding of failure to submit a SIP. The EPA
has missed that two year deadline here and therefore, lacks authority
under the CAA to issue the proposed FIP. Furthermore, where a State
submits a SIP after the issuance of a deficiency finding by the
Administrator, the EPA must act on the SIP submittal before issuing a
FIP.

Pursuant to 42 U.S.C. § 7410(c)(l)(B):

(1) The Administrator shall promulgate a Federal implementation plan at
any time within 2 years after the Administrator –

(A) finds that a State has failed to make a required submission or finds
that the plan or plan revision submitted by the State does not satisfy
the minimum criteria established under subsection(k)(l)(A) of this
section, or

(B) disapproves a State implementation plan submission in whole or part,
unless the State corrects the deficiency, and the Administrator approves
the plan or plan revision, before the Administrator promulgates such
Federal implementation plan.

In this Supplemental Proposal, the EPA has not followed the CAA mandate
requiring that prior to promulgating a FIP, the EPA must find that a
State has failed to make a required submission, or find that the plan or
plan revision submitted by the State does not satisfy minimum criteria,
or must disapprove a SIP submission in whole or part. As set forth
above, the EPA has not taken any action in regards to the May 1, 2007,
Interstate Transport SIP that addressed the requirements of CAA
1l0(a)(2)(D)(i). In addition, the EPA has not taken action to approve or
disapprove the December 2007 State of Oklahoma SIP revision and
certification that was submitted pursuant to an October 2007, EPA
guidance document. In that submission, Oklahoma certified (following the
dictates in the EPA Guidance Document) in December 2007 that Oklahoma's
Interstate Transport SIP met all the requirements of CAA section
11O(a)(I) and (2), which includes CAA section 110(a)(2)(D)(i)(I).
Therefore, the EPA does not have the authority under the CAA to issue
the FIP it is proposing in this Supplemental Rule under the CSAP
Program.  [Oklahoma Attorney General (EPA-HQ-OAR-2009-0491-4592), p.
3-6; Oklahoma Gas & Electric Company (OGE) (EPA-HQ-OAR-2009-0491-4590),
p. 3-5; AES Shady Point (EPA-HQ-OAR-2009-0491-4595), p. 3-4; Western
Farmers Electric Cooperative (EPA-HQ-OAR-2009-0491-4589), p. 5-6.]

RESPONSE:   For the reasons explained in section IV.C.2 of the preamble
to the final Transport Rule (76 FR 48208, August 8, 2011) and in the TSD
entitled “Status of CAA 110(a)(2)(D)(i)(I) SIPs Supplemental Rule”
TSD (December 2011), EPA has the legal obligation to promulgate all of
the FIPs in this rule.  This conclusion is based on the plain language
of the Clean Air Act as applied to the specific circumstances of each
state.  

Section 110(c)(1) of the Act requires EPA to promulgate FIPs within two
years after finding that a state has failed to make a required SIP
submission or disapproving a SIP.  Specifically, section 110(c)(1)
states that the Administrator shall promulgate a Federal Implementation
Plan within two years after the Administrator "(A) finds that a State
has failed to make a required submission or finds that the plan or plan
revision submitted by the State does not satisfy the minimum criteria
established under subsection (k)(1)(A) of this section or (B)
disapproves a State implementation plan submission in whole or in part."
 42 U.S.C. § 7410 (c)(1)(A) & (B).  EPA is relieved of the obligation
to promulgate a FIP only if the state corrects the deficiency and EPA
approves the SIP before promulgating a FIP.  42 U.S.C. § 7410(c)(1). 
The Act does not create any exceptions to this rule.  EPA has not
approved a SIP for Oklahoma that addresses this requirement; rather, EPA
has proposed to disapprove Oklahoma’s SIP submittal that addresses
110(a)(2)(D)(i)(I).  EPA is promulgating FIPs only in those
circumstances where the Clean Air Act explicitly provides that it shall
do so. 

The commenter does not identify any provision that would give EPA
authority to alter the deadlines established in the statute or delay
implementation of the Transport Rule as requested.  In this case, the
Act explicitly provides that the SIP submission deadline runs from the
date of promulgation or revision of a NAAQS and the FIP clock starts
when EPA finds that a state has failed to make a required SIP submission
or EPA disapproves a SIP submission.

Nothing in the Act requires EPA, where it has already made findings of
failure to submit or disapproved SIP submissions, to promulgate a SIP
Call to give the states another opportunity to promulgate a SIP before
promulgating a FIP.  Section 110(a)(2)(D)(i)(I) SIP submittals are due 3
years after promulgation or revision of a NAAQS.  See 42 U.S.C. §
7410(a)(1).  Thus, 110(a)(2)(D)(i)(I) SIPs for the 1997 ozone NAAQS were
due in 2000.  While the statute gives EPA authority to prescribe a
shorter period of time for states to make SIP submissions, it does not
give EPA authority to extend the 3-year deadline established by statute.
 See 42 U.S.C. § 7410(a)(1).  Moreover, there is no requirement that
EPA promulgate a rule or issue guidance regarding the specific
requirements of section 110(a)(2)(D)(i)(I) in advance of the SIP
submittal deadline.

The plain language of section 110(c)(1) of the Act provides that EPA
shall promulgate a FIP at any time within 2 years after the
Administrator finds that a state has failed to make a required
submission or after disapproving a state implementation plan submission
in whole or part, unless the state corrects the deficiency, and the
Administrator approves the plan or plan revision, before the
promulgation of a FIP. See 42 U.S.C. § 7410(c)(1).  Far more than 2
years has now elapsed since EPA first made these findings with respect
to the 1997 ozone and 1997 PM2.5 NAAQS.  There is nothing in the plain
language reading of the Act that says that EPA’s obligation
“expires” at the end of the 2 years.

Finally, nothing in the Act requires EPA, where it has already made
findings of failure to submit or disapproved SIP submissions, to
promulgate a SIP Call to give the states another opportunity to
promulgate a SIP before promulgating a FIP.  Section 110(a)(2)(D)(i)(I)
SIP submittals are due 3 years after promulgation or revision of a
NAAQS.  See 42 U.S.C. § 7410(a)(1).  While the statute gives EPA
authority to prescribe a shorter period of time for states to make SIP
submissions, it does not give EPA authority to extend the 3-year
deadline established by statute.  See 42 U.S.C. § 7410(a)(1).

Comment:  The EPA's Supplemental Rule Proposal for Oklahoma under the
CSAP Program intrudes upon the role generally left to the States under
the CAA. In Train v. Natural Resources Defense Council, Inc., 421 U.S.
60 (1975), the United States Supreme Court found that although the CAA
plainly charges the EPA with the responsibility for setting the national
ambient air quality standards, the Act, just as plainly, relegates the
EPA to a secondary role in the process of determining and enforcing the
specific, source-by-source emission limitations that are necessary if
the national standards are to be met. According to the Court, the Act
gives the agency no authority to question the wisdom of a State's choice
of emission limitations if they are part of a plan that satisfies the
standards of section 110(a)(2), and the agency may devise and promulgate
a specific plan of its own only if a State fails to submit an
implementation plan that satisfies those standards. To this point, the
Court stated: "So long as the ultimate effect of a State's choice of
emission limitations is compliance with the national standards for
ambient air, the State is at liberty to adopt whatever mix of emission
limitations it deems best suited to its particular situation." Id. at
79.

The CAA then "establishes a partnership between EPA and the states for
the attainment and maintenance of national air quality goals." See
Natural Res. Def  Council, Inc. v. Browner, 57 F.3d 1122, 1123 (D.C.
Cir. 1995). "Air pollution prevention ... at its source is the primary
responsibility of States and local governments...." 42 U.S.C. §
7401(a)(3). Congress "carefully balanced State and national interests by
providing for a fair and open process in which State and local
governments, and the people they represent, will be free to carry out
the reasoned weighing of environmental and economic goals and needs."

In the current Supplemental Proposal, the EPA is proposing a FIP in
Oklahoma regarding summertime NOX reductions under the CSAP ozone-season
control program. The EPA is proposing to promulgate this FIP despite the
fact that Oklahoma has submitted a SIP that meets the requirements of
the CAA and despite the fact that EPA has not taken any action on
Oklahoma's SIP, as required in the CAA. Such action by the EPA is an
impermissible intrusion upon the authority reserved to Oklahoma and
exceeds the authority granted to the EPA under the CAA.

Another commenter makes similar arguments, citing the same case law. 
The commenter also notes that The Act expressly allocated to states the
initial responsibility for determining the manner in which air quality
standards are to be achieved. Train v. Natural Resources Defense
Counsel, 421 U.S. 60,64 (1975). EPA "may devise and promulgate a
specific plan of its own only if a State fails to submit an
implementation plan which satisfies those standards." Commonwealth of
Virginia, 108 F.3d at 1482.  The CAA is an experiment in federalism "and
the EPA may not run roughshod over the procedural prerogatives that the
Act has reserved to the states ... " Id at 1408 (internal citation
omitted).  [Oklahoma Attorney General (EPA-HQ-OAR-2009-0491-4592), p.
6-8; Western Farmers Electric Cooperative (EPA-HQ-OAR-2009-0491-4589),
p. 4.]

RESPONSE:  For the reasons explained in section IV.C.2 of the preamble
to the final Transport Rule (76 FR 48208, August 8, 2011) and in the TSD
entitled “Status of CAA 110(a)(2)(D)(i)(I) SIPs Supplemental Rule”
TSD (December 2011), EPA has the legal obligation to promulgate all of
the FIPs in this rule.  This conclusion is based on the plain language
of the Clean Air Act as applied to the specific circumstances of each
state.  

Section 110(c)(1) of the Act requires EPA to promulgate FIPs within two
years after finding that a state has failed to make a required SIP
submission or disapproving a SIP.  Specifically, section 110(c)(1)
states that the Administrator shall promulgate a Federal Implementation
Plan within two years after the Administrator "(A) finds that a State
has failed to make a required submission or finds that the plan or plan
revision submitted by the State does not satisfy the minimum criteria
established under subsection (k)(1)(A) of this section or (B)
disapproves a State implementation plan submission in whole or in part."
 42 U.S.C. § 7410 (c)(1)(A) & (B).  EPA is relieved of the obligation
to promulgate a FIP only if the state corrects the deficiency and EPA
approves the SIP before promulgating a FIP.  42 U.S.C. § 7410(c)(1). 
The Act does not create any exceptions to this rule.  EPA has not
approved a SIP for Oklahoma that addresses this requirement; rather, EPA
has proposed to disapprove Oklahoma’s SIP submittal that addresses
110(a)(2)(D)(i)(I).  EPA is promulgating FIPs only in those
circumstances where the Clean Air Act explicitly provides that it shall
do so.  

EPA has considered the specific circumstances of each state in
determining whether it has an obligation to promulgate a FIP for that
state with respect to a particular NAAQS.  More detailed information
regarding Oklahoma-specific actions taken by EPA and the status of its
110(a)(2)(D)(i)(I) SIP appears in the TSD entitled, Status of CAA
110(a)(2)(D)(i)(I) SIPs Supplemental Rule TSD (December 2011).  The TSD
explains the current status of the Oklahoma 110(a)(2)(D)(i)(I) SIP.

Further, the cases cited by the commenter do not support the commenter's
conclusion that it is improper for EPA to promulgate the FIPs in this
rule.  The cases discussing the system of cooperative federalism
established by the Clean Air Act explain both the primary role given to
states with regard to the protection of air quality and EPA’s
obligation to promulgate federal plans “if a State fails to submit an
implementation plan which satisfies [the standards of 110(a)(2)]”. 
Train, 421 U.S. at 79.  

Comment:  Commenter incorporates by reference comments filed by Oklahoma
Gas and Electric and Western Farmers Electric Cooperative.  [Oklahoma
Utility Group (EPA-HQ-OAR-2009-0491-4591), p. 3.]

RESPONSE:  See above responses in this section.

Wisconsin

No comments received on this issue. 

Ozone Season NOX Emission Budgets 

3.1. General

No comments received on this issue. 

3.2 Iowa

No comments received on this issue. 

3.3 Kansas

 (A) Commenters(s) raise issues with the budgets established for Kansas.

Comment:  The allowances allocated to sources in these rules are
severely short for the SO2 and NOX programs. This presents difficult
compliance challenges for Empire District as well as other sources in
Kansas that have either maintained compliance with previous rules via
banked or purchased allowances and have not previously been subject to
the Clean Air Interstate Rule and subsequently do not have the needed
controls installed. 

Empire District compared 2010 SO2 actual emissions with the 2012 SO2
allowance allocations and the initial analysis shows that SO2 Group 1
allocations will be approximately 600,000 tons short and SO2 Group 2, of
which the state of Kansas belongs, allocations will be short
approximately 300,000 tons. Since it is apparent there will be a
shortage of SO2 allowances in 2012 available to purchase for compliance,
it is critical that sources have more time to install the controls
needed to comply with the CSAPR and supplemental rule to the CSAPR. 

According to the technical supporting document for Final Transport Rule
analysis in EPA’s application of the Integrated Planning Model,
Appendix 3-3 New Source Review settlements in Base Case v 4.10_F
Transport (05-16-11), the state of Kansas SO2 budgets were developed
without considering the allowance restrictions contained in the consent
agreement for Westar’s Jeffrey Energy Center. Because of the consent
agreement restrictions there will be approximately 13,700 SO2 allowances
unavailable for trading. This egregious error severely restricts Kansas
utilities ability to comply with the CSAPR via the SO2 allowance trading
program and represents approximately 34 percent of the total state SO2
budget. 

Empire District questions how many other consent agreements were ignored
or erroneously missed. Furthermore, what is the true number of total
allowances available across Group 1 and Group 2 if EPA considers all
consent agreements in its respective state budgets? This is a serious
issue that must be addressed prior to the commencement of the CSAPR
going into effect in January 2012.  [Empire District Electric Company
(EPA-HQ-OAR-2009-0491-4581), p. 2]

RESPONSE:  These comments are outside the scope of this rulemaking.  EPA
is addressing the role of consent decrees in unit-level allowance
allocation in a separate rulemaking.  See Revisions to Federal
Implementation Plans to Reduce Interstate Transport of Fine Particulate
Matter and Ozone, 76 FR 63860, October 14, 2011. 

3.4 Michigan

No comments received on this issue. 

3.5 Missouri

No comments received on this issue. 

Oklahoma

(A)  EPA's Supplemental Proposal requires an overly drastic reduction in
ozone season NOX emissions for commenter and the State of Oklahoma
generally.

Comment:  EPA's Supplemental Proposal marks a significant change in the
Agency's treatment of Oklahoma compared to the 2010 Proposed interstate
Transport Rule, and also would require a drastic reduction in the
State's actual ozone season NOX emissions.  Commenter include a table
that summarizes the differences between the ozone season NOX emissions
budget for Oklahoma in the Supplemental Proposal, the same budget in the
2010 Proposed Rule, and the State's actual NOX emissions from EGUs
during the 2010 ozone season.

As shown in the table, the Supplemental Proposal assigns to Oklahoma a
statewide ozone season NOX emissions budget of just 21,835 tons per
year, which represents a reduction of more than 15,000 tons (roughly
41%) from the state's budget in the 2010 Proposed Transport Rule.
Compared to 2010 actual emissions, this budget requires a reduction of
more than 13,000 tons (37%) statewide. The budget proposed for Oklahoma
is inadequate because EPA relies on a collection of flawed assumptions
about unit shutdowns and curtailments.  For example, OG&E's Mustang
units are allocated no allowances and are therefore assumed to retire. 
OG&E does not expect shutdown of these units to be feasible while still
meeting consumer demand.

For OG&E's units, the total emission reduction required under the
Proposal would be roughly 6,000 tons or 38.6% percent less than actual
2010 emissions and 6,800 tons or 41.5% percent less than the allowances
provided to OG&E's units under the 2010 Proposed Rule. Certain OG&E
units would bear the bulk of this impact. As an example, OG&E's Muskogee
and Sooner units (the base load units of OG&E's generation) are assumed
to achieve a 50% reduction in NO, emissions during the ozone season.
Simply put, this would mean that these base load units could only run
during 50% of the ozone season, which is the hot period in Oklahoma.
Based on the demand for recent operations, OG&E believes that these
reductions are not feasible.

EPA claims that the drastic reductions in ozone season NOX emissions
called for by the Supplemental Proposal can be achieved by installing
emission controls, or by purchasing allowances to cover emissions. It is
not realistic, however, to expect EGUs to install controls prior to the
proposed compliance deadline of May I, 2012. Nor has EPA accounted for
the likelihood of allowance shortfalls in light of the drastic emission
reductions that EGUs would be required to achieve by May 2012.
Furthermore, as discussed in more detail in Section IV, the compliance
schedule may not allow enough time for OG&E to purchase power to replace
what would be lost through curtailments.  [Oklahoma Gas & Electric
Company (OGE) (EPA-HQ-OAR-2009-0491-4590), p. 6-8; Oklahoma Utility
Group (EPA-HQ-OAR-2009-0491-4591), p. 5.]

RESPONSE:  See the preamble to this final rule, section III.B.iv. and
section III.E.

Comment:  Oklahoma’s ozone-season NOX budget is unreasonably low. In
the proposed rulemaking for the Transport Rule (published in the Federal
Register on August 2, 2010), the ozone-season NOX budget for the state
of Oklahoma was set at 37,087 tons (Table IV.E-2 of the preamble to the
proposed rule, 75 FR 45291).  In the SNPR, the ozone-season NOX budget
for the state of Oklahoma was set at 21,835 tons (Table I.C-1 of the
SNPR, 76 FR 40666).  Between the issuance of the proposed rule in 2010
and the issuance of the SNPR in 2011, EPA reduced the state budget by
41.1%. In the proposed rule, the state budgets were set using a process
that started with a baseline developed using the Integrated Planning
Model (IPM), based on historic emissions data and projections about
future growth in demand for electricity.  After setting the emissions
baseline, EPA used an air quality assessment tool (AQAT) and cost curves
to determine the amount of NOX reduction that would be required to
comply with the “good neighbor” requirements of the Clean Air Act.
Using this method, EPA set the state ozone-season NOX budget at 37,087
tons for Oklahoma.  Despite some concerns about some aspects of the
process, we believed the state budget established in the proposed rule
to be fair and achievable.  In the final rule and in the SNPR, EPA made
a radical and previously unannounced departure from the previous method.
 In the preamble for the final rule, EPA stated that, based on emissions
inventory data for 2009, they were changing the ozone-season emissions
baseline.  In the final rule, EPA acknowledged this change, but did not
provide a clear description of the process.  Was the new baseline based
almost exclusively on the 2009 emissions inventory data?  Did the EPA
take into consideration the reduced demand associated with the recession
as mentioned in section IV?  How much did the IPM analysis performed
previously factor into the new state budgets?  The criteria used to set
the new baseline remain unclear. We believe this change in methodology
fails to meet the standards for prudent rulemaking and that, by failing
to re-propose the rule and accept comments on this aspect of the rule,
EPA has acted in an arbitrary and capricious manner.  [Oklahoma
Department of Environmental Quality (EPA-HQ-OAR-2009-0491-4593), p.
5-6.]

RESPONSE:   First, EPA has finalized in this action an ozone-season NOX
budget for the state of Oklahoma that is significantly larger than the
budget in the proposed Supplemental rulemaking.  This higher budget for
2012 addresses the commenter’s concerns about compliance with the 2012
proposed budget.  In regards to questions about 2009 data informing the
baseline, the 2009 data was one of several variables that influenced the
final rule emissions baseline for 2012.  For instance, lower natural gas
prices are another variable that were updated that would also impact the
baseline between proposal and final.  EPA did consider the recessionary
impact of 2009 on emissions levels.  We did not use emission levels in
our modeling, only emission rates — which are minimally impacted by
any recessionary impacts.

Comment:  The Oklahoma emission budget is not based on accurate data and
operational  assumptions. EPA may not impose requirements on sources or
states without authorization by Congress through the CAA. See, e.g.,
Michigan v. EPA, 268 F.3d 1075, 1081 (D.C. Cir. 2001) ("It is axiomatic
that an administrative agency's power to promulgate legislative
regulations is limited to the authority delegated by Congress.")
(quoting Bowen v. Georgetown Univ. Hosp., 488 U.S. 204, 208 (1988». EPA
has not demonstrated that the severely restrictive budget proposed for
Oklahoma would achieve only those emissions reductions required by
Section 110 of the Act. Emissions reductions beyond what is required to
eliminate significant contribution to downwind nonattainment are an
abuse of EPA discretion and in excess of the Agency's authority under
the Act. 

The unit-specific data and operational assumptions relied on by EPA to
establish the

State emissions budget contains numerous inaccuracies.  Comments on the
2010 Proposal submitted by the Oklahoma Department of Environmental
Quality ("ODEQ") identified a significant number of errors in EPA's
Integrated Planning Model ("IPM") files. See Docket No.
EPA-HQ-OAR-2009-049I -2662. ("The IPM model run results are very
erratic, retiring both coal and natural gas facilities that have no
plans to retire, increasing and decreasing electricity production at
various facilities, and adding solar power to Oklahoma when none has
been planned.")  Similarly, WFEC notified EPA that the IPM files
contained no heat input data for units 4, 5, and 6 at its Anadarko
facility.  See Docket No. EPA-HQ-OAR-2009-0491-2642.  The heat input of
EGUs within a state play a crucial role in EPA's methodology for
establishing the state's emissions budget.  EPA's failure to correct
these errors "bespeaks a 'let them eat cake' attitude that ill-becomes
an administrative agency whose obligation to the public it serves is
discharged if only it avoids being arbitrary and capricious." Chemical
Mfrs. Ass'n v. EPA, 28 F.3d at 1266. The emissions budget for the State
of Oklahoma cannot accurately reflect emissions reductions necessary to
eliminate only those emissions that significantly contribute to downwind
pollution because it is not based on the most accurate data available to
the Agency.

EPA cannot impose emissions reductions beyond those required to address
Oklahoma's significant contribution to downwind pollution.  In North
Carolina v. EPA, the D.C. Circuit rejected EPA's attempt to require some
states "to exceed the mark" in reducing upwind emissions because the Act
gave "EPA no authority to force an upwind state to share the burden of
reducing other upwind states' emissions." 531 F.3d 896, 921.  Because
the state budgets at issue in North Carolina "shifted the burden of
emissions reductions," the Court invalidated the budgets as arbitrary
and capricious.  Id In CSAPR, EPA claims that the methodology used to
set state emissions budgets evaluates only cost and state-specific
contributions to downwind pollution. 76 Fed. Reg. 48,211-12.  EPA
completely disregards the fact that any methodology would produce
inaccurate emissions budgets if fed inaccurate data and assumptions. 
Because the emissions budget proposed for Oklahoma is based on
inaccurate data, modeling and assumptions, there is every reason to
believe that it too "exceeds the mark."  [Western Farmers Electric
Cooperative (EPA-HQ-OAR-2009-0491-4589), p. 9-10.]

RESPONSE:   Regarding the “solar” issue, EPA is not aware of any
projected new solar in the IPM outputs, and the commenter does not
specify what solar in the IPM outputs they are referring to.  There are
only 7.5 MW of Oklahoma solar generating capacity (or about 0.03% of
more than 20,000 MW of total Oklahoma generating capacity) in the NEEDS
input files, reflecting what is reported to be recently installed
Southern Great Plains Commercial Solar.  However, this appears to
already exist and the source did not comment on its existence when
depicted in the proposed modeling, so this is unlikely what commenter is
addressing.  Moreover, the relative size of these existing solar units
would not have any meaningful impact on emissions modeled at the state
level.  In regards to IPM projections of unit level heat input, they are
just that – projections of what would likely be the least-cost
response to the state budget or cost threshold modeled.  EPA was very
clear about how IPM does not reflect how the unit may ultimately decide
to operate.  CSAPR is not a unit-level control program and the
allocations are not based on its projections.  Therefore, the projection
has no bearing on how the unit chooses to operate.  The commenter’s
suggestion that simply substituting historic data from a period when a
source was not subject to an ozone season emission reductions program
for a period when it is subject to an emission reductions program is not
reasonable and fundamentally inconsistent with the notion of emission
reductions.

In regards to the viability of the state budget in 2012, this final rule
addresses those concerns by increasing the budget to a level where no
combustion controls are assumed.

 (B) There are errors in EPA’s methodology as applied to calculation
of Oklahoma’s budget.

Comment:  DEQ’s analysis of the application of EPA’s methodologies
for determining state budgets and allocations to individual units
demonstrates clear errors were made in determining Oklahoma’s
allowances. By changing the method of calculation for the state
allowance budgets from using the IPM modeling to a reliance on 2009
emissions inventory data, EPA made a number of errors that have resulted
in assigning an ozone-season NOX allowance budget for the State of
Oklahoma which is lower than is warranted. The Excel spreadsheet titled
“Proposed CSAPR Unit Level Allocations under the SNPR and Underlying
Data” provides baseline data used to calculate the state budgets. A
number of units are listed in the spreadsheet, because they will be
subject to the Cross-State Air Pollution Rule even though they are not
currently subject to the Acid Rain Program (ARP). These units do not
report emissions to the Clean Air Markets Division (CAMD), and their
emissions are not included in the CAMD database. EPA obtained some
historical emissions data for these units, and those data are included
in the spreadsheet. However, for the years 2008, 2009, and 2010, many of
these units appear in the table as having zero emissions for the ozone
season. The absence of any emissions data for these units has the effect
of lowering the baseline from which EPA set the state budget. The agency
provides a in the comment letter to provide a comparison of EPA and DEQ
data for the total emissions from all subject units for the 2008 through
2010 ozone seasons.

Two key observations are noteworthy: (1) due to the recession, the 2009
ozone season represented an anomalously low year for ozone-season
emissions and (2) the failure to capture year-to-year trends in
emissions from units not subject to the ARP yielded a state budget
baseline that is nearly 3,000 tons too low. In addition, we calculate
that utilities operating in Oklahoma will need to retrofit 43 individual
combustion units with low-NOX burners to comply with the requirements of
the new rule. This number is substantially larger than the 30 units in a
four-state area predicted in EPA’s TSD, “Installation Timing for Low
NOX Burners (LNB).” To remedy these failings, we recommend that EPA
increase the state ozone-season NOX allowance budget by 1,500 tons and
that the compliance date for units operating in Oklahoma be pushed back
to the 2015 ozone season. 

We believe that our request for this increase in the state budget is
justified for four reasons: (1) the 2009 emissions estimate data are
unrepresentative due to the recession, (2) the failure to include
non-ARP sources in the budget baseline lead to a baseline nearly 3,000
tons below actual values, (3) assuming a 50% control efficiency for
low-NOX burners applied to a baseline elevated by 3,000 tons would yield
an additional 1,500 tons of allowances for the state budget, and (4)
this increase in the budget still retains over 90% of the reductions
from the initial proposal. We believe the fourth point to be
particularly salient, and we would ask EPA to use their AQAT to
determine whether, with a state ozone-season NOX budget of 23,335 tons,
Oklahoma would be expected to make a significant contribution to the
ozone levels in Allegan Co., Michigan.  [Oklahoma Department of
Environmental Quality (EPA-HQ-OAR-2009-0491-4593), p. 3-4.]

RESPONSE:  The 2012 state budget has been increased.  See section
III.B.iv and section III.E. in the preamble of the final rule.  EPA has
finalized an ozone-season NOX budget for the state of Oklahoma that is
significantly larger than the budget in the proposed Supplemental.  This
higher budget for 2012 addresses the commenter’s concerns about
compliance with the 2012 proposed budget.  In regards to questions about
2009 data informing the baseline, the 2009 data was one of several
variables that influenced the final rule emissions baseline for 2012. 
For instance, lower natural gas prices are another variable that were
updated  that would also impact the baseline between proposal and final.
 EPA did consider the recessionary impact of 2009 on emissions levels. 
We did not use emissions levels in our modeling, only emission rates —
which are minimally impacted by any recessionary impacts.

3.7 Wisconsin

No comments received on this issue. 

Proposed Unit-Level Ozone Season NOX Allowance Allocations (data inputs
& new unit set-asides)

4.1 General

No comments received on this issue. 

4.2 Iowa

No comments received on this issue. 

4.3 Kansas

No comments received on this issue. 

4.4 Michigan

 (A) Commenter disagrees with set-aside for Indian Country in Michigan.

Comment:  The MDEQ believes that the proposed new unit set-asides of 26
allowances for 2012 and 25 allowances for 2014 are not necessary and
should be returned to the allocation pools for the electric generating
units main allowances. At this time and into the foreseeable future, the
MDEQ is not aware of any proposed power plant installations on Michigan
tribal lands.

In addition, the EPA has indicated that even if a state received federal
approval for a SIP, the EPA will continue to allocate allowances to
Indian Country units.

The MDEQ believes it is more appropriate for states to have the ability
to process any new source requests and allowance determinations for the
entire state, including tribal lands, rather than having a split
allocation system once the state's SIP is approved.  [Michigan
Department of Environmental Quality (EPA-HQ-OAR-2009-0491-4606), p.
1-2.]

RESPONSE:  EPA retains administration of new unit set-asides designated
for Indian country because EPA, rather than the states, has the
authority and responsibility of administering the Transport Rule with
regard to new units that locate in Indian country.  Under the FIP, any
allowances from the Indian country new unit set-aside that are not
allocated in a given control period are redistributed to the state’s
new unit set-aside for allocation to new units not in Indian country. 
Any allowances not allocated from the state’s new unit set-aside are
redistributed to existing units, which are not in Indian country.  
Under a SIP, the state may determine how unallocated allowances in the
Indian country set-aside are to be allocated within the state.  For
details, see section VII.D.2.e of the preamble of the final Transport
Rule and see section XX.D of the Transport Rule Primary Response to
Comments document [EPA-HQ-OAR-2009-0491-4513].

4.5 Missouri

No comments received on this issue. 

4.6 Oklahoma

 (A) Commenter(s) state that facility-specific allowance allocations are
based on flawed and/or incomplete data.

Comment:  The data used in the proposed allowance allocations in the
NODA for the proposed rule and supporting technical information are in
error. The allowance allocations in the Technical Support Documents are
based on flawed and incomplete historical data and emissions level model
inputs. These errors were identified in previous comments regarding this
rulemaking by the AES Corporation, but have thus far not been corrected.

The NOX allocations provided for the AES Shady Point facility do not
reflect historical actual NOX emissions.  As indicated in the Unit
Allocation Table in the Administrative Record, the source of the NOX
data for AES Shady Point is Energy Information Administration (EIA)
data. However, in years 2005 - 2006, for the AES Shady Point facility,
despite non zero data entry entries for heat input, data entries for NOX
emissions were omitted, and apparently treated as zero within the
allocation methodology.

This omission must be a clerical error since it cannot be assumed that
the facility generated heat without generating NOX.  The facility was
unquestionably in existence and operating at the time. More
specifically, the State of Oklahoma authorized NOX emissions for this
facility (Permit Number) and included AES Shady Point facility emissions
in its State Implementation Plan (SIP) emissions inventory. For example,
for calendar year 2009, the facility reported a total of 1315.7 tons NOX
(2009 Point Source Emissions, ODEQ, August 22, 2011,
http://www.deq.state.ok.us/AQDnew/emissions/index.htm).  Based on an
engineering review utilizing actual facility NOX emission data (in place
of the omitted data) and the proposed USEPA methodology, these data
omissions appear to yield erroneously low NOX allocations.  A copy of
these engineering calculations including proposed corrected NOX
allocations are attached.

AES asks that the above clerical error be corrected and the facility's
actual emissions be recognized in the allocation methodology. If the
clerical error is not corrected AES will be significantly and unfairly
impacted. While NOX emissions from other company's facilities would be
recognized-the NOX emissions from AES would not-resulting in a
significant competitive disadvantage to AES. Such an unfair  application
of the law would appear to be both arbitrary and capricious, and a
Federal due process violation.  [AES Shady Point
(EPA-HQ-OAR-2009-0491-4580), p. 1.]

RESPONSE:  The AES Shady Point facility’s historic data has been
updated in this final rulemaking to reflect the data submitted during
the comment period and described above.  The final unit level
allocations reflect these updates and can be seen in the allocation
Technical Support Document, “Final Unit-Level Ozone Season NOX
Allowance Allocations to Existing Units in 5 States:  Supplemental Final
Rule,” in the docket to this rulemaking.

4.7 Wisconsin

No comments received on this issue. 

Other In-Scope Issues 

5.1 Implementation Issues (e.g., timing for installation of controls)

Commenter supports EPA’s compliance timetable.

Comment:  Sierra Club also supports EPA’s proposal to implement the
NOX emissions reductions on the same schedule for the 20 other states
subject to the transport rule.  Each state has been well aware that it
could be subject to ozone season NOX emission requirements.  Four of the
states were subject to the ozone season NOX requirements in the Clean
Air Interstate Rule, and three were subject to 2010 cross state
proposal.

Another compelling reason why these six states must start making
reductions immediately is that the proposed cross state rule only aims
to achieve compliance with 1997 ozone standard of .08 ppm. The standard
was decreased in 2008 to .075 ppm and EPA has proposed to lower the
standard even further to .06 to .07 ppm.  Indeed, Sierra Club’s
position continues to be that the transport rule, and this supplement,
is deficient because it fails to prevent significant contribution to
nonattainment of 2008 ozone NAAQs.  [Sierra Club
(EPA-HQ-OAR-2009-0491-4587), p. 2-3.]

RESPONSE:  EPA concurs with the comments supporting the application of
the same schedule under the Transport Rule ozone season NOX program to
Iowa, Michigan, Missouri, Oklahoma, and Wisconsin as to other states. 
To the extent the comment addresses the 2008 Ozone NAAQS, the comment is
outside the scope of this rulemaking, which addresses the 1997 ozone
NAAQS.

Comment:  We also support EPA’s proposal to implement the ozone season
NOX emission requirements for the six states subject to the SNPR
(including SIP submissions) on the same schedule as those requirements
for the 20 other states subject thereto, that is, beginning May 1, 2012.
Those six states and power plants located therein should be able to meet
this schedule, as each state has been subject to (or on notice that it
could be subject to) ozone season NOX emission requirements.  Of the six
states subject to the SNPR, four (Iowa, Michigan, Missouri and
Wisconsin) were subject to the ozone season NOX requirements in the
Clean Air Interstate Rule (“CAIR,” which CSAPR replaces), while
three (Kansas, Michigan, and Oklahoma) were subject to the 2010 CSAPR
proposal.  [Clean Air Task Force, et al. (EPA-HQ-OAR-2009-0491-4588), p.
3.]

RESPONSE:  EPA concurs with the comments supporting the application of
the same schedule under the Transport Rule ozone season NOX program to
Iowa, Michigan, Missouri, Oklahoma, and Wisconsin as to other states. 

Commenter(s) questions the feasibility and/or validity of the compliance
deadline.

Comment:  How does EPA expect the States to comply with this rule by the
2012 compliance date?  In the Fact Sheet for the Cross-State Air
Pollution Rule, EPA notes that “This action builds on more than
fifteen years of progress…”  The six States in the Supplemental Rule
will have approximately 6 months to prepare and implement a method of
compliance.   [Oklahoma Municipal Power Authority (OMPA)
(EPA-HQ-OAR-2009-0491-4573), p. 2.]

RESPONSE:  See the preamble of final Transport Rule, section VII.C and
the "Transport Rule Engineering Feasibility Response to Comments"
document in the docket to this rulemaking (Docket ID No.
EPA–HQ–OAR–2009–0491).  The Transport Rule does not apply
unit-specific requirements, thus provides flexibility for compliance
across units with various control capacities. While the NOX ozone season
program begins in May 2012, sources have until September 30, 2012 to
achieve emission reductions and can purchase allowances to cover
emissions until December 1, 2012 when they have to comply by holding
allowances sufficient to cover their ozone season NOX emissions.

Comment:   Empire District believes the short time frame whereby
affected sources must comply with the provisions of both the final CSAPR
as well as the proposed supplemental rule to the CSAPR is not realistic.
 Affected sources are allowed less than five months to comply with the
annual NOX and SO2 emissions requirements in the final CSAPR and allowed
less than nine months ( to comply with the ozone season NOX emission
requirements in the proposed supplemental rule to the CSAPR. 

In the proposed Transport Rule Empire District commented that during a
2010 study of our Asbury facility to determine the Air Quality Control
Systems (ACQS) needed for the future, including a scrubber, that the
actual projected timeline from project planning to project completion is
approximately 60 months. Using this as an example, it will take more
time than allowed for by these rules for sources to install the controls
that will be required for compliance.  [Empire District Electric Company
(EPA-HQ-OAR-2009-0491-4581), p. 1.]

RESPONSE:  See the preamble of final Transport Rule, section VII.C.2 and
the "Transport Rule Engineering Feasibility Response to Comments"
document in the docket to this rulemaking (Docket ID No.
EPA–HQ–OAR–2009–0491).  The Transport Rule does not apply
unit-specific requirements, thus provides flexibility for compliance
across units with various control capacities.

Comment:  EPA indicates that it expects to finalize this rulemaking on
or before November 1, 2011 which allows for only a six-month lead time
prior to the 2012 ozone season which commences on May 1, 2012. See 76
Fed. Reg. 40667, Section I(E). KCP&L has coal-fired units that are part
of what EPA describes as the “vast majority of covered sources already
have combustion controls installed”, but also has units that are part
of the “small number of sources will need to install combustion
controls to comply”. Id. KCP&L does not concur that for this later
group of sources it is “practical” to install combustion controls to
achieve compliance “within the time period” allowed. Id. 

Installation of these combustion controls on KCP&L’s coal-fired units
is a time-consuming, expensive process. Steps in this process include:
initial design, development of project specifications, procurement
(identification of bidders, solicitation, review of bids, and contract
negotiations), permitting including in some cases New Source Review
Presentation of significant Deterioration permits, regulatory approvals,
obtaining financing, final design, mobilization of the needed workforce,
construction, process tie-ins and process startup testing. These
activities cannot commence until a final rule is issued and the
compliance requirements known. KCP&L believes several of these steps
alone can exceed six months much less all the steps in total.  [Kansas
City Power & Light (EPA-HQ-OAR-2009-0491-4584), p. 2.]

RESPONSE:  See the preamble of final supplemental rule, section
III.B.iv.  Also see the preamble to the final Transport Rule, section
VII.C.2 and the "Transport Rule Engineering Feasibility Response to
Comments" document in the docket to this rulemaking (Docket ID No.
EPA–HQ–OAR–2009–0491).  The Transport Rule does not apply
unit-specific requirements, thus provides flexibility for compliance
across units with various control capacities.

Comment:   In the Supplemental Proposal, EPA says that it expects to
finalize a rule for Oklahoma and the other five states on or before
November 1, 2011, which would allow approximately six months of lead
time before the start of the 2012 ozone season. EPA claims that only "a
small number" of sources would need to install combustion controls in
order to comply with the CSAPR, and that this small number of required
installations can be achieved in just six months.

EPA's assumptions are incorrect for Oklahoma and the surrounding region.
To comply with the Supplemental Proposal (without considering other EPA
rulemakings on air emissions), one commenter anticipates that it would
need to install Low NOX Burners ("LNBs") on all five of its coal-fired
units, which represent a significant portion of the generating capacity
in Oklahoma. Another utility group commenter states broadly that it will
need to install controls on many of its affected units. EPA itself
estimates that LNBs will need to be installed on 30 coal-fired boilers,
mostly in Oklahoma and surrounding states. In Oklahoma alone, EPA
assumes that LNBs will be responsible for 9,900 tons of ozone season NO,
emission reductions in Oklahoma during 2012. Commenters therefore
disagree with EPA's notion that LNB retrofits will impact a "small
number" of units in the region.

Furthermore, it will not be possible to procure, permit, and install the
necessary controls on these units before May of 2012. Some commenters
estimate that it will need roughly three years from the date of an
appropriate final rule before the required controls could be operational
on the appropriate units. Overall, this process requires project
scoping, budgeting, engineering and design, bidding, permitting, outage
planning, equipment fabrication, delivery, and installation.

One utility group commenter estimates that it will need somewhere
between 2 ½ to 3 years, before the required controls can be installed
and are functionally operational. This is typical for this kind of
installation of air emission controls. It will take 12-18 months to bid
and acquire the necessary equipment. Assuming permits and regulatory
approvals can be obtained simultaneously with this procurement process,
the timing to apply for and receive the required permits and approvals
will require approximately 12-18 months. Another 12-18 months of
construction time may be needed once equipment, permits and approvals
are available for the controls to be operational. In some instances the
controls may require additional tuning over a period of four (4) to six
(6) months to achieve the designed reductions. Overall, this process
requires project scoping, budgeting, engineering and design, bidding,
permitting, outage planning, equipment fabrication, delivery, and
installation. While not included in this estimate, another timing
requirement will be to schedule and determine how the units will be
allowed to be taken off-line for the installation – whether during
scheduled outages or if other timing involved with the SPP approval.
This comprehensive planning and schedule coordination will require
additional time and may prevent the controls being installed as quickly
as they might without these important considerations for electricity
reliability on the grid.

Another commenter states that it is estimated that affected units would
need 12-18 months to acquire the necessary equipment and permits, and
that it is estimated that 21 months of construction time would be
necessary once equipment and permits are available.

EPA's two anecdotal instances of LNB installations in Texas and Georgia
from the two page TSD for this rulemaking do not support a six-month
schedule for installing LNBs on commenters’ units.  Two commenters
note that as an initial matter, the "similar schedule" mentioned in the
TSD for the Southern Company units was actually eight months, not six.
Furthermore, these examples do not account for the entire process
leading up to operation. Specification, selection, procurement, and
permitting are not discussed in one or both examples. Optimization is
also not taken into account. Reliant Energy, for example, installed LNBs
in May but then spent through September tuning the controls to achieve
performance guarantees.

One of the commenters notes that unit-specific characteristics
distinguish EPA's two examples from its units. The document cited for
Reliant Energy mentions additional boiler modifications that would be
required when burning Powder River Basin coal, which commenter uses in
its units. For the Southern Company units, the retrofit system consisted
of burners only and did not require the design or installation of
overfire air systems, as would commenter's units. The examples were
1990-2001 retrofits, whereas commenter would need to install controls
under different circumstances and at a time when contractor and
equipment availability will be at a premium. Commenter also differs from
EPA's two examples in that it will need to proceed on its units in
phases for demand reliability reasons. For all these reasons, EPA has no
basis for setting a compliance schedule based on two isolated examples
that are not comparable to LNB installation schedules for EGUs in
Oklahoma.

Some commenters also noted in conjunction with these examples cited by
EPA that, contrary to those examples, it is infeasible for WFEC to
install controls by May 1, 2012 that will be necessary to achieve the
emissions reductions required by the proposed unit-specific allowance
allocation for its units. Under the Proposal, WFEC must reduce its
overall emissions by 30 percent. Installation of dry low NOX burners on
Anadarko units 4, 5, and 6 is the quickest and most cost effective way
for WFEC to achieve the required reductions through controls. EPA makes
the unfounded assumption in its technical support documents that low NOX
burners can be purchased, installed, and tuned within 6 months. EPA's
assumption is based on vendor papers developed as marketing tools. These
papers are not an unbiased technical assessment representative of
industry practices. EPA cannot rely on promotional materials to
determine the feasibility of installing controls.  

WFEC has solicited bids for low NOX burners from several vendors. Simply
obtaining the equipment will take WFEC between 12 and 18 months. After
WFEC obtains the necessary equipment, there are numerous other factors
that will further delay installation of the controls. First, WFEC must
obtain approval from Southwest Power Pool ("SPP") before it takes the
units offline for an extended outage (it will take approximately six to
eight weeks to install the low NOX burners at each unit). SPP has
members in nine states; eight of those states are affected by either
CSAPR or the Proposal. EGUs in each of these states also will need to
install controls. Second, WFEC must undertake extended outages during
shoulder seasons. Due to increased demand, it cannot shut down a unit
for a significant amount of time during the winter and summer months.
Finally, WFEC must obtain the necessary permits before it begins
installation of the low NOX burners. The permitting process can take
between 12 and 24 months, depending on the potential affect of the
proposed project on emissions. Even if WFEC beings the process of
obtaining the necessary equipment and permits now, it would be
impossible for the company to install controls by the compliance
deadline for the first ozone-season NOX trading program.

Another commenter (GRDA) stated that it must reduce its overall NOX
emissions by over 50 percent. Installation of low NOX burners with over
fire air and computerized combustion controls is the most expedient and
technically established method for GRDA. To implement such a customized
solution requires substantially more time than the six (6) months
allowed by the Proposal. GRDA has carefully studied EPA's explanation of
its decision to allow Oklahoma utilities only six (6) months to
implement multiple and complex NOX reduction projects, and it is
apparent EPA's six month allowance is unreasonable, and not consistent
with EPA's own statement of facts. For example, EPA expressly
acknowledges on page 48281 of the published rule that "historical
projects suggest a more typical schedule would be 12 to 16 months for
the contractors' portion of the work.  A plant owner's project planning
and procurement work in advance of a contract award would typically
involve several additional months."  The coal fired units in Oklahoma
are the primary and most reliable base load providers of electricity. 
Of these, nine units totaling over 4,000 megawatts will require design
and installation of customized low NOX burner systems.  The installation
and startup time for such systems is a minimum of several months, and
electric system needs would necessitate that all nine Oklahoma units be
scheduled between March 20 - May 1, 2012.  Successfully mobilizing the
necessary experienced engineers and technicians for installation and
successful startup would be in grave doubt. Industry experience has
clearly shown that tuning of new low NOX systems often takes several
months. GRDA has consulted with a highly experienced engineering firm
(Black & Veatch), and also obtained preliminary conceptual proposals
from vendors of low NOX burner systems. Based on these consultations, an
achievable and prompt implementation schedule for customized and
effective low NOX burner systems in Oklahoma (and GRDA) needs to include
a number of elements totaling 36 months as outlined in a full detailed
list in the commenter’s letter.

Another commenter states that it will need roughly three years from the
date of an appropriate final rule before the required controls can be
installed and operation. This amount of time is typical for the kind of
work required, and other EGUs are likely to be in a similar position.
Therefore, in addition to revising the Supplemental Proposal to set
forth a reasonable degree of emission reductions, the Agency should
allow no less than three years from the date of a final rule for sources
to comply.  [Oklahoma Gas & Electric Company (OGE)
(EPA-HQ-OAR-2009-0491-4590), p. 8-9; Oklahoma Utility Group
(EPA-HQ-OAR-2009-0491-4591), p. 6-7; AES Shady Point
(EPA-HQ-OAR-2009-0491-4595), p. 5; Western Farmers Electric Cooperative
(EPA-HQ-OAR-2009-0491-4589), p. 10-11; Grand River Dam Authority
("GRDA") (EPA-HQ-OAR-2009-0491-4586), p. 6-8.]

RESPONSE:  See the preamble of final supplemental rule, section
III.B.iv.  Also see the preamble to the final Transport Rule, section
VII.C.2 and the "Transport Rule Engineering Feasibility Response to
Comments" document in the docket to this rulemaking (Docket ID No.
EPA–HQ–OAR–2009–0491).  The Transport Rule does not apply
unit-specific requirements, thus provides flexibility for compliance
across units with various control capacities.

Comment:  EPA's claim of two (2) low NOX burner projects requiring only
six (6) months is incorrectly assessed and should not be utilized for
the following reasons:

In the case cited by EPA (subscript 62) by Riley Power, this project
only included low NOX burners, and was specifically implemented without
the use of overfire air (OFA) or burner sizing changes. The actual NOX
emissions attained after this example project was only a pitiful .36
lb/MMBtu. In the case of GRDA, this .36 level is already surpassed and
GRDA must achieve a reduction to approximately 0.16 lb/MMBtu to meet
Proposal allowances. Clearly, much more sophisticated, and customized,
burner technology systems will be needed than this EPA example. Lastly,
this "paper" cited by EPA was obviously written by the vendor as a sales
promotional tool. This "paper" should not be relied on by EPA as
representing an independent and impartial assessment of the facts, and a
basis for issuing enforceable regulations.

In the case cited by EPA (subscript 63) by Foster Wheeler, this "paper"
is again obviously a sales promotional tool, and should not be relied on
by EPA as representing an independent and impartial assessment of the
facts, and a primary basis for establishing enforceable regulations. The
paper claims 16 weeks to engineer and fabricate low NOX burners, and 7
weeks to install them. However, no facts are presented that demonstrated
the project only took six months total and no mention was made of how
this base load unit was scheduled for its nearly two month outage. In
addition, the paper states "The system was fully operational in May of
2000 and was tuned by September to consistently achieve a NOX
performance guarantee". This statement positively shows an additional
four months were needed to actually get burners to work properly, which
is not unusual for low NOX systems. The "paper" states the design was
simply copied from a similar Foster Wheeler design and it is therefore
clear the engineering time was minimal. The paper states the burners
were for tangential style burners which are known to be easier to
achieve results than those at GRDA. Clearly, this case does not warrant
EPA establishing a six (6) month implementation regulation for the
Oklahoma base load coal units.  [Grand River Dam Authority ("GRDA")
(EPA-HQ-OAR-2009-0491-4586), p. 6-7.]

RESPONSE:  EPA is not basing this final rule’s 2012 state budget for
Oklahoma on the installation of new combustion controls by that control
period.  See the preamble of final supplemental rule, section III.B.iv. 
Also see the preamble to the final Transport Rule, section VII.C.2 and
the "Transport Rule Engineering Feasibility Response to Comments"
document in the docket to this rulemaking (Docket ID No.
EPA–HQ–OAR–2009–0491).  The Transport Rule does not apply
unit-specific requirements, thus provides flexibility for compliance
across units with various control capacities.  In addition, sources in
four of the states covered by this final rule (Iowa, Michigan, Missouri,
and Wisconsin) are already covered by the final Transport Rule’s
annual NOX program, which was finalized almost a year in advance of the
2012 ozone season; these sources should have already begun implementing
any desired NOX combustion controls well before finalization of this
rule.

Comment:  EPA'S referenced Technical Support Document (TSD)
"Installation Timing for Low NOX Burners (LNB)” states that EPA
"analyzed" the installation of NOX burners on 30 boilers in the four
affected states that are not in CAIR, with Oklahoma being one of these
four states.  This TSD is defective in several aspects.  First of all,
the TSD assumes the proposal will be finalized in June 2011 and burner
modifications will be performed during fall of 2011 and spring of 2012. 
However, the Proposal final rule is now planned for October 2011.  The
TSD was not updated to consider this four month delay of the rule. No
implementation of low NOX burners can be done in the fall of 2011 which
means all 30 boilers must be modified in the spring of 2012.  These 30
boilers are the primary base load electric generators for the involved
utilities. In particular, taking out the affected boilers in Oklahoma,
Kansas and Nebraska at the same time would be irresponsible and
threatens public safety and health.  EPA could have learned this with a
few simple phone calls, but no evidence is included this was done.  EPA
has apparently instead relied on a single telephone discussion with a
vendor, and two sales promotion "papers" issued by vendors.  No apparent
effort was made to obtain independent and non-biased assessment from a
competent engineering firm. There is no evidence that EPA contacted
utilities, the Southwest Power Pool or NERC. As such, EPA has failed to
meet its required consideration of such impacts on the electric energy
system. Secondly, the referenced paper by Foster Wheeler contains no
statement the entire project was done in a total of six months. In fact,
the paper admits that tuning to meet burner guarantees took four months
(during the critical summer months). Thirdly, both referenced examples
were clearly non-typical situations where minimal boiler modifications
and engineering was performed. EPA should not base their rule making on
such non-typical examples. Fourthly, EPA's overly "aggressive" six month
schedule will prevent GRDA and other Oklahoma utilities from obtaining
competitive and fair pricing, and is therefore not in the public
interest. Vendors will be able to charge exorbitant prices, overtime and
support fees. GRDA's ability to negotiate NOX performance guarantees
will be impaired and this is not in the public interest.

In summary, GRDA strongly objects to the six (6) month implementation
and requests EPA to utilize the 36 month implementation schedule
described earlier.  [Grand River Dam Authority ("GRDA")
(EPA-HQ-OAR-2009-0491-4586), p. 7-8.]

RESPONSE:  EPA is not basing this final rule’s 2012 state ozone season
budget for Oklahoma on the installation of new combustion controls by
that control period.  See the preamble of final supplemental rule,
section III.B.iv.  Also see the preamble to the final Transport Rule,
section VII.C.2 and the "Transport Rule Engineering Feasibility Response
to Comments" document in the docket to this rulemaking (Docket ID No.
EPA–HQ–OAR–2009–0491).  The Transport Rule does not apply
unit-specific requirements, thus provides flexibility for compliance
across units with various control capacities. 

Comment:  Even if EPA retains the allowance budget that is currently
proposed for Oklahoma, an initial compliance deadline of May 1, 2012 is
not reasonable and must be revised. The Agency should allow no less than
three years for utilities to evaluate generating options, obtain
allowances, arrange for the purchase of replacement power where
possible, and install controls to achieve the actual emission reductions
that will drive the trading program in Oklahoma.

Commenters state that a compliance schedule of three years is especially
appropriate in light of the fact that Oklahoma was not included in CAIR
and has relatively tenuous modeled impacts compared to other states. As
evidenced by the shifts in EPA's modeling and emission budgets for
Oklahoma over the years, affected sources in the State have faced
significant uncertainty regarding whether, and to what extent, they
would be subject to interstate transport requirements. This uncertainty
translated into Oklahoma sources having less notice of the requirements
than sources in other states.

One of the commenters also argues that EPA should consider that capital
expenditures for environmental controls cannot be included in customer
rates until state regulatory agencies authorize cost recovery. The
commenter notes that controls were not going to be necessary for its
units based on the 2010 Proposed Rule. As a final alternative, if EPA
retains the allowance budget and the May 1, 2012 deadline in the
Supplemental Proposal, EPA should at the very least delay application of
the assurance provisions until 2014 so that interstate trading can
provide more allowances to Oklahoma EGUs while the markets for instate
allowances develop. Otherwise, if faced with inadequate lead time for
installing controls and a shortfall of emission allowances, commenter
and other electric utilities would be forced to curtail operations to
avoid noncompliance with CSAPR. As noted, the commenter's coal-fired
units could need to curtail operations by roughly 50% during the ozone
season, compared to their operations in 2010, if sufficient allowances
are not available. EPA has not provided an adequate assessment of the
impacts on electric reliability from this level of curtailment. 
[Oklahoma Gas & Electric Company (OGE) (EPA-HQ-OAR-2009-0491-4590), p.
10; Oklahoma Utility Group (EPA-HQ-OAR-2009-0491-4591), p. 7.]

RESPONSE:  EPA has proposed to adjust the effective date of the
Transport Rule assurance provisions in a separate rulemaking (see 76 FR
63860, October 14, 2011).  EPA believes that this final rule’s 2012
state budget for Oklahoma represents feasible emission reductions in
that timeframe (see section III.B.iv of the preamble to this final
rule).  The Transport Rule does not apply unit-specific requirements,
thus provides flexibility for compliance across units with various
control capacities.  See comments in other areas of this Response to
Comment document regarding EPA’s assessment of current and future
market liquidity and allowance availability based on current reports
from the market place.

Comment:  The timeline for compliance with the rule’s requirements is
not realistic. The final rule (published in the Federal Register on
August 8, 2011) establishes a compliance schedule for both the PM-2.5
and ozone portions of the rule that makes sense only for those states
that were previously covered by the Clean Air Interstate Rule (CAIR).
The final CAIR was published in March of 2005, allowing utilities owning
or operating units subject to the rule years to come into compliance
with the cap-and-trade system set up by the rule. Key components of any
successful market-based program include (1) sending clear signals to the
market regarding the number of allowances which will be issued and (2)
allowing the companies owning and operating units subject to the rule
sufficient time to plan for future needs. DEQ recognizes that the
vacatur and associated holding by the Court of Appeals for the District
of Columbia Circuit put the EPA in a position which necessitated that
EPA set the compliance date for states already covered by CAIR such that
there would be no gap between compliance with CAIR and compliance with
CSAPR. DEQ appreciates that this was needed to retain the reductions in
emissions associated with the installation of new controls and other
changes required by CAIR. A seamless transition from CAIR to CSAPR makes
sense for facilities located in those states; however there is no
similar justification for the short time period between the issuance of
the final rule and the requirements for compliance for states like
Oklahoma that were not part of CAIR. Based on the state budgets
announced in the Proposed Transport rule, it appeared that EPA was
taking those concerns into account by establishing budgets that could
realistically be achieved by the non-CAIR states in general and Oklahoma
in particular.

However, if the SNPR is finalized by the target date of October 1, 2011,
that will only allow utilities seven months to plan, acquire capital,
design, and install the control systems that will be required to allow
those units to operate in compliance with the rule. Even the EPA’s own
Technical Support Document (TSD), “Installation Timing for Low NOX
Burners,” (issued July 2010), assumes that there are only 30 boilers
located in the four states not already covered by CAIR and that
utilities would be able to install low NOX burners on those boilers if
EPA was able to finalize the rule by June 2011. EPA has missed that
target. The rule covering the six states (including Oklahoma) will not
likely be finalized until October of 2011; therefore, even EPA’s own
documents show the false premises upon which the assumption of a smooth
transition to compliance with CSAPR were constructed.

For Oklahoma utilities who were first excluded from CAIR and then led to
believe based on the allowances in the proposed Transport Rule that
compliance could reasonably be achieved, seven months is not enough time
to accomplish the required measures based on the new timeline and new
allowances in the SNPR. First, before installation can even begin,
utilities will have to have the capital expenditures approved, get
designs in place, and obtain the required permits. If PSD is triggered,
permit issuance could possibly take 9-12 months, 3-6 months if PSD is
not triggered. Secondly, Oklahoma utilities must obtain permission from
the Southwest Power Pool to undertake the substantial downtimes that
will be required to install control equipment (upwards of 40-50 days).
Based on power demand and the short timeline, brown-outs could very
likely occur.

The pace and aggressiveness of these environmental regulations should be
adjusted to reflect and consider the overall risk to the bulk power
system. EPA, FERC, DOE, and state utility regulators, both together and
separately, should employ the array of tools at their disposal to
moderate reliability impacts, including, among other things, granting
required extensions to install emission controls.  [Oklahoma Department
of Environmental Quality (EPA-HQ-OAR-2009-0491-4593), p. 2-3.]

RESPONSE:  See the preamble of final supplemental rule, section
III.B.iv.  Also see the preamble to the final Transport Rule, section
VII.C.2 and the "Transport Rule Engineering Feasibility Response to
Comments" document in the docket to this rulemaking (Docket ID No.
EPA–HQ–OAR–2009–0491).  The Transport Rule does not apply
unit-specific requirements, thus provides flexibility for compliance
across units with various control capacities.

Comment:  Agency commenter calculates that utilities operating in
Oklahoma will need to retrofit 43 individual combustion units with
low-NOX burners to comply with the requirements of the new rule. This
number is substantially larger than the 30 units in a four-state area
predicted in EPA’s TSD, “Installation Timing for Low NOX Burners
(LNB).” To remedy these failings, agency recommends that the
compliance date for units operating in Oklahoma be pushed back to the
2015 ozone season. We believe that our request for a minimum of a
three-year extension of the compliance date is justified for four
reasons: (1) that is the minimum time necessary between issuance of the
final rule (which will set the allowance budget) and the time the
utilities will need to be able to plan for compliance with the rule –
this is an essential component of any market-based approach, (2) the
failure of the EPA to abide by the assumptions identified in their own
TSD regarding the time required to install low-NOX burners, (3) the
large number of units (43) that will need to be retrofitted in Oklahoma
and the burden that will place on engineers, equipment suppliers, and
the utilities, and (4) the need to allow for utilities to apply for PSD
permits as well as the time required to evaluate and draft the permits
and that needed for the public and EPA to review them.  [Oklahoma
Department of Environmental Quality (EPA-HQ-OAR-2009-0491-4593), p.
4-5.]

RESPONSE:  See the preamble of final supplemental rulemaking, section
III.B.iv.  Also see the preamble to the final Transport Rule, section
VII.C.2 and the "Transport Rule Engineering Feasibility Response to
Comments" document in the docket to this rulemaking (Docket ID No.
EPA–HQ–OAR–2009–0491).  The Transport Rule does not apply
unit-specific requirements, thus provides flexibility for compliance
across units with various control capacities.

Comment:  Should the EPA proceed with this rule making I would request
that Oklahoma be granted a more reasonable implementation timeline.
Because we are so close to the effective date, it will be impossible for
Oklahoma sources to comply and be fitted in time with the appropriate
emission controls, which require custom design, engineering, and
installation specific to each source. With an unrealistic implementation
timeline such as the one proposed, the reliability of electrical
generation in the state of Oklahoma and the livelihood of many
Oklahomans is at stake.  [Governor, State of Oklahoma
(EPA-HQ-OAR-2009-0491-4594), p. 1.]

RESPONSE:  See the preamble of final supplemental rulemaking, section
III.B.iv.  In addition, note that the Transport Rule does not apply
unit-specific requirements, thus provides flexibility for compliance
across units with various control capacities.

Comment:  Our preliminary review of the proposed rule and the supporting
documentation brings us to the same conclusion for both the final rule
and the proposed supplemental rule, which is there is insufficient time
for Kansas utilities to comply with either rule.  Of the 28 states that
will be subject to the Cross State Air Pollution rule (CSAPR) starting
in 2012, all but three were subject to the Clean Air Interstate Rule
(CAIR) prior to CSAPR. The three non-CAIR states are Kansas, Nebraska,
and Oklahoma. CAIR states have had several years to prepare plans,
obtain construction permits, and commence or complete construction of
projects to reduce emission of nitrogen oxides and sulfur dioxide. Until
very recently, Kansas and the other non-CAIR states have shared
uncertainty with respect to interstate transport and future control
requirements. Kansas finds itself at a disadvantage, when compared to
the CAIR states, when faced with a 2012 compliance deadline. In section
VII C. 2. of the CSAPR preamble (76 FR 48279), EPA states:

"In 2012, the industry will largely meet the rule's NOX requirements by:
Operating an extensive existing set of combustion and post-combustion
controls on fossil fuel-fired generators; dispatching lower emitting
units more often; and installing and operating a limited amount of
relatively simple NOX pollution controls in states not previously
subject to CAIR."

EPA acknowledges that utilities in non-CAIR states likely will need to
install "relatively simple NOX pollution controls," such as low-NOX
burners and overfire air, to meet the state emissions budget for 2012.
EPA further posted a NOX cost threshold of$500 per ton for states under
the Cross-State Air Pollution Rule. Installing combustion NOX pollution
controls is not a simple task and cannot be completed in the six months
time frame the proposed rule would allow. We have current real world
costs for installation of such controls and they are higher than the
$500 cost threshold. The rush for Kansas utilities to comply with a 2012
deadline will result in increased costs and the potential for stranded
investments resulting from interim measures or "temporary fixes." For a
utility to install low NOX burners, a prevention of significant
deterioration (PSD) permit generally will be required due to an increase
in carbon monoxide emissions. The PSD construction permitting process
alone would normally take more than five months. It is highly unlikely
that Kansas utilities will be able to install controls and meet the
state's emissions budget for 2012.

Many Kansas facilities are in the planning stages for installation of
additional controls to be completed between 2012 and 2015, and have
received permits or have applications in process. Emission reductions
from approved projects now underway and in 2012 will reduce over 14,000
tons of NOX and 18,133 tons of SO2. Once these controls are in place,
the State of Kansas should be able to meet the assurance levels
established in the Cross State Rule. Kansas needs the time for these
plans to be put in place to comply with the proposed rule. We believe
that the assumptions made by EPA in terms of cost of compliance and
ability to complete pollution control projects in time for the 2012
ozone season are fundamentally flawed and therefore do not accurately
reflect the actual impacts the proposed rule will have on the Kansas
economy. We formally request that EPA provide the State of Kansas and
its affected utilities two additional years to comply with the
provisions of the proposed supplemental Cross-State Air Pollution Rule. 
[Kansas Department of Health & Environment (EPA-HQ-OAR-2009-0491-4582),
p. 2.]

RESPONSE:  See the preamble of final supplemental rule, section III.C
regarding the status of Kansas in the final rule.  Also see the preamble
for the final Transport Rule, section VII.C.2 and the "Transport Rule
Engineering Feasibility Response to Comments" document in the docket to
this rulemaking (Docket ID No. EPA–HQ–OAR–2009–0491) regarding
timing and feasibility of controls installation.  Note that the
Transport Rule does not apply unit-specific requirements, thus provides
flexibility for compliance across units with various control capacities.

Comment:  PSO and AEP believe that if these states are required to
reduce emissions, the level of reductions and the schedule for the
emission reductions can and should be relaxed. There is no immediate
need associated with potential health impacts since the Michigan County
at issue is already in attainment and predicted to remain in attainment.
 This clearly illustrates that there is no need to rush the emission
reductions EPA proposed for 2012. The proposed time schedule for
emission reductions will greatly impact the availability of existing
generating sources in these states by requiring controls be added prior
to the 2012 ozone season. Very few units will be able to install
controls in order to comply with the 2012 ozone season and generation
may not be available during high demand periods. Given the uncertain
availability of NOX allowances and the compliance risk associated with
utilizing allowances above the assurance level, sources will likely
curtail operation to achieve compliance. The result could be either
insufficient generating capacity to meet load demands during peak summer
conditions and/or much higher prices for power purchases during these
periods which will significantly impact customer costs. We are also
uncertain of the availability and price of NOX allowances for the 2012
ozone season.  We expect the State as a whole to require allowances
greater than the 21% assurance level above the state budget which will
add an undue burden on the rate payers of Oklahoma.  By extending the
compliance dates by 2-years or more, the industry would have time to
properly plan and install any needed controls in Oklahoma and other
states.  [Public Service Company of Oklahoma
(EPA-HQ-OAR-2009-0491-4600), p. 3-4.]

RESPONSE:  See section III.B.iv in the preamble to the final
supplemental rule.  Also, see the preamble of final Transport Rule,
section VII.C.2 and the "Transport Rule Engineering Feasibility Response
to Comments" document in the docket to this rulemaking (Docket ID No.
EPA–HQ–OAR–2009–0491).  Note in addition that the Transport Rule
does not apply unit-specific requirements, thus provides flexibility for
compliance across units with various control capacities.

(C)  Commenter(s) disagree with statement that ability to purchase
allowances helps address concerns over compliance timetable.

Comment:  EPA recognizes in the Supplemental Proposal that some sources
will need more than six months to procure, permit and install combustion
controls. In this situation, EPA says that sources may comply through
the purchase of additional allowances. Commenters have not been able to
identify how sufficient allowances could be available in Oklahoma while
still allowing EGUs to meet consumer demand. They believe there will be
a shortfall of allowances, especially during the 2012 ozone season. With
no system of information available with which to discern allowances and
their pricing, commenters assert that more time is needed before the
start of CSAPR in order to appropriately manage the supply (and cost) of
allowances available for purchase.

Table A in the comments shows that a reduction of close to 13,000 tons,
or roughly 40%, from actual 2010 state-wide NO, emissions would be
needed in order to make sufficient in-state allowances available to EGUs
in Oklahoma. The group of commenters represents roughly 80% of the
generating capacity in the State of Oklahoma and, as noted, will be
unable to install controls by May 2012. The group of commenters would
need to purchase roughly 16,000 tons of ozone season NOX allowances
prior to that date.  One of the commenters states that it will have to
purchase up to 6,000 allowances, mostly from EGUs that make up the other
50% of the State's generating capacity. Those other units, however,
would themselves be required to reduce (or obtain allowances for) an
additional 7,000 tons or so of NOX emissions. Because many of these
units will be unable to reduce emissions in less than six months time,
there is not likely to be a sufficient supply of in-state allowances
available for purchase in the early days of CSAPR in Oklahoma.

Moreover, even if some units reduce emissions in 2012, they could choose
to bank rather than sell their allowances in light of EPA's stated
intent to promulgate further interstate transport requirements in the
future. Allowances that are available for purchase are likely to be at a
premium price. While commenters agree that allowance banking is a
cost-effective and important feature of any emissions trading program,
EPA should recognize that banking also can lead to an allowance
shortfall (and high prices) at the beginning of a trading program when
drastic emission reductions are required without sufficient time to
install controls, and further, unspecified reductions are likely to be
required in the future.

Interstate trading will be of limited use to commenters and other
utilities that must obtain thousands of allowances at the start of the
CSAPR program, if EPA applies the assurance provisions starting in 2012.
The assurance provisions would mean that interstate trading could
provide no more than 4,585 tons of ozone season NOX allowances (i.e. the
variability limit) to EGUs in Oklahoma during 2012, or else penalties
would be imposed on the State and its sources. One of the commenters
notes that it alone will require more allowances than interstate trading
could provide.

Given the lack of sufficient allowances, EPA should revise the
Supplemental Proposal to allow a reasonable allowance budget. Once a
reasonable budget is established, utilities will need at least three
years from the date of a final rule to comply, which is the amount of
time needed to install controls for achieving actual emission
reductions. One of the commenters notes here, echoing similar remarks
summarized above, that EPA should at the very least delay application of
the assurance provisions until 2014 so that interstate trading can
provide more allowances to Oklahoma EGUs while the markets for instate
allowances develop. Otherwise, if faced with inadequate lead time for
installing controls and a shortfall of emission allowances, commenter
would be forced to curtail operations to avoid noncompliance with CSAPR.
EPA has not provided an adequate assessment of the impacts on electric
reliability from this level of curtailment.  [Oklahoma Gas & Electric
Company (OGE) (EPA-HQ-OAR-2009-0491-4590), p. 9-10; Oklahoma Utility
Group (EPA-HQ-OAR-2009-0491-4591), p. 7.]

RESPONSE:  Market liquidity is already developing in the Transport Rule
(TR) emission markets, including in the ozone season NOX market relevant
to this rule. Facilities have been trading allowances since EPA formally
recorded 2012 Transport Rule allowances into designated facility
accounts on October 18, 2011.  In fact, several trades of Transport
Rule allowances took place prior to the formal distribution of
allowances. Initial transactions have been followed by further trades
and options contracts in all of the trading programs.

Further, on Thursday, November 17, 2011, First Energy held a successful
auction of a minimum of a 100,000 SO2 allowances, 50,000 each from 2012
and 2013 markets, along with an undisclosed amount of annual and ozone
season NOX allowances. This auction is a significant injection of
liquidity into the market by a single entity, further indicating that a
sufficient numbers of allowances will be available to meet the needs of
sources.  They are planning to hold another auction on April 30, 2012.

Market liquidity has developed rapidly in all of the previous emission
trading programs that EPA has implemented, including the Acid Rain
Trading Program, the NOX SIP Call, the NOX Budget Program, and most
recently CAIR, the predecessor to the Transport Rule.  In both CAIR and
the Transport Rule, individual units have the flexibility to supplement
their own allowance allocations with the acquisition from the
marketplace of any additional allowances needed to cover emissions under
the Transport Rule programs.  Healthy markets for the CAIR annual NOX,
CAIR ozone-season NOX, and Acid Rain (SO2) program allowances have been
in existence for many years and delivered compliance costs below EPA’s
estimates.  Further, as mentioned, trades have already taken place for
allowances in the seasonal NOX market and other TR markets in
anticipation of program implementation.  Also see “Sam Napolitano
Omnibus Declaration” in the docket to this final rulemaking.

EPA is addressing the timing of the assurance provisions in a separate
rulemaking (76 FR 63860, October 14, 2011).

 

Comment:  EPA says in the Supplemental Proposal that individual sources
may comply through the purchase of additional allowances in the event
that it takes a particular source more than six (6) months to install
combustion controls. EPA fails to acknowledge the potential for
allowance shortfalls, especially leading up to the proposed 2012
compliance deadline.  EPA allowance allocation shows that a 37.5%
reduction in actual state-wide NOX emissions are needed for sufficient
in-state allowances to be available. EPA has not suggested any means for
achieving this sort of reduction in the few short months before the
beginning of the 2012 ozone season.

Interstate trading is limited by assurance provisions that restrict EGU
emissions within each state to the state's budget plus the variability
limit. The number representing a state's budget + variability limit is
referred to as the "assurance level." For Oklahoma, the assurance level
in 2012 for ozone season NOX from EGUs is 21,835 (budget) + 4,585
(variability limit) = 26,420. The purchase of allowances by individual
sources cannot authorize the state to go above 26,420 tons of ozone
season NOX emissions.  Otherwise, penalties will be imposed on the state
and its sources. Note that EPA originally proposed to apply the
assurance provisions starting in 2014, but the final rule implements the
provisions starting in 2012. Commenters favor a 2014 start date for
assurance provisions.

If faced with inadequate lead time for installing controls and a
shortfall of emission allowances, utilities would be forced to shut down
some units or curtail their operation to avoid noncompliance with CSAP
program requirements. If curtailment becomes necessary for CSAP
compliance, the only other alternative to serve customers is to purchase
power from out-of state generators. It is unknown if sufficient
additional generating capacity will be available from units within the
Southwest Power Pool. Additionally power which is available to be
supplied by the Southwest Power Pool would increase NOX emissions in the
generating state as well. [AES Shady Point (EPA-HQ-OAR-2009-0491-4595),
p. 5-6.]

RESPONSE:  Market liquidity is already developing in the Transport Rule
(TR) emission markets, including in the ozone season NOX market relevant
to this rule. Facilities have been trading allowances since EPA formally
recorded 2012 Transport Rule allowances into designated facility
accounts on October 18, 2011.  In fact, several trades of Transport
Rule allowances took place prior to the formal distribution of
allowances. Initial transactions have been followed by further trades
and options contracts in all of the trading programs.

Further, on Thursday, November 17, 2011, First Energy held a successful
auction of a minimum of a 100,000 SO2 allowances, 50,000 each from 2012
and 2013 markets, along with an undisclosed amount of annual and ozone
season NOX allowances. This auction is a significant injection of
liquidity into the market by a single entity, further indicating that a
sufficient numbers of allowances will be available to meet the needs of
sources.  They are planning to hold another auction on April 30, 2012.

Market liquidity has developed rapidly in all of the previous emission
trading programs that EPA has implemented, including the Acid Rain
Trading Program, the NOX SIP Call, the NOX Budget Program, and most
recently CAIR, the predecessor to the Transport Rule.  In both CAIR and
the Transport Rule, individual units have the flexibility to supplement
their own allowance allocations with the acquisition from the
marketplace of any additional allowances needed to cover emissions under
the Transport Rule programs.  Healthy markets for the CAIR annual NOX,
CAIR ozone-season NOX, and Acid Rain (SO2) program allowances have been
in existence for many years and delivered compliance costs below EPA’s
estimates.  Further, as mentioned, trades have already taken place for
allowances in the seasonal NOX market and other TR markets in
anticipation of program implementation.

EPA is addressing the timing of the assurance provisions in a separate
rulemaking (76 FR 63860, October 14, 2011.

 

Comment:  It is equally infeasible for WFEC to comply with the required
emissions reduction by purchasing allowances. EPA anticipates that EGUs
that cannot install controls by the compliance deadline will avoid
violating the regulatory program by purchasing allowances. However,
EPA's expectations for the liquidity of the allowance market are greatly
inflated. ODEQ analyzed the state emissions budget information provided
by EPA in its technical support documents for CSAPR and determined that,
of the 26 states included in the ozone season NOX trading program, only
five states are predicted to have excess allowances. See ODEQ Analysis,
Attached. None of these states are in the SPP. The remaining states are
expected to have allowance shortfalls that range between 4.5 and 42
percent. ld.

Accordingly, the Proposal would place WFEC in the untenable position of
having to choose between violating the requirements of the program or
derating or shutting down units. Derating or shutting down units would
jeopardize WFEC's ability to fulfill its contractual obligation to
supply power to the cooperatives it serves and ultimately to their
retail, residential, commercial and industrial consumers. WFEC currently
relies on some bilateral and market purchases to satisfy its
obligations. If WFEC is forced to derate its EGUs to avoid violating
CSAPR, it will need to significantly increase its power purchases. WFEC
is concerned that if the Proposal is finalized, it will be unable to
purchase sufficient power to cover the shortfall in generation. Eight of
the nine states in the SPP are affected by either CSAPR or the Proposal
and, thus, are unlikely to generate excess power for sale. Derating or
shutting down units also would pose a significant risk to the public
health and welfare. The cooperatives served by WFEC do not have access
to other sources of power. If WFEC cannot respond to demand, the rural
areas served by the cooperatives will experience potentially dangerous
brownouts and blackouts. 16 WFEC urges EPA to seek additional
information from SPP regarding the potential affect of the Proposal on
reliability before issuing a final rule.  

GRDA has studied the NOX allowances for EGU's under the Proposal and is
deeply concerned that sufficient NOX allowance will not be available in
2012 and that Oklahoma utilities will be gouged exorbitant millions of
dollars to purchase allowances from eastern utilities which have had
ample opportunity to accumulate such NOX allowances. To force Oklahoma
utilities like GRDA to transfer many millions of dollars of wealth is
patently unfair and serves no environmental benefit. GRDA estimates that
its public power customers will express great outrage if such a wealth
transfer was forced on them. Again, GRDA urges EPA to reconsider the
Proposal schedule and allow a minimum of 36 months for implementation by
Oklahoma EGU's.  [Grand River Dam Authority ("GRDA")
(EPA-HQ-OAR-2009-0491-4586), p. 5-6]

RESPONSE: Market liquidity is already developing in the Transport Rule
(TR) emission markets, including in the ozone season NOX market relevant
to this rule. Facilities have been trading allowances since EPA formally
recorded 2012 Transport Rule allowances into designated facility
accounts on October 18, 2011.  In fact, several trades of Transport
Rule allowances took place prior to the formal distribution of
allowances. Initial transactions have been followed by further trades
and options contracts in all of the trading programs.

Further, on Thursday, November 17, 2011, First Energy held a successful
auction of a minimum of a 100,000 SO2 allowances, 50,000 each from 2012
and 2013 markets, along with an undisclosed amount of annual and ozone
season NOX allowances. This auction is a significant injection of
liquidity into the market by a single entity, further indicating that a
sufficient numbers of allowances will be available to meet the needs of
sources.  They are planning to hold another auction on April 30, 2012.

Market liquidity has developed rapidly in all of the previous emission
trading programs that EPA has implemented, including the Acid Rain
Trading Program, the NOX SIP Call, the NOX Budget Program, and most
recently CAIR, the predecessor to the Transport Rule.  In both CAIR and
the Transport Rule, individual units have the flexibility to supplement
their own allowance allocations with the acquisition from the
marketplace of any additional allowances needed to cover emissions under
the Transport Rule programs.  Healthy markets for the CAIR annual NOX,
CAIR ozone-season NOX, and Acid Rain (SO2) program allowances have been
in existence for many years and delivered compliance costs below EPA’s
estimates.  Further, as mentioned, trades have already taken place for
allowances in the seasonal NOX market and other TR markets in
anticipation of program implementation.

EPA is addressing the timing of the assurance provisions in a separate
rulemaking (76 FR 63860, October 14, 2011.

Comment:  EPA indicates that, “Individual sources may comply through
other measures (such as purchasing additional allowances) in the event
that it takes a particular source more than six months for installation
of a given combustion control.” See 76 Fed. Reg. 40668, Section I(E).
KCP&L does not agree with this statement. For example, KCP&L operates
the La Cygne Generating Station in Kansas. This Station is the only
coal-fired units operated by KCP&L in Kansas. The proposed 2012 ozone
season allowance allocation shortfall is substantially greater than the
Station’s share of the Kansas variability limit.  So even if KCP&L was
able to procure allowances to cover its shortfall at the Station, it
would be exceeding its share of the Kansas variability limit.  This puts
KCP&L at risk of a significant penalty if the state exceeds it assurance
level even though KCP&L La Cygne Generating Station had allowances to
cover its emissions. See 76 Fed. Reg. 48296.  This risk of penalty
illustrates that EPA’s statement that the sources may purchase
additional allowances is not a viable option to comply with our
allowance shortfall at the Station.  [Kansas City Power & Light
(EPA-HQ-OAR-2009-0491-4584), p. 2-3.]

RESPONSE:  EPA is not finalizing the FIP for Kansas at this time.  See
the preamble to the final supplemental, section III.C, for a discussion
on Kansas’ current status.

(D) The rule presents reliability concerns.

Comment:  In order to provide electrical service to our customers Empire
District must maintain system reliability. Considering the severe
shortage of allowances that will be available for purchase in 2012,
sources that do not receive sufficient allocations permitting them to
operate normally must install pollution control equipment. And given the
fact that sources will not have enough time to install the proper
pollution control equipment before January 1, 2012, to meet customer
electrical demand will be forced to de-rate the unit(s) in order to
comply with these rules. This will have an effect on system reliability.
[Empire District Electric Company (EPA-HQ-OAR-2009-0491-4581), p. 2.]

RESPONSE:  EPA rejects the commenter’s assertion that there will be no
allowances available to purchase in 2012.  As explained above, ample
evidence already exists that viable allowance markets for Transport Rule
compliance are already developing and trading has already begun. 
Furthermore, sources have the benefit of accessing the allowance market
not just ahead of the commencement of the programs, but also over the
entire time horizon ahead of the allowance transfer deadlines by which
sources must demonstrate compliance (December 1, 2012 for the 2012 ozone
season and March 1, 2013 for the annual programs in 2012).  As a result,
the Transport Rule offers substantial unit-level flexibility that will
allow sources to continue to meet whatever obligations they may have
concerning electric reliability.  See also “Sam Napolitano Omnibus
Declaration” in the docket to this rulemaking.

	

Comment:  The Attorney General is also concerned about the effect of
compliance upon the availability of electricity and the reliability of
the grid in this region. To the knowledge of the Attorney General, there
has been no complete or formal electric reliability assessment in
cooperation with FERC for the CSAP Program, either alone or in
combination with EPA's other proposed rules. FERC's informal analysis
regarding the CSAP Program indicates that over 80 gigawatts of
coal-fired generation may be retired in response to EPA regulations.
Instead of obtaining a complete reliability assessment from FERC, EPA
instead relied upon emission projections and analyses developed using
its Integrated Planning Model ("IPM"). However, a major limitation of
the IPM model is that it assumes adequate transmission capacity exists
to deliver any resource located in, or transferred to, the region in
question and thereby fails to accurately evaluate impacts on reliability
resulting from EPA programs such as CSAP, or combinations of EPA
programs. The Attorney General is concerned that deficiencies in the IPM
modeling could result in a shortage of replacement purchased power
during the times in which either generating units are down or usage is
curtailed due to efforts to comply with CSAP or other EPA rules.
Electricity shortages could also result from the inability of utilities
to timely obtain needed permits and control equipment. Oklahoma electric
utilities doubt that they can take the measures required to attain
compliance with CSAP Rules within the six-month time frame currently
allowed by EPA. Instead, they estimate needing twelve to eighteen months
to acquire all necessary permits and equipment is given, with a total of
twenty-one months estimated for the controls to be operational. Failure
to obtain needed permits and control equipment by May 2012 would result
in the utilities' de-rating its non-compliant generating units to a
point necessary to comply with ozone standards, meaning each of those
units would be operating at far less than its rated capacity. As a
result, uncertainty currently exists regarding the availability and
price of purchased power, even assuming adequate transmission
capabilities, due to the widespread imposition of CSAP requirements
across many states. [Oklahoma Attorney General
(EPA-HQ-OAR-2009-0491-4592), p. 2-3.]

RESPONSE:  The commenter erroneously attributes “FERC’s informal
analysis” of 80 GW of potential retirements to the Transport Rule,
while in reality the commenter is citing a preliminary FERC estimate
that was based on a wide spectrum of several regulations, many of which
had not even been proposed, and about which FERC made highly
conservative assumptions about stringency of required controls that were
not borne out by the rules as they were subsequently proposed.  FERC’s
own chairman has emphasized that FERC staff had to make assumptions
about EPA regulations before they were proposed and that these
regulations have changed , has characterized this document as only “an
adequate back-of-the-envelope first assessment” and has stated that it
“cannot be relied on to determine specific effects on system
reliability.  Therefore, it  is “inadequate to use as a basis for
decision making.”    EPA disagrees with the commenter’s unsupported
allegation that its analysis, included in the final Transport Rule,
“fails to accurately evaluate impacts on reliability.”  EPA’s
modeling includes interregional transmission limitations based on
parameters published by the North American Electric Reliability
Corporation, and those modeling assumptions were subject to public
comment in the Transport Rule proposal as well as that rule’s IPM
NODA.  As explained in section III.B.iv of this rule’s preamble, as
well as in the Determination of State Budgets for the Final Ozone Season
Supplemental of the Transport Rule” TSD in the docket to this final
rule, EPA has carefully examined the feasibility of emission reductions
required for Oklahoma under this final rule, and EPA’s analyses show
that the Transport Rule does not threaten the availability or
reliability of electricity for Oklahoma’s citizens.

	

(E)  A more complete electric reliability assessment is needed for
Oklahoma and the surrounding region.

Comment:  EPA has yet to provide an appropriate electric reliability
assessment for the CSAPR and EPA's other rulemaking initiatives for the
electric power sector. Requiring compliance with the Supplemental
Proposal by May 2012 without a more complete assessment of reliability
would create an unacceptable risk, especially if 2012 brings weather
patterns similar to those experienced in Oklahoma this year.
Commenters’ generating resources are currently operating at maximum
capacity to meet consumer demands during a record-breaking heat wave. A
more complete reliability assessment is critically important to evaluate
whether Oklahoma EGUs could meet these kinds of demands if curtailments
were required as a result of EPA rules.

According to the Federal Electric Reliability Commission ("FERC"), "the
appropriate vehicles for addressing the impact on electric reliability
of the EPA rules in detail are the planning processes used by utilities
to identify and plan for the infrastructure and resources they will need
to meet future needs." The processes FERC is referring to are
implemented by regional planning authorities, including the Southwest
Power Pool ("SPP") for the States of Arkansas, Kansas, Louisiana,
Mississippi, Missouri, Nebraska, New Mexico, Oklahoma, and Texas. EPA
should coordinate with FERC and the SPP to prepare a more complete
electric reliability assessment for the implementation of CSAPR and
other major EPA rulemakings in Oklahoma.

In response to comments submitted on the final CSAPR, EPA claims that it
has "been in contact" with many of the regional planning authorities
concerning the impact of CSAPR on electric reliability. EPA does not
elaborate on its consultations with the SPP or other regional
authorities, but instead relies on the fact that regional authorities
"have not filed comments or expressed concerns over reliability in the
record for this rule. . . .  EPA's response misses the point. Numerous
entities have expressed concern over the lack of a complete assessment
of electric reliability impacts from EPA rules, especially during the
early years of CSAPR. The way to obtain a complete assessment is for EPA
to prepare one in consultation with FERC and the regional authorities.
It is not sufficient for EPA to point to a lack of comments from these
entities in the official record.

Rather than performing a complete reliability assessment, EPA claims
that the flexibility allowed by an interstate trading program structure
will inherently promote electric reliability, and that "basic
reliability requirements are already incorporated into the analysis" of
its IPM. With respect to the structure of CSAPR, allowance trading alone
will not ensure electric reliability in Oklahoma during the initial
phase of the CSAPR for the reasons discussed previously in commenters’
letters. Rather, unit curtailments are possible on a scale that could
impact electric reliability, if sufficient replacement power is not
available for purchase and transmission into and within the State.
Commenters are concerned that the IPM does not sufficiently account for
restrictions on purchase power availability and transmission
capabilities.

In particular, the IPM sets certain limits on the ability to transfer
power between model regions using the bulk power transmission system.
One of the commenters recently performed a Linear First Contingency
Incremental Transfer Capability Single Transfer Analysis to determine
the deliverability of a resource in another region and found that the
capacity from that resource was not deliverable, even though the
capacity studied was less than the Joint Capacity and Energy Limit used
by the IPM. Furthermore, the IPM's inter-regional limits were determined
from the 2004 NERC Reliability Assessments, which were performed at a
time when energy sales in Oklahoma were 7% less than in 2010.

The IPM also assumes that, within each model region, "adequate
transmission capacity exists to deliver any resources located in, or
transferred to, the region. . . This is a major limitation on the
model's ability to accurately evaluate impacts on electric reliability
in Oklahoma from major federal rules like the CSAPR. While the IPM can
determine that a given region could meet reserve requirements with
imported resources, it performs no assessment of the reliable delivery
of the needed capacity. Regarding the legacy systems of its members in
the 2010 State of the Market Report, the SPP stated that "[w]hile these
systems may be functionally sufficient for their respective areas, they
may not be sufficient for macro transfers across the region." Also, of
the 20 most congested flowgates identified by the SPP in the July 2011
Monthly State of the Market Report, 18 were within the SPP region as
designated in the IPM.

EPA should consider that purchasing replacement power of this magnitude
may require OG&E to go through a competitive procurement bid process
pursuant to the Oklahoma Corporation Commission ("OCC") electric utility
rules and also may require OCC approval. Depending on the complexity of
the issues involved and the timelines prescribed by OCC rules, this
process could require one year or more to complete, which could
frustrate commenters' ability to arrange for replacement power if needed
during the 2012 ozone season. Commenters are also concerned that an
increase in demand for purchase power of this magnitude could impact the
electricity prices during 2012.

Finally, the IPM does not substitute for an operational study of the
generation capacity that would be available in the SPP during the ozone
season if the Supplemental Proposal is finalized. The assumptions made
in the IPM tend to overestimate unit availability, and commenters are
concerned that EPA has not adequately evaluated how the Supplemental
Proposal would impact the ability to meet peak loads on a month-to-month
basis during the summer time. In addition, while it is theoretically
possible that generation capacity might be available within the region,
it is also necessary to consider potential limitations in gas supply and
transportation deliverability to ensure that fuel switching from coal to
natural gas can be accomplished, along with the costs.

These are select examples of commenters’ concerns about the IPM as a
means of evaluating electric reliability for the Supplemental Proposal,
especially during the 2012 ozone season. More time is needed to review
the model's assumptions and incorporate additional considerations
related to generation capacity and transmission in the SPP. It will not
be possible to prepare a sufficient assessment in time for compliance by
May 2012. Furthermore, requiring compliance by that date without an
appropriate assessment would create an unacceptable risk to electric
reliability.

Commenter AES Shady Point raises some of the same concerns in its
comments.  [Oklahoma Gas & Electric Company (OGE)
(EPA-HQ-OAR-2009-0491-4590), p. 11-13; Oklahoma Utility Group
(EPA-HQ-OAR-2009-0491-4591), p. 8-10; AES Shady Point
(EPA-HQ-OAR-2009-0491-4595), p. 6-7.]

RESPONSE:  As the commenters note in quoting FERC, utility planning
processes are the appropriate vehicle for addressing the impact on
reliability of EPA rules.   EPA has been working with RTOs, utilities,
state commissions, and other appropriate entities in developing this
rule, and others, and will continue working with these agencies as the
rules are finalized and implemented.   EPA considers all rules that are
in effect when it evaluates each proposal, so that the cumulative impact
of rules are evaluated as they are put into effect.  EPA evaluates all
comments that are filed on each proposed rule in finalizing the rule.  

EPA believes the limitations on inter-regional power transfers used in
IPM are appropriate; these limits are evaluated and modified when
information is provided that is sufficient to update these limits.  
Broad references, such as those in these comments, do not provide
sufficient information to update these limits.  Although the
commenters’ information that sales in Oklahoma are 7% higher today
than they were in 2010 is relevant to the determination of resource
adequacy, IPM already captures these impacts through its use of reserve
margin requirements.

EPA believes commenters are wrong to dismiss the flexibility of the rule
and then claim reliability problems.  The rule’s inherent flexibility
permits all utilities many alternatives to comply with the requirements
of the rule: to purchase allowances from the market, to choose the best
units to dispatch, to change fuel supply, or to import power from other
regions.  These comments provide no substantive reasons why these
alternatives – either singly or in combination – are not viable ways
of meeting the rule’s requirements.

With respect to comments on the operational limitation of the IPM model,
IPM is not intended to be a model of operating capability within a
region, nor does it need to be for the purposes used in developing the
rule.  IPM is an economic model of the operation of the overall power
system in the lower 48 states; it projects generation and emissions in
each region to meet electricity demand using engineering, environmental,
and economic information on generating units and information on the
inter-regional transmission grid.  The rule does not set operating
requirements on individual units and provides operators of units the
flexibility to choose how units should be operated.  IPM models the
loads in each region using a load curve that represents the amount of
load that must be served at each load level; load levels are designed to
reflect key operational peak, intermediate, and baseload conditions,
thus capturing differing monthly load levels where these are important
for operation of the system.  The model also contains a detailed
representation of the natural gas pipeline system, and incorporated the
cost of increased supply of natural gas and additional pipeline capacity
in deciding whether fuel switching is the best economic choice.  EPA has
developed a balanced set of data on unit availabilities to accurately
represent plant availabilities in the IPM modeling:  coal unit
availabilities are based on historical data used by the Energy
Information Administration, and other availabilities for other types of
units are based on the GADS data (see IPM Documentation in the docket to
this rulemaking).

Statutory and Executive Order Reviews (e.g., ICR)

  Commenter questions how the health benefits of this rule were
determined.

Comment:  How does EPA justify/calculate the numerous health benefits to
this rule?  Limiting emissions from any source will of course provide
health benefits; however, the numbers being issued with this rule seem
to have no basis in fact.  They appear to be based on estimations such
as number of cases, severity of cases, and costs associated with
treatment.  Using those estimations to build a case to justify the
action does not seem fair to the States that will pay the actual costs
for estimated benefits.    [Oklahoma Municipal Power Authority (OMPA)
(EPA-HQ-OAR-2009-0491-4573), p. 2.] 

RESPONSE:  The comment is outside the scope of the SNPR because it
addresses issues that were within the scope of the request for comments
in the proposed Transport Rule but not in the SNPR.  While EPA is not
reconsidering or reopening the issue, we note that the issue was
previously addressed in section VIII.C of the preamble for the final
Transport Rule and in chapter 5 of the final Transport Rule Regulatory
Impact Analysis, in the docket for this rulemaking.

Rule does not take into consideration inability of independent power
producers to recover compliance costs.

Comment:  Additionally and most importantly, the FIP does not take into
consideration that the AES Shady Point facility, as a contracted
independent power producer under long term contracts, does not currently
have mechanisms in place to recoup the implementation costs of this
rule. This lack of a cost sharing mechanism unfairly penalizes
independent power producers, and thereby jeopardizing facility
profitability, economic viability, regional grid stability, and
employment of 90 workers, not including trickling economic impacts. 
[AES Shady Point (EPA-HQ-OAR-2009-0491-4580), p. 1-2.]

RESPONSE:  The comment is outside the scope of the SNPR because it
addresses issues that were within the scope of the request for comments
in the proposed Transport Rule but not in the SNPR.  While EPA is not
reconsidering or reopening the issue, we note that the issue was
previously addressed in the under the subject of long term contracts in
the final Transport Rule RTC.  See section V.D.2.a.,
Applicability/Opt-in Units in the Transport Rule Primary Response to
Comments for more details.

 (C) Commenter questions finding of no significant or unique impacts on
small governments.

Comment:  BPU questions EPA's finding that "[n]either the final
Transport Rule nor the provisions in this SNPR have regulatory
requirements that might significantly or uniquely affect small
governments." 76 FR at 40669/1.  BPU, an administrative agency of the
Unified Government of Wyandotte County/Kansas City, Kansas, operates an
electric utility on a not-for-profit basis for the benefit of
approximately 65,000 residential, commercial, and industrial customers
in Wyandotte County/Kansas City, Kansas. The final Transport Rule and
the SNPR will significantly affect BPU by requiring it to spend
substantial sums of money to comply with these rules. Raising those
amounts through bonds, as BPU must do, places additional burdens on the
regulatory and political process. Moreover, the final Transport Rule and
the SNPR require instantaneous results in comparison to the amount of
time needed to obtain bond financing thus significantly affecting the
ability of a small government to timely react to such requirements
without facing the possibility of enforcement of the new rules against
BPU. Finally, any increased costs related to new emission controls
required by the rule will strain the finances of all BPU's customers,
which include the county, the city and their citizens. [Kansas City
Board of Public Utilities (EPA-HQ-OAR-2009-0491-4585), p. 4-5.]

RESPONSE:  As already stated, EPA believes that no unfunded mandates
have been created by the Transport Rule program inclusive of this
supplemental rule. This comment does not provide evidence to demonstrate
that this rule uniquely affects small governments or that the magnitude
of the impact should be considered “significant.”  EPA has not
changed its finding based on the fact that most of the costs are to be
incurred by the private sector.  In the economic analysis for this rule,
EPA’s projected impacts on electricity prices are small—around 1% on
average nationwide.   See Chapter 8 of the RIA for more on economic
impacts.

(D) The cost of compliance creates a substantial burden upon consumers
and may affect the availability of electricity and the reliability of
transmission within the region.

Comment:  The Attorney General is concerned about the cost of compliance
for utilities within Oklahoma and the effect those costs will have on
consumers when they are inevitably transferred to ratepayers. The cost
of compliance for just the two largest electric utilities in Oklahoma is
estimated to be approximately $150 million. Those costs, when combined
with the substantial costs of compliance with other EPA mandates and
proposed rules create an unprecedented burden upon the electric energy
consumers of Oklahoma. The EPA should be mindful of the fact that
electricity consumers are not only paying the cost of compliance with
its rules, they are also responsible for additional costs mandated by
other entities for transmission development, demand-side management and
energy efficiency programs, renewable energy programs, etc.  [Oklahoma
Attorney General (EPA-HQ-OAR-2009-0491-4592), p. 2.]

RESPONSE:  EPA’s projected impacts on electricity prices are
small—around 1% on average nationwide.   These are the incremental
rate increases attributable to compliance with the Transport Rule.  The
total retail electricity prices forecasted by EPA do take into account
non-Transport Rule variables such as those cited above.  EPA appreciates
the comment and believes its modeling exemplifies its awareness of the
millions of variables factored into electricity prices.  See Chapter 8
of the RIA in the docket of this rulemaking for more on economic
impacts.

 (E) Commenter believes that EPA has failed to consider costs and
impacts properly n applying the rule to Iowa.

Comment:  The 2012 and 2014 base case modeling runs conducted to
determine, among other things, NOX emissions in Iowa and their purported
interference with the maintenance of the 1997 8-hour ozone standard in
Allegan County, Michigan utilized NOX values from IPM that did not
account for the Utility HAPs MACT or the Boiler MACT, both of which will
serve to lower emissions nationwide through the installation of controls
or shutdown of facilities.  Utilization of old base case data that fails
to consider the real emission reductions achieved by the CAIR (as
discussed above) and the emission reductions that will be achieved by
the MACT standards, MEC does not believe EPA has, in this instance,
achieved the requirements of Executive Order 13563 of January 18, 2011,
to “tailor its regulations to impose the least burden on society,
consistent with obtaining regulatory objectives, taking into account,
among other things, and to the extent practicable, the costs of
cumulative regulations” in requiring ozone season NOX reductions from
Iowa sources.  [MidAmerican Energy Company (EPA-HQ-OAR-2009-0491-4597),
p. 4.]

RESPONSE:  The comment is outside the scope of the SNPR because it
addresses issues that were within the scope of the request for comments
in the proposed Transport Rule but not in the SNPR.  While EPA is not
reconsidering or reopening the issue, we note that the issue regarding
the baseline was previously addressed in the preamble to the final
Transport Rule, section VI.  Also see the Regulatory Impact Assessment
for the final Transport Rule in the docket to this rulemaking.

Comment:  EPA is quick to point out that since Iowa and other states
subject to the supplemental notice provisions are already subject to
other requirements to reduce NOX emissions under the final Transport
Rule, there are, effectively, no additional costs to comply with the
ozone season requirements. EPA’s proposed allocation of seasonal NOX
allowances is insufficient to cover Iowa’s compliance burdens without
additional controls, leaving Iowa sources with a choice to spend money
on controls or in the purchase of allowances. Further, there are
incremental administrative costs associated with maintaining compliance
under the ozone season NOX reduction requirements, both from a reporting
and allowance management perspective. EPA’s analysis fails to account
for these incremental costs (as it does for the so-called “sunk
costs” associated with compliance with the legally enforceable
requirements of CAIR); MEC believes that the EPA should be required to
inform the public in general, and MEC’s customers, of the full costs
associated with reducing emissions in Michigan at an ozone monitoring
station that indicates the county is in attainment of the standard. 
[MidAmerican Energy Company (EPA-HQ-OAR-2009-0491-4597), p. 4-5.]

RESPONSE:  While EPA is not reconsidering or reopening the issue, we
note that the issue regarding the treatment of CAIR was previously
addressed in section V.B of the preamble of Federal Implementation
Plans: Interstate Transport of Fine Particulate Matter and Ozone and
Correction of SIP Approvals (EPA–HQ–OAR–2009–0491) and Transport
Rule Primary Response to Comments (EPA-HQ-OAR-2009-0491-4513) Section
IV.A.  In regards to the assertion that Iowa will need to install
additional controls or purchase additional allowances, the commenter has
not provided significant evidence of that.  For instance, the 2010
ozone-season emissions in Iowa were 17,179 tons; the preliminary 2011
emissions for ozone-season NOX in Iowa are 17,179 tons, and their 2012
budget is 16,532 tons.  If Iowa continues to lower emissions at the rate
it has been, its 2012 emissions will be well below its 2012 budget
(meaning the state will likely have a surplus of allowances, not a
shortage).  EPA does not anticipate any additional controls or allowance
purchase for the state.  In fact, EPA modeling projects the state to be
a seller of allowances in 2012.  As for the incremental cost of
operating controls that would not otherwise operate in the base case,
that type of cost is captured in EPA’s analysis.  Additionally, the
$1.6 billion in capital cost already sunk into controls for CAIR is also
cited in EPA’s rollout of the final Transport Rule.  In regards to any
administrative or reporting cost, these are not reflected in the model,
as they are negligible.

 (F) EPA’s cost thresholds do not support retrofit environmental
controls.

 

Comment:  EPA's analysis suggests that additional combustion controls
can be added to many units within Oklahoma in 2012 at or below the cost
effective threshold of $500/ton of seasonal NOX. However, based on the
data contained within the IPM v4.10 documentation and the NEEDs input
file, the cost of some of the projected controls exceeds the
cost-effective threshold. As an example, Northeastern unit 3 is
projected to change from a NOX rate of 0.382 lb/MMBtu in the base case
to a NOX rate 0.251 lb/MMBtu in the policy case. According to EPA
documentation this project would cost almost $13 million and remove
approximately 860 tons of NOX per year. Using the retrofit capital
charge rate of 11.3%, this project would effectively cost more than
$1,600 per ton, before even consideration of O&M impacts. Therefore, it
is unclear why this project would be selected in the policy case. There
appears to be a fundamental discrepancy with the IPM model logic
surrounding control cost-effectiveness.  [Public Service Company of
Oklahoma (EPA-HQ-OAR-2009-0491-4600), p. 4.]

RESPONSE:  The commenter does not provide its underlying calculations
for the $13 million in capital cost referenced in the comment.  EPA’s
calculations reveal a much smaller number for total capital cost and
$/ton figures.  It appears that the commenter may have erred by
reflecting the total capital cost of “LNC3” controls in its
assessment.  However, the only cost incurred to achieve incremental
tonnage reductions noted above are the incremental cost of upgrading
combustion controls from LNC1 to LNC3 figures.  Table 5-5 of EPA’s
IPMv4.10 documentation provides the capital cost and scaling factor
assumption necessary to calculate the upgrade cost at this particular
unit.  Therefore, EPA believes that the reductions can be made
cost-effectively and that the commenter has misinterpreted EPA data in
forming its capital cost projection.  EPA believes that the capital
costs postulated by the commenter are drastically overstated.  However,
even if one were to assume that all the commenter’s calculations are
correct, their observation should be of no surprise.  The EPA remedy
modeling that the commenter is referencing projects an ozone season
allowance price that exceeds $1600/ton in future years, which would make
the upgrade decision for the Northeastern 3 unit economic even at the
conservative parameters assumed in the comment.  Finally, the commenter
assumes that the emissions avoided are 860 tons annually (as shown in
2012).  However, this does not take into account the ability of the unit
to increase utilization in future years and thus generate even more
emission reductions thereby further lowering the $/ton cost of the
control upgrade.  Finally, in this IPM remedy scenario, the state
emissions are less than the budget.  That is, EPA does not assume that
this control is part of necessary emission reductions to reach the
budget level, but rather part of a strategy that allows the state to
outperform its budget (emit less than its budget).  There are no source
specific requirements that the unit install any controls under Transport
Rule.  Moreover, in this rulemaking, EPA has removed any such assumption
that these controls are installed in 2012 in Oklahoma.  

5.3 Miscellaneous Issues

(A)  Several commenters claimed that EPA has not provided a valid
opportunity for public review and comment of the supplemental proposal. 
 [Note:  RESPONSE at end of this set of comments]

Comment:  Commenters who submitted follow-up comments had earlier
submitted requests for an extension of time to submit comments based on
the size of the rulemaking record and the late date by which the input
and output files were provided for CAMx modeling used to support the
Supplementary Proposal.  [AES Shady Point (EPA-HQ-OAR-2009-0491-4578),
Oklahoma Gas & Electric Company (EPA-HQ-OAR-2009-0491-4577), Westar
Energy (EPA-HQ-OAR-2009-0491-4575).]

Comment:  For the reasons set forth in earlier extension request
letters, commenters request that EPA extend the applicable comment
period to provide sufficient time for commenter to thoroughly review and
further inform its comments on this important proposal. The need for
additional time is especially compelling for Oklahoma sources. As
mentioned above, Oklahoma was not previously included in CAIR, and
whether or not Oklahoma would be included in any interstate transport
rule, and the basis for such inclusion, has been a moving target. EPA
included Oklahoma in its 2010 Proposed Transport Rule on the basis that
sources within the State contributed to nonattainment with NAAQS in
Tarrant County, Texas, with those allowances and actual emissions not
requiring that any control equipment be installed. EPA now proposes to
include Oklahoma based on modeled impacts in Allegan County, Michigan,
and also proposes to require a nearly 40% reduction in emissions in a
six month period.

Oklahoma's purported connection to Allegan County, Michigan is based on
new modeling data. The CAMx input and output files for this modeling
were not available online like other technical support documents, and it
took more than a week to receive the 8 TB of data once requested.
Additionally, it can take weeks to re-run the same CAMx model runs that
EPA completed in support of the CSAPR and the 42 day comment period is
not enough time to allow a thorough review. Given the nature of this
data and the tenuous link between Oklahoma and Allegan County
purportedly established by it, it is important that commenters be
provided additional time in which to conduct an in-depth analysis of the
methodology used by EPA as the basis for including Oklahoma in the
Supplemental Proposal. The 42 days that EPA allows for public comment is
not enough.

Furthermore, the comment period allowed by EPA is inconsistent with
Executive Order 13563, which was signed by the President on January 18,
2011, as well as with other recent rulemakings by EPA. Executive Order
13563 (Improving Regulation and Regulatory Review), provides that in
order to facilitate public participation in the regulatory process,
"each agency shall afford the public a meaningful opportunity to comment
... with a comment period that should generally be at least 60 days."
Although EPA followed this mandate in other rulemakings that also
required time-intensive review, such as the 2010 Proposed Transport Rule
and the Utility MACT, EPA inexplicably fails to do the same here.
Commenters request that EPA reconsider its August 19, 2011 decision not
to extend the comment period for the Supplemental Proposal.

Commenter AES Shady Point raises some of the same concerns in its
comments.  [Oklahoma Gas & Electric Company (OGE)
(EPA-HQ-OAR-2009-0491-4590), p. 13; Oklahoma Utility Group
(EPA-HQ-OAR-2009-0491-4591), p. 10-11; AES Shady Point
(EPA-HQ-OAR-2009-0491-4595), p. 7.]

Comment:   EPA has failed to provide an adequate notice and opportunity
for public comment on the Proposal. The Administrative Procedures Act
("APA") requires the EPA to provide notice of a proposed rulemaking
"adequate to afford interested parties a reasonable opportunity to
participate in the rulemaking process." Florida Power & Light Co. v.
United States, 846 F.2d 765, 771 (D.C.Cir. 1988); SENATE JUDICIARY
COMMITTEE, ADMINISTRATIVE PROCEDURE ACT, S. REP., No. 752, 77th Cong.,
1st Sess. 14 (1945) ("Agency notice must be sufficient to fairly apprise
interested parties of the issues involved, so that they may present
responsive data or argument relating thereto."). This requirement serves
both (1) "to reintroduce public participation and fairness to affected
parties after governmental authority has been delegated to
unrepresentative agencies"; and (2) to assure that the "agency will have
before it the facts and information relevant to a particular
administrative problem." National Ass'n of Home Health Agencies v.
Schweiker, 690 F.2d 932, 949 (D.C. Cir. 1982); see also, MCI
Telecommunications Corp. v. FCC, 57 F.3d 1136, 1141 (D.C. Cir. 1995)
(same). The Proposal does not provide potentially affected parties with
sufficient opportunity to understand and respond to the data and highly
technical modeling utilized by EPA and, thus, does not satisfy the
notice and comment requirements of the APA.

The Proposal does not provide affected source with sufficient notice and
opportunity to comment. The Proposal is just eight pages long and does
not have its own technical support documents. EPA directs sources to the
preamble and the technical support documents associated with CSAPR See
76 Fed. Reg. at 40,666. Sources affected by the Proposal do not have a
similar short-cut available for parsing through the relevant information
and developing responsive data and arguments. CSAPR covers 27 states and
establishes four different trading programs. The amount of information
underlying the rule is voluminous. The CSAPR preamble is 146 pages long
and has over 147 technical support documents. EPA has made no effort to
organize the CSAPR materials in a way that clearly identifies
information relevant to the Proposal. Additionally, the Proposal itself
lacks any meaningful explanation of EPA's significant contribution
analyses, state emissions budget determination, or allowance allocation
calculation. EPA's cursory treatment of the Proposal undermines "the
values of public participation, fairness, and informed agency decision
making that the notice-and-comment process is designed to foster." City
of Idaho Falls v. FERC, 629 F.3d 222,229 (D.C. Cir. 2011).

The comment period for the Proposal is too short for affected sources
and states to verify the accuracy of the data and analyses relied on by
EPA. To develop meaningful comments on the Proposal, affected sources
faced the daunting tasks of (1) locating and verifying the data EPA
attributes to their EGUs; (2) understanding the complicated methodology
used by EPA to determine unit-specific emissions allowances; (3)
determining whether EPA correctly calculated the number of allowances to
be allocated to their EGUs; (4) understanding the complicated
methodology used by EPA to calculate state emissions budgets; (5)
locating all the relevant information for EGUs in the state of interest
and determining whether EPA correctly calculated the state emissions
budget; and (6) understanding the highly technical model used by EPA to
determine that the state of interest significantly contributes to
downwind pollution. Even with the assistance of expert consultants, it
is impossible to complete all analyses necessary to fully address the
Proposal in the 43-day comment period.

EPA's failure to provide a reasonable comment period is inconsistent
with Executive Order 13563. The order requires EPA to "afford the public
a meaningful opportunity to comment" on proposed rules, "with a comment
period that should generally be at least 60 days." 7'6 Fed. Reg. 3,821
(Jan. 21, 2011). EPA has had nearly three years since the decision in
North Carolina to develop the Proposal. It expects the general- public
and affected sources, which are far less equipped than EPA to conduct
technical analyses and air quality modeling, to develop comments in just
43 days. Given the scope of the technical analysis required, that
timeframe is clearly insufficient and not in accordance with the mandate
of Executive Order 13563 to allow for meaningful public comment. 
[Western Farmers Electric Cooperative (EPA-HQ-OAR-2009-0491-4589), p.
12-13.]

Comment:  EPA set a 30-day comment period ending August 22, 2011.
Despite EPA's efforts to post everything related to this rule on the EPA
website, key decisions and assumptions made in the rulemaking,
particularly changes from proposed to final, have not been readily
accessible. Although the emissions data are available in the docket, the
modeling input and output files are not. As late as a month after
signing the rule, information is still being posted to the website. The
technical and supporting documentation for the final and proposed rules
is over 132,000 pages. Parsing how and why Kansas fits into the proposed
rule has been an extremely difficult task for KDHE staff. In the
proposed rule, Kansas is linked only to maintenance of the ozone
standard in Allegan County, Michigan. The proposed rule is the first
opportunity for KDHE to review the technical information supporting
Kansas' significant contribution to Allegan County. A complete analysis
of these data will take more time than the current comment period
allows. KDHE requests an extension of time to file a more complete set
of comments from the currently set date of August 22, 2011 to October 1,
2011.  [Kansas Department of Health & Environment
(EPA-HQ-OAR-2009-0491-4582), p. 1-2.]

RESPONSE:  EPA disagrees with the commenters for the following reasons. 
Section 307(d)(5) of the CAA requires the Administrator to give an
opportunity for written or oral comments.  The Act does not specify the
length of time, other than the record must be open 30 days after holding
public hearings - with which EPA has complied.  There was no request
for a public hearing on the Supplemental Notice of Proposed Rulemaking
(SNPR).

EPA posted the signed version of the Proposed Transport Rule to the web
when it was signed on July 6, 2010.  The proposal was published in the
Federal Register on August 2, 2010, and the public comment period closed
on October 1, 2010.  This provided a 60 day comment period (and also
provided 90 days of public notice from the date posted to the web).

EPA posted the signed version of the first Transport Rule Notice of Data
Availability (NODA) addressing its power sector model (IPM) to the web
when it was signed on August 25, 2010.  The first NODA was published in
the Federal Register on September 1, 2010, and the public comment period
closed on October 15, 2010.  This provided a 45 day comment period (or
52 days from the date posted to the web).  EPA posted the signed
version of the second Transport Rule NODA (addressing emissions
inventories) to the web when it was published in the Federal Register on
October 27, 2010.  The public comment period closed on November 26,
2010, which provided a 30 day comment period.  EPA posted the signed
version of the third Transport Rule NODA (allocations and related
matters) to the web when it was signed on December 30, 2010.  The third
NODA was published in the Federal Register on January 7, 2011.  The
public comment period closed on February 7, 2011, which provided a 30
day comment period (or 38 days from the date posted to the web).  EPA
posted the signed version of the Supplemental Notice of Proposed
Rulemaking to the web when it was signed on July 6, 2011.  The SNPR was
published in the Federal Register on July 11, 2011.  The public comment
period closed on August 22, 2011, which provided a 6 week comment period
from publication in the Federal Register (and 47 days from availability
of the proposal on the web).

Given the timeframe EPA provided to the public for submission of
comments, and indicative of the fact that the Agency received several
thousand substantive comments, commenters did, in fact, have a
reasonable opportunity to submit their comments for consideration. 

Commenter states broadly that for all of the various reasons in the
comments, EPA has acted arbitrarily and capriciously in the rulemaking
as it affects Oklahoma.

Comment:   While DEQ is in support of reducing NOX emissions in
Oklahoma, it is our belief that the implementation of this rule in this
manner undermines the credibility of the EPA and that this particular
rulemaking has crossed the line between prudent rulemaking and practices
that can only be characterized as arbitrary and capricious. We believe
that the best way forward would be for the EPA to remove the State of
Oklahoma from the states included in the CSAPR.

DEQ wishes to express its support for emission reductions in the state
of Oklahoma. Unfortunately, it appears that compliance with the heavily
reduced emission budget and the tight timeline will be extremely
difficult, if not impossible for Oklahoma utilities. EPA has acted in a
manner that is arbitrary and capricious in: (1) changing the methodology
used to set state allowance budgets without allowing public comment, (2)
failing to make those procedures clear and unambiguous in a timely
manner, and (3) delaying the final rulemaking which will set state
ozone-season budgets for the six states (including Oklahoma) covered by
the Supplemental Notice of Proposed Rulemaking (SNPR). This is
particularly burdensome for utilities operating units located in the
state of Oklahoma. In fact, Oklahoma is the only state covered by the
SNPR that has no requirements whatsoever in the CSAPR as finalized on
August 8, 2011. EPA’s actions have jeopardized the market-based
benefits of the rule by failing to issue clear and unambiguous signals
regarding the allowances that will be allocated on a unit-by-unit basis
and by holding the potentially subject units hostage to another proposed
rulemaking. As of the date these comments were prepared (August 22,
2011), utilities operating units located in Oklahoma do not know whether
those units will be subject to the rule at all and, if they are subject,
how many allowances each unit will be issued for the 2012 ozone season
beginning in May of that year.

After the issuance of the proposed rule in 2010, many Oklahoma utilities
pointed out inconsistencies in the IPM assumptions and problems with
specific allocations, but, in general, the utilities expected to be able
to comply with the rule. All of these assumptions were upended this
summer and, based on the changes between 2010 and 2011, it would be
imprudent for the utilities to assume consistency between the SNPR and
whatever will be included in the rule when it is finalized this fall. We
believe that the best way to remedy these problems would be for the EPA
to remove the State of Oklahoma from this version of the CSAPR
altogether. We acknowledge that we may well need to be included in a
future extension of CSAPR in a future rulemaking. If and when that
occurs, it is essential that EPA provide sufficient time between the
publication of the final rule and the beginning of the compliance period
to allow utilities operating subject units to adapt to the new
requirements. By failing to provide that time, EPA undercuts its
credibility and makes it more difficult for states to defend rules
which, if implemented properly, would provide substantial benefits to
our citizens. 

DEQ respectfully requests that EPA exclude Oklahoma from the rule
altogether, but requests that at the very least, EPA increase the state
ozone-season NOX allowance budget by 1,500 tons and that the compliance
date for units operating in Oklahoma be pushed back to the 2015 ozone
season.  [Oklahoma Department of Environmental Quality
(EPA-HQ-OAR-2009-0491-4593), p. 5-7.]

RESPONSE:  EPA disagrees with the commenter for the following reasons. 
Section 307(d)(5) of the CAA requires the Administrator to give an
opportunity for written or oral comments.  The Act does not specify the
length of time, other than the record must be open 30 days after holding
public hearings - with which EPA has complied.  There was no request
for a public hearing on the Supplemental Notice of Proposed Rulemaking
(SNPR).

EPA posted the signed version of the Proposed Transport Rule to the web
when it was signed on July 6, 2010.  The proposal was published in the
Federal Register on August 2, 2010, and the public comment period closed
on October 1, 2010.  This provided a 60 day comment period (and also
provided 90 days of public notice from the date posted to the web).

EPA posted the signed version of the first Transport Rule Notice of Data
Availability (NODA) addressing its power sector model (IPM) to the web
when it was signed on August 25, 2010.  The first NODA was published in
the Federal Register on September 1, 2010, and the public comment period
closed on October 15, 2010.  This provided a 45 day comment period (or
52 days from the date posted to the web).  EPA posted the signed
version of the second Transport Rule NODA (addressing emissions
inventories) to the web when it was published in the Federal Register on
October 27, 2010.  The public comment period closed on November 26,
2010, which provided a 30 day comment period.  EPA posted the signed
version of the third Transport Rule NODA (allocations and related
matters) to the web when it was signed on December 30, 2010.  The third
NODA was published in the Federal Register on January 7, 2011.  The
public comment period closed on February 7, 2011, which provided a 30
day comment period (or 38 days from the date posted to the web).  EPA
posted the signed version of the Supplemental Notice of Proposed
Rulemaking to the web when it was signed on July 6, 2011.  The SNPR was
published in the Federal Register on July 11, 2011.  The public comment
period closed on August 22, 2011, which provided a 6 week comment period
from publication in the Federal Register (and 47 days from availability
of the proposal on the web).

Given the timeframe EPA provided to the public for submission of
comments, and indicative of the fact that the Agency received several
thousand substantive comments, commenters did, in fact, have a
reasonable opportunity to submit their comments for consideration. 

See section III.B.iv of the preamble to this final rule regarding the
Oklahoma budget.  EPA has considered the comment concerns with respect
to the compliance schedule.   In the final rule, EPA is setting the
Oklahoma 2012 ozone season NOX budget at a level that reflects emission
reductions achievable through actions (such as changes in generation
unit dispatch) that do not include additional LNB installations.  EPA is
setting the Oklahoma ozone season NOX budget for 2013 and beyond at the
level that was proposed, i.e., to reflect NOX levels achievable with
additional LNB installations that can be completed before the 2013 ozone
season without necessitating the shutdown of units during the summer
peak demand period in 2012.   

 (C) EPA’s refusal to consider data is improper.

Comment:   EPA’s supplemental notice of proposed rulemaking seeks to
limit the scope of public input, nothing that: 

EPA is not taking comment on any aspect of the final Transport Rule,
including any aspect of the methodology used to identify receptors for
nonattainment; the methodology used to identify receptors for
maintenance; the methodology used to identify any specific state’s
significant contribution and interference with maintenance; the
methodologies used to establish state budgets, variability limits, and
state assurance levels; or the methodologies used to allocate allowances
to existing units, to establish new unit set-asides and Indian country
new unit set-asides, or to allocate allowances in these set-asides. 

EPA’s refusal to take comment on these issues is premised on its
theory that “EPA provided an adequate opportunity for public comment
on all of these issues during the comment period for the proposed
Transport Rule and during the comment periods for the associated Notices
of Data Availability.” MEC respectfully disagrees that EPA has
provided an adequate opportunity for public comment on the issue of the
methodology used to identify receptors for maintenance. EPA’s previous
proposal to include “all six states in at least one of the Transport
Rule trading programs” does not (and in this case, cannot) mean that
entities impacted by this supplemental notice had the same opportunity
to review the potential impacts at a receptor that EPA itself admits was
not identified or included in the July 6, 2010, proposed rule. 
[MidAmerican Energy Company (EPA-HQ-OAR-2009-0491-4597), p. 2.]

RESPONSE:  EPA disagrees with the commenter for the following reasons. 
Section 307(d)(5) of the CAA requires the Administrator to give an
opportunity for written or oral comments.  The Act does not specify the
length of time, other than the record must be open 30 days after holding
public hearings - with which EPA has complied.  There was no request
for a public hearing on the Supplemental Notice of Proposed Rulemaking
(SNPR).

EPA posted the signed version of the Proposed Transport Rule to the web
when it was signed on July 6, 2010.  The proposal was published in the
Federal Register on August 2, 2010, and the public comment period closed
on October 1, 2010.  This provided a 60 day comment period (and also
provided 90 days of public notice from the date posted to the web).

EPA posted the signed version of the first Transport Rule Notice of Data
Availability (NODA) addressing its power sector model (IPM) to the web
when it was signed on August 25, 2010.  The first NODA was published in
the Federal Register on September 1, 2010, and the public comment period
closed on October 15, 2010.  This provided a 45 day comment period (or
52 days from the date posted to the web).  EPA posted the signed
version of the second Transport Rule NODA (addressing emissions
inventories) to the web when it was published in the Federal Register on
October 27, 2010.  The public comment period closed on November 26,
2010, which provided a 30 day comment period.  EPA posted the signed
version of the third Transport Rule NODA (allocations and related
matters) to the web when it was signed on December 30, 2010.  The third
NODA was published in the Federal Register on January 7, 2011.  The
public comment period closed on February 7, 2011, which provided a 30
day comment period (or 38 days from the date posted to the web).  EPA
posted the signed version of the Supplemental Notice of Proposed
Rulemaking to the web when it was signed on July 6, 2011.  The SNPR was
published in the Federal Register on July 11, 2011.  The public comment
period closed on August 22, 2011, which provided a 6 week comment period
from publication in the Federal Register (and 47 days from availability
of the proposal on the web).

Given the timeframe EPA provided to the public for submission of
comments, and indicative of the fact that the Agency received several
thousand substantive comments, commenters did, in fact, have a
reasonable opportunity to submit their comments for consideration. 

6.  Out of Scope Comments

Significant Contribution Methods

Commenter requests additional clarification on the data used to
determine that Oklahoma is a significant contributor to downwind ozone
nonattainment.  

How were the wind patterns developed?  For example, there are numerous
monitoring locations between Oklahoma and Michigan (including large
metropolitan areas such as Kansas City and St. Louis).  None of these
monitors showed a significant contribution from Oklahoma.  [Oklahoma
Municipal Power Authority (OMPA) (EPA-HQ-OAR-2009-0491-4573), p. 2.]

RESPONSE:  The model meteorological estimates from MM-5 used as inputs
for CAMx are described in EPA-HQ-OAR-2009-0491-4140, the "Air Quality
Modeling Final Rule Technical Support Document."  Furthermore, examining
the state-by-state ozone contribution estimates from the 2012 base case
air quality modeling for the final rule, Oklahoma clearly contributes at
or above the 1% contribution threshold to a number of monitor-receptors
in downwind states (e.g., Texas, Missouri, Indiana, Illinois, and
Michigan) (see the “Contributions of 8-hour ozone, annual PM2.5, and
24-hour PM2.5 from each state to each monitoring site” excel workbook
accessible on the EPA transport rule website for the contributions. 
This is part of EPA-HQ-OAR-2009-0491-4228 in the docket to this rule). 
However, with the exception of the Allegan, Michigan receptor, these
monitors were modeled not to have nonattainment and/or maintenance
issues.  Thus, they were not further used to examine significant
contribution and/or interference with maintenance in the Transport Rule.

Comment:  Have there been any third party evaluations of the results of
the CAMx model?  The results of this model will subject the State of
Oklahoma, cities within the state, and businesses to numerous
requirements costing billions of dollars.  The results of this model
need to be verified.  [Oklahoma Municipal Power Authority (OMPA)
(EPA-HQ-OAR-2009-0491-4573), p. 2.]

RESPONSE:  There have been a number of third party evaluations of CAMx
model predictions.  Several of these are listed below.  Additional
information on the development and use of CAMx can be found on the web
site:   HYPERLINK "http://www.camx.com"  www.camx.com .

(1) Tesche, T. W.; Morris, R.; Tonnesen, G.; McNally, D.; Boylan, J.;
Brewer, P.,

CMAQ/CAMx annual 2002 performance evaluation over the eastern US.
Atmospheric

Environment 2006, 40(26), 4906-4919.

(2) Hogrefe, G., Civeroio, K.L., Hao, W., Ku, J-Y., Zalewsky, E.E., and
Sistla, G., Rethinking the Assessment of Photochemical Modeling Systems
in Air Quality Planning Applications. Air & Waste Management Assoc.,
58:1086-1099, 2008.

(3) Implementation of State-of-Science PM Modules into the PMCAMx
Photochemical Grid Model.R.E. Morris, G. Yarwood, C. Emery, Spyros
Pandis, and F. Lurmann. Presented at the 96th Annual Conference and
Exhibition of the A&WMA, San Diego, California (June 2003).

(4) Arunachalam, S. Peer Review of Source Apportionment Tools in CAMx
and CMAQ, EP-D-07-102. University of North Carolina, Institute for the
Environment, August 2009.

Budget/Variability/Assurance Methods

No comments received on this issue. 

Allocation Methods

No comments received on this issue. 

Emission Inventories

Commenter asserts that the types of sources affected by the SNPR should
be expanded so that the burden of precursor emissions reduction is more
widely shared.  

The Midwest has numerous Title V stationary sources that rely on
coal-generated heat for process boilers. Notable among these are grain
handling and processing facilities, of which there are several in Iowa.
For Iowa and other Midwestern states, the focus on fossil fuel-fired
energy facilities is too narrow. One major source, a value-added grain
products facility, is proposing to increase the stack height of a main
stack in order to solve local problems. However, the NOX and volatile
organics from this source will continue to blow downwind, barring
upgrades to NOX controls under other rules and the current SIP call for
Muscatine.  [Iowa Environmental Council (EPA-HQ-OAR-2009-0491-4572), p.
2.]

RESPONSE:  This issue is beyond the scope of the Supplemental
rulemaking.

Although the cost curves presented in the Transport Rule only included
EGU reductions, EPA also assessed the cost of NOX emission reductions
available for source categories other than EGUs in the rulemaking.  This
assessment for the Transport Rule concluded that there are little or no
reductions available from non-EGUs sources, such as those noted by the
commenter, at costs lower than the $500/ton threshold for NOX.  

Kansas Proposed SIP Call

No comments received on this issue. 

Other Out of Scope Comments

Commenter expresses concern that the SNPR does not include Nebraska and
South Dakota for ozone season NOX emissions.  

At the current 75 ppb standard, Iowa does not have any nonattainment
areas for ozone, which is the primary driver for the rule and the
Supplemental Action. However, a 65 ppb standard could mean nonattainment
in several Iowa cities, and an even lower standard of 60 ppb would put
much of the state in nonattainment. Clearly Iowa’s own upwind air shed
plays a role in the State’s ozone levels, and these may become
problematic when the new NAAQS standard comes into play—as it should,
given the health benefits. [Iowa Environmental Council
(EPA-HQ-OAR-2009-0491-4572), p. 1-2.]

RESPONSE:  The Transport Rule and the SNPR address ozone transport under
the 1997 ozone standard.  These comments regarding transport to areas at
potential 60 or 65 ppb standards are thus beyond the scope of the rule.

Commenter expresses concern that the SNPR does not anticipate or address
scenarios associated with a revised 8-hour NAAQS for ozone.    

EPA’s online materials (fact sheets, presentations) do not make
mention of any scenarios under a revised eight-hour NAAQS. And yet these
clearly are factors in determining ozone nonattainment. It is likely
that many more downwind counties than the ones shown in the presentation
at http://www.epa.gov/airtransport/pdfs/CSAPRPresentation.pdf will be
affected by nonattainment designations once the NAAQS is changed.  [Iowa
Environmental Council (EPA-HQ-OAR-2009-0491-4572), p. 2.]

RESPONSE:  The Transport Rule and the SNPR address ozone nonattainment
and maintenance for the 1997 ozone standards.  Because revised NAAQS are
beyond the scope of this rulemaking, and because the levels of the NAAQS
under these revisions are not yet known, it would not be appropriate to
include consideration of them in our online materials for this rule. 
Other communications materials will address ozone attainment under
subsequent standards.  For example, EPA will be developing online
materials for area designations under the 2008 ozone standard of 75 ppb.
 

(C)  EPA’s opportunity for comment should allow for changes in
coverage in the other direction (adding states). 

(1) Finally, we note that EPA’s SNPR provides opportunity for comment
for changes in coverage of the rule only in one direction. In
particular, opportunity for comment is provided with respect to states
linked to downwind monitors showing nonattainment and maintenance
problems in the revised analysis for CSAPR that were not so linked in
the prior analysis in the August 2010 proposed rule, but not for the
reverse. That is, EPA has not provided an opportunity for interested
persons to challenge the removal of a state from the rule’s
requirements as proposed last August on the grounds that the downwind
monitors to which those states were then linked did not evidence
nonattainment and maintenance problems in the revised CSAPR analysis. We
believe that EPA should adopt an even-handed approach to changes in
requirements of the rule resulting from revised analyses, both with
respect to subjecting additional states to the rule and to removing
states from coverage of the rule. While we are not requesting that EPA
reopen this rulemaking solely to address this issue, we do request the
Agency to consider this issue of fundamental fairness in future
rulemakings (or in further iterations of this one).  [Clean Air Task
Force, et al. (EPA-HQ-OAR-2009-0491-4588), p. 3.]

RESPONSE:  Comment noted and appreciated.  EPA will take this comment
into consideration in future rulemakings.

 First Energy. FirstEnergy Successfully Concludes Emission Allowance
Auction. Press release. November 18, 2011. Available on the Internet at
<http://www.fescsaprauction.info/Portals/0/Documents/News/2011-11-18%20F
ES%20Successfully%20Concludes%20Emission%20Allowance%20Auction.pdf>.

 First Energy. FirstEnergy Successfully Concludes Emission Allowance
Auction. Press release. November 18, 2011. Available on the Internet at
<http://www.fescsaprauction.info/Portals/0/Documents/News/2011-11-18%20F
ES%20Successfully%20Concludes%20Emission%20Allowance%20Auction.pdf>.

 First Energy. FirstEnergy Successfully Concludes Emission Allowance
Auction. Press release. November 18, 2011. Available on the Internet at
<http://www.fescsaprauction.info/Portals/0/Documents/News/2011-11-18%20F
ES%20Successfully%20Concludes%20Emission%20Allowance%20Auction.pdf>.

 Testimony of Chairman Jon Wellinghoff, Federal Energy Regulatory
Commission, Before the House Subcommittee on Energy and Power of the
Committee on Energy and Commerce United States House of Representatives,
September 14, 2011.

 PAGE 1 

