 6560-50-P

ENVIRONMENTAL PROTECTION AGENCY

40 CFR Parts 51 and 52

[Docket ID No. EPA-HQ-OAR-2005-0163; FRL- xxxx-x]

RIN-2060-AN28

Supplemental Notice of Proposed Rulemaking for Prevention of Significant
Deterioration and Nonattainment New Source Review:  Emission Increases
for Electric Generating Units

AGENCY:  Environmental Protection Agency (EPA).

ACTION:  Supplemental Notice of Proposed Rulemaking.

SUMMARY:  This action is a supplemental notice of proposed rulemaking
(SNPR) to EPA’s October 20, 2005 notice of proposed rulemaking (NPR). 
In the October 2005 NPR, EPA (we) proposed to revise the emissions test
for existing electric generating units (EGUs) that are subject to the
regulations governing the Prevention of Significant Deterioration (PSD)
and nonattainment major New Source Review (NSR) programs (collectively
“NSR”) mandated by parts C and D of title I of the Clean Air Act
(CAA or Act).  We proposed three alternatives for the emissions test: a
maximum achievable hourly emissions test, a maximum achieved hourly
emissions test, and an output-based hourly emissions test.  In the NPR,
we did not propose to include, along with any of the revised NSR
emissions tests, any provisions for computing a significant increase or
a significant net emissions increase, although we solicited comment on
retaining such provisions.  In addition, we solicited comment on
whether, if we revised the NSR test to be a maximum achieved emissions
test or output-based emissions test, we should revise the New Source
Performance Standards (NSPS) regulations to include a maximum achieved
emissions test or an output-based emissions test.  This action recasts
the proposed options so that the output test, instead of being an
alternative to the maximum hourly achieved or maximum hourly achievable
tests, becomes an alternative method for sources to implement those two
tests.  This action includes proposed rule language and supplemental
information for the October 2005 proposal as it relates to the major NSR
regulations, including an examination of the impacts on emissions and
air quality that would result were we to finalize one of the
applicability tests proposed in the        October 2005 proposal or in
today’s SNPR, as described below.  This action also proposes two
additional options that were not included in the October 2005 rule.  For
one of the additional options proposed today, which we characterize as
Option 2, we are proposing that an hourly emissions increase test
(either maximum achieved or maximum achievable, each with an
output-based option) would include the significant net emissions
increase test in the current major NSR rules, which is calculated on an
actual-to-projected-actual annual emissions basis.  As Option 3, we are
proposing to retain the current applicability test, but to extend the
period for determining baseline actual emissions at EGUs from 5 to 10
years.  

These proposed regulations interpret the emissions increase component of
the modification test under CAA section 111(a) (4), in the context of
NSR, for existing EGUs.  The proposed regulations would promote the
safety, reliability, and efficiency of EGUs.

Consistent with the primary purpose of the major NSR program, the
proposed regulations balance the economic need of sources to utilize
their existing physical and operating capacity with the environmental
benefit of regulating those emissions increase related to a change. 

	DATES:  Comments.  Comments must be received on or before [INSERT DATE
60 DAYS AFTER PUBLICATION IN THE FEDERAL REGISTER].  Under the Paperwork
Reduction Act, comments on the information collection provisions must be
received by the Office of Management and Budget (OMB) on or before
[INSERT DATE 30 DAYS AFTER PUBLICATION IN THE FEDERAL REGISTER].

Public Hearing:  If anyone contacts us requesting to speak at a public
hearing on or before [INSERT DATE 20 DAYS AFTER PUBLICATION IN THE
FEDERAL REGISTER], we will hold a public hearing approximately 30 days
after publication in the Federal Register.

ADDRESSES:  Submit your comments, identified by Docket ID No.
EPA-HQ-OAR-2005-0163 by one of the following methods:

	  HYPERLINK "http://www.regulations.gov"  http://www.regulations.gov : 
Follow the on-line instructions for submitting comments.

	E-mail: a-and-r-docket@epa.gov.

	Mail:  Attention Docket ID No. EPA-HQ-OAR-2005-0163, U.S. Environmental
Protection Agency, EPA West (Air Docket), 1200 Pennsylvania Avenue, NW,
Mail code: 6102T, Washington, DC 20460.  Please include a total of 2
copies.  In addition, please mail a copy of your comments on the
information collection provisions to the Office of Information and
Regulatory Affairs, Office of Management and Budget (OMB), Attn: Desk
Officer for EPA, 725 17th Street, NW, Washington, DC 20503.  

	Hand Delivery:  U.S. Environmental Protection Agency, EPA West (Air
Docket), 1301 Constitution Avenue, Northwest, Room B102, Washington, DC
20004, Attention Docket ID No. EPA-HQ-OAR-2005-0163.  Such deliveries
are only accepted during the Docket's normal hours of operation, and
special arrangements should be made for deliveries of boxed information.

Instructions.  Direct your comments to Docket ID No.
EPA-HQ-OAR-2005-0163.  EPA's policy is that all comments received will
be included in the public docket without change and may be made
available online at   HYPERLINK "http://www.regulations.gov" 
http://www.regulations.gov  including any personal information provided,
unless the comment includes information claimed to be Confidential
Business Information (CBI) or other information whose disclosure is
restricted by statute.  Do not submit information that you consider to
be CBI or otherwise protected through www.regulations.gov or e-mail. 
The   HYPERLINK "http://www.regulations.gov"  http://www.regulations.gov
  website is an “anonymous access” system, which means EPA will not
know your identity or contact information unless you provide it in the
body of your comment.  If you send an e-mail comment directly to EPA
without going through   HYPERLINK "http://www.regulations.gov" 
http://www.regulations.gov , your e-mail address will be automatically
captured and included as part of the comment that is placed in the
public docket and made available on the Internet.  If you submit an
electronic comment, EPA recommends that you include your name and other
contact information in the body of your comment and with any disk or
CD-ROM you submit.  If EPA cannot read your comment due to technical
difficulties and cannot contact you for clarification, EPA may not be
able to consider your comment.  Electronic files should avoid the use of
special characters, any form of encryption, and be free of any defects
or viruses.  For additional instructions on submitting comments, go to
section B. of the SUPPLEMENTARY INFORMATION section of this document.

Docket.  All documents in the docket are listed in the   HYPERLINK
"http://www.regulations.gov"  http://www.regulations.gov 

index.  Although listed in the index, some information is not publicly
available, i.e., CBI or other information whose disclosure is restricted
by statute.  Certain other material, such as copyrighted material, is
not placed on the Internet and will be publicly available only in hard
copy form.  Publicly available docket materials are available either
electronically in   HYPERLINK "http://www.regulations.gov" 
http://www.regulations.gov  or in hard copy at the U.S. Environmental
Protection Agency, Air Docket, EPA/DC, EPA West Building, Room B102,
1301 Constitution Ave., NW, Washington, DC.  The Public Reading Room is
open from 8:30 a.m. to 4:30 p.m., Monday through Friday, excluding legal
holidays.  The telephone number for the Public Reading Room is (202)
566-1744, and the telephone number for the Air Docket is (202) 566-1742.


NOTE:  The EPA Docket Center suffered damage due to flooding during the
last week of June 2006.  The Docket Center is continuing to operate. 
However, during the cleanup, there will be temporary changes to Docket
Center telephone numbers, addresses, and hours of operation for people
who wish to make hand deliveries or visit the Public Reading Room to
view documents.  Consult EPA's Federal Register notice at 71 FR 38147
(July 5, 2006) or the EPA website at www.epa.gov/epahome/dockets.htm for
current information on docket operations, locations and telephone
numbers.  The Docket Center’s mailing address for U.S. mail and the
procedure for submitting comments to www.regulations.gov are not
affected by the flooding and will remain the same.

FOR FURTHER INFORMATION CONTACT:  Ms. Janet McDonald, Air Quality Policy
Division (C504-03), U.S. Environmental Protection Agency,  Research
Triangle Park, NC  27711, telephone number: (919) 541-1450; fax number:
(919) 541-5509, or electronic mail e-mail address: 
mcdonald.janet@epa.gov.

SUPPLEMENTARY INFORMATION:

I.  General Information  TC "I.  General Information" \f C \l "1"  

A.  Does this action apply to me?  TC "A.  What are the regulated
entities?" \f C \l "2"  

Entities potentially affected by the subject rule for this action are
fossil-fuel fired boilers and turbines serving an electric generator
with nameplate capacity greater than 25 megawatts (MW) producing
electricity for sale.  Entities potentially affected by the subject rule
for this action also include State, local, and tribal governments. 
Categories and entities potentially affected by this action are expected
to include:

Industry Group	SICa	NAICSb

Electric Services	491	221112

Federal government	22112	Fossil-fuel fired electric utility steam
generating units owned by the Federal government

State/local/Tribal government	22112	Fossil-fuel fired electric utility
steam generating units owned by municipalities.  Fossil-fuel fired
electric utility steam generating units in Indian country.

a	Standard Industrial Classification

b	North American Industry Classification System.

B.  Where can I get a copy of this document and other related
information? tc \l2 "B.Where can I get a copy of this document and other
related information? 

In addition to being available in the docket, an electronic copy of this
proposal will also be available on the WWW.  Following signature by the
EPA Administrator, a copy of this notice will be posted in the
regulations and standards section of our NSR home page located at  
HYPERLINK "http://www.epa.gov/nsr"  http://www.epa.gov/nsr .

C.  What should I consider as I prepare my comments for EPA? tc \l2 "C.
How should I submit Confidential Business Information (CBI) to the
Agency? 

1.  Submitting CBI.  Do not submit this information to EPA through  
HYPERLINK "http://www.regulations.gov"  http://www.regulations.gov  or
e-mail.  Clearly mark the part or all of the information that you claim
to be CBI.  For CBI information in a disk or CD ROM that you mail to
EPA, mark the outside of the disk or CD ROM as CBI and then identify
electronically within the disk or CD ROM the specific information that
is claimed as CBI.  In addition to one complete version of the comment
that includes information claimed as CBI, a copy of the comment that
does not contain the information claimed as CBI must be submitted for
inclusion in the public docket.  Information so marked will not be
disclosed except in accordance with procedures set forth in 40 CFR part
2.  Send or deliver information identified as CBI only to the following
address: Roberto Morales, OAQPS Document Control Officer (C404-02), U.S.
EPA, Research Triangle Park, NC 27711, Attention Docket ID No.
EPA-HQ-OAR-2005-0163.

2.  Tips for Preparing Your Comments.  When submitting comments,
remember to:

Identify the rulemaking by docket number and other identifying
information (subject heading, Federal Register date and page number).

Follow directions - The agency may ask you to respond to specific
questions or organize comments by referencing a Code of Federal
Regulations (CFR) part or section number.

Explain why you agree or disagree; suggest alternatives and substitute
language for your requested changes.

Describe any assumptions and provide any technical information and/or
data that you used.

If you estimate potential costs or burdens, explain how you arrived at
your estimate in sufficient detail to allow for it to be reproduced.

Provide specific examples to illustrate your concerns, and suggest
alternatives.

Explain your views as clearly as possible, avoiding the use of profanity
or personal threats.

Make sure to submit your comments by the comment period deadline
identified.

D.  How can I find information about a possible hearing?

People interested in presenting oral testimony or inquiring if a hearing
is to be held should contact Ms. Pamela S. Long, New Source Review
Group, Air Quality Policy Division (C504-03), U.S. EPA, Research
Triangle Park, NC 27711, telephone number (919) 541-0641.  If a hearing
is to be held, persons interested in presenting oral testimony should
notify Ms. Long at least 2 days in advance of the public hearing. 
Persons interested in attending the public hearing should also contact
Ms. Long to verify the time, date, and location of the hearing.  The
public hearing will provide interested parties the opportunity to
present data, views, or arguments concerning these proposed rules.

E.  How is the preamble organized?

The information presented in this preamble is organized as follows:

I.  General Information

A.  Does this action apply to me?

B.  Where can I get a copy of this document and other related
information?

C.  What should I consider as I submit comments to EPA?

D.  How can I find information about a possible public hearing?

E.  How is the preamble organized?	

II.   Overview

A.  Option 1:  Hourly Emissions Increase Test

B.  Option 2:  Hourly Emissions Increase Test Followed by Annual
Emissions Test

C.  Option 3:  10-Year Baseline Period in EGU Annual Emissions Test

III.   Analyses Supporting Proposed Options

A.  The Integrated Planning Model

B.  CAIR/CAMR/CAVR NSR Scenarios – Description of the Scenarios

C.  CAIR/CAMR/CAVR NSR Scenarios – Discussion of Results

IV.   Additional Regulatory History, Analyses, and Legal Basis
Supporting Option 3- Annual Emissions Increase Test with 10-Year
Baseline 

A.  Regulatory History for 5-Year Baseline Period for EGUs

B.  New Information Concerning the Appropriate Baseline Period for
Utilities

C.  EGU Business Cycle

D.  Rationale and Legal Basis

V.  Proposed Regulations for Option 1:  Hourly Emissions Increase Test

A.  Test for EGUs Based on Maximum Achieved Emissions Rates

B.  Test for EGUs Based on Maximum Achievable Emissions

VI.   Proposed Regulations for Option 2:  Hourly Emissions Increase Test
Followed by Annual Emissions Test

VII.   Proposed Regulations for Option 3:  Annual Emissions Test with
10-Year Baseline Period

VIII.   Legal Basis and Policy Rationale

IX.   Statutory and Executive Order Reviews

A.  Executive Order 12866(Regulatory Planning and Review tc \l2 "A. 
Executive Order 12866Regulatory Planning and Review 

B.  Paperwork Reduction Act tc \l2 "B.  Paperwork Reduction Act 

C.  Regulatory Flexibility Act (RFA) tc \l2 "C.  Regulatory Flexibility
Act (RFA) 

D.  Unfunded Mandates Reform Act tc \l2 "D.  Unfunded Mandates Reform
Act 

E.  Executive Order 13132:  Federalism tc \l2 "E.  Executive Order
13132Federalism 

F.  Executive Order 13175:  Consultation and Coordination with Indian
Tribal Governments tc \l2 "F.  Executive Order 13175Consultation and
Coordination with Indian Tribal Governments 

G.  Executive Order 13045:  Protection of Children from Environmental
Health Risks and Safety Risks tc \l2 "G.  Executive Order
13045Protection of Children from Environmental Health Risks and Safety
Risks 

H.  Executive Order 13211:  Actions Concerning Regulations That
Significantly Affect Energy Supply, Distribution, or Use tc \l2 "H. 
Executive Order 13211Actions Concerning Regulations That Significantly
Affect Energy Supply, Distribution, or Use	 

I.  National Technology Transfer and Advancement Act

X.   Statutory Authority

II.   Overview

This action is a SNPR to EPA’s October 20, 2005 (70 FR 61081) NPR.  In
the October 2005 NPR, EPA (we) proposed to revise the emissions test for
existing EGUs that are subject to the regulations governing the PSD and
nonattainment major NSR programs (collectively “NSR”) mandated by
parts C and D of title I of the CAA.  We proposed three alternatives for
the emissions test: a maximum achievable hourly emissions test, a
maximum achieved hourly emissions test, and an output-based hourly
emissions test.  In the NPR, we did not propose to include, along with
any of the revised NSR emissions tests, any provisions for computing a
significant increase or a significant net emissions increase, although
we solicited comment on retaining such provisions.  In addition, we
solicited comment on whether, if we revised the NSR test to be a maximum
achieved emissions test or output-based emissions test, we should revise
the NSPS regulations to include a maximum achieved emissions test or an
output-based emissions test.  This action recasts the proposed options
so that the output test, instead of being an alternative to the maximum
hourly achieved or maximum hourly achievable tests, becomes an
alternative method for sources to implement those two tests. 
Specifically, we propose today that each of the two tests would be
implemented through (i) an input method (as defined below), (ii) the
output method, or (iii) at the source’s choice, either the input or
output method.  This action includes proposed rule language and
supplemental information for the October 2005 proposal as it relates to
the major NSR regulations, including an examination of the impacts on
emissions and air quality that would result were we to finalize one of
the applicability tests proposed in the October 2005 proposal or in
today’s SNPR, as described below. 

This action also proposes two additional options that were not included
in the October 2005 rule.  For convenience, this action characterizes
the tests contained in the October 2005 NPR, described above,  as Option
1 (with the maximum hourly achieved test characterized as Alternatives
1-6 and the maximum hourly achievable test characterized as Alternatives
7-8 within that Option 1, and with each of those tests including
output-based alternatives).  For one of the additional options proposed
today, which we characterize as Option 2, we are proposing that an
hourly emissions increase test (either maximum achieved or maximum
achievable, each with output-based alternatives) would include the
significant net emissions increase test in the current major NSR rules,
which is calculated on an actual-to-projected-actual annual emissions
basis.  As Option 3, we are proposing to retain the current
applicability test, but to extend the period for determining baseline
actual emissions at EGUs from 5 to 10 years.  That is, we are proposing
that the owner or operator would select any consecutive 24-month period
within the 10-year period immediately preceding when the owner or
operator begins actual construction of the project.  To reiterate, this
Option 3—the 10-year baseline period for EGUs—is alternate to
adopting a maximum hourly emissions increase test under either Option 1
or Option 2.  

When we proposed a revised emissions test for EGUs in October 2005, we
referenced United States v. Duke Energy Corp., 411 F.3d 539 (4th Cir.)
rehearing den. __ F.3d __ (2005), cert. granted ___ U.S. ___ (2006).  In
that case, which was handed down on June 15, 2005, the Fourth Circuit
Court of Appeals ruled that EPA must use a consistent definition of the
term “modification” under CAA section 111(a)(4) for the purposes of
both the NSPS program under section 111 of the Act and the PSD program
under part C of the Act.  The Court further ruled that because EPA had
defined the term first in the NSPS regulations through a test based on
increases in a plant’s hourly rate of emissions, the PSD regulations
had to be interpreted to include a consistent hourly test.  We continue
to respectfully disagree with the Fourth Circuit’s decisions in Duke
Energy and continue to believe that we have the authority to define
“modification” differently in the NSPS and NSR programs.  However,
we believe that the options for a maximum hourly test that we proposed
in our October 2005 NPR and in today’s SNPR are an appropriate
exercise of our discretion.  In addition, we believe that we have
authority to propose the 10-year baseline option in this action.  

At the time of our proposal, the Fourth Circuit had denied the United
States’ petition for rehearing on the decision in Duke Energy, but the
deadline for filing a petition for certiorari to the United States
Supreme Court had not yet run.  Subsequently, on December 28, 2005,
Intervenor plaintiffs Environmental Defense Fund, North Carolina Sierra
Club, and North Carolina Public Interest Research Group filed a petition
for certiorari asking the court to address several matters.  On May 15,
2006 the United States Supreme Court granted the petition for a writ of
certiorari.  Of course, it is unclear at present, what, if any, impact a
Supreme Court decision would have on the rulemaking on which we take
supplemental action today.  We continue to believe that providing an
hourly emissions test has particular merit for EGUs.  Accordingly, we
continue to pursue the viability of imposing an hourly emissions test on
EGUs for purposes of major NSR applicability.  We will, of course,
conform our final rule, to the extent required, to the decision the
Supreme Court renders. 

  SEQ CHAPTER \h \r 1 In May 2001, President Bush’s National Energy
Policy Development Group issued findings and key recommendations for a
National Energy Policy.  This document included numerous recommendations
for action, including a recommendation that the EPA Administrator, in
consultation with the Secretary of Energy and other relevant agencies,
review NSR regulations, including administrative interpretation and
implementation.  The recommendation requested that we issue a report to
the President on the impact of the regulations on investment in new
utility and refinery generation capacity, energy efficiency, and
environmental protection.  Our report to the President and our
recommendations in response to the National Energy Policy were issued on
  June 13, 2002.  A copy of this information is available at   HYPERLINK
"http://www.epa.gov/nsr/publications.html" 
http://www.epa.gov/nsr/publications.html . 

 In that report we concluded:

As applied to existing power plants and refineries, EPA concludes that
the NSR program has impeded or resulted in the cancellation of projects
which would maintain and improve reliability, efficiency and safety of
existing energy capacity.  Such discouragement results in lost capacity,
as well as lost opportunities to improve energy efficiency and reduce
air pollution.  (New Source Review Report to the President at pg. 3.)

On December 31, 2002, we promulgated final regulations that implemented
several of the recommendations in the New Source Review Report to the
President.  However, that action left the NSR regulations as they
related to utilities largely unchanged.  This action continues to
address the recommendations in the New Source Review Report to the
President as they relate to electric utilities specifically and in light
of the regulatory requirements for EGUs that have been promulgated since
our 2002 regulations. 

The regulations proposed in the October 2005 NPR and today would promote
the safety, reliability, and efficiency of EGUs.  The proposed
regulations are consistent with the primary purpose of the major NSR
program, which is not to reduce emissions, but to balance the need for
environmental protection and economic growth.  The proposed regulations
reasonably balance the economic need of sources to use existing physical
and operating capacity with the environmental benefit of regulating
those emissions increases related to a physical or operational change. 
This is particularly true in light of the substantial national EGU
emissions reductions that other programs have achieved or are expected
to achieve, which we described in detail at 70 FR 61083.  Moreover, as
the analyses included in today’s SNPR demonstrate, the proposed
regulations would not have an undue adverse impact on local air quality.

This section gives an overview of our proposed actions for major NSR
applicability at existing EGUs, including the proposals in the NPR, as
recast today, for the maximum hourly emissions tests and today’s
additional proposals.  Each of the options would promote the safety,
reliability, and efficiency of EGUs.  Each of the options would also
balance the economic need of sources to use existing physical and
operating capacity with the environmental benefit of regulating those
emissions increases related to a change, considering the substantial
national emissions reductions other programs have achieved or will
achieve from EGUs.  Each option also has additional merits that we
articulate in sections II A though C of this preamble.  The additional
merits among the various options differ.  For this reason, we are not
identifying today a preferred option.  We will select the final option
after weighing the public comments on the Options.  Table 1 summarizes
our three Options.  

Table 1.  Proposed Options for Major NSR Applicability for Existing EGU

Option 1	Step 1:  Physical Change or Change in the Method of Operation

Step 2:  Hourly Emissions Increase Test

Alternative 1 – Maximum achieved hourly emissions; statistical
approach; input basis

Alternative 2 – Maximum achieved hourly emissions; statistical
approach; output basis

Alternative 3 – Maximum achieved hourly emissions; two-in-5-year
baseline; input basis

Alternative 4 – Maximum achieved hourly emissions; two-in-5-year
baseline; output basis

Alternative 5 – Maximum achieved hourly emissions; one-in-5-year
baseline; input basis

Alternative 6 – Maximum achieved hourly emissions; one-in-5-year
baseline; output basis

Alternative 7 – NSPS test – maximum achievable hourly emissions;
input basis

Alternative 8 – NSPS test – maximum achievable hourly emissions;
output basis

Option 2	Step 1:  Physical Change or Change in the Method of Operation

Step 2:  Hourly Emissions Increase Test

Alternative 1 – Maximum achieved hourly emissions; statistical
approach; input basis

Alternative 2 – Maximum achieved hourly emissions; statistical
approach; output basis

Alternative 3 – Maximum achieved hourly emissions; two-in-5-year
baseline; input basis

Alternative 4 – Maximum achieved hourly emissions; two-in-5-year
baseline; output basis

Alternative 5 – Maximum achieved hourly emissions; one-in-5-year
baseline; input basis

Alternative 6 – Maximum achieved hourly emissions; one-in-5-year
baseline; output basis

Alternative 7 – NSPS test – maximum achievable hourly emissions;
input basis

Alternative 8 – NSPS test – maximum achievable hourly emissions;
output basis

Step 3: Significant Emissions Increase Determined Using the
Actual-to-Projected-Actual Emissions Test as in the Current Rules

Step 4:  Significant Net Emissions Increase as in the Current Rules

Option 3	Step 1:  Physical Change or Change in the Method of Operation

Step 2: Significant Emissions Increase Determined Using the
Actual-to-Projected-Actual Emissions Test with 10-Year Baseline

Step 3:  Significant Net Emissions Increase as in Current Rules



We request public comment on all aspects of this action.  Many
commenters have already indicated a strong preference for an annual
emissions increase test, either in conjunction with an hourly emissions
increase test or in lieu of an hourly emissions increase test.  Among
the three alternatives that we proposed for the hourly emission increase
test, many commenters preferred the maximum achievable hourly emissions
increase test.  Therefore, to fully evaluate public preferences among
the three Options, we solicit input on whether commenters prefer Option
1 or Option 3.  In particular, we solicit comment on whether commenters
prefer Option 1, Alternatives 1-6, Maximum Achieved Hourly Emissions
Test, or Option 3, Annual Emissions Test with 10-Year Baseline.  We
intend to finalize either Option 1, Option 2, or Option 3.  If we
finalize Option 1 or Option 2 we will also finalize either the maximum
achieved or the maximum achievable alternative.  We intend to respond to
public comments on the October 20, 2005 NPR and this actions in a single
Federal Register Notice and Response to Comments Document.

A.  Option 1:  Hourly Emissions Increase Test

In this action, we are providing regulatory language, data, and
additional information in support of our proposed rule, published by
notice dated October 20, 2005, “Prevention of Significant
Deterioration, Nonattainment Major New Source Review, and New Source
Performance Standards: Emissions Test for Electric Generating Units.” 
   (70 FR 61081.)  In the October 2005 NPR, we proposed to revise the
emissions test for existing EGUs that are subject to the regulations
governing the major NSR programs mandated by parts C and D of title I of
the CAA.  We proposed to adopt an hourly emissions increase test and to
remove the requirement to compute a significant emissions increase and a
significant net emissions increase on an annual basis.  We also proposed
three alternatives for an hourly emissions test: a maximum achievable
hourly emissions test, a maximum achieved hourly emissions test, and an
output-based hourly emissions test.  In today’s SNPR, we have grouped
those alternatives in our October 2005 proposal into Option 1.  In doing
so, we have recast the output option, as described above. 

The proposed maximum hourly achieved test would streamline NSR
applicability determinations.  The proposed maximum hourly achievable
test provides even more streamlining by conforming NSR applicability
determinations to NSPS applicability determinations.  We also note the
achieved and achievable tests eliminate the burden of projecting future
emissions and distinguishing between emissions increases caused by the
change from those due solely to demand growth, because any increase in
the emissions under the hourly emissions tests would logically be
attributed to the change.  Both the achieved and achievable tests reduce
recordkeeping and reporting burdens on sources because compliance will
no longer rely on synthesizing emissions data into rolling average
emissions.  Option 1 would reduce the reviewing authorities’
compliance and enforcement burden.

Consistent with our policy goal of improving energy efficiency, we are
proposing both an input and output based format for both the maximum
achievable and maximum achieved hourly emissions increase test options. 
Specifically, we are proposing the alternatives of (i) use of
input-based methodology for each test, (ii) use of output-based
methodology for each test, or (iii) allowing the source to choose
between input- or output-based methodology.  Some commenters strongly
opposed an output-based format, believing that it would encourage
emissions increases.  We believe these concerns are mitigated in a
system where total annual emissions are capped nationally.  Other
commenters supported the output based format, noting that it would
encourage energy efficiency.

We agree that an output-based test encourages efficient units, which has
well-recognized benefits.  The more efficient an EGU, the less it emits
for a given period of operation.  For example, a 50 MW combustion
turbine that operates 500 hours a year, for 25,000 MWh per year at an
emission rate of 75 ppm, would emit 46 tons per year at 25 percent
efficiency, 41 tons per year at 28 percent efficiency, 37 tons per year
at 31 percent efficiency, and 34 tons per year at 34 percent efficiency.

Furthermore, we have established pollution prevention as one of our
highest priorities.  One of the opportunities for pollution prevention
is maximizing the efficiency of energy generation.  An output-based
standard establishes emission limits in a format that incorporates the
effects of unit efficiency by relating emissions to the amount of useful
energy generated, not the amount of fuel burned.  By relating emission
limitations to the productive output of the process, output-based
emission limits encourage energy efficiency because any increase in
overall energy efficiency results in a lower emission rate.  Allowing
energy efficiency as a pollution control measure provides regulated
sources with an additional compliance option that can lead to reduced
compliance costs as well as lower emissions.  The use of more efficient
technologies reduces fossil fuel use and leads to multi-media reductions
in environmental impacts both on-site and off-site.  On-site benefits
include lower emissions of all products of combustion, including
hazardous air pollutants, as well as reducing any solid waste and
wastewater discharges.  Off-site benefits include the reduction of
emissions and non-air environmental impacts from the production,
processing, and transportation of fuels.

While output-based emission limits have been used for regulating many
industries, input-based emission limits have been the traditional method
to regulate steam generating units.  However, this trend is changing as
we seek to promote pollution prevention and provide more compliance
flexibility to combustion sources.  For example, in 1998 we amended the
NSPS for electric utility steam generating units (40 CFR part 60,
subpart Da) to use output-based standards for NOx (40 CFR 63.44a, 62 FR
36954, and  63 FR 49446).  We recently promulgated new output-based
emission limits for SO2 and NOx under subpart Da of 40 CFR part 60 (71
FR 9866) and for combustion turbines.   (71 FR 38482.)  

The proposed maximum hourly achieved and maximum achievable tests would
streamline and simplify NSR applicability determinations by reducing the
number of steps involved.  The proposed maximum hourly achievable test
provides more streamlining by conforming NSR applicability
determinations to NSPS applicability determinations.  We also note that
Option 1 (both achieved and achievable alternatives) eliminates the
burden of projecting future emissions and distinguishing between
emissions increases caused by the change from those due solely to demand
growth, because any increase in the emissions under the maximum
achievable emissions test would logically be attributed to the change. 
It reduces recordkeeping and reporting burdens on sources because
compliance will no longer rely on synthesizing emissions data into
rolling average emissions.  Option 1 would reduce the reviewing
authorities’ compliance and enforcement burden.

In the October 2005 NPR, we also solicited comment on whether, if we
revised the NSR test to be a maximum achieved emissions test or
output-based emissions test, we should revise the NSPS regulations to
include a maximum achieved emissions test or an output-based emissions
test.  This SNPR concerns the emissions test for existing EGUs in the
major NSR programs.  It does not address the emissions test for existing
EGUs under the NSPS program.  

B.  Option 2:  Hourly Emissions Increase Test Followed by Annual
Emissions Test

In the NPR, we did not propose to include, along with any of the revised
NSR emissions tests, any provisions for computing a significant
emissions increase or a significant net emissions increase, although we
solicited comment on retaining such provisions.  Many commenters
believed netting is required under the Alabama Power Court decision, and
supported options retaining netting.  Under Option 2, today we are
proposing that major NSR applicability would include an hourly emissions
increase test as described in Option 1 above, followed by the current
regulatory requirements for the actual-to-projected-actual emissions
increase test to determine significance, and the significant net
emissions increase test.  Specifically, under Option 2, the major NSR
program would include a four-step process as follows:  (1) physical
change or change in the method of operation; (2) hourly emissions
increase test (as described under Option 1); (3) significant emissions
increase as in the current major NSR regulations; and (4) significant
net emissions increase as in the current major NSR regulations.  Section
VI of this preamble describes Option 2 in more detail.  

We recognize that Option 2 does not offer the benefits of streamlining
major NSR program to the extent that would occur under Option 1.  We
propose Option 2 for the purpose of maintaining the current significant
net emissions increase component of the emissions increase test.

C.  Option 3:  10-Year Baseline Period in EGU Annual Emissions Test

Many commenters suggested an annual emissions test as an alternative to
an hourly emissions test.  In this action, we are also proposing an
additional option that was not included in the October 2005 NPR, which
is to allow EGUs to use the baseline period that other major stationary
sources use.  Specifically, we are proposing to retain the annual
actual-to-projected-actual emissions increase test for electric utility
steam generating units (EUSGUs) in the current regulations, but to
extend the period in which baseline actual emissions are calculated from
5 to 10 years.  That is, we are proposing that for EGUs, like other
major stationary sources, baseline actual emissions would be determined
as average annual emissions over a 24-month period within the 10-year
period immediately preceding when the owner or operator begins actual
construction of the project.

Many commenters preferred an annual emissions test to an hourly
emissions test.  As discussed below, we believe a 10-year baseline
period more closely mirrors the business cycle for EGUs, and therefore
is a better period than the current 5-year period to allow EGUs to
determine baseline emissions.  This is particularly so in light of the
fluctuations likely in EGU utilization during the CAIR implementation
period, but it is also generally true considering the variations in
weather, economic activity, and system-wide demand.

Option 3 has the benefit of retaining netting, and in a manner less
complex than that of Option 2.  It also has the benefit of conforming
the applicability test for EGUs to that for other source categories.  

III.   Analyses Supporting Proposed Options

We examined how our three proposed options for major NSR applicability
for EGUs would affect control technology installation, emissions, and
air quality.  We conducted two separate analyses using the Integrated
Planning Model (IPM).  Our analyses show that none of the proposed
options would have a detrimental impact on county-level emissions or
local air quality.  This section discusses our analyses and findings. 
More extensive information on our analyses is available in the Technical
Support Document, which is available in Docket ID No.
EPA-HQ-OAR-2005-0163.

A.  The Integrated Planning Model  

We use the IPM to analyze the projected impact of environmental policies
on the electric power sector in the 48 contiguous states and the
District of Columbia.  The IPM is a multi-regional, dynamic,
deterministic linear programming model of the entire electric power
sector.  It provides forecasts of least-cost capacity expansion,
electricity dispatch, and emission control strategies for meeting energy
demand and environmental, transmission, dispatch, and reliability
constraints.  We have used the IPM extensively to evaluate the cost and
emissions impacts of proposed policies to limit emissions of sulfur
dioxide (SO2), nitrogen oxide (NOx), carbon dioxide (CO2), and mercury
from the electric power sector.  The IPM was a key analytical tool in
developing the Clean Air Interstate Regulation (CAIR; see 70 FR 25162). 
However, the IPM capabilities and results are not limited to projections
for CAIR states.  It includes data for and projects emissions and
controls for the electric sector in the contiguous United States.

Each IPM run is based on emissions controls on existing units, state
regulations, cost and performance of generating technologies, SO2 and
NOx heat rates, natural gas supply and prices, and electricity demand
growth assumptions.  This input is updated on a regular basis.  We most
recently used the IPM to project EGU SO2 and NOx controls, emissions,
and air quality in 2020 considering projected emission controls under
the Clean Air Interstate Rule, Clean Air Mercury Rule, and Clean Air
Visibility Rule.  For convenience, we refer to this projection as the
CAIR/CAMR/CAVR 2020 Base Case Scenario or, more simply, the Base Case
Scenario.  The IPM model used for this scenario is IPM v.2.1.9. 
Complete documentation for IPM, including the Base Case Scenario, is
available at   HYPERLINK "http://www.epa.gov/airmarkets/epa-ipm" 
http://www.epa.gov/airmarkets/epa-ipm . 

The IPM v 2.1.9 contains specific information about 13,814 EGUs,
including 1,242 coal-fired EGUs.  This represents all existing EGUs in
the contiguous United States as of 2004, as well as new units that are
already planned or committed, and new units that are projected to come
online by 2007.  The IPM v.2.1.9 contains geographic location, fuel use,
emissions control, and other data on each existing EGU based on the
National Electric Energy Data System (NEEDS).  NEEDS data for existing
EGUs comes from information submitted to the Department of Energy’s
Energy Information Agency, on Forms EIA 860 and 767.  That is, each
existing EGU in the IPM v.2.1.9 is an actual EGU currently in operation
and the information on that EGU in the IPM is real data that has been
submitted to the DOE.  

The IPM v.2.1.9 model also accounts for growth in the EGU sector that is
projected to occur through new builds, including both planned committed
units, and potential units.  Planned-committed EGUs are those that are
likely to come online because ground has been broken, financing obtained
or other demonstrable factors indicate a high probability that the EGU
will come online.  Planned-committed units in IPM v.2.1.9 were based on
two information sources: RDI NewGen database (RDI) distributed by Platts
(www.platts.com) and the inventory of planned-committed units assembled
by DOE, Energy Information Administration, for their Annual Energy
Outlook.  Potential EGUs are those units that may be built at a future
date in response to electricity demand.  In IPM v.2.1.9, potential new
units are modeled as additional capacity and generation that may come
online in each model region.	

IPM v.2.1.9 also accounts for emission limitations due to State
regulations and enforcement actions.  It includes State regulations that
limit SO2 and NOx emissions from EGUs.  These are included Appendix 3-2,
available at   HYPERLINK "http://www.epa.gov/airmarkets/epa-ipm/" 
http://www.epa.gov/airmarkets/epa-ipm/ .  The IPM v.2.1.9 includes NSR
settlement requirements for the following six utility companies: SIGECO,
PSEG Fossil, TECO, We Energies (WEPCO), VEPCO and Santee Cooper.  The
settlements are included as they existed on March 19, 2004.  A summary
of the settlement agreements is included in Appendix 3-3 of the IPM
documentation and is available at   HYPERLINK
"http://www.epa.gov/airmarkets/epa-ipm/" 
http://www.epa.gov/airmarkets/epa-ipm/ .  

Absent specially designed inputs, IPM does not project that any existing
EGUs will make physical or operational changes that increase their
efficiency, generation, or emissions.  Such predictions for individual
EGUs are beyond the capabilities of IPM.  Therefore, IPM does not
predict which sources will be subject to the major NSR applicability
requirements.  However, as discussed below, EPA has specially designed
inputs to IPM that provide useful information directly related to major
NSR applicability requirements. 

Reliability is a critical element of power plant operation.  Reliability
is generally defined as whether an EGU is able to operate over sustained
periods at the level of output required by the utility.  One measure of
reliability is availability, the percentage of total time in a given
period that an EGU is available to generate electricity.  An EGU is
available if it is capable of providing service, regardless of the
capacity level that can be provided.  Availability is generally measured
using the number of hours that an EGU operates annually.  For example,
if an EGU operated 8760 hours in a particular year, it was 100 percent
available.  Each year, EGUs are not available for some number of hours
due to planned outages, maintenance outages, and forced outages.  

IPM v.2.1.9 uses information from the North American Electric
Reliability Council (NERC)’s Generator Availability Data System (GADS)
to determine the annual availability for EGUs.  The GADS database
includes operating histories—some dating back to the early
1960’s—for more than 6,500 EGUs.  These units represent more than 75
percent of the installed generating capacity in the United States and
Canada.  Each utility provides reports, detailing its units’ operation
and performance.  The reports include types and causes of outages and
deratings, unit capacity ratings, energy production, fuel use, and
design information.  GADS provides a standard set of definitions for
determining how to classify an outage on a unit, including planned
outages, maintenance outages, and forced outages.  The GADS data are
reported and summarized annually.  A planned outage is the removal of a
unit from service to perform work on specific components that is
scheduled well in advance and has a predetermined start date and
duration (for example, annual overhaul, inspections, testing.)  Turbine
and boiler overhauls or inspections, testing, and nuclear refueling are
typical planned outages.

A maintenance outage is the removal of a unit from service to perform
work on specific components that can be deferred beyond the end of the
next weekend, but requires the unit be removed from service before the
next planned outage.  Typically, maintenance outages may occur any time
during the year, have flexible start dates, and may or may not have
predetermined durations.  For example, a maintenance outage would occur
if an EGU experiences a sudden increase in fan vibration.  The vibration
is not severe enough to remove the unit from service immediately, but
does require that the unit be removed from service soon to check the
problem and make repairs.

A forced outage is an unplanned component failure or other breakdown
that requires the unit be removed from service immediately, that is,
within 6 hours, or before the end of the next weekend.  A common cause
of forced outages is boiler tube failure.  

Each EGU must report the number of hours due to planned outages,
maintenance outages, and forced outages to NERC annually.  NERC
summarized the data for all EGUs over the period from 2000 - 2004 in its
Annual Unit Performance Statistics Report, available at   HYPERLINK
"http://www.nerc.com/~gads/"  http://www.nerc.com/~gads/ .  For the
years 2001 - 2004, the average annual planned outage hours for all EGUs
was 572.09 (about 23 days.), the average annual maintenance outage hours
for all EGUs was 156.27 (about 6 days), and the average annual forced
outage hours for all coal-fired EGUs was 348.75 (about 14 days).  The
total annual unavailable hours were 1,087.57, which is 15.1 percent of
the total annual hours of 8760.  Based on this data, the IPM v.2.1.9
assumed coal-fired EGUs were 85 percent available.  As just noted, of
the 1,087.57 total unavailable hours, 348.75 were forced outage hours,
which means that coal-fired EGUs were unavailable due to forced outages
approximately 4 percent of the hours in a year for the years 2000 -
2004.

We recently released a graphic presentation of electric power sector
results under CAIR/CAMR/CAVR.  Entitled “Contributions of
CAIR/CAMR/CAVR to NAAQS Attainment: Focus on Control Technologies and
Emission Reductions in the Electric Power Sector,” it is available at 
 HYPERLINK "http://www.epa.gov/airmarkets/cair/analyses.html" 
http://www.epa.gov/airmarkets/cair/analyses.html .  As this presentation
shows, under the CAIR/CAMR/CAVR 2020 Base Case Scenario, local SO2 and
NOx emissions generally decrease, average SO2 and NOx emission rates
decrease, national SO2 and NOx emissions decrease, and most coal-fired
EGUs will have scrubbers by 2020.  Approximately half of the coal-fired
EGUs will have SCR or SNCR by 2020, and the majority of the remaining
coal-fired EGUs will have combustion control. These effects occur
throughout the contiguous 48 states, not just in the CAIR States.

B.  CAIR/CAMR/CAVR NSR Availability Scenarios – Description of the
Scenarios

We developed IPM scenarios to examine the effects of our proposed
regulations, including the maximum hourly emissions increase tests
(achievable and achieved, on an input and output basis) and the 10-year
annual emissions test, on EGU emissions and control technologies.  These
new IPM scenarios incorporate the parameters used in the IPM model
v.2.1.9 that we describe above, including information for the electric
sector in the contiguous United States.  Thus, these new IPM scenarios
revise the parameters in the CAIR/CAMR/CAVR 2020 Base Case Scenario
consistent with the way EGUs might operate under the proposed major NSR
applicability changes.  

The primary difference between the current applicability test and the
proposed tests is that under the proposed tests, sources could more
readily make repairs or improvements that prevent forced outages, and
thereby allow the source to operate more hours.  These repairs allow the
source to operate at the higher availability level that it achieved
before its equipment degraded so much as to cause more forced outages. 

Some commenters emphasized this difference between the current
applicability test and our proposals in the NPR.  They explained that
because, as we noted at 70 FR 61100, hours of operation are considered
in determining annual emissions under the actual-to-projected-actual
test in the current major NSR program but have no role in any of our
proposed hourly emissions increase test options, an EGU could make a
change that does not increase the maximum hourly emissions rate, but
does allow the source to run more hours.  This change would not trigger
review under a maximum hourly emissions increase test in any case, but
in some cases might trigger review under the current major NSR emissions
increase test based on annual emissions with a 5-year baseline period. 
These commenters assert that the proposed applicability tests could
allow substantial increases in annual emissions without triggering NSR.

For several reasons, we believe commenters have overstated the
likelihood that substantial increases in annual emissions and resulting
deterioration in air quality would occur under the proposed maximum
hourly emissions tests, as opposed to the current annual emissions,
5-year baseline test.  First, an EGU can increase its hours of operation
under the current regulations, as long as it does not make a physical
change or change in the method of operation.  Information from the RBLC
confirms that most EGUs are already permitted to run 8760 hours
annually.  Thus, increases in hours of operation at various EGUs
frequently occur under the current regulations without triggering major
NSR.  Second, increases in actual emissions stemming from increases in
hours of operation that are unrelated to the change, are not considered
in determining projected actual emissions.  To the extent that changes
resulting in increased hours would occur under the proposed regulatory
scheme, any resulting increases in emissions will be diminished as the
CAIR and BART programs are implemented and the SO2 and NOx emissions for
most EGUs are capped.  As we described in detail in the NPR, 70 FR
61087, national and regional caps limit total actual annual EGU SO2 and
NOx emissions.  These caps greatly reduce the significance of hours of
operations on actual emissions from the sector nationally.  Furthermore,
as we indicated in our recent report of the CAIR/CAMR/CAVR, the more
hours an EGU operates, the more likely it is to install controls. 
Moreover, existing synthetic minor limits to avoid major NSR and
enforceable limits on hours of operation on a particular EGU as a result
of netting would remain in place under any revised emissions increase
test.  We thus believe the opportunities for many EGUs to significantly
increase their emissions through higher hours of operation under a
maximum hourly emissions increase test, as compared to the current
annual emissions increase test with a 5-year baseline period, are
generally limited.  

Increasing the baseline period to 10 years, as we are proposing, could
also afford the opportunity for an EGU owner/operator to select a
baseline period with higher hours of operation than would be available
in a 5-year period.  However, we do not believe that this proposal will
result in significantly higher emissions than would occur under the
current applicability test.  The same reasons as cited immediately above
apply here, as well.  In addition, under the current regulations EGUs
can select a baseline period beyond the 5-year period that is more
representative of normal operation, including higher hours of operation.
 We thus believe opportunities for EGUs to significantly increase hours
of operation under a 10-year versus a 5-year baseline period are also
limited. 

Nonetheless, we want to comprehensively examine the outcomes of a
maximum hourly emissions increase test or an annual emissions increase
test with a 10-year baseline period, using a robust methodology based on
conservative (that is, protective of the environment) estimates.  We
therefore developed two IPM scenarios, which we call the CAIR/CAMR/CAVR
NSR Availability Scenarios, or, more simply, the NSR Availability
Scenarios, to examine how changes to major NSR applicability under the
proposed regulations could, by allowing sources to make repairs or
improvements that increase hours of operation, affect emissions and
control technology installation.  These IPM scenarios are based on the
CAIR/CAMR/CAVR 2020 Base Case Scenario, which employs the IPM v. 2.1.9
model that we describe above, including information for the electric
sector in the contiguous United States.  

The parameters in the IPM model are based on availability for 6,500 EGUs
over the 5-year period from 2000 - 2004.  In the NSR Availability
scenarios, however, we changed the parameters in IPM v.2.1.9 consistent
with the way EGUs might operate under the more flexible regulations that
we are proposing.  That is, we assumed that some owner/operators might
make changes that increase the hours of operation of some EGUs.  It is
unlikely that an owner/operator would be able to make changes that
reduce the hours that an EGU is unavailable due to a planned outage or a
maintenance outage.  However, EGUs would be able to make changes that
increase their hours of operation as a result of a reduction in the
number and length of forced outages.  Specifically, with more
flexibility concerning the number of hours EGUs operate annually, EGU
owner/operators may replace broken down equipment in an effort to reduce
the number of forced outages.  Such actions would increase the safety,
reliability, and efficiency of EGUs, consistent with one of our primary
policy goals for today’s proposed regulations.  

Therefore, in the CAIR/CAMR/CAVR NSR Availability Scenario, we assumed
that coal-fired EGUs would be able to make changes that affect forced
outage hours in two, alternative, ways: (1) coal-fired EGUs would reduce
their forced outage hours by half (2 percent increase in availability);
and (2) coal-fired EGUs would have no forced outage hours (4 percent
increase in availability).  Therefore, in the first model run, we
increased the coal-fired availability by 2 percent, from 85 percent to
87 percent annually.  In the second NSR EGU run, we increased coal-fired
availability by 4 percent, to 89 percent annually.  We believe it is
unlikely that an EGU would be able to make repairs that completely
eliminate forced outage hours.  However, we wanted a robust examination
of changes that could impact emissions and air quality. All other
information in both the CAIR/CAMR/CAVR NSR Availability Scenarios is the
same as that in IPM v.2.1.9 used for the Base Case Scenario.  We request
comment whether a 4% increase in availability, as assumed in the
CAIR/CAMR/CAVR NSR Availability Scenario, would encourage projects that
would maintain or improve the safety, reliability, and efficiency of
EGUs.  In particular, we request comment on examples of such projects
that it might encourage.

The NERC GADS calculates the average availability for an EGU by taking
the actual total number of unavailable hours in a given year for all
EGUs and dividing it evenly among the total number of EGUs.  Based on
the GADS data, the IPM assumes an upper bound of 85% availability for
coal-fired EGUs.  In GADS data for the years 2000-2004, some EGUs
actually had more than 85% availability and some actually had less.  The
particular EGUs that had greater than 85% availability and less than 85%
varied from year to year.  Similarly, by eliminating forced outages,
some EGUs could increase their availability by more than 2-4% and some
EGUs could increase their availability by less than 2- 4%.  Likewise,
the particular EGUs that were able to reduce their forced outage hours
would also vary from year to year.  For modeling purposes, it thus makes
more sense to assume an average availability than to determine
unit-by-unit availabilities for each and every EGU in a given year. 

Our approach based on average availability is also consistent with
actual historical operations at particular EGUs and plantsites, which
are most directly related to local emissions and air quality.  Variation
in actual annual hours of operation at a given EGU and at given
plantsites do occur under current major NSR applicability.  It is not
uncommon for actual hours of operation for a particular EGU to vary by
348 hours (4% availability) or more from year to year.  It is also not
uncommon for the variation in actual hours of operation to occur among
EGUs at a particular plantsite by 4% or more from year to year.  For
example, in one year Unit A might run 7800 hours and Unit B might run
7400 hours.  In the next year Unit B might run 7800 hours and Unit A
7400 hours.  This pattern further supports an approach based on average
availability for estimating local emissions.  Changes in average
availability, rather than the absolute availability of any given EGU,
thus is appropriate for analyzing the impact of proposed changes to
major NSR applicability.  

C.  CAIR/CAMR/CAVR NSR Scenarios – Discussion of Results  

1.  Control Device Installation

As Table 2 shows, the CAIR/CAMR/CAVR NSR Availability Scenario projects
retrofitting of more control devices than under the CAIR/CAMR/CAVR 2020
Base Case Scenario.  This result occurs whether hours of operation
increase by 2 percent or by 4 percent.  Significantly, under the 4
percent scenario, more GWs of electric capacity are controlled than
under the 2 percent scenario.  These results are consistent with what
IPM generally projects, as noted above, that is, the more hours an EGU
operates, the more likely it is to install controls.  We thus conclude
that the more hours an EGU operates, the more likely it is to install
controls, regardless of whether the major NSR applicability test is on
an hourly basis or an annual basis with a 10-year—as opposed to a
5-year—baseline.

Table 2. 2020 Retrofitted Capacity (GWs) Under CAIR/CAMR/CAVR NSR
Availability Scenario

	Additional Retrofits Compared to 2004 Base Case	Additional Retrofits
Compared to CAIR/CAMR/CAVR 2020 Base Case

	NSR – Avail.2%	NSR – Avail. 4%	NSR – Avail. 2%	NSR – Avail. 4%

FGD	109.62	111.53	1.71	3.63

SCR	73.47	73.92	0.62	1.07



2.  National Emissions

As Table 3 shows, the CAIR/CAMR/CAVR NSR Availability Scenarios project
essentially no changes in SO2 or NOx emissions nationally by 2020 as
compared to emissions under the Base Case Scenario.  This result is
consistent with the fact that under the NSR Availability Scenarios, the
amount of controls increases, compared to the Base Case, and we find
that these associated emissions decreases are offset by the emissions
increases associated with the reduced forced outages and higher
production levels.

Table 3.  National EGU Emissions Under CAIR/CAMR/CAVR NSR Availability
Scenarios Compared to Base Case

	Total Emissions Under Base Case 	Total Emissions Under NSR – Avail
(4%)	Total Emissions Under NSR – Avail (2%)	Emissions Change Under NSR
Avail. (4%) Compared to Base Case  (tpy)	Emissions Change Under NSR
Avail. (2%  Compared to Base Case  (tpy)

SO2	4,063,000 tpy	4,046,000 tpy	4,043,000 tpy	(17,000) tpy

.5% decrease	(20,000) tpy

.5% decrease

NOx	1,940,000 tpy	1,967,000 tpy	1,954,000 tpy	27,000 tpy

~1% increase	14,000 tpy

~1 % increase



As noted above, the NSR Availability Scenarios examine emissions changes
based on conservative estimates developed using actual historical hours
of operation for 6,500 EGUs over the years 2000 - 2004.  We conclude
that to any extent that EGU hours of operation increase under a maximum
hourly or an average annual 10-year baseline test, as opposed to the
current average annual 5-year baseline test, such increased hours of
operation would not increase national EGU SO2 and NOx emissions.  This
conclusion as to emissions in the contiguous 48 states supports
extending the proposed rules nationwide, instead of limiting them to the
states in the CAIR region.

3.  Local Emissions Impact

To examine the effect of the maximum hourly and 10-year baseline tests
on local air quality, we compared 2020 county-level EGU SO2 and NOx
emissions under the CAIR/CAMR/CAVR 2020 Base Case and NSR Availability
(4 Percent) Scenario.  Tables 4 and 5 show these comparisons. 

Table 4.  Changes in County-level SO2 Emissions CAIR/CAMR/CAVR NSR
Availability (4 Percent) Scenario Compared to CAIR/CAMR/CAVR 2020

	Number of Counties with Changes in SO2 Emissions

Total number of counties with decreases	65

Decreases between 20,000 and 36,941 tpy	2

Decreases between 3,000 and 20,000 tpy	13

Decreases between 1,000 and 3,000 tpy	12

Decreases between 40 and 1000 tpy	31

No change in EGU emissions	780

Increases between 40 and 1000 tpy	255

Increases between 1,000 and 3,000 tpy	47

Increases between 3,000 and 6,801 tpy	6

Total number of counties with increases	338

Total number of counties	1183

 

Table 5.  Changes in County-level NOx Emissions CAIR/CAMR/CAVR NSR
Availability (4 Percent) Scenario Compared to CAIR/CAMR/CAVR 2020

	Number of Counties with Changes in NOx Emissions

Total number of counties with decreases	238

Decreases between 3,000 and 10,720 tpy	2

Decreases between 1,000 and 3,000 tpy	9

Decreases between 40 and 1000 tpy	61

No change in EGU emissions	540

Increases between 40 and 1000 tpy	269

Increases between 1,000 and 3,000 tpy	9

Increases between 3,000 and 3,172 tpy	1

Total number of counties with increases	405

Total number of counties	1183



As Tables 4 and 5 show, the proposed revised NSR applicability tests
would, under the highly conservative assumptions described above, result
in a somewhat different pattern of local emissions, with some counties
experiencing reductions, some experiencing increases, and some remaining
the same.  This pattern is consistent with the fact that most coal-fired
EGUs are in the CAIR region and therefore subject to regulations
implementing the CAIR cap.  It is also consistent with the structure of
the electric utility industry.  That is, the utility sector is unique
among other sectors in that there is no distinguishing feature of a
kilowatt hour of electricity other than its cost.  Thus, economics drive
the sale and purchase of power.  The United States has several power
markets that are interconnected by a transmission grid.  Electric power
that is generated in one market may be dispatched in another power
market.  Generators use their own financial models and methods to bid to
supply electricity to serve the load.  On a given day, a purchaser may
find it cheaper to buy from another transmission area than from the
transmission area within which the purchaser is located, and vice versa.
 Thus, electricity is dispatched according to the least costly means of
serving load at different locations, based on actual operating
conditions existing on the power system and on the prices at which
members have offered to supply energy in the power market.

For these reasons, an increase in emissions will also result in a
decrease in emissions elsewhere.  Of course, as noted elsewhere, the
presence of a regionwide cap, under CAIR, results in the same phenomenon
within the CAIR region, that is, an increase in emissions in one area
results in a decrease elsewhere.  This dynamic occurs regardless of the
major NSR applicability test for existing EGUs.  Nonetheless, the NSR
Availability Scenario demonstrates that this pattern continues to occur
when increased availability is assumed, such as we assume for present
purposes would occur under the maximum hourly and 10-year baseline tests
proposed today. 

As Table 4 shows, in counties with an SO2 emissions increase of at least
40 tons per year, emission increases ranged from 43 to 6,801 tpy.  The
degree of county-level emission decreases was higher than that of the
increases, ranging from 10 to 36,941 tpy.  This pattern also occurred
with NOx emissions.  As Table 5 shows, in counties with a NOx emissions
increase of at least 40 tons per year, emission increases ranged from 41
to 3,172 tpy.  The degree of county-level emission decreases was higher
than that of the NOx increases, ranging from 1 to 10,720 tpy.  The
increases and decreases occurred in CAIR and non-CAIR states.

 EGUs had emissions increases of ≥ 3,721 tpy NOx as compared to 2003. 
In 2004, 37 EGUs had emissions increases of ≥ 6,801 tpy SO2 as
compared to 2003.  Thus, the highest county-level projected emissions
increases for SO2 and NOx under CAIR/CAMR/CAVR NSR Availability (4
Percent) Scenario are less than the emissions increases that actually
occurred, measured using CEMS, at individual EGUs over the period of
2003 - 2004.  As this perspective shows, the local emissions increases
that the IPM results indicate could theoretically occur from this action
are not large.

4.  Air Quality Impact

As we discussed above, projected emissions changes under proposed
revised NSR applicability tests would result in a somewhat different
pattern of local emissions, with some counties experiencing reductions,
some experiencing increases, and some remaining the same.  As we also
noted, the degree and pattern of these changes is consistent with those
under CAIR/CAMR/CAVR 2020.  Moreover, the emission changes under the NSR
Availability Scenario are projected assuming extremely conservative
assumptions, as described above. 

We would also expect patterns of air quality changes to be consistent
with projections under CAIR/CAMR/CAVR in 2020 based on the spatial
distribution of SO2 and NOx emissions changes as shown in Figures 1 and
2, respectively, under the NSR Availability (4 Percent) Scenario. 
Projected increases in emissions of these pollutants due to increased
hours of operation at EGUs are small in magnitude and sparse across the
continental U.S.  Therefore, we would expect these increases to cause
minimal local ambient effect.  Because many counties experience
decreases in emissions, we would further expect any local ambient
effects from increased emissions to be somewhat diminished because of
the emissions decreases elsewhere that yield regionwide improvements in
air quality.  We also expect similar outcomes with respect to the NSR
Availability (2 Percent) Scenario where the emissions changes are
smaller and constitute a pattern of increases and decreases that is
similar to that of the NSR Availability (4 Percent) Scenario.  

We thus believe that the local air quality under today’s proposed
regulations would be commensurate with that under the CMAQ modeling
based on CAIR/CAMR/CAVR 2020 Base Case Scenario emissions projections.  
Similarly, we have examined the effect of the county level emissions
increases projected to occur under the CAIR/CAMR/CAVR NSR Availability
(4% and 2%)  Scenarios on air quality values under the PSD requirements.
 Based on the size of the emissions increases, we do not believe that
local areas will exceed their SO2 increment or the SO2 NAAQS, nor do we
believe that local areas will exceed the NO2 increment or NAAQS.  For
more information, see [cite to TSD with figures 3 and 4.]

Figure 1.  2020 CAIR/CAMR/CAVR NSR Availability (4 Percent) Scenario: 
County-level Changes in SO2 with a 4 Percent Increase in EGU Operating
Hours

Figure 2.  2020 CAIR/CAMR/CAVR NSR Availability (4 Percent) Scenario: 
County-level Changes in NOx with a 4 Percent Increase in EGU Operating
HoursIV. Additional Regulatory History, Analyses, and Legal Basis
Supporting Option 3 – Annual Emissions Increase Test with 10-Year
Baseline

Option 3 of this action is a proposal that for EGUs, like other major
stationary sources, baseline actual emissions would be determined within
the 10-year period immediately preceding when the owner or operator
begins actual construction of the project.  This proposal would replace
current regulations, under which the baseline period is 5 years, unless
a longer period of up to 10 years is more representative.  The analyses
in Section III of this preamble support this revision of the average
annual emissions test.  This section contains additional regulatory
history, analyses, and legal basis specific to this proposal to extend
the baseline period to 10 years.  

A.  Regulatory History for 5-Year Baseline Period for EGUs 

Our regulations for EGUs have long provided flexibility in the time
period for determining baseline actual emissions.  On June 14, 1991 (56
FR 27630), we proposed to allow EGUs to determine pre-change emissions
on the basis of the 2-year period prior to the change or a different
consecutive 2-year period within the 10 years, where the permitting
authority determines that such period is more representative of normal
source operations.  (56 FR 27630.)  In response to this proposal, some
commenters indicated that we should “allow utilities to use the
maximum utilization in any 1 year within at least the last 10 years,
since 10 years is a more relevant capacity investment planning horizon
than 5 years.”  (See 57 FR 32323/3.)  On July 21, 1992 we promulgated
final regulations that presumed pre-change actual emissions would be
based on a 2-year period within the previous 5 years, but that allowed
the use of a longer period, not to exceed 10 years, if the reviewing
authority determines such a period to be more representative of normal
source operations.  We based our decision on “considerations that
electricity demand and resultant utility operations fluctuate in
response to various factors such as annual variability in climatic or
economic conditions that affect demand, or changes at other plants in
the utility system that affect the dispatch of a particular plant.” 
(See 57 FR 32325.)  This 1992 rulemaking was closely related to an NSR
applicability action involving WEPCO, in which the 5-year period prior
to the change was considered the appropriate baseline period.  It should
be noted, however, that in the 1992 rulemaking, we did not cite specific
reasons why a 5-year period, as opposed to a different period, was
consistent with those factors.

In 1996, we further considered the appropriate time period for
establishing pre-change baseline actual emissions.  To allow sources to
determine applicability based on their highest level of utilization, we
proposed a 10-year baseline for all source categories, including EUSGUs.
 We stated:

  SEQ CHAPTER \h \r 1 The EPA is today proposing to extend the time
period for determining baseline in the definition of actual emissions to
10 years for all source categories and to allow sources to base their
actual emissions on the highest consecutive 12 months during this
10-year period.    SEQ CHAPTER \h \r 1 The EPA's intent is to allow
sources to determine applicability based on their highest level of
utilization and not necessarily their highest emissions rate.  The
emissions rate of units at issue may be subject to any number of current
Federal or State restrictions (e.g., RACT, MACT, BACT, LAER, NSPS,
national emission standard for HAP (NESHAP)) as well as voluntary limits
(e.g., reductions used for netting, offsets, Emission Reduction Credits
creation) and these limits may have been imposed since the time the
source achieved its highest emissions level.  Therefore, these limits
must be included in establishing the baseline emissions.  For this
reason, the EPA is today proposing that sources calculate the baseline
by using their current emissions factor in combination with the
utilization level from the 12-month time period selected.  This
safeguard insures that no significant loss of environmental protection
will result from the proposed change.  

61 FR 38258/9, July 23, 1996.

Many commenters supported our proposed 10-year baseline period for all
source categories.  Utility commenters noted that a 10-year baseline
period would provide flexibility, especially in situations where low
utilization rates have persisted for extended periods due to economic
constraints.  However, some commenters believed that a 10-year period
would be too long.  To address these concerns, we sought to better
understand what time period best represents an industry’s normal
business cycle.  Therefore, in 1997, we conducted a study of nine
diverse industries that would be representative of business cycles
across all sectors.

In our December 2002 final rules, we finalized extension of the baseline
period to 10 years for all source categories except for utilities.  We
concluded that each of the nine source categories we examined, not
including utilities, exhibited a business cycle of between three and
eight years, and that a 10-year period would be reasonable to capture
the entire cycle of each industry.  We nonetheless retained the 5-year
baseline period for EUSGUs, concluding that “any 2-year period out of
the preceding 5 years is a sufficient period of time to capture normal
business cycles at a EUSGU.  We do not believe that any information
received during the public comment period for this final rule adequately
supports a different conclusion.”  (See 67 FR 80200.)  We did not
provide any specific information indicating that the business cycle for
utilities should be a 5-year period.  Nor did we explain why, on the
basis of the conclusions we came to for the nine source categories
studied, we treated every source category as having a business cycle
that merited a 10-year baseline period, except that we continued to give
utilities, and utilities alone, a 5-year baseline period.

B.  New Information Concerning the Appropriate Baseline Period for
Utilities

Several events since December 2002, coupled with further analysis,  have
caused us to reevaluate whether 5 years is in fact adequate to capture a
baseline period that is representative of normal operations for EGUs. 
In particular, we promulgated CAIR in March 2005.  As a result of CAIR,
we project installation of scrubbers on an additional 64 GW of existing
coal-fired generation capacity for SO2 control and SCR on an additional
34 GW of existing coal-fired generation capacity for NOx control by
2015.  By 2020, we expect installation of scrubbers on an additional 82
GW of existing coal-fired generation capacity for SO2 control and SCR on
an additional 33 GW of existing coal-fired generation capacity for NOx
control.  As we noted at 70 FR 25197, the number of air pollution
control retrofits that will result from the CAIR are quite significant
and cannot be immediately installed.  We recognized that the regulated
industry will need to secure large amounts of capital to meet the
control requirements while managing an already large debt load, and is
facing other large capital requirements to improve the transmission
system.  In addition, in light of labor constraints for implementing
controls, we allowed approximately 10 years, until 2015, for the
effective date of the final phase of the CAIR emissions reductions,
although we expect that the availability of banked allowances will
result in controls being phased in past 2020.  During this period of
control installation, we expect the capacity utilization of individual
EGUs to vary as the industry plans for and installs controls to meet the
CAIR requirements.  We therefore believe it is reasonable to provide a
period longer than 5 years and instead to allow a 10-year baseline
period to allow EGU owner/operators to accommodate these changes. 
Moreover, as the National Association of Regulatory Utility
Commissioners recently noted, much of the electricity infrastructure is
aging and lead times for planning for additional generation range from 5
- 10 years.

C.  EGU Business Cycle

In addition, we have conducted further research into the length of the
business cycle for electric utilities, which calls into question the
conclusions in our 2002 rule that a 5-year period is appropriate.  This
research also sheds light on whether there is a basis for treating
utilities differently than all other source categories for these
purposes.

For the 2002 rule, we completed a report that examined business cycles
in nine major emitting industry sectors.  As we noted in that report,
business cycles are made up of four separate phases: 1) the peak-output
is assumed to be at or near full capacity; 2) the recession - output and
employment decline; 3) the trough-output and employment bottom out at
their lowest levels; and 4) the recovery-output and employment expand. 
We examined gross product originating (GPO) by industry using data from
the Bureau of Economic Analysis (BEA).  Using this data, we examined
peak-to-peak and trough-to-trough business cycles.  We concluded that
industry business cycles of 5 years or more are sufficiently frequent
that a 5-year period would probably be insufficient to provide
reasonable certainty that industry fluctuations had been captured. 
Since the longest business cycle identified among these industries was 8
years, we concluded that a minimum of 10 years would be required to
capture an entire industry cycle.  As noted above, we did not undertake
this business cycle analysis for electric utilities, and we did not
explain why although the nine-sector study was sufficient to establish a
10-year baseline for all industries (including those whose business
cycle was not studied), the study was not sufficient to establish a
10-year baseline for utilities.

For today’s proposal, however, we have undertaken this business cycle
analysis.  To determine a business cycle for EGUs, we considered aspects
unique to the electric power sector.  Unlike other sectors, until
relatively recently, retail prices for electric power were tightly
regulated by state and federal agencies.  There was no competitive
market for electric power.  In recent years, partial de-regulation has
occurred.  Therefore, for the utility sector, retail sales and more
generally the patterns of business cycles are not directly analogous to
those in other sectors, where market economies have always existed. 

More specifically, up to the 1970s (and in some states, currently);
Public Utility Commissions (PUCs) regulated the rates that residential
commercial and industrial customers paid for utility services. 
Beginning in the 1970s, government policy shifted against traditional
regulatory approaches and in favor of deregulation for many important
industries, including transportation, communications, and energy, which
were all thought to be natural monopolies (prior to 1970) that warranted
governmental control of pricing.  Some of the primary drivers for
deregulation of electric power included the desire for more efficient
investment choices, the possibility of lower electric rates, reduced
costs of combustion turbine technology that opened the door for more
companies to sell power, and complexity of monitoring utilities’ cost
of service and establishing cost-based rates for various customer
classes. 

The national Energy Policy Act of 1992 accelerated competition in the
electric power industry.  The Energy Policy Act revised earlier
legislation to include a new class of power producers, called exempt
wholesale generators.  It also required utilities to allow wholesale
electricity generators to access their power lines.  Although the Energy
Policy Act of 1992 paved the way for competition in the electric utility
industry, the electric utility industry is still undergoing a transition
to competition.  Some states have already restructured their electric
utility markets.  Others are in the process of instituting retail
competition in their respective jurisdictions.  The pace of
restructuring in the electric power industry slowed significantly in
response to market volatility and financial turmoil associated with
bankruptcy filings of key energy companies in California.  By the end of
2001, restructuring had either been delayed or suspended in eight states
that previously enacted legislation or issued regulatory orders for its
implementation.  Another 18 states that had seriously explored the
possibility of deregulation in 2000 reported no legislative or
regulatory activity in 2001 (DOE, EIA, 2003a).  Currently, there are 17
states where price deregulation of generation (restructuring) has
occurred.  The effort is more or less at a standstill.  However, at the
federal level, there are efforts in the form of proposed legislation and
proposed Federal Energy Regulatory Commission (FERC) actions aimed at
reviving restructuring.  For states that have not begun restructuring
efforts, it is unclear when and at what pace these efforts will proceed.


This history of the utility industry makes clear that compared to other
sectors, for which market economies have always existed, retail sales
and more generally the patterns of business cycles for EGUs are not
directly analogous to those in other sectors.  In light of this, to
analyze business cycles for utilities, we examined net generation in
kilowatt hours.  We did not monetize net generation, even though, in the
2002 NSR Rule, we monetized other industries’ output.  Net electricity
generation is a reasonable proxy to measure gross product from power
generating facilities.  It is defined as the "amount of gross generation
less the electrical energy consumed at the generating station(s) for
station service or auxiliaries." (U.S. Energy Information
Administration, Electric Power Monthly, May 2006).  This measure
incorporates output of the major product from such facilities as it is
delivered to the public via the power grid, and it is net of inputs such
as fuels and materials.  Net electricity generation is influenced
heavily by activity along a business cycle.

Our analysis indicates that   SEQ CHAPTER \h \r 1 the business cycles
for the electric power industry, measured as net electricity generation,
are similar to the general business cycles for the U.S. economy, which
is measured as real Gross Domestic Product (GDP).  Specifically, since
the end of World War II, growth in electricity use has coincided with
growth in GDP.  Electricity use, and thus net electricity generation,
influences GDP, and there is a strong relationship between energy use
and economic output.  Real GDP is viewed by the National Bureau of
Economic Research (NBER) as “the single best measure of aggregate
economic activity.”  Also, “[i]n determining whether a recession has
occurred and in identifying the approximate dates of the peak and the
trough [of a business cycle], the [NBER] ... places considerable weight
on the estimates of real GDP issued by the Bureau of Economic Analysis
of the U.S. Department of Commerce.” 

Figure 5.  Electricity Net Generation and GDP, 1949-2004The NBER
defines a peak and trough, for purposes of identifying an economy-wide
business cycle, as follows:  “A recession involves a substantial
decline in output and employment.  In the past 6 recessions, industrial
production fell by an average of 4.6 percent and employment by 1.1
percent.  The Bureau waits until the data show whether or not a decline
is large enough to qualify as a recession before declaring that a
turning point in the economy is a true peak marking the onset of a
recession.”  NBER rarely offers a precise benchmark for determining a
peak or trough for a business cycle; rather, NBER reviews real GDP and
other economic data and determine as a committee when a peak or trough
occurs.  Based on this definition, NBER has identified, from 1945 to
2001, 11 troughs and 11 peaks.  The average of the trough-to-trough and
peak-to-peak cycles is 63 months.  Five of the 11 have been longer than
60 months (5 years).  One has been longer than 10 years, and that has
been 128 months.  http://www.nber.org/cycles/.

Examination of net electricity generation reveals comparable peaks and
troughs.  We compared net electricity generation to the general business
cycle for the U.S. economy for the years 1949-2004, as defined by the
dates of peaks and troughs established by the NBER during that period of
time.  Precise comparisons are difficult because NBER analyzes and
presents data for GDP on a month-to-month basis, but data for net
electricity generation are available on only an annual basis.  Even so,
broad comparisons are possible.

There are only 3 years in which net electricity generation declined from
the year before - 1982, 1992, and 2001.  The NBER established that there
was a trough in the business cycle in November 1982, March 1991, and
March 2001.  All of these years are part of a recession.  Thus, the
years in which net electric generation actually fell generally coincide
with troughs in the business cycle. 

Furthermore, periods of smaller increases in net generation also
correspond to business cycle troughs for the U.S. economy.  A peak
occurred in August 1957 and a trough occurred in April 1958.  The net
electric generation only increased 2.3 percent between 1958 and 1957,
the smallest yearly increase (to that point) since this dataset began in
1949.  The amount of increase in net electric generation from 1973 to
1974 was only 0.4 percent, the lowest increase since the dataset began
in 1949 and the increase in net electric generation from 1974 to 1975
was only 2.7 percent.  These periods of low increase in net electric
generation are consistent with the GDP business cycle peak in November
1973 and a trough in March 1975.  In addition, the amount of increase in
net electric generation from 1980 to 1981 was only 0.3 percent (even
lower than the small increase from 1973 to 1974.)  This period coincides
with the peak in GDP in January 1980 and a trough in July 1980.  Thus,
peaks and troughs in net electric generation are comparable to those in
the general economy's business cycle as defined by NBER.  

A 10-year period reasonably encompasses all of the business cycles of
the GDP since World War II.  As noted above, of the GDP’s 11
trough-to-trough and 11 peak-to-peak cycles, the average length was 63
months; and only one was longer than 10 years, and that was longer by
only the brief period of 8 months.  Because the business cycle of
utilities should be considered to parallel that of GDP, we conclude that
a 10-year period reasonably characterizes the business cycle of
utilities.

Another reason for concluding that utilities should be considered to
have a business cycle that is consistent with a 10-year baseline period
is that in the 2002 NSR Reform rule, we decided that 10 years for a
baseline period fits the business cycle periods of the nine industries
studied.  We concluded that those nine industries were representative of
all industries, so that the 10-year baseline period is appropriate for
all industries, including those not studied,  We believe it is
reasonable to treat utilities the same way as all other industries.  All
other industries collectively comprise the GDP and utilities’ net
electric generation broadly mirrors GDP.

D.  Rationale and Legal Basis

In New York v. EPA, the Court upheld the provisions of the 2002 NSR
Reform rule that extended the baseline period to 10 years for all source
categories other than EGUs. 413 F.3d 3, 47 (D.C. Cir.), rehearing and
rehearing en banc denied, __ F.3d __ (2005).  The Court stated that CAA
section 111(d) (4) mandates that determining whether a physical or
operational change increases emissions must be based on actual
emissions.  However, the Court emphasized that the provision is silent
on how to determine the amount of the increase, including the length of
the period over which the pre-change amount of emissions may be
calculated.  As a result, EPA has discretion to determine the length of
that period, under Chevron v. NRDC, 467 U.S. 837 (1984). 

The purpose of the baseline period for EGUs, as for other source
categories, is to allow an EGU to consider the amount of existing
capacity that it utilized during a full business cycle in determining
whether there will an emissions increase from a physical or operational
change.  Generally, an EGU’s operations cover a range of operating
(and emissions) levels, and not simply a single level of utilization. 
This is especially the case under the current market-based system, in
which utilization of particular EGUs is determined by system wide
demand, by the efficiency of the EGU in relation to other EGUs, and by
climatic variations that are beyond the control of the particular EGU. 
Consequently, the baseline period ensures that an EGU seeking to make
changes at its facility at a time when utilization may not be at its
highest can use a baseline period that encompasses its business cycle. 
This baseline period allows the EGU to identify capacity actually used
in order to determine an average annual emissions rate from which to
calculate any projected actual emissions resulting from the change.  

Based on the research described above, we believe that use of a fixed
10-year baseline period, and not the 5-year period that we reiterated in
the 2002 NSR rule, is consistent with the business cycle for electric
utilities.  In addition, we believe that a 10-year baseline period is
necessary to establish normal operation in light of the significant
changes in the utility sector that are expected as a result of the CAIR
in the next 10-20 years. 

Further, we note that one of our reasons for extending the baseline
period to 10 years for other source categories in the 2002 rule was to
provide clarity and certainty to the process of selecting an appropriate
utilization/emissions level that is representative of a source’s
normal operation.  Until the 2002 rule, other sources could utilize a
baseline period longer than 5 years only upon a showing that such longer
period was representative of normal source operations.  That approach
resulted in uncertainty and disputes.  We solicit comment on whether the
current rule for utilities, which similarly allows a baseline period
longer than 5 years only upon a showing that such longer period is
representative of normal source operations, also results in uncertainty
and disputes.

As with other source categories, baseline actual emissions calculated
from the consecutive 24-month period selected cannot yield a higher
pollution level than a unit is currently allowed to emit.  Therefore,
the regulations require sources to determine whether any legally
enforceable limitations currently exist that would prevent the affected
EGU from emitting a pollutant at the levels calculated from the 24-month
baseline period.  The approach that we propose today allows EGUs to
reference plant capacity that has actually been used, but not emissions
levels that are not legally allowed at the time the modification is to
occur.

We note that in its ruling, the Court cited our statements in the record
that 90 percent of the environmental benefits of the NSR program have
come from new sources, modifications at electric utilities,
modifications at sources where emissions have been highest in recent
years, and modifications at sources where emissions have been relatively
stable, none of which are affected by revising the baseline period to 10
years.  The record for the 2002 rule included more specific information
regarding emissions reductions from existing sources, including EGUs. 
As we noted in the Supplemental Environmental Analysis cited by the
Court, “We believe that the majority of the emissions
reduction/prevention benefits from NSR come from new sources and new
units.  According to recent PSD permitting data, more than 80 percent of
the PSD benefits come from these categories.”  Thus, only 20 percent
of the benefits of NSR come from modifications generally, of which EGU
modifications are one component.  Moreover, we believe that in the past
few years, more EGUs have implemented programs that make physical and
operational changes, but at times and in a manner that do not trigger
major NSR.  Thus, we believe that the amount of emissions reductions
from NSR major modifications of utilities is small and diminishing.  We
believe that the 10-year baseline will allow EGUs to determine baseline
emissions using existing capacity, particularly in light of the
fluctuations likely in EGU utilization during the CAIR implementation
period, but more generally considering the variations in climate,
economic activity, and system wide demand. 

V.  Proposed Regulations for Option 1:  Hourly Emissions Increase Test	

This section contains details on the proposed regulatory language for
Option 1, the hourly emissions increase test.  We are proposing that
Option 1 would apply to all existing EGUs.  As we noted at 70 FR 61093,
however, we are also requesting comment on whether Option 1 should be
limited to the geographic area covered by CAIR, or to the geographic
area covered by both CAIR and BART.  We are also proposing that the
Option 1 would apply to all regulated NSR pollutants.  However, we also
request comment on whether Option 1 should be limited to increases of
SO2 and NOx emissions. 

In today’s SNPR, for Option 1 we are proposing to exempt EGUs from the
procedures in the current regulations for determining a significant
emissions increase and a significant net emissions increase. 
Specifically, we are proposing to exempt EGUs from the applicability
procedures based on a significant emissions increase and significant net
emissions increase in the current regulations at 40 CFR 51.165, 51.166,
52.21, and 52.24 and in appendix S to 40 CFR part 51, although the
amendatory language in today’s proposed rule does not include specific
provisions to effect this change.  For example, under Option 1 the
provisions for determining a significant emissions increase and a
significant net emissions increase in §51.166(a) (7) (iv) would be
modified to exempt EGUs.

In place of the applicability procedures in the current regulations,
Option 1 applies an hourly emissions test to EGUs, for which we are
proposing regulatory language.  If a physical or operational change at
an existing EGU is found to be a modification according to this hourly
emissions test, the EGU would then be subject to all the substantive
major NSR requirements of the existing regulations.  Accordingly, we are
also proposing to revise the substantive provisions in all the current
major NSR regulations that apply to major modifications to apply also to
modifications at EGUs.  The amendatory language in today’s proposed
rule does not include specific provisions for these changes.  For
example, the substantive provisions to be amended would include, but not
be limited to, the provisions in §51.166(a)(7)(i) though (iii), (b)(9),
(b)(12), (b)(14)(ii), (b)(15), (b)(18), (i)(1) through (9), (j)(1)
through (4), (m)(1) through (3), (p)(1) through (7), (r)(1) though (7),
and (s)(1) through (4).  

We are also proposing to add a definition of the “increases”
component of “modification” to the major NSR rules.  Under Option 1,
we are proposing to define the “increases” component to mean maximum
hourly emissions rate achieved.  That is, if a physical change or change
in the method of operation (as defined under existing regulations, which
we are not proposing to change), is projected to result in an increase
in the maximum hourly emissions rate expected to be achieved over the
maximum hourly emissions rate actually achieved at the EGU prior to the
change, a modification would occur.  In the alternative, we are
proposing the maximum hourly achievable test, and therefore we are also
proposing in the alternative to add a definition of the “increases”
component of “modification” that substantially mirrors the
definition of the “increases” component of “modification” in the
NSPS provisions, which are found in 40 CFR 60.2.

Specifically, under Option 1, we are proposing to add two new sections
to the major NSR program rules that would include the two-step
provisions for modifications.  The first, 40 CFR 51.167, would specify
the requirements that State Implementation Plans must include for major
NSR applicability at existing EGUs, including those for both attainment
and nonattainment areas.  (Proposed rule language for 40 CFR 51.167
accompanies today’s SNPR.)  The second, 40 CFR 52.37, would contain
the requirements for major NSR applicability for existing EGUs where we
are the reviewing authority.  Although the proposed amendatory language
is for 40 CFR 51.167, we are proposing that the same requirements would
apply under 40 CFR 52.37, differing only in that the Administrator is
the reviewing authority, rather than the State, local, or tribal agency.
 Although this notice does not contain specific regulatory language, we
are proposing that either 40 CFR 51.167 or 40 CFR 52.37, as appropriate,
would contain the requirements for emissions increases at EGUs for all
sections of the Code of Federal Regulations that contain the major NSR
program, including 40 CFR 51.165, 51.166, 52.21, 52.24, and appendix S
of 40 CFR part 51, as well as any regulations we finalize to implement
major NSR in Indian Country.  We are also proposing to make the same
changes where necessary to conform the general provisions in parts 51
and 52 to the requirements of the major NSR program, such as in the
definition of modification in 40 CFR 52.01.  In addition, we are
proposing to remove all applicability requirements for existing EUSGUs
in all sections of the CFR that contain the major NSR program, as the
EGU requirements would supersede these requirements.  

In the NPR, we proposed three alternatives for the hourly emissions
increase test- the NSPS maximum achievable hourly emissions test,
maximum achieved hourly emissions, and an output-based measure of hourly
emissions.  As some commenters noted, we did not give much detail about
the output-based measure of hourly emissions.  In today’s SNPR, we are
recasting what we proposed in the NPR for the output-based methodology. 
In today’s SNPR, both the maximum achieved hourly emissions test and
the maximum achievable hourly emissions test include output-based
alternatives.  Specifically, we are proposing two broad approaches under
Option 1:  (1) a maximum achieved hourly emissions test; and (2) a
maximum achievable hourly emissions test.  If we adopt the maximum
achieved hourly emissions test, we may require that it be expressed in
an input-based format (lb/hr) or an output-based format (lb/MWh). 
Alternatively, and as we did in our recently promulgated NSPS for
combustion turbines (40 CFR part 60, subpart KKKK, July 6, 2006), we may
also adopt both an input and output based format.  If we adopt both
formats, sources, at their choice, would be able to implement the hourly
emissions test in either input- or output-based formats.  Likewise, if
we adopt the maximum achievable hourly emissions test, it may be
expressed in an input-based format (lb/hr), an output-based format
(lb/MWh), or both.  We are also proposing three methods for computing
maximum achieved emissions: (1) statistical approach; (2) two-in-5-year
baseline; and (3) one-in-5-year baseline.  In terms of the regulatory
language that accompanies today’s notice, we are proposing eight
alternatives for determining whether a physical or operational change at
an EGU is a modification.  These alternatives are summarized in Table 6
and can be found at proposed §51.167(f) (1).

Table 6.  Major NSR Applicability for Existing EGUs Under Option 1

Option 1	Step 1:  Physical Change or Change in the Method of Operation

Step 2:  Hourly Emissions Increase Test

Alternative 1 – Maximum achieved hourly emissions; statistical
approach; input basis

Alternative 2 – Maximum achieved hourly emissions; statistical
approach; output basis

Alternative 3 – Maximum achieved hourly emissions; two-in-5-year
baseline; input basis

Alternative 4 – Maximum achieved hourly emissions; two-in-5-year
baseline; output basis

Alternative 5 – Maximum achieved hourly emissions; one-in-5-year
baseline; input basis

Alternative 6 – Maximum achieved hourly emissions; one-in-5-year
baseline; output basis

Alternative 7 – NSPS test – maximum achievable hourly emissions;
input basis

Alternative 8 – NSPS test – maximum achievable hourly emissions;
output basis



In Sections V.A and B, we describe our two broad approaches for the
hourly emissions increase test in more detail.  The regulatory language
proposed today for these approaches (that is, maximum achieved and
maximum achievable hourly emissions increase tests) would apply under
both Option 1 and Option 2.  Option 1 would eliminate the significance
and netting steps that are included under current applicability
regulations, whereas Option 2, as described below in Section VI, would
not eliminate the significance and netting steps.  This action includes
proposed rule language for Option 2 in this respect (see proposed
§51.167(c)).

A.  Test for EGUs Based on Maximum Achieved Emissions Rates  TC \l2 "C. 
Test for EGUs Based on Maximum Achieved Hourly Emissions 

As one approach, we are proposing that the hourly emissions increase
test would be based on an EGU’s historical maximum hourly emissions
rate.  We call this approach the maximum achieved hourly emissions test.
 Under this approach, an EGU owner/operator would determine whether an
emissions increase would occur by comparing the pre-change maximum
actual hourly emissions rate to a projection of the post-change maximum
actual hourly emissions rate.  We request comment on all alternatives
for the maximum achieved hourly emissions increase test (see proposed
Alternatives 1 through 6 for §51.167(f) (1)), as well as on other
possible approaches for determining maximum achieved hourly emissions. 
In particular, we request comments on whether the proposed maximum
achieved methodologies would account for variability inherent in EGU
operations and air pollution control devices.

1.  Determining the Pre-Change Emissions Rate

The pre-change maximum actual hourly emissions rate would be determined
using the highest rate at which the EGU actually emitted the pollutant
within the 5-year period immediately before the physical or operational
change.  Thus, the maximum achieved emissions test is based on specific
measures of actual historical emissions during a representative period.

We are proposing six alternatives for determining the pre-change maximum
hourly emissions rate actually achieved, which we denote here and in the
proposed rule language as Alternatives 1 through 6.  As shown above in
Table 6, these alternatives consist of three different methods for
determining the pre-change maximum emissions rate (i.e., the statistical
approach, the two-in-5-year baseline approach, and the one-in-5-year
baseline approach), each of which can be applied on an input (lb/hr)
basis or output (lb/MWh) basis.  In addition to these six alternatives,
which are included in the proposed 

rule language at §51.167(f) (1), we a proposing that the source would
have a choice of implementing the test on either an input- or
output-basis.

Proposed Alternatives 1 and 2 (input basis and output basis,
respectively) utilize a statistical approach for you to use to analyze
CEMS or PEMS data from the 5 years preceding the physical or operational
change to determine the maximum actual pollutant emissions rate.  The
statistical approach utilizes actual recorded data from periods of
representative operation to calculate the maximum actual emissions rate
associated with the pre-change maximum actual operating capacity in the
past 5 years.  The maximum actual emissions rate is expressed as the
upper tolerance limit (UTL).  The UTL concept and equations are derived
from work conducted by the National Bureau of Standards (now the
National Institute of Standards and Technology (NIST)). 

In conducting the analysis, you would select a period of 365 consecutive
days from the 5 years preceding the change.  Next, you would compile a
data set (for example, in a spreadsheet) for the pollutant of interest
with the hourly average CEMS or PEMS (as applicable) measured emissions
rates (in lb/hr for Alternative 1, or lb/MWh for Alternative 2) and
corresponding heat input data for all of the EGU operating hours in that
period.  From that data set, you would delete selected hourly data from
this 365-day period in accordance with certain data limitations. 
Specifically, you would delete data from periods of startup, shutdown,
and malfunction; periods when the CEMS or PEMS was out of control (as
described below); and periods of noncompliance, according to proposed
§51.167(f) (2) as explained below in Section V.A.3 on data limitations.

The next step in the procedure is to sort the data set for the remaining
operating hours by heat input rates.  You would then extract the hourly
data for the 10 percent of the data set corresponding to the highest
heat input rates for the selected period.  The next step is to apply
basic statistical analyses to the extracted CEMS or PEMS hourly
emissions rate data, calculating the average emissions rate, the
standard deviation, and finally the UTL.  See the proposed rule language
for Alternatives 1 and 2 at §51.167(f) (1) for the specifics of the
calculations.  As included in the proposed rule, Alternatives 1 and 2
calculate the UTL for the 99.9th percentile of the population (of hourly
emissions rate readings) at the 99 percent confidence level.  That is,
under the proposed methodology we would expect, with a 99 percent
confidence level, 99.9 percent of the hourly emissions rate data to be
less than the UTL value.  We are also proposing a 90 percentile of the
population (of hourly emissions rate readings).  We request comment on
these proposed levels.  In particular we request comment on whether a 99
or 90 percentile of the population (of hourly emissions rate readings)
would be more appropriate.  We also request comment on whether a 95 or
90 percent confidence level would be more appropriate.

Alternatives 1 and 2 focus on EGU emissions during periods of
representative operation at the greatest actual operating capacity of
the unit, as demonstrated over the preceding 5 years (that is, the
capacity that the unit actually utilized in the preceding 5 years).  We
believe that this is appropriate for a test with the purpose of,
essentially, determining whether a physical or operational change
increases the capacity of the unit, or the capacity utilization of the
unit, over that achieved in the past 5 years.  We further believe that
the statistical approach properly accounts for the variability inherent
in EGU operations and air pollution control technology.  This approach
helps to ensure that the emissions from an EGU will not exceed its
pre-change maximum achieved hourly emissions rate simply through the
random variability of the system, when a change has not expanded the
capacity of the unit.  Thus, the statistical approach utilizes actual
recorded data from periods of representative operation to calculate the
maximum actual hourly emissions rate in the past 5 years.  We expect
that for the most part, this rate will be associated with the pre-change
maximum actual operating capacity during this period.

Because Alternatives 1 and 2 can be used only if one has CEMS or PEMS
data, we cannot adopt these alternatives alone.  That is, if we elect to
include either or both of these alternatives in the final rule, we will
also finalize another alternative to be used for emissions of any
regulated NSR pollutants that a source does not measure directly with a
CEMS or PEMS. 

While we believe that the statistical approach would be best applied to
hourly emissions data from the periods of highest heat input rates, we
also propose and request comment on the option of sorting and extracting
data based on the hourly emissions rate itself in lb/hr or lb/MWh, as
applicable.  In this alternative method for conducting the statistical
approach, you would compile a data set in the same manner as in
Alternatives 1 and 2.  As in Alternatives 1 and 2, you would delete
selected hourly data from this 365-day period in accordance with the
same data limitations.  Specifically, you would delete data from periods
of startup, shutdown, and malfunction; periods when the CEMS or PEMS was
out of control (as described below); and periods of noncompliance, as
defined in proposed §51.167(f) (2).  However, the data would then be
sorted by the recorded hourly average emissions rates, rather than by
heat input rates.  You would then extract the hourly data for the 10
percent of the data set corresponding to the highest hourly emissions
rate readings for the selected period.  You would next apply basic
statistical analyses to the extracted CEMS or PEMS hourly emissions rate
data, calculating the average emissions rate, the standard deviation,
and finally the UTL.  Under this alternate statistical method based on
recorded hourly emissions rates, we are proposing a 99.9 percentile of
the population (of hourly emissions rate readings) at a 99 percent
confidence level.  That is, under the proposed methodology we would
expect, with a 99 percent confidence level, 99.9 percent of the hourly
emissions rate data to be less than the UTL value.  We are also
proposing a 90 percentile of the population (of hourly emissions rate
readings).  We request comment on these proposed levels.  In particular
we request comment on whether a 99 or 90 percentile of the population
(of hourly emissions rate readings) would be more appropriate.  We also
request comment on whether a 95 or 90 percent confidence level would be
more appropriate.

Proposed Alternatives 3 and 4 for determining the pre-change maximum
actual emissions rate involves calculating the average of the two
highest hourly readings in any period of 24 consecutive months (selected
by the source) within the 5-year period immediately before the physical
or operational change.  This baseline period for pre-change emissions is
analogous to the consecutive 24-month period used to determine baseline
actual emissions for the actual-to-projected-actual applicability test
in the existing major NSR program.  The source would determine the
highest emissions rate (in lb/hr for Alternative 3, or lb/MWh for
Alternative 4) actually achieved for 1 hour in the first 12 months and
for 1 hour in the second 12 months of the selected 24-month period,
where the two 1-hour periods must also fall in different calendar years.

Proposed Alternatives 5 and 6 use the highest emissions rate (in lb/hr
and lb/MWh, respectively) actually achieved for any hour within the
5-year period immediately before the physical or operational change. 
That is, the pre-change maximum emissions rate could be an emissions
rate that was actually achieved for only 1 hour in the 5-year period.  

Under Alternatives 3, 4, 5, and 6, the highest hourly emissions rate
would be determined based on historical actual emissions.  You must
determine the highest pre-change hourly emissions rate for each
regulated NSR pollutant using the best data available to you.  You must
use the highest available source of data in the hierarchy presented
below, unless your reviewing authority has determined that a data source
lower in the hierarchy will provide better data for your EGU:

Continuous emissions monitoring system.

Approved PEMS.

Emission tests/emission factor specific to the EGU to be changed.

Material balance.

Published emission factor (such as AP-42). 

Under this hierarchy, most EGUs will use CEMS to measure the highest
hourly SO2 and NOx emissions.  Some EGUs are currently equipped with
CEMS to measure CO, and would thus use CEMS to measure historical hourly
CO emissions.  For other pollutants, we anticipate most EGUs would
measure historical actual emissions using emission tests, site-specific
emission factors, or mass balances (where applicable).  We request
comment on appropriate measures of historical actual emissions for all
regulated NSR pollutants for all EGUs.  In particular, we request
comment on appropriate measures of historical actual emissions of CO,
VOC, and lead, as turbines may not have significant emissions of these
regulated NSR pollutants.  We also request comment on whether emission
factors that are not site-specific, such as those in AP-42, would be
appropriate measures of historical actual emissions.

As discussed above, proposed Alternatives 1, 3, and 5 provide specific
proposed rule language for the input-based (lb/hr) alternatives. 
Proposed Alternatives 2, 4, and 6 provide specific proposed rule
language for the output-based (lb/MWh) alternatives, largely repeating
the proposed language for Alternatives 1, 3, and 5, respectively.  For
purposes of the output-based alternatives, the proposed language for
their input-based counterparts is adjusted in the following ways:

Emissions rates would be expressed in terms of lb/MWh, rather than
lb/hr.  

For EGUs that are cogeneration units, emissions rates would be
determined based on gross energy output.  For other EGUs, emissions
rates would be determined based on gross electrical output. 

Actual and projected emissions rates in lb/MWh would be determined over
a 1-hour averaging period (that is, a period of one hour of continuous
operation, rather than an instantaneous spike).

We are proposing a gross output basis for this test, rather that net
output, due to the difficulties involved in determining net output. 
This gross output basis is consistent with our recent revisions to the
NSPS for EUSGUs (40 CFR part 60, subpart Da; 71 FR 9866) and stationary
combustion turbines (40 CFR part 60, subpart KKKK; 71 FR 38487).  

For the output-based alternatives, we propose to cite the definitions in
the CAIR rule at §51.124(q) for the definitions of “cogeneration
unit” and numerous other terms used in that definition.  We propose to
include definitions in §51.167(h) (2) of this rule for “gross
electrical output” and “gross energy output.”  We propose to add
definitions for “gross power output” and “useful thermal energy
output,” which are terms used in the proposed definition of “gross
energy output.”  We invite comment on the output-based approach in
general, the proposed output-based alternatives, and the related
definitions we are proposing.

2.  Determining the Post-Change Emissions Rate

We are proposing the same approach to post-change emissions for
Alternatives 1 through 6.  Specifically, for each regulated NSR
pollutant, you must project the maximum emissions rate that your EGU
will actually achieve in any 1 hour in the 5 years following the date
the EGU resumes regular operation after the physical or operational
change.  An emissions increase results from the physical or operational
change if this projected maximum actual hourly emissions rate exceeds
the pre-change maximum actual hourly emissions rate.  Regardless of any
preconstruction projections, you must treat an emissions increase as
occurring if the emissions rate actually achieved in any 1 hour during
the 5 years after the change exceeds the pre-change maximum actual
hourly emissions rate.

3.  Data Limitations in Determining Emissions Rates

We are proposing four limitations on the data used to determine
pre-change and post-change maximum emissions rates under the maximum
achieved hourly emissions test (see proposed §51.167(f)(2)(i)).  The
proposed limitations are identical for Alternatives 1 through 6.  For
purposes of determining maximum emissions rates under the maximum
achieved test, we propose that you must not use the following types of
data in your calculations: 

Emissions rate data associated with startups, shutdowns, or malfunctions
of your EGU, as defined by applicable regulation(s) or permit term(s),
or malfunctions of an associated air pollution control device.  A
malfunction means any sudden, infrequent, and not reasonably preventable
failure of the EGU or the air pollution control equipment to operate in
a normal or usual manner.

Continuous emissions monitoring system (CEMS) or predictive emissions
monitoring system (PEMS) data recorded during monitoring system
out-of-control periods.  Out-of-control periods include those during
which the monitoring system fails to meet quality assurance criteria
(for example, periods of system breakdown, repair, calibration checks,
or zero and span adjustments) established by regulation, by permit, or
in an approved quality assurance plan.

Emissions rate data from periods of noncompliance when your EGU was
operating above an emission limitation that was legally enforceable at
the time the data were collected.

Data from any period for which the information is inadequate for
determining emissions rates, including information related to the
limitations listed above.

The first two of these limitations are based on requirements of the NSPS
General Provisions in subpart A of part 60.  The prohibition of data
from periods of startup, shutdown, and malfunction is found in the
section on performance tests, specifically §60.8(c), which states, in
pertinent part:

Operations during periods of startup, shutdown, and malfunction shall
not constitute representative conditions for the purpose of a
performance test nor shall emissions in excess of the level of the
applicable emission limit during periods of startup, shutdown, and
malfunction be considered a violation of the applicable emission limit
unless otherwise specified in the applicable standard.

The principle set out in this paragraph is that emissions during periods
of startup, shutdown, and malfunction are not representative and
typically should not figure into emission calculations.  We propose to
apply this principle to all data required to comply with the
requirements in this action, and not limit it to performance test data. 
We do not believe that emissions during startup, shutdown, or
malfunction are a reasonable basis for determining whether a physical or
operational change at an EGU would result in an hourly emissions
increase.  It is more appropriate to focus on emissions during normal
operations, which are expected to correlate more closely with the actual
operating capacity of the EGU than would emissions during periods of
startup, shutdown, or malfunction.  The proposed rule language also
expands slightly on the language of §60.8(c) to clarify the meanings of
startup, shutdown, and malfunction in the context of this action.

The second data limitation reflects §60.13(h), which states that
“data recorded during periods of continuous system breakdown, repair,
calibration checks, and zero and span adjustments shall not be included
in data averages computed under this paragraph.”  We do not believe
that this type of unrepresentative CEMS or PEMS data, which may bear no
relationship to actual emissions, should be included in calculations of
maximum achieved emissions rates.  The proposed rule language refers to
and defines “monitoring system out-of-control periods,” in keeping
with more current terminology for monitoring systems.

The third proposed data limitation listed above would prohibit the use
of emissions rate data from periods of noncompliance when your EGU was
operating above an emission limitation that was legally enforceable at
the time the data were collected.  This reflects existing requirements
under the major NSR program, specifically the definition of “baseline
actual emissions” that is used in the actual-to-projected-actual
applicability test.  (See, for example, §51.166(b) (47) (i) (b).)  

The fourth proposed data limitation reflects existing requirements under
the major NSR program, again in the definition of “baseline actual
emissions” that is used in the actual-to-projected-actual
applicability test.  (See, for example, §51.166(b) (47) (i) (d).)  This
limitation would preclude the use of data from periods where there is
inadequate information for determining emissions rates, including
information related to the other three data limitations.  This provision
is simply intended to ensure that you generate reliable, defensible
values for pre-change and post-change emissions rates.

4.  Recordkeeping and Reporting Requirements

Finally, we are proposing to incorporate provisions that are generally
analogous to those in §60.7(a) (4) and §60.7(f) concerning
notifications of proposed physical changes and changes in the method of
operation (proposed §51.167(g) (1) (i)), and records of such changes
(proposed §51.167(g) (2)).  The proposed requirements are identical for
Alternatives 1 through 6.

Specifically, you must provide a notification to the reviewing authority
at least 6 months before commencing construction on any physical or
operational change to an existing EGU that may increase the emissions
rate of any regulated NSR pollutant.  The notification must contain the
information in proposed §51.167(g) (1) (i) and the reviewing authority
may request additional relevant information after receiving the
notification.  

Although §60.7(a)(4) requires a notification only 60 days, or as soon
as practicable, before a change is commenced, we are proposing a 6-month
advance notification for Alternatives 1 through 6.  We believe that the
maximum achieved hourly emissions tests that we are proposing under
Alternatives 1 through 6 may entail relatively complex analyses, and
that reviewing authorities may need more than 60 days to review and
evaluate the notifications and analyses.  Further, we believe that it is
reasonable to require a notification 6 months in advance of a change
because changes to EGUs must be planned for a shutdown of the unit,
which are typically planned more than a year in advance.  We invite
comment on the practicality of requiring notifications 6 months in
advance under the maximum achieved hourly emissions test alternatives
proposed today.

You must also maintain a file of all information related to
applicability determinations that you make under this section, including
the specific information in proposed §51.167(g) (2).  These proposed
recordkeeping requirements are drawn from the requirements of §60.7(f).
 We are proposing that you must maintain the records until the later of
the following dates:  (1) 5 years after the EGU resumes regular
operation after the physical or operational change, and (2) 5 years
after the record was recorded.  This expands on the 2-year requirement
in §60.7(f) to be consistent with more recent recordkeeping
requirements, such as in the title V operating permit program (see 40
CFR 70.6(a) (3) (ii) (B)).

Under proposed Alternatives 1 through 6, regardless of any
preconstruction projections, an emissions increase has occurred if the
emissions rate actually achieved in any one hour during the 5 years
after the change exceeds the pre-change maximum actual hourly emissions
rate (see, for example §51.167(f)(1)(iii) under Alternative 1).  Most
EGUs are already reporting hourly SO2 and NOx emissions through CEMS
data to EPA.  Therefore, the majority of emissions increases of
regulated NSR pollutants will be transparent to the Agency and to the
public.  However, we request comment on whether additional recordkeeping
and reporting requirements for post-change emissions should be required
where EGUs are not using CEMS to measure emissions.

B.  Test for EGUs Based on Maximum Achievable Emissions Rates

As we stated in our October 2005 NPR (70 FR 61090), we are proposing to
allow existing EGUs to use the same maximum achievable hourly emissions
test applied in the NSPS to determine whether a physical or operational
change results in an emissions increase under the major NSR program. 
This test is based on a comparison of pre-change and post-change
emissions rates in pounds per hour (lb/hr).  Today we are proposing an
additional variation on the NSPS test, which would compare pre-change
and post-change achievable emissions rates in pounds per megawatt-hour
(lb/MWh).  In the discussion that follows and in the proposed rule
language, we refer to these two approaches as Alternatives 7 and 8,
respectively.  

1.  Determining Pre-Change and Post-Change Emissions Rates

Under Alternative 7, the major NSR regulations would apply at an EGU if
a physical or operational change results in any increase above the
maximum hourly emissions achievable at that unit during the 5 years
prior to the change.  Under this alternative, we are proposing to
incorporate provisions similar to those in §60.14(h) into the new
§51.167(f) (1).  We propose that this regulatory language would
substantially mirror, but would not be identical to, §60.14(h).  As
with the definition of modification that we are proposing for
§51.167(h) (2), there are differences between the two programs that
prevent a wholesale adoption of the NSPS modification provisions of
§60.14(h).  Specifically, our proposed rule language addresses the full
range of pollutants regulated under the major NSR program by referring
to the “regulated NSR pollutants,” while the NSPS provisions limit
the analysis to those pollutants regulated under an applicable NSPS. 
Also, as we previously explained at 70 FR 61090, we are proposing that
the emissions increase test would apply to EGUs, rather than to EUSGUs. 
Under Alternative 7, §51.167(f) (1) would read as follows:

Emissions increase test.  For each regulated NSR pollutant, compare the
maximum achievable hourly emissions rate before the physical or
operational change to the maximum achievable hourly emissions rate after
the change.  Determine these maximum achievable hourly emissions rates
according to §60.14(b) of this chapter.  No physical change, or change
in the method of operation, at an existing EGU shall be treated as a
modification for the purposes of this section provided that such change
does not increase the maximum hourly emissions of any regulated NSR
pollutant above the maximum hourly emissions achievable at that unit
during the 5 years prior to the change.

As stated in this proposed rule language, pre-change and post-change
hourly emissions rates would be determined according to the NSPS
provisions in §60.14(b).  That is, hourly emissions increases would be
determined using emission factors, material balances, continuous monitor
data, or manual emission tests.

Alternative 8 is also based on the NSPS “maximum achievable” test,
but is modified to an energy output (lb/MWh) basis.  Under Alternative
8, §51.167(f) (1) would read as follows:

Emissions increase test.  For each regulated NSR pollutant, compare the
maximum achievable emissions rate in pounds per megawatt-hour (lb/MWh)
before the physical or operational change to the maximum achievable
emissions rate in lb/MWh after the change.  Determine these maximum
achievable emissions rates according to §60.14(b) of this chapter,
using emissions rates in lb/MWh achievable over 1 hour of continuous
operation in place of mass emissions rates.  For EGUs that are
cogeneration units, determine emissions rates based on gross energy
output.  For other EGUs, determine emissions rates based on gross
electrical output.  No physical change, or change in the method of
operation, at an existing EGU shall be treated as a modification for the
purposes of this section provided that such change does not increase the
maximum emissions rate of any regulated NSR pollutant above the maximum
emissions rate achievable at that unit during the 5 years prior to the
change.

To maintain an hourly basis for the emissions rate, the proposed
language specifies that the maximum achievable emissions rate in lb/MWh
is to be determined based on what is achievable over 1 hour of
continuous operation (that is, a 1-hour averaging period rather than an
instantaneous spike).  In addition, as noted above in the discussion of
the output-based alternatives under the maximum achieved hourly
emissions test (Alternatives 2, 4, and 6), we propose to cite the
definition in the CAIR rule at §51.124(q) for the definitions of
“cogeneration unit” and related terms.  We propose to include
definitions in §51.167(h) (2) of this rule for “gross electrical
output,” “gross energy output,” “gross power output,” and
“useful thermal energy output.” 

2.  Data Limitations in Determining Emissions Rates

We are proposing three limitations on the data used to calculate the
pre-change and post-change emissions rates under the maximum achievable
hourly emissions test (see proposed §51.167(f) (2) (ii)).  The proposed
limitations are identical for Alternatives 7 and 8.  For purposes of
determining maximum emissions rates under the maximum achievable test,
we propose that you must not use the following types of data in your
calculations: 

Emissions rate data associated with startups, shutdowns, or malfunctions
of your EGU, as defined by applicable regulation(s) or permit term(s),
or malfunctions of an associated air pollution control device.  A
malfunction means any sudden, infrequent, and not reasonably preventable
failure of the EGU or the air pollution control equipment to operate in
a normal or usual manner.

Continuous emissions monitoring system (CEMS) or predictive emissions
monitoring system (PEMS) data recorded during monitoring system
out-of-control periods.  Out-of-control periods include those during
which the monitoring system fails to meet quality assurance criteria
(for example, periods of system breakdown, repair, calibration checks,
or zero and span adjustments) established by regulation, by permit, or
in an approved quality assurance plan.

Data from any period for which there is inadequate information for
determining emissions rates, including information related to the
limitations listed above.

These proposed data limitations are the same as three of the four data
limitations that we are proposing for the maximum achieved tests
(Alternatives 1 through 6).  See Section V.A.3. above for the discussion
of these three data limitations.	 

3.  Recordkeeping and Reporting requirements

We are proposing nearly the same recordkeeping and reporting
requirements for the maximum achievable test (Alternatives 7 and 8) that
we propose for the maximum achieved hourly emissions test (Alternatives
1 through 6).  The only difference is in the amount of advance
notification required.  See Section V.A.4 of this preamble for the
discussion of the common requirements.  The proposed rule language is
found at §51.167(g) (1) (ii) and (2).

The notification requirements are based on the requirements found in
§60.7(a) (4).  These NSPS provisions require a notification 60 days, or
as soon as practicable, before the change is commenced.  Although we
propose to require a 6-month advance notification under the maximum
achieved test for the reasons discussed above in Section V.A.4 of this
preamble, we propose to retain the shorter notification requirement of
§60.7(a)(4) for the maximum achievable test.  Because determinations of
pre-change and post-change maximum hourly emissions rates typically are
made using emission factors under the maximum achievable test, we
believe that a notification 60 days, or as soon as practicable, in
advance of commencing the change is adequate to allow the reviewing
authority to review and evaluate the notification.

VI.   Proposed Regulations for Option 2:  Hourly Emissions Increase Test
Followed By Annual Emissions Test  

In the NPR, we did not propose to include, along with any of the revised
NSR emissions tests, any provisions for computing a significant increase
or a significant net emissions increase, although we solicited comment
on retaining such provisions.  Many commenters preferred to retain an
annual emissions increase test in addition to the hourly emissions
increase test.  Today, we are proposing Option 2, in which the hourly
emissions increase test under Option 1 above (whether the hourly
achieved test or the hourly achievable test) would be followed by the
actual-to-projected-actual emissions increase test and the significant
net emissions increase test in the current regulations.

Specifically, under this Option 2, the major NSR program would continue
to include a four-step process (with the second step revised as proposed
today, and with no proposals concerning the other steps):  (1) physical
change of change in the method of operation as in the current major NSR
regulations; (2) hourly emissions increase test (maximum achieved hourly
emissions rate or maximum achievable hourly emissions rate, each with
output-based alternatives, all as described under Option 1); (3)
significant emissions increase as in the current major NSR regulations;
and (4) significant net emissions increase as in the current major NSR
regulations.

Under Option 2, Steps 1 and 2 would be the same as under Option 1.  That
is, for a modification to occur, under Step 1, a physical change or
change in the method of operation must occur, and, under Step 2, that
change must result in an hourly emissions increase at the existing EGU. 
Unlike Option 1, however, Option 2 retains the requirements for a
significant emissions increase and a significant net emissions increase.
 Therefore, if a post-change hourly emission increase is projected,
under Step 3, the owner/operator would determine whether an emissions
increase would occur using the actual-to-projected-actual annual
emissions test in the current regulations.  There would be no conversion
from annual to hourly emissions.  Finally, in Step 4, as in the current
regulations, if a significant emissions increase is projected to occur,
the source would still not be subject to major NSR unless there was a
determination that a significant net emissions increase would occur. 
Table 7 summarizes these four steps, with today’s proposal
substituting Option 2 for the current annual emissions increase test.

Table 7.  Major NSR Applicability for Existing EGUs Under Option 2

Option 2	Step 1:  Physical Change or Change in the Method of Operation

Step 2:  Hourly Emissions Increase Test

Alternative 1 – Maximum achieved hourly emissions; statistical
approach; input basis

Alternative 2 – Maximum achieved hourly emissions; statistical
approach; output basis

Alternative 3 – Maximum achieved hourly emissions; two-in-5-year
baseline; input basis

Alternative 4 – Maximum achieved hourly emissions; two-in-5-year
baseline; output basis

Alternative 5 – Maximum achieved hourly emissions; one-in-5-year
baseline; input basis

Alternative 6 – Maximum achieved hourly emissions; one-in-5-year
baseline; output basis

Alternative 7 – NSPS test – maximum achievable hourly emissions;
input basis

Alternative 8 – NSPS test – maximum achievable hourly emissions;
output basis

Step 3: Significant Emissions Increase Determined Using the
Actual-to-Projected-Actual Emissions Test as in the Current Rules

Step 4:  Significant Net Emissions Increase as in the Current Rules



We solicit comment on whether we should adopt a 10-year baseline period
under Step 3 of Option 2.

As with Option 1, under Option 2 we are proposing to develop two new
sections (40 CFR 51.167 and 52.37) to the major NSR program rules that
would include the two-step provisions for modifications at EGUs.  Thus,
the amendatory language in this action applies to Option 2.

Unlike Option 1, however, Option 2 would not eliminate—and therefore
would have no effect on—the provisions in the current major NSR
regulations pertaining to a significant emissions increase and a
significant net emissions increase.  Therefore, the regulations would
retain the definitions of net emissions increase, significant, projected
actual emissions at and baseline actual emissions.  [See §51.166(b)
(3), §51.166(b) (23), §51.166(b) (40), §51.166(b) (47), and analogous
provisions in 40 CFR 51.165, 52.21, 52.24, and appendix S to 40 CFR part
51.]  The regulations would also retain all provisions in the current
regulations that refer to major modifications, including, but not
limited to,  those in §51.166(a)(7)(i) though (iii), (b)(9), (b)(12),
(b)(14)(ii), (b)(15), (b)(18), (i)(1) through (9), (j)(1) through (4),
(m)(1) through (3), (p)(1) through (7), (r)(1) though (7), and (s)(1)
through (4) analogous provisions in 40 CFR 51.165, 52.21, 52.24, and
appendix S to 40 CFR part 51.  

We are proposing that Option 2 would apply to all EGUs.  As with Option
1, however, we are also requesting comment on whether Option 2 should be
limited to the geographic area covered by CAIR, or to the geographic
area covered by both CAIR and BART.  We are also proposing that the
Option 2 would apply to all regulated NSR pollutants.  However, we also
request comment on whether Option 2 should be limited to increases of
SO2 and NOx emissions. 

Because Option 2 contains an hourly emissions increase test followed by
an annual emissions increase test, it does not provide the
simplification benefits as Option 1 does.  It also would not provide a
single applicability test for the major NSR and NSPS programs, as the
achievable alternative for Option 1 would.  However, Option 2 does
provide much the same benefits of facilitating improvements for
efficiency, safety, and reliability, without adverse air quality effects
(as the above discussion of the IPM and air quality analyses indicates).
 In addition, it includes netting.

VII. Proposed Regulations for Option 3:  Annual Emissions Test with
10-Year Baseline Period

In this action, we are proposing an additional option that was not
included in the NPR.  Specifically, we are proposing to retain the
actual-to-projected-actual emissions increase test for EUSGUs in the
current regulations, but to extend the period in which baseline actual
emissions are determined from 5 to 10 years.  That is, we are proposing
that the owner or operator would select any consecutive 24-month period
within the 10-year period immediately preceding when the owner or
operator begins actual construction of the project.  

Under Option 3, we are proposing that all of the requirements in the
major NSR program that pertain solely to EUSGUs would now apply to EGUs.
 The current regulations apply to EUSGUs, rather than EGU.  As we noted
at 70 FR 61090, the definition of EGU is broader than the definition of
EUSGU currently found in the NSR regulations.  The EGU definition
includes simple cycle gas turbines that are not included in the EUSGU
definition.  Under the current regulations, the 10-year baseline period
thus already applies to simple cycle gas turbines.  We believe it is
advisable to have one definition for the electric utility sector across
the CAIR, NSPS, and NSR programs.  Therefore, we are proposing to
incorporate the term EGU, as opposed to EUSGU, in all the major NSR
programs.  Specifically, we are proposing to remove the definition of
EUSGUs in §51.165(a) (1) (xx) and replace it with a definition of EGU
consistent with that in the CAIR and NSPS programs.  We would also
replace the term EUSGU with the term EGU in the provisions for
determining baseline actual emissions for EUSGUs in §51.165(a) (1)
(xxxv) (A), the requirements for reporting EUSGU projected actual
emissions in §51.165(a) (6) (C) (ii).  We would replace the term EUSGU
with EGU in analogous provisions in all sections of the Code of Federal
Regulations that contain the major NSR program, including 40 CFR 51.165,
51.166, 52.21, 52.24, and appendix S of 40 CFR part 51, as well as any
regulations we finalize to implement major NSR in Indian Country.  We
are also proposing to make the same changes where necessary to conform
the general provisions in parts 51 and 52 to the requirements of the
major NSR program.

VIII.   Legal Basis and Policy Rationale 

This section supplements the legal arguments in our October 2005
proposal.     (70 FR 70565.)  In that action, we provided our legal
basis and rationale for the proposed maximum achievable hourly emissions
test and our alternative proposal, the maximum achieved hourly emissions
test.  We noted that the key statutory provisions provide, in relevant
part, that a “modification” that triggers NSR occurs when a physical
change or change in the method of operation “increases the amount of
any air pollutant emitted” by the source.  Although the Court in New
York v. EPA held that the quoted provision refers to increases in actual
emissions, the Court further indicated that the statute was silent as to
the method for determining whether increases occur.  

When a statute is silent or ambiguous with respect to specific issues,
the relevant inquiry for a reviewing court is whether the Agency(s
interpretation of the statutory provision is permissible.  Chevron
U.S.A., Inc. v. NRDC, Inc. 467 U.S. 837, 865 (1984).  Accordingly, we
have broad discretion to propose a reasonable method by which to
calculate emissions increases for purposes of NSR applicability.  We
believe that our discretion extends to determining whether emissions
increases occur by reference to maximum hourly emissions rate and to
retaining an annual emissions test but extending the baseline period to
10 years.

This action continues to propose both the maximum achievable hourly
emissions increase test and the maximum achieved hourly emissions
increase test.  We set forth legal basis and rationale in the NPR for
these two tests.  In today’s SNPR, however, we provide additional
legal and policy basis for the hourly emissions increase tests, on both
an input and output basis, as well as for the option to retain the
annual emissions test but extend the baseline to 10 years.

We believe that a test based on maximum actual hourly emissions is a
reasonable measure of actual emissions.  It measures actual emissions at
peak, or close to peak, physical and operational capacity.  For reasons
described elsewhere, and summarized below, we believe this approach
implements sound policy objectives.

As we noted at 70 FR 61091, we believe that a test based on maximum
achievable hourly emissions remains a test based on actual emissions. 
The reason is that, as noted in the October 2005 proposal, as a
practical matter, for most, if not all EGUs, the hourly rate at which
the unit is actually able to emit is substantively equivalent to that
unit’s historical maximum hourly emissions.  That is, most, if not all
EGUs will operate at their maximum actual physical and operational
capacity at some point in a 5-year period.  In general, highest
emissions occur during the period of highest utilization.  As a result,
both the maximum achievable and maximum achieved hourly emissions
increase tests allow an EGU to utilize all of its existing capacity, and
in this aspect the hourly rate at which the unit is actually able to
emit is substantively equivalent under both tests.  

Some commenters took issue with this statement, arguing that maximum
achievable emissions could differ from maximum achieved emissions for a
given EGU for any given period as a result of factors independent of the
physical or operational change, including variability of the sulfur
content in the coal being burned.   

We have long recognized that the highest hourly emissions do not always
occur at the point of highest capacity utilization, due to fluctuations
in process and control equipment operation, as well as in fuel content
and firing method.  In fact, we justified an emission factor approach as
our preferred approach when we proposed the NSPS regulations at §60.14
in 1974.  (See 39 FR 36947.)  As we also noted in developing these NSPS
provisions for modifications, “measurement techniques such as emission
tests or continuous monitors are sensitive to routine fluctuations in
emissions, and thus a method is needed to distinguish between
significant increases in emissions and routine fluctuations in
emissions.”  (39 FR 36947.)  At that time, we proposed a statistical
method for use with stack tests and continuous monitors to measure
actual emissions to address this issue.  

In light of these concerns, we developed a statistical approach for the
maximum achieved hourly emissions increase test to assure that it
identifies the maximum hourly pollutant emissions value (for example
maximum lb/hr NOx during a specific one-year period).  The statistical
procedure would provide an estimate of the highest value (99.9
percentage level) in the period represented by the data set.  We believe
that this approach mitigates some of the uncertainty associated with
trying to identify the highest hourly emissions rate at the highest
capacity utilization.  We thus believe that, over a period that is
representative of normal operation, in general the maximum achievable
and maximum achieved hourly emissions test would lead to substantially
equivalent results. 

Option 1 does not include steps for determining whether significant net
emissions increases have occurred.  We recognize that the D.C. Circuit,
in the seminal case, Alabama Power v. EPA, 636 F.2d 323 (D.C. Cir.
1980), which was handed down before Chevron, held that failure to
interpret “increases” to allow netting  would be “unreasonable and
contrary to the expressed purposes of the PSD provisions....”  Id. at
401.  As we noted at 70 FR 61093, it is important to place this ruling
in the context of the rules before the Court at that time.  Our 1978
regulations required a source-wide accumulation of emissions increases
without providing for an ability to offset these accumulated increases
with any source-wide decreases.  In finding that we must apply a bubble
approach, the Court held that we could not require sources to accumulate
increases without also accumulating decreases.  It is unclear whether
the Court would have reached the same conclusion if the emissions test
before the Court only considered the increases from the project under
review and not source-wide increases from multiple projects. 

Moreover, the Court's rationale focused on the ability to “net” as
it relates to the addition of new units, rather than to changes existing
units, which is the concern of our proposed action.   Specifically, the
court stated:

         It is important first to recognize that alternations of almost
any plant occur continuously; whether to replace depreciated capital
goods, to keep pace with technological advances, or to respond to
changing consumer demands.  This dynamic aspect of American industry was
not disputed by the parties.  To [construe “increases” not to allow
netting], however, would require PSD review for many such routine
alterations of a plant; a new unit would contribute additional
pollutants, these increases could not be set off against the decrease
resulting from abandonment of the old unit, and thus the change would be
come a ‘modification’ subject to PSD review.  Not only would this
result be extremely burdensome, it was never intended by Congress in
enacting the Clean Air Act Amendments....  Congress wished to apply
[PSD]..., only where industrial changes might increase pollution in an
area, not where an existing plant changed its operations in ways that
produced no pollution increase....

The Court further stated that Congress intended to “generate
technological improvement in pollution control, but this approach
focused upon ‘rapid adoption of improvements in technology as new
sources are built,’ not as old ones [plants] were changed without
pollution increases.”  Id at xxx.  Importantly, some have argued that
the netting approach may have impeded Congress’ objective of promoting
“rapid adoption of improvements in technology as new sources are
built.”  This is because it allows construction of new units at
existing facilities without emissions controls, while requiring major
NSR for large greenfield sources.  Nevertheless, because today's action
does not change the emissions test that applies when a major stationary
sources adds a new emissions unit, the existing regulations continue to
satisfy the Alabama Power Court's interpretation of Congressional
intent, irrespective of today's proposed changes.   We request comment
on our observations related to the Alabama Power Court’s decision
related to netting and whether a major NSR program without netting can
be supported under the Act.  

With respect to the significance levels, which, like netting, are not
included under Option 1, we recognize that Alabama Power also upheld
significance levels as a “permissible ... exercise of agency power,
inherent in most statutory schemes, to overlook circumstances that in
context may fairly be considered de minimis.”  Id. At 360.  It is
clear, however, that the Court considered the establishment of
significance levels as discretionary.  We believe that significance
levels are not important to include in the rules proposed in Option 1
because under those rules, relatively minor changes for which the
significance levels might come into play would not increase the maximum
hourly rate.  By comparison, the changes that do increase the maximum
hourly rate are likely to be capacity increases that should not, by
their nature, be considered de minimis. 

Each of today’s proposed options would promote the safety,
reliability, and efficiency of EGUs.  Each of the options would balance
the economic need of sources to use existing operating capacity with the
environmental benefit of regulating those emission increases related to
a change, considering the substantial national emissions reductions
other programs have achieved or will achieve from EGUs.  The proposed
regulations are consistent with the primary purpose of the major NSR
program, which is not to reduce emissions, but to balance the need for
environmental protection and economic growth.  As the analyses included
in today’s SNPR demonstrate, the proposed regulations would not have
an undue adverse impact on local air quality.  Furthermore, as our
analyses demonstrate, increases in hours of operation at EGUs, to the
extent they may change under a maximum hourly rate test or an annual
emissions increase test with a 10-year, as opposed to 5-year, baseline
period, do not increase national SO2 and NOx emissions.  Consistent with
earlier analyses, our analyses demonstrate that in a system where
national emissions are capped, the more hours an EGU operates, the more
likely it is to install controls.

Moreover, each of the proposed options also offers additional benefits
consistent with our overall policy goals.  The proposed maximum hourly
tests would streamline major NSR by reducing applicability
determinations complexity.  The proposed maximum hourly achievable test
provides more streamlining by conforming them to NSPS applicability
determinations.  We also note that Option 1 (both achievable and
achieved alternatives) eliminates the burden of projecting future
emissions and distinguishing between emissions increases caused by the
change from those due solely to demand growth, because any increase in
the emissions under the maximum hourly achievable emissions test would
logically be attributed to the change.  In addition, Option 1 reduces
recordkeeping and reporting burdens on sources because compliance will
no longer rely on synthesizing emissions data into rolling average
emissions.  Option 1 would also reduce the reviewing authorities’
compliance and enforcement burden.

We recognize that Options 2 and 3, which retain an annual emissions
increase test, would not streamline the major NSR program as Option 1
would.  We propose Option 2 for the purpose of maintaining the current
significant net emissions increase component of the emissions increase
test.  In light of the additional complexity that netting adds, we
solicit comment on whether netting and significance levels would retain,
in combination with an hourly test, the usefulness they have under an
annual test.  Option 3 has the benefit of retaining netting, albeit in a
manner less complex than that of Option 2.  Moreover, Option 3 conforms
the applicability test for EGUs to that of all other source categories.

Consistent with our policy goal of encouraging efficient use of existing
energy capacity, we are continuing to propose an output-based format for
the hourly emissions increase tests.  The output-based format is a
measure of actual emissions that relates emissions to the amount of
useful energy generation.  We acknowledge that an output-based format
may not be as effective a measure of existing capacity utilization in
some instances as our input based options.  However, the more efficient
an EGU, the less it emits for a given period of operation.  We believe
the output based standard is likely to result in lower emissions in many
instances because any increase in overall energy efficiency results in a
lower emission rate.  This is especially true in a system where total
annual emissions are capped, as in the case of the acid rain program
(nationally) and CAIR (regionally).  Furthermore, allowing energy
efficiency as a pollution control measure provides regulated sources
with an additional compliance option that can lead to reduced compliance
costs as well as lower emissions.  The use of more efficient
technologies reduces fossil fuel use and leads to multi-media reductions
in environmental impacts both on-site and off-site. 

We request comment on all aspects of our legal and policy basis.

IX. Statutory and Executive Order Reviews tc \l1 "VI.  Statutory and
Executive Order Reviews 

A.  Executive Order 12866:  Regulatory Planning and Review tc \l2 "A. 
Executive Order 12866Regulatory Planning and Review 

Under Executive Order (EO) 12866 (58 FR 51735, October 4, 1993), this
action is a "significant regulatory action."  The action was identified
as a "significant regulatory action" because it raises novel legal or
policy issues.  Accordingly, EPA submitted this action to the Office of
Management and Budget (OMB) for review under EO 12866 and any changes
made in response to OMB recommendations have been documented in the
docket for this action.

B.  Paperwork Reduction Act tc \l2 "B.  Paperwork Reduction Act 

The information collection requirements in this proposed rule have been
submitted for approval to the Office of Management and Budget (OMB)
under the Paperwork Reduction Act, 44 U.S.C. 3501 et seq.  The
Information Collection Request (ICR) document prepared by EPA has been
assigned EPA ICR number 1230.19.

Certain records and reports are necessary for the State or local agency
(or the EPA Administrator in non-delegated areas), for example, to: (1)
confirm the compliance status of stationary sources, identify any
stationary sources not subject to the standards, and identify stationary
sources subject to the rules; and (2) ensure that the stationary source
control requirements are being achieved.  The information would be used
by the EPA or State enforcement personnel to (1) identify stationary
sources subject to the rules, (2) ensure that appropriate control
technology is being properly applied, and (3) ensure that the emission
control devices are being properly operated and maintained on a
continuous basis.  Based on the reported information, the State, local
or tribal agency can decide which plants, records, or processes should
be inspected.

The proposed rule would reduce burden for owners and operators of major
stationary sources.  We expect the proposed rule would simplify
applicability determinations, eliminate the burden of projecting future
emissions and distinguishing between emissions increases caused by the
change from those due solely to demand growth, and reduce recordkeeping
and reporting burdens.  Over the 3-year period covered by the ICR, we
estimate an average annual reduction in burden for all industry entities
that would be affected by the proposed rule.  For the same reasons, we
also expect the proposed rule to reduce burden for State and local
authorities reviewing permits when fully implemented.  However, there
would be a one-time, additional burden for State and local agencies to
revise their SIPs to incorporate the proposed changes.  

Burden means the total time, effort, or financial resources expended by
persons to generate, maintain, retain, or disclose or provide
information to or for a Federal agency.  This includes the time needed
to review instructions; develop, acquire, install, and utilize
technology and systems for the purpose of responding to the information
collection; adjust existing ways to comply with any previously
applicable instructions and requirements; train personnel to respond to
a collection of information; search existing data sources; complete and
review the collection of information; and transmit or otherwise disclose
the information.

An agency may not conduct or sponsor, and a person is not required to
respond to, a collection of information unless it displays a currently
valid OMB control number.  The OMB control numbers for EPA(s regulations
are listed in 40 CFR parts 9.

To comment on the Agency(s need for this information, the accuracy of
the provided burden estimates, and any suggested methods for minimizing
respondent burden, including use of automated collection techniques, EPA
has established a public docket for this rule, which includes this ICR,
under Docket ID number EPA-HQ-OAR-2005-1063.  Submit any comments
related to the ICR for this proposed rule to EPA and OMB.  See
(Addresses( section at the beginning of this notice for where to submit
comments to EPA.  Send comments to OMB at the Office of Information and
Regulatory Affairs, Office of Management and Budget, 725 17th Street,
Northwest, Washington, DC 20503, Attention:  Desk Officer for EPA. 
Since OMB is required to make a decision concerning the ICR between 30
and 60 days after [INSERT DATE OF PUBLICATION IN THE FEDERAL REGISTER],
a comment to OMB is best assured of having its full effect if OMB
receives it by [INSERT DATE 30 DAYS AFTER DATE OF PUBLICATION IN THE
FEDERAL REGISTER].  The final rule will respond to any OMB or public
comments on the information collection requirements contained in this
proposal.

C.  Regulatory Flexibility Act (RFA) tc \l2 "C.  Regulatory Flexibility
Act (RFA) 

The RFA generally requires an agency to prepare a regulatory flexibility
analysis of any rule subject to notice and comment rulemaking
requirements under the Administrative Procedure Act or any other statute
unless the agency certifies that the rule will not have a significant
economic impact on a substantial number of small entities.  Small
entities include small businesses, small organizations, and small
governmental jurisdictions. 

For purposes of assessing the impacts of today's notice on small
entities, small entity is defined as: (1) a small business that is a
small industrial entity as defined in the U.S. Small Business
Administration (SBA) size standards.  (See 13 CFR 121.201); (2) a small
governmental jurisdiction that is a government of a city, county, town,
school district, or special district with a population of less than
50,000; or (3) a small organization that is any not-for-profit
enterprise that is independently owned and operated and is not dominant
in its field.

After considering the economic impacts of today(s notice on small
entities, I certify that this action will not have a significant
economic impact on a substantial number of small entities.  In
determining whether a rule has a significant economic impact on a
substantial number of small entities, the impact of concern is any
significant adverse economic impact on small entities, since the primary
purpose of the regulatory flexibility analyses is to identify and
address regulatory alternatives (which minimize any significant economic
impact of the proposed rule on small entities.(  5 U.S.C. sections 603
and 604.  Thus, an agency may certify that a rule will not have a
significant economic impact on a substantial number of small entities if
the rule relieves regulatory burden, or otherwise has a positive
economic effect, on all of the small entities subject to the rule.

We believe that today(s proposed rule changes will relieve the
regulatory burden associated with the major NSR program for all EGUs,
including any EGUs that are small businesses.  This is because the
proposed rule would simplify applicability determinations, eliminate the
burden of projecting future emissions and distinguishing between
emissions increases caused by the change from those due solely to demand
growth, and by reducing recordkeeping and reporting burdens.  As a
result, the program changes provided in the proposed rule are not
expected to result in any increases in expenditure by any small entity. 


We have therefore concluded that today(s proposed rule would relieve
regulatory burden for all small entities.  We continue to be interested
in the potential impacts of the proposed rule on small entities and
welcome comments on issues related to such impacts.

D.  Unfunded Mandates Reform Act tc \l2 "D.  Unfunded Mandates Reform
Act 

Title II of the Unfunded Mandates Reform Act of 1995 (UMRA), P.L. 104-4,
establishes requirements for Federal agencies to assess the effects of
their regulatory actions on State, local, and tribal governments and the
private sector.  Under section 202 of the UMRA, EPA generally must
prepare a written statement, including a cost-benefit analysis, for
proposed and final rules with "Federal mandates" that may result in
expenditures to State, local, and tribal governments, in the aggregate,
or to the private sector, of $100 million or more in any one year. 
Before promulgating an EPA rule for which a written statement is needed,
section 205 of the UMRA generally requires EPA to identify and consider
a reasonable number of regulatory alternatives and adopt the least
costly, most cost-effective or least burdensome alternative that
achieves the objectives of the rule.  The provisions of section 205 do
not apply when they are inconsistent with applicable law.  Moreover,
section 205 allows EPA to adopt an alternative other than the least
costly, most cost-effective or least burdensome alternative if the
Administrator publishes with the final rule an explanation why that
alternative was not adopted.  Before EPA establishes any regulatory
requirements that may significantly or uniquely affect small
governments, including tribal governments, it must have developed under
section 203 of the UMRA a small government agency plan.  The plan must
provide for notifying potentially affected small governments, enabling
officials of affected small governments to have meaningful and timely
input in the development of EPA regulatory proposals with significant
Federal intergovernmental mandates, and informing, educating, and
advising small governments on compliance with the regulatory
requirements. 

We have determined that this rule would not contain a Federal mandate
that would result in expenditures of $100 million or more by State,
local, and tribal governments, in the aggregate, or the private sector
in any 1 year.  Although initially these changes are expected to result
in a small increase in the burden imposed upon reviewing authorities in
order for them to be included in the State(s SIP, these revisions would
ultimately simplify applicability determinations, eliminate the burden
of reviewing projected future emissions and distinguishing between
emissions increases caused by the change from those due solely to demand
growth, and reduce the burden associated with making compliance
determinations.  Thus, this action is not subject to the requirements of
sections 202 and 205 of the UMRA.	

For the same reasons stated above, we have determined that today(s
notice contains no regulatory requirements that might significantly or
uniquely affect small governments.  Thus, this action is not subject to
the requirements of section 203 of the UMRA.

E.  Executive Order 13132:  Federalism tc \l2 "E.  Executive Order
13132Federalism 

Executive Order 13132, entitled (Federalism( (64 FR 43255, August 10,
1999), requires EPA to develop an accountable process to ensure
(meaningful and timely input by State and local officials in the
development of regulatory policies that have federalism implications.( 
(Policies that have federalism implications( is defined in the Executive
Order to include regulations that have (substantial direct effects on
the States, on the relationship between the national government and the
States, or on the distribution of power and responsibilities among the
various levels of government.(  

This proposed rule does not have federalism implications.  It will not
have substantial direct effects on the States, on the relationship
between the national government and the States, or on the distribution
of power and responsibilities among the various levels of government, as
specified in Executive Order 13132.  We estimate a one-time burden of
approximately 2,240 hours and $83,000 for State agencies to revise their
SIPs to include the proposed regulations.  However, these revisions
would ultimately simplify applicability determinations, eliminate the
burden of reviewing projected future emissions and distinguishing
between emissions increases caused by the change from those due solely
to demand growth, and reduce the burden associated with making
compliance determinations.  This will in turn reduce the overall burden
of the program.  Thus, Executive Order 13132 does not apply to this
rule. 

In the spirit of Executive Order 13132, and consistent with EPA policy
to promote communications between EPA and State and local governments,
EPA specifically solicits comment on this proposed rule from State and
local officials. 

F.  Executive Order 13175:  Consultation and Coordination with Indian
Tribal Governments tc \l2 "F.  Executive Order 13175Consultation and
Coordination with Indian Tribal Governments 

Executive Order 13175, entitled (Consultation and Coordination with
Indian Tribal Governments( (65 FR 67249, November 9, 2000), requires EPA
to develop an accountable process to ensure (meaningful and timely input
by tribal officials in the development of regulatory policies that have
tribal implications.(  This proposed rule does not have tribal
implications, as specified in Executive Order 13175.  There are no
Tribal authorities currently issuing major NSR permits.  To the extent
that today(s proposed rule may apply in the future to any EGU that may
locate on tribal lands, tribal officials are afforded the opportunity to
comment on tribal implications in today(s notice.  Thus, Executive Order
13175 does not apply to this rule.  

Although Executive Order 13175 does not apply to this proposed rule, EPA
specifically solicits comment on this proposed rule from tribal
officials.  We will also consult with tribal officials, including
officials of the Navaho Nation lands on which Navajo Power Plant and
Four Corners Generating Plant are located, before promulgating the final
regulations.  In the spirit of Executive Order 13132, and consistent
with EPA policy to promote communications between EPA and State and
local government, EPA specifically solicits comment on this proposed
rule from State and local governments.

G.  Executive Order 13045:  Protection of Children from Environmental
Health Risks and Safety Risks tc \l2 "G.  Executive Order
13045Protection of Children from Environmental Health Risks and Safety
Risks 

Executive Order 13045: (Protection of Children from Environmental health
Risks and Safety Risks( (62 FR 19885, April 23, 1997) applies to any
rule that: (1) is determined to be (economically significant( as defined
under Executive Order 12866, and (2) concerns an environmental health or
safety risk that EPA has reason to believe may have a disproportionate
effect on children.  If the regulatory action meets both criteria, the
Agency must evaluate the environmental health or safety effects of the
planned rule on children, and explain why the planned regulation is
preferable to other potentially effective and reasonably feasible
alternatives considered by the Agency.

The EPA interprets Executive Order 13045 as applying only to those
regulatory actions that are based on health or safety risks, such that
the analysis required under section 5-501 of the Order has the potential
to influence the regulation.  This rule is not subject to Executive
Order 13045, because we do not have reason to believe the environmental
health or safety risks addressed by this action present a
disproportionate risk to children.  We believe that, based on our
analysis of electric utilities, this rule as a whole will result in
equal environmental protection to that currently provided by the
existing regulations, and do so in a more streamlined and effective
manner. 

H.  Executive Order 13211:  Actions Concerning Regulations That
Significantly Affect Energy Supply, Distribution, or Use tc \l2 "H. 
Executive Order 13211Actions Concerning Regulations That Significantly
Affect Energy Supply, Distribution, or Use	 

This rule is not a (significant energy action( as defined in Executive
Order 13211, (Actions Concerning Regulations That Significantly Affect
Energy Supply, Distribution, or Use( [66 FR 28355 (May 22, 2001)]
because it is not likely to have a significant adverse effect on the
supply, distribution, or use of energy.  In fact, this rule improves
owner/operator flexibility concerning the supply, distribution, and use
of energy.  Specifically, the proposed rule would increase
owner/operators( ability to utilize existing capacity at EGUs.

I.  National Technology Transfer and Advancement Act

Section 12(d) of the National Technology Transfer and Advancement Act of
1995 

 ((NTTAA(), Public Law No. 104-113, 12(d) (15 U.S.C. 272 note) directs
EPA to use voluntary consensus standards in its regulatory activities
unless to do so would be inconsistent with applicable law or otherwise
impractical.  Voluntary consensus standards are technical standards (for
example, materials specifications, test methods, sampling procedures,
and business practices) that are developed or adopted by voluntary
consensus standards bodies.  The NTTAA directs EPA to provide Congress,
through OMB, explanations when the Agency decides not to use available
and applicable voluntary consensus standards. 

Today(s proposed rule does not involve technical standards.  Therefore,
EPA is not considering the use of any voluntary consensus standards.

X.  Statutory Authority

	The statutory authority for this action is provided by sections 307(d)
(7) (B), 101, 111, 114, 116, and 301 of the CAA as amended (42 U.S.C.
7401, 7411, 7414, 7416, and 7601).  This notice is also subject to
section 307(d) of the CAA (42 U.S.C. 7407(d)).

List of Subjects 

40 CFR Part 51

	Environmental protection, Administrative practice and 

procedure, Air pollution control, Electric Generating Unit, Nitrogen
oxides, Sulfur dioxide

40 CFR Part 52

	Environmental protection, Administrative practice and 

procedure, Air pollution control, Electric Generating Unit, Nitrogen
oxides, Sulfur dioxide.

____________________

Dated:

_____________________

Stephen L. Johnson,

 Administrator.  SEQ CHAPTER \h \r 1 For the reasons set out in the
preamble, title 40, chapter I of the Code of Federal Regulations is
proposed to be amended as follows:

PART 51 - [Amended]

	1.  The authority citation for part 51 continues to read as follows:

	Authority: 23 U.S.C. 101; 42 U.S.C. 7401 - 7671q.

Subpart I - [Amended]

	2.  Add §51.167 to read as follows:

§51.167  Preliminary major NSR applicability test for electric
generating units (EGUs)  TC "§51.167  Major modification procedures for
electric generating units (EGUs)" \f C \l "1"  .

(a)  What is the purpose of this section?  TC "(a)  What is the purpose
of this section?" \f C \l "2"    State Implementation Plans and Tribal
Implementation Plans must include the requirements in paragraphs (b)
through (h) of this section for determining (prior to or after
construction) whether a change to an EGU is a modification for purposes
of major NSR applicability.  Deviations from these provisions will be
approved only if the State or Tribe demonstrates that the submitted
provisions are at least as stringent in all respects as the
corresponding provisions in paragraphs (b) through (h) of this section.

	(b)  Am I subject to this section?  TC "(b)  Am I subject to this
section?" \f C \l "2"    You must meet the requirements of this section
if you own or operate an EGU that is located at a major stationary
source, and you plan to make a change to the EGU.

	(c)  What happens if a change to my EGU is determined to be a
modification according to the procedures of this section?    TC "(c) 
What happens if a change to my EGU is determined to be a major
modification according to the procedures of this section?" \f C \l "2" 
If the change to your EGU is a modification according to the procedures
of this section, you must determine whether the change is a major
modification according to the procedures of the major NSR program that
applies in the area in which your EGU is located.  That is, you must
evaluate your modification according to the requirements set out in the
applicable regulations approved pursuant to §51.165 and/or §51.166,
depending on the regulated NSR pollutants emitted and the attainment
status of the area in which your EGU is located for those pollutants. 
Section 51.165 sets out the requirements for State nonattainment major
NSR programs, while §51.166 sets out the requirements for State PSD
programs.  

	(d)  What is the process for determining if a change to an EGU is a
modification?  TC "(d)  What is the process for determining if a change
to an EGU constitutes a major modification?" \f C \l "2"    The two-step
process set out in paragraphs (d)(1) and (2) of this section is used to
determine (before beginning actual construction) whether a change to an
EGU located at a major stationary source is a modification.  Regardless
of any preconstruction projections, a modification has occurred if a
change satisfies both steps in the process.

	(1)  Step 1.  Is the change a physical change in, or change in the
method of operation of, the EGU?  (See paragraph (e) of this section for
a list of actions that are not physical or operational changes.)  If so,
go on to Step 2 (paragraph (d)(2) of this section).

	(2)  Step 2.  Will the physical or operational change to the EGU
increase the amount of any regulated NSR pollutant emitted into the
atmosphere by the source (as determined according to paragraph (f) of
this section) or result in the emissions of any regulated NSR
pollutant(s) into the atmosphere that the source did not previously
emit?  If so, the change is a modification.

	(e)  What types of actions are not physical changes or changes in the
method of operation?  (Step1)  TC "(e)  What types of physical or
operational changes are exempted from consideration under this section? 
(Step1)" \f C \l "2"    For purposes of this section, a physical change
or change in the method of operation shall not include:

	(1)  Routine maintenance, repair, and replacement;

	(2)  Use of an alternative fuel or raw material by reason of an order
under sections 2(a) and (b) of the Energy Supply and Environmental
Coordination Act of 1974 (or any superseding legislation) or by reason
of a natural gas curtailment plan pursuant to the Federal Power Act;

	(3)  Use of an alternative fuel by reason of an order or rule under
section 125 of the Act;

	(4)  Use of an alternative fuel at a steam generating unit to the
extent that the fuel is generated from municipal solid waste;

	(5)  Use of an alternative fuel or raw material by a stationary source
which the source is approved to use under any permit issued under 40 CFR
52.21 or under regulations approved pursuant to §51.165 or §51.166, or
which:

	(i)  For purposes of evaluating attainment pollutants, the source was
capable of accommodating before January 6, 1975, unless such change
would be prohibited under any federally enforceable permit condition
which was established after January 6, 1975 pursuant to 40 CFR 52.21 or
under regulations approved pursuant to 40 CFR part 51 subpart I or
§51.166; or

	(ii)  For purposes of evaluating nonattainment pollutants, the source
was capable of accommodating before December 21, 1976, unless such
change would be prohibited under any federally enforceable permit
condition which was established after December 21, 1976 pursuant to 40
CFR 52.21 or under regulations approved pursuant to 40 CFR part 51
subpart I or §51.166;

	(6)  An increase in the hours of operation or in the production rate,
unless such change is prohibited under any federally enforceable permit
condition which was established after January 6, 1975 (for purposes of
evaluating attainment pollutants) or after December 21, 1976 (for
purposes of evaluating nonattainment pollutants) pursuant to 40 CFR
52.21 or regulations approved pursuant to 40 CFR part 51 subpart I or
§51.166;

	(7)  Any change in ownership at a stationary source;

	(8)  The installation, operation, cessation, or removal of a temporary
clean coal technology demonstration project, provided that the project
complies with:

	(i)  The State Implementation Plan for the State in which the project
is located; and

	(ii)  Other requirements necessary to attain and maintain the national
ambient air quality standard during the project and after it is
terminated;

	(9)  For purposes of evaluating attainment pollutants, the installation
or operation of a permanent clean coal technology demonstration project
that constitutes repowering, provided that the project does not result
in an increase in the potential to emit of any regulated pollutant
emitted by the unit.  This exemption shall apply on a
pollutant-by-pollutant basis; or

	(10)  For purposes of evaluating attainment pollutants, the
reactivation of a very clean coal-fired EGU.

	(f)  How do I determine if there is an emissions increase?  (Step 2) 
TC "(f)  How do I determine if my physical or operational change will
increase the amount of a regulated NSR pollutant emitted into the
atmosphere by my source?  (Step2)" \f C \l "2"    You must determine if
the physical or operational change to your EGU increases the amount of
any regulated NSR pollutant emitted to the atmosphere using the method
in paragraph (f)(1) of this section, subject to the limitations in
paragraph (f)(2) of this section.  If the physical or operational change
to your EGU increases the amount of any regulated NSR pollutant emitted
into the atmosphere or results in the emission of any regulated NSR
pollutant(s) into the atmosphere that your EGU did not previously emit,
the change is a modification as defined in paragraph (h)(2) of this
section.

Alternative 1 for paragraph (f)(1):

	(1)  Emissions increase test.  For each regulated NSR pollutant for
which you have hourly average CEMS or PEMS emissions data with
corresponding fuel heat input data, compare the pre-change maximum
actual hourly emissions rate in pounds per hour (lb/hr) to a projection
of the post-change maximum actual hourly emissions rate in lb/hr,
subject to the provisions in paragraphs (f)(1)(i) through (iii) of this
section.

	(i)  Pre-change emissions.  Determine the pre-change maximum actual
hourly emissions rate as follows:

	(A)  Select a period of 365 consecutive days within the 5-year period
immediately preceding when you begin actual construction of the physical
or operational change.  Compile a data set (for example, in a
spreadsheet) with the hourly average CEMS or PEMS (as applicable)
measured emissions rates and corresponding heat input data for all of
the hours of operation for that 365-day period for the pollutant of
interest.

	(B)  Delete any unacceptable hourly data from this 365-day period in
accordance with the data limitations in paragraph (f)(2) of this
section.

	(C)  Extract the hourly data for the 10 percent of the remaining data
set corresponding to the highest heat input rates for the selected
period.  This step may be facilitated by sorting the data set for the
remaining operating hours from the lowest to the highest heat input
rates.  

	(D)  Calculate the average emissions rate from the extracted (i.e.,
highest 10 percent heat input rates) data set, using Equation 1:  

 					Equation 1

Where:

 = average emissions rate, lb/hr;

	n = number of emissions rate values; and

  = ith emissions rate value, lb/hr

	(E)  Calculate the standard deviation of the data set, s, using
Equation 2:

 			Equation 2

	(F)  Calculate the Upper Tolerance Limit, UTL, of the data set using
Equation 3:

 Equation 3

Where:

	Z1-p = 3.090, Z score for the 99.9 percentage of interval; and

	Z1-q =  2.326, Z score for the 99 percent confidence level. 

	(G)  Use the UTL calculated in paragraph (f)(1)(i)(F) of this section
as the pre-change maximum actual hourly emissions rate.

	(ii)  Post-change emissions—preconstruction projections.  For each
regulated NSR pollutant, you must project the maximum emissions rate
that your EGU will actually achieve in any 1 hour in the 5 years
following the date the EGU resumes regular operation after the physical
or operational change.  An emissions increase results from the physical
or operational change if this projected maximum actual hourly emissions
rate exceeds the pre-change maximum actual hourly emissions rate.

	(iii)  Post-change emissions—actually achieved.  Regardless of any
preconstruction projections, an emissions increase has occurred if the
hourly emissions rate actually achieved in the 5 years after the change
exceeds the pre-change maximum actual hourly emissions rate.

Alternative 2 for paragraph (f)(1):

	(1)  Emissions increase test.  For each regulated NSR pollutant for
which you have hourly average CEMS or PEMS emissions data with
corresponding fuel heat input data, compare the pre-change maximum
actual emissions rate in pounds per megawatt-hour (lb/MWh) to a
projection of the post-change maximum actual emissions rate in lb/MWh,
subject to the provisions in paragraphs (f)(1)(i) through (iii) of this
section.  For EGUs that are cogeneration units, emissions rates are
determined based on gross energy output.  For other EGUs, emissions
rates are determined based on gross electrical output.

	(i)  Pre-change emissions.  Determine the pre-change maximum actual
emissions rate as follows:

	(A)  Select a period of 365 consecutive days within the 5-year period
immediately preceding when you begin actual construction of the physical
or operational change.  Compile a data set (for example, in a
spreadsheet) with the hourly average CEMS or PEMS (as applicable)
measured emissions rates in lb/MWh and corresponding heat input data for
all of the hours of operation for that 365-day period for the pollutant
of interest.

	(B)  Delete any unacceptable hourly data from this 365-day period in
accordance with the data limitations in paragraph (f)(2) of this
section.

	(C)  Extract the hourly data for the 10 percent of the remaining data
set corresponding to the highest heat input rates for the selected
period.  This step may be facilitated by sorting the data set for the
remaining operating hours from the lowest to the highest heat input
rates.  

	(D)  Calculate the average emissions rate from the extracted (i.e.,
highest 10 percent heat input rates) data set, using Equation 1:  

 					Equation 1

Where:

 = average emissions rate, lb/MWh;

	n = number of emissions rate values; and

  = ith emissions rate value, lb/MWh

	(E)  Calculate the standard deviation of the data set, s, using
Equation 2:

 			Equation 2

	(F)  Calculate the Upper Tolerance Limit, UTL, of the data set using
Equation 3:

 Equation 3

Where:

	Z1-p = 3.090, Z score for the 99.9 percentage of interval; and

	Z1-q =  2.326, Z score for the 99 percent confidence level. 

	(G)  Use the UTL calculated in paragraph (f)(1)(i)(F) of this section
as the pre-change maximum actual hourly emissions rate.

	(ii)  Post-change emissions—preconstruction projections.  For each
regulated NSR pollutant, you must project the maximum emissions rate
that your EGU will actually achieve over any period of 1 hour in the 5
years following the date the EGU resumes regular operation after the
physical or operational change.  An emissions increase results from the
physical or operational change if this projected maximum actual
emissions rate exceeds the pre-change maximum actual emissions rate.

	(iii)  Post-change emissions—actually achieved.  Regardless of any
preconstruction projections, an emissions increase has occurred if the
emissions rate actually achieved over any period of 1 hour in the 5
years after the change exceeds the pre-change maximum actual emissions
rate.

Alternative 3 for paragraph (f)(1):

	(1)  Emissions increase test.  For each regulated NSR pollutant,
compare the pre-change maximum actual hourly emissions rate in pounds
per hour (lb/hr) to a projection of the post-change maximum actual
hourly emissions rate in lb/hr, subject to the provisions in paragraphs
(f)(1)(i) through (iv) of this section.

	(i)  Pre-change emissions—general procedures.  The pre-change maximum
actual hourly emissions rate is the highest rate at which the EGU
actually emitted the pollutant at any time during the 5-year period
immediately prior to the physical or operational change, determined as
follows:

	(A)  Select a period of 24 consecutive months within the 5-year period
immediately preceding when you begin actual construction of the physical
or operational change;

	(B)  Determine the highest emissions rate (lb/hr) actually achieved for
1 hour in the first 12 months and for 1 hour in the second 12 months of
the selected 24-month period, where the two 1-hour periods also fall in
different calendar years; and

	(C)  Calculate the arithmetic average of these two values. 

	(ii)  Pre-change emissions—data sources.  You must determine the
highest pre-change hourly emissions rate for each regulated NSR
pollutant using the best data available to you.  Use the highest
available source of data in the following hierarchy, unless your
reviewing authority has determined that a data source lower in the
hierarchy will provide better data for your EGU:

	(A)  Continuous emissions monitoring system (CEMS).

	(B)  Approved predictive emissions monitoring system (PEMS).

	(C)  Emission tests/emission factor specific to the EGU to be changed.

	(D)  Material balance calculations.

	(E)  Published emission factor.

	(iii)  Post-change emissions—preconstruction projections.  For each
regulated NSR pollutant, you must project the maximum emissions rate
that your EGU will actually achieve in any 1 hour in the 5 years
following the date the EGU resumes regular operation after the physical
or operational change.  An emissions increase results from the physical
or operational change if this projected maximum actual hourly emissions
rate exceeds the pre-change maximum actual hourly emissions rate.

	(iv)  Post-change emissions—actually achieved.  Regardless of any
preconstruction projections, an emissions increase has occurred if the
hourly emissions rate actually achieved in the 5 years after the change
exceeds the pre-change maximum actual hourly emissions rate.

Alternative 4 for paragraph (f)(1):

	(1)  Emissions increase test.  For each regulated NSR pollutant,
compare the pre-change maximum actual emissions rate in pounds per
megawatt-hour (lb/MWh) to a projection of the post-change maximum actual
emissions rate in lb/MWh, subject to the provisions in paragraphs
(f)(1)(i) through (iv) of this section.  For EGUs that are cogeneration
units, emissions rates are determined based on gross energy output.  For
other EGUs, emissions rates are determined based on gross electrical
output.

	(i)  Pre-change emissions—general procedures.  The pre-change maximum
actual emissions rate is the highest rate at which the EGU actually
emitted the pollutant at any time during the 5-year period immediately
prior to the physical or operational change, determined as follows:

	(A)  Select a period of 24 consecutive months within the 5-year period
immediately preceding when you begin actual construction of the physical
or operational change;

	(B)  Determine the highest emissions rate (lb/MWh) actually achieved
over a period of 1 hour in the first 12 months and over a period of 1
hour in the second 12 months of the selected 24-month period, where the
two 1-hour periods also fall in different calendar years; and

	(C)  Calculate the arithmetic average of these two values. 

	(ii)  Pre-change emissions—data sources.  You must determine the
highest pre-change emissions rate for each regulated NSR pollutant using
the best data available to you.  Use the highest available source of
data in the following hierarchy, unless your reviewing authority has
determined that a data source lower in the hierarchy will provide better
data for your EGU:

	(A)  Continuous emissions monitoring system (CEMS).

	(B)  Approved predictive emissions monitoring system (PEMS).

	(C)  Emission tests/emission factor specific to the EGU to be changed.

	(D)  Material balance calculations.

	(E)  Published emission factor.

	(iii)  Post-change emissions—preconstruction projections.  For each
regulated NSR pollutant, you must project the maximum emissions rate
that your EGU will actually achieve over any period of 1 hour in the 5
years following the date the EGU resumes regular operation after the
physical or operational change.  An emissions increase results from the
physical or operational change if this projected maximum actual
emissions rate exceeds the pre-change maximum actual emissions rate.

	(iv)  Post-change emissions—actually achieved.  Regardless of any
preconstruction projections, an emissions increase has occurred if the
emissions rate actually achieved over any period of 1 hour in the 5
years after the change exceeds the pre-change maximum actual emissions
rate.

Alternative 5 for paragraph (f)(1):

	(1)  Emissions increase test.  For each regulated NSR pollutant,
compare the pre-change maximum actual hourly emissions rate in pounds
per hour (lb/hr) to a projection of the post-change maximum actual
hourly emissions rate in lb/hr, subject to the provisions in paragraphs
(f)(1)(i) through (iv) of this section.

	(i)  Pre-change emissions—general procedures.  The pre-change maximum
actual hourly emissions rate for the pollutant is the highest emissions
rate (lb/hr) actually achieved by the EGU for 1 hour at any time during
the 5-year period immediately preceding when you begin actual
construction of the physical or operational change.

	(ii)  Pre-change emissions—data sources.  You must determine the
highest pre-change hourly emissions rate for each regulated NSR
pollutant using the best data available to you.  Use the highest
available source of data in the following hierarchy, unless your
reviewing authority has determined that a data source lower in the
hierarchy will provide better data for your EGU:

	(A)  Continuous emissions monitoring system (CEMS).

	(B)  Approved predictive emissions monitoring system (PEMS).

	(C)  Emission tests/emission factor specific to the EGU to be changed.

	(D)  Material balance calculations.

	(E)  Published emission factor.

	(iii)  Post-change emissions—preconstruction projections.  For each
regulated NSR pollutant, you must project the maximum emissions rate
that your EGU will actually achieve in any 1 hour in the 5 years
following the date the EGU resumes regular operation after the physical
or operational change.  An emissions increase results from the physical
or operational change if this projected maximum actual hourly emissions
rate exceeds the pre-change maximum actual hourly emissions rate.

	(iv)  Post-change emissions—actually achieved.  Regardless of any
preconstruction projections, an emissions increase has occurred if the
hourly emissions rate actually achieved in the 5 years after the change
exceeds the pre-change maximum actual hourly emissions rate.

Alternative 6 for paragraph (f)(1):

	(1)  Emissions increase test.  For each regulated NSR pollutant,
compare the pre-change maximum actual emissions rate in pounds per
megawatt-hour (lb/MWh) to a projection of the post-change maximum actual
emissions rate in lb/MWh, subject to the provisions in paragraphs
(f)(1)(i) through (iv) of this section.  For EGUs that are cogeneration
units, emissions rates are determined based on gross energy output.  For
other EGUs, emissions rates are determined based on gross electrical
output.

	(i)  Pre-change emissions—general procedures.  The pre-change maximum
actual emissions rate for the pollutant is the highest emissions rate
(lb/MWh) actually achieved by the EGU over any period of 1 hour during
the 5-year period immediately preceding when you begin actual
construction of the physical or operational change.

	(ii)  Pre-change emissions—data sources.  You must determine the
highest pre-change emissions rate for each regulated NSR pollutant using
the best data available to you.  Use the highest available source of
data in the following hierarchy, unless your reviewing authority has
determined that a data source lower in the hierarchy will provide better
data for your EGU:

	(A)  Continuous emissions monitoring system (CEMS).

	(B)  Approved predictive emissions monitoring system (PEMS).

	(C)  Emission tests/emission factor specific to the EGU to be changed.

	(D)  Material balance calculations.

	(E)  Published emission factor.

	(iii)  Post-change emissions—preconstruction projections.  For each
regulated NSR pollutant, you must project the maximum emissions rate
that your EGU will actually achieve over any period of 1 hour in the 5
years following the date the EGU resumes regular operation after the
physical or operational change.  An emissions increase results from the
physical or operational change if this projected maximum actual
emissions rate exceeds the pre-change maximum actual emissions rate.

	(iv)  Post-change emissions—actually achieved.  Regardless of any
preconstruction projections, an emissions increase has occurred if the
emissions rate actually achieved over any period of 1 hour in the 5
years after the change exceeds the pre-change maximum actual emissions
rate.

Alternative 7 for paragraph (f)(1):

	(1)  Emissions increase test.  For each regulated NSR pollutant,
compare the maximum achievable hourly emissions rate before the physical
or operational change to the maximum achievable hourly emissions rate
after the change.  Determine these maximum achievable hourly emissions
rates according to §60.14(b) of this chapter.  No physical change, or
change in the method of operation, at an existing EGU shall be treated
as a modification for the purposes of this section provided that such
change does not increase the maximum hourly emissions of any regulated
NSR pollutant above the maximum hourly emissions achievable at that unit
during the 5 years prior to the change.

Alternative 8 for paragraph (f)(1):

	(1)  Emissions increase test.  For each regulated NSR pollutant,
compare the maximum achievable emissions rate in pounds per
megawatt-hour (lb/MWh) before the physical or operational change to the
maximum achievable emissions rate in lb/MWh after the change.  Determine
these maximum achievable emissions rates according to §60.14(b) of this
chapter, using emissions rates in lb/MWh achievable over 1 hour of
continuous operation in place of mass emissions rates.  For EGUs that
are cogeneration units, determine emissions rates based on gross energy
output.  For other EGUs, determine emissions rates based on gross
electrical output.  No physical change, or change in the method of
operation, at an existing EGU shall be treated as a modification for the
purposes of this section provided that such change does not increase the
maximum emissions rate of any regulated NSR pollutant above the maximum
emissions rate achievable at that unit during the 5 years prior to the
change.

	(2)  Data limitations for maximum emissions rates.  For purposes of
determining pre-change and post-change maximum emissions rates under
paragraph (f)(1) of this section, the following limitations apply to the
types of data that you may use:

	(i)  Data limitations for Alternatives 1 – 6 .

	(A)  You must not use emissions rate data associated with startups,
shutdowns, or malfunctions of your EGU, as defined by applicable
regulation(s) or permit term(s), or malfunctions of an associated air
pollution control device.  A malfunction means any sudden, infrequent,
and not reasonably preventable failure of the EGU or the air pollution
control equipment to operate in a normal or usual manner.

	(B)  You must not use continuous emissions monitoring system (CEMS) or
predictive emissions monitoring system (PEMS) data recorded during
monitoring system out-of-control periods.  Out-of-control periods
include those during which the monitoring system fails to meet quality
assurance criteria (for example, periods of system breakdown, repair,
calibration checks, or zero and span adjustments) established by
regulation, by permit, or in an approved quality assurance plan.

	(C)  You must not use emissions rate data from periods of noncompliance
when your EGU was operating above an emission limitation that was
legally enforceable at the time the data were collected.

	(D)  You must not use data from any period for which the information is
inadequate for determining emissions rates, including information
related to the limitations in paragraphs (f)(2)(i)(A) through (C) of
this section.

	(ii)  Data limitations for Alternatives 7 and 8 .

	(A)  You must not use emissions rate data associated with startups,
shutdowns, or malfunctions of your EGU, as defined by applicable
regulation(s) or permit term(s), or malfunctions of an associated air
pollution control device.  A malfunction means any sudden, infrequent,
and not reasonably preventable failure of the EGU or the air pollution
control equipment to operate in a normal or usual manner.

	(B)  You must not use continuous emissions monitoring system (CEMS) or
predictive emissions monitoring system (PEMS) data recorded during
monitoring system out-of-control periods.  Out-of-control periods
include those during which the monitoring system fails to meet quality
assurance criteria (for example, periods of system breakdown, repair,
calibration checks, or zero and span adjustments) established by
regulation, by permit, or in an approved quality assurance plan.

	(C)  You must not use data from any period for which the information is
inadequate for determining emissions rates, including information
related to the limitations in paragraphs (f)(2)(ii)(A) and (B) of this
section.

	(g)  What are my requirements for notifications and recordkeeping?  You
must submit the notifications described in paragraph (g)(1) of this
section and keep the records described in paragraph (g)(2) of this
section.

	(1)  Notifications.  You must send a notification to your reviewing
authority for any physical or operational change to an existing EGU
which may increase the emissions rate of any regulated NSR pollutant
(that is, may trigger Step 2 as set out in paragraphs (d)(2) and (f) of
this section).  The following provisions apply to these notifications:

	(i)  Notifications requirements for Alternatives 1 – 6 for paragraph
(f)(1) of this section.

	(A)  The notification must be postmarked no later than 6 months before
the change is commenced.

	(B)  The notification must include information describing:

	(1)  The precise nature of the change;

	(2)  The present and proposed emission control systems;

	(3)  The productive capacity of the EGU before and after the change;
and 

	(4)  The expected completion date of the change. 

	(C)  The reviewing authority may request additional relevant
information subsequent to this notification.

	(ii)  Notification requirements for Alternatives 7 and 8 for paragraph
(f)(1) of this section.

	(A)  The notification must be postmarked 60 days or as soon as
practicable before the change is commenced.

	(B)  The notification must include information describing:

	(1)  The precise nature of the change;

	(2)  The present and proposed emission control systems;

	(3)  The productive capacity of the EGU before and after the change;
and 

	(4)  The expected completion date of the change. 

	(C)  The reviewing authority may request additional relevant
information subsequent to this notification.

	(2)  Recordkeeping.  You must maintain a file of all information
related to determinations that you make under this section of whether a
change to an EGU is a modification, subject to the following provisions:

	(i)  The file must include, but is not limited to, the following
information recorded in permanent form suitable for inspection:

	(A)  Continuous monitoring system, monitoring device, and performance
testing measurements; 

	(B)  All continuous monitoring system performance evaluations; 

	(C)  All continuous monitoring system or monitoring device calibration
checks;

	(D)  All adjustments and maintenance performed on these systems or
devices; and

	(E)  All other information relevant to any determination made under
this section of whether a change to an EGU is a modification. 

	(ii)  You must retain the file until the later of:

	(A)  The date 5 years following the date the EGU resumes regular
operation after the physical or operational change; and

	(B)  The date 5 years following the date of such measurements,
maintenance, reports, and records.  

	(h)  What definitions apply under this section?  TC "(j)  What
definitions apply under this section?" \f C \l "2"    The definitions in
paragraphs (h)(1) and (2) of this section apply.  Except as specifically
provided in this paragraph (h), terms used in this section have the
meaning accorded them under §51.165(a)(1) or §51.166(b), as
appropriate to the situation (for example, the attainment status of the
area where your source is located for a particular regulated NSR
pollutant of interest).  Terms not defined here or in §51.165(a)(1) or
§51.166(b) (as appropriate) have the meaning accorded them under the
applicable requirements of the Clean Air Act, 42 U.S.C. 7401, et seq.

	(1)  Terms related to EGUs that are defined in §51.124(q).  The
following terms are as defined in §51.124(q):  

	Boiler.

	Bottoming-cycle cogeneration unit.

	Cogeneration unit.

	Combustion turbine.

	Electric generating unit or EGU.

	Fossil fuel.

	Fossil-fuel-fired.

	Generator.

	Maximum design heat input.

	Nameplate capacity.

	Potential electrical output capacity.

	Sequential use of energy.

	Topping-cycle cogeneration unit.

	Total energy input.

	Total energy output.

	Useful power.

	Useful thermal energy.

	Utility power distribution system. 

	(2)  Other terms defined for the purposes of this section.

	Attainment pollutant means a regulated NSR pollutant for which your EGU
may be subject to the PSD program that is applicable in the area where
your EGU is located.  In general, attainment pollutants are the
regulated NSR pollutants listed in the PSD program for which there is no
NAAQS or for which the area in which your EGU is located is designated
as attainment or unclassifiable according to part 81 of this chapter. 
However, pollutant or precursor transport considerations may cause such
regulated NSR pollutants to be treated as nonattainment pollutants as
defined in this paragraph (h)(2) (for example, if your EGU is located in
an ozone transport region).

	Gross electrical output means the electricity made available for use by
the generator associated with the EGU.  

	Gross energy output means, with regard to a cogeneration unit, the sum
of the gross power output and the useful thermal energy output produced
by the cogeneration unit.  

	Gross power output means, with regard to a cogeneration unit,
electricity or mechanical energy made available for use by the
cogeneration unit.  

	Modification, for an EGU, means any physical change in, or change in
the method of operation of, an EGU which increases the amount of any
regulated NSR pollutant emitted into the atmosphere by that source or
which results in the emission of any regulated NSR pollutant(s) into the
atmosphere that the source did not previously emit.  An increase in the
amount of regulated NSR pollutants must be determined according to the
provisions in paragraph (f) of this section.  For purposes of this
section, a physical change or change in the method of  operation shall
not include the types of actions listed in paragraph (e) of this
section.

	Nonattainment pollutant means a regulated NSR pollutant for which your
EGU may be subject to the nonattainment major NSR program that is
applicable in the area where your EGU is located.  In general,
nonattainment pollutants are the regulated NSR pollutants listed in the
nonattainment major NSR program for which the area in which your EGU is
located is designated as nonattainment according to part 81 of this
chapter.  However, pollutant or precursor transport considerations may
cause such regulated NSR pollutants to be treated as attainment
pollutants as defined in this paragraph (h)(2).

	Useful thermal energy output means, with regard to a cogeneration unit,
the thermal energy made available for use in any industrial or
commercial process, or used in any heating or cooling application, that
is, total thermal energy made available for processes and applications
other than electrical or mechanical generation.  Thermal output for this
section means the energy in recovered thermal output measured against
the energy in the thermal output at 15 degrees Celsius and 101.325
kilopascals of pressure.

 Establishments owned and operated by Federal, State, or local
government are classified according to the activity in which they are
engaged.

 For clarity, this table lists all of the steps in the applicability
determinations under the various options and alternatives.  These steps
include, as Step 1, the determination of whether a physical change or
change in the method of operation has occurred.  This Step 1 is included
in the table solely for purposes of clarity; neither the October 2005
NPR nor this action proposes any action of any type (or makes any
re-proposal) concerning the regulations defining physical change or
change in the method of operation.  Similarly, the steps also include,
as Steps 3 and 4, the current net significance test; and today’s SNPR
does not propose any action of any type (or make any re-proposal)
concerning the current net significance test.  Finally, this action does
not propose any action of any type (or make any re-proposal) concerning
the current applicability test for EGUs, except, under Option 3, to
revise the baseline from a 5-year period to 10 years. 

 In this context, we use the term “input” as a convenient way to
refer to the hourly emission rate test, and to distinguish it from the
output test, which is calculated on the basis of hourly emissions per
kilowatt hour of generation.

 We are also proposing note that the 10-year baseline approach would
apply to EGUs, rather than to EUSGUs.  The differences in the definition
of EUSGUs and EGUs are discussed in detail in Section VII of this
preamble.

 See the NEEDS 2004 documentation for IPM v.2.1.9 is in Exhibit 4-6,
which can be found at
http://www.epa.gov/airmarkets/epa-ipm/section4genres.pdf.

 See our report, “Contributions of CAIR/CAMR/CAVR to NAAQS Attainment:
Focus on Control Technologies and Emission Reductions in the Electric
Power Sector,” on pages 39 and 43.  The report is available at
http://www.epa.gov/air/interstateairquality/charts.html.

 While we believe it is most likely that an EGU would increase its hours
of operation under today’s proposed regulations due to reducing the
number of hours that the EGU is unavailable due to forced outage hours,
the analysis is applicable to increases in hours of operation for other
reasons.

 See our report, “Contributions of CAIR/CAMR/CAVR to NAAQS Attainment:
Focus on Control Technologies and Emission Reductions in the Electric
Power Sector,” on pages 39 and 43.  The report is available at  
HYPERLINK "http://www.epa.gov/air/interstateairquality/charts.html" 
http://www.epa.gov/air/interstateairquality/charts.html .

 Footnote as to docket item.

 See Technical Support Document for the Prevention of Significant
Deterioration and Nonattainment Area New Source Review Regulations, pg.
I-2-1, available at http://www.epa.gov/nsr/actions.html.

 “Business Cycles in Major Emitting Source Industries.”  Eastern
Research Group; September 25, 1997. 

 U.S. EPA, Regulatory Impact Analysis for the CAIR at p. 7-5.  See item
0022 in OAR-2005-0163.

 Over the past 2 years, energy policy makers and regulators in many
areas have realized that much of their electricity infrastructure is
aging, that their infrastructure has become significantly dependent on
natural gas-fired generation, and that natural gas prices are high and
likely to remain so for the next several years—or longer.  This
awareness has triggered intense interest in relying more on other
generation technologies to meet growing electricity needs.  Most of
these technologies, however, involve the development of new generation
distant from load centers, have larger associated transmission
requirements, and typically involve 5- to 10-year lead-times.  As a
result, multi-state, regional-scale generation and transmission planning
has emerged as an important energy policy priority.  See National
Association of Regulatory Utility Commissioners at
http://www.naruc.org/displayindustryarticle.cfm?articlenbr=28812.

 See Business Cycles in Major Emission Sources Industries,” Eastern
Research Group, September 30, 1997.  This report is available as docket
item

 See Regulatory Analysis for the Final Clean Air Interstate Rule, EPA
-452-05-002, pg. 6-4, available at 

 Warkentin, Denise.  Electric Power Industry in Nontechnical Language,
PennWell, pg. 65.  

 Electricity Utilities:  Tipple School of Management, University of
Iowa, April 20, 2006.

 Cleveland, C.J., et al.  “Aggregation and the role of energy in the
economy.”  Ecological Economics, 32 (2000), pp. 301-317.

 “The NBER’s Business-Cycle Dating Procedure,” Business Cycle
Dating Committee, National Bureau of Economic Research, October 21,
2003.  

 The use of a 24-month period within the past 10 years to establish an
average annual rate is intended to adjust for unusually high short-term
peaks in utilization.  

 Supplemental Environmental Analysis at F-3.

 Mary Gibbons Natrella (1963).  “Experimental Statistics,” NBS
Handbook 91, U.S. Department of Commerce.  This work is available on the
internet at   HYPERLINK
"http://www.itl.nist.gov/div898/handbook/prc/section2/prc263.htm" 
http://www.itl.nist.gov/div898/handbook/prc/section2/prc263.htm .

 In the NSPS regulations, emissions rates are compared in terms of
kilograms per hour.  We use English units in today’s proposed
rulemaking in keeping with longstanding practice in the major NSR
program, where annual emissions are generally computed using the lb/hr
rate and hours of operation.

 See “Do Combined Cycle Gas Turbine Systems Qualify as ‘Electric
Utility Steam Generating Units’ for Purposes of Determining
Applicability of NSR, “August 6, 2001, available at
http://nlquery.epa.gov/epasearch/epasearch.

 Commenters stated that the maximum achieved test is difficult to comply
with due to fluctuations in equipment
an⁤潣瑮潲⁬敤楶散瀠牥潦浲湡散琠慨⁴牡⁥敢潹摮琠
敨挠湯牴汯漠⁦桴⁥䝅⁕睯敮⽲灯牥瑡牯‮഍ഃЍ഍ഃ
Ѝ഍湉整湲污愠摮䐠汥扩牥瑡癩⁥牄晡⵴䐠⁯潮⁴畱瑯
ⱥ挠瑩ⱥ挠灯ⱹ漠⁲楤瑳楲畢整䄍牰汩㌠‬〲㘰഍倓䝁
⁅ᔠ

¢

®

¯

A

ç

摧㐹

hÆ_

 hü

혈F됃ကﰎ쀒!将

혈F됃ကﰎ쀒!将

葠ː摧䑰ç

hï

kd1

愀Ĥ摧䏀

hl

)

kdz

kdÙ

葠ː摧烥<Ѐ'摧᷻U

h÷+

hl

摧䤁

萑ː搒Ǡ葠ː摧䣁"Ѐ'摧᷻U

hl

hl

h

j¤

 hð

葠ː摧ӌÓЀ)摧䤁

h

h

h

hå

Ѐ)摧䤁

葠ː摧⎼Ѐ)摧䤁

Ѐ)摧䤁



(

:

=

V

W

X

Y

Z

_

`

…

Š

”

 

¼

¾

Ç

2

b

d

´

º

ü

þ

摧䤁

h

萑ː搒Ǡ葠ː摧䉥è

h

h

h

h

⤆㄀$摧煦a✆㄀$摧煦a

☊଀ᝆഀ߆퀁ĂŨༀ梄ሁā㄀$葞Ũ摧煦a	ሀā㄀$摧⫙{

 hê

☊଀ᙆഀ߆ꀁąŨༀ梄ሁā㄀$葞Ũ摧煦a⤆㄀$摧煦a

฀킄༂킄㄂$葝ː葞ː摧煦a✆㄀$摧煦a⤆㄀$摧煦a

U

U

ì

ö

 hê

U

摧䤁

⤆㄀$摧煦a

h

h

摧䤁

I

hÕ

h

h 

h¼s

I

I

I

I

j

j

#

l

o

 hê

Prevention of Significant Deterioration, Nonattainment New Source
Review, and New Source Performance Standards:  Emission Increases for
Electric Generating Units – Page   PAGE  110  of   NUMPAGES  133 

 

