Visibility, Regional Haze, and the Clean Air Act: Status of Implementation

CRS Report for Congress
Visibility, Regional Haze, and the Clean Air Act:
Status of Implementation
Updated October 23, 2006
Larry Parker
Specialist in Energy Policy
Resources, Science, and Industry Division
John Blodgett
Deputy Assistant Director
Resources, Science, and Industry Division


Congressional Research Service ˜ The Library of Congress

Visibility, Regional Haze, and the Clean Air Act:
Status of Implementation
Summary
Section 169A of the Clean Air Act (CAA) sets “as a national goal the prevention
of any future, and the remedying of any existing, impairment to visibility” in
designated “class I areas” (e.g., national parks and wilderness areas). It requires 26
categories of major stationary sources of pollution — including electric generating
units (EGUs) — in existence on the date of enactment (1977), but not more than 15
years old as of that date, to install “best available retrofit technology” (BART) if the
state determines the source may reasonably be anticipated to cause or contribute to
any impairment of visibility in any class I area. A key contributor to regional haze
is very fine particles (PM2.5), to which sulfur dioxide (SO2) and nitrogen oxides
(NOx) are important contributors. EGUs are major emitters of SO2 and NOx.
The Environmental Protection Agency (EPA) was directed to issue regulations
to assure that State Implementation Plans (SIPs) required (1) reasonable progress
toward meeting the national goal and (2) compliance with specific provisions,
including the BART requirements. However, EPA delayed issuing regional haze
rules, and in 1990 Congress amended the CAA’s visibility requirements. EPA issued
the final regional haze rule on July 1, 1999. Among its provisions, the rule required
“reasonable progress” toward visibility improvement and a state BART
implementation plan. For BART, states could alternatively propose a trading
program — but only if it achieved greater progress in improving visibility.
The BART requirement’s interaction with other air pollution control programs
has become an issue — most notably its relation to the Clean Air Interstate Rule
(CAIR) designed to reduce emissions crossing state lines and hindering compliance
with National Ambient Air Quality Standards (NAAQS). CAIR involves controls
on SO2 and NOx, focuses on EGUs as the most cost-effective source to control, and
proposes using a trading mechanism to accomplish reductions. At issue is how the
model CAIR trading program for EGUs interacts with the BART requirement for
EGUs. In 2005, EPA made a final determination to exempt EGUs subject to the
CAIR trading program from the Section 169A visibility BART program. Critics of
EPA’s proposal point out that Section 169A specifies protection of individual class
I areas and that BART requirements would be more stringent than CAIR for
individual sources; and they claim that overall, visibility improvements attributable
to CAIR would not be adequate to meet CAA goals.
EPA’s effort to meld the visibility program with CAIR is consistent with its
expressed desire to redirect CAA compliance strategies toward a market-oriented,
cap-and-trade program, viewed by many as more cost-effective than direct regulation
(such as BART). The Administration has proposed “Clear Skies” legislation to create
a more integrated trading process for addressing SO2 and NOx emissions from
EGUs, but it failed to be reported out of committee in the Senate. CAIR represents
a regulatory initiative to achieve a step in coordinating certain CAA programs, but
it may be that a statutory solution will be necessary.
This report will not be updated.



Contents
Background to the Regional Haze Rule.............................1
The Regional Haze Rule........................................3
The Regional Haze Rule and Very Fine Particulates...................5
The Regional Haze Rule and the Clean Air Interstate Rule..............7
The Clean Air Interstate Rule (CAIR)..........................7
BART, CAIR, and Electric Generating Units....................8
Questions ...............................................12
Implications .................................................13
List of Tables
Table 1. Schedule Outline for Section 308 Regional Haze Program...........6
Table 2. EPA Estimates of Regional NOx and SO2 Emissions...............7



Visibility, Regional Haze, and the
Clean Air Act: Status of Implementation
Background to the Regional Haze Rule
When amending the Clean Air Act in 1977, Congress added provisions focused
on protecting the quality of clean air areas, and especially of national parks and other
important national sites.
Codifying regulations developed by EPA in 1974 and 1975, the Prevention of
Significant Deterioration (PSD) program focuses on preventing further deterioration
of air quality in pristine areas of the country by specifying how much increase in
pollution levels is permitted.1 Mandatory class I areas — those areas that receive the
maximum amount of protection — include most national parks, national wilderness
areas, and national memorial parks, currently 156 areas. PSD regulations apply to
emissions of sulfur dioxide (SO2), particulates (PM), and nitrogen oxides (NOx) from
new and modified sources of air pollution.2
Along with the PSD program for new sources, the Congress also added a new
Section 169A, setting “as a national goal the prevention of any future, and the
remedying of any existing, impairment to visibility in mandatory class I Federal3
areas....” PSD and Section 169A act in tandem, with PSD controlling new sources
of impairment and Section 169A reducing emissions from existing sources of
impairment. Under PSD, major new or modified sources in PSD areas must undergo
preconstruction review and must install “best available control technology” (BACT);
more stringent controls can be required if modeling indicates that BACT is
insufficient to avoid violating an allowable PSD increment or the National Ambient
Air Quality Standard itself. Under Section 169A, 26 categories of major stationary
sources of pollution in existence on the date of enactment (1977), but not more than

15 years old as of that date, must install “best available retrofit technology” (BART)


if the state determines the source may reasonably be anticipated to cause or contribute
to any impairment of visibility in a class I area. Included in the list are electric
generating units (EGUs).
Implementing these provisions protecting visibility has not been easy,
particularly Section 169A respecting existing sources. First, EPA had to define what
visibility was. In general, visibility impairment from human activities manifests
itself in two ways: (1) plume blight, where a clearly identifiable plume of smoke


1 1977 Clean Air Act Amendments, Part C, Title 1. P.L. 95-95.
2 EPA could administratively set PSD requirements for other pollutants for which National
Ambient Air Quality Standards (NAAQS) have been established.
3 Section 169A, Clean Air Act, 42 U.S.C. 7492 (italics added).

emanates from one or more sources; and (2) regional haze, where a uniform
reduction in visual range occurs, or a layered discoloration by hovering bands of air
tinged brown, yellow, or red. Second, EPA had to promulgate regulations within 24
months of enactment to assure that State Implementation Plans (SIPs) required (1)
reasonable progress toward meeting the national goal mentioned earlier, and (2)
compliance with several very specific provisions, including the Best Available
Retrofit Technology (BART) requirements for existing sources.
EPA promulgated rules in 1980 to address visibility impairment that was
“reasonably attributable” to a single source or small group of sources (i.e., plume
blight).4 As with many air pollution regulations, these visibility regulations are
implemented by states through SIPs. In general, the 36 states with mandatory class
I areas were required to revise their SIPs to assure reasonable progress toward the
national visibility goal. The major elements of the regulation were (1) identifying
existing sources causing visibility impairment and creating procedures for
determining which existing stationary sources should be subject to BART
requirements; (2) assessing potential adverse impacts from proposed new sources (or
modified old sources) and recommending remedial actions via the New Source
Review (NSR) process and the PSD program; (3) developing a 10-15 year long-term
strategy to make “reasonable progress” toward the visibility goal; and (4) conducting
visibility monitoring in mandatory class I areas.
As noted, these regulations deal with plume blight only — regional haze
reduction was explicitly delayed until some future date. This lack of aggressive
implementation of Section 169A extended to the implementation of the 1980
regulations as well. After 35 of 36 states missed the September 1981 deadline for
final visibility plans, the Environmental Defense Fund sued the EPA in 1982 to
implement the plume blight regulations. The suit was settled in 1984 with the EPA
developing a phased-in schedule for compliance with a December 1986 deadline for
states to revise their SIPs to include controls on existing sources that hinder visibility
goals.5 This sequential implementation of plume blight regulations actually extended
through 1989. So far, the only BART installation to occur under the 1980 regulations
has been the installation of sulfur dioxide scrubbers at the Navajo Generating Station
in Arizona in 1991.6
EPA’s lack of initiative on visibility during the 1980s prompted the Congress
to revisit the issue in the 1990 amendments to the Clean Air Act. Those actions
included a new Title IV, controlling precursors of acid rain and regional haze,7 and
a new Section 169B. In some ways, Section 169B was a triggering mechanism to
force EPA to move on Section 169A with respect to regional haze. Specifically, the


4 45 Federal Register 80084 (December 2, 1980); 40 CFR 51.300-51.307.
5 Environmental Defense Fund v. Gearstick, No. CO2-6850 (N.D. CA) (April 20, 1984). See

49 Federal Register 20647 (May 16, 1984).


6 56 Federal Register 50172 (October 3, 1991); 40 CFR 52.
7 As noted by Section 401(a)(1): “the presence of acidic compounds and their precursors in
the atmosphere and in deposition from the atmosphere represents a threat to natural
resources, ecosystems, materials, visibility, and public health.”

1990 Amendments required EPA to establish a Grand Canyon Visibility Transport
Commission (GCVTC) within 12 months of enactment (and other commissions upon
its own discretion or petition from at least two states). Commissions were required
to assess the scientific, technical, and other data available on visibility impairment
from potential or projected emissions growth in the region. Based on those data, the
commissions were to issue reports within four years to EPA recommending what
measures, if any, should be taken to remedy such impairment. Within 18 months of
receiving a commission’s report, EPA was to carry out its responsibilities under
Section 169A, including criteria for measuring “reasonable progress” toward the
national goal. Finally, states affected by any regulations promulgated under Section

169A were required to revise their SIPs within 12 months of such promulgation.


In 1991, a Visibility Transport Commission for the region affecting visibility in
Grand Canyon National Park was established. In June 1996, this commission
(consisting of the governors of Arizona, California, Colorado, Nevada, New Mexico,
Oregon, Utah, and Wyoming, and the leaders of five Indian tribes) approved a set of
recommendations for improving western vistas.8 There were nine primary
recommendations, including increased energy conservation, use of renewable energy,
and emission reductions from stationary sources.9 The commission’s Baseline
Forecast anticipated that current regulatory programs would reduce emissions of
sulfur dioxide from stationary sources (power plants, smelters, and other industrial
sources) 13% by the year 2000, although additional measures under consideration
might reduce emissions 20%-30%. In light of this uncertainty about the effects of
current programs and the fact that emissions were projected to decline in the short
term without additional regulation, the commission agreed to set only regional targets
for sulfur dioxide emissions in the year 2000. The ultimate targets would be in the
range of 50%-70% reduction by the year 2040, but “interim targets may also be
needed to ensure steady and continuing emission reductions and to promote
investment in pollution prevention.”10 If the targets are exceeded, this would trigger
a regulatory program, probably including a regional cap on emissions, with market-
based trading.
The Regional Haze Rule
The 1990 Clean Air Act Amendments required the EPA Administrator to take
action under Section 169A within 18 months of receipt of a commission report. The
proposed rule appeared in the Federal Register on July 31, 1997.11


8 Recommendations for Improving Western Vistas, Report of the Grand Canyon Visibility
Transport Commission to the United States Environmental Protection Agency, June 10,
1996. A ninth state, Idaho, was included in the region, but chose not to participate in the
commission.
9 The recommendations are summarized in ibid., pp. i-iii.
10 Ibid., pp. 34-35.
11 62 Federal Register 41138 (July 31, 1997).

The final regional haze rule was published on July 1, 1999.12 The regional haze
program represents a nationwide effort to protect 156 PSD class I areas from
visibility impairment from manmade air pollution. All 50 states are included under
the program — including those that do not have any class I areas within their
boundaries — since pollution causing haze can travel beyond a state’s boundaries
and contribute to impaired visibility in a class I area located elsewhere. The rule
encourages regional approaches. Indeed, the final rule includes special provisions
(Section 309 program)13 that permit the former member-states of the Grand Canyon
Visibility Transport Commission to implement their specific recommendations
within the framework of the national regional haze program (Section 308 program).
States are required under Section 169A to develop SIPs that ensure reasonable
progress toward the national goal. Under Section 308 of the rule, SIPs must contain
the following:
!Reasonable progress goals. States must establish goals expressed14
in deciviews that provide for reasonable progress toward achieving
natural visibility conditions in class I areas by 2064.
!Calculations of baseline and natural visibility conditions. States
must determine baseline conditions expressed in deciviews for the
most impaired and least impaired days during 2000-2004.
!Long-term planning. States must submit a long-term strategy to
address regional haze for each class I area within the state or affected
by emissions within the state. The strategy must include compliance
schedules, enforceable emission limitations, and other measures
necessary to achieve reasonable progress goals.
!Monitor strategy. States must submit with the SIP a strategy for
measuring, characterizing, and reporting regional haze.
!Best Available Retrofit Technology (BART). States must submit
a BART implementation plan, including emission limitations and15
compliance schedules for each BART-eligible source that “may
reasonably be anticipated” to contribute to visibility impairment in
a class I area. States may choose to use a trading program or other
alternative, if that alternative will achieve greater reasonable
progress to natural visibility conditions than BART.
!Tracking Progress. SIPs must include several provisions to ensure
the adequacy of the SIP. In particular, the SIP must include
requirements for submitting SIP revisions to EPA every 10 years,


12 64 Federal Register 35714-35774 (July 1, 1999)
13 Referring to section 309 of the rule (not of the statute).
14 A measure of clarity of the air.
15 Defined at Clean Air Act, Section 169A(g)(7).

beginning in 2018. Progress reports tracking the state’s reasonable
progress efforts are due every five years. Reports must include a
determination of the adequacy of the state’s SIP.
An alternative program is provided in Section 309 as an option for nine former
members of the Grand Canyon Visibility Transport Commission (GCVTC).16 Five
states chose to meet the EPA deadline for inclusion under this option.17 Based on the
commission’s 1996 report, Section 309 allows states to choose to follow the
commission’s recommendations for reducing visibility impairment in the 16 class I
areas in Colorado rather than the Section 308 program, up to the year 2018.18
Focused primarily on SO2 emissions, which are a major component of regional haze,
states set voluntary “SO2 milestones,” instead of requiring BART. If the milestones
are not achieved, then a back-up mandatory emissions trading program would be
activated to ensure compliance with the milestones. The successor organization to the
GCVTC, the Western Regional Air Partnership (WRAP) submitted to EPA an annex
to the commission’s report in 2000 that identifies the voluntary SO2 reduction
milestones out to the year 2018, along with the back-up trading program details. The
2018 milestone of 510,000 tons would represent a reduction of 320,000 tons from

1990 emissions of 830,000 tons. EPA approved the annex in 2003.19


The Regional Haze Rule and Very Fine Particulates
While working on the regional haze rule, EPA was also proposing to implement
a new National Ambient Air Quality Standard (NAAQS) for very fine particulates
(PM2.5), which are key contributors to regional haze. To implement the 1997 PM2.5
NAAQS, a monitoring network had to be established and three years of data collected
before states could identify PM2.5 nonattainment areas and begin the development of
SIPs. Adhering to the separate schedules could lead some states to revising SIPs
twice, once for visibility and then, a year or two later, for PM2.5 attainment. As a
result, EPA proposed that states preparing SIPs for attaining the 1997 PM2.5 NAAQS
combine it and their submittal of the regional haze SIP revisions.20 In P.L. 105-178,
enacted June 9, 1998, Congress codified this proposal and also extended deadlines
for areas not designated nonattainment. The enacted language stipulates that SIPs
implementing the regional haze rule be submitted on the same schedule as those for
PM2.5 nonattainment areas.21


16 Arizona, California, Colorado, Idaho, Nevada, New Mexico, Oregon, Utah, and Wyoming.
17 Arizona, New Mexico, Oregon, Utah, and Wyoming.
18 Recommendations for Improving Western Vistas, Report of the Grand Canyon Visibility
Transport Commission to the United States Environmental Protection Agency, June 10,

1996.


19 68 Federal Register 33764-33791 (June 5, 2003)
20 For discussion of the PM2.5 SIP deadlines, see CRS Report 97-8, Air Quality: Background
Analysis of EPA’s 1997 Ozone and Particulate Matter Standards, pp. 22-23 (note 54).
21 Section 6102(c)(2), The Transportation Equity Act for the 21st Century.

This linking of the implementation schedules of regional haze and PM2.5 rules
effectively extended the regional haze actions. Under the nationwide Section 308
program, states classified as attainment under the 1997 PM2.5 NAAQS have one year
after that designation (which occurred on December 17, 2004) to submit to EPA their
revisions to SIPs to implement the regional haze requirements. But states classified
as nonattainment under the 1997 PM2.5 NAAQS will have three years after that
designation to submit to EPA their revised SIP, allowing them to combine
implementation of the regional haze rule with the 1997 PM2.5 NAAQS compliance.
Optional SIP schedules are provided for states that chose to develop a regional,
coordinated approach to regional haze. Likewise, states choosing to follow the
recommendations of the GCVTC have an alternative compliance schedule. Table 1
provides a rough implementation schedule for the regional haze rule based on EPA’s22
latest estimated schedule for PM2.5 compliance.
Table 1. Schedule Outline for
Section 308 Regional Haze Program
DateRegulatory Action
April 5, 2005Effective date of final PM2.5 NAAQS area
designations.
April 5, 2006, or one year afterStates submit haze plans for areas designated
the final PM2.5 designation dateattainment or unclassifiable under PM2.5 NAAQS.
April 5, 2008States submit haze plans for areas designated
nonattainment under PM2.5 NAAQS. States
participating in regional planning submit haze
plans.
2011-2013 (five years afterSources subject to BART required to install and
approval of haze plans)operate BART.
2013 (and every five yearsStates submit progress report on reasonable
thereafter)progress goals and adequacy of haze plans.
Before 2018Sources comply with any emission trading or
alternative control measures.

2018 (and every 10 yearsStates complete revised haze plans.


thereafter)
Source: Environmental Protection Agency.


22 In September 2006, EPA promulgated a revised PM2.5 NAAQS. With an implementation
schedule five years later than the 1997 standard, there maybe some future effort to
coordinate the revised standard’s implementation with the regional haze rule’s schedule.

The Regional Haze Rule and the Clean Air Interstate Rule
The regional haze and PM2.5 programs interact with other air quality programs
as well — notably EPA’s finalized Clean Air Interstate Rule (CAIR).23
The Clean Air Interstate Rule (CAIR).24 Published May 12, 2005, CAIR
addresses the effect of interstate transport of air pollutants on nonattainment of the
NAAQS for fine particulates (PM2.5) and the 8-hour ozone standard. For PM2.5, the
rule finds that the interstate transport of SO2 and NOx from 23 states and the District
of Columbia contribute significantly to downwind nonattainment; for ozone, the rule
finds that interstate transport of NOx from 25 states and the District of Columbia
contribute significantly to downwind nonattainment of the 8-hour ozone standard.
Both SO2 and NOx are involved in regional haze and PM2.5, with SO2 playing a25
particularly major role, so all three programs ultimately deal with some of the same
sources of pollution — of which electric generating units are a major one.
To remedy the situation, CAIR generally follows (with some important26
exceptions) the methodology EPA employed with the NOx SIP Call, a regulation
addressing regional ozone nonattainment. With CAIR, EPA proposes a region-wide
emissions cap for NOx and SO2 to be implemented in two phases — 2010 (2009 for
NOx) and 2015. Based on the methodology employed in the rule, EPA’s estimates
of emissions under the caps are provided in Table 2. EPA determined the caps by
applying “highly cost effective” pollution controls on electric generating units.
Table 2. EPA Estimates of Regional NOx and SO2 Emissions
(million tons)
NOxNOxSO2 EmissionsSO2 Emission
YearEmissionsEmissions(no cap)(with cap)
(no cap)(with cap)
2010 (2009 for NOx)2.71.58.75.1
2015 2.8 1.3 7.9 4.0
2020 2.8 1.3 7.7 3.3
Source: Environmental Protection Agency, 2005.


23 Environmental Protection Agency, Rule to Reduce Interstate Transport of Fine
Particulate Matter and Ozone (Clean Air Interstate Rule); Revisions to Acid Rain Program;
Revisions to NOx SIP Call; Final Rule (70 Federal Register 25162-25405, May 12, 2005).
24 For more on CAIR, see CRS Report RL32927, Clean Air Interstate Rule: Review and
Analysis, by Larry Parker; also CRS Report RL32273, Air Quality: EPA’s Proposed
Interstate Air Quality Rule, by Larry Parker and John Blodgett (available from the authors).
25 SO2 is the subject of numerous provisions of the Clean Air Act: these include the SO2
NAAQS, New Source Performance Standards (NSPS), Prevention of Significant
Deterioration (PSD), Acid Precipitation provisions (Title IV), and mobile source provisions.
26 For background and discussion of the NOx SIP Call, see CRS Report 98-236 ENR, Air
Quality: EPA’s Ozone Transport Rule, OTAG, and Section 126 Petitions — A Hazy
Situation? by Larry Parker and John Blodgett (available from the authors).

BART, CAIR, and Electric Generating Units. Both the regional haze rule
and CAIR address emissions of SO2 and NOx. Although each could control
emissions from any major source of these emissions, CAIR is focused on electric
generating units, while the regional haze rule is focused on 26 different categories of
sources. Therefore, as major sources of SO2 and NOx, electric generating units
become a critical point of interaction between CAIR and the regional haze rule. The
contentious issue has been whether BART for EGUs can be and should be
superseded by CAIR for affected EGUs.
Determining BART under the Regional Haze Rule. The Clean Air Act
explicitly states that BART decisions are to be made according to their impact on
visibility. As stated in Section 169A:
...each major stationary source ... which, as determined by the State ... emits any
air pollutant which may reasonably be anticipated to cause or contribute to any
impairment of visibility in any such area, shall procure, install, and operate, as
expeditiously as practicable (and maintain thereafter) the best available retrofit
technology, as determined by the State ... for controlling emissions from such27
source for the purpose of eliminating or reducing any such impairment....
EPA originally proposed guidelines to assist states in determining BART in

2001.28 After portions of the regional haze rule were remanded by the court in the29


American Corn Growers v. EPA decision, EPA revised and re-proposed its BART
determination guidelines in May 2004.30 In particular, the proposed revisions
focused on state determinations of individual source contributions, rather than on the
collective contribution to visibility impairment as contained in the proposed regional
haze rule and 2001 guidelines: “... this reproposal focuses on the use of single source
emission modeling for assessing the degree of improvement in visibility from various31
BART control levels.”
Under Section 169A, BART is a plant-by-plant determination made by the state
— except for EGUs over 750 Mw in capacity, for which EPA makes the
determination. When EPA proposed its May 2004 revisions to the regional haze rule,
it proposed to set the default 750 Mw EGU SO2 reduction requirement at 95%
removal or emission limitations in the range of 0.1 to 0.15 lb. SO2 per million Btu.
For units between 250 Mw and 750 Mw, EPA proposed a rebuttable presumption
that states should require the same limitations. As stated by EPA:


27 Clean Air Act, section 169A(4)(b)(2)(A).
28 66 Federal Register 38108-38135 (July 20, 2001).
29 American Corn Growers Association v. EPA, 291 F. 3d 1 (May 24, 2002, D.C. Cir.). In
that case, the court ruled that it is the states, not EPA, who must determine which BART-
eligible sources should be subject to BART. Further, the court stated that the regional haze
rule tied the states’ hands and forced them to require BART controls at sources “without any
empirical evidence of the particular source’s contribution to visibility impairment in a Class
I area.”
30 69 Federal Register 25184-25232 (May 5, 2004).
31 69 Federal Register 25203 (May 5, 2004).

This presumption would apply unless the State has persuasive evidence that an
alternative determination is justified. Our intent is that it should be extrememly
[sic] difficult to justify a BART determination less than the default control level
for a plant greater than 750 Mw, and just slightly less difficult for a plant 75032
Mw or smaller.
On July 6, 2005, EPA finalized its guidelines for determining BART.33 For
coal-fired EGUs greater than 200 Mw, the BACT presumptive emissions limit for
SO2 was set at 95% removal or an emissions rate of 0.15 lb. SO2/mmBtu. NOx
BACT presumptive limits for coal-fired EGUs are based on the coal-type burned and
the firing configuration. The limits range from 0.15 lb./mmBtu NOx for tangential-
fired boilers using subbituminous coal to 0.62 lb./mmBtu NOx for wet-bottom
tangential-fired boilers using bituminous coal.34
The 1999 regional haze rule also allowed for a trading program for
implementing BART if the state requesting a trading program submitted analyses
demonstrating “that the emissions trading program or other alternative measure will
achieve greater reasonable progress than would have resulted from the installation
and operation of BART at all sources subject to BART in the State.”35 Under the
1999 regional haze rule, the specific requirements for substituting emissions trading
for BART were as follows:36
!“The State must demonstrate that this emission trading program ...
will achieve greater reasonable progress than would be achieved
through the installation and operation of BART.” This
demonstration must be based on analysis of the visibility
improvement that would be achieved in class I areas.
!The trading program must apply to all BART-eligible sources unless
the source has an enforceable emission limitation that the EPA and
state determines meets BART.
!Emission reductions must occur by 2018 (the first long-term strategy
period).
!“A demonstration that the emission reductions resulting from the
emission trading program ... will be surplus to those reductions
resulting from measures adopted to meet requirements of the CAA
as of the baseline date of the SIP.”


32 69 Federal Register 25199 (May 5, 2004).
33 Environmental Protection Agency, Regional Haze Regulations and Guidelines for Best
Available Retrofit Technology (BART) Determinations; Final Rule, 70 Federal Register

39103-39172 (July 6, 2005).


34 70 Federal Register 39172 (July 6, 2005).
35 64 Federal Register 35768 (July 1, 1999).
36 64 Federal Register 35768 (July 1, 1999).

The proposed 2001 BART guidelines also proposed guidelines for states to
assist them in determining the appropriate state emission budgets (or caps) for their
trading program to ensure it met the greater reasonable progress requirement.37 The
proposed guidelines would have required dispersion modeling of BART and the
trading program to ensure better visibility. Specifically, the modeling should identify
(1) the difference in visibility conditions under both approaches for each class I area;
and (2) the average difference in visibility over all class I areas affected by the
region’s emissions. The analysis would demonstrate greater reasonable progress if
(1) visibility does not decline in any class I area; and (2) there is overall improvement
in visibility as determined by comparing the average differences over all affected
class I areas. These trading program guidelines were re-proposed on May 5, 2004,
essentially unchanged.38
These alternative program guidelines were not included in the final rule because
of a D.C. Circuit Court decision vacating EPA’s approval of the WRAP alternative
trading program under Section 309 of the regional haze rule (the WRAP Annex
Rule).39 Instead, on July 20, 2005, the EPA proposed new requirements for an
emissions trading program that responded to the objections raised by the court. The
final rule was published on October 13, 2006.40 With respect to the alternative
trading program, the primary change to the existing guidelines was to bring the
program requirement into compliance with the American Corn Growers v. EPA
decision.41 Specifically, the revision permits states to use the same BART
determination approach to develop a baseline estimate of BART in the alternative
program, as it allows for source-by-source BART. As stated by EPA:
In short, to demonstrate that a trading program or other alternative program
makes greater reasonable program than BART, the State can develop an estimate
of BART emissions reductions using the same approach that it would use to
establish source-by-source BART emission limitations under the BART42
guidelines.
Substituting CAIR for BART. CAIR is designed to assist states in meeting
the PM2.5 and 8-hour ozone NAAQS by mitigating interstate air pollution. As a
preferred implementation strategy, EPA encourages states to use a trading program
to reduce emissions in a cost-effective manner. To set allocations, EPA compared
the costs of various control strategies to determine the most cost-efficient allocation
scheme. That cost analysis indicated that electric generating units were the most
cost-effective source of emission reductions. Thus, like the NOx SIP Call before it,


37 66 Federal Register 38108-38135 (July 20, 2001).
38 69 Federal Register 25184-25232 (May 5, 2004).
39 Center for Energy and Economic Development v. EPA (398 F. 3d 653, 2005).
40 71 Federal Register 60612-60634 (October 13, 2006).
41 American Corn Growers Association v. EPA, 291 F. 3d 1 (May 24, 2002, D.C. Cir.).
42 71 Federal Register 60612-60634 (October 13, 2006), p. 60615.

the emissions allocations under the CAIR proposed trading program are based on
cost-effectiveness criteria.43
EPA opened the issue of substituting CAIR for BART in a supplemental
proposed rule published June 10, 2004, that detailed the proposed CAIR model
trading program.44 Among its provisions, the proposed supplemental rule would have
permitted electric generating units to use the emission trading program under CAIR
to meet the BART requirement imposed by the regional haze rule. To achieve this,
EPA proposed to amend the trading program requirements under the regional haze
rule. The proposed supplemental rule would have amended and revised the regional
haze regulation to exempt electric utility sources that comply with the CAIR from the
regional haze regulation’s BART requirement. Specifically, the CAIR would have:45
!Revised Section 308(e)(2) so that sources participating in the CAIR
trading program would have been excluded from the requirement
that a state demonstrate that its regional haze emission trading
program “will achieve greater reasonable progress than would be
achieved through the installation and operation of BART.”
!Inserted a renumbered Section 308(e)(3) providing that a state’s
BART-eligible electric generating units that participate in the CAIR
trading program would not have to install and operate BART.
Thus, the proposed supplemental CAIR revisions to the regional haze rule
would have done two things: (1) exempted states from having to demonstrate that
sources complying with the CAIR through its proposed trading program would
achieve greater reasonable progress than would be achieved through the installation
and operation of BART; and (2) exempted such sources from BART.
In the final CAIR, EPA decided to defer the decision on substituting CAIR for
BART for affected units until the BART guidelines are finalized. As stated by EPA:
The results clearly indicate that the CAIR will achieve greater reasonable
progress than BART as proposed, measured by the proposed better-than-BART
test. At this time, we can foresee no circumstances under which BART for
EGUs could produce greater visibility improvement than the CAIR. However,
for the reasons noted in section IX.C.1 above, we are deferring a final
determination of whether the CAIR makes greater reasonable progress than
BART until the BART guidelines for EGUs and the criteria for BART-46


alternative programs are finalized.
43 Although, unlike the NOx SIP Call, CAIR focuses solely on electric generating units,
where the NOx SIP Call included other sources of cost-effective NOx reductions.
44 69 Federal Register 32683-32772 (June 10, 2004).
45 69 Federal Register 32738 (June 10, 2004).
46 Clean Air Interstate Rule, p. 25304.

In the final BART rule, EPA finalized its determination that CAIR achieves
greater progress than BART and may be used by states as a BART substitute.47 In
making this determination, EPA notes that “we are not constraining the discretion of
States to determine which sources are subject to BART and to make BART
determinations. CAIR-affected States are not required to accept our determination
that CAIR may substitute for BART.”48
EPA’s Justification. The proposal by EPA to declare CAIR to be better than
BART for individual BART-eligible electric generating units had been strongly
hinted in its May 2004 proposed revisions to BART guidelines. In that proposal,
EPA included a strong statement of support for both employing trading programs to
address regional haze, and the use of CAIR as a “better than BART” alternative. As
stated by EPA:
Based on our current evaluation, we believe the [Interstate Air Quality Rule, later
CAIR] ... as proposed, is clearly better than BART for those affected EGUs in
the affected States which we propose to cover under the IAQR. We thus expect
that the final IAQR would satisfy the BART requirements for affected EGUs that49
are covered pursuant to the final IAQR.
Analysis to support this declaration was provided in the June 2004 proposal
supplemental rule for the CAIR trading program, and in the final CAIR.50 The Bush
Administration uses a regional analysis of the visibility improvement resulting from
BART and CAIR to justify exempting BART-eligible electric generating units from
BART and from the requirement that trading rather than installing BART must yield
greater reasonable progress. The two-part test examined the effects of the two
programs on 116 class I areas with respect to potential visibility degradation. The
analysis concludes that “CAIR emissions reductions in the East produce significantly51
greater visibility improvements than source-specific BART.” On a nationwide
basis, EPA states:
... the visibility modeling shows that for all 116 class I areas evaluated, the
average visibility improvement, on the 20 percent worst days, in 2015 was 0.5
dv [deciview] under the CAIR cap-and-trade program in the East and BART in
the West, but only 0.2 dv under the nationwide source-specific BART52
approach.
Questions. This linking of CAIR to the regional haze rule is based on the
programs’ common characteristic of controlling sulfur dioxide and nitrogen oxides.
EPA use of a proposed collective methodology designed to assist states in
determining state emission budgets to justify excluding individual units from


47 70 Federal Register 39137 (July 6, 2005).
48 70 Federal Register 39143 (July 6, 2005).
49 69 Federal Register 25204 (May 5, 2004).
50 70 Federal Register 25299-25304 (May 12, 2005).
51 70 Federal Register 25303 (May 12, 2005).
52 70 Federal Register 25303 (May 12, 2005).

undergoing individual state-led BART review has proven contentious. Questions
include the following:
!Visibility impacts on Class I Areas individually and collectively.
Critics contend that EPA’s analysis suggesting that “nationwide” the
“average” visibility would improve more under a CAIR/BART
program than a BART program is inadequate. They observe that
Section 169A discusses BART in terms of visibility impairment of
“any” class I area — not an average of all 156 class I areas or the
116 (29 in the East) class I areas EPA analyzed for its CAIR
determination. EPA cites data limitations for not including other
class I areas (5 in the East) in its analysis.53 With the final BART
determination guidelines permitting such an analysis, litigation is
likely.
!Stringency of BART versus CAIR. On an individual EGU basis,
the 95% reduction requirement contained in EPA’s BART
guidelines is substantially more stringent than the overall 67%
reduction in SO2 emissions from a future 2015 baseline envisioned
in CAIR. However, the scope of the two scenarios is different. For
EGUs, BART is required nationwide on powerplants over 250
million Btu (thermal input basis) operating in 1977 but not more
than 15 years old (1962) that may reasonably be anticipated to cause
or contribute to any impairment of visibility in any class I area.
CAIR’s model trading program is a regional scheme focused on all
EGUs that are greater than 25Mw within the 23 state PM2.5 region.54
!Achieving visibility goals. As noted above, using an analysis that
grafted the CAIR trading program onto an individual BART
program for the other 25 BART-eligible categories, EPA argues that
CAIR is better than BART. Opponents argue that the analysis is
insufficient — that a 1-2 deciview improvement will not achieve the
CAA visibility goal. Instead, opponents assert that a 7-9 deciview
improvement is necessary to achieve the CAA’s visibility goal of
preventing any future, and remedying any existing, visibility
impairment in class I areas. Achieving such a goal will require
BART controls on the level of EPA’s proposed default levels, plus
controls on additional EGUs such as required under CAIR.
Implications
The Clean Air Act has evolved over time in response to a developing
understanding of the environment, new technologies, and changes in the nation’s
transportation, energy, and industrial sectors. The result has been a patchwork of
requirements that are not always consistent — and may even be incompatible — at


53 69 Federal Register 32705, note 34 (June 10, 2004).
54 69 Federal Register 32709 (June 10, 2004).

any given moment. Moreover, implementing regulations change and are added to
over time. Although the evolution of the act has resulted in a structure that some
consider unwieldy, emissions of most air pollutants have substantially declined, and
the number of persons living in areas where pollution exceeds standards has
diminished. 55
From a policy standpoint, EPA has presented the Clean Air Interstate Rule —
and the accompanying Mercury (Hg) rule — as a “suite of integrated air actions” to
reduce emissions of three pollutants: SO2, NOx, and Hg. By promulgating guidelines
to help states determine appropriate state emissions budgets for their trading program
and to exempt sources subject to the CAIR from the individual BART determinations
required by Section 169A (visibility impairment), EPA appears to be trying to extend
the “suite” to the visibility protection provisions of the CAA. In other words, EPA
is endeavoring to transform CAIR from another layer on the already multilayered
cake called the Clean Air Act to an integrative program that simplifies the layers.
As discussed, this effort to meld Section 169A (visibility) and Section 109
(NAAQS) implementation strategies based on their common characteristic of
controlling sulfur dioxide and nitrogen oxides raises numerous issues. In the Clean
Air Act, their only procedural link is the SIP process, but they have different scopes,
purposes, and requirements. Because of the happenstance that the timing of the PM2.5
NAAQS and the regional haze rules partly coincided, EPA proposed and Congress
legislated that the initial implementation schedules of the SIP process for the 1997
PM2.5 NAAQS and the SIP process for the regional haze rule be coordinated.
Now EPA is linking the two programs by determining that certain CAIR
program provisions can in effect substitute for related but different visibility
requirements. Whether it can do this while accomplishing the express purposes and
requirements of both Section 109 and Section169A is subject to debate and possible
litigation. By using a collective analysis designed to assist states in determining state
emission budgets to justify excluding individual units from undergoing individual
state-led BART review, EPA concludes that the CAIR program adequately meets
visibility requirements — a conclusion that is contentious. Indeed, opponents of the
attempt have described it as regulatory “bait and switch.”56 This conflict is not
surprising as EPA is attempting to integrate regulatory provisions that are separate
in many essential respects.
It appears the Administration’s goal is to redirect CAA compliance strategies
toward a market-oriented cap-and-trade program — viewed by many observers as a
more cost-effective approach to pollution control than direct regulation (such as the
BART program). Such a redirection of compliance approaches has been proposed
— and the Title IV acid rain provisions of the CAA are often cited as the preeminent
example of its application. Several proposals have focused on electric generating


55 For data on air quality trends, see EPA’s website at [http://www.epa.gov/airtrends/].
56 Felicity Barringer, “Critics Say Clean-Air Plan May Be a Set Back for Parks,” New York
Times, May 31, 2004.

units.57 One approach is a “multi-pollutant” strategy — a framework based on a
consistent set of emissions caps, implemented through emissions trading. In
February 2002, the Bush Administration announced a “Clear Skies” multi-pollutant
proposal that would amend the Clean Air Act to place emission caps on electric
utility emissions of SO2, NOx, and Hg. Implemented through a tradable allowance
program, the emission caps would generally be imposed in two phases: 2008 and
2018. Although different in geographic scope, the Administration’s Clean Air
Interstate Rule and mercury rule are very similar in terms of reduction requirements
as Clear Skies. However, unlike EPA’s “suite of integrated air action,” Clear Skies
contains significant conforming language to avoid conflicts with other CAA
provisions such as Section 169A. The Administration has stated its preference for
Clear Skies over its regulatory approach.
However, the Congress has yet to move any multi-pollutant proposal to the
floor, nor has it given EPA broad authority to reconstitute regulatory approaches into
market-oriented ones.58 EPA’s combining of CAIR and BART represents a
regulatory initiative to achieve at least a partial step in coordinating regulatory
programs under a market-oriented approach. It is possible, however, that a statutory
solution could be necessary.


57 See CRS Report RL33755, Air Quality: Multi-Pollutant Legislation in the 109th Congress,
by Larry Parker.
58 In the 109th Congress, S. 131, introduced by Senator Inhofe, is modeled on the
Administration’s Clear Skies proposal. However, after committee markup, the Senate
Environment and Public Works Committee failed to report S. 131 to the floor on a 9-9 tiedth
vote (March 9, 2005). See CRS Report RL33552, Clean Air Act Issues in the 109
Congress, by James E. McCarthy.