Amicus Curiae Brief — Environmental Defense v. Duke Energy Corporation

6

Power Company v. Reilly, 893 F.2d 901, 905 (7th Cir. 1990)

(“WEPCO”), United States v. Ohio Edison Co., 276 F. Supp.

2d 829, 862 (S.D. Ohio 2003); United States v. Southern

Indiana Gas and Electric Co., 245 F. Supp. 2d 994, 998

(S.D. Ind. 2003) (““SIGECO”).

In EPA’s PSD enforcement action against Duke

Energy, the Fourth Circuit’s role is to apply the “modifica-

tion” test specified in EPA’s PSD regulations and deter-

mine if the company’s plant renovations triggered PSD

requirements. Yet, completely disregarding section 307’s

clear requirements, the Fourth Circuit announced that

“the language and various interpretations of the PSD

regulations, on which the district court partially based its

holding and which the parties exhaustively discuss, are

largely irrelevant to the proper analysis of this case.”

United States v. Duke Energy, 411 F.3d 539, 547 n.3 (4th

Cir. 2005). Instead of interpreting and applying EPA's

regulations, the Fourth Circuit focused exclusively on

ascertaining what it thought the statute requires, and

based on that statutory reading, announced that the PSD

program must apply the same test as in the NSPS regula-

tions for determining when a “modification” occurs. See id.

at 550. (“Congress’ decision to create identical statutory

definitions of the term ‘modification’ has affirmatively

mandated that this term be interpreted identically for the

two programs.”). Without making any attempt to explain

how EPA’s PSD regulations could be read to establish an

hourly emissions rate test, the court declared that the PSD

test must turn on hourly emissions because “[n]o one

disputes that prior to enactment of the PSD statute, the

EPA promulgated NSPS regulations that define the term

‘modification’ so that only a project that increases a plant’s

hourly rate of emissions constitutes a ‘modification.’” /d.

at 550 (emphasis in original).

The Fourth Circuit tries to explain away the section

307 requirements by stating that “lolur choice of this

interpretation of the PSD regulations — as required under

the statute — over the EPA's interpretation is not an

invalidation of those regulations.” Jd. at 549 n.7. But the

Fourth Circuit did not interpret the regulations; if it did, it

would have no choice but to conclude that the PSD regula-

tions define modification in terms of whether a physical or

operational change will result in an annual emissions

increase, taking into account increased hours of operation

enabled by the change.’ To the contrary, the Fourth Circuit

ignored the PSD regulations and substituted its own test

based on its reading of the statute — something it has no

jurisdiction to do. See Adamo Wrecking v. United States,

434 U.S. 275, 284 (1978) (The district court is not in

violation of section 307 because it “did not presume to

judge the wisdom of the regulation”). See also WEPCO,

893 F.2d at 914 n.6 (“As a preliminary matter, we note

that WEPCO has not asked us to review the propriety of

the NSPS regulations themselves. Indeed, we have no

such jurisdiction to conduct such an inquiry: 42 U.S.C.

section 7607(b\(1) reserves such questions for the [DC

Circuit].... WEPCO simply requests that we consider

whether the EPA properly applied these regulations to the

... generating units. We have jurisdiction to undertake

such an inquiry.”); United States v. Ho, 311 F.3d 589, 607

(5th Cir. 2002); Puerto Rican Cement Co. v. Environmental

Protection Agency, 889 F.2d 292, 299 (1st Cir. 1989).

* See infra pp. 14-15 for further discussion of the plain language of

the PSD regulations.

8

Duke is well aware that a challenge to the validity of

nationally applicable regulations must occur in the D.C.

Circuit; indeed, a host of industry petitioners, including

Duke, challenged various provisions of the 1980 regula-

tions directly in the D.C. Circuit shortly after those regu-

lations were promulgated. See Chemical Manufacturers

Ass’n v. Environmental Protection Agency, Nos. 79-1112 et

al. That case was stayed several times and was recently

revived and consolidated with additional challenges to

EPA’s 2002 rules in New York I. 413 F.3d at 3. In New York

I, after stating that the “Industry petitioners also chal-

lenge the 1980 rule’s definition of modification in the NSR*

context to the extent that it differs from the NSPS defini-

tion,” the court held, “[wje are not convinced.” Jd. at 18.

The D.C. Circuit found no indication in the statute or

legislative history that Congress intended to incorporate

the definition of “modification” in the NSPS regulations

into the PSD program. /d. at 19. Further, the court noted

that the NSPS regulations in place as of the 1977 Act

amendments “used two different (and possibly inconsis-

tent) definitions of modification,” and that “[gliven the two

quite differently worded regulatory definitions of ‘modifi-

cation’ within the NSPS program at the time of the 1977

amendments, it would take a rather pointed indication

from Congress to support the idea that it expressly

adopted one of them for NSR. No such indication exists.”

* NSR or New Source Review is the overarching program that

includes PSD review, for areas in attainment with the National

Ambient Air Quality Standards (“NAAQS”), and Nonattainment New

Source Review, for areas that are not in attainment with the NAAQS.

This case concerns only PSD as the areas where Duke's plants are

located are in attainment for the NAAQS.

* In New York I, the D.C. Circuit rejected industry’s congressional

incorporation argument — that when Congress amended the Clean Air

(Continued on following page)

9

Id. at 19-20 (emphasis in original). In sum, :ust as directed

by section 307, Duke challenged the validity of EPA’s PSD

regulations in the D.C. Circuit. After failing in that chal-

lenge, Duke cannot have a second bite at the apple in this

case.

B. Uniform Judicial Review of Nationally

Applicable Regulations Is Critical to

EPA’s Ability to Properly Carry Out the

Directives of the Clean Air Act.

This case provides a prime example for why assurance

that “the substantive provisions of ... {regulations are]

uniformly applied and interpreted and that the circum-

stances of ... adoption [should] be quickly reviewed by a

single court” is of critical importance to EPA's ability to

implement the Act. Adamo Wrecking, 434 U.S. at 284;

United States v. Ethyl Corp., 761 F.2d 1153, 1156 n.7, 1157

(5th Cir. 1985). Following the enactment of the Clean Air

Act in 1970, EPA was tasked with establishing National

Ambient Air Quality Standards (“NAAQS”), by “iden-

tify{ing] the maximum airborne concentration of a pollut-

ant that the public health can tolerate, decreas[ing] the

concentration to provide an ‘adequate’ margin of safety,

and set(ting] the standard at that level.” Whitman v.

American Trucking Ass’ns, 531 U.S. 457, 465 (2001). The

Act in 1977 and added the PSD program Congress intended to incorpo-

rate the existing NSPS regulations into the PSD program. 413 F.3d at

20. As to the slightly different argument Duke attempts to make here,

the court stated “[wje express no opinion as to whether Congress

intended to require that EPA use identical regulatory definitions of

modification across the NSPS and NSR programs. That argument was

not made by industry petitioners in their opening brief and is therefore

waived.” /d. (internal citations omitted).

10

PSD program is crucial to the Act because it “[s]eek[s} to

prevent backsliding in regions whose air quality [meets

the] NAAQS [by requiring] sources undertaking modifica-

tions to obtain preconstruction permits.” New York I, 413

F.3d at 12. Without uniform, national regulations for the

PSD program, EPA’s ability to approve state implementa-

tion plans (“SIPs”), to write permits and to bring appropri-

ate enforcement actions will be significantly disrupted.

First, the stability and effectiveness of SIPs — which

lie at the core of delegated state enforcement actions —

would be undermined if the Fourth Circuit’s decision is

allowed to stand. The Act requires EPA to approve SIPs,

designed by and for each individual state, provided that

the SIPs “include a program to provide for the . . . regula-

tion of the modification and construction of any stationary

source within the areas covered by the plan as necessary

to assure that the national ambient air quality standards

are achieved, including a permit program.” 42 U.S.C.

§ 7410(a)(2)(C). If the lower courts can invalidate the PSD

regulations at any time, thereby potentially forcing EPA to

redraft the regulations time and again, EPA (and the

states’) could never be certain that the SIPs will actually

contain the appropriate requirements to limit emissions

and achieve the NAAQS. The states would, in essence, be

required to revisit and revise the SIPs every time the

regulations were invalidated and rewritten. 40 C.FR.

§$§ 51.166(a)(3), (4). Since development and adoption of the

regulations is an incredibly costly and time consuming

endeavor, the SIPs would likely be left in limbo for lengthy

* See State of New Jersey’s amicus brief for further discussion of

the importance of uniform PSD regulations in conjunction with the

states’ ability to formulate adequate SIPs.

11

periods of time.’ Section 307 was added to the Act precisely

to prevent such an inefficient scenario. See Adamo Wreck-

ing, 434 U.S. at 284. See also Lubrizol Corp., 547 F.2d at

315, 317.

In addition, without uniform and final PSD regula-

tions, the preconstruction permits will be written under a

veil of uncertainty for both EPA and industry. The PSD

program was established “to assure that any decision to

permit increased air pollution in any area... is made only

after careful evaluation of all the consequences of such a

decision.” 42 U.S.C. § 7470(5). To carry out this goal, the

Act provides that “[n]o major emitting facility ... may be

constructed [or modified] in an area |that has attained the

NAAQS] unless ... the required analysis has been con-

ducted in accordance with regulations promulgated by the

Administrator.” 42 U.S.C. §§ 7475(a\(2), 7479(2\(c). Such

regulations must “assurle] that maximum allowable

increases ... [of pollutants] shall not be exceeded.” 42

U.S.C. § 7473(a). EPA and the states, when issuing per-

mits, bear the responsibility of ensuring that modifications

to existing sources do not “cause, or contribute to, air

pollution in excess of” these thresholds. 42 U.S.C.

§ 7475(aX(3).

" For example, when the 1978 PSD regulations were successfully

challenged in Alabama Power under section 307 in the D.C. Circuit, it

took EPA two years (until August, 1980) to issue amended regulations.

See Alabama Power Co. v. Costle, 636 F.2d 323 (D.C. Cir. 1980). See also

43 Fed. Reg. 26,380 (June 19, 1978); 45 Fed. Reg. 52,676 (Aug. 7, 1980).

When EPA's 1997 rules for ozone NAAQS were successfully challenged

under section 307 in the D.C. Circuit (and then appealed to this Court),

it took EPA seven years (until April, 2004) to issue amended regula-

tions. See American Trucking Ass’ns, 531 U.S. 457. See also 62 Fed.

Reg. 38,856 (July 18, 1997); 69 Fed. Reg. 23,951 (April 30, 2004).

12

The potential for changing regulations would create

competitive disparity across industry if some facilities are

permitted under old regulations, while others are permit-

ted under new regulations.” EPA should not be tasked with

the prospect of redrafting industry permits throughout the

life of the regulations in response to industry challenges to

the regulations. Such a scenario will provide incentive for

regulated entities to challenge regulations any time they -

are unhappy with their permit or are targeted in an

enforcement action. Although EPA is certainly able to

revisit regulations on its own, due to changed circum-

stance or otherwise, any revisions to the regulations would

be subject to notice and comment rulemaking, not the

unilateral decision of a lower court judge.

Finally, the Clean Air Act’s enforcement scheme would

break down in the face of an affirmation of the Fourth -

Circuit’s decision. “The process [for completing a new

source review case] can take years due to: the size of

utilities; the amount and complexity of historical informa-

tion involved; the type of utility expertise needed to

effectively develop an NSR case; and the limited EPA,

State, and local agency enforcement resources.” New

Source Review Rule Change Harms EPA's Ability to En-

force Against Coal-fired Electric Utilities, Office of Inspec-

tor General, Report No. 2004-P-00034, Sept. 30, 2004, pp.

22-23 (available at http://www.epa.gov/oig/reports/2004/

20040930-2004-P-00034.pdf). See also Air Pollution: EPA

Should Improve Oversight of Emissions Reporting at Large

* EPA believes that the Fourth Circuit’s “holding creates a poten-

tial disparity in the way we interpret the program in States in the

Fourth Circuit compared to States in other Circuits in the country.” 70

Fed. Reg. 61,081, 61,083 (Oct. 20, 2005).

13

Facilities, Government Accounting Office, April 2001, pp.

10-11 (available at http://www.gao.gov/new.items/d0146.

pdf). EPA will usually begin by submitting a section 114

request for documents to the regulated entity and can

spend years reviewing these documents before attempting

settlement, filing its own administrative complaint or

referring the case to the Department of Justice for en-

forcement. See id. See also 42 U.S.C. § 7413(a\(1); 42

U.S.C. § 7414. Discovery in this case alone “produced

approximately 46 million pages of documents ... {and}

lengthy briefs [were] accompanied by thousands of pages

of exhibits.” United States v. Duke Energy, 278 F. Supp. 2d

619, 622 (M.D.N.C. 2003). If EPA has to change course in

the middle of building an enforcement case in response to

the invalidation of its own regulations, enforcement

actions will become overwhelmingly inefficient, and in

some circumstances enforcement will simply fall away.

Such a scenario is not what Congress envisioned when the

Clean Air Act was enacted.

II. EVEN IF THE FOURTH CIRCUIT HAD JURIS-

DICTION, ITS REGULATORY INTERPRETATION

DISREGARDS THE PLAIN LANGUAGE OF THE

PSD REGULATIONS AND VIOLATES THE ACT.

industry’s hourly emissions test for triggering PSD

requirements is contrary to the purposes of the Act, EPA’s

consistent PSD regulations, and related case law. The

plain language of the PSD regulations clearly provide for

an actual annual emissions test taking into account

increased hours of operation enabled by a change at a

facility. In addition, only an annual emissions test will

fulfill the purposes of the PSD program to prevent back-

sliding in regions where the air quality is adequate. And

14

the fact that industry, including Duke, and EPA entered

into a settlement agreement that required EPA to propose

an hourly test after the applicable 1980 PSD regulations

were finalized provides further proof that an hourly test

does not apply to this case. Since an hourly test does not

capture total actual annual emissions increases from a

source, such a test simply cannot be sanctioned.

A. Industry’s Hourly Test Is Contrary to the

Plain Language of EPA’s PSD Regulations.

Contrary to the Fourth Circuit’s decision, EPA’s PSD

regulations unmistakably require an annual test, not an

hourly test, for calculating an emissions increase that will

trigger the PSD permitting requirements. The Act requires

the owner or operator of a major stationary source to

obtain a permit before undertaking a “modification,” which

the Act defines as “any physical change in, or change in

the method of operation of, a stationary source which

increases the amount of any air pollutant emitted by such

source or which results in the emission of any air pollutant

not previously emitted.” 42 U.S.C. §7411l(a\4). The

regulations clarify that PSD requirements are triggered

only for “major” modifications. 40 C.F.R. § 51.166(i2). A

major modification occurs where a “physical change ...

would result in a significant net emissions increase.” 40

C.F.R. § 51.166(b\2\i). A significant net emissions in-

crease can be broken into a four-part test:

1. Emissions increase: “[a}ny increase in actual

emissions from a particular physical change or

change in the method of operation at a stationary

source.” 40 C.F.R. § 51.166(b\3)iXa) (emphasis

added).

15

2. Actual emissions: “the actual rate of emis-

sions of a pollutant from an emissions unit .. . In

general, actual emissions as of a particular date

shall equal the average rate, in tons per year, at

which the unit actually emitted the pollutant

during a two-year period which precedes the par-

ticular date and which is representative of nor-

mal source operation. ... Actual emissions shall

be calculated using the unit’s actual operating

hours, production rates, and types of materials

processed, stored, or combusted during the select

time period.” 40 C.F.R. §51.166(b)\21)i), Ui)

(emphasis added).

3. Netting: combine step two with: “|a|ny other

increases and decreases in actual emissions at

the source that are contemporaneous with the

particular change and are otherwise creditable.”

40 C.F_R. § 51.166(b)(3)(b).

4. Significant: The net emissions increase must

exceed pollutant specific annual emissions

thresholds expressed in tons per vear. For exam-

ple, the net emissions increase for sulfur dioxide

must exceed 40 tons per year to trigger the PSD

requirements. 40 C.F.R. § 51.166(b)(23)(1).

The lower courts’ interpretation of this regulatory

language as warranting an hourly emissions test is com-

pletely unfounded. The words “hourly emissions” exist

nowhere in the text of the regulations. Thus, it is unsur-

prising that the Fourth Circuit’s decision omits any

explanation for how the regulations could be read to

establish an hourly test. In light of the plain regulatory

language establishing an actual annual emissions test for

determining PSD applicability, the Fourth Circuit's con-

clusion that EPA must apply an hourly test should be

rejected.

16

B. The EPA Regulations and Related Case Law

Regarding PSD Confirm that Duke Was Re-

quired to Consider Increased Emissions

Caused by Increased Hours of Operation

When Evaluating Whether PSD Requirements

Applied to its Plant Renovations.

The “net emissions increase” step of the “major modi-

fication” test clearly requires a comparison of the pre- and

post-project annual emissions, including “actual operating

hours.” 40 C.F.R. § 51.166(b)(21)(ii). The lower courts’

determination that the pre-project hours of operation must

remain constant in calculating the emissions increase is

simply incorrect. See Duke Energy, 411 F.3d at 545, Duke

Energy, 278 F. Supp. 2d at 640. The so-called “hours of

operation exclusion,” which states that “[a] physical

change ... shall not include ... [a]n increase in hours of

operation,” does not apply to this case. See 40 C.F.R.

§ 51.166(b)(2 iii). Rather, this exclusion simply clarifies

that a facility that is not operating at capacity may in-

crease its hours of operation “to take advantage of favor-

able market conditions,” 45 Fed. Reg. 52,676, 52,704 (Aug.

7, 1980), without triggering PSD requirements, as long as

this increase is achieved without an associated “physical

change in or change in the method of operation of,” 40

C.F.R. § 51.166(b)2)i), the source. In other words, an

increase in hours of operation, on its own, does not consti-

tute an NSR-triggering physical or operational change.

The exclusion does not apply to the present case because

the increase in hours of operation occurred as a result of

the physical changes that Duke made to its facilities.

EPA and the courts have consistently interpreted the

exclusion in this manner. EPA confirmed the narrow

application of the hours of operation exclusion when it

17

issued the applicability determination for the proposed

modifications at the WEPCO plants. WEPCO argued that

since “any emissions increases would be due to increased

production rates or hours of operation rather than higher

emissions per unit of production.... these increases

should be excluded from consideration in determining

whether a net significant emissions increase and, hence, a

major modification, would occur.” Applicability of Preven-

tion of Significant Deterioration Requirements (PSD) and

New Source Performance Standards (NSPS) Requirements

to the Wisconsin Electric Power Company (WEPCO) Port

Washington Life Extension Project, Memorandum From:

Don R. Clay, Acting Asst. Admin. For Air and Radiation,

To: David A. Kee, Director, Air and Radiation Division,

Region V, Sept. 9, 1988, p. 8 (Joint Appendix “JA” 257-58).

EPA responded to WEPCO’s argument by stating that:

The WEPCO is incorrect in this regard.... the

exclusions cited by WEPCO are intended to apply

where a source increases emissions by simply

combusting a larger amount of fuel, or processing

a larger amount of raw materials during a given

time period, or by expanding its hours of opera-

tion ‘to take advantage of favorable market con-

ditions’ (see 45 FR 52704) ... it is obvious that

WEPCO’s plans to increase production rate or

hours of operation are inextricably intertwined

with the physical changes planned under the life

extension project... . [A]ccepting WEPCO’s inter-

pretation of the major modification regulations

would serve to exclude from consideration all

physical or operational changes except those which

cause increased emissions per unit of production.

Clearly, EPA never intended this result. It would

allow, through substantial capital investment,

significant expansion of the pollution-emitting

18

capacity and longevity of major industrial facili-

ties without PSD review of the impacts on air

quality.

Id. (emphasis added).

In a follow up letter to the Vice President of WEPCO,

EPA staff again confirmed the narrow application of the

exclusion by stating that “an increase in any one of these

three factors [emission rate, production rate or capacity

utilization, and hours of operation], if attributable to a

physical or operational change, can trigger an emissions

increase for PSD purposes .. . EPA explicitly assumed that

emissions increases ... would come ... from increases in

production rate or hours of operation.... The hourly

capacity demonstration for NSPS purposes is not relevant

to the PSD analysis.” Letter From: Don R. Clay, Acting

Asst. Admin. For Air and Radiation, To: John W. Boston,

Vice President of WEPCO, February 15, 1989, pp. 9-10 (JA

294-95). EPA’s applicability determination in regard to the

hours of operation exclusion was ultimately confirmed by

the Seventh Circuit where the court decided, inter alia,

that “[djespite WEPCO’s protestations, we note initially

that the EPA’s refusal to apply the ‘production rate/hours

of operation exclusion’ was proper. This exclusion ... was

provided to allow facilities to take advantage of fluctuating

market conditions, not construction or modification activ-

ity.” WEPCO, 893 F.2d at 916 n.11 (internal citations

omitted) (emphasis added). See also Puerto Rican Cement,

889 F.2d at 298.

This application of the exclusion was again memorial-

ized in the preambles to EPA’s 1992 and 2002 rules: “Al-

though a source may vary its hours of operation or

production as part of its everyday operations, an increase in

emissions attributable to an increase in hours of operation or

19

production rate which is the result of a construction-

related activity is not excluded from review.” 57 Fed. Reg.

32,314, 32,328 (July 21, 1992). “|A]n increase in utilization

should not trigger the major NSR requirements unless it is

related to a physical or operational change. .. . [T]he CAA

only applies the major NSR requirements to emissions

increases that are the result of a physical or operational

change. Thus, we do not believe that the major NSR

requirements should apply to a utilization increase unless

the increase is related to the modification.” 67 Fed. Reg.

80,186, 80,203 (Dec. 31, 2002).

Failing to cite a single EPA regulation or judicial

decision in support of its broad-based interpretation of the

hours of operation exclusion, Duke relies solely on an

internal EPA memorandum and a ietter written by a

single EPA staff member in the early 1980s. See Memo-

randum From: Edward E. Reich, Director Division of

Stationary Source Enforcement, To: Charles Whitmore,

Chief, Technical Analysis Section, Region VII, Jan. 22,

1981 (JA 35-37). See also Letter From: Edward E. Reich,

To: Amasjit S. Gill, General Electric, June 24, 1981 (JA 27-

28). Given that the consistent EPA documents (discussed

above) were written both before and after the conflicting

memorandum and letter, as the court in Puerto Rican

Cement similarly held, “these materials do not show a

significant, legally recognizable ‘conflict’ within the

agency.” 889 F.2d at 299. The Fourth Circuit cannot simply

retract the plain language of the regulations and EPA's

and the courts’ historically consistent, narrow application

of the exclusion based on one EPA staff member’s misstep.

See id. See also Ohio Edison Co., 276 F. Supp. 2d at 876-77

(rejecting reliance on Reich letters for same proposition).

Indeed, because “(t]he purpose of the ‘consistency’ doctrine

20

in administrative law is . . . to prevent the agency . . . from

significantly changing those policies without conscious

awareness of, and consideration of the need for, change,”

such a deviation in the application of the regulations

would require EPA to undergo notice and comment rule-

making.’ Jd. The lower courts’ reliance on these documents

for applying the hours of operation exclusion to the pre-

sent facts, and the resulting hourly emissions test, is

therefore improper.

Since EPA and the courts have repeatedly and consis-

tently described the appropriate application of the hours of

operation exclusion, this narrow application should come

as no surprise to Duke. As the Southern District of Indi-

ana court explained in a similar PSD enforcement case,

Congress sweepingly defined modification . . . [for

PSD purposes], and the goal of the CAA was ‘to

speed up and intensify’ the war against pollution.

Moreover, the D.C. Circuit rejected the EPA’s ear-

lier attempts to make broad, categorical exclu-

sions from the CAA’s definition of modification.

Sez Ala. Power v. Costle, 204 U.S. App. D.C. 51,

636 F.2d 323 (D.C.Cir.1979) []. With this regula-

tory context in mind, a context that a sophisti-

cated entity like SIGECO [or in this case Duke]

was surely aware of, it would be inconsistent for

the EPA to broadly define a regulatory exemption

(for PSD].

SIGECO, 245 F. Supp. 2d at 1014.

* In fact, EPA is currently taking comments on a proposed rule for

an hourly emissions test for PSD. 70 Fed. Reg. 61,081 (Oct. 20, 2005).

See infra, pp. 26-27. Therefore, EPA recognized that notice and com-

ment rulemaking was necessary if such a departure is going to occur.

21

In sum, there is no merit to the Fourth Circuit’s

conclusion that EPA’s PSD regulations authorize sources

to exclude emissions resulting from increased hours of

operation made possible by a physical change when

assessing whether that change triggers PSD review. Here,

Duke’s massive renovations plainly enabled its plants to

increase their hours of operation, dramatically increasing

the actual amount of pollution they release into the air

each year. By discounting those emission increases and

proceeding without PSD review, Duke violated the Act.

C. An Hourly Emissions Rate Test for PSD Would

Undermine the Purposes of the Act’s PSD Pro-

visions.

The NSPS and PSD programs were added to the Act

at different times and serve different purposes. In 1970,

Congress included the NSPS program in the Act and EPA

was then required to promulgate technology-based emis-

sion limits for categories of sources without regard to site-

specific factors. 42 U.S.C. § 7411(a)(1). In 1977, when the

Act was amended, Congress added the PSD program to,

inter alia, prevent the quality of the air from deteriorating

in areas that are in attainment with the NAAQS. 42

U.S.C. § 7470. The NSPS program is based solely on the

particular type of equipment or facility emitting the

pollutants, whereas the PSD program establishes control

requirements on a case-by-case basis taking into account

site-specific factors such as the specific environmental

impact a new or modified source will have upon the area

where it will be located. See 57 Fed. Reg. 32,314, 32,315-

316 (July 21, 1992). See also Northern Plains Resource

Council, 645 F.2d at 1356. EPA reinforced these differences

by promulgating different tests for triggering PSD and

22

NSPS requirements: “In contrast [to the NSPS hourly

emissions rate test and] in light of the air quality planning

component of the NSR program, the NSR regulations

examine total actual annual emissions to the atmosphere.

Consequently, normal operations over a period of time is

considered for purposes of determining a source’s impact

on ambient air.” 57 Fed. Reg. at 32,331. Only an annual

emissions test will fulfill the purpose of the PSD program

to prevent backsliding in regions where the air quality is

adequate.

To prevent facilities from significantly deteriorating

the air quality, PSD statutory and regulatory provisions, in

conjunction with the SIPs, establish and assure compliance

with “increments” or “maximum allowable increases” of

pollutants. 42 U.S.C. § 7475(a), (d), 7473(a), (b); 40 C.FR.

§ 51.166(c). An owner or operator of a facility seeking to make

modifications must demonstrate that “emissions from con-

struction or operation of such facility will not cause, or con-

tribute to, air pollution in excess of” these increments. 42

U.S.C. § 7475(a\(3). The PSD permitting process is the mecha-

nism under which EPA examines appropriate consumption of

these increments. 42 U.S.C. § 7475(a)(2). Since an hourly test

does not capture the total actual emissions increases result-

ing from a change, EPA will not be able to accurately quan-

tify adverse impacts on air quality when doling out

preconstruction permits in PSD areas. The preamble to

EPA's 1980 regulations supports this analysis:

Source applicants ... must ... perform an analysis

to compute how much of the PSD increment re-

mains available to them. In general the amount of

increment that is available depends on certain

changes in actual emissions.... Accordingly,

cleanup adds to the available growth margin while

23

new emissions diminish it. ... Once the question

of how much increment remains is resolved, then

the applicant must demonstrate that his pro-

posed new emissions would not exceed the re-

maining PSD increment. Where a_ proposed

project would cause a new violation of the incre-

ment or contribute to an existing violation, it

cannot be approved. Existing violations must be

entirely corrected before PSD sources which af-

fect the area can be approved.

45 Fed. Reg. 52,676, 52,678 (Aug. 7, 1980).

Recognizing that not all emissions increases should

trigger the PSD requirements, EPA established “signifi-

cant net emissions increase” thresholds calculated in “tons

per year.”” 40 C.F.R. § 51.166(b)(23)i) (emphasis added).

Before establishing these thresholds, EPA analyzed the

“cumulative effect on increment consumption of multiple

sources in an area each making the maximum de minimis

emissions increase (thereby going unreviewed under PSD

at the time of the change).” 45 Fed. Reg. 52,676, 52,707

(Aug. 7, 1980). EPA understood that if too many unac-

counted for insignificant modifications occurred, EPA

would not be able to enforce the emission increments

through the PSD permitting requirements. Yet, this is

exactly what will happen if the Fourth Circuit’s hourly test

is sanctioned. Projects that do not increase hourly emis-

sions but do significantly increase hours of operation, and,

in turn, annual emissions, would escape PSD review.

Unless EPA is able to consider actual annual emissions

increases, including emissions resulting from increases in

These significance thresholds are the fourth step in the “major

modification” definition. See supra p. 15.

24

hours of operation, the emissions inventory will be se-

verely disrupted, and the PSD program will not function

properly.

D. Industry, Including Duke, Was Well Aware

that the Test for PSD Was Annual Emis-

sions, Not Hourly Emissions.

Following industry’s initial challenge to the 1980 PSD

regulations, a settlement agreement ensued, whereby EPA

agreed to propose a regulatory amendment that would

allow sources to “calculate emissions ... based on either

the actual emissions methodology in the existing rules or

the unit’s potential emissions, measured in terms of hourly

emissions.” 61 Fed. Reg. 38,250, 38,268 (July 23, 1996)

(discussing and proposing language from 1982 Settlement

Agreement). In other words, by entering into this settle-

ment agreement, EPA and industry effectively agreed that

the 1980 rules already included projects that increase

actual annual emissions but do not increase potential

hourly emissions and that EPA would propose and take

public comment on an amendment to exclude such pro-

jects. If industry believed that such projects were already

excluded from the 1980 PSD regulations, as they allege in

th.: case, then there would be no incentive for industry,

including Duke, to settle in 1982 based on EPA's agree-

ment to propose their desired test for public comment.

In 1996, EPA proposed for public comment the lan-

guage agreed to in the 1982 Settlement Agreement:

“sources may calculate emissions increases and decreases

... Measured in terms of hourly emissions (i.e., pounds of

pollutant per hour).” 61 Fed. Reg. at 38,268. EPA noted

that “[t)he primary effect of an hourly potential test is to

eliminate a source's level of operations as a factor when

25

determining whether a proposed change will result in an

increase. Past and future level of utilization of the source

are completely disregarded.” Jd. at 38,269. After a prelimi-

nary analysis of the proposed rule, EPA concluded that:

[T]here is concern for the environmental conse-

quences. For example, assume the emissions unit

at the widget factory that is emitting 10 pounds

an hour but has historically operated at 40 per-

cent capacity due at first to operating cost, but

with age, reduced efficiency and reliability. Un-

der the [hourly potential test], the owner could

modernize the unit ... [and] allow the owner to

use the machine at much higher levels (e.g., more

hours per day or week) than it had in the past.

As a result actual emissions (measured in tpy

- [tons per year]) could more than double due to

the increase in utilization even though hourly po-

tential emissions remain the same.... [O]ne of

the most troubling side effects of the . . . proposal

is that it could ultimately stymie major new

source growth by allowing unreviewed increases

of emissions from modifications of existing

sources to consume all available increment in

PSD areas.... (U)nder the ... [proposal] an old

grandfathered source could experience a ‘signifi-

cant’ net increase in annual actual emissions, yet

it would not necessarily be subject to review.

Id. at 38,269, 38,270.

In 2002, after receiving comments on the hourly

emissions test for triggering PSD review, EPA rejected the

test. 67 Fed. Reg. 80,186, 80,205 (Dec. 31, 2002). EPA

reasoned that an hourly test

26

could sanction greater actual emissions increases

to the environment ... without any preconstruc-

tion review. In addition, actual emissions increases

resulting from unreviewed projects could go largely

undocumented until a PSD review is performed by

a new or modified facility that ultimately must un-

dergo review. By that time, however, a violation of

an increment could have unknowingly occurred. .. .

We agree that ... [the proposed hourly test] could

lead to unreviewed increases in emissions that

would be detrimental to air quality and could make

it difficult to implement the statutory require-

ments for state-of-the-art controls.

Id.

Although EPA's latest proposed PSD regulations depart

from the lengthy historic consistency described above, it only

further illustrates that the regulations applicable in this case

provide for an actual annual emissions test. In October, 2005,

EPA requested public comment on amendments to the PSD

program that would allow existing power plants, such as

Duke’s, “to use the same maximum achievable hourly |

emissions test we apply under NSPS to determine whether a

physical change in or change in the method of operation .. .

results in an emissions increase.” 70 Fed. Reg. 61,081, 61,088

(Oct. 20, 2005). Indeed, EPA states in the 2005 proposed rule

that “EPA entered into a Settlement Agreement which

required us to propose an NSPS-like, hourly-potential-to-

hourly-potential emissions increase test for modification,”

and that the current PSD rules require “a source [to] look{]

at whether a project will result in a significant emissions

increase on an annual basis.” Jd. EPA asks for comment on

this proposed rule a minimum of 25 times in the 22-page

document; EPA recognizes that the proposed rule is a

27

stark departure from the previous long-standing actual

annual emissions test.”

EPA’s use of an actual annual emissions test, rather

than an hourly potential test, should come as absolutely no

surprise to Duke. EPA was quite capable of writing an hourly

emissions test if it so desired, as it plainly did for the NSPS

program 30 years ago and as it has only now proposed for the

PSD program. If industry and EPA believed that an hourly

rather than annual test already existed, the 1982 settlement,

the resulting 1996 proposal, the 2002 rejection of that

proposal, and the subsequent 2005 proposal would not be

necessary.

e

CONCLUSION

The decision of the Fourth Circuit Court of Appeals

should be reversed in its entirety.

Dated: July 21, 2006

Respectfully submitted,

HOLLY D. GORDON*

DEBORAH A. SIVAS

STANFORD LAW SCHOOL

ENVIRONMENTAL LAW CLINIC

Crown Quadrangle

559 Nathan Abbott Way

Stanford, California 94305-8610

Telephone: (650) 725-8571

Facsimile: (650) 723-4426

*Counsel of Record

“ The 2005 proposal clearly states that “|t}he proposed rule would

only apply prospectively.” 70 Fed. Reg. at 61,081 (Uct. 20, 2005).