Amicus Curiae Brief — Environmental Defense v. Duke Energy Corporation

4

a petition to review the validity of EPA’s regulations, based on

interpretation of the CAA. See 42 U.S.C. § 7607(b). As New

Jersey and other states expiain in their amicus brief, in this

enforcement action, the Fourth Circuit’s opinion exceeded the

jurisdictional limits of § 7607(b).

The Fourth Circuit justified this overreach by a clearly

erroneous interpretation of applicable EPA regulations. At issue

is the meaning of the term “modification,” as used in the CAA’s

PSD provisions’ — specifically, whether that term includes

projects that, though they do not result in any increase in the

hourly rate of emissions during operation, will result in increases

in the hours of operation of a plant. EPA has adopted regulations

that directly address this interpretive question; they clearly

provide that a project that increases a unit’s hours of operation

but not its maximum hourly rate of emissions may still be a

“modification” for PSD purposes where it will lead to increased

annual emissions. The Fourth Circuit, in holding that such a

project is not a “modification,” mandated use of a distinct test

set forth in EPA’s regulations governing a different regulatory

program, the New Source Performance Standards (“NSPS”).

That test ignores any increases in hours of operation and

considers only an increase in the maximum hourly rate of

emissions during operations. (EPA refers to this test as an “hourly

potential emissions” test.) In mandating this test for purposes

of the PSD program, the court effectively — and improperly —

invalidated EPA’s interpretation of the CAA in its PSD

regulations, a course only open to the D.C. Circuit.

3. The PSD » ovisions of the CAA, 42 U.S.C. §§ 7470-92, cover

areas that have already attained the National Ambicnt Air Quality

Standards ("NAAQS"). Although only the PSD requirements are at issue

in this case, the Fourth Circuit's decision also applies to areas that have

not yet attained the NAAQS, subject to the nonattainment New Source

Review (“NNSR”) provisions of the CAA, 42 U.S.C. § 7503. The PSD

and nonattainment NSR programs are together referred to as the New

Source Review (“NSR”) Program.

5

The Fourth Circuit, recognizing that it lacked jurisdiction

to invalidate the PSD regulations, protested that its adoption of

an hourly potential emissions test was “not an invalidation of

those regulations.” Duke, 411 F.3d at 549 n. 7. In a conclusory

analysis that does not address, let alone analyze, the language

of the PSD regulations and is relegated to a footnote, the court

justified this assertion by holding that those regulations “can”

be interpreted consistently with an hourly potential emissions

test. /d. That conclusion flies in the face of the plain language

of the PSD regulations, which, on their face, require an increase

in actual annual emissions. Under the regulations, as common

sense would suggest, actual annual emissions may increase

either because of an increase in the hourly rate of emissions

during operation or because of an increase in the hours of

operation. Moreover, the PSD regulations’ focus on actual

annual emissions reflected a deliberate decision by EPA, when

it revised its PSD regulations, to abandon a potential emissions

test virtually identical to the emissions test under the NSPS

regulations. In the course of notice-and-comment rulemaking,

EPA explicitly concluded that such a potential emissions test

should not be used for the PSD program.

Because the plain language of the PSD regulations mandates

an actual annual emissions test, thus foreclosing the Fourth

Circuit’s preferred hourly potential emissions test, the court was

required to end its analysis there; no direct analysis of the CAA

was necessary. Even if it had been proper, though, for the court

to look beyond the plain text of the PSD regulations, EPA’s

application of the PSD regulations below is consistent with the

relevant statutory language and congressional intent. The hourly

potential emissions test that the lower court adopted is a test

designed to determine whether there will be a change in the

emissions that would potentially be emitted if a plant operated

at maximum emissions capacity. The test looks only at the

theoretical maximum rate of emissions during operation; it

ignores hours of operation. But this is contrary to the statutory

definition of “modification” under 42 U.S.C. § 7411(a)(4), as

6

incorporated in the PSD provisions. The CAA is clear that PSD

requirements are triggered by physical changes resulting in

increases in “actual” emissions of a source, as opposed to

“potential” emissions. See New York v. EPA (“New York I”),

413 F.3d 3, 19-20, 39-40 (D.C. Cir. 2005). Furthermore, in

enacting the PSD provisions of the CAA in 1977, Congress

clearly expressed its concern with actual annual emissions

increases, not changes in hourly potential emissions.

ARGUMENT

I. The Plain Language Of The PSD Regulations Requires

Increases In Actual Annual Emissions, Not Hourly

Potential Emissions.

In this EPA enforcement action, the Fourth Circuit has

jurisdiction to determine only whether EPA is properly

interpreting and applying the applicable PSD regulations. As

explained by petitioners, 42 U.S.C. § 7607(b) grants the D.C.

Circuit exclusive jurisdiction over any challenge to EPA’s

interpretation of the CAA in its regulations, and thus the validity

of those regulations.’ Despite this jurisdictional limitation, the

Fourth Circuit found the hourly potential emissions test

applicable to Duke’s projects based on its analysis of the

language and legislative history of the CAA and conclusion that

Congress had mandated that the emissions tests under the NSPS

and PSD regulatory programs must be the same. Duke, 411 F.3d

at 546-51. The court’s exercise of jurisdiction to interpret the

4. To the extent that EPA regulations interpreting the CAA “could

have been” challenged in the D.C. Circuit, such regulations “shall not

be subject to judicial review in civil or criminal proceedings for

enforcement.” 42 U.S.C. § 7607(b)(2); see also Wis. Elec. Power Co. v.

Reilly (*WEPCo”), 893 F.2d 901, 914, n. 6 (7" Cir. 1990) (review limited

by § 7607(b)(1) to “whether the EPA properly applied [the] regulations”

at issue); United States v. Ethyl Corp., 761 F.2d 1153, 1157 (S" Cir.

1985) (“section 307(b)’s clear language and Congress's manifest intent

[was] to avoid protracted and inconsistent adjudications over the

validity” of nationally applicable CAA regulations), cert. denied, 474

U.S. 1070 (1986).

7

CAA was premised on its assumption that the PSD regulations

“can” be interpreted to institute a test based solely on hourly

potential emissions. Based on that premise, the court concluded

that its decision was “not an invalidation of those regulations.”

Id. at 548 n. 7.

In support of this critical determination that the PSD

regulations can be interpreted to direct an hourly potential

emissions test, the Fourth Circuit conducted no analysis of the

language of those regulations or the history of their adoption.

An evaluation of these issues demonstrates that the court’s

interpretation cannot stand, and that EPA’s interpretation and

application of its regulations is correet.

A. The PSD Regulatory Emissions Test Cannot Be

Interpreted As An Hourly Potential Emissions Test.

The enforcement action in United States v. Ohio Edison,

Co., 278 F. Supp. 2d 829 (S.D. Ohio 2003), provides a vivid

illustration of how a power plant project that increases hours of

operation may substantially increase the plant’s actual annual

emissions, even if the project does not increase its hourly

emissions capability. The defendant in that case undertook boiler

component replacements — similar to those at issue below — to

increase the hours of operation of the units. /d. at 858. Persistent

boiler tube leaks at one such unit caused it to be shut-down a

total of 2,134 hours per year. The court found that replacing the

failed boiler components triggered PSD permitting requirements

because the increased hours of operation projected by the

defendant to result from the replacement would yield 5,200 tons

per year of increased SO, emissions, id. at 869-70,° which is

5. Similarly, the D.C. Circuit, using the facts in Puerto Rican

Cement Co. v. EPA, 889 F.2d 292 (1* Cir. 1989), has aptly illustrated

this point:

In that case, a factory sought to . . . replace old cement

kilns that operated 60% of the time with a new kiln that

would emit fewer pollutants per hour. . . . Under the

(Cont'd)

8

well above the 40 ton per year emissions increase threshold,

see 40 C.F.R. § 51.166(b)(23)(1). The boiler component

replacements at issue here, like those in Ohio Edison, were

intended to increase the future hours of operation of the plants

without increasing their hourly emission rate, Duke, 411 F.3d

at 544, resulting in projected increases of actual annual

emissions. As explained below, the PSD regulations clearly reach

such increases.

1. Unlike the NSPS Regulations, the 1980 PSD

Regulations Require Emissions Increases to Be

Based on Changes in Actual Annual Emissions.

As other courts have correctly held, the plain language of

the 1980 r-gulations, as well as the context in which they were

adopted, directly contradicts the Fourth Circuit’s determination

that the PSD regulations can and must be “interpreted” to require

an hourly potential test. First, the PSD regulations, by their plain

language, require a finding of an increase in actual annual

emissions, whether because of an increase in hours of operation

or because of an increase in the hourly rate of emissions during

operation. The PSD pre-construction requirements apply, inter

alia, to “any major modification” which, in turn, is defined as a

physical change that “would result in a significant net emissions

increase... .” 40-C.F.R. § 51.166(b)(2)(4), (4)(2), G)(3).

The definition of “net emissions increase” denotes the positive

sum of any increase in “actual emissions” at the unit resulting

from the physical or operational change and any other

(Cont'd)

interpretation urged by industry petitioners, . . . the company

had not undergone an ‘increase’ in emissions — and thus

would not trigger NSR — since the kiln would have a lower

hourly emissions rate than the old ones. Siding with EPA,

the First Circuit agreed that the company had to obtain an

NSR permit to make the intended change [because of the

projected increase in annual emissions resulting from the

change].

New York 1, 413 F.3d at 15.

9

contemporaneous decreases or increases in “actual emissions”

at other units within the source. /d. § 51.166(b)(3). The

regulations make clear that actual emissions are measured on

an annual, not hourly, basis. For the pre-change baseline period,

the regulations define “‘actual emissions” to mean “the average

rate, in tons per year, at which the unit actually emitted the

pollutant.” /d. § 51.166(b)(21)(ii) (emphasis added). Likewise,

the calculation of post-change emissions is also based on “the

actual rate of emissions from the unit.”° Jd. § 51.166(b)(21)(i).

If the net emissions increase is positive, the next step is to

determine if the emissions increase is “significant,” defined

under the regulations as “a rate of emissions that would equal

or exceed” different thresholds of pollutant emissions, once

again measured in “tons per year.” Jd. § 51.166(b)(23)(i).

The language of the PSD regulations contrasts starkly with

the language of the NSPS regulations. The NSPS emissions

test does not measure either actual or annual emissions increases.

The regulatory term “modification,” which triggers the NSPS

requirements, is defined in terms of an increase in an “emission

rate,” id. § 60.14(a), measured in kilograms of pollutant

emitted per hour at maximum unit capacity, as specified in

§ 60.14(b)(2).” The NSPS emissions test is thus a paper and

6. Under 40 C.F.R. § 51.166(b)(21)(iv) (1987), “[flor any

emissions unit which has not begun normal source operations on the

particular date, actual emissions shall equal the potential to emit of the

unit on that date.” Thus conversely, where the unit has begun normal

operations at the time of the physical or operational change, the unit’s

potential to emit cannot be used. See 56 Fed. Reg. 27,630, 27,633/1

(June 14, 1991) (“linchpin . . . for predicting future emissions after a

modification is thus whether the unit has ‘begun normal operations” ).

Under these circumstances, § 51.166(b)(21)(i) (1987) — which defines

“actual emissions” — is applicable. See WEPCo, 893 F.2d at 917 (finding

post-change emissions for unit that has begun normal operations based

on “realistic assessment of [a change’s] impact on ambient air quality

levels”) (internal quotation marks omitted).

7. This brief references the 1975 NSPS regulations, which first

employed the hourly potential emissions test.

10

pencil exercise that computes changes in hypothetical hourly

emissions at maximum unit capacity — regardless of whether

the unit was ever actually operated at that capacity, with both

the historical and projected utilization of the unit irrelevant.

Accordingly, EPA has consistently described the NSPS

emissions test as a calculation of changes in hourly potential

emissions: “[P]ursuant to longstanding EPA interpretations, the

emission rate before and after a physical change is evaluated at

each unit by comparing the hourly potential emissions under

current maximum capacity to emissions at maximum capacity

after the change.” Joint Appendix (“J.A.”) 259 (Applicability

of PSD and NSPS Requirements to WEPCO Life Extension

Project, Sept. 9, 1988, p. 9) (emphasis added); see also WEPCo,

893 F.2d at 913 (“EPA compares the hourly emissions of the

unit at its current maximum capacity to its potential emissions

at maximum capacity after the change” (emphasis added));

Requirements for Preparation, Adoption and Submittal of

Implementation Plans, 57 Fed. Reg. 32,314, 32,330/2 (EPA July

21, 1992) (“Under current regulations, the emissions rate before

and after a physical or operational change is evaluated at each

unit by comparing the current hourly potential emissions at

maximum operating capacity to hourly emissions at maximum

capacity after the change.”’).*

8. Even the regulated industry understood that a modification

can occur under the PSD regulafions absent an increase in hourly

emissions capacity. In their original challenge to the 1980 PSD

regulations, industry petitioners interpreted the regulations in the same

manner as EPA does here, stating that (i) the regulations “provide that a

modification occurs whenever a series of contemporaneous changes at

a source result in a significant increase in actual emissions, even though

the source's net capacity to emit remains constant or declines,” and (11)

the “question of whether there is a net increase in actual emissions

requires that variations in the source’s hours and rates of operation be

taken into account.” See Brief for Industry Petitioners at 5-6, 28-29,

Chem. Mfrs. Ass'n v. EPA, No. 79-1112 (D.C. Circ. Feb. 12, 1981):

New York 1, 413 F.3d at 14-15 (original challenge to 1980 PSD

(Cont'd)

Second, the Fourth Circuit’s assumption that the NSPS and

PSD regulatory emissions tests “can” be read identically

disregards the circumstances of EPA’s adoption of the actual

annual emissions test under the 1980 PSD regulations,

demonstrating that EPA intended and understood the PSD test

to be different from the NSPS hourly potential test. Significantly,

when EPA adopted the actual annual emissions test, it abandoned

a capacity-based test, similar to the NSPS emissions test, that it

had used in an earlier version of the PSD program in effect

between 1978-80. Unlike the 1980 regulations at issue here,

the 1978 PSD regulations defined “major modification” as any

physical or operational change “which increases the potential

emission rate of any air pollutant regulated under the act” by

either 100 or 25U tons per year (depending on source category).

43 Fed. Reg. 26,380, 26,403-04 (June 19, 1978) (emphasis

added).’ EPA defined “potential to emit” in the 1978 regulations

as “the capability at maximum capacity to emit a pollutant in

(Cont'd)

regulations stayed), see also Settlement Agreement, Ex. B at 1, Chemical

Mfrs. Ass'n v. EPA, No. 79-1112 (D.C. Cir. Feb. 22, 1982) (settlement

between industrial polluters, including Duke, and EPA, requiring EPA

to propose changing PSD emissions test to allow emissions increases

to be based on “source's potential to emit (as calculated in terms of

pounds of pollutant emitted per hour)’).

9 EPA interpreted “modification” in its 1974 PSD regulations to

be “consistent with” the NSPS regulatory definition of that term. 39

Fed. Reg. 42,510, 42,513/1 (Dec. 5, 1974) (interpreting 40 C.F.R.

§ 52.01(d)). The NSPS definition in effect at that time, however, applied

simply to “increases [in] the amount of any air pollutant,” with no

reference to hourly emission rates. 36 Fed. Reg. 15,704, 15,705/1 (Aug.

17, 1971); Duke, 411 F.3d at 542. The 1974 PSD regulatory definition

of “modification” was subsequently superseded and made moot by the

PSD emissions test adopted by EPA in 1978 to administer the 1977

CAA amendments. Ala. Power Co. v. Costle, 636 F.2d 323, 348 n. 24

(D.C. Cir. 1979); see also Envtl. Defense Fund, Inc. v. EPA, 898 F.2d

183, 184 n. | (D.C. Cir. 1990) (explaining history of PSD regulations’

promulgation).

12

the absence of air pollution control equipment.” /d. at 26,404/1

(emphasis added). In Alabama Power, however, the D.C. Circuit

vacated that definition of “major modification” under the 1978

PSD regulations, 636 F.2d at 399-400, finding that the related

PSD statutory definition of “major emitting facility” under 42

U.S.C. § 7479(1) requires the measurement of emissions to be

based on “some measure of actual emissions.” /d. at 353.

EPA promulgated the 1980 regulations “[i]n response to”

Alabama Power. 45 Fed. Reg. 52,676/1 (Aug. 7, 1980). EPA

re-examined the statutory definition of “modification”

incorporated into the PSD provisions, finding that the words

“the amount of any air pollutant emitted by such source or which

results in the emissions of any air pollutant not previously

emitted’. . . appear to refer to what the source is actually emitting

at a particular time.” /d. at 52,700/2; see also New York I, 413

F.3d at 40 (EPA in “the preamble of the 1980 rule” interpreted

$ 7411(a)(4) to require that emissions increases be based on

actual emissions). EPA also found Alabama Power's discussion

of the meaning of “major emitting facility” controlling in this

respect, noting that the Court “used language which, like the

section |11(a)(4) definition, suggest changes in actual

emissions.” 45 Fed. Reg. at 52,700/2. Thus, “[flollowing the

lead” of Alabama Power, EPA “shifted the focus of its regulatory

definitions from” a capacity-based, potential emissions test,

similar to the test the Fourth Circuit decision requires, to the

test based on actual annual emissions that is now reflected in

the plain language of the PSD regulations. /d. at 52,700/3.

Because, for purposes of the PSD program, EPA explicitly

rejected a potential emissions test like the one used for the NSPS

program in favor of an actual annual emissions approach, the

PSD regulations cannot be interpreted to provide for an hourly

potential test. Cf County of Los Angeles v. Shalala, 192 F.3d

1005, 1014 n. 2 (D.C. Cir. 1999) (“adding and deleting certain

words” gives amended provision “different meaning from its

predecessor provision”), cert. denied, 530 U.S. 1204 (2000);

DIRECTV, Inc. v. Brown, 371 F.3d 814, 817 (11th Cir. 2004)

13

(“[w]here the words of a later statute differ from those of a

previous one on the same or related subject, then Congress must

have intended them to have a different meaning” (quoting

Muscogee (Creek) Nation v. Hodel, 851 F.2d 1439, 1444 (D.C.

Cir. 1988))).

Accordingly, the Fourth Circuit stands alone as the only

circuit court to find that the NSPS and PSD emissions tests

must or even can be interpreted identically. In New York /, Duke

and other industry petitioners argued that “the 1980 regulation

provided that an emissions ‘increase’ occurs only if the

maximum hourly emissions rate goes up as a result of the

physical or operational change.” 413 F.3d at 15. Noting that

Duke’s test could only be instituted if the existing PSD

regulations were invalidated, the D.C. Circuit reyected industry’s

contention that EPA must and can interpret the 1980 PSD

regulations so that only a project that increases a plant’s hourly

emissions at maximum unit capacity constitutes a modification.

See id. at 19-20."° See also New York v. EPA (“New York IT’),

443 F.3d 880, 889 (D.C. Cir. 2006) (rejecting industry’s

“rel[iance] on NSPS regime to reargue their position that

‘modifications’ [under PSD] require an increase in maximum

emission rates”); WEPCo, 893 F.2d at 913, 915 (finding that

NSPS and PSD regulations “measure[] emissions in a

fundamentally distinct manner,” with the NSPS test “concerned

primarily with . . . emissions rates, expressed in kilograms per

hour,” and the PSD test “concerned with changes in total annual

emissions, expressed in tons per year”); Puerto Rican Cement,

889 F.2d at 298 (PSD requirements triggered by change allowing

plant to operate more hours even though hourly emissions

10. The D.C. Circuit’s decision carries special weight in these

matters because it has exclusive jurisdiction to review EPA’s “nationally

applicable” air pollution regulations. 42 U.S.C. § 7607(b)(1). Moreover,

no party has petitioned this Court for certiorari in New York 1, meaning

that the validity of the emissions test under the 1980 PSD regulations

can no longer be questioned.

14

remain unchanged or reduced).'' The other circuits have

correctly interpreted the PSD regulations.

2. The Fourth Circuit’s Conclusion that the NSPS

and PSD Regulations Regarding Modifications

Must Be Interpreted Identically Ignores

Numerous Differences Between the Two

Regulatory Programs.

Any attempt to read the NSPS and PSD regulations as

consistent would require papering over of numerous obvious

differences between the two programs. One important difference

between the two emissions tests, for example, is that the PSD

test allows netting (or “bubbling”) of emissions increases and

decreases across units within a source in determining whether

the modification requirements are triggered, while the NSPS

test does not allow such netting. Compare 40 C.F.R.

§ 51.166(b)(2\(i), (3) (“major modification” defined in terms

of change resulting in “net emissions increase”), with id.

§ 60.14(a) (“modification” defined in terms of change resulting

in “increase in the emission rate” at unit only). As explained in

Chevron U.S.A., Inc. v. Natural Res. Def. Council, 467 U.S.

837, 840, 860 (1984) (in nonattainment New Source Review,

or “NNSR,” context) and Alabama Power, 636 F.2d at 400-401

(in PSD context), EPA based these divergent approaches on

different interpretations of the single statutory definition of

“stationary source,” under 42 U.S.C. § 741 1(a)(3), which ts a

component term of the statutory definition of “modification,”

id § 741 1(a)(4) (“term ‘modification’ means any physical change

in... a stationary source which increases the amount of any air

_ In Puerto Rican Cement, EPA compared ™ the actual historical

amount or pollutants that Kilns 3 and 6 emitted . . . with the amount of

pollutants that the [single] converted kiln would be capabie of emitting

in the future.” 889 F.2d at 296 (emphasis added). The First Circuit upheld

EPA's application of the actual-to-potential emissions test there based

on its finding that the converted kiln at issue constituted a “new”

emissions unit that had not begun normal operations. /d. at 292, 296-

97: see also WEPCo, 893 F.2d at 917 n.12 (distinguishing Puerto Rican

Cement on this basis).

15

pollutant emitted by such source . . .””) (emphasis added). EPA

applied these different interpretations of “stationary source” to

distinguish “between the technology-based provisions of NSPS

and the air quality-based provisions of [PSD].” See 57 Fed. Reg.

at 32,316/1.

At the time the PSD regulations at issue here were adopted

in 1980, there were at least three other key differences between

the two programs: (1) While the PSD regulations require

examination of a representative two-year period in determining

how the emissions baseline is measured, the NSPS baseline is

based on the maximum achievable emission rate immediately

preceding the change; (2) While the PSD emissions test requires

an increase above an annual “significan[ce]” threshold, NSPS

requirements are triggered by any increase in maximum

achievable emissions; and (3) The NSPS, but not PSD,

regulations exempt pollution control projects from the

modification requirements.'* See 40 C.F.R. § 51.166(b)(2)(1),

(21 (ii), (23); 40 C.F.R. § 60.14(e)(5), (h); WEPCo, 893 F.2d at

914-15.

Additionally, the Fourth Circuit finding that EPA can and

must adopt identical NSPS and PSD definitions of

“modification” is both flatly inconsistent with and unworkable

in light of the decisions of this Court in Chevron and the D.C.

Circuit in Alabama Power, both of which approve an approach

to modifications under NNSR and PSD that differs from that

used in the NSPS program. Those decisions upheld EPA’s

interpretation of “stationary source” to allow for netting in

determining whether a source has been modified under NSR

but not under NSPS. See Chevron, 467 U.S. at 863-64 (allowing

netting in determining what is a modification under NNSR

programs, stating that “agency primarily responsible for

administering this important legislation has consistently

interpreted it flexibly — not in a sterile textual vacuum, but in

12. EPA’s pollution control project exclusion in the 2002 PSD rule

changes was vacated by the D.C. Circuit as being beyond EPA’s authority.

New York 1, 413 F.3d at 40-42.

16

the context of implementing policy decisions in a technical and

complex arena”); Alabama Power, 636 F.2d at 402 (“bubble

concept would be contrary to the intent of the NSPS provisions,

but such is clearly not the case with regard to the PSD

provisions”); see also ASARCO Inc. v. EPA, 578 F.2d 319 (D.C.

Cir. 1978) (prohibiting netting under NSPS program). Indeed,

the holding in Alabama Power was also based on “two possible

ways to construe the term ‘increases’” — one of the component

terms of § 741 1(a)(4) at issue here; one that allows for netting

and one that does not. 636 F.2d at 401."

Thus, under Chevron, a power plant unit is not considered

to be modified under PSD and NNSR if actual annual emission

increases resulting from a physical change at the unit are offset

by emission decreases elsewhere at the plant, but the same unit

may still be considered modified under NSPS (where netting is

unavailable) if the unit increases its potential-to-emit. While

Chevron and Alabama Power authorize EPA to interpret

“modification” to allow for netting under PSD but not NSPS to

implement the different purposes of the programs, the Fourth

Circuit’s decision now requires EPA to interpret “modification”

identically under the two programs. The Fourth Circuit’s holding

directly conflicts with the holdings in Chevron and Alabama

Power.

B. The Other Arguments Relied Upon By the Fourth

Circuit are Erroneous.

The Fourth Circuit failed to explain how the PSD

regulations could be construed identically with the NSPS ~

regulations, despite the clear differences between them. Instead,

relegating its entire analysis of the meaning of the regulations

to a partial sentence in a footnote, the Fourth Circuit explained

that “the PSD regulations can be interpreted consistently with

13. Of course, as discussed infra, pp. 22-24, as used in the PSD

context, the phrase “increases the amount of any air pollutant emitted

by such source,” § 741 1(a)(4), must be interpreted as requiring “actual,”

as opposed to “potential,” emission increases.

17

pre-existing principles — the NSPS regulations — as the district

court demonstrated and as EPA’s Director of the Division of

Stationary Source Enforcement twice opined shortly after

promulgation of the PSD regulations.” See 411 F.3d at 549 n. 7.

However, the arguments set forth in the district court’s decision

provide no basis for this construction of the PSD regulation.

First, the notion that Congress incorporated the NSPS

regulatory definition of “modification” into the PSD provisions

when it amended the CAA in 1977, see United States v. Duke

Energy Corp., 278 F. Supp.2d 619, 629 (M.D.N.C. 2003), is

erroneous. This issue bears on the validity of the 1980 PSD

regulations, not their interpretation, and thus under 42 U.S.C. §

7607(b)(2) could only have been, and indeed was, raised in the

D.C. Circuit in New York I. Moreover, as the D.C. Circuit

correctly found in New York J, there are no “indications in the

statutory language or history to infer that Congress intended to

incorperate” the NSPS regulatory definition of “modification”

into the PSD provisions. 413 F.3d at 19-20; see also Demarest

v. Manspeaker, 498 U.S. 184, 190 (1991) (“administrative

interpretation of statute” not incorporated into reenacted statute

where “[t]here is no indication that Congress was aware of

administrative construction”); cf, Bragdon v. Abbott, 524 U.S.

624, 632 (1998) (finding statutory incorporation of regulation

where Congress “adopted a specific statutory provision” that

expressly adopted an existing regulaiory provision); FDIC v.

Philadelphia Gear Corp., 476 U.S. 426, 437 (1986) (finding

FDIC regulation defining statutory term “deposit” incorporated

into reenacted statute where regulation had been in existence

for 50 years and Congress had “expressly designed to incorporate

the FDIC’s rules and regulations on ‘deposits””).

Second, the exemption for increases in hours of

operation from what constitutes a physical change, 40 C.F.R.

§ 51.166(0)(2)(11i)(f), in no way precludes EPA from considering

increases in hours of operation in calculating emissions increases

attributable to a physical change. See Duke, 278 F. Supp.2d at

641. By its terms, § 51.166(b)(2)(iii)(f) only precludes EPA from

18

finding that an “increase in the hours of operation” alone

constitutes a physical change; it has no application, however,

to situations where — as in this case — there is a physical change

such as a power plant renovation that enables a source to operate

more hours. As explained in WEPCo, EPA recognized in the

preamble to the 1980 PSD rules that the “hours of operation”

exemption was “provided to allow facilities to take advantage

of fluctuating market conditions, not construction or

modification activity.” 893 F.2d at 916 n. 11 (citing 45 Fed.

Reg. at 52,704/2). Thus, EPA’s contemporaneous interpretation

of the “hours of operation” exemption is that, while an increase

in such hours by itself does not constitute a “modification,” the

exemption does not apply to actual construction that allows a

source to increase its hours of operation. Accordingly, WEPCo

found that “EPA’s refusal to apply ‘[this] production rate/hours

of operation’ exclusion was proper” because the exclusion did

not cover the “modification” activity at issue in the case.'* /d.;

see also Puerto Rican Cement, 889 F.2d at 298 (“hours of

operation” exemption inapplicable to determining emission

increase attributable to physical change).

Third, the statements of a single agency employee made in

a letter to an industry source and in an internal memorandum

cannot support an interpretation of the PSD regulation as

requiring application of the NSPS-hourly emissions test.

See 411 F.3d at 549 n. 7; see also Duke, 278 F. Supp.2d at 641.'°

14. WEPCo does not require PSD emissions increases to be based

on changes in “the maximum hourly rate of emissions.” Cf. 278 F.

Supp.2d at 644-45. WEPCo holds that the NSPS and PSD programs

“measure|] emissions in a fundamentally distinct manner” and that EPA

may consider increased hours of operation in determining whether a

physical change will increase emissions, see 893 F.2d at 913, 916 n. 11.

1S. It is far from clear that the documents cited by the district

court, and thus implicitly relied upon by the Fourth Circuit, actually

support this proposition. For_cxample, the district court misconstrued a

January 22, 1981 memorandum for a finding that “increased hours of

(Cont'd)

19

This conclusion is flatly inconsistent with the regulations’ plain

language, EPA’s express abandonment of a potential emissions

test in the 1980 regulations, and EPA’s contemporaneous

interpretation as set forth in the regulatory preamble.

See Cinergy, 384 F. Supp.2d at 1278 (memoranda “are contrary

to the plain language of the CAA and EPA’s regulations”); Ohio

Edison, 276 F. Supp.2d at 877 (reliance on memoranda “‘is

misplaced because they are contrary to the plain language of

[CAA]”); cf. Puerto Rican Cement, 889 F.2d at 298 (rejecting

similar argument where “EPA materials written both before,

and after, the deviant letter are consistent with [EPA’s] present

interpretation’’).

Reliance on these memoranda is also contrary to EPA’s

consistent interpretation of the 1980 PSD regulations as

requiring an actual annual emissions test that considers increases

in hours of operation attributable to a physical change. Indeed,

as discussed supra at 10-11 n. 8, at the time the two documents

were written, EPA was defending against an industry lawsuit,

challenging the regulations on the grounds that they did not

include a maximum hourly rate test and had entered into a

settlement agreement requiring it to propose changing the PSD

(Cont'd)

operation, even when coupled with a physical or operational change,

would not be considered a modification.” Duke, 278 F. Supp.2d at 641-

42. The actual memorandum, however, considered whether a plant

converting to a new fuel must undergo PSD permitting. with EPA’s

determination that the “fuel switch” was exempt from being a “physical

change” under 40 C.F.R. § 51.166(b)(2)(iii Ne). See Ohio Edison, 276

F. Supp.2d at 877. Thus, as the court in Ohio Edison explained. absent

a physical change, the source was allowed to increase hours 0! —

without triggering PSD requirements. /d. Additionally, the cou

a June 24, 1981 applicability letter as follows: “*PSD applicability

is determined by evaluating any change in the [hourly] emissions

rates caused by’ the physical or operational change being examined.”

Duke, 278 F. Supp.2d at 641. Notably, the word “hourly” is not in the

text (see JA28), and the specified emission rate can alternatively refer

to the annual rate required under PSD.

recited

20

rules back to the pre-1980 approach. See also J.A. 255

(Applicability of PSD to the WEPCO Life Extension Project,

Sept. 9, 1988, p. 6 (“exclusion for increases in hours of operation

... does not take the project beyond the reach of PSD coverage

if those increases do not stand alone but rather are associated

with non-excluded physical or operational changes”)); Letter

from EPA to Ohio Edison (Jan. 1990) (“post-demonstration

emissions increase attributable solely to an increase in the hours

of operation . . . of the unit” does not trigger PSD “to extent that

the emissions increase did not result from a physical or operation

change”), quoted in Ohio Edison, 276 F. Supp.2d at 876; 57

Fed. Reg. at 32,328/1 (“increase in emissions attributable to an

increase in hours of operation or production rate which is the

result of a construction-related activity is not excluded from

review’); 61 Fed. Reg. 38, 250, 38,269/2 (July 23, 1996)

(expressing concern with industry request to change test to one

based on increases in hourly emissions as allowing source

“owner to use the [operating unit] at much higher levels

(e.g., more hours per day or week) than it had in the past” and

thus “increase . . . utilization even though hourly potential

emissions remain the same.’’).

Moreover, as a matter of law, a single agency official cannot

change the plain meaning and EPA’s contemporaneous

interpretation of the 1980 PSD regulations absent notice-and-

comment rulemaking required to make such a change.

See Paralyzed Veterans of Am. v. D.C. Arena L.P., 117 F.3d

579, 586 (D.C. Cir. 1997) (“Once an agency gives its regulation

an interpretation, it can only change that interpretation as it

would formally modify the regulation itself: through the process

of notice and comment rulemaking.”), cert. denied sub nom.,

Pollin v. Paralyzed Veterans of Am., 523 U.S. 1003 (1998). The

States are in charge of implementing the PSD program through

their SIPs and rely in this regard on the language of EPA’s

regulations and official agency statements made in rulemaking

documents; they simply cannot be bound by a “secret” agency

“law” memonialized in a letter to a single industry source or in

a

21

an EPA internal memorandum. See Train v. NRDC, 421 U.S.

60, 87 (1975) (deferring to EPA’s official interpretation of statute

where “there has undoubtedly been reliance upon its

interpretation by the States”).

Finally, there are post-1992 claims at issue in this case and

no analysis that the Fourth Circuit has relied upon can establish

that the applicable 1992 PSD regulations direct an hourly

potential emissions test. There is simply no way that anyone

could have come away from EPA’s 1992 rulemaking proceeding

believing that the PSD and NSPS emissions tests could be

interpreted identically. EPA reiterated in the regulatory preamble

that, with respect to determining emissions increases, (i) “the

applicable [NSPS and PSD] rules branch apart,” with

“{e]missions increases for NSPS purposes . . . determined by

changes in the hourly emissions rates at maximum physical

capacity” and emissions increases under PSD determined by

changes in “total emissions to the atmosphere . . . expressed in

tons per year,” and (11) “an increase in emissions attributable to

an increase in hours of operation or production rate which is

the result of a construction-related activity is not excluded from

review (see WEPCo, 893 F.2d at 916 n. 11; Puerto Rican

Cement, 889 F.2d at 298).” 57 Fed. Reg. at 32,316/1, 32,328/1.

EPA also specified that power plant operators are required to

consider increases in utilization in performing an emissions

increase analysis. See 40 C.F.R. § 51.166(b)(33) (“representative

actual annual emissions” defined as “average rate, in tons per

year, at which the source is projected to emit a pollutant . . .,

considering the effect any such change will have on increasing

or decreasing the hourly emissions rate and on projected capacity

utilization” (emphasis added)).'°

16. The district court found erroneously that Duke “opted out” of

the 1992 regulatory requirements “by failing to satisfy the regulatory

prerequisite of submitting emissions data for a five-year period following

the physical change.” Duke, 278 F. Supp.2d at 647 n. 25. There is no

provision in the regulations that allows Duke or any other utility to

(Cont'd)

22

In sum, the Fourth Circuit ruling is inconsistent with the

plain language of the 1980 PSD regulations, the context in which

they were promulgated, including the EPA’s contemporaneous

interpretation, and EPA’s consistent interpretation of the

regulations for the last 26 years.

Il. The PSD Regulations Requiring Emission Increases To

Be Based On Changes In Actual Annual Emissions

Effectuate Congressional Intent.

Contrary to the Fourth Circuit’s erroneous conclusion, 41 |

F.3d at 548-51, the statutory language and legislative history of

the CAA establish that the PSD regulatory emissions test

correctly implements the plain language of the PSD provisions.

A. The Statutory Definition of “Modification”

Requires Actual Emissions Increases.

The actual annual emissions test under the PSD regulations

follows directly from the plain language of the statutory

definition of modification as applied in the PSD context.

Congress defined “modification” in the 1970 CAA amendments

as:

[A]ny 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. § 7411(a)(4) (emphasis added).

In New York I, the D.C. Circuit employed “traditional tools

of statutory interpretation” in correctly finding “that the CAA

unambiguously defines ‘increases’ [under § 741 1(a)(4)] in terms

(Cont'd)

opt out of ‘regulatory requirements. Instead, as specified under

§ 51.166(b)(21 iv) and (v), EPA may assume that a source owner that fails

to meet the emissions record-keeping requirement of § 51.166(b)(33)

was emitting at maximum capacity following the change. Despite this, EPA

is applying the more lenient actual-to-projected-actual test in this case.

23

of ‘actual emissions’” — not “potential” or “allowable”

emissions. 413 F.3d at 39."’ The court’s ruling gave effect to the

plain meaning of the term “emitted,” which, without a modifying

phrase such as “potentially,” refers to pollution that a source

has actually generated. See Webster’s Third New International

Dictionary 742 (3d ed. 1967) (defining “emit” as, inter alia, to

“send out: discharge, release”). Similarly, the term “amount,”

as used in the statutory definition of modification, signifies

Congress’s concern with the actual quantity of pollutants emitted

rather than with the source’s emissions “rate” or “potential.”

See New York I, 413 F.3d at 40 (“phrase ‘the amount of any air

pollutant emitted by [the] source’ plainly refers to actual

emissions”). Thus, analysis of the ordinary meaning of these

terms demonstrates that “increases [in] the amount of any air

pollutant emitted” means increases in the actual emissions of

the source, not some theoretical measure of changes in plant

capacity.

This interpretation of the statutory definition of

“modification” is compelling when examined in the context of

the PSD provisions of the CAA considered as a whole. Indeed,

the other uses of the terms “emitted” and “emission” in the latter

part of the definition of “modification” also refer to actual

emissions: “. . . or which results in the emission of any air

pollutant not previously emitted.” 42 U.S.C. § 7411(a)(4). This

language would make no sense if “emitted” and “emission”

referred to potential emissions; by using the term “previously”

to characterize “emitted,” Congress clearly was referring to

actual emissions. Moreover, it is unlikely that Congress could

have intended the single word “emitted” to have different

meanings within the same sentence.

17. Applying this interpretation, the court vacated an exemption

based on “clean unit” status under the 2002 NSR regulatory changes.

This exemption would have allowed sources that installed pollution

controls or instituted certain work practices to avoid PSD permitting

for 10 years, even if the source undertook a physical change that

significantly increased actual emissions during that time period. 413

F.3d at 38-40.

24

That Congress meant the term “emit” to refer to actual

emissions is also demonstrated by Congress’s “juxtaposition

of the terms ‘emit’ and ‘potential to emit,’” as part of the

definition of “major emitting facility,” 42 U.S.C. § 7479(1),

which “indicates that when Congress enacted the NSR program

in 1977, it was conscious of the distinction between actual and

potential emissions.” New York J, 413 F.3d at 39; see also

Alabama Power, 636 F.2d at 353 (holding that the term “emit”

in the definition of “major emitting facility,” 42 U.S.C.

§ 7479(1), refers to “some measure of actual emissions” rather

than potential emissions). Likewise, the juxtaposition of the

terms “emission limitation” and “emitted” in the context of the

definition of “best availabie control technology” under 42 U.S.C.

§ 7479(3), shows that Congress understood the difference

between those terms, with the latter referring exclusively to

“actual” emissions. New York J, 413 F.3d at 39. As the D.C.

Circuit explained, “[i]f Congress had intended for ‘increases’

in emissions to be measured in terms of potential or allowable

emissions, it would have added a reference to ‘potential to emit’

or ‘emission limitations.’ The absence of such a reference must

be given effect.” /d. at 40; see also Alabama Power, 636 F.2d at

354, 355 (legislative history of the 1977 CAA amendments

indicates that “Congress was concerned with large industrial []

major actual emitters of air pollution” and “that only major

sources of actual emissions would be covered by the PSD permit

requirements” (emphasis added)).

Thus, the common understanding of “emitted” and

Congress’s use of that term under other PSD provisions supports

an actual emissions test. Accordingly, the D.C. Circuit’s reading

of § 7411(a)(4) in New York J, requiring changes in “actual”

emissions as the event that triggers modification, is the only

natural reading of the statutory definition of “modification” in

the context of the PSD provisions. See Engine Mfrs. Ass'n v.

S. Coast Air Quality Mgmt. Dist., 541 U.S. 246, 252 (2004);

Bluewater Network v. EPA, 370 F.3d 1, 13 (D.C. Cir. 2004).

25

B. The Actual Annual Emissions Test Under the PSD

Regulations Furthers Congressional Emphasis on

Actual Air Quality.

The actual annual emissions test under the PSD and NNSR

regulations also plays an integral role in furthering Congressional

intent to achieve and maintain levels of air quality that protect

public health and welfare. See, e.g., 42 U.S.C. § 7470(1). Prior

to 1977, many areas of the country were failing to attain the

NAAQS; for example, 78 areas still were not in compliance

with the ozone standard. S. Rep. No. 101-228, at 10 (1989), as

reprinted in 1990 U.S.C.C.A.N. 3385, 3396; see also id. at 11,

as reprinted in 1990 U.S.C.C.A.N. at 3397 (even as of 1989,

150 million people were living in areas failing to meet NAAQS

for one or more pollutants). As with later failures to meet CAA

deadlines, the “paper demonstrations of attainment” in some

SIPs bore “little relation to the likelihood of actual attainment”

and, in conjunction with the NSPS and mobile source emissions

controls required under the 1970 CAA amendments, failed to

translate into sufficient emissions reductions to achieve

attainment. /d., as reprinted in 1990 U.S.C.C.A.N. at 3397.

Congress intended the 1977 amendments, which included the

NSR provisions, to address the insufficient progress being made

toward clean air. See, e.g, Senate ENv'T AND PuBLic Works

Como., S. Rep. No. 95-127, at 12 (May 10, 1977) (“several

thousand major industrratsources will not meet the deadlines

of the act”).

The NSR provisions added in 1977 reflect Congress's focus

on actual air quality. In particular, the NSR provisions are

concemed with maintaining (in the case of PSD) and improving

(in the case of NNSR) actual air quality in areas where new or

modified plants are located. The PSD provisions thus require a

new or modified source to obtain a pre-construction permit,

demonstrating that it will not cause a violation of an air quality

“increment” (designed to prevent air quality from deteriorating

significantly), as well as to install emission controls that

represent BACT. 42 U.S.C. § 7475. The NNSR provisions

26

require a new or modified source to obtain a pre-construction

permit, obtain emissions offsets so that total “actual” emissions

in the nonattainment area are reduced when the source begins

operation, and comply with the Lowest Achievable Emission

Rate (LAER). /d. § 7503.'* EPA’s determination under NSR to

treat actual annual emission increases as the triggering

mechanism for “modifications” thus promotes Congressional

intent to focus on protecting and improving actual air quality.

One of Congress’s primary mechanisms for ensuring

continued industrial development without harming air quality

is the increment consumption analysis a new or modified source

must undertake before obtaining a PSD permit. A new or

modified source’s emissions violate a region’s “increment” if

its projected pollutant emissions, when added to emissions

increases and decreases at other sources in the same area, will

exceed “maximum allowable” pollution above a “baseline

concentration” level. See id. §§ 7473, 7475(a)(3)(A); 45 Fed.

Reg. at 52,718/1-2. This analysis is based on the measurement

of “actual” air quality conditions. The D.C. Circuit in Alabama

Power found that the statutory definition of “baseline

concentration,” 42 U.S.C. § 7479(4), requires the measurement

of “actual ambient pollution levels existing at the time of the

first [PSD] permit application.” 636 F.2d at 376 (emphasis

added); see also id. (“baseline is to include all emissions actually

being made by major facilities”) (emphasis added).'’ Consistent

with the statutory language, EPA decided in its 1980 PSD

18. Non-attainment SIPs must include “a comprehensive, accurate,

current inventory of actual emissions from all sources.” 42 U.S.C.

§ 7502(c\3) (emphasis added).

19. An earlier version of the House Bill premised the calculation

of the “baseline concentration” of pollutants on “plant capacity in

existence.” Houst Comm. ON INTERSTATE AND FortiGn Commerce, Report

95-295, at 397 (to accompany H.R. 6161, 95" Cong. 1* Sess. 397) (May

12, 1977). However, as noted in Alabama Power, the House Bill's

reference to plant capacity was deleted from the enacted version of the

1977 CAA amendments. See 636 F.2d at 380-81.

27

regulations to use “actual source emissions” in determining the

baseline concentration in order to “reflect actual air quality in

an area.” 45 Fed. Reg. at 52,714/2-3, 52,718/1.

In adopting the 1980 regulations, EPA followed the same

“actual emissions” approach to determining the amount of

allowable increment that remains available to a proposed major

source or modification: “[T]he most reasonable approach,

consistent with the statute, is to use actual source emissions . .

. to calculate increment consumption or expansion.” 45 Fed.

Reg. at 52,718/1. In so finding, EPA explained that “[i]ncrement

calculations based on the best prediction of actual emissions

links PSD permitting more closely to actual air quality

deterioration than calculations of allowable ‘paper’ emissions.”

Id. EPA emphasized the need to maintain consistency with the

actual emissions test for modifications: “[U]se of actual

emissions for increment consumption is consistent with using

an actual emissions baseline for defining a major modification.”

Id.*°

The NSR regulatory emissions trigger for modifications is

also consistent with the statutory focus on annual emissions as

the basis for NSR applicability. For example, the PSD

requirements apply to “major emitting facilities,” defined in

terms of sources that emit pollutants above a threshold measured

in “tons per year.” 42 U.S.C. §§ 7475, 7479(1). For power plants,

the PSD requirements govern all new sources with emissions

above 100 tons per year. /d. § 7479(1). Similarly, 42 U.S.C. §

7475(b) provides an exception to one of the PSD pre-

construction requirements that is applicable “. . . in the case of

an expansion or modification of a major emitting facility . . .

whose allowable emissions of air pollutants . . will be less than

20. A new or modified source in an attainment area is also required

to undertake “an analysis . . . of the ambient air quality at the proposed

site . . . for each pollutant subject to regulation under the Act which will

be emitted from such facility.” 42 U.S.C. § 7475S(e). The court in

Alabama Power found that Congress intended this requirement to ensure

“a realistic projection of air quality.” 636 F.2d at 372.

28

fifty tons per year.” Id. § 7475(b) (emphasis added). Further,

each of the PSD increments over the baseline concentration 1s

based on an “annual geometric mean” concentration in addition

to 24-hour and 3-hour maximum concentrations. /d. § 7473(b)

(emphasis added). Finally, the requirement that non-attainment

SIPs provide “reasonable further progress” toward attainment

is based on “annual incremental reductions in emissions of the

relevant air pollutant.” /d. § 7501(1). These provisions provide

strong evidence that Congress intended the NSR provisions to

apply to physical changes that increase annual emissions of

pollutants.

The actual annual emissions test also furthers Congress’s

intent to promote responsible industrial growth in a manner that

prevents deterioration of air quality. Congress believed that

subjecting existing sources to control requirements when they

are modified would reduce the pollution burden of existing

sources, furthering Congressional intent to allow continued

industrial development without harming air quality. See id.

§ 7470(3) (Congressional intent “to insure that economic growth

will occur in a manner consistent with the preservation of

existing clean air resources”); House COMM. ON INTERSTATE AND

FoREIGN Commerce, Report 95-295, at 133 (to accompany H.R.

6161, 95" Cong. 1“ Sess. 133) (May 12, 1977) (“if each new or

modified major source is located, constructed, and operated so

as to minimize its impact on available clean air resources, then

more and bigger plants will be able to locate in the same area

without serious air quality degradation”). The Fourth Circuit

decision undermines this careful economic/environmental

balance established by Congress under PSD because it allows

physical changes that increase actual annual emissions which,

in turn, may consume the PSD increment and thereby limit room

for economic growth. See WEPCo, 893 F.2d at 909 (“too

restrictive interpretation of ‘modification’ might upset the

economic-environmenta! balance” established under PSD).

In sum, Congress intended the PSD program to be built

upon a structure of actual annual emissions, from measuring

ambient pollution levels, to determining whether a source

is “major” or triggers the “modification” requirements, to

implementing the various mechanisms for ensuring compliance

with the NAAQS. The Fourth Circuit’s decision should therefore

be reversed.

CONCLUSION

For the reasons set forth above and in the petition, the Court

should reverse the Fourth Circuit decision.

Peter H. LEHNER

Chief, Environmental

Protection Bureau

RosBert ROSENTHAL

J. JaRED SNYDER

Micuaet J. Myers

Assistant Attorneys General

* Counsel of Record

Respectfully submitted,

E.iot Spitzer

Attorney General of the

State of New York

CAITLIN J. HALLIGAN*

Solicitor General

ANDREW BING

Danie. J. CHEPAITIS

Assistant Solicitors General

120 Broapway

New York, NY 10271

(212) 416-8016

Bit. Lockyer

Attorney General

State of California

CALIFORNIA DEPARTMENT

OF JUSTICE

1300 | Street

P.O.Box 944255

Sacramento, CA 94244

RICHARD BLUMENTHAL

Attorney General

State of Connecticut

P.O. Box 120

Hartrorb, CT 06141-0120

Cart C. DANBERG

Attorney General

State of Delaware

DeLAWARE DePARTMENT

OF JUSTICE

Carvel State Building

820 N. French Street

Wilmington, DE 19801

Lisa MADIGAN

Attorney General

State of Illinois

188 West Randolph Street

Suite 2001

Chicago, IL 60601

Tuomas J. MiLLer

Attorney General

State of lowa

1305 E. Walnut Street

Des Moines, IA 50319

30

G. Steven Rowe

Attorney General

State of Maine

6 State House Station

August, ME 04333-0006

J. JosepH Curran, Jr.

Attorney General

State of Maryland

200 St. Paul Place

Baltimore, MD 21202

THOMAS F. REILLY

Attorney General

Commonwealth

of Massachusetts

200 Portland Street

3" Floor

Boston MA 02114

Mike Hatcu

Attorney General

State of Minnesota

MINNESOTA ATTORNEY

GENERAL’S OFFICE

900 Bremer Tower

445 Minnesota St.

St. Paul, MN 55101

Ketty AYOTTE

Attorney General

State of New Hampshire

OFFice OF ATTORNEY

GENERAL

33 Capitol Street

Concord, NH 03301

Patricia A. MAprRip

Attorney General

State of New Mexico

P.O. Drawer 1508

Santa Fe, NM 87504-1508

(S05) 827-6939

HARDY MYERS

Attorney General

State of Oregon

1162 Court St. N.E.

Salem, OR 97301

Patricia LYNCH

Attorney General

State of Rhode Island

150 South Main Street

Providence, RI 02903

WILLIAM H. Sorrece

Attorney General

State of Vermont

109 State Street

Montpelier, VT 05609-1001

SUSAN SHINKMAN

Chief Counsel

Rosert A. REeiLey

Assistant Counsel,

Commonwealth of Pennsylvania,

DEPARTMENT OF ENVIRONMENTAL

PROTECTION

400 Market Street

Harrisburg, PA 17101-2301

Attorneys for Amici Curiae