Opposition Brief — Environmental Defense v. Duke Energy Corporation

public health and welfare with an adequate margin of safety.

42 U.S.C. §7409.7 The states, in turn, develop State

Implementation Plans (“SIPs’’), setting source-by-source

emissions limits to meet the NAAQS. /d. §74100 These

limits are premised on a permitted source operating at its

maximum emitting capacity (/.e., all year at maximum

production capacity) without exceeding the NAAQS.°

B. New Source Programs.

I. The New Source Performance Standards

(“NSPS”) program did not regulate existing

emissions capacity.

The 1970 Amendments also directed EPA to issue NSPS to

minimize the impact of adding new emissions capacity.’

Congress debated whether to subject existing emissions

capacity to NSPS, but chose not to do so." Thus, NSPS apply

only to new sources—i.e., newly-constructed emission units

or “modifications” of existing units. /d. §7411(a)(2) (“new

source” is one on which “construction or modification ...is

commenced after” a given date). Modification is defined 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.” /d.

* See also Train vy. NRDC, 421 U.S. 60, 66-67 (1975).

* See also Union Elec. Co. vy. EPA, 427 U.S. 246, 249-50 (1976).

" See, e.g., Cleveland Elec. IMluminating Co. v. EPA, 572 F.2d 1150,

1160 (6th Cir. 1978) (“The model |used to calculate required SIP limits] ts

operated on the assumption that the plants concerned operate 24 hours a

day at full capacity and predictions are made for every day of the year.”).

“The legislative history of §111 [NSPS] ... reveals that Congress was

most concerned that new plants—new sources of pollution-— would have

to be controlled to the greatest degree practicable if the national goal of a

cleaner environment was to be achieved.” Essex Chem. Corp. vy.

Ruckelshaus, 486 F.2d 427, 434 n.14 (D.C. Cir. 1973).

* See S. Rep. No. 91-1196, at 15-17 (1970), reprinted in Legal Compi-

lation: Statutes and Legislative History, EPA (1973) (noting the high cost

and difficulty of installing new control technologies on existing sources).

§741 1(a)(4).

The NSPS regulatory § definition of “modification”

promulgated in 1971 tracked the statutory definition almost

verbatim. 36 Fed. Reg. 24,876, 24,877 (Dec. 23, 1971). The

rules further specified that “[rJoutine maintenance, repair, and

replacement shall not be considered physical changes,” and a

“change in the method of operation” does not include: (1) an

“increase in the production rate ... [within the facility's]

operating design capacity;” (2) an “increase in hours of

operation,” or (3) the “use of an alternative fuel” that the

facility was “designed to accommodate” before a given date.

Id. (codifying 40 C.F.R. §60.2(h)). Thus, under these rules,

only activities that increase a facility’s design emitting

capacity triggered NSPS.”

In 1975, EPA revised the NSPS rules to “clarify the phrase

... ‘increases the amount of any air pollutant’” in the statutory

and general regulatory (40 C.F.R. §60.2) definition of

“modification.” EPA stated this phrase refers to an increase

“in emission rate,” which the regulations “expressed as

kg/hr.” 40 Fed. Reg. 58,416, 58,419 (Dec. 16, 1975). As

EPA explained, the use of this measurement unit clarified that

the modification rule would be sensitive to “increased

production capacity and to overall increases in total emissions

to the atmosphere” (/.e., to new capacity to emit), while

“automatically allowing] increases in operating hours as

intended by one of the existing exemptions under 40 C.F.R.

§60.2(h).” 39 Fed. Reg. 36,946, 36,947 (Oct. 15, 1974)

(emphasis added). Thus, EPA recognized, the existing

exclusion in the NSPS rule for increased “hours of operation”

by definition meant that only activity that increased the

facility’s existing hourly emissions rate—i.e., its actual

emitting capacity—is a “modification” under §111 of the Act.

” The NSPS provisions were “designed to prevent new air pollution

problems.” National Asphalt Pavement Ass'n v. Train, 539 F.2d 775, 783

(D.C. Cir. 1976); see also H.R. Rep. No. 1146, at 3 (1970), reprinted in

1970 U.S.C.C.A.N. 5356, 5358.

i)

2. The New Source Review (“NSR”) programs

adopted the NSPS concept of “modification.”

a. Congress adopted the NSPS “modification”

concept into the 1977 statutory NSR program.

In 1977, Congress further amended the CAA by enacting

two more new source programs: the PSD and Non-

Attainment New Source Review (NNSR) programs.'”

Basically, the 1977 Amendments codified a then-existing

1974 regulatory PSD program.'' In doing so, Congress

specifically considered the various provisions of the

regulatory program and decided to “tighten[]” some of them.

See Ala. Power, 636 F.2d at 349-50. Congress did not,

however, change the applicability provisions of these

regulations (which were the same as NSPS); it simply

incorporated the NSPS definition of modification into the

NSR definition of modification.

Thus, Congress defined “modification” under NNSR to

“mean the same as the term ‘modification’ as used in section

7411(ay(4) [NSPS].” 42 U.S.C. §7501(4) (emphasis added).

"PSD applies to all new emissions capacity in areas meeting the

NAAQS (attainment areas), while NNSR applies to all new emissions

capacity in areas not meeting the NAAQS (non-attainment areas). Here.

EPA had alleged violations of PSD because the areas in North and South

Carolina where Duke's plants operate were in attainment for all relevant

pollutants during the periods of interest. This brief refers to the NSPS,

PSD, and NNSR programs collectively as the “New Source programs.” It

refers to the PSD and NNSR programs together as “NSR.”

'' 42 U.S.C. §7478(a) (“Until such time as an applicable implemen-

tation plan is in effect ... which ... meets the requirements of this part ...

applicable regulations under this chapter prior to August 7, 1977 shall

remain in effect ... except as otherwise provided in subsection (b)"). Prior

to Congress’ amending the Act to include NSR requirements, EPA—

prompted by a court's preliminary injunction—had promulgated rules

adopting a PSD program. See generally Ala. Power Co. v. Costle, 636

F.2d 323, 346-49 (D.C. Cir. 1979) (describing the statutory and regulatory

history of PSD). EPA specified that the “general definition of modified

source” in its PSD rules was meant to be “consistent with the definition

used in Part 60 [/.e., NSPS]. 39 Fed. Reg. 42,510, 42.513 (Dec. 5, 1974).

As originally enacted, the PSD program reierred only to the

“construction” of major sources, and not to “modification” as

did the NNSR program. Three months later, Congress fixed

this oversight by adding “modification” as defined under

NSPS to the “construction” activity subject to PSD. See id.

§7479(2(C) (“The term ‘construction’ ... includes the

modification (as defined in section 7411(a) [NSPS] ...).”).

In support of this “technical amendment,” Congress

explained that it intended to incorporate the “usage” of the

NSPS definition of “modification.”'? EPA has repeatedly

acknowledged that “[t}he phrase ‘usage in other parts of the

Act’ ... refers, not only to section 111(a)(4), but also to the

EPA regulations implementing section 111] that were in effect

at that time.””

b. The 1980 PSD rules incorporated the NSPS

concept of “modification.”

EPA’s 1978 PSD rules defined “major modification” as a

“physical change [or] change in the method of operation”

which “increases the potential emission rate” (in tons per

year) of regulated pollutants from a “major stationary source.”

'? 123 Cong.Rec. 36331 (1977) (technical amendments “{ijmplement| |

conference agreement to cover ‘modification’ as well as ‘construction’ by

defining ‘construction’ in Part C to conform to usage in other parts of the

Act.”). Congress emphasized it was “not the purpose of these [technical]

amendments to re-open substantive issues in the [CAA].” 123 Cong.Rec.

36252 (1977). Congress had resolved the substantive issue of the

applicability of the new statutory NSR programs to existing units by

leaving the applicability provisions of the 1974 PSD regulations

unchanged and providing that “modification” under NSR is the same “as

used” under the pre-existing NSPS program.

' 49 Fed. Reg. 43.211, 43,213 (Oct. 26, 1984) ; see also 43 Fed. Reg.

26,380, 26,394 (Jun. 19, 1978); 68 Fed. Reg. 61,248, 61,269 (Oct. 27,

2003) (1977 CAA “refer|red] to our preexisting rules interpreting the term

‘modification’ in the NSPS context.”); Pet. App. 35a-36a. See also

Braydon v. Abbott, 524 U.S. 624, 631 (1998) Congress’ repetition of a

well-established term carries the implication that Congress intended the

term to be construed in accordance with pre-existing regulatory

interpretations.) (emphasis added).

43 Fed. Reg. at 26,403 (emphasis added). Because these rules

assumed the same “potential” hours of operation before and

after the change, only a change that increased an existing

unit’s hourly emissions rate (/.e., an NSPS “modification”)

could trigger NSR. Virtually all of the 1978 rules, except that

aspect of the definition of “major modification,” were

challenged. See Ala. Power Co., 636 F.2d 323. In

implementing the court’s remand requiring a “bubble”

concept for defining “source” under PSD,"* the final 1980

rules defined “major modification” as “any physical [or

operational} change ... of a major stationary source that would

result in a significant net emissions increase” in the source’s

“actual emissions” (in tons per year)."”

For there to be a “major modification,” the 1980 PSD rules

require that an increase in “actual emissions” in tons per year

result from the proposed change. For existing units that have

begun normal operations, “actual emissions” is defined by

reference to “actual operating hours [and] production rate” for

a historic (pre-change) period “which is representative of

normal source operation.” /d. §51.166(b)(21)(ii). That is, for

these units, only one period of “representative” conditions is

required to characterize those “normal source operations.”

Accordingly, post-project “actual emissions” must be

calculated using the same “representative” conditions of

operation (/.e., hours of operation) that existed before the

change. This is confirmed by the definition of “modification,”

which states that the “increase” in actual emissions must result

from the “physical [or operational] change.” which by

definition excludes “[a]n increase in the hours of operation or

'* Under the “bubble” concept, the “source” subject to NSR is an entire

plant, which may include several “emissions units.” By contrast, each

individual emissions unit is a separate “source” under NSPS.

" 40 C.P.R. §§51.166(b 21), (b3)i), (b21). The NSR rules

applicable here are the PSD rules adopted in the state SIPs, as approved by

EPA. The applicability provisions of the state rules are identical to the

federal 1980 PSD rules, 40 C.F.R. §51.166 (1987). All references to the

1980 PSD rules are to the 1987 C.F.R., except where otherwise noted.

in the production rate” (unless restricted by permit). /d.

§51.166(b)(2)(iii)(f) (emphasis added).'® Thus, as under all of

the prior modification rules, a change at an existing unit

would result in an increase in “actual emissions” on/y if it first

increases the unit’s hourly emissions rate—i.¢., only if itis a

“modification” under NSPS, 42 U.S.C. §741 1(a)(4)."”

By contrast, for units that have not begun normal operations

(e.g., new units or completely reconstructed units), the 1980

rules require a different assumption for post-change operating

hours: since the unit has no operating history, one must

assume the full “potential to emit” (8,760 hours, unless hours

were limited). /d. §51.166(b)(21)(iv). This different test for

units that have not begun “normal operation” is known as the

“actual-to-potential” test.

3. Implementation of the PSD program.

a. EPA’s contemporaneous interpretation of

“modification” in the PSD rules is consistent

with the NSPS definition of “modification.”

Shortly after finalizing the 1980 PSD rules, EPA confirmed

that the definition of “net emissions increase” meant just what

it said. See Pet. App. 58a-59a. In 1981, EPA confirmed that

“PSD applicability [at a previously operating unit] is

determined by evaluating any change in [hourly] emissions

rates caused by” the physical or operational change under

review. Letter of Reich, EPA, to Gill, General Elec. (June 24,

1981) (“GE Determination”), CA App. 234-41. Absent a

change in the per-hour emission rate, EPA concluded, actual

emissions “could increase only if there is an increase in tlie

The 1980 rules’ preamble explains, “The first step ... is to determine

whether the physical or operational change in question would i/se/f result

in an increase in ‘actual emissions.”” 45 Fed. Reg. 52,676, 52,698 (Aug.

7, 1980) (emphasis added). “|T]hose phrases do not encompass ... any

increase in the hours or rate of operation of a source, so long as the

increase would require no change in any ... permit condition.” /d.

'” Pet. App. 58a-60a: see WEPCo, 893 F.2d at 918 0.14.

production rate or hours of operation, both of which are

specifically exempt from PSD review.” Id. (citing 40 C.F.R.

§52.21(b)(2)(iii)(f))."" Therefore, the unit's production rate

and hours of operation are kept constant in calculating an

“emission increase” under the “actual-to-actual” test required

by the 1980 PSD rules. See 40 C.F.R. §51.166(b)(2)(iii(f).”

b. The WEPCo court rejected EPA’s attempt to

reinterpret the 1980 rules.

Only once in the 30-year history of new source programs

before EPA’s new “enforcement initiative” against coal-fired

utilities did EPA determine that a power plant repair and

replacement project would trigger PSD and NSPS.~” In 1988,

Wisconsin Electric Power Company (“WEPCo”) proposed a

“renovation” project at its Port Washington Plant, which had

operated over 10 years at a maximum achievable capacity

well below its original capacity. Mem. from Kee, EPA

Region V, to Emison, EPA (Mar. 25, 1988), CA App. 256-

57... EPA found the replacements WEPCo proposed to

increase “|p|lantwide capacity ... about 40 percent above

current levels to 400 megawatts,” id., “highly unusual, if not

unprecedented.” Initial WEPCo Determination, CA App. 807.

Given both the magnitude and uniqueness of these

replacements, EPA determined WEPCo’s project was not

'* This determination was consistent with a previous one dated January

22, 1981. Letter of Reich, EPA, to Whitmore, EPA Region VII Ulan. 22

1981), CA App. 242-43 (“Cargill Determination”). In both

determinations, physical and operational changes were being made to the

units; these were not mere stand-alone increases in operating hours.

” 40 CER. §51,166(b)( 211i F) is identical to §52.21(b\2 iii Mf) that

the GE Determination cites. See also Pet. App. 4lan.11.

° See Mem. from Clay, EPA, to Kee, EPA Region V (Sept. 9, 1988)

(“Initial WEPCo Determination”), CA App. 804-15; Letter trom Thomas,

EPA Adm’'r, to Boston, WEPCo (Oct. 14, 1988), CA App. 248-55; Letter

from Clay, EPA, to Boston, WEPCo (Feb. 15, 1989). CA App. 830-41.

*' Several years before the proposed project, the WEPCo units had been

formally “derated™ from their full capacity of 80 MW cach to operating

capacities ranging from 49 MW to 78 MW. CA App. 256.

10

“routine maintenance, repair and replacement” within the

exclusion provided under the NSPS rules and incorporated

into the NSR rules. /d. The Seventh Circuit upheld that

determination. WEPCo, 893 F.2d at 910-13.

Regarding emissions increase, EPA found an NSPS

emissions increase at three WEPCo units. EPA, however,

also found a PSD emissions increase at all five units, by

comparing their actual annual emissions during a pre-repair

representative period to their theoretical, maximum annual

emissions (/.¢., “potential to emit”), derived by multiplying

the units’ maximum emissions rates by a full year (8,760

hours )}—the “‘actual-to-potential” test.

The court upheld EPA’s NSPS finding, but reversed EPA’s

PSD determination. Under the rules, the “potential to emit”

concept plainly applies only to a “unit which has not begun

normal operations.” 40 C.F.R. §51.166(b)(21 iv). Refusing

to “defer to agency interpretations that ... appear to assume

what they seek to prove,” the court rejected EPA’s circular

reasoning—that, “in order to demonstrate that [WEPCo’s]

like-kind replacement project constitutes a modification,”

EPA “applies the potential to emit concept (to show an

increase in emissions),” and, “in order to apply the potential to

emit concept to like-kind replacement, the EPA assumes that

the plant is a ‘modified’ unit.” WEPCo, 893 F.2d at 917.

On remand, the court ordered EPA to determine “whether

the renovated plant would cause a significant net emissions

increase if it were operated under present hours and

conditions,” id. at 918 n.14 (emphasis added)—i.e., calculate

post-project emissions on the basis of the restored hourly

emissions rate of the repaired plant under the actual hours and

conditions of operation during the representative baseline

period. By holding pre- and post-project hours and conditions

constant at a level representative of normal operations, any

“net emissions increase” could be due only to an increase in

the units’ hourly rates, not a result of more hours of operation.

Thus, the court confirmed the rules’ actual-to-actual emissions

test promulgated in 1980 and applied by EPA in 1981.7>

c. EPA ignored the court’s remand instructions

in settling WEPCo.

On remand, EPA was precluded from applying the actual-

to-potential test for the WEPCo units that had “begun normal

operations.” But EPA refused to follow the court’s remand

instructions—that EPA determine on the basis of “present

hours and conditions” whether WEPCo’s project would result

in an annual emissions increase—summarily dismissing them

as “incorrect.””* Instead, for settlement purposes and with no

reference to the regulatory language whatsoever, EPA

devised an emissions calculation methodology comparing the

units’ representative baseline annual emissions (/.e., pre-

project emissions) to their “estimated future actual emis-

sions.” Remand WEPCo Determination, CA App. 271-72. In

the Initial WEPCo Determination, when WEPCo suggested

such a test as an alternative argument, EPA categorically

responded: “The [1980] PSD regulations provide no support

for” an actual-to-projected-actual test. CA App. 810 n.4.

In the 1992 “WEPCo Rule.” EPA promulgated a rule

containing the new actual-to-projected-actual methodology for

which the 1980 rules “provide no support.” Significantly,

EPA did not change the “base” 1980 rules governing

emissions increase, and EPA made this new 1992 “actual-to-

* After WEPCo, EPA abandoned its re-interpretation of the rule that

the Seventh Circuit had rejected, explaining that “|alt least for changes

that are ‘like kind replacements,” ‘normal operations’ have begun, and the

actual-to-potential test is impermissible” 57 Ved. Reg. 32,314, 32.317

(Jul. 21, 1992) (emphasis added). Yet, here, EPA returned in its summary

judgment papers to this discredited interpretation, only to abandon 1 again

at the summary judgment hearing. See Pet. App. 58a n.17.

** Letter. from Rosenberg, EPA. to Boston, WEPCo Gun. 8, 1990)

(“Remand WEPCo Determination”), CA App. 265-78.

* fd. at 270. Although EPA asked for an extension of time for tiling a

motion for clarification or reconsideration of the court's remand, sce Foote

Affidavit in WEPCo (CA App. 279-82), it never filed it, choosing instead

simply to ignore the court's clear instructions. See a/so Pet. App. 70a.

12

projected-actual”™ test available only to utilities that elect to

undertake additional reporting. 40 C.F.R. §§51.166(b)(21)(v)

& (b)(32) (1993).*° Otherwise, the activity would be eval-

uated using the emissions increase test of the 1980 rules.”°

4. EPA Reverses Course In 1999.

Throughout the last three decades, the utility industry's

maintenance practices—including component replacements

sometimes referred to as “life extension”—were a matter of

public knowledge. Before 1999, except for the unusual and

unprecedented WEPCo project, no federal or state agency, or

any environmental group for that matter, ever suggested that

such projects triggered NSR. Indeed, as the district court

recounted at length, EPA was fully aware of “life extension”

practices as early as the 1980s. Pet. App. 50a n.13. EPA-

directed studies, which were provided to Congress and formed

the basis of the Acid Rain Program enacted in the 1990 CAA

Amendments, assumed that existing coal-fired power plants

would continue to operate at original capacity for 55 to 65

years on average, being “refurbished” around age 30, without —

being subject to New Source programs. /d.”’ After WEPCo,

and throughout the 1990s, EPA continuously reaffirmed its

"Thus. the WEPCo Rule's actual-to-projected-actual test may be used

by electing utilities, even where a project increases an existing unit's

maximum hourly emissions rate. See 57 Fed. Reg. at 32,324-26 (explai-

ning the new test may be applied to any change at a utility, except the

construction of a “greenfield” unit or the replacement of an existing one).

Thus, after 1992. the NSR rules provided three emissions increase

tests: (1) “actual-to-actual™ for units that have begun normal operations; (ii)

“actual-to-potential” test for units that have not begun normal operations:

and (itt) “actual-to-projected actual.” which can be used by certain

operating utilities, provided they submit necessary reports. Recently, EPA

promulgated new rules rewriting the NSR emissions increase provisions.

See 67 Fed. Reg. 80,186 (Dec. 31, 2002). The parties agree this case ts

governed by the 1980 rules.

~' See also “1989 EPA Base Case Forecasts” (ICF for EPA), Appendix

C, CA App. 364-77; Letter trom K. Schweers. ICF, to R. Beck, Edison —

Elec. Institute July 26, 1989), CA App. 378-79.

13

position, telling industry and Congress most power plant

projects and activities would not trigger NSR. /d. at Sla-52a.

Starting in 1999, EPA reversed course. It filed nine actions

against midwestern and southern coal-fired utilities, including

Duke Energy, and an administrative action against the

Tennessee Valley Authority.” In the enforcement initiative,

EPA alleged near universal non-compliance with NSR,

claiming hundreds of utility repair and replacement projects

over the past twenty years violated NSR.~” According to EPA

and Petitioners, Duke Energy misunderstood (or misapplied)

the 1980 PSD rules for the past quarter of a century (and EPA,

the States, and organizations such as petitioners were all

asleep at the switch). Duke Energy defended on the ground

that EPA was seeking in this enforcement action to re-write

and apply, retroactively, new and unlawful interpretations of

the 1980 PSD rules.

Hl. FACTUAL AND PROCEDURAL BACKGROUND

The court of appeals’ and district court’s decisions and

orders state the relevant facts. See Pet. App. Sa-7a; 23a-3 1a,

50a-53a, 60a-62a; 87a-95a. These facts are summarized

briefly below.

Duke Energy has provided electricity to North and South

Carolina since the early 1900s. Its system includes nuclear,

hydroelectric, gas-fired combustion turbine, and coal-fired

plants. Duke Energy operates thirty coal-fired units at cight

plants in the Carolinas. /d. at 22a, 24a. Virtually all of Duke

Energy's coal-fired units are at tssuc here.

Each coal-fired unit consists of three primary components:

the boiler, the steam turbine, and the generator. The boilers

are large, building-like structures six to twenty stories tall,

containing thousands of steel tubes in which water is heated to

superheated steam with temperatures in excess of 9OO°F. In

general, a boiler consists of collections of tube assemblies. /d.

™ See Dist. Ct. App. DX48-55 (complaints filed by United States).

” See CA App. 493-511 (projects alleged to violate NSR in all actions).

14

at 24a. Because these tubes are subject to corrosion and

extreme pressures and temperatures, they periodically fail and

must be repaired or replaced so that the utility can meet its

statutory duty to provide the public a reliable electricity

supply. See, e.g., N.-C.GEN.STAT. §62-2.

Here, EPA alleged that Duke Energy modified generating

units at all eight of its coal-fired power plants without

obtaining necessary PSD pre-construction permits. The

earliest of the 29 targeted projects took place in 1988. Most

of the projects consisted of replacing one or more of four

boiler tube assemblies that had deteriorated due to normal

wear and tear (/.¢., economizers, portions of waterwalls,

superheaters and reheaters). See Pet. App. 24a-25a.

Most of the projects occurred during regularly-scheduled

planned outages. During such outages, maintenance, repair

and replacement tasks are performed inside the unit that

cannot be performed during operation. These outages occur

every 48-84 months, depending upon the unit's condition and

load requirements of the system. Typically, outages are plan-

ned for spring or fall when electricity demand is lower, and

they can last from several weeks to months. CA App. 518410.

The other projects at issue occurred during periods of

“extended cold shutdown” (“ECS”)—when the units were not

needed to meet demand, but during which Duke Energy

“made definitive plans for preserving and conditioning the

units so that they could be returned to service when demand

dictated.” Pet. App. 27a-28a.

Some time after the units were placed in ECS, Duke Energy

developed an enhanced maintenance program, the “Plant

Modernization Program” (“PMP”). While separate projects

were consolidated under PMP, the work was similar in nature

to the boiler and turbine component repair and replacements

commonly undertaken during regularly scheduled outages by

Duke Energy (and industry generally). CA App. 55999.

During this period, Duke Energy was in frequent contact

with state regulators and told them about the work to be done

at its units. State regulators, who were responsible for

1S

determining CAA compliance, regularly inspected Duke

Energy's plants while the units were in ECS and PMP, and

knew of the work being performed on them. The inspection

reports uniformly certified the plants were in compliance with

all applicable air quality regulations. CA App. 623-79

(collecting reports). For example, in the course of renewing

the state permits for Duke Energy's Buck plant in 1987, the

North Carolina regulator wrote that “Unit Nos. 3, 4 and 5 are

currently in the plant modernization project (PMP) until

1989,” and he explicitly noted "PSD, NSR, NSPS. and

NESHAP do not apply.” CA App. 654.

EPA was well aware of industry repair and replacement

practices, including “the utility industry practice of engaging

in life extension projects.” Pet. App. 50a n.13. Indeed, EPA

knew of specific Duke Energy life extension projects, and

raised no questions about them. See id. In the last two

decades, EPA conducted thousands of utility — plant

inspections.” EPA has visited plants while boiler component

replacements were ongoing, but EPA never cited them even as

potential violations of NSR. /d.

Although Duke Energy's existing units are not regulated as

“new sources, they are extensively regulated under the CAA.

They have emission limits set to protect public health (with an

adequate margin of safety) under the NAAQS, based on the

conservative assumption that the units would operate at

maximum capacity every hour in the year. Supra 3. In

addition, they are regulated under (1) the CAA “Acid Rain”

Program, which requires existing utilities to make further

reductions in sulfur dioxide (SO») and nitrogen oxides (NO,)

emissions; (2) the “NO, SIP Call Program” and the Clean Air

Interstate Rule, which require additional reductions in utility

SO, and NO, emissions in the eastern half of the U.S.. see

Michigan vy. EPA, 213 F.3d 663 (D.C. Cir, 2000), 70 Fed.

" See, e.g, EPA, Profile of the Fossil Fuel Electric Power Generation

Industry, 123, 125 (Sep. 1997) htp://www.epa.gov/compliance/resources/

publications/assistance/sectors/notebooks/fossil. him.

16

Reg. 25,162 (May 12, 2005); and (3) North Carolina’s 2002

Clean Smokestacks law, one of the nation’s most stringent air

emission control laws, N.C.GEN.STAT. §143-215.107D. Duke

Energy's program to meet this new state law alone involves

the installation of additional controls on most of its units at an

expected cost of $1.5 billion. CA App. 523-24 998-11.

None of Duke Energy’s units have violated any of the

emission limits or control requirements under these CAA and

state programs. The parties have stipulated that none of the

projects increased the maximum achievable hourly emission

rate of any of the units. Pet. App. 88a-89a. That is, none of

the projects at issue created any new or unregulated emissions

at any unit. Rather, Plaintiffs’ allegations below were “based

solely on their contention that the projects would have been

projected to result in increased utilization of the units at

issue.” /d. (emphasis added)."'

On cross-motions for summary judgment, the court had

before it three different interpretations of “actual emissions

increase” under the 1980 PSD rules. Each of these

interpretations had been advanced by EPA at one point in the

regulatory history: (1) EPA’s contemporaneous interpretation

of the rules providing an “actual-to-actual” test (see Pet. App.

58a-60a); (2) an “actual-to-potential” interpretation that EPA

announced as the basis for this enforcement action, but then

abandoned later in the district court (see id. at 58a n.17); and

(3) an “actual-to-projected-actual” interpretation, which EPA

then asserted in this action. The court upheld EPA’s actual-

to-actual interpretation, finding it compelled by the rules’

plain language.’ and consistent with EPA’s historic interpre-

tation of the rules, the statute, and precedent. /d. at S8a-72a.

After reviewing the two interpretations of the rules that the

' Duke Energy disputes this contention, and believes none of the

projects resulted in an emissions increase, even under Plaintiffs’ theories.

* Pet. App. 60a (also rejecting EPA's attempt “‘under the guise of

interpreting a regulation, to create de facto a new regulation."”) (quoting

Christensen v. Harris County, 529 U.S. 576, S88 (2000)), id. at 70a

(EPA’s enforcement interpretation is “{njowhere in the regulations.”’).

17

parties pressed on appeal, the Fourth Circuit agreed with EPA

that the “critical first ‘question at issue’ was “whether the

EPA ‘can interpret the statutory term “modification” under

PSD differently from how EPA interpreted that term’ in the

NSPS.” Pet. App. |0a-1 la (quoting U.S. Brief at 1). Because

the answer was no, and Plaintiffs’ litigation interpretation of

the 1980 PSD rules rendered those rules inconsistent with the

NSPS rules defining “modification,” the unanimous panel

(Judges Motz, Luttig, and Wilson—by designation) concluded

that interpretation must be rejected in favor of EPA’s

contemporancous interpretation that makes the two regulatory

programs consistent, as Congress intended.

REASONS FOR DENYING THE WRIT

I. THE FOURTH CIRCUIT’S CONSTRUCTION OF

THE RULES AND THE STATUTE IS NOT

INCONSISTENT WITH THOSE OF OTHER

COURTS OF APPEALS OR THIS COURT.

The Fourth Circuit construed in this case the 1980 PSD

rules, as applied to Duke Energy's existing units—units that

had long ago “begun normal operations.” The court rejected

the Plaintiff's litigation position, and affirmed instead EPA’s

own contemporancous interpretation of these rules (see supra

8-9). That construction is not contrary to any other court of

appeals’ construction of these rules.

Puerto Rican Cement Co. v. EPA, 889 F.2d 292 (Ist Cir.

1989) did not address the 1980 rules for units that fave begun

normal operations. The very first sentence of that opinion

explains, the “Company wishes to build a new cement kiln,

replacing older kilns ....° /d. at 293 (emphasis added). A new

kiln is a unit that has net begun normal operations, and for

which the actual-to-potential test plainly applies under the

1980 rules. As then-Judge Breyer observed, “[T]here ts no

logical contradiction in rules that, on the one hand, permit

firms using existing Capacity simply to increase their output

and, on the other, use the potential output of new capacity as a

18

basis for calculating an increase in emissions levels.” /d. at

298 (underlines added). But, that is neither disputed by the

Fourth Circuit nor contrary to its construction of the 1980

rules for units that have begun normal operations. Cf.

WEPCo, 893 F.2d at 917 n.12 (“unlike the case at issue here,

Puerto Rican Cement involved the construction of a new

emissions unit at an existing source.) (underlines added).

The Seventh Circuit in WEPCo rejected EPA’s attempt to

interpret the rules as requiring an actual-to-potential test for

units that had begun normal operations. /d. at 917. The

court’s remand instructions were clear: NSR is triggered for

existing units that have begun normal operations where “the

renovated plant would cause a significant net emissions

increase if it were operated under present hours and

conditions” —i.e., if hours of operation were held constant.

Id. at 918 n.14 (emphasis added). This is the actual-to-actual

test applied by EPA in 1981, adopted by the district court, and

affirmed by the Fourth Circuit. Indeed, given that the district

court discussed WEPCo at length and relied on it in part to

reach its construction of the 1980 rules, Pet. App. 67a-70a,

Petitioners’ argument that the decision below is contrary to

WEPCo defies credulity.

Contrary to Petitioners’ claim, nothing the D.C. Circuit said

in New York conflicts with the Fourth Circuit's construction

of the 1980 rules or the statute. The D.C. Circuit stated: “We

express no opinion as to whether Congress intended to require

that EPA use identical regulatory definitions of modification

across the NSPS and NSR programs. C/. [Duke Energy].”

New York, 413 F.3d at 20 (emphasis added).

Despite this express acknowledgement, Petitioners strain to

identify a conflict between Duke Energy and New York. They

fail. New York principally addressed the lawfulness of the

revised NSR rules promulgated by EPA in 2002, rules that

were not at issue in this case and to which the Fourth Circuit

did not speak at all. And while industry petitioners challenged

EPA’s contention in the preamble of the 2002 rule that,

“(prior to today, the [1980 PSD] regulations applied ... an

19

actual-to-potential applicability test for all ... emissions

units,’ the D.C. Circuit explicitly declined to reach that issue

as well, finding it not ripe for review and best addressed in a

more concrete factual setting. According to the court, “[i]f

there are still pending applications of the 1980 rule in which

EPA attempts to employ the disputed” interpretation, “judicial

proceedings addressed to the application could solve the

problem of any affected firm.” /d. at 21. That is precisely

what the Fourth Circuit did here. In short, the D.C. Circuit

did not adopt avy interpretation of the 1980 PSD rules, much

less one in “conflict” with the court’s decision here. — It

specifically left the debate over the meaning of the 1980 rules

for resolution in the local circuits in the enforcement context.

Unable to find a conflict between the Fourth Circuit and

otiier circuits on the actual issue before the court below (i.c.,

the meaning of the rules), Petitioners claim the court's reason-

ing (/.¢., its application of statutory construction principles)

presents such a conflict and is inconsistent with this Court's

teachings on statutory construction. Petitioners are wrong.

Both courts below recognized that, in some circumstances,

courts — indeed the Fourth Circuit itself — have agreed that

EPA may define the same term differently under PSD and

NSPS. See Pet. App. 16a-18a (discussing PEPCo v. EPA, 650

F.2d 509 (4th Cir. 1981) and Northern Plains Res. Council v.

EPA, 645 F.2d 1349 (9th Cir. 1981)): Pet. App. 64a-67a

(discussing Alabama Power Co. v. Costle, 636 F.2d 323, 401

(D.C.Cir. 1979) and Northern Plains). But they both

explained that these courts’ decisions turned on the fact that

Congress had defined the terms at issue there (“source™ in

PEPCO and Alabama Power, and “commenced” in Northern

Plains) differently under PSD than NSPS. Here, by contrast,

Congress explicitly “linked the PSD definition of the term

ae

See 67 Bed. Reg. at 80.199. Notably, that ts an interpretation

different from the actual-to- projected-actual interpretation EPA advanced

simultaneously before the Fourth Circunt, further undercutting Petitioners’

claim of the court's rejection of EPA’s “authoritative” interpretation.

20

[modification] to its use in the NSPS provisions of the

statute.” Pet. App. 17a.

Nor did the Fourth Circuit ignore or misconstrue this

Court’s precedents. The court recognized that the task of

statutory construction, as “holistic [an] endeavor” as it may

be, does not start with the court’s (or the parties’) conception

of the “different purposes” of PSD and NSPS. Rather, that

task starts with the language of the statute, see e.g., Barnhart

v. Sigmon Coal Co., 534 U.S. 438 (2002), and the “different

purposes of the NSPS and PSD programs cannot override

|{Congress’] mandate” as expressed in the statute.” Pet. App.

17a-18a; cf. FDA v. Brown & Williamson Tobacco Corp., 529

U.S. 120, 161 (2000) (“In our anxiety to effectuate the

congressional purpose of protecting the public, we must (ake

care not to extend the scope of the statute beyond the point

where Congress indicated it would stop.”) (internal quotation

and citation omitted).

The court below expressly recognized that while “the same

word or phrase will generally be presumed to have the same

meaning when used in different parts of the statute,” that

“presumption ... relents’ when there is ‘a variation in the

connection in which the words are used as reasonably to

warrant the conclusion that they were employed in different

parts of the act with different intent.” Pet. App. 17a (quoting

Gen. Dynamics Land Sys., Inc. v. Cline, 540 U.S. 581, 595

(2004)). Indeed, the court explained, decisions like PEPCO

* Koons Buick-Pontiac GMC vy. Nigh, 125 8. Ct. 560, 566-67 (2004),

quoted in Pet. at 25.

* The district court explained that construing the PSD and NSPS

modification rules congruently such that PSD is triggered only if there ts

NSPS modification activity at an existing unit, and then only if the source-

wide annual emissions exceed a de minimis threshold, is not only

consistent with the statutory text and intent but also effectuates the air

quality purposes of PSD and thus “does not ignore the objectives of each

program.” Pet. App. 64a. As EPA agrees, the structure of the CAA makes

clear that NSR is not an emission reduction program for existing sources,

but rather a program to manage emissions “growth” (.¢., new emissions).

69 Fed. Reg. 23,951, 23.986 (Apr. 30, 2004) (emphasis added).

et

and Northern Plains illustrate that principle. /d. But the court

looked here, as it must, at the particular statutory language at

issue, and it recognized that this language is different from the

language construed in those cases, and must be given cffect.

It correctly concluded that, as this Court’s decision in Rowan

illustrates, it means something when Congress does not

merely use the same term in different parts of the statute but

also defines that single term in substantially the same way.”

Pet. App. | la-15a (discussing Rowan Cos. v. United States,

452 U.S. 247 (1981)). It defies common sense to conclude in

such circumstances that despite Congress’ identical statutory

definition of that single term, the agency can give that term

two different meanings. This reasoning is even stronger here.

“ Contrary to Petitioners’ claim, the similarities between this case and

Rowan are not only that Congress defined the same term the same way in

two related parts of the statute. The Rowan Court noted that Congress

chose the concept of “wages” as the “base” for measuring tax obligations,

and used substantially the same language to define that concept—tirst in

1938 in FICA and FUTA, and then seven years later in the income tax

withholding statute. Here, in 1970, Congress chose the concept of

“modification” as the “base” for the type of activity that would transform

an existing unit into a new one under NSPS. Seven years later, Congress

made NSPS “modification” a trigger for pre-construction review of an

existing unit under NSR. As Rowan held, “|iJn view of this sequence of

consistency. the plain language of the statutes is strong evidence that

Congress i ~ded [the commonly defined term| to mean the same.” 452

U.S. at 255. Hf there is a difference between Rowan and this case, it is that

the plain language of the CAA, incorporating by reference the NSPS

definition of modification into NSR, is “even stronger evidence” that

Congress imended the term to be interpreted consistently. Pet. App. | 4a.

The court also contirmed the legislative history “does not in any way

suggest that Congress intended these identical statutory definitions to

receive different interpretation,” and “indicates congressional concern with

the same sort of simplicity and consistency” evident in Rowan. Id. at |4a-

15a. Finally, EPA’s contemporaneous construction of the statute's PSD

provisions is reflected in the 1978 rules, which defined “modification”

under PSD congruently with NSPS. see supra 6-7, as did EPA’s

comtemporancous construction of the L980 PSD rules, sce supra 8-9. Cf.

Rowan, 452 U.S. at 253 (discussing the agency's. contemporancous

interpretation of the statute ).

22

lt defies common sense even more for petitioners to argue that

by expressly stating in one part of the statute that a term

means the same as in another part of the statute, Congress

meant to authorize the agency to give that term different

meanings.

The language of the statute is plain: for an activity at an

existing unit to trigger PSD pre-construction review, it must

be a “modification [Jas defined in section 7411.” 42 U.S.C.

§7479(2)(C). It is only under these circumstances that the

Fourth Circuit held that if Congress” words in the statute are

to be given effect, the presumption of uniform usage had

become “effectively irrebutable.” Pet. App. 18a.

Despite the Fourth Circuit's invitation, Petitioners have still

failed to cite “a single case in which any court has held that

identical statutory definitions can be interpreted differently by

the agency charged with enforcement of the statute.” Pet.

App. 18a. Gen. Dynamic Land Systems and United States vy.

Cleveland Indians Baseball Co., 532 U.S. 200 (2001), plainly

do not fit that category. The fact is that only one other court

was squarely presented with the particular statutory

construction issue that the Fourth Circuit resolved.*’ That

court flatly refused to engage the issue. See New York, 413

F.3d at 20. There is no conflict for this Court to resolve,

Ht. CERTIORARI SHOULD NOT BE GRANTED BE-

CAUSE THIS CASE INVOLVES AN ISSUE THAT

IS PERCOLATING IN THE LOWER COURTS.

Petitioners’ claims that this case presents “recurring issues

of national importance” are based primarily on their assertions

that (1) the Fourth Circuit’s decision is inconsistent with that

of other circuits—an assertion that is clearly incorrect as

discussed in section |, supra; and (2) that the Fourth Circuit

” It is not surprising that there is no precedent addressing the particular

issue Of statutory construction presented here. The language of the CAA,

in which Congress explicitly and by cross-reference adopted a pre-existing

definition of a term (here “modification”) for a new program, is to our

knowledge umique. This fact also counsels against certiorari in this case.

23

exceeded its jurisdiction—an assertion that is also clearly

incorrect, as discussed in section III, infra.

Petitioners also note that EPA recently proposed a further

amendment of the PSD rules that would adopt provisions

consistent with the Fourth Circuit's construction of the rules.

70 Fed. Reg. 61,081 (Oct. 20, 2005). Contrary to Petitioners’

suggestion, see Pet. at 27-29, EPA’s recent action counsels

against this Court granting certiorari and devoting its own

scarce resources to reviewing the Fourth Circuit’s decision.

That is particularly true here because EPA’s legal rationale for

the proposed revision to the PSD rules does not rely on the

Fourth Circuit’s decision or reasoning (rather EPA reiterates

its disagreement with it). See 70 Fed. Reg. at 61,098-101,

61,090. Nor does EPA’s policy rationale for the proposed

rule rely on the Duke Energy decision. See id. at 61,093-95."

EPA’s proposed rule, once finalized, will rise or fall on EPA’s

legal and policy rationale for the rule, irrespective of whether

this Court decides to review the Duke Energy decision.

Petitioners nevertheless insist the meaning of the 1980 PSD

rules “will retain considerable importance” because these

rules will continue governing numerous projects undertaken

before their amendment. Pet. at 29 n.15. Even if that were

true, it is no reason for this Court to grant review. Petitioners

identify several pending actions in which the issues decided

by the Fourth Circuit in this case are squarely before district

courts in different circuits. To date, two of these courts have

issued interlocutory orders ruling on the issue here. United

* EPA explained: “We believe it is desirable to change the approach to

major NSR. The current approach discourages sources trom replacing

components, and encourages them to replace components with inferior

components or to artificially constrain production in other ways. This

behavior does not advance the central policy goals of the major NSR

program as applied to existing sources. ... We also do not believe the

outcomes produced by the approach we have been taking |/e.. the

interpretation reflected in the enforcement initiative} have significant

environmental benefits compared with the approach we are proposing

today.” 70 Fed. Reg. at 61,094.

24

States. v. Ala. Power Co., 372 F.Supp.2d 1283 (N.D. Ala.

2005), followed Duke Energy, adopting EPA’s contempora-

neous interpretation of the 1980 PSD rules. United States v.

Cinergy reached a contrary conclusion and adopted EPA’s

new actual-to-projected-actual test. The Seventh Circuit

recently granted Cinergy’s motion for interlocutory appeal.

384 F.Supp.2d 1272, 1274 (S.D. Ind. 2005), appeal docketed,

No. 06-8029 (7th Cir. Jan, 3, 2006). Accordingly, the issue

addressed by the Fourth Circuit is percolating in the different

circuits, though there is currently no conflict among them.

For this reason alone, even if Petitioners were correct that the

issue the Fourth Circuit decided is a “recurring” one, this fact

counsels against Supreme Court review in the context of this

case. Sec, e.g.. Brown v. Texas, 522 U.S. 940 (1997) (the

likelihood that the issue will be resolved correctly may

increase if this Court allows other tribunals to serve as

laboratories in which the issue receives further study before it

is addressed by this Court”) (internal quotation and citation

omitted).

Il. THE FOURTH CIRCUIT DECISION TO REJECT

EPA’S ENFORCEMENT INTERPRETATION OF

THE 1980 RULES AND AFFIRM INSTEAD THE

DISTRICT COURT'S APPLICATION OF EPA’‘S

CONTEMPORANEOUS INTERPRETATION OF

THESE RULES PRESENTS NO CONFLICT AND

WAS WITHIN THE COURT'S JURISDICTION.

Petitioners argue at length that the Fourth Circuit exceeded

its jurisdiction by “invalidat|ing]” the 1980 PSD rules. Pet. at

13. This argument is based on a false premise. The court did

not invalidate the rules. While 42 U.S.C. §7607(b) provides

that challenges to nationally applicable CAA rules may only

be brought in the D.C. Circuit, the enforcement action brought

by EPA against Duke Energy did not involve a challenge to

any rule. Rather, at issue was the permissibility of the

interpretation of the 1980 PSD rules on which EPA based its

claims of liability in a case brought in the Middle District of

25

North Carolina, an issue that, on appeal, may be resolved only

in the Fourth Circuit. 42 U.S.C. §7413(b); 28 U.S.C. § 1291.

The district court and the Fourth Circuit had before them

three different interpretations of “actual emissions increase”

under the PSD rules. These interpretations had ech been

advanced by EPA previously. See supra 8-12 & 1.26. Of

these interpretations, the Fourth Circuit, like the district court,

upheld the “actual-to-actual” interpretation, the interpretation

that EPA had announced almost contemporaneously with its

promulgation of 1980 PSD rules. The court explained: “Our

choice of this interpretation of the PSD regulations—as

required by statute—over the EPA’s interpretation is not an

invalidation of those regulations. The PSD regulations remain

fully intact and enforceable” under EPA’s contemporaneous

interpretation, which is fully consistent with the statute. Pet.

App. ISa n.7. Petitioners have presented no compelling

reasons for certiorari.

First, Petitioners halt-heartedly suggest a conflict, citing

myriad cases for the proposition that the court's decision here

is “an abrupt departure” from other courts of appeals that

“have held that contentions that CAA regulations are contrary

lo statute are cognizable in the D.C. Circuit, and are barred

from enforcement proceedings,” Pet. at 14 & n.7, 0.8. These

cases are inapposite. The vast majority of these cases did not

involve an agency's interpretation of the rule, as was the case

here.” None held an agency's interpretation of its rule ts

”

In their zeal to locate decisions from multiple circuits that

purportedly support their contention. Petitioners cite cases that are oardly

relevant to the issue here. Enviro Tech Intl, Inc. v. EPA, 371 F.3d 370

(7th Cir, 2004), was a FOIA case, involving no challenge to any rule or its

imterpretation. Several cases involved untimely. collateral attacks on

permits or final agency action, not a regulation or its interpretation. See

Chemical Weapons Working Group, Inc. vo U.S. Dep't of the Army, 111

F.3d 1485. 1492 (10th Cir, 1997) (collateral attack on RCRA permit):

Hawaiian Elec. Co. v. EPA, 723 F.2d 1440, 1445 (9th Cir, 1984) Cuntimely

challenge to PSD permit); Pan Am. Grain Mfg. Co., v. EPA, 9S &.3d LOL,

103 (ist Cir, 1996) (untimely challenge to nonattainment designation);

Greenpeace, lnc. v. Waste Tech. Indus... 9 F.3d 1174, 1180 (6th Cir. 1993)

26

reviewable only in the D.C. Circuit.” Petitioners fail to cite

the only other circuit decision that generally presents facts

similar to this case. That decision reached the same

conclusion as the Fourth Circuit below. See League of

Wilderness Defenders/Blue Mountains Biodiversity Project v.

Forsgren, 309 F.3d 1181, 1190 n.8 (9th Cir. 2002) (stating, in

response to an argument similar to Petitioners’ here, “we do

not invalidate the regulation. Rather, we reject the [agency's]

interpretation of the regulation and give it a construction

consistent with its administrative history, case law, and the

governing statute”): id. at 1190 (“An agency simply may not

interpret a regulation in a way that contravenes a statute.”).

For the same reason, the Fourth Circuit's decision here is

not in “tension,” much less conflict, with this Court’s decision

in Adamo Wrecking Co. v. United States, 434 U.S. 275

(1978), where the Court looked at the rule itself and held that

rule altogether outside the ambit of §7607(b). No claim that

(collateral attack on RCRA permit). Two cases involved direct attacks on

the validity of regulations and merely recognized. as did the Fourth Circuit

here, Pet. App. 1Sa n.7, that such attacks must be addressed to the D.C.

Circuit. See United States v. Kune-Shou Ho, 311 F.3d 589, 607-08 (Sth

Cir. 2002) (retusing to entertain a defendant's procedural challenge to a

regulation); United States v. Walsh, 8 F.3d 659, 664 (9th Cir. 1993)

(noting that defendant challenged the rules themselves and the limitation

on review set forth in 42 U.S.C. §7607(b)). Petitioners even misquote one

case, claiming that Siege/man v. EPA, 911 F.2d 499 (Lith Cir. 1990),

stands for the proposition that a “claim that “EPA’‘s regulation conflicts

with an act of Congress’ may not be entertained in enforcement

proceeding.” Pet. at 14.0.7. That case in fact says the opposite: “we do

have the jurisdiction to consider the argument that EPA’s regulation

conflicts with an act of Congress.” /d. at SO2 n.6.

” Petitioners cite only one case involving a challenge to an agency's

interpretation of the rules. While Petitioners cite that court's unremarkable

footnote noting the company in that case did not challenge the validity of

the NSPS rules and that the court. in that event. would not have had

jurisdiction to entertain such a challenge. see WEPCo, 893 F.2d at 914

n.6, they fail to note that defendant in that case challenged the interpre-

tation of the 1980 PSD rules that EPA sought to apply and the court set it

aside. See id. at 918 (rejecting EPA’s “actual-to-potential” interpretation).

the 1980 rules are not subject to §7607(b) has been made here.

Second, Petitioners’ attack on the court’s construction of

the 1980 rules in the course of their ostensible argument based

on §7607(b) reveals that argument for what it is: not a

jurisdictional argument, but merely a disagreement with the

Fourth Circuit’s decision on the merits. The fundamental

premise of Petitioners’ argument is found in a footnote: that

the district court’s construction of the rules that the Fourth

Circuit affirmed is inconsistent with the rules’ language. Pet.

at 17 n.9. According to Petitioners, the court effectively

invalidated the rules because they cannot be read consistently

with NSPS, which “targets ‘“cmissions rates’ measured in

kilogram per hour whereas the PSD regulations address

‘actual emissions’ in “tons per year.”” Pet. at 18.

Contrary to Petitioners’ contention, the Fourth Circuit did

not construe the 1980 rules as mandating a maximum hourly

emissions rate test to the exclusion of an annual emissions

test. See Pet. at 19. The court affirmed the district court's

(and EPA’s contemporaneous) construction, which explicitly

held that to be a PSD “modification,” “Duke|’s] projects

[must] cause|] an increase in annual net emissions, assuming

constant hours and conditions.” Pet. App. 83a (emphasis

added); see WEPCo, 893 F.2d at 918 n.14. Thus, an annual

net emissions increase can result only from an increase in the

hourly emissions rate, Pet. App. 58a-59a, and only NSPS

“modifications” can be also PSD “modifications.” This const-

ruction is required by the statute, as the Fourth Circuit held. It

is also consistent with the plain language of the rules and the

WEPCo remand instructions. See id. at S8a-60a, 67a-69a.

Regardless, that Petitioners disagree with the court below

on the merits is no reason for this Court to grant review, for

the reasons discussed earlier in Parts I and I.

Third, Petitioners are simply wrong on the merits. That the

Fourth Circuit relied primarily on the statute for its rejection

of EPA’s (and Petitioners’) interpretation of the 1980 rules in

favor of EPA’s own contemporaneous interpretation of these

rules does not turn the court's ruling into an “invalidation” of

28

the rules. Rather, it is a straightforward application of this

Court's “often stated” admonition that deference is due to an

agency's regulatory interpretation that is not plainly erroneous

or inconsistent with the rules “provided |that interpretation]

does not violate the Constitution or a federal statute.” Stinson

v. United States, 508 U.S. 38, 45 (1993) (emphasis added).

The Fourth Circuit, faced with multiple EPA interpretations of.

the 1980 PSD rules, thus recognized that the first step in

discharging its duty in this case — /.e., determining which EPA

interpretation might command the deference that this Court’s

decisions require — is to determine which of these interpreta-

tions is consistent with the statute.*' See Pet. App. 10a-1 la.

The court correctly deemed it unnecessary to proceed beyond

that first step when it determined that only one of EPA’s

interpretations—the contemporaneous interpretation upheld

by the district court—is consistent with the statute.”

In short, the court's review of EPA’s enforcement interpre-

tation falls squarely within its jurisdiction. Whereas the D.C.

Circuit has jurisdiction to review the validity of nationally

applicable CAA rules, the Fourth Circuit has jurisdiction to

review the validity of regulatory interpretations that EPA

asserts in actions arising in the Fourth Circuit.”

" Both EPA and Petitioners “expressly acknowledge[d| ... these

principles” in their briefs below, and EPA conceded that “the critical first

‘question at issue’” was the consistency of its interpretation of the

regulation with the statute. Pet. App. 10a-lla. It was not until the court

requested supplemental briefing that the Plaintiffs claimed that answering

that “critical first question” somehow violates 42 U.S.C. §7607(b).

© The district court also relied on the statute, see Pet. App. 63a-64a

(citing Stinson), but went on to explain EPA's enforcement interpretation

of the 1980 rules was inconsistent with their plain language, id. at S8a-60a.

* See United States y. Ohio Department of Highway Safety, 635 F.2d

1195, 1197 n.2 (6th Cir. 1980) ('[The rule] was not challenged by a

petition for review. Thus, we treat it as a lawful regulation and are

concemed only with the proper interpretation of the regulation ....°); see

also PEPCo v. EPA, 650 F.2d SOY, 513 (4th Cir, 1981) (This court ... has

jurisdiction to review only the EPA’s interpretation of those regulations

. h Forsgren, 309 F.3d at 1190 0.8.

29

Finally, Petitioners’ statement that “authoritative”

interpretations of the regulation are subject to §7607(b)(2) is

wide off the mark. First, if Petitioners’ argument is correct,

no district court would have any authority to engage in any

review of the agency's interpretation for that case; the court

must simply take the agency’s litigation position at face value

~ even if that interpretation is plainly inconsistent with the

regulatory language — and mechanically apply it to the case at

hand. That is not, and never has been, the law.

Second, what is EPA’s “authoritative” interpretation of the

1980 PSD rules? Petitioners ignore EPA’s contemporaneous

interpretation and blithely assert EPA’s “formal” interpreta-

tion was announced, and upheld, in the WEPCo matter. Pet.

at 7-8. In fact, the “actual-to-potential” interpretation that

EPA advanced in WEPCo was rejected by the Seventh Circuit

and was acknowledged by EPA itself as impermissible in

1992. Supra 10-11 & n.22. Moreover, the “locally

applicable” final agency action that was reviewed, and set

aside, in WEPCo was EPA’s PSD applicability determination

in that case.

Section 7607(b) requires prompt challenges to nationally

applicable CAA rules in the D.C. Circuit and other final

actions in the regional circuits. It does not provide a means

for EPA to promulgate ambiguous rules, change its interpreta-

tions at will, and claim that whatever interpretation it chooses

to apply in a given enforcement case is immune from review.

As the Fourth Circuit explained, it scrupulously followed

the requirements of the CAA, reviewing the interpretations

presented to it by the parties and not the validity of the rules.

Pet. App. 1San.7. This is a straightforward application of the

statute. Certiorari is not warranted.

CONCLUSION

The petition for writ of certiorari should be denied.

Respectfully submitted.

PAUL R. TOURANGEAU

GARRY S. RICE

DUKE ENERGY CORPORATION

Legal Department

526 South Church Street,

ECO3T

Charlotte, NC 28242

(704) 382-8111

DEAN M. MOESSER

DUKE ENERGY CORPORATION

5400 Westheimer Court

Houston, TX 77056-5310

(713) 627-5298

March 8, 2006

HENRY V. NICKEL

F. WILLIAM BROWNELL

(Counsel of Record)

MARK B. BIFRBOWER

MAKRAM B. JABER

HUNTON & WILLIAMS LLP

1900 K Street, N.W.

Washington, D.C. 20006

(202) 955-1500

T. THOMAS COTTINGHAM, III

NASH E, LONG, III

HUNTON & WILLIAMS LLP

101 South Tryon Street,

Suite 3500

Charlotte, NC 28280

(704) 378-4700

Counsel for Respondent