Reply Brief — Alaska Dept. of Environmental Conservation v. EPA

No. 02-658

IN THE

Supreme Court of the Gnited States

STATE OF ALASKA, DEPARTMENT OF ENVIRONMENTAL

CONSERVATION,

Petitioner,

v.

UNITED STATES ENVIRONMENTAL PROTECTION AGENCY, ef al.,

Respondents.

On Writ of Certiorari to the

United States Court of Appeals

for the Ninth Circuit

REPLY BRIEF FOR TECK COMINCO ALASKA

INCORPORATED IN SUPPORT OF PETITIONER

ROBERT T. CONNERY LAWRENCE L. HARTIG

MARCY G. GLENN ROBERT J. MAHONEY*

HOLLAND & HART LLP HARTIG RHODES HOGE

555 Seventeenth Street & LEKISCH

Suite 3200 717 K Street

Denver, Colorado 80202 Anchorage, Alaska 99501

(303) 295-8000 (907) 276-1592

* Counsel of Record Counsel for Teck Cominco

Alaska Incorporated

WILSON-EPES PRINTING Co., INC. — (202) 789-0096 -— WASHINGTON, D. C. 20001

——

TABLE OF CONTENTS

I. BOTH THE LAW AND THE FACTS

SUPPORT ADEC’S DECISION ...........:::0:0004

II. COMINCO’S PERMIT WILL NOT RE-

SULT IN HARM TO THE ENVIRON-

Page

13

Respondent’s Rule 29.6 Statement appears at Resp. Br. ii.

(i)

il

TABLE OF AUTHORITIES

CASES:

Alabama Power Co. v. Costle, 636 F.2d 323 (D.C.

Cit. 1DCD)......csccrccsccscovesossscsosesssoonsssovscsesenenensensenneenee

Chevron U.S.A. Inc. v. Natural Resources Defense

Council, Inc., 467 U.S. 837 (1984) ..cccceeeeeneeeenes

Motor Vehicle Mfrs. Ass'n v. State Farm Mut. Auto.

Pee w (7) ee

STATUTE:

42 USL. § PAPI scoxccceessssicccessescectamasusi

REGULATIONS:

YAS SY \\)

40 CBR. © S1.2GG6OD ceconcercocecesecsessnmsssmaniinea

61 Fed. Reg. 1800 (1996) ......sssscssssssscssesseessessscessessnseee

ADMINISTRATIVE PROCEEDING:

In re Inter-Power of New York, Inc., PSD Appeal

Nos. 92-8 & 92-9, 1994 WL 114949 (E.A.B.

OTHER AUTHORITIES:

NANA & Cominco, Ltd., Shared Values, Common

Goals, Exceptional Results: The Red Dog Mine

Story (available at: http://www.teckcominco.

com/articles/operations/rd-sharedvalues.pdf).........

Page

hl a a -

iil

OTHER AUTHORITIES:

Richard W. Phelps, Leader of the Pack — Red Dog

Mine “Grows Up,” Eng’g & Mining J. (May

1998) (available at: http:/e-mj.com/ar/mining_

nr

Teck Cominco Alaska Incorporated, A Unique

Partnership at the Red Dog Mine: A Model of

Cooperation (available at: http://www.teck

cominco.com/presentations/dh-rd-partner-

CE

U.S. Census Bureau, U.S. Census 2000: Alaska:

Ee nne

Alaskan Communities: Northwest Arctic Census

Region (available at: www.ilovealaska.com/

alaska/boroughs.cfim? bid=28) .............cccccccceeeees

Communities: Red Dog Mine (Nov. 15, 2002)

(available at: http://www.beringsea.com/ com-

munities/index.php?community=3 1 2)...............+.

IN THE

Supreme Court of the Anited States

No. 02-658

STATE OF ALASKA, DEPARTMENT OF ENVIRONMENTAL

CONSERVATION,

Petitioner,

v.

UNITED STATES ENVIRONMENTAL PROTECTION AGENCY, ef al.,

Respondents.

On Writ of Certiorari to the

United States Court of Appeals

for the Ninth Circuit

REPLY BRIEF FOR TECK COMINCO ALASKA

INCORPORATED IN SUPPORT OF PETITIONER

ARGUMENT

The Alaska Department of Environmental Conservation

(“Alaska” or “ADEC”) has explained in detail why the

governing statute does not provide EPA with unilateral veto

power over discretionary BACT determinations that Con-

gress entrusted to the State. As Alaska has shown, the proper

remedy for any party concerned about a State’s exercise of

the discretion afforded by the Clean Air Act is completion of

the State’s own administrative and judicial review proce-

dures. Such procedures have been, and will continue to be,

2

fully adequate to address all of EPA’s concerns about arbi-

trary agency action.

Cominco will not repeat those arguments here. Instead,

Cominco responds to clarify certain assertions made by EPA

and its amici regarding Cominco and its permit. As shown

below, ADEC’s decision is fully justified in both law and

fact. And as evidenced by the reasonable permit proposals

made by Cominco and accepted by ADEC, Cominco takes

seriously its commitments to preserve the environment in

which it operates and to promote balanced economic devel-

opment in the local region.

Had ADEC’s decision gone through the administrative and

judicial review procedure afforded by Alaska law, Cominco

is confident the decision would have been sustained in full.

But since EPA proceeded instead with a unilateral veto auth-

ority it does not have, Cominco and ADEC were deprived of

the opportunity to create a full factual administrative record,

and ADEC’s decision was stripped of the substantial defer-

ence it would have been owed on judicial review. All this is

further reason why the Court should enforce the statute as

written and hold that EPA does not have such authority. The

EPA’s interpretation not only destroys Congress’s coopera-

tive federalism scheme, but subjects applicants like Cominco

to intolerable uncertainty by continually raising the prospect

that EPA might by fiat unilaterally override a State determi-

nation even after that determination becomes final under state

law. See Cominco Br. 5-7.!

! Cominco advised the Ninth Circuit that the record tendered by

EPA, on which the agency purportedly based its enforcement

orders, was inadequate to evaluate the merits of ADEC’s underly-

ing decision. See Cominco Alaska Incorporated’s Response to the

Court’s August 3, 2001 Order 5-6 (Aug. 17, 2001); see also

Petitioners’ Combined Reply Brief 1-3 (Sept. 5, 2000). Had the

matter proceeded through Alaska’s de novo administrative review

process, all parties would have had the opportunity to create a

3

I. BOTH THE LAW AND THE FACTS SUPPORT

ADEC’S DECISION

At bottom, EPA asserts that it has authority to override

ADEC’s permitting decision because the decision was, in

EPA’s view, arbitrary and capricious. The statute, however,

makes clear that the State is entrusted with the discretion to

make BACT determinations, and the concerns raised by EPA

are fully addressed by the available state administrative and

judicial review procedures, which EPA itself requires as a

precondition to approving a state CAA implementation plan

in the first place. See 61 Fed. Reg. 1800, 1882 (1996). But

in any event, even if EPA’s own view of its authority were

correct, its determination that ADEC acted arbitrarily or

unreasonably is incorrect and EPA thus lacks veto authority

even under its own view of the law.

1. EPA’s argument that ADEC acted arbitrarily rests

ultimately on a single premise: that ADEC had a legal duty

to determine Cominco’s ability to pay for SCR and to impose

SCR if the technology was economically feasible for the

company. EPA Br. 26-27, 49-50; see also Environmental

Defense Br. 27-28 & n.28. EPA has taken a position that

neither the Clean Air Act, its legislative history, nor EPA’s

own prior guidance on the statute can support. Cominco did

not open its books to ADEC - it did not attempt to prove that

installing SCR would force it into bankruptcy — both because

it was unwilling to divulge its confidential financial affairs

and because that information was not in fact a necessary part

of the agency’s economic analysis.2 ADEC appropriately

complete factual record. Even on the present incomplete record,

however, ADEC’s decision is fully supportable.

2 EPA requested Cominco to provide “an analysis of the spe-

cific economic impacts of requiring SCR at the Red Dog Opera-

tions with a projection of the cost per unit of production.” J.A.

134. Cominco declined to provide that information in part due to

“concerns related to confidentiality.” Jd. However, Cominco did

4

considered the economic impacts of selecting SCR as BACT

in accordance with EPA’s published guidelines. Under that

correct analysis of economic impacts, ADEC’s selection of

Low NOx as BACT was not arbitrary or capricious.

The Clean Air Act requires the state permitting authority

that determines BACT to take into account, inter alia,

“economic impacts and other costs.” 42 U.S.C. § 7479(3).

Nothing in this language can be read as requiring a state to

evaluate the economic impact of a particular control technol-

ogy upon the individual permittee. Nothing in the Act’s

legislative history supports any such mandate. In fact, EPA’s

own guidance document counsels against consideration of

“the economic situation of the individual source.” Rec. 71-

115 (EPA New Source Review Workshop Manual).

Rather, as EPA admitted prior to this case, “[i]n the eco-

nomic impacts analysis, primary consideration should be

given to quantifying the cost of control.” Jd. As stated by

EPA’s Environmental Appeals Board (“EAB”):

Cost-effectiveness usually involves two considerations.

First, the permit writer must evaluate whether the total cost

per ton of control for the pollutant is within the range of

costs being borne by similar sources also charged with

controlling the pollutant. Second, the permit writer should

evaluate the comparative cost-effectiveness of serious

control options to determine their incremental cost-

effectiveness.

review for EPA the extent of its then-current debt ($400 million),

its losses at the Mine due to a drop in zinc prices (over $150

million), and its prior investment in environmental improvements

at the Mine (over $50 million). /d at 134-35. Separately,

Cominco told EPA that the cumulative cost of installing and

operating SCR over the 40 remaining years in the projected life of

the Mine would be “very significant, substantially affecting

Cominco’s cost of production.” Rec. 56-003.

5

In re Inter-Power of New York, Inc., PSD Appeal Nos. 92-8

& 92-9, 1994 WL 114949, at *4-5 (E.A.B. 1994).

Despite the clarity of EPA’s position on the proper eco-

nomic analysis, EPA chose this permit proceeding for a new

approach. Thus, contrary to the guidance in its New Source

Review Manual, in late 1999 — after release of both the

Preliminary TAR and the draft Final TAR — EPA’s regional

office in Seattle told ADEC:

[Ijn order to justify economic infeasibility, the Region

believes that the economic impact analysis conducted in the

draft permit should have gone beyond a review of cost

effectiveness to include an analysis of whether requiring

Cominco to install and operate the more effective control

strategies [i.e., SCR] would have any adverse economic

impacts upon Cominco specifically.

Rec. 50-004 — 50-005.

At the same time, EPA told other owners and operators in

Alaska that the agency would not “make the same demands

for PSD permits for rural utilities which install similar

generators,” but instead “will continue to be sensitive to

affordability of environmental control technologies for future

permit applicants.” Rec. 62-001 — 62-002. In short, EPA’s

new economic analysis turns on the size and profitability of

the owner/operator. The presumably richer ones, like

Cominco, must pay for the most stringent control technolo-

gies, regardless of how the costs of those technologies com-

pare to those imposed on comparable sources. The presuma-

bly poorer ones, like the Alaska electric utilities, will be

permitted to use less stringent and less costly technologies.

Although EPA’s current economic impact analysis is un-

supported by either the statutory language or history, it is

arguably the EPA’s prerogative to change its own view of

economic factors that it would consider, if it were the deci-

sionmaker, in making a discretionary BACT decision. But

6

cf. Motor Vehicle Mfrs. Ass'n v. State Farm Mut. Auto. Ins.

Co. 463 U.S. 29, 42 (1983) (“[A]n agency changing its

course by rescinding a rule is obligated to supply a reasoned

analysis for the change beyond that which may be required

when an agency does not act in the first instance.”). It is

quite another thing, however, for the agency to impose its

changed views on a state permitting authority that has relied

on EPA’s longstanding approach and has discharged its duty

to select BACT painstakingly and in good faith.

Here, the record makes clear that ADEC undertook the

economic analysis outlined in EPA’s New Source Review

Manual and EAB decisions. The essential inquiry in the

economic analysis is how the cost-effectiveness of SCR on

MG-17 compares to the cost-effectiveness of SCR or other

NOx-reduction controls found as BACT for other similar

sources. See supra at 4-5. Yet, as ADEC worked on that

comparative analysis, EPA insisted that no technology may

be rejected as too expensive if it costs less than $10,000 per

ton of NOx removed — without reference to how the actual

costs compare to costs allowed in other similar circum-

stances. The arbitrary $10,000 cut-off stemmed not from the

statute or its legislative history, not from any EPA regula-

tions, not from any EPA guidance documents, and not from

any court decisions. Rather, it was articulated in a January

1999 letter from an EPA staffer. See Rec. 21-013 — 21-014,

21-020 (ADEC internal document referencing and relying

upon letter); J.A. 126 (September 1999 EPA letter to ADEC

continuing to rely upon $10,000 figure); see also Rec. 22-035

(ADEC internal memorandum documenting conversation in

which author of letter reiterated $10,000 figure).

There is no legal basis for any across-the-board BACT

cost-effectiveness cutoff, much less for the arbitrary $10,000

figure. The apparent references for the claimed cutoff were

two examples involving different types of sources and a

different pollutant, at least one of which was not based on a

BACT determination but on an EPA allegation that was later

7

withdrawn. Rec. 22-035; Petitioners’ Supplemental Excerpts

of Record 463-64. Moreover, ADEC stated in the Final TAR

that it had contacted each of EPA’s regional offices to see

whether there is a standardized cost that is considered

excessive for BACT, but “the Department found that there

does not appear to be a uniform cost policy that is used

throughout the country.” J.A. 206.

Although the figure was unsupportable, it became a pri-

mary basis for EPA’s challenge to ADEC’s cost-

effectiveness analysis and, in turn, for EPA’s intervention in

this case. J.A. 126. EPA’s brief to this Court is notable for

its failure to mention, much less defend, the $10,000 pur-

ported baseline that figured so prominently in EPA’s override

of ADEC’s discretionary BACT determination. These

circumstances — a single staffer’s imposition on a state

permitting agency of an arbitrary cutoff for the key economic

test for BACT — underscore the baselessness of EPA’s

enforcement orders.

2. For the first time in this proceeding, EPA takes issue

with ADEC’s statement that “because Cominco had agreed to

install Low NOx on all its generators, ADEC’s permit dec-

ision was expected to result in ower overall NOx emissions

than would occur if SCR — the EPA’s preferred technology —

were installed on only the MG-17 generator.” ADEC Br. 13

(quoted in part in EPA Br. 45) (emphasis in original).

It is undeniable that two of Cominco’s preexisting engines

— MG-2 and MG-6 — were not subject to BACT and, there-

fore, that Cominco was not required to install any new

control technology on those engines. J.A. 156-57. More-

over, ADEC reasonably found that if SCR were required as

BACT for the new MG-17 generator, Cominco would be

unlikely to use MG-17 except on a standby basis in light of

its significantly higher operating costs. As ADEC recog-

nized, “(t]he higher operating cost of SCR . . . provides an

economic incentive for Cominco to reduce the operating time

8

of the SCR unit. Following this reasoning, Cominco may

choose one of the two units fitted with SCR as the standby

unit in order to reduce operating costs.” J.A. 87.5 Thus, if

SCR were required for MG-17, Cominco would operate four

generators full-time with Low NOx controls (MG-1, and

MG-4 through MG-6), and two generators full-time with a

less stringent control (MG-2 and MG-6), with one largely

unused standby unit (MG-17). That standby unit would

result in little or no emissions reductions, and there would be

no reductions at all from the two exempt full-time units. By

contrast, if all seven units were equipped with Low NOx,

with six operating full-time, overall emissions would unques-

tionably be reduced because of the reductions realized from

the more stringent controls on the MG-2 and MG-6 units.

Ultimately, ADEC did not include the emissions-netting

approach in the Final TAR. But that was in direct response

to EPA’s policy change on the appropriateness of using such

an analysis. In March 1999, when ADEC first proposed a

calculation based on combined NOx emissions, EPA ap-

proved of that approach. Six months later, after ADEC re-

leased its Preliminary TAR, EPA withdrew its support of that

rationale. J.A. 243-44; Petitioners’ Excerpts of Record

(“PER”) 388, 430. EPA’s opposition to the concept, how-

ever, finds no support in the statute. Thus, in considering the

reasonableness of ADEC’s action, it is entirely appropriate to

recognize that the result of that action will be cleaner air,

which is the ultimate goal of the statute.

In determining whether a source will increase emissions so

as to trigger BACT review, it is well-settled that a particular

device is not considered in isolation; rather, emissions may

3 As of May 1999, when ADEC issued the Preliminary TAR,

the agency still was considering BACT for two engines - MG-5

and MG-17. Ultimately, Cominco agreed to an emissions cap for

certain generators, which avoided triggering BACT review for

MG-S. J.A. 149, 197.

9

be “bubbled” so as to consider the entire facility’s overall

impact:

[T]he PSD provisions express a purpose of ensuring that

economic growth occurs in a manner consistent with pres-

ervation of clean air. The bubble concept is precisely

suited to preserve air quality within a framework that al-

lows cost-efficient, flexible planning for industrial expan-

sion and improvement.

Alabama Power Co. v. Costle, 636 F.2d 323, 402 (D.C. Cir.

1980) (footnote omitted). See also Chevron U.S.A. Inc. v.

Natural Resources Defense Council, Inc., 467 U.S. 837

(1984) (upholding “bubble” concept as applied to other

provisions of Clean Air Act).

There is no reason why the same concept cannot apply

when reviewing whether a State has acted arbitrarily or

capriciously in rendering a discretionary BACT determina-

tion. The statute specifically directs a weighing of the

“environmental . . . impacts,” among other factors, of a

particular technology choice. 42 U.S.C. § 7479(3). Thus,

when considering the reasonableness of Alaska’s technology

choice in this case, the Court should reject EPA’s invitation

to disregard overall environmental benefits that technology

will have. Indeed, as next shown, Cominco’s permit will

have no adverse effect on the surrounding environment no

matter how the case is considered.

II. COMINCO’S PERMIT WILL NOT RESULT

IN HARM TO THE ENVIRONMENT

The facts do not warrant the subtext in EPA’s and its sup-

porters’ briefs that the permit granted to the Red Dog Mine

will be detrimental to both the environment and the Native

Alaskan population. The Mine has led to limited develop-

ment in an otherwise undeveloped part of the state, but the

PSD program expressly contemplates industrial development

within clean-air areas of the nation, and it leaves those

10

“growth management decisions . . . for resolution by the

states.” Alabama Power, 636 F.2d at 364. Here, ADEC

made a reasonable accommodation of economic growth in a

rural area that will have no impact at all on the air quality in

surrounding areas.

As noted, ADEC’s decision will result in greater emissions

reductions than if EPA’s preferred solution were employed.

But even if one considered only the emissions from the

MG-17 generator, the Cominco permit will have no impact at

all on the surrounding environment. Using Low NOx

controls on MG-17, the total NOx emissions would remain

well below the NAAQS threshold for public health and

welfare, and also below the PSD increment (which is set at

25% of the NAAQS threshold). See 40 C.F.R. § 50.11(b); id.

§ 51.166(c); J.A. 155, 230. In fact, as the former ADEC

Commissioner explained to EPA, the increase in emissions

from using Low NOx instead of SCR “is basically undetect-

able at the facility boundary — \ess than 1 microgram per

cubic meter. No people would be affected by using SCR

instead of low NOx.” PER 430 (emphasis added). Thus, any

increased emissions will not stray beyond the uninhabited

mine.4

4 EPA errs in describing the Mine as a population center with

“({mJany workers . . . housed within its boundaries.” EPA Br. 4.

There are no permanent residences at or in the vicinity of the

Mine; rather, workers spend intervals of time working on-site

before returning, by plane, to their homes. See Communities: Red

Dog Mine (Nov. 15, 2002) (available at: http://www.beringsea.

com/communities/index.php?community=312). While on-site, and

particularly in the frigid conditions that exist much of the year,

employees spend as much time as possible indoors. See Richard

W. Phelps, Leader of the Pack — Red Dog Mine “Grows Up,”

Eng’g & Mining J. (May 1998) (available at: http:/e-mj.com/

ar/mining leader_packred_dog/). Nor is there any community,

large or small, “nearby” the Mine. EPA Br. 4. Kivalina (popula-

tion 377) and Noatak (population 428) are, respectively, approxi-

mately 50 miles and 35 miles away. U.S. Census Bureau, U.S.

11

Similarly, there is no factual basis for the contention of

several amici that the projected increase in NOx emissions

could harm nitrogen-sensitive vegetation in the Noatak

National Preserve. Kivalina Br. 4; Environmental Defense

Br. 20 n.22. ADEC has reviewed two years of on-site

vegetation studies and five years of off-site vegetation studies

at a comparable site, both of which showed no detectable

impact on the tundra vegetation. PER 430.5

Finally, ADEC properly took account of Cominco’s special

role as one of the few sources of economic activity for a dis-

advantaged community in a remote and inhospitable locale.

The agreement between Cominco and the NANA Regional

Corporation, Inc. (“NANA”) for development and operation

of the Mine is unique in both its commitment to the local

Native Alaskan population and the control vested in NANA

Census 2000: Alaska: 2000 (“U.S. Census”) 13, 15 (June 2003);

Rec. 17-125; Alaskan Communities: Northwest Arctic Census

Region (available at: www.ilovealaska.com/alaska/boroughs.cfm?

bid=28). The Mine is located 90 miles north of Kotzebue (popula-

tion 3,082), the next closest town, and 100 miles north of the

Arctic Circle. U.S. Census at 14; Rec. 17-067. Denali National

Park, located 420 miles southeast, is the nearest Class I attainment

area — the most protected “clean air” area under the Clean Air Act.

Rec. 17-009. In short, contrary to EPA’s suggestion, the Mine is

in an extremely remote and sparsely populated location. See also

ADEC Br. 7-9.

5 One of EPA’s amici raises unrelated environmental compli-

ance issues. Kivalina Br. 3-5. Dust pollution from the haul road

between the port and Mine, and allegations of impacts of Mine

operations on downstream waters, have nothing to do with

ADEC’s authority to determine Low NOx controls as BACT for

the MG-17 generator. Thus, it is not surprising that EPA has not

relied on any of these allegations and concerns — all of which have

been or are being cooperatively addressed by EPA and ADEC

through voluntary actions, agreements, revised permits, and studies

— as the basis for EPA’s intervention in ADEC’s permitting

decision.

12

to ensure that local social and environmental needs are met.

The 1982 agreement between Cominco and NANA provides

for payment of a substantial annual royalty to NANA,

protection of Native Alaskan subsistence resources, training,

hiring, contract and purchasing preferences for NANA

members, and NANA approval of changes in the Mine’s size.

See Teck Cominco Alaska Incorporated, A Unique Partner-

ship at the Red Dog Mine: A Model of Cooperation (avail-

able at: http://www.teckcominco.com/presentations/dh-rd-

partner-may02.pdf). Red Dog is a model of sustainable

development due to its approximately 40 years of reserves

and Cominco’s commitment to NANA for the life of the

Mine. See NANA & Cominco, Ltd., Shared Values, Com-

mon Goals, Exceptional Results: The Red Dog Mine Story 7,

9 (available at: http://www.teckcominco.com/articles/opera-

tions/rd-shared values.pdf).

It true that economic and environmental impacts “must

always be weighed” against each other in the BACT analysis.

Kivalina Br. 5. But that is precisely what ADEC did in this

case, after considering the comments of all interested parties.

The question is whether that weighing is to be done by the

approved State agency that is necessarily closest to the facts

and circumstances, or a federal agency operating from offices

thousands of miles away. Congress provided a clear answer.

— + ee >

13

CONCLUSION

The Ninth Circuit’s judgment should be reversed.

ROBERT T. CONNERY

Marcy G. GLENN

HOLLAND & HART LLP

555 17th Street

Suite 3200

Denver, Colorado 80202

* Counsel of Record

Respectfully submitted,

LAWRENCE L. HARTIG

ROBERT J. MAHONEY*

HARTIG RHODES HOGE

& LEKISCH

717 K Street

Anchorage, Alaska 99501

(907) 276-1592

Counsel for Teck Cominco

Alaska Incorporated