Petition for Writ of Certiorari — Ohio Power Co. v. United States Environmental Protection Agency

Supreme Court brief1984

Ask Donna

What actually matters in this document.

Text

Office -Supreme Court, US.

Fill kED

AUé’S 1984

84-195

No. ——_—— ; ;

ee eT racenpenacieiasicicsichiadteriemasiedans “3

| _ CLERK

IN THE

Supreme Court of the United States

‘OCTOBER TERM, 1984

OHIO PowER COMPANY, - —

Petitioner,

Vv.

UNITED STATES ENVIRONMENTAL PROTECTION AGENCY,

WILLIAM D. RUCKELSHAUS, ADMINISTRATOR,

Respondents.

PETITION FOR A WRIT OF CERTIORARI TO THE.

UNITED STATES COURT OF APPEALS

FOR THE SIXTH CIRCUIT

?

EDWARD BERLIN

Counsel of Record for Petitioner

FRANCIS S. BLAKE

SWIDLER, BERLIN & STRELOW,

CHARTERED

1000 Thomas Jefferson St., N. W.

Washington, D.C. 20007

; (202) 342-5500

. A. JOSEPH DowD

’ JEFFREY P. WHITE

OHIO POWER COMPANY

E. DONALD ELLIOTT

Box 401A Yale Station

New Haven, Connecticut 06520

Of Counsel

WILSON - EPES PRINTING Co., INC. - 789-0096 - WASHINGTON, D.C. 20001

aa

QUESTION PRESENTED

May a court, which has held an administrative rule

arbitrary and capricious, authorize an agency to limit’

its inquiry on remand to the “further study” of a single

“issue,” notwithstanding the mandate of the Administra- -

tive Procedure Act that agency action held to be’ arbi-

trary must be set aside and returned to the\agency for

repromulgation and the requirement of Article III of the

‘Constitution that courts must review the agency action in

controversy, not abstract issues?

*

(i)

ii

PARTIES TO THE PROCEEDING BELOW

Parties before the Court of Appeals in addition to

Petitioner and Respondents named in the caption of this

case were Columbus & Southern Ohio Electric Company,

_ Dayton Power & Light Company, Peabody Coal Company,

and the Commonwealth of Pennsylvania.*

LIST OF PARENT COMPANIES, |

‘SUBSIDIARIES AND AFFILIATES

A listing of parent companies, subsidiaries and af-

filiates of Petitioner as required by Rule 28.1 of the

Supreme Court Rules is as follows:

Petitioner Ohio Power Company is a subsidiary of

American Electric Power Company, the affiliates. and

subsidiaries (other than wholly owned subsidiaries) of

which are:

Appalachian Power Company

Cardinal Operating Company

Columbus & Southern Ohio Electric Company

Indiana Kentucky Electric Corporation

Ohio Valley Electric Corporation

* Additional parties appeared before the Court of Appeals in

the initial proceeding leading to the remand of an “issue” to the

agency (Shell Oil Co., Austin Powder Co. and E.I. duPont

deNemours & Co.). Those parties were not involved in the deci-

sion leading to the final judgment at issue in this case.

TABLE OF CONTENTS

I ie occvecenncsoecnaconiacens

I his

CONSTITUTIONAL AND STATUTORY PROVI-

SN ssid nnccdckianaciscswnssdinessnashctsbsieonnssssoeess

STATEMENT OF THE CASE W002...

REASONS FOR GRANTING THE PETITION ..........

The Decisions Below Raise Important and Unset-

tled Questions of Administrative Law on Which

There Is Conflict among the Lower Federal Courts..

A. The Lower Court’s Failure to Set Aside and

Remand for Repromulgation Is Inconsistent

with Congress’s and this Court’s Delineation

of the Proper Relationship between Courts

and Administrative Agencies ........00000000022- 0.

B. The Lower Court’s Limitation of Judicial Re-

view to the “Issue” Remanded to the Agency

Is Inconsistent with the Proper Function of

Judicial Review under Article III of the Con-

a a salsa behlpnenencpaiaduns

Ea a

APPENDIX: .

Opinions of the Court of Appeals —.........0.0000000000000...

Decisions of the United States Environmental Pro-

RETR EERE

Judgments of the Court of Appeals

Order of the Court of Appeals Denying Rehearing

Se cece

Constitutional and Statutory Provisions Involved..

(iii)

11

16

18

94a

98a

99a

iv

TABLE OF AUTHORITIES ~

Cases: Page

Action on Smoking and Health v. CAB, 713 F.2d

oR + _; RENT eNO RE ene eam 7,10

American Meat Institute v. EPA, 526 F.2d 442

A I ci sss te atcnmbbiineebe Redeigaiens 7

Buckeye Power, Inc. v. EPA, 481 F.2d 162 (6th

Cir. 1973), cert. denied, 425 U.S. 934 (1974)... 12

Burlington Truck Lines v. United States, 371 U. S.

OU i ae al cahasintis 15.

California Cartage Co. v. United enema 721 F.2d

I I soe ics dateaietenacaccctennsectianimnsicoes 7

Camp v. Pitts, 411 U.S. 135 (i978) OPENS AS 7-11

Central Power & Light Co. v. United States, 634

¢ # 8 ge Th. : eeieie eee ce

Cincinnati Gas & Electric Co. v. EPA, 578 F.2d

660 (6th Cir. 1978), cert. denied, 489 U.S. 1114

¢. : | Eaaeneee POY ILD OL SIRS IO ee ORAL TE 2, passim

Citizens to Preserve Overton Park v. Volpe, 401

Oa a eee 8-11

Cleveland Electric Illuminating Co. v. EPA, 572

F.2d 1150 (6th Cir. 1978), cert. denied, 439 U.S.

i caiaediiclinien 3

Columbus & Southern Ohio Electric Co. v. ‘Guile,

638 F.2d 910 (6th Cir. 1980) ...........002220000222...2...- 5

EDF v. Costle, 657 F.2d 275 (D.C. Cir. 1981)........ 7

FCC v. Pottsville Broadcasting Co., 309 U.S. 134

RAB. SR SR eI LTE TY PR 13, 17-18

FCC v. Schrieber, 381 U.S. 279 (1965) .................. 13

Federal Radio Comm. v. General Electric €o., 281

ee Sg ENE By nen es rene oe eaten 17

FMC Corp. v. Train, 539 F.2d 973 (4th Cir. 1976)... 7

Ford Motor Co. v. NLRB, 305 U.S. 364 (1939)... 15

PC v. Transcontinental Gas Pipe Line Corp., 423

Oe 9,13

‘Harrison v. PPG Industries, Inc., 446 U.S. 578

Bee, aR EE RR Ae NIeTe tare Pe RCN eC IG ear 8-9

__) Montana-Dakota Utility Co. v. FERC, 631 F.2d

Ie 7

National Lime Assoc. v. EPA, 627 F.2d 416 (D.C.

aerate crt ace ER th eae ah ee RROD 12

Vv

TABLE OF AUTHORITIES—Continued

Page

NRDC v. EPA, 494 F.2d 519 (2d Cir. 1974) . eee 7

Port Norris Express Co. v. ICC, 687 F.2d 803

SE WIN: I can id onan eka crscaaaausuaiunienioncianenns 7

Public Power Council v. Johnson, 674°F.2d 791

I ON an ubdaucianiomnniaines 9

Rogers v. Hill, 289 U.S. 582 (19338) -................00...... 2

SEC v. Chenery, 332 U.S. 194 (1947) -........00000000.... 14, 15

Tabor v. Joint Board for Enrollment of Actuaries,

ee &, Ble A. ys Ren cae 7

Valley Forge College v. Americans United for

Separation of Church and State, 454 U.S. 464

I aN ae 17

Vermont ‘Yankee Nuclear Power Corp. v. NRDC, |

EE NE TE FD oi inven datnonscrcisincnasaptebacodlbauson 11-13

Statutes and Regulations:

Administrative Procedure Act, 5 U.S.C. § 500 et

seq.

Section 10(e), 5 U.S.C. § 706 .........................- 11-12

Clean Air Act, 42 U.S.C. § 7401 et seq:

Section 109(b) (2), 42 U.S.C. § 7409 (b) (2) ..... 11

Section 110(c), 42 U.S.C. § 7410 (c) _20022- 3

Section 165(e) (3) (d), 42 U.S.C. § 7475 (e),

(3) (d) ......-.- PRM OS EE AN AS Oe OTS 14

Section 302 (h), 42 U.S.C. § 7602 (h) ....000.......... 11

Section 307 (d), 42 U.S.C. § 7607 (d) _....000. 3, 12, 16

Section 820, 42 U.S.C. § 7620 _........................4.... 14

' US. Constitution, Article III, Section 2 .......... -10, 16-18

Miscellaneous:

EPA Promulgation of State Implementation Plan

for the Control of Sulfur Dioxide for the State

of Ohio (41 Fed. Reg. 36324 (August 27, 1976) ;

40 C.F.R. Part 52, Subpart KK, § 1881(b)

OI vcivciskacilastaatnaetaditce Mek n a ees ee ee Ee ae

EPA’s “Reconsideration of the Use of Pasquill-

Gifford Dispersion Coefficients for Stability

Class A in Setting Emission Limitations for

Four Ohio Power Plants” :

vi

TABLE OF AUTHORITIES—Continued

Page

Comments Solicited (44 Fed. Reg. 7798 (Feb-

SUE Ta TORO. taiicicccnbinicrenisssssadminiiibasioiagninsainn 4

Evaluation of Public Comments (45 Fed. Reg.

41501 (June 19, 1980) ) W000... 4

Denial of Petitions for Reconsideration (45

Fed. Reg. 74041 (November 7, 1980) )........ 4-5

Restart of Compliance Schedules (45 Fed. Reg.

_ 73927 (November 7, 1980)) ....................-.-- 4

K. Davis, Administrative Law Treatise, §§ 7.1, 7.2

I i a ee 14

IN THE

Supreme Court of the Huited States

OCTOBER TERM, 1984

No. ——

OHIO POWER COMPANY,

. Petitioner,

UNITED STATES ENVIRONMENTAL PROTECTION AGENCY,

WILLIAM D. RUCKELSHAUS, ADMINISTRATOR,

Respondents.

PETITION FOR A WRIT OF CERTIORARI TO THE

UNITED STATES COURT OF APPEALS

FOR THE SIXTH CIRCUIT

- Ohio Power Company petitions for a writ of certiorari

to review the judgment of the United States Court of

Appeals for the Sixth Circuit in this case.

OPINIONS BELOW

The opinion of the Court of Appeals remanding an

issue to the United States Environmental Protection

Agency is reported at 578 F.2d 660 (Petitioner’s Appen-

dix at 9a-27a) (hereafter referred to as App.). The

opinion of the Court of Appeals upholding the agency’s

action on remand is reported at 729 F.2d 1096. App. 1la-

8a. The order of the Court of Appeals denying rehear-

ing and rehearing en banc is reprinted at App. 98a. The

decisions of the United States Environmental Protection

Agency are reported at 45 Fed. Reg. 41501, 73927, 74041.

App. 28a-938a.

2

JURISDICTION

The judgment of the Court of Appeals (App. la) was

entered on March 22, 1984. Rehearing was denied on

May 8, 1984. App. 98a. This petition is filed within

90 days of the date of denial of rehearing. The jurisdic-

tion of this Court is invoked under 28 U.S.C. § 1254(1).

The decision of the Court of Appeals remanding an “is-

sue” to the U.S. Environmental Protection Agency for

“further study” was issued on June 29, 1978. Since that

decision did not constitute a final judgment with respect

to the controversy presented here, the decision is properly

before the Court at this time under Rule 18 of the Su-

preme Court Rules; see 28 U.S.C. § 2101(e); see also

Rogers Vv. Hill, 289 U.S. 582 (1933). Other aspects of the

Court of Appeals’ 1978 decision that were final are not at

issue in this case. See Cincinnati Gas & Electric Co. V.

EPA, 578 F.2d 660 (6th Cir. 1978), cert. denied, 4389

U.S. 1114 (1979).

CONSTITUTIONAL AND

STATUTORY PROVISIONS INVOLVED

The following constitutional and statutory provisions

are involved in this case and are set out verbatim in the

appendix, App. 99a-107a:

U.S. Constitution

Article I{I, Section 2.

Administrative Procedure Act, 5 U.S.C. § 500 et seq.,

Section 10(e), 5 U.S.C. § 706.

Clean Air Act, 42 U.S.C. § 7401 et seq.,

Section 307(d), 42 U.S.C. § 7607(d).

STATEMENT OF THE CASE

This case involves Ohio Power Company’s attempt to

seek judicial review of the United States Environmental

Protection Agency’s (EPA’s) action establishing an

emission limitation for one of Ohio Power Company’s

3 ;

powerplants. EPA promulgated the emission limitation

at issue in 1976 pursuant to its authority under the Clean

Air. Act."

Ohio Power Company first sought judicial review of the

emission limitation following its promulgation in 1976.

Ohio Power argued that the emission limit was arbitrary

and capricious because the computer model EPA used in |

establishing the limit was not supported by the record in

the rulemaking proceeding. In general, EPA uses com-

puter models to predict the air quality impact of a plant’s

emissions and then, based-on those predictions, establishes

an emission limitation designed to ensure that the plant’s

emissions do not violate the National Ambient Air Qual-

ity Standards. See generally, Cleveland Electric Illumi-

nating Co. v. EPA, 572 F.2d 1150, 1160 (6th Cir. 1978),

cert. denied, 439 U.S. 910 (1979). In this case, Ohio

Power contended that a critical set of assumptions (called

the “Class A” coefficients” )- contained in EPA’s computer

_.model were in error and had been discredited by the

scientific community.

The Court of Appeals for the Sixth Circuit, noting the

scientific community’s rejection of the Class A coefficients,

held: “On this present record we conclude that United

States EPA’s employment of the Class A [coefficients]

. . 1s not a rational decision and is arbitrary and capri-

cious.” App. 16a.

141 Fed. Reg. 36324 (August 27, 1976); 40 C.F.R. Part 52,

§ 1881 (b) (47) (1984). Under section 110(c) of the Clean Air Act

(codified at that time in 42 U.S.C. § 1857c-5(c) ), EPA promulgated

a plan for control of sulfur dioxide emissions, including specific

emission limitations for individual sources, for the State of Ohio.

The Clean Air Act was substantially amended in 1977 and recodi-

fied at 42 U.S.C. § 7401 et seq. (Supp. 1979). The agency’s author-

ity to promulgate emission limitations was left intact by the 1977

amendments, see 42 U.S.C. § 7410(c), but the amendments did

significantly expand the statutory procedures applicable to EPA’s

promulgation of an emission limit, see id., § 7607(d). a

4

Having found the agency’s action arbitrary and capri-

cious, the court then described its mandate to EPA:

“Our ultimate action on this score is simupiy to

remand this issue to United States EPA for surther

_ study.” App. 17a.

The court did not state in its decision that the emission

limitation established by EPA was set aside.”

- In the administrative proceedings that followed, EPA

read the court’s mand. te as meaning that it could meet

its obligations on remand by merely supplementing the

record to support its use of the Class A coefficients. EPA

~ announced its intent to develop this supplemental record

in February, 1979, and solicited comments for inclusion

in the record. 44 Fed. Reg. 7798-7805. In June, 1980,

EPA announced its final development of the supplemental

record. App. 28a-62a. EPA contended at that point that

it had fully discharged its legal obligations on remand by

conducting a “further study” of the Class A issue. The

agency rejected Ohio Power Company’s argument that it

had to promulgate an emission limitation to replace the

emission limitation which had been found arbitrary and

capricious. In EPA’s view, the 1976 limitation had never

been set aside, and the court’s mandate authorized it to

conduct a separate proceeding to “study” the Class A

issue without connecting that proceeding to the promul-

gation of an emission limitation. App. 58a-59a. As a

‘consequence, EPA believed it was only required to show

that the supplemental record supported the Class A

coefficients, and not the 1976 emission limit. Jd.

Ohio Power Company’s petition for administrative re-

consideration of EPA’s decision was denied (App. 71la-

93a), and at the same time, the agency announced that

2 The Court of Appeals issued a subsequent decision “clarifying”

its earlier order, but again did not indicate that it had set aside

the emission limitation. App. 23a-27a.

5

the schedule for Ohio Power Company’s bringing its

powerplant into compliance with the 1976 limitation was

“restarted,” effective June, 1980 (App. 60a-70a).

Following these decisions, Ohio Power Company again

petitioned the Court of Appeals for review. Ohio Power

argued that the emission limitation had necessarily been

set aside by the court’s prior ruling that the ageicy’s

action was arbitrary and capricious.* Ohio Power re-

quested that the case be remanded to EPA for promulga-

tion of an emission limitation.

This request was not merely a matter of form. The

agency, in its efforts to establish the propriety of the

Class A coefficients on remand, used a different modeling

procedure than it had used in establishing the emission

limitation.* It did not, however, use that revised model-

ing procedure to reevaluate the propriety of the emission

limitation—the most significant matter at issue. Had it

done so, Ohio Power argued, a substantially less stringent

emission limitation would have resulted. The revised

modeling procedure employs the Class A coefficients but

results in an emission limitation with which the Com-

pany’s plant is already in compliance. Thus, promulga-

tion of an emission limitation to replace the one that

was held to be arbitrary would moot the entire contro-

versy. The costs involved in EPA’s refusal to repromul-

gate an emission limitation are substantial—approxi-

mately $29 million annually for Ohio Power Company.

*Ohio Power Company pointed out that, in another case, the

same panel that remanded the Class A issue to EPA had described

its remand in this case as having “set aside EPA’s emissions

limits.” Columbus & Southern Ohio Electric Co. v. Costle, 638 F.2d

910, 911 (6th Cir. 1980) (emphasis added). Based on that descrip-

tion, it appeared possible—at least prior to the court’s March, 1984,

decision—that the court’s failure to state that it had set aside the

emission limitation was inadvertent and that the court would reject

EPA’s interpretation of the remand order.

* The revised modeling procedure was based on both a new model

and different input data. App. 87a-90a.

6

The Court of Appeals, however, refused to address any

issues relating to the emission limitation. It ruled that

the emission limitation was not at issue in the case, since

it had “carefully limited the remand to a single issue

and we should now decide only that issue.” App. 6a

(emphasis added.) That issue was: “[{W]hether or not

the EPA, since the remand, has properly carried out this

court’s mandate to undertake further study of the basis

for the use of Class A coefficients.” App. 5a. Having

narrowed its scope of review to the single issue of

whether the agency had conducted “further study” of the

Class A issue, the court then found that EPA “did re-

consider and introduce into the record additional support

for its original finding that the . . . Class A coefficients

should be used and we now affirm.” App. 6a.

REASONS FOR GRANTING THE WRIT

This case presents an important and unsettled aber

of administrative law: Whether a court must set aside an

agency rule that it has found to be arbitrary and capri-

cious and remand it to the agency for repromulgation in

vompliance with the relevant statutory requirements. The

two Sixth Circuit decisions at issue here® stand for the

proposition that a court has the authority to remand

selected “issues” to an agency for “further study” with-

out setting aside the rule found to be arbitrary and ca-

pricious and without requiring the agency’s rule to be

repromulgated. |

This approach conflicts with that taken by the majority

of lower federal courts, which have held that the proper

5 As indicated in the Jurisdictional Statement, both decisions of

the Sixth Circuit are appropriately before this Court for review

since the 1978 decision was not a final judgment. See p. 2, supra.

7

remedy for an arbitrary and capricious agency action is

“to vacate the action and to remand the matter back to

the agency for further consideration.” EDF v. Costle,

657 F.2d 275, 285 (D.C. Cir. 1981); see also California

Cartage Co. v. United States, 721 F.2d 1199, 1202 (9th

Cir. 1983); Action on Smoking and Health v. CAB, 713

F.2d 795 (D.C. Cir. 198%); Port Norris Express Co. Vv.

ICC, 687 F.2d 803 (3d Cir. 1982); Central Power &

Light Co. v. United States, 634 F.2d 137, 150 (5th Cir.

1980) ; Montana-Dakota Utility Co. v. FERC, 631 F.2d

557, 562-63 (8th Cir. 1980); Tabor v. Joint Bd. for En-

_ rollment of Actuaries, 566 F.2d 705, 711-12 (D.C. Cir.

1977). Decisions from other lower federal courts, how-

- ever, have taken an approach similar to that of the Sixth

Circuit. See FMC Corp. v. Train, 539 F.2d 973, 985 (4th

Cir. 1976) (standards set under Clean Water Act “re-

manded to the Agency for further documentation’) ;

American Meat Institute v. EPA, 526 F.2d 442 (7th Cir.

1975 (remand for the Agency to consider whether evi-

dence already in the record would support rule); NRDC

v. EPA, 494 F.2d 519, 526 (2d Cir. 1974) (remanding

agency action based on “unrealistic” computer model for

further study on specific issues) .®

' This Court has not directly addressed the issue of the

type of remand necessary after; an agency rule is invali-

dated. The Court’s decision in Camp v. Pitts, 411 U.S.

It should be noted that the circumstances presented in the

above cases vary. In American Meat Institute, for example, the

court offered the agency the alternative of supporting its arbitrary

and capricious regulation with evidence already in the record

but not previously relied on. The agency would be allowed to follow

that course of action without repromulgating the regulation. The

court also made it clear, however, that if the agency developed new

evidence for the record, it had to repromulgate the rule. 526 F.2d

at 467. The Seventh Circuit’s action, therefore, is a more moderate

version of the Sixth Circuit approach, which authorized the agency

to introduce new evidence into the record without repromulgation

of the rule.

g

135 (1973) (per curiam), holds that an arbitrary agency

decision “must be vacated,” id. at 148. However, since

the agency action at issue in Camp v. Pitts did not in-

volve an exercise of agency rulemaking authority, the

Court did not deal with the scope of the ensuing remand.

The Court should grant certiorari in this case to resolve

the confusion among the lower federal courts and to

clarify the principle stated in Camp v. Pitts for agency

rulemaking. The statutory, constitutional and policy

issues involved are substantial. The scope of a court’s

remand authority relates directly to the question of the

proper boundaries of the relationship between courts and

administrative agencies; the proper interpretation of the

Administrative Procedure Act; and the “case or con-

troversy” requirement of Article III of the Constitution.

The present case is, moreover, a suitable one for the res-

olution of these issues, since the Sixth Circuit’s failure

to set aside the agency’s action and remand for reprom-

ulgation has a substantial practical impact, amounting to

nearly $30 million per year for Ohio Power Company.

THE DECISIONS BELOW RAISE IMPORTANT AND

UNSETTLED QUESTIONS OF ADMINISTRATIVE

LAW ON WHICH THERE IS CONFLICT AMONG

THE LOWER FEDERAL COURTS

This Court’s decisions bracket, but do not directly

address, the issue of the proper scope of a remand order

following a decision that an agency has acted arbitrarily

and capriciously.

In Citizens to Preserve Overton Park v. Volpe, 401 U.S.

402 (1972), the Court held that a limited remand for an

explanation of agency action is appropriate if the grounds

for the agency’s decision were not adequately disclosed on

the record. See Camp Vv. Pitts, 411 U.S. 138, 143 (distin-

guishing Overton Park); see also Harrison v. PPG In-

dustries, Inc., 446 U.S. 578, 594 (1980) (reviewing court” » >

may remand for further administrative proceedings if

2 nee Ne Ome eae beee: ML

ae een oo. er

9

the record is inadequate to permit “informed judicial

review”). The purpose of the limited remand in that

circumstance is ‘to aid the court in its review, and the

limited remand is issued without the court’s reaching a

determination on the arbitrarmess or rationality of the

rule. See, e.g., Public Power Council v. Johnson, 674 F.2d

791 (9th Cir. 1982).

The Court’s decision in Overton Park, however, does

not sanction limited remands for supplementation where

the record before the court and the agency’s explanations

for its findings and actions are sufficiently developed to

allow judicial review. This distinction was clearly set out

in Camp V. Pitts. The Court in Camp stated that, where

the record is adequate for judicial review “[t]he validity

of the [agency’s] action must stand or fall on the propri-

ety of that finding, judged, of course, by the appropriate

standard of review.” Camp Vv. Pitts, supra, 411 U.S. at

143; see also FPC v. Transcontinental Gas Pipe Line

Corp., 423 U.S. 326 (1976) (per curiam).

This case falls squarely within the situation described

by Camp. The record before the court was fully adequate

to permit judicial review, and the court never suggested

otherwise. The court’s assessment, based on the record,

was that EPA’s use of the Class A coefficients to estab-

lish the emission limit was not a “rational decision”

and was “arbitrary and capricious.” App. 16a. Having

reached that conclusion, however, the court’s next step

was to remand the Class A “issue” for “further study.”

Its apparent conclusion—subsequently made explicit in its

decision after the remand proceedings—was that the

agency could enforce its original rule if it supplemented

the record on that one issue.

In Camp v. Pitts, the Court was not faced with the

issue of determining the scope of reconsideration on re-

mand necessary to resurrect an invalid rule. The agency

action at issue in Camp was not subject to rulemaking

10

procedures. As the Court noted, “neither the National

Bank Act nor the APA requires the Comptroller to hold

a hearing or to make formal findings on the hearing

records when passing on applications for new banking

authorities.” 411 U.S. at 140-41. .

A recent case interpreting Camp has pointed out:

“Where, as in Camp ..., the agency’s explanation is

required to be responsive to the purposes of the en-

abling statute, rather than to a record developed

through mandatory hearings or public comments,

post hoc rationalizations, while undesirable, are not

fatal.”

Action on Smoking and Health v. C.A.B., 713 F.2d 795,

798 n.2 (D.C. Cir. 1983) (citations omitted).

The court in Action on Smoking and Health went on to

address the open issue of what “steps an agency must

take to enter a valid new order... .” Jd. at 798. The

D.C. Circuit concluded that the agency had to conduct a

new inquiry, adhering to the same statutory require-

ments applicable to the initial promulgation. Facing the

situation left open in Camp, the court reasoned that

repromulgation was necessary to prevent post hoc ration-

alizations on the part of the agency and to provide the

parties on remand the full benefit of the applicable statu-

' tory provisions. Id.

The Sixth Circuit’s contrary conclusion in this case

raises two distinct, though interrelated, issues: First,

whether the failure to set aside an invalid rule and

remand for repromulgation is consistent with Congress’s

and this Court’s delineation of the proper relationship

between courts and administrative agencies; and, second,

whether the lower court’s subsequent limitation of judi-

cial review to the “issue” remanded to the agency is con-

sistent with Article III of the Constitution and the prin-

ciples governing judicial review of administrative action.

u

A. The Lower Court’s Failure to Set Aside and Remand

for Repromulgation Is Inconsistent with Congress’s

and this Court’s Delineation of the Proper Relation-

ship between Courts and Administrative Agencies.

The lower court viewed its reihand for supplementa-

tion as flowing from the proper rélationship between

courts and administrative agencies as outlined by this

Court’s decisions.’ The court noted that “decision-making

(particularly in this highly technical’area) is the primary

responsibility of the agency and not the responsibility of

this court,” App. 16a, citing Vermont Yankee Nuclear

Power Corp. v. NRDC, 435 U.S. 519 (1978), and Citi-

zens to Preserve Overton Park v. Volpe, 401 U.S. 402

(1971). The court apparently viewed a remand for “fur-

- ther study” as a less “intrusive” judicial action, and thus

more in keeping with the spirit of Vermont Yankee and

Overton Park, than setting aside the agency action and

requiring repromulgation of the emission limit.

The Court’s approach, however, was misguided and

contrary to the spirit of Vermont Yankee.

First, the court’s action was inconsistent with the

Administrative Procedure Act (APA). The Sixth Cir-

cuit’s initial review of EPA’s action was governed by the

7 The court may also have been influenced in its choice of a re-

mand approach by its perception of the health hazards posed by

powerplant emissions. App. 2a-4a; 6a. It is important to note, how-

ever, that the court misunderstood the purpose of the emission limi-

tation at issue in this case. The limitation was set to protect the

“secondary” national ambient air quality standard, not the “pri-

mary” standard. The primary standard is designed to protect

health; the secondary standard is designed only to protect other

general “welfare” concerns (e.g., vegetation). 42 U.S.C. §§ 7409

(b) (2); 7602(h) (defining “welfare” considerations). Indeed,

EPA never claimed that the limitation was necessary to protect

public health and stayed enforcement of the limitation pending its

Class A “study.” App. 64a.

12

judicial review provisions of the APA, 5 U.S.C. § 706,”

which state:

The reviewing court shall—

. (2) hold unlawful and set aside agency action,

findings and conclusions found to be—

(A) arbitrary, capricious, an abuse of

discretion or otherwise not in accordance

with law; ....

5 U.S.C. § 706(2) (A) (emphasis added).

There is nothing in the Sixth Circuit’s opinion, how-

ever, to suggest that the court recognized that the lan-

guage of the APA constrained its choice of a remand

order. Yet the Congressional instruction is unambiguous

that the court “shall set aside” arbitrary and capricious

rules. The APA certaintly contains no Congressional

authorization for a limited remand of an arbitrary and

capricious rule for “further study.”

This Court has previously noted that the requirements

of the APA represent a “statutory scheme through which

Congress enacted ‘a formula upon which opposing social

and political forces have come to rest.’” Vermont Yankee

Nuclear Power Corp. v. NRDC, supra, 435 U.S. at 547

(citation omitted). The Act’s requirement that arbitrary

rules be set-aside serves several vital purposes: It acts as

a restraint on judicial review by ensuring that agency

8 The court’s review of EPA’s actions under the Clean Air Act,

as in effect when EPA promulgated the emission limitation at

issue, was governed by the judicial review provisions of the APA.

5 U.S.C. §§ 701-706. See Buckeye Power, Inc. v. EPA, 481 F.2d

162, 171 (6th Cir. 1973), cert. denied, 425 U.S. 934 (1974). The

Clean Air Act Amendments of 1977 contain their own judicial

review provisions modeled after the APA. 42 U.S.C. § 7607(d).

The language in the Clean Air Act now states that the reviewing

court is to “reverse any .. . action found to be—arbitrary, capri-

cious... .” Id. § 7607(d)(9)(A). The legislative history of the

1977 Amendments clearly indicates that this language was in-

tended to parallel the APA, see National Lime Ass’n v. EPA, 627

F.2d 416, 452 and nn. 126-27 (D.C. Cir. 1980).

13

action will not lightly be found arbitrary; it obviates the

Due Process concerns that would be raised by an agency’s

attempting to enforce an arbitrary rule during the pen-

dency of a remand; and it protects the integrity of the

administrative process. The latter point is particularly

important in this case.

In the past, this Court has struck down lower courts’

attempts to impose particular procedures on remand. See

Vermont Yankee Nuclear Power Corp. v. NRDC, supra,

435 U.S. at 542-49 (court order requiring rulemaking

process on remand to include more extensive procedures) ;

FPC v. Transcontinental Gas Pipe Line Corp., supra, 423

U.S. at 333 (court order requiring agency to study issue

of natural gas shortage); FCC v. Schrieber, 381 U.S.

279, 290 (1965) (court order requiring confidential pro-

ceeding on remand) ; FCC v. Pottsville Broadcasting Co.,

309 U.S. 134 (1940) (court order requiring agency to

conduct proceedings on remand based on then-existing

record).

In this case, rather than imposing a special procedure

on the agency, the court authorized a special procedure.

The remand order was then used by EPA as justification

for limiting its scope of inquiry to the Class A issue and

for ignoring the ultimate question of the validity of the

emission limitation.? The court’s authorization of the

truncated inquiry on remand harms the administrative

process as much as the imposition of special procedures

in cases such as Vermont Yankee.

It is not the province of a reviewing court to assume

that an arbitrary and capricious rule can be cured by

® For example, in denying Ohio Power Company’s petition for

reconsideration, EPA briefly discussed the differences between the

modeling procedures it used in setting the limitation and those it

used to “study” the Class A coefficients. In that discussion, the

agency simply concluded that the differences were insubstantial for

purposes of its “study,” but never addressed whether those differ-

ences "vould result in a changed emission limitation. App. 87a-90a.

14

studying and supplementing the record on one particular

factual issue. An agency rule is an amalgam of factual

findings and poliey considerations applied to yield a final

action—here, an emission limitation. K. Davis, Adminis-

trative Law Treatise, §$ 7.1. 7.2 (1979). Removal of

one finding from the jackstraw pile of findings and

policy considerations that comprise the agency’s final

action can have effects that the court, lacking in technical

expertise, cannot anticipate. See SEC Vv. Chenery, 332

U.S. 194, 195-97 (1947). That is well-illustrated in

this case.

The computer models EPA uses to set emission limita-

tions are, in essence, sophisticated methods for processing

facts (data) in accordance with a given set of assump-

tions (such as the Class A coefficients). Changes in either

the input data or the method of processing the data can

alter the resulting emission limitation. The Class A co-

efficients can be used to produce any of a variety of

results depending on the data and computer model used.”

What the court’s remand authorized EPA to do was

to “study” and justify the Class A coefficients using a

model with a different set of data and assumptions than

those used in the model that produced the emission limi-

tation. By failing to set aside the emission limitation

and remand for repromulgation, the court invited EPA

to pull one set of assumptions out of the statutory deci-

sion-making context and deprived Ohio Power Company

of the opportunity to challenge the connection between the

‘agency’s finding of fact (i.e., the computer model used to

process data) and the agency’s action (the emission

limitation) .™

10 See p. 5, supra.

11 Computer models are an increasingly important tool in agency

rulemaking. Congress’s concern over the potential abuses of com-

puter modeling in the context of the Clean Air Act led it to re-

quire EPA to establish uniform modeling techniques and review

and update the models periodically as science develops. 42 U.S.C.

§§ 7475(e) (8) (D) and 7620.

15

In the context of adjudicatory proceedings, this Court

has emphasized that an order remanding an agency ac-

tion should not constrict the parties’ opportunity to raise

any and all objections to an agency’s action on remand.

See Ford Motor Co. v. NLRB, 305 U.S. 364, 374-76

(1939) (a proper remand “does not encroach upon ad-

ministrative functions” or restrict the parties’ right to

raise all objections to the agency’s action since “the set-

ting aside of the findings and order carrie[s] with it

the opportunity for reconsideration and the making of

a new record”); see also SEC v. Chenery, supra, 332

U.S. at 201 (“after the remand was made.. ., the

Commission was bound to deal with the problem afresh,

performing the function delegated to it by Congress’).

The same principle should apply in the context of in-

formal rulemaking.

A contrary conclusion would invite post hoc rationaliza-

tions. But see Burlington Truck Lines v. United States,

371 U.S. 156, 165-69 (1962). In this case, the court’s re-

mand encouraged post hoc findings. It allowed EPA to

keep the 1976 emission limitation in place, on the condi-

tion that the “further study” of the Class A coefficients

resulted in an affirmation of those coefficients. App. 17a.

Post hoc supplementation of the record was the path of

least resistance, allowing the agency to avoid reconsider-

ation of the emission limitation itself. .

The failure to set aside invalid agency action thus

provides the agency with an excuse for truncating the

administrative inquiry. The agency is invited to design

the inquiry on remand to satisfy the court’s limited or-

der, not its statutory mandate. That is what oceurred in

this case, and it affected both the proceedings on remand

and the subsequent judicial review.

16

B. The Lower Court’s Limitation of Judicial Review to

the “Issue” Remanded to the Agency Is Inconsistent

with the Proper Function of Judicial Review under

Article III of the Constitution.

In reviewing EPA’s response to its remand, the Sixth

Circuit held: “Thus the sole issue of practical signifi-

cance before this court at this time is the question as to

whether or not the EPA, since the remand, has properly

carried out this court’s mandate to undertake further

study of the basis for the use of Class A coefficients. .. .”

App. 5a. In fact, the Class A coefficients in isolation have

no practical significance; it is the emission limitation that

determines what Ohio Power must do. The issue the

court reviewed was whether EPA complied with its man-

date of “further study,” not. whether the agency com-

plied with Congress’s mandate that its emission limita-

tion be rational and supported on the record. 42 U.S.C.

§ 7607(d). Similarly, the court reviewed whether EPA’s

action on remand was adequate in terms of the “further

study,” but not whether its action was adequate for the

promulgation of an emission limitation. App. 6a, n.2.”

In short, the court substituted its interest in the Class A

issue for the Congressionally-mandated focus on the ra-

tionality of the emission limitation. -

In failing to review the issue of the emission limita-

tion, the lower court treated judicial review of agency

rulemaking as though it were a collaborative process in

which the court affirms findings and identifies problem

12 The Court briefly addressed the issue of whether EPA had to

comply with the procedural requirements of the Clean Air Act

relating to the promulgation of an emission limitation. App. 6a,

n. 2. The court held that EPA did not, since its “study” did not

involve promulgation of an emission limit and that, in any event,

the “study” was done with adequate notice and comment, making

the procedural error “harmless.” Jd. The court clearly did not

address whether the agency properly reconsidered the emission

limitation itself.

17

areas and the agency, in response, must only address the

problem areas.’* That approach may be appropriate in

the context. of judicial review of a lower court decision

but it is inappropriate for judicial review of agency

action. See FCC v. Pottsville Broadcasting Co., 309 U.S.

134, 140-44 (1940). As Justice Frankfurter stated in

Pottsville :

What is in issue is not the relationship of federal

courts inter se—a relationship defined largely by the

courts themselves—but the due observance by courts

of the distribution of authority made by Congress as

between its power to regulate commerce and the re-

viewing power which it has conferred upon the courts

under Article III of the Constitution.

Id, at 141,

At heart, the error in the decisions below was that the

Sixth Circuit perceived its judicial function as though it

were reviewing the resolution of the issue of the Class A

coefficients and not the action of the agency in imposing

an emission limitation. Courts, however, have no author-

ity to review issues”; their authority is to review “cases

or controversies”. U.S. Const. Art. III § 2, cl. 1; see

Valley Forge College v. Americans United For Separation

of Church and State, 454 U.S. 464, 472 (1982); Federal

Radio Comm. v. General Electric Co., 281 U.S. 464, 469

(1930) (a court on judicial review “cannot give decisions

which are merely advisory; nor can it exercise or partici-

pate in the exercise of functions which are essentially . . .

administrative’) . .

18 The Sixth Circuit was of the view that it had “approved the

EPA plan in all respects except one. The exception pertained to

the EPA’s employment of the Class A assumption... .” App. 4a.

The court did not explain how it could have “approved” the ’

emission limitation while striking down the agency’s use of the

Class A coefficients, particularly since the Class A coefficients, as

employed in the agency’s previous model, were the determining

factor in establishing the emission limitation. See App. 14a.

18

The requirement that arbitrary agency action be set

aside and remanded for repromulgation implements the

“case or controversy” limitation by ensuring that the

court reviews and acts upon the contested agency action

and that the proceedings on remand are directed to the

resolution of the relevant statutory issues, not the court’s

interest in an issue in the abstract.

As evidenced by the Sixth Circuit’s decisions, it is en-

ticing for a court to limit its remand to “issues.” That

approach appears to provide a short cut through what

may be perceived as a cumbersome administrative process

and a method of limiting the issues for subsequent judi-

cial review. But as the court stated in Pottsville, it is

an enticement that must be resisted.

It is . . . urged upon us that if all matters of

administrative discretion remain open for determina-

tion on remand after reversal, a succession of single

determinations upon single legal issues is possible

with resulting delay and hardship. . . . But courts

are not charged with general guardianship against

all potential mischief. . . . Congress which creates

and sustains these agencies must be trusted to cor-

rect whatever defects experience may reveal. Inter-

ference by the courts is not conducive to the develop-

ment of habits of responsibility in administrative

agencies. 3

FCC v. Pottsville Broadcasting Co., supra, 309 U.S.

at 146.

CONCLUSION

In several recent cases, this Court has attempted to

correct the inclination of the lower federal courts to in-

sert themselves into the administrative process and “im-

prove” upon the statutory structure created by Congress.

This case presents a comparable problem in a different;

though equally important, context. The Court should ad-

dress the issue of the proper scope of a remand order

after agency action is found arbitrary and capricious.

19

The issue is one that arises frequently and the lower

courts have reached conflicting conclusions. Several

strains from this Court’s prior decisions—the concerns

expressed over intrusive judicial review, post hoc agency

rationalizations, and the proper role of the courts under

Article I[1]—all suggest that the Sixth Circuit’s resolu-

tion of the issue left open in Camp was in error. Ohio

Power Company has been substantially harmed by that

error.

This Court should grant certiorari to resolve the issue

of the proper scope of a remand order and to clarify the .

‘ application of the principles announced in Camp v. Pitts

and Overton Park for agency rulemakings.

Respectfully submitted,

EDWARD BERLIN

Counsel of Record for Petitioner

FRANCIS S. BLAKE

SWIDLER, BERLIN & STRELOW,

CHARTERED

1000 Thomas Jefferson St., N.W.

Washington, D.C. 20007

(202) 342-5500

A. JOSEPH DowD

JEFFREY P. WHITE

OHIO POWER COMPANY

E. DONALD ELLIOTT

Box 401A Yale Station

New Haven, Connecticut 06520

Of Counsel

August 3, 1984

APPENDIX

=~ —-

la

APPENDIX

UNITED STATES COURT OF APPEALS

FOR THE SIXTH CIRCUIT

Nos. 80-3561, 80-3730, 80-3732, 80-3813,

77-1367, 76-2090

CHIO POWER COMPANY;

CINCINNATI GAS & ELECTRIC COMPANY, et al.,

Petitioners,

Vv.

UNITED STATES ENVIRONMENTAL PROTECTION AGENCY

and ANNE M. GorsucH, Administrator,

Respondents.

THE COMMONWEALTH OF PENNSYLVANIA, (80-3813),

Petitioner,

v. :

UNITED STATES ENVIRONMENTAL PROTECTION AGENCY

and ANNE M. GorSUCH, Administrator,

Respondents,

OHIO POWER COMPANY and

COLUMBUS AND SOUTHERN OHIO ELECTRIC COMPANY,

Intervenors.

On Petitions to Review Final Action of the

Administrator of the United States

Environmental Protection Agency

Decided and Filed March 22, 1984

2a

Before: EDWARDS, Circuit Judge; PHILLIPS and PECK,

Senior Circuit Judges.

EDWARDS, Circuit Judge. This appeal concerns the

problem of airborne sulphur dioxide emanating from

three major power plants located in rural areas in the

State of Ohio.t The pollution of the ambient air by

sulphur dioxide (produced by burning high sulphur coal),

is a medically recognized threat to human health. This

is particularly true as to infants, the ill and the aged.

It also contributes to acid rain with its subsequent dele-

terious impact on the plant and fish life in the lakes and

rivers of Ohio and neighboring states. Indeed it is al-

leged to contribute to the latter type of pollution in down-

wind areas as far removed as New England and East-

ern Canada.

The fundamental issues of this case were dealt with

and decided over five years ago in Cleveland Electric

Illuminating Co. v. EPA, 572 F.2d 1150 (6th Cir. 1978)

and Cincinnati Gas & Electric Co. v. EPA, 578 F.2d

660 (6th Cir. 1978), cert. denied, 489 U.S. 1114 (1979).

In those two opinions, this court approved the EPA’s

plan for controlling the ambient air affected by four

power plants and the Supreme Court denied certiorari.

This court’s basic rationale for its approval of the EPA’s

air quality standards was (and remains) concern about

human hezlth and safety. In our earlier Cleveland Elec-

tric decision, we said:

1(1) Stuart (Dayton Power & Light)—now in compliance, but

remaining in litigation;

(2) Cardinal (Ohio Power Co.)—limits not contested, but if

P-G Class A coefficients are again vacated, the limits are once again

in question; and

(3) Muskingum River (Ohio. Power Co.)—only plant in which

emission limitations are still contested.

One petition involved in the original appeal (that of Columbus

& Southern Ohio Electric Company’s Conesville plant) has been

voluntarily dismissed.

3a

The federal Clean Air Act program which pro-

duced these standards is based primarily upon the

adverse effect which sir poiution has upon human

life and health. 3

Acute episodes of high pollution have clearly re-

sulted in mortality and morbidity. Often the effects

of high pollutant concentrations in these episodes

have been combined with other environmental fea-

tures such as low temperatures or epidemic diseases

(influenza) which may in themselves have serious

or fatal consequences. This has sometimes made it

difficult to determine to what extent pollution and

temperature extremes are responsible for the effects.

Nevertheless, there is now no longer any doubt that

high levels of pollution sustained for periods of days

can kill. Those aged 45 and over with chronic dis-

eases, particularly of the lungs or heart, seem to be

predominantly affected. In addition to these acute

episodes, pollutants can attain daily levels which

have been shown to have serious consequences to

city dwellers.

a * o *

There is a large and increasing body of evidence

that significant health effects are produced by long-

term exposures to air pollutants. Acute respiratory

infections in children, chronic respiratory diseases in

adults, and decreased levels of ventilatory lung func-

tions in both children and adults have been found

~ to be related to concentrations of SO, and particu-

lates, after apparently sufficient allowance has been

made for such confounding variable as smoking and

socioeconomic circumstances.

Rall, Review of the Health Effects of Sulfur Oxides,

8 Env’tal Health Perspectives 97, 99 (1974).

* * * *

Two other facts should be added from the exten-

sive technical record in this case before we turn to

4a

the specific legal issues. The first is that sulfur

dioxide emitted from plant stacks reacts with other

elements in the atmosphere to form sulfuric acid

mist and various suspended sulfates which are in

fact the irritants which adversely affect human

health. T. Lewis, M. Amdur, M. Fritzhand & K.

Campbell, Toxicology of Atmospheric Sulfur Dioxide

Decay Products 17 (1972).

The second important fact is that these derivatives

from sulfur dioxide tend to be airborne for days.

They affect areas at great distances downwind, even

when in the original sulfur dioxide form they were

emitted from a high power plant stack.

Rall, Review of the Health Effects of Sulfur Oxides,

8 Env’tal Health Perspectives 97, 106 (1974).

572 F.2d at 1153, 1155.

While three of the original petitioners are parties to

this appeal, we are advised that two of the four are now

in compliance with the EPA standards which we ap-

proved in 1978. As to those two appellants, this appeal

is in effect a petition to roll back emission standards

which are currently being met, and thus to increase

SO, emissions in the ambient air in Ohio and elsewhere.

As indicated above, the plant not in compliance is

Ohio Power Company’s Muskingum River plant.

This court, in its opinion in Cincinati Gas & Electric

Co. v. EPA, supra, approved the EPA plan in all respects

except one. The exception pertained to the EPA’s em-

ployment of the Class A assumption in determining pol-

lution dispersion under least stable wind conditions in

rural areas. As to this issue, we remanded saying as

follows:

On this present record we conclude that United

States EPA’s employment of the Class A assumption

in determining pollution dispersion under “least

~

——— ee

«

>

5a

stable” wind conditions in rural areas and areas of

complex terrain is not a rational decision and is ar-

bitrary and capricious.

As we read EPA’s position on this point, it is that

no better solution has been proposed. This answer,

however, ignores the Enviroplan and Smith-Singer

studies. More importantly, it ignores the conclusions

of the experts’ conference convened by Argonne Na-

tional Laboratory at United States .EPA’s own re-

quest. The report of that conference suggested

“elimination of the A curve and the use of the B

curve for both A and B stability categories.” See

Report of the Specialists’ Conference on the EPA

Modeling Guideline, Feb. 22-24, 1977, Chicago, II-

linois, at § 2.7.5: “Vertical Dispersion Estimates.”

mandate its acceptance by United States EPA. Our

ultimate action on this score is simply to remand

this issue to United States EPA for further study.

This might result in the writing of a new record

which supplies the now missing support for the use

of Class A coefficients for the least stable wind con-

dition in rural counties, or it might result in United

States EPA’s adoption of use of the Class B co-

efficients for the two least stable wind conditions, or

it might result in a new record which supports a:

wholly new solution.

578 F.2d at 663-64.

Thus the sole issue of practical significance before this

court at this time is the question as to whether or not

the EPA, since the remand, has properly carried out this

court’s mandate to undertake further study of the basis

for the use of Class A coefficients and whether, if so, the

record now supports EPA’s renewed determintion that

the Class A coefficients should continue to be used for

power plants in rural areas.

By pointing out this proposed solution, we do not

6a

We now have this case back before us with many

newly stated arguments from EPA’s opponents® and

some from Intervenor State of Pennsylvania.

After a lengthy hearing (which included the State of

Pennsylvania’s contentions), the court concludes that we

carefully limited the remand to a single issue and that

we should now decide only that issue. This record shows

that EPA did re¢onsider and did introduce into the

record additional support for its original finding that the

Pasquill-Gifford Class A coefficients should be used and

we now affirm.

Since the beginning, this case has been the subject of

careful study by this court because of 1) the complexity

and technicality of the record; 2) the economic impor-

tance of the decision to the industry concerned; and 3)

the health hazards to the general public. Yet our review

is limited by a very narrow standard. Congress has

provided :

(9) In the case of review of any action of the Ad-

ministrator to which this subsection applies, the court

may reverse any such action found to be—

(A) arbitrary, capricious, an abuse of discretion, or

otherwise not in accordance with law;

42 U.S.C. § 7607(d) (9) (A) (Supp. I 1977).

The Supreme Court has elaborated upon this standard:

Under the “arbitrary and capricious” standard the

scope of review is a narrow one. A reviewing court

2 Petitioners also contend that EPA was required on this remand

to follow the formal notice requirements of § 307(d) of the Clean

Air Act, 42 U.S.C. § 7607(d). We agree with EPA, however, that

the reconsideration ordered by this court did not constitute either

promulgation or revision of an implementation plan. Additionally,

if we were to assume the applicability of § 307(d) to this proceed-

ing, any deviation from the § 307(d) requirements under the record

contained here would be harmless within the meaning of 42 U.S.C.

§ 7607 (d) (9).

a ee ee Ps

Ta

must “consider whether the decision was based on a

consideration of the relevant factors and whether

there has been a clear error of judgment. .... Al-

though this inquiry into the facts is to be searching

and careful, the ultimate standard of review is a

narrow.one. The court is not empowered to sub-

stitute its judgment for that of the agency.” Citizens

to Preserve Overton Park v. Volpe, supra, at 416.

The agency must articulate a “rational connection

between the facts found and the choice’ made.”

Burlington Truck Lines v. United States, 371 U.S.

156, 168 (1962). While we may not supply a rea-

soned basis for the agency’s action that the agency

itself has not given, SEC v. Chenery Corp., 332 US.

194, 196 (1947), we will uphold a decision of less

than ideal clarity if the agency’s path may reason-

ably be discerned. Colorado Interstate Gas Co., V.

FPC, 324 U.S. 581, 595 (1945).

Bowman Transportation, Inc. Vv. Arkansas-Best

Freight System, Inc., 419 U.S. 281, 285-86 (1974).

Undertaking the review called for concerning the issue

in this case, we find that the agency took the following

actions:

1) At a meeting of EPA Region V staff representatives

on August 15, 1978, this court’s remand of the Pasquill-

Gifford dispersion coefficients was considered. EPA there-

after decided to continue to use the Class A coefficients

“until an alternative approach is sufficiently demonstrated

to. be more adequate.”

2) February 7, 1979, EPA published a Notice of Re-

consideration in 44 Fed. Reg. 7,798 (1979), concerning

1) the use of Class A coefficients at the four power plants

then concerned; and 2) extending the time necessary for

compliance with emission limitations.

3) EPA also considered and rejected the solution pro-

posed by appellants, namely, the use of Class B dispersion

8a

coefficients “as contrary to both dispersion theory and

available data.” It based the rejection in part upon data

taken from studies at the Karlsruhe Nuclear Research

Center in Germany. These studies, EPA found, supported

use of the Class A coefficients.

4) A period followed when EPA received and con-

sidered objections filed by the appellants claiming 1) in-

applicability of the Karlsruhe data and EPA’s mis-

interpretation of it; 2) EPA’s failure to consider all

alternatives; and 3) its failure to take into proper ac-

count “The enormous economic costs involved.”

5) On June 19, 1980, EPA announced and published

its intention to continue to employ the Pasquill-Gifford

Class A coefficients (See 45 Fed. Reg. 41,501 (1980) ).

It subsequently denied appellants’ motions to reconsider.

This court, of course, has no responsibility for deter-

mining the merits (or lack thereof) of the technical

' standard EPA has adopted. Our duty ends when we are

able to say that the agency has acted lawfully and that

the decision is not “arbitrary, capricious or an abuse of

discretion.” We find no violation of this standard on

review of this record. The instant petitions for review

are dismissed. :

ee a ‘ —

9a

UNITED STATES COURT OF APPEALS

SIXTH CIRCUIT

Nos. 76-2090, 77-1367, 76-2232, 77-1361, 76-2241,

77-1357, and 76-2278

CINNCINNATI GAS & ELECTRIC Co., COLUMBUS & SOUTH-

ERN OHIO ELECTRIC Co., DAYTON POWER & LIGHT Co.,

OHIO EDISON Co., OHIO POWER Co., SHELL OIL Co.,

AUSTIN POWDER Co., E. I. du PONT de NEMOURS & Co.,

Petitioners,

Vv.

ENVIRONMENTAL PROTECTION AGENCY and DoucGLas M.

CosTLE, ADMINISTRATOR OF THE ENVIRONMENTAL PRO-

TECTION AGENCY,

Respondents.

Argued Oct 21, 1977

Decided June 29, 1978

Louis E. Tosi, C. Randolph Light, Fuller, Henry, Hodge

& Snyder, Toledo, Ohio, Daniel W. Kemp, Cincinnati,

Ohio, for Cincinnati Gas & Elec. Co., Shell Oil Co., et al.

James Van Carson, Squire, Sanders & Dempsey, Cleve-

land, Ohio, for Austin Powder Co.

Carl B. Everett, Legal Dept., E. I. du Pont de Nemours

& Co., Wilmington, Del., for E. I. du Pont de Nemours &

Co.

i ee ee i ee

= —— - =

10a

Ronald C. Hausmann, E.P.A., Paul M. Kaplow, Dept.

of Justice, Washington, D.C., Mary Ann Muirhead, Region

V, E.P.A., Chicago, Ill., Ned Williams, Director, Ohio

E.P.A., Columbus, Ohio, for respondents.

Before PHILLIPS, Chief Judge, and EDWARDS and

PECK, Circuit Judges.

EDWARDS, Circuit Judge.

This opinion deals with certain additional issues pre-

sented in 23 industry petitions covering 32 major power

and industrial companies in Ohio objecting to the United

States EPA’s plan for control of SO, pollution in Ohio.

These additional issues concern only point sources of SO,

pollution in Ohio’s rural areas or areas with complex

terrain. This opinion should be read as supplementary to

the opinion of this court dated February 13, 1978, Cleve-

land Electric Illuminating Co., et al. v. Environmental

Protection Agency, et al., 572 F.2d 1150 (6th Cir. 1978).

Our focus herein is upon the following petitioners and

the designated facilities belonging to them. These peti-

tioners protest certain features of the United States EPA

model (MAXT-24) employed for predicting plant pollu-

tion in rural and complex terrain areas: .-

Nos. 76-2090, 77-1367:

Cincinnati Gas & Electric Co., all facilities (Hamil-

ton & Clermont Counties).

Columbus & Southern Ohio Electric Co., all facilities

(Athens, Coshocton, Pickaway Counties).

Dayton Power & Light Co., all facilities not covered

by opinion dated February 13, 1978 (Adams County).

Ohio Edison Co. (Jefferson County) Sammis Plant

only.

Ohio Power Co., all facilities (Washington and Mor-

gan Counties).

lla

No. 76-2278:

E. I. du Pont de Nemours & Co., all facilities (Hamil-

ton County).

Nos. 76-2282, 77-1361:

Shell Oil Co., all facilities (Washington County).

Nos. 76-2241, 77-1357: |

Austin Powder Co., all facilities (Vinton County).

The MAXT-24 model (Second Maximum 24-Hour Dis-

persion Model with Terrain Adjustments) is designed for

use in predicting SO, pollution resulting from single

sources located in rural areas. Unlike the RAM model

employed in urban areas, which we dealt with in Cleve-

land Electric Illuminating Co., supra, MAXT-24 does not

provide estimates of comparative contributions to total

SO, pollution from a number of point sources. The

MAXT-24 model treats each point source as an isolated

problem, and only general background SO, pollution data

are added into the formula.

In other respects the MAXT-24 model strongly re-

sembles the RAM model.* Thus, like RAM, MAXT-24

starts with a solid ascertainable data base, namely, the

established design capacity of the power or steam gen-

erating plants in question related to the sulfur content

of the fuel used by such plants. Emissions data are

developed from these factors. Subsequently, stack height,

wind, weather, and terrain data are added. Like RAM,

MAXT-24 employs a Gaussian plume formula and as-

sumes vertical and horizontal dispersion of the pollution

plume. It employs the Pasquill-Gifford stability classifi-

cations and coefficients.

Like RAM, the MAXT-24 model was designed by United

States EPA largely as a result of industry criticism of

1 See Cleveland Electric Illuminating Co., supra, Section 3, 572

F.2d at 1160-64.

, 12a

the use of rollback modeling. As was true in relation to

the RAM results, the results of use of MAXT-24 were

generally less strict than those contemplated by the 1972

and 1974 Ohio EPA SO, regulations.”

Indeed, the comments this court made in Cleveland |

Electric Illuminating Co., supra, in Section 3 of the

opinion are largely applicable to EPA’s adoption of

MAXT-24 and we cite and rely on said Section 3 in hold-

2 The following summary is drawn from Cleveland Electric Il-

luminating Co., supra, Appendix A, 572 F.2d at 1165-74:

OhioEPA Ohio EPA

US EPA 1976-77 1972 1974

MAX regs are: regs for: regs for:

1. less strict than 19 16 of petitioners’

facilities *

stricter than 1 5 -

ambiguous ”

compared with 3 2

2. less strict than 31 27 of Ohio counties

modeled entirely

with MAX

stricter than 3 *

ambiguous ”

compared with 5 6 es

3. less strict than 38 82 of Ohio counties

in which MAX

: was employed

stricter than 3 6 7

ambiguous ”

compared with 12 15 -

*Including facilities to the regulation of which petitioners do not

object.

> /.e., stricter for some stacks or facilities and less strict for others;

or employing different units of measurement, rendering comparison

impossible.

The 1972 Ohio EPA plan was submitted to United States

EPA on January 30, 1972, but was “withdrawn” by the Gov-

ernor of Ohio on August 27, 1972. The 1974 Ohio EPA plan

was submitted to United States EPA on September 22, 1974,

and was withdrawn on July 16, 1975. See Cleveland Electric

Illuminating Co., supra. 572 F.2d at 1156.

13a

ing that in general (and with one exception noted below)

the EPA’s adoption and use of the MAXT-24 model is

not arbitrary or capricious and, like the use of the RAM

model, must be affirmed by this court.

Despite the discussion above, we are not certain that

any of the petitions we deal with in this opinion seriously

disputes the general] validity of the MAXT-24 model.

What these petitioners clearly do contend is that the

MAXT-24 model results are badly skewed to their great

economic disadvantage by 1) the Class A assumption em-

ployed to estimate pollution dispersion in the least stable

wind condition, and 2) the failure of EPA to employ the

half ground displacement theory in estimating pollution

impact on hillly terrain.

I THE.CLASS A ASSUMPTION ISSUE

The MAXT-24 model makes use of a set of six co-

efficients for determining plume dispersal. The classes

of coefficients employed were based upon six different

weather conditions. The term Class A is employed to

describe both the least stable weather condition and the

set of assumptions which is based on the most direct and

- quickest impact of the pollution plume upon ground level

with the least prior dispersal.

The six Pasquill-Gifford coefficients employed in MAXT-

24 are derived from a Nebraska study made in the 1950’s

and are referred to by United States EPA as “time-

tested” What this defense appears to ignore, however, is

that petitioners in this instance (contrary to the general

attack upon the six coefficients employed in RAM) are

not objecting to the use of the coefficients; they are attack-

ing the accuracy of one set of them—the Class A set

associated with “gusty winds.” Specifically they claim

that the Class A assumption is fallacious in that it as-

sumes a longer period of downward draft than occurs in

fact and fails to make allowance for the lateral dispersal

which would accompany such a vertical wind at the point

of impact.

l4a

The lead brief for the utilities presents the case thus:

In all modeling of rural power plants, EPA utilized

dispersion coefficients under Class A stability eondi-

tions which have no support in data, which have been

repudiated by most modelers and which are demon-

strated inaccurate by this record. As applied to this

rulemaking, this seemingly simple assumption is ex-

ceedingly important because, for almost 1% of the

power plants in Ohio, it was the determining factor

in establishing emission limits.

The meaning of “Class A.”

Diffusion models can account for thousands of bits

of data. Most important are meterological data of

which stability classes are an aspect.

Specifically, stability classes are categorizations of

the atmosphere’s ability to disperse plumes. These

classes are divided into six categories ranging from

extreme dispersion of plumes (Class A) to minimum

dispersion (Class F). Under Class A, a plume is

assumed to disperse very rapidly to the ground level -

before there is any substantial dilution. This, in

turn, leads to predictions of high ground level con-

centrations. The fundamental issue, therefore, is

whether the Class A assumption describes the manner

in which plumes disperse at rural power plants and

whether the phenomena it depicts really occur.

Brief of Utility Petitioners at 31-32 (emphasis in

original).

Petitioners then detail the results of three separate

studies which they claim attack and undermine the va-

lidity of the Class A coefficients, and generally urge

substitution of Class B coefficients. These studies are

the privately financed study by Environplan, Inc., a

similarly produced study by Smith-Singer Meteorologists,

SSE eS ee ee erhrh Oc lL

15a

and a strongly critical report resulting from the Spe-

cialists’ Conference of February 22-24, 1977, sponsored

by United States EPA itself through the Argonne Na-

tional Laboratory.

To this argument the EPA’s response is as follows:

EPA properly determined that the “Class A” sta-—

bility factors should continue to be used until new

field data proved them incorrect.

Petitioners argue that EPA should have changed

the dispersion coefficients used in the rural MAX

(CRSTER) model for analyzing ground level con-

centrations caused by a source in very unstable

weather, known as “Class A” conditions. In the

remand comment period, the utility petitioners pre-

sented various theories that the model did not ac- °

curately reflect the way wind patterns in such

weather conditions affect dispersion patterns and

that therefore the model might be overestimating

ground level concentrations for a 3-hour analysis.

Petitioners argue that it was arbitrary or capricious

for EPA not to accept theories presented in their

comments.

EPA recognized in the STSD [Supplemental Tech-

nical Support Document] at 55 that there was a

growing concern among atmospheric modeling scien-

tists about the issue. EPA determined, however,

that until further studies could be done to substanti-

ate the theories, there was no experimental or field

data to justify changing the dispersion curves or to

determine how the dispersion equations should be

changed. And since petitioners did not submit any

data, no change could be made in the equations used.

Id.

EPA Brief at 48-49.

164

We are, of course, aware that decision-making (par-

ticularly in this highly technical area) is the primary

responsibility of the agency and not the responsibility of —

this court. See Vermont Yankee Nuclear Power Corp. v.

NRDC, 435 U.S. 519, 98 S.Ct. 1197, 55 L.Ed.2d 460

(1978). As we said in Cleveland Electric Illuminating

Co., supra:

Our standard of review of the actions of United

States EPA is whether or not the action of the

agency is “arbitrary, capricious, an abuse of discre-

tion, or otherwise not in accordance with law.”

Clean Air Act Amendments of 1977, Pub.L. No. 95-

95, §305(a), 91 Stat. 775 (to be codified as 42

U.S.C. § 7607(d) (9) (A)). Thus, we are required

to affirm if there is a rational basis for the agency

action and we are not “empowered to substitute

[our] judgment for that of the agency.” Citizens to

Preserve Overton Park v. Volpe, 401 U.S. 402, 416

[91 S.Ct. 814, 28 L.Ed.2d 136] (1971).

572 F.2d at 1161.

On this present record we conclude the United States

EPA’s employment of the Class A assumption in deter-

mining pollution dispersion under “least stable’ wind

conditions in rural areas of complex terrain is not a

rational decision and is arbitrary and capricious.

As we read EPA’s position on this point, it is that no

better solution has been proposed. This answer, however,

ignores the Enviroplan and Smith-Singer studies. More

importantly, it ignores the conclusions of the experts’

conference convened by Argonne National Laboratory at

United States EPA’s own request.* The report of that

* This conference was initiated by United States EPA and one

of the participants was the Director of EPA Region V, which

Region includes Ohio. The conference occurred during the remand

Way

conference suggested “elimination of the A curve and the

use of the B curve for both A and B stability categories.”

See Report of the Specialists’ Conference on the EPA

Modeling Guideline, Feb. 22-24, 1977, Chicago, Illinois, at

§ 2.7.5: “Vertical Dispersion Estimates.”

By pointing out this proposed solution, we do not man-

date its acceptance by United. States EPA. Our ultima’e

action on this score is simply to remand this issue to

United States EPA for further study. This might result

in the writing of a new record which supplies the now

missing support for the use of Class A coefficients for

the least stable wind condition in rural counties, or it

might result in United States EPA’s adoption of use of

the Class B coefficients for the two least stable wind

conditions, or it might result in a new record which

supports a wholly new solution.

II THE TERRAIN ADJUSTMENT PROBLEM

The original MAXT-24 model assumed that the pollu-

tion plume moves downwind in a straight line from a

point determined by the height of the smoke stack plus

plume rise (“effective stack height”). Under that as-

sumption, if effective stack height is 600 feet and there

is a hill 800 feet high downwind, a receptor site located

600 feet up that ‘hill will therefore in theory receive the

full impact of the pollutants in the center of the plume.

Petitioners’ attack upon this set of assumptions is, we ©

believe, best stated by the Shell Oil Co. brief:

The reference to “the effect of the terrain on the

plume” is an easily understood concept. Since the

wind which blows the plume toward a hill (terrain)

period of this litigation, and three months before the finally

amended regulations were promulgated. We consider the Confer-

ence Report to be properly a part of the appellate record.

18a

cannot blow through the hill, it must blow up and

over the hill. This effect causes all the layers of air

above the wind at ground level, and hence the plume

itself, to be carried up and over the hill rather than

hitting directly into its side. A widely recognized

means of accounting for such a situation, and one

which EPA has used, is to incorporate mathematical

changes in the model which reduce the receptor height

by the one-half difference between stack base and

receptor elevation and limit the approach of the

center line of the plume to ten meters above the

receptor.

Shell and its consultant, Enviroplan,, recommended

this change to U.S. EPA. Moreover, this change

was supported in the modeling literature by two

other independent experts—Briggs and Egan. Also,

another consultant, Environmental Research & Tech-

nology, Inc. (“ERT”), recommended the same ad-

justment in a report submitted to EPA during the

comment period for Columbus & Southern Ohio Elec-

tric Co. See “A Technical Review of the U.S. EPA

Ohio State Implementation Plan for Sulfur Dioxide,”

January 1977, prepared for C&SOE. (App. 210-213.)

Indeed, this approach is so well recognized that EPA,

Region II, approved its use in sustaining a revision

of the implementation plan for Puerto Rico. See

40 Fed. Reg. 52410 (1975).

Brief of Shell Oil Co., at 19-20..

As to this argument, the EPA brief contains this

comment and admission:

Contrary to petitioners’ claims, EPA utilized avail-

able monitoring data wherever possible. As set forth

in detail in both the Final and the Supplemental

Technical Support Documents, EPA conducted valida-

tion studies of the dispersion model used to set

emission limitations for isolated, rural power plants.

19a

See STSD at 53-55, and the FTSD at 27-34. The

validation studies compared model predictions of SO.

ground level concentrations to actual air quality

monitor data. These comparisons indicated that for

sources located on flat terrain, the correlation be-

tween monitor data and predictions was quite good

with the model tending to underpredict, but that for

power plants located on hilly terrain, the compari-

sons showed consistent overpredictions.

EPA Brief at 45-46 (footnote omitted).

The EPA brief then goes on to assert that cootain

adjustments have been made in the model “so that it

could handle dispersion in hilly terrain more accurately,”

and then cited the Supplemental Technical Support Docu-

ment at page 55. The STSD materiai referred to follows:

The validation studies which compared model pre-

dictions of SO, ground level concentrations to actual

air quality monitor data indicated that in certain

situations the model overpredicted and needed modi-

fication. The problem usually occurred when air

quality monitors were at elevations higher than the

top of the stack. To correct this, terrain data used

in the model was limited in such a way that terrain

features were always assumed to be no higher than

the stack height of the source stack in question.

This was deemed to be an appropriate adjustment

because the validation study showed a high degree

of correlation between model’ predictions and sample

readouts from monitors positioned on terrain. lower

than stack height in elevation. When this assump-

tion was mathematically incorporated into the model,

the validation studies showed that the. model ac-

curately predicted the ground level concentrations

observed by the monitors.

Petitioners proposed a different method for modi-

fying the model to account for complex terrain situa-

20a

tions, but the proposal is not based on any valida-

tion studies of the CRSTER model. The Agency

has no way of determining if the proposal is a better

modification to the model than the modification

made by the Agency after the validation studies.

The Agency, therefore, has determined that the

model does not need further modifications because of

any information presented by the petitioners.

EPA Brief at 54-55.

While the record does not establish conclusively that

this adjustment made by United States EPA in the re-

mand period will prove a satisfactory solution to the

problem posed, neither does the record offer evidence to

the contrary. We note, of course, that United States EPA

has disowned the apparent implication in its brief that

it had made validation studies of this latest adjustment

for hilly terrain. And, in fact, our holding on this issue

should not be read as this court’s rejection of petitioners’

half ground displacement theory in favor of the United

States EPA adjustment outlined above. There may well

be occasion for the agency to continue to review this is-

sue. All we hold is that on the present record, we can-

not find that United States EPA’s present terrain ad-

justment in MAXT-24 is “arbitrary or capricious.”

For the reasons indicated above, the petitions of Day-

ton Power & Light Co. (Adams County facilities only),

Ohio Power Co. (all facilities), Columbus & Southern

Ohio Electric Co. (Coshocton County facility only), and

Austin Powder Co. (all facilities) are remanded to the

United States EPA for reconsideration of the employ-

ment of Class A coefficients in least stable wind condi-

tions in rural counties.

All petitions referred to at the beginning of this opin-

ion are denied to the extent that they attack the MAXT-

24 model as to the terrain adjustment feature.

2la

III OTHER ISSUES

that there is no merit to objections based

e to calibrate the MAXT-24 model (or failure to

reject its results because of claims of overprediction as

demonstrated by some monitor readings). See Cleveland

Electric Illuminating Co., supra, 572 F.2d at 1163-64,

numbered paragraph 7.

Shell contends that its emission limitation should be

expressed in terms of pounds SO, per hour rather than

pounds SO, per million British Thermal Units. The

Shell proposal would require the EPA either to assume

that stack gas temperature and exit velocity (the impor-

tant factors bearing on plume rise and thus ultimately

on ground level SO, concentration) are relatively con-

stant, or alternatively to monitor stack gas temperature

and exit velocity. Clearly, EPA considers policing such

a system to be an impossible task. EPA’s formula, by

contrast, requires only the use of fixed, easily ascertain-

able date—the plant’s design-rated capacity. We regard

EPA’s choice of formula, which minimizes administra-

tive costs while obeying the Clean Air Act’s command to

“insure attainment and maintenance” of national am-

bient air standards, 42 U.S.C. § 1857c-5(a) (2) (B)

(1970),* to be within the range of the agency’s dis-

cretion.

We have considered the other issues raised by Shell

and find them to be without merit.

Disputes between petitioners and EPA concerning ap-

propriate SO, background levels, emission data, or other

fact issues will not be decided by this court until com-

pletion of the administrative review of such issues which

was suggested by this court and agreed upon by the

parties.

4 Newly recodified at 42 U.S.C.A. § 7410(a) (2)(B) (1977 Pam-

phlet) .

22a

Based upon what has been said by this court in Cleve-

land Electric Illuminating Co. v. EPA, supra, and in this

opinion, and finding no other material issues, we dismiss

the following petitions in toto: Cincinnati Gas & Electric

Co:, Shell Oil Co.

Final dispositions in the petitions of Columbia & South-

ern Ohio Electric Co., Ohio Edison Co., and E. I. du

Pont de Nemours & Co. will be entered on resolution

of the remaining issues therein.

23a

UNITED STATES COURT OF APPEALS

SIXTH CIRCUIT

Nos. 76-2090, 77-1367

THE CINCINNATI GAS AND ELECTRIC COMPANY, et al.,

Petitioners,

V.

ENVIRONMENTAL PROTECTION AGENCY and

DouGLas M. CosTLe, Administrator of the

ENVIRONMENTAL PROTECTION AGENCY,

Respondents.

June 29, 1978

Petition for Review of an Order of the

Environmental Protection Agency

Louis E. Tosi, C. Randolph Light, Fuller, Henry,

Hodge & Snyder, Toledo, Ohio, Daniel W. Kemp, Cin-

cinnati, Ohio, for petitioners.

Ronald C. Hausmann, E.P.A., Paul Kaplow, Land and

Natural Resources Div., Pollution Control Section, Dept.

of Justice, Washington, D.C., Mary Ann Muirhead, Re-

gion V, E.P.A., Chicago, Ill., for respondents.

John W. Edwards, Lane, Alton & Horst, Columbus,

Ohio, for amicus curiae Ohio Mining and Reclamation

Ass’n.

William W. Wehr, Freifield, Bruzzese, Wehr, Morland

& England, LPA, Stuebenville, Ohio, for amicus curiae

Ohio Coal Operators’ Ass’n, Inc.

24a

William J. Brown, Atty. Gen. of Ohio, Environmental

Law Section, David E. Northrop, Asst. Atty. Gen., Co-

lumbus, Ohio, for intervenor, State of Ohio.

Before PHILLIPS, Chief Judge, and EDWARDS and

PECK, Circuit Judges.

ORDER

On receipt and consideration of a motion for clari-

fication of this court’s opinion, dated February 13, 1978;

and

Believing that said opinion is self-explanatory, but un-

derstanding petitioners’ desire for certainty, we reprint

below the specific dispositive paragraphs of Cleveland

Electric Illuminating Co. v. EPA, 572 F.2d 1150 (6th

Cir. 1978) (Nos. 76-2090 et al.), numbered in sequence

of their appearance in the opinion.

1) The cases dealt with in this opinion’ present the

major general issues. Other individual cases, in ad-

1 This decision dismisses the objections to the regulations that

apply to the following facilities :

(a) Cleveland Electric Illuminating Co.—all facilities.

(b) Dayton Power & Light Co.—Montgomery County facilities

only.

(ec) Ohio Edison Co.—Lorain County facilities only.

(d) Toledo Edison Co.—all facilities.

(e) The Timken Co.—all steam generating units.

(f) White-Westinghouse Corp.—all facilities. (Although there

was some confusion on this point in the briefs, the record makes

clear that White-Westinghouse’s Franklin County facility is

subject to the RAM model. See EPA Final Technical Support

Document at IV-57.)

(g) Standard Oil Co. of Ohio—Lucas County steam generating

unite. ;

(h) Interlake, Inc.—all steam generating units.

(i) Coulton Chemical Corp.—all steam generating units.

id., at 1152 (emphasis added).

25a

dition to presenting one or more of the general issues,

also present specific issues of fact. These are reserved

pending a review of and reports on the factual dis-

putes between the United States EPA and the in- ©

dividual petitioners. |

2) No other material issues are presented.

3) One petition pending before this court from the

Northern Ohio Lung Association attacks United

States EPA’s failure to promulgate a separate im-

plementation plan for the “secondary standards” for

the ambient air. This petition will be the subject of

separate consideration.

4) Simliarly, this opinion does not govern any petitions

where the RAM model waz not used. We do not de-

cide any specific fact disputes raised by any petitioner

as to plants other than those treated in this opinion.®

Decision of these cases will follow.

* See note 1, supra.

5) For the reasons stated above, the decision of the Ad-

ministrator in imposing the SO, contro] plan is af-

firmed subject to the reservations indicated above.

Id., at 1165 (emphasis added).

The petitioners’ Motion to Clarify, filed February 27,

1978, states accurately that this court’s opinion dis-

posed of five issues, which it stated as follows:

(1) The appropriateness of the use of RAM (raised

in the joint RAM brief and incorporated by ref-

erence in the utilities’ brief),

(2) The claims for needed procedural safeguards of

cross examination on remand (raised in the utility

brief),

(3) The economic and technical arguments (raised in

the utilities’ brief), and

26a

(4) The urban/rural designation issue (raised in the

utilities, brief) .

(5) The urban dispersion coefficients used in RAM

(raised in the joint RAM brief).

Subsequently, said petition lists the following five is-

sues as unresolved:

(1) Failure of U.S. EPA to utilize monitor data in

regulation development (utilities’ brief),

(2) Improper use of Class A Assumption in the rural

model (utilities’ brief) ,

(3) Improper terrain adjustment features to the rural

mode! ( utilities’ brief),

(4) Failure of EPA to correct background and data

errors (utilities’ brief), and

(5) Invalidity of the plan for vagueness, and failure

to consider sulfur-in-coal variability (utilities’

reply brief).

Issues 2 and 3 are disposed of by this court’s opinion

issued this date in Cincinnati Gas & Electric Co. v. EPA,

578 F.2d 660 (6th Cir. 1978).

Issues numbered 1 and 5 above we deem disposed of

by affirmance of EPA’s employment of the RAM and

MAXT-24 models and the discussion in Cleveland Elec-

tric Illuminating Co. v. EPA, and Cincinnati Gas & Elec-

tric Co. v. EPA.

Issue number 4 we consider to be fact questions which

are subject to agreed-on administrative review. See Cin-

cinnati Gas & Electric Co. v. EPA.

For the information of these petitioners, and others,

we call attention to the fact that by order dated Feb-

ruary 9, 1978, the petition of the Northern Ohio Lung

Association (No. 76-2282) was remanded to the United

States EPA for further consideration. Northern Ohio

27a

Lung Ass’n v. EPA, 572 F.2d 1182 (6th Cir. 1978).

Additionally, as of today, an order is being entered deny-

ing the petition of the Ohio Mining & Reclamation As-

sociation and the Ohio Coal Operators’ Association, Inc.

(No. 77-3290) because it seeks relief which can only be

had from the Congress of the United States.

As of this present date, this court considers that the

United States EPA control plan for all Mahoning and

Summit County facilities belonging to certain petitioners

in these cases is under voluntary re-evaluation by United

States EPA due to factual disputes. As these and other

fact disputes are resolved, stipulations for dismissal!

should be submitted promptly to this court. When factual

disputes cannot be resolved, the parties should make

every effort promptly to stipulate the disputed issues, or.

failing that, to stipulate the opposing positions of the

parties on the issue or issues.

28a

[FRL 1579-5]

RECONSIDERATION OF THE USE OF PASQUILL-GIFFORD

DISPERSION COEFFICIENTS FOR STABILITY CLASS A IN

SETTING EMISSION LIMITATIONS FOR FOUR OHIO POWER

PLANTS; EVALUATIONS OF PUBLIC COMMENTS

On February 7, 1979, the Environmental Protection

Agency (EPA or the Agency) published a notice in the

Federal Register soliciting public comment on the agen-

cy’s reconsideration of the use of the Pasquill-Gifford

(P-G) dispersion coefficients (or curve) for class A

meteorological conditions in setting emission limitations

for four Ohio power plants. 44 FR 7798. The agency

published the notice in response to the decision of the

United States Court of Appeals for the Sixth Circuit in

Cincinnati Gas & Electric Co. v. EPA, 578 F. 2d 660

(1978). On June 29, 1978, the Court remanded to the

Agency its decision to use the P-G dispersion coefficients

for stability class A in modeling isolated rural power

plants.

The modeling had been performed by EPA to set emis-

sion limitations for sources in Ohio to assure attainment

and maintenance of the -national ambient air quality

standards for the pollutant sulfur dioxide. See 41 FR

36324, 41 FR 52455, and 42 FR 27588. The Court found

that the agency had not developed an adequate record to

support the use of the P-G curve for class A conditions.

Moreover, the Court held that the agency had not ade-

quately considered an alternative to the use of the P-G

class A curve proposed by the utilities during the rule-

making.

1In the February 7, 1979 notice, EPA solicited comment on

whether the four power plants whose emission limits were stayed by

the agency’s reconsideration would need more time to come into

compliance. The agency will propose new compliance schedules for

those plants in a separate Federal Register notice.

29a

In response to the Court’s decision, the agency recon-

sidered the use of the P-G dispersion coefficients for class

A conditions and evaluated the utilities’ proposal and the

P-G class B dispersion coefficients be used in place of the

class A dispersion coefficients. Based on field data not

previously considered by the agency and current disper-

sion theory, the agency found that the utilities’ proposal

would severely underestimate ground level concentrations

(44 FR 7798, February 7, 1979). Conversely, the agency

found that dispersion theory and the Karlsruhe field data

confirmed the agency’s use of the P-G class A dispersion

coefficients in setting emission limitations to assure at-

tainment and maintenance of the standards. The agency,

therefore, concluded that it is reasonable to use the P-G

class A curve to set emission limitations for the four

power plants at issue.”

Several utilities and utility consultants commented on

the agecny’s proposed decision to continue to use the P-G

class A dispersion coefficients. In general, commenters

challenged the agency’s proposal by criticizing the Karls-

ruhe field study data and continuing to argue that use

of the P-G class A curve on modeling sources with ele-

vated release points such as power plants will result in

unrealistically high ground level concentrations at loca-

tions unrealistically close to the source. However, in ad-

dition to their original hypothesis that the sigma-z com-

ponent of the P-G class A coefficients is too large, the

utilities and their consultants also argued for the first

time in their comments that the sigma-y component of

the P-G class A curve is too smail.

2 EPA review of its rural modeling analysis identified four power

plants where the class A conditions determined the emission limita-

tions. The emission limitations for all other rural power plants in

Ohio were not determined by class A conditions. The four plants

are Stuart (Dayton Power & Light Co.) Conesville (Columbus and

Southern Chio Electric Co.) Cardinal and Muskingum River (Ohio

Power).

30a

The utilities originally urged only one alternative to

the P-G class A curve, namely to substitute the P-G

class B curve for the P-G class A curve. See 44 FR

7798 (February 7, 1979). In response to the agency’s

notice, commenters submitted new proposed alternatives

to the P-G class A curve. Specifically, they urge the use

of the Brookhaven class B, curve, the Julich class A

curve, the Briggs curve, the F. B. Smith curve, the flutu-

ating plume model, and the smaller averaging times

method.

Commenters also criticized the agency for not evaluat-

ing actual air quality data available from power plant

monitoring systems. In response to this, the agency re-

viewed monitoring data from the John Sevier, Widows

Creek, Cumberland, Muskingum River, Big Bend, East

Bend, Conesville and Gibson power plants.

The agency finds that the utility comments are with-

out merit and that the utilities have not provided any

technical basis to support changing the agency’s proposed

finding. In fact, the monitor data which the utilities

asked EPA te review support the agency’s determination

that it should continue to use the P-G class A curve.

Moreover, the data establish that all of the utilities’ pro-

posals would underestimate maximum ground level con-

centrations and therefore would not assure attainment

and maintenance of the standards.

In addition to the utility comments and data, the

Agency reviewed the comments and recommendations of

the Specialists’ Conference on the EPA Modeling Guide-

line, ANL (1977), the American Meterological Society

(AMS) Workshop on Stability Classification Schemes

and Sigma Curves, Hanna et al. (1977), and the Na-

tional Commission on Air Quality Atmospheric Disper-

sion Modeling Panel, NCAQ (1980). These scientific

groups recommended additional study of the effects of tall

stack plume characteristics on ground level concentrations

and the suitability of the P-G curves for tall stack

sources. The Specialists Conference suggested that a pos-

3la

sible result from such additional study “might include

the: elimination of the A curve and the use of the B

curve for both A and B stability categories.” See § 2.7.5

“Vertical Dispersion Estimates.”. The AMS Workshop,

on the other hand, recommended the use of the Brook-

haven curves, rather than the P-G curves, for elevated

sources.

The agency has studied the use of the P-G class A dis-

persion coefficients and evaluated both suggested alterna-

tives as part of its response to the Court’s remand. Based

on EPA’s study, the agency has determined that the

available data demonstrate that the P-G class A dis-

person coefficients are reasnable means of setting emis-

sion limitations that will assure attainment and main-

tenance of the national air quality standards. In con-

trast, the available data establish that the suggested al-

ternatives underpredict ground level concentrations and

therefore can not provide a means to set emission limita-

tions adequate to assure attainment and maintenance of

the standards.

As stated in the February 7 notice, it is not the agen-

cy’s contention that better methods of representing dis-

persion should not be used when available and justified.

44 FR 7805. In fact, the agency continues to research

this area. However, the agency has not yet developed

a more accurate means for predicting ground level con-

centrations near rural power plants during class A me-

teorological conditions than the P-G class A dispersion

_ coefficients.

A. Original Utility Challenges, the Class B Alternative,

and the EPA Proposal

During the rulemaking on the sulfur dioxide emission

limitations, the utilities contended that the P-G class

A dispersion coefficients, which were developed from field

studies of emissions released near ground level, were in-

Eee a

32a

appropriate for use in modeling power plants with tall

stacks

Specifically, the utilities hypothesized that -the P-G

sigma-z values (controlling vertical dispersion) for class

A conditions were too large for estimating ground level

pollution concentrations caused by sources with tall

stacks.* The utilities contended that use of the P-G class

A sigma-z values for power plant modeling resulted in

overpredicting ground level concentrations.

However, as was explained’ in the February 7, 1979

notice, experimental research suggests that the sigma-z

values should be as large or larger for elevated releases

as compared to the sigma-z values for near ground level

releases. See Panofsky (1978). Vogt, et al. (1978) and

Weil (1978). Moreover, current dispersion theory is that

sigma-z values for emissions over rough terrain should

be equal to or larger than the sigma-z values for emis-

sions over smooth terrain. See F.B. Smith (1973) and

Pasquill (1974). 44 FR 7798. This is significant in that

the original field experiments used to derive the P-G

sigma-z values were conducted over prairie grass fields

(relatively smooth terrain with a characteristic surface

roughness length of less than 0:03 meters (m)) which

are much smoother than the average terrain where the

Ohio power plants are situated (surface roughness length

is in the range of 1.0 m).

In evaluating the utilities’ challenge to the class A

sigma-z values and their recommendation of P-G class

B, the agency reviewed the sigma-z values for the most

3 Each set of dispersion coefficients include a value representing

the vertical dispersion of the pollutant (sigma-z value) and a value

representing the horizontal dispersion (sigma-y value) in EPA’s

CRSTER modeling each class of meteorological conditions is asso-

ciated with a different set of dispersion coefficients Class A meteoro-

logical conditions are the most unstable(very light and variable

winds, clear sky and well developed mixing layer). Class B through

F represent increasingly more stable atmospheric conditions.

33a

unstable meteorological conditions which were determined

by three different experimenters in three different dis-

persion field studies. Table 1 in the February notice

sets forth the sigma-z values derived by the Brookhaven,

' . Julich and Karlsruhe experiments. 44 FR 7800. Unlike

the Pasquill-Gifford experiment, each of these experi-

ments involved elevated releases of tracer (approximately

100 m) over terrain with surface roughness length of

approximately 1.0 meters. These experiments, therefore,

more closely approximated the elevated release points and

terrain characteristics of the Ohio power plants.

The comparison of sigma-z values from these three

experiments with the sigma-z values in the P-G class A

and B curves showed that under unstable meteorological

conditions all three experiments measured sigma-z values

larger than the sigma-z values associated with the P-G

class B curve.* See Table 1, 44 FR 7800. The data dem-

onstrate that use of the P-G class B curve as a substitute

for class A would necessarily underpredict maximum

ground level concentrations reached during class A con-

ditions. Emission limitations based on the class B curve,

therefore, would not assure attainment and maintenance

of the standards. In short, without an analysis using the

P-G class A dispersion coefficients, the agency would not

be sure of setting emission limitations which would pro-

tect the standards during periods of unstable meteorologi-

cal conditions.

In addition, the agency noted that the Karlsruhe field

experiments determined sigma-z values during class A

conditions as large or larger than the P-G class A

sigma-z values. The Karlsruhe data therefore confirm

the P-G class A sigma-z values and the resulting ground

4 The Brookhaven sigma-z values set forth in Table 1 are for BNL

class B, meteorological conditions which are more stable than class

A conditions. 44 Fed. Reg. 7800 Brookhaven experimenters did not

derive sigma-z values for the most unstable stability class (class A).

34a

level concentrations. The Karlsruhe experiments also con-

firm the general locations of maximum ground level con-

centrations predicted by using the P-G class A dispersion

coefficients. 44 FR 7801. The agency therefore found

that the Karlsruhe data supported the use of the P-G

‘class A dispersion coefficients in setting emission limita-

tions for rural power plants.

B.. Utility Comments on EPA’s Proposal :

During the comment period, the utilities repeated their

earlier contention that use of the P-G class A curve for

rural power plants would result in model overprediction

and overly stringent emission limitations. The utilities

criticized the agency’s support for the use of the P-G

class A curve and suggested other alternatives which like

their original P-G class B proposal, would calculate lower

maximum ground level concentrations than using the

P-G class A dispersion coefficients. The utilities also

urged the agency to review any available monitoring data

recorded near power plants as a means of verifying the

accuracy of the P-G class A dispersion coefficients.

In response to this comment, the agency gathered the

available monitoring data suggested by comments and

reviewed it. Based on that review, the agency deter-

mined that most of the available data is not relevant for

evaluating the predictive accuracy of the P-G class A

dispersion coefficients because of the placement of the

monitors or because a supplementary control system was

used. However, the agency did compare all of the rele-

_ vant utility monitoring data with P-G class A modeling

predictions. The agency found that actually measured

sulfur dioxide concentrations during class A conditions

are as high or higher than the maximum predictions cal-

culated for the same monitor site using the P-G class A

dispersion coefficients. Actual utility data, therefore, sup-

ports the agency’s use of the P-G class A dispersion co-

efficients for modeling power plants in rural areas.

ee

35a

1. Utility Monitoring Data

Agency review of power plant monitoring data as sug-

gested by the commenters demonstrated that few of the

monitors were located at distances less than 2 km from

the source where maximum concentrations under class A

meteorological conditions are expected to occur. EPA

model anlyses using the P-G dispersion coefficients for the

four Ohio power plants predicted maximum concentra-

tions under class A conditions at distances 1.0 to 1.3 km

from the plant.’ Therefore, to verify the accuracy of

P-G class A model predictions, measurements from moni-

tors located close to the source are required.

The agency also found that several of the power plants

having monitors within 2 km of their stacks use sup-

plementary control systems.* This means that the plants

reduce their emission rate during periods of expected

high ground level pollution. An effective supplementary

control system reduces ground level concentrauons that

would otherwise occur under class A conditions without

the use of such a system. Appropriate comparison of

model predictions with monitor observations for these

plants (where operating loads are reduced) requires the

use of data reflecting the actual operation of the plants.

Such data, including actual hourly emissions and stack

effluent characteristics data and on-site meteorology, are

not readily available. Each of the model analyses the

agency has performed in evaluating available data has

relied on maximum operating load and national weather

service meteorological data. The agency, therefore, could

not use the data influenced by the use of supplementary

control systems to evaluate the predictive accuracy of the

P-G class A curve. |

5 The Karlsruhe data confirmed that maximum concentrations

under class A conditions occur within 1-5 km of the release point.

44 FR 7796. ?

6 The TVA plants at Widows Creek and Cumberland and the PSI

Gibson plant use such systems.

36a

Based on these considerations, the agency determined

that data from two power plants—TVA’s John Seiver

power plant and Ohio Power’s Muskingum River plant—

could be used to evaluate the predictive accuracy of mod-

eling with the P-G class A dispersion coefficients. See

Table 2. In addition, the agency found that data from

the monitoring system at the Muskingum River plant

could be used to check the model’s prediction of the loca-

tions of maximum concentrations under class A con-

ditions.

a. Accuracy of model predicted maximum concentra-

tions using P-G class A dispersion coefficients.—In order

to evaluate the accuracy of model predictions using the

P-G dispersion coefficients for class A stability, the

agency modeled the John Seiver ‘and Muskingum River

power plants. The agency used the MPTER model with

full load operating data provided by the power plants

and at least six years of meteorological data.? The agency

calculated the highest and second highest 3-hour ground

level concentrations at the monitor sites for each year of

meteorological data. See Table 3. The agency then com-

pared these model predictions with the highest and second

highest concentrations actually observed in a 3-hour pe-

riod during each of monitor operation. This comparison

showed that the monitors recorded maximum concentra-

tions of sulfur dioxide as high or higher than the maxi-

mum concentrations calculated by the model using the

P-G class A dispersion coefficients.

At John Sevier, the monitor has recorded sulfur di-

oxide concentrations for six years. In four of the six

years, the monitor recorded higher maximum concentra-

tions than the model calculated for any of the six years

of meteorological data.* Maximum levels of. 45, .40, .43,

8 Comparisons of monitor observations with model calculations

using meteorological data for identical years were performed for

1973 and 1974 at the John Sevier monitor Meteorological data were

not available for identical years to make such comparisons for addi-

tional years.

87a

.34, .28 and .17 ppm (parts per million) were actually

recorded. See Table 3. The model using the P-G class A

curve predicated a maximum of .32 ppm at the same lo-

cation. Second highest maximum concentration levels of

.39, .39, .31, .30, .22, and .15 ppm were recorded in the

six years of monitoring compared to the highest model

predicted second high maximum of .32 ppm.

At Muskingum River, the monitor has recorded sulfur

dioxide measurements for only two years. In those two

years, the monitor measured maximum concentrations of

.63 and .39 ppm, while the model calculated a maximum

concentration of .55 ppm based on seven years of mete-

orological data. The monitor recorded second high maxi-

mum concentrations of .41 and .20 ppm, compared to the

highest second high model prediction of .43 ppm.

Actual power plant measurements, therefore, compare

well with model calculations of maximum ground level

concentrations using P-G class A dispersion coefficients.

Moreover, the comparison establishes that the agency’s

modeling using P-G class A curves does not predict un-

realistically high concentraticns; rather, the agency’s

modeling accurately predicts sulfur dioxide concentrations

for class A meteorological conditions.

b. Accuracy of model predicted locations of maximum

ground level concentrations using P-G class A dispersion

coefficients.—The agency’s modeling using the P-G class

A dispersion coefficients predicted that maximum ground

level concentrations would occur between 1 and 1.3 km

from the source. Utilities challenged this contending that

for elevated-sources maximum concentrations will occur

further out from the-source. The utility monitoring data,

however, confirm the P-G class A predictions.

Data from the Muskingum River monitor system can

be used to investigate the location of maximum concen-

trations during class A meteorological conditions. At the

Muskingum River plant, four of the six continuous SO,

38a

monitors are located along a relatively straight line in

a northeasterly direction from the plant. The monitors

are located at distances of 1.7 km, 4.6 km, 8.5 km and

20.7 km from the plant along this radial. See Table 4.

Review of the monitoring data shows that the highest 3-

hour ground level concentration recorded by these moni-

tors was observed at the monitor closest to the plant.

See Table 5. Moreover, review of meteorological condi-

tions during this period shows that the maximum con-

centration occurred during class A conditions. In con-

trast, the maximum impacts observed at monitors located

further from the plant generally occurred during periods

of more stable (greater cloud cover and/or higher wind

speed) meteorological conditions.

The Muskingum River data, therefore, establish that

maximum concentrations do occur as close to the source

as predicted by the model using the P-G dispersion co-

efficients under class A conditions. Based on review of

the utility monitoring data, the agency finds that model-

ing with the P-G dispersion coefficients accurately pre-

dicts the magniture and location of maximum ground

level concentrations from power plants with tall stacks.

2. The Utility Challenge to the Karlsruhe Data

As explained in the February 7, 1979 notice, the agency

found that data from the Karlsruhe field experiments

support the use of the class A dispersion coefficients. 44

CR 7801. The Karlsruhe field experiments as reported by

Thomas et al. (1976) and Thomas and Nester (1976),

measured ground level concentrations of two dispersion

tracers, tritium (H,) and halogenated hydrocarbons

(CCl,), released from 10 meter heights over terrain

similar to the terrain surrounding the four Ohio power

plants. Based on the measured ground level concentra-

tions, sigma-z and sigma-y values were derived. Since

the original utility challenge focused on the sigma-z

values for the P-G class A curve, the agency compared

39a

sigma-z values and found that the Karlsruhe experi-

menters derived sigma-z values for class A conditions

comparable to the P-G class A sigma-z values. See Table

1 and Figures 1-3, 44 FR 7800, 7800-8. The agency con-

cluded that the Karlsruhe experimental data confirm the

‘general level and location of maximum ground level con-

centrations predicted by using the P-G dispersion coeffi-

cients (sigma-z and sigma-y) for class A meteorological

conditions. See Figures 4-6, 44 FR 7803-4.

The: utilities commented that the agency should not

rely on the Karlsruhe data. They fault the Karlsruhe

experiments for not using what the commenters contend

are more advanced procedures and methodology in de-

termining ground level concentrations and in deriving

the dispersion coefficients. They commented that the

Karlsruhe data are therefore suspect. However, com-

menters did not point to any data to support their con-

tention that the procedures actually used in the Karlsruhe

experiments would result in unreliable data. On the

contrary, as discussed above, utility monitors have actu-

ally measured maximum concentrations as high or higher

and as close to the source as predicted by the P-G class

A dispersion coefficients. This monitor data confirm the

Karlsruhe results as well as the use of the P-G class A

dispersion coefficients.

Moreover, additional Karlsruhe dispersion field studies

reported in 1979 employed some of the experimental pro-

cedures suggested by commenters and released tracer

from 195 meters. This data confirmed the earlier Karls-

ruhe results. Therefore, there is no basis for concluding

that the Karlsruhe experimental procedures in any way

“ In fact, the utility criticisms are contrary to their willingness

to accept the results of the Brookhaven and Julich experiments or

even the Pasquill-Gilford (for class B curves), all of which used

experimental procedures similar to or less advanced than the

Karisruhe procedures.

40a

resulted in false measurements. See Kiefer and Koelzer

(1979).

Specifically, commenters criticized the Karlsruhe ex-

periments for using non-bouyant tracer materials, a 100

meter release height, a manual sampling and collection

system instead of an automated system, for analyzing air

samples dissolved in toluene rather than directly sampling

the air for measuring ground level concentrations and

deriving .the sigma-z values from those measurements,

rather than directly measuring sigma-z values, and for

the mathematical analysis procedures chosen to derive the

sigma-z and sigma-y values. Each of these criticisms is

discussed below.

a. Non-buoyant tracer and 100 m release height.—

The utilities criticized the Karlsruhe experiments for

using a non-bouyand tracer released at 100 m meters.

The utilities point out that bouyant plumes, such as those

released from power plant stacks, normally attain heights

much higher than 100 meters (200-1,000 meters or

more). The utilities commented that unstable tempera-

ture gradients and building downwash would effect a non-

buoyant tracer released at 100 meters, increasing the

verticle dispersion of the tracer. Commenters argue that

the vertical dispersion rates (sigma-z values) derived

from the experiments would be too large for power plant

plumes.

However, the utility comment is contrary to both actual

monitor data and dispersion theory. First, the utility

monitor data demonstrate that under class A conditions,

maximum concentrations as high or higher than calculated

using the P-G class A curve and as close to the power

plant as predicted by the P-G class A curve have been

recorded. This data therefore confirm the use of the

sigma-z values derived by the Karlsruhe experiments as

well as the P-G class A sigma-z values.

4la

Second, experimental research indicates that the verti-

cal dispersion of power plant plumes increases with

height. The rate of verticle dispersion of the plume does

not decrease once the plume. rises to levels where the

temperature gradients are more stable but rather remains

constant or continues to increase. See Irwin (1979). The

sigma-z values therefore should not be smaller for sources

with elevated releases as compared to source releases near

ground level. See also Panofsky (1978), Vogt et al.

(1978), 44 FR 7799. a.

Finally, the later Karlsruhe dispersion field studies,

using a 195 meter release height, confirmed the large

sigma-z values measured by the earlier Karlsruhe experi-

ments and the P-G class A sigma-z values. In these later

Karlsruhe experiments, tracer was released from 195

meters, a higher release point than used for any of the

alternative curves proposed by the-utilities. See Table 1.

The 195 m Karlsruhe experiments therefore are the most

appropriate experimental work for evaluating dispersion

coefficients for power plants. These later experiments

measured vertical dispersion rates (sigma-z values) at

distances greater than 1.3 km which were larger than the

rates derived from the earlier 100 meter experiments.

The data therefore contradicts the utility contention and

support the earlier Karlsruhe results and the P-G class

A sigma-z values. See Kiefer and Koelzer (1979).

b. Manual sampling and collection system.—Comment-

ers criticized the Karlsruhe experiments for using a

manual sampling and collection system.. However, com-

menters offer no data to show that an automated system

would result in different measurements. In fact, the

1° Additional experimental studies were also conduced at 80, 100

and 160 meters. See Kiefer and Koelzer (1979). Results from the

195 m studies most closely approximate dispersion of pollutants

from tall stack sources such as power plants. The agency therefore

has relied upon the results of the 195 meter experiments in prefer-

ence to the others.

42a

later Karlsruhe field studies used an automated sampling

and collection system and the results confirm the earlier

Karlsruhe results. See Kiefer and Koelzer (1979).

ce. Use of toluene in sample analysis —Commenters

criticized the sample analysis method used by the Karls-

ruhe experiments. Specifically, the commenters criticized

the technique of dissolving the sampling in toluene before

analysis and commented that a direct analysis would be

more reliable. Commenters, however, do not demonstrate

that direct analysis of the air sample would be more

reliable than the technique used in Karlsruhe. ‘The use of

toluene in the analysis technique is necessary in order’ to

remove all of the tracer from the coilecting glass vessel

and assure measurement of the entire sample. In fact,

the methodology used by the Karlsruhe experimenters is

an accepted experimental technique and there is no basis

to conclude that it results in faulty data.

d. Deriving sigma-z values from ground level concen-

trations.—Commenters also criticized the Karlsruhe ex-

perimenters for deriving the dispersion coefficients from

ground level concentration measurements rather than

making direct measurements of the sigma-y and sigma-z

values. However, none of the dispersion field studies .

reviewed by the agency, including Brookhaven and Julich,

measured sigma-z and sigma-y values directly. Com-

menters do not point to any dispersion studies which

made direct measurements of sigma-y and sigma-z values

under unstable conditions. It is simply too costly to make

direct measurements under unstable meteorological con-

ditions. Therefore investigators have traditionally de-

rived the sigma-z and sigma-y values from ground level

tracer concentration measurements.

Commenters also failed to provide any data or theory

to establish that direct measurements would lead to dif-

ferent results. In short, while it might be preferable to

determine dispersion coefficient values under very un-

43a

stable conditions directly, as a practical matter, it is not

done.

e. Mathematical analysis procedures.—One commenter

criticized the mathematical method the Karlsruhe experi-

menters used to derive the sigma-y and sigma-z values.

The Karlsruhe experimenters used a method of determin-

ing dispersion coefficient values which smoothed out the

variation in ground level concentrations expected under

class A conditions. See Thomas and Nester (1978). The

commenters applied two different mathematical methods

to the Karlsruhe data and derivied smaller sigma-y values

than the Karlsruhe experimenters. However, the com-

menter chose methods which do not take into account the -

expected variations in concentrations under class A con-

ditions. Both the second moments and the cross-wind in-

tegrated ground level concentrations methods are more

appropriate for data sets with little variation. The in-

appropriateness of these methods is confirmed by the

commenter’s calculation of sigma-y values which de-

creased with distance from the source. Such a result is

contrary to the law of atmopheric physics that the hori-

zontal distribution of a plume can not decrease with

distance.

f. Other comments on the Karlsruhe data.—Comment-

ers made several other comments criticizing the Karlsruhe

data, One commenter cited the fact that in experiment

number 19.2, the Karlsruhe experimenters measured dif-

ferent ground level concentrations of simultaneously re-

leased tracers at co-located samplers. However, com-

menters do not provide any data or theory to establish

that the different measurements indicate faulty data. In

fact, the 10-30 percent variation in sigma-z values

derivied from these two measurements establish that the

variations in concentrations was small.

One commenter cited the fact that high background

levels of the tracer were measured during time periods

44a

when the experimenters indicated that the wind was not

transporting tracer in the direction of the monitor. The

commenter argues that this fact demonstrates the in-

accuracy of the experimental sampling system. However,

analysis of the cited time periods shows that while tracer

was not transported directly to the sampling site, tracer

materials were transported to the site by an indirect path

resulting from the changing wind direction. Therefore,

the tracer concentrations measured during these periods

of changing wind direction are as would be expected and

do not indicate any experimental inaccuracy.

Another commenter criticized the Karlsruhe experi-

ments because the distribution of crosswind concentra-

tions for the individual experimentation periods did not

fit perfect Gaussian curves and the maximum measured

concentrations did not fall on a straight line from the

tracer release point. Both of these criticisms assume un-

realistic experimental results. Given the variations ex-

pected in measurements made in a single experimentation

period of 20 to 30 minutes, a perfect Gaussian distribu-

tion is simply not expected. Moreover, given the chang-

ing transport wind associated with class A meteorological

conditions, maximum concentrations would not be ex-

pected to follow a straight line. Rather, concentration

measurements in a single experimentation period would

be expected to vary from a perfect Gaussian curve and

from a straight line from release point and this is ex-

actly what happened.

Another commenter criticized the Karlsruhe data be-

cause some measurements fell outside the confidence limits

specified by the experimenters. By definition, a percent-—

age of the measurements are expected to fall outside the

confidence limits. There is, however, no indication and

the commenter does not establish that the number of .

measurements falling outside confidence limits indicates

that the experiments were deficient.

45a

One commenter criticized the Karlsruhe data because

the ratio of sigma-z values to sigma-y values is greater

than one. The commenter contends that this indicates

that the sigma-z values are too large. However, the com-

menter does not support his assumption that this ratio

should equal one for the unstable meteorological condi-

tions associated with class A stability. Under the com-

menter’s assumption, the vertical distribution of the plume

would equal the horizontal distribution. However, under

class A conditions greater vertical than horizontal turbu- .

lence would be expected due to the convective forces

associated with unstable meteorological conditions. There-

fore, a ratio of sigma-z to sigma-y values which is greater

than one would be expected and is exactly what was

determined at Karlsruhe.

One commenter used two different theories to calculate

sigma-z values from.some of the Karlsruhe data. Again,

the result was to derive smaller sigma-z values from the

Karlsruhe data. The commenter based the derivation on

the publications of Irwin (1979) and Smith (1968).

Both Irwin and Smith have developed a relationship be-.

tween the standard deviation of the wind elevation angle

and the plume dispersion coefficients. The coramenter

applied the theories to elevation angle data from the

Karlsruhe experiments and derived smaller sigma-z values

than the Karlsruhe experimenters derived from their

measurements. However, the utility monitor- data dis-

cussed above demonstrate that the sigma-z values derived

by Karlsruhe and the P-G class A sigma-z values are not

too large. In other words, use of smaller sigma-z values

would conflict with actual monitor data.

Another commenter used the Karlsruhe data to derive

different dispersion coefficients than those derived by the

Karlsruhe experimenters. The commenter used his ver-

sion of the Karlsruhe dispersion coefficients to make a

comparison with P-G class A and class B dispersion co-

efficients. Mowever, the commenter does not explain or

46a

support his reworking of the Karlsruhe data. In the

February notice, the agency used the Karlsruhe disper-

sion coefficients exactly as derived and reported by the

Karlsruhe experimenters. Moreover, the utility monitor

data confirms the Karlsruhe results. There is no reason

to manipulate the Karlsruhe data and comparisons based

on the manipulated data are inappropriate. |

Commenters also criticized the agency for considering

Karlsruhe experiment’ number 19.2 and not considering

experiments number 19.1 and 19.3. Commenters contend

that experiment number 19.2 was not conducted under

class A conditions and that experiments number 19.1

and 19.3 were both class A experiments which should

have been considered.

In determining which experiments to consider as class

A experiments, the agency used the classification of ex-

periments set forth by the Karlsruhe experimenters. The

Karlsruhe experimenters listed the class A experiments

in Table 1, Thomas and Nester (1976). Experiment 19.2

was listed as class A based on the wind profile, even

_ though the vertical wind fluctuation and temperature

gradients factors were more class-B like. Moreover, the

dispersion coefficients derived from experiment 19.2

clearly fit in the class A category. See Table 1, 44 FR

7800 (February 7, 1979). Therefore, it was appropriate

for the agency to consider the results of experiment 19.2.

Experiments 19.1 and 19.3, on the other hand, were not

listed by the Karlsruhe experimenters as valid class A

experiments. During the sampling period, the wind car-

- vied the tracer material away from the sampling net-

work. Therefore, dispersion coefficients could not be de-

veloped from the ground level concentration measure-

ments because the tracer material was transported where

_there were no samplers. The Karlsruhe experimenters

simply could not use these two experiments. See Thomas

and Nestor (1976).

47a

Finally, commenters criticized the small number of

class A experiments conducted at Karlsruhe. In the

earlier Karlsruhe experiments, only two numbered ex-

periments (9 and 19) were conducted for which sigma-y

and sigma-z curves were developed by the investigators.

Commenters suggested relying on the Brookhaven or

Julich sets of curves which they contend were based on

larger numbers of experiments. However, the Brook-

haven curves were not based on any class A experiments

and the Julich curves were derived from two class A

experiments.

Review of the literature on the Brookhaven experi-

ments indicates that no curves were developed for the

most unstable class of conditions, class A. See Singer and

Smith (1966) and Smith (1973). According to the origi-

nal investigators, because the experiment site is located

on Long Island near the ocean, both the frequency and

duration of class A conditions is diminished. That is, the

sea breeze prevents extended periods of low wind speed

necessary for class A conditions. Singer and Smith how-

ever did develop a set of curves for sigma-y and sigma-z

which are based on more stable atmospheric conditions.

The most unstable class of conditions for which curves

were developed by the Brookhaven investigators was the

class labeled B2. While the Julich experimenters de-

veloped class A curves, they were based on the same num-

ber of experiments as the early Karlsruhe experiments.

Neither the Brookhaven nor Julich curves are based

on more experiments than the Karlsruhe curves. Fur-

thermore, the more recent 195 m Karlsruhe experiments,

which better approximate the tall stacks of the Ohio

power plants than either the Brookhaven or Julich ex-

periments, added to the number of Karlsruhe experiments

on Class A. The 195 m experiments confirm the results

of the earlir 100 m experiments at Karlsruhe. There-

fore, the Karlsruhe class A curves are based on more

48a

experiments and better approximate the heights of tall

stack plumes than any of the available alternatives.

8. The Challenge to the Sigma-y Values

In addition to the original hypothesis that the P-G

class A sigma-z values are too large, commenters sub-

mitted a new hypothesis: the sigma-y values (the hori-

zontal dispersion factors) in the P-G class A curve are

too small.‘ Commenters base their hypothesis on the

contention that larger sigma-y values are necessary to

accurately calculate one hour average maximum ground

level concentrations. Again, this contention conflicts with

the utility monitor data. In fact, the utility monitors

have recorded levels higher than the levels predicted by

using the P-G class A dispersion coefficients. Therefore,

based on the utility monitor data, use of larger sigma-y

values would necessarily underestimate maximum ground

level concentrations and could not be used to set emission

limitations that would assure attainment and mainte-

nance of the standards.

Commenters’ criticism of the P-G class A sigma-y

values rests on the fact that the values were derived from

tracer measurements of three to fifteen minutes in length.

Commenters postulate that one-hour tracer measurements

(resulting in larger sigma-y values) would more accu-

rately estimate maximum one-hour average ground level

concentrations. Based on this hypothesis, commenters

urge the agency to use a time scaling factor to increase

the P-G class A sigma-y values. However, the utility

monitor data establish that sigma-y values derived from

three to fifteen minute measurements accurately estimate

11 Larger sigma-y valves mean greater horizontal dispersion of

the pollutant. Smaller sigma-z values mean less vertical dispersion

and more channeling of the pollutant out to further distances before

it reaches ground level. Either greater horizontal dispersion or less

vertical dispersion of a plume would result in lower ground level

concentrations.

49a

maximum one-hour ground level concentrations. Based

on the utility monitor data, as well as the Karlsruhe data,

the agency finds that sigma-y values larger than the

P-G class A sigma-y values should not be used to set

emission limitations for Ohio power plants.

4. The Proposed Alternatives

In response to the agency’s February notice, com-

menters have also suggested new alternatives to the

P-G class A dispersion coefficients. Previously, the utili-

ties proposed the use of the P-G class B curve in place

of the P-G class A curve. In addition to the P-G class B

curve, commenters suggested that the Brookhaven class

B, or the Julich class A curve also would be acceptable

substitutes for the P-G class A dispersion coefficients.

Commenters also proposed using curves developed by

Briggs and F. B. Smith. One commenter suggested the

use of a fluctuating plume model and a method of cal-

culating concentrations with smaller averaging times as

a substitute for the agency’s modeling using the P-G

class A curve.

Ail of the proposed alternative curves would calculate

lower maximum ground level concentrations than the

P-G class A dispersion coefficients would calculate. The

fluctuating plume model and the smaller averaging time

methodology also calculate lower maximum ground level

concentrations than the agency’s modeling using the P-G

class A dispersion coefficients. The agency, however, finds

that its use of the P-G class A dispersion coefficients has

been confirmed by the utility monitoring data. There is

simply no basis for substituting the P-G class A curve

with any of the suggested alternatives. Nevertheless, the

agency has evaluated each of the proposed alternatives.

That evaluation is set forth below.

a. Brookhaven class B, and Julich class A curves.—

Both the Brookhaven class B, and Julich class A curves

50a

would calculate lower maximum ground level concentra-

tions than the P-G class A curve calculates. This is be-

cause both experiments derived sigma-z values smaller

than the P-G class A sigma-z values and sigma-y values

as large or larger than the P-G class A sigma-y values.

A smaller sigma-z value means less vertical spread of

the plume, namely, the plume is carried farther from the

source before it reaches ground level. The concentration

of the pollutant decreases with the distance traveled.

A larger sigma-y value means greater horizontal spread

of the plume which also decreases concentration of the

pollutant.

In the February 7, 1979 notice, the agency discussed

sigma-z values derived by the Brookhaven, Julich and

Karlsruhe experiments for unstable meteorological condi-

tions. The agency compared the sigma-z values with the

P-G class A and class B sigma-z values. Table 1, 44 FR

7780, and Figure 2. This comparison established that

the Karlsruhe experiments had derived sigma-z values

as large or larger than the P-G class sigma-z values. In

contrast, the Julich and the Brookhaven experiments

derived smaller sigma-z values for unstable conditions

than either the Karlsruhe or the P-G class 4 sigma-z

values.

In response to comments on the February 7 notice,

the agency made a similar comparison of the sigma-y

values determined by these same experimenters. See

Figure 1. This comparison establishes that the Julich

class A sigma-y values are much larger than either the

Karlsruhe or P-G class A sigma-y values for distances

greater than about 0.4 km. In contrast, Brookhaven class

B, sigma-y values are in fairly close agreement with the

P-G class A sigma-y values. ,

In order to evaluate the effects of the different sigma-y

and sigma-z values on calculations of ground level con-

centrations, the agency compared the maximum normal-

ized concentrations and the distance to maximum impacts

4

%,

i

5la

calculated by each set of dispersion coefficients. See Fig-

ure 3. The results of these comparisons demonstrate that

the P-G class B, Julich class A and Brookhaven class B,

all calculate lower maximum ground level concentrations

than calculated by either the P-G or Karlsruhe class A

curves. In addition, these curves calculate maximum

concentrations at distances farther from the source than

calculated by either the P-G or Karlsruhe class A curves.

See Figure 4. Based on the utility monitor data which

confirm the agency’s use of the P-G dispersion coefficients,

the agency finds that use of the Brookhaven class B, or

the Julich class A curves would underestimate maximum

ground level concentrations and could not be used to set

emission limitations which would assure attainment and

maintenance of the standards.

b. Briggs and F. B. Smith curves.—Commenters also

suggested that the agency use curves developed by Briggs

and F. B. Smith in place of the P-G class A curves.

However, the agency finds that both suggestions are with-

out merit. According to Briggs (1979), the Briggs curve

is based on the Brookhaven class B, curve for distances

greater than 500 meters from the source (where sigma-z

equals about 100 meters). Since maximum concentra-

tions under class A conditions are expected to occur at

distances greater than 500 meters, the Briggs curve is

equivalent to the Brookhaven class B, curve and is un-

acceptavle for the reasons discussed above.

The F. B. Smith curve is a theoretically developed

curve which was specifically developed for ground level

sources. It utilizes smaller sigma-z values than the P-G

class A sigma-z values. As discussed above, use of smaller

sigma-z values than the P-G class A values would result

in the underprediction of maximum ground level concen-

trations. The agency, therefore, could not rely on the

F. B. Smith curve to set emission limitations which

would assure attainment and maintenance of the

standards.

52a

c. A fluctuating plume model and smaller averaging

times method.—One commenter proposed the use of a

“fluctuating plume model” as the use of averaging times

smaller than one hour to calculate concentrations. Both

would result in lower maximum ground level concentra-

tions than the P-G class A curves. Based on the utility

monitor data, the agency finds that these suggestions

therefore are inappropriate.

A fluctuating plume model is a Gaussian-type disper-

sion model (like the agency’s CRSTER model) ; however,

modeling assumptions in the fluctuating plume model in-

crease the horizontal dispersion of the plume and result

in lower calculated maximum ground level concentrations

than would be calculated using CRSTER with the P-G

class A dispersion coefficients. In effect, the fluctuating

plume model is like using the agency’s model with a

larger sigma-y dispersion coefficient. As discussed above,

sigma-y values larger than the P-G class A sigma-y

values would result in lower ground level concentrations.

The utility monitor data establishes that larger sigma-y

values would necessarily cause underpredictions of sulfur

dioxide levels.

The sam? commenter also suggested a technique of

caleulatirg one-hour ground level concentrations by using

segmented averaging periods smaller than one hour.

The commenter’s method calculates a one-hour average

by summing and averaging 20-minute average concen-

trations rather than calculating one-hour average con-

centrations. This technique results in lower ground level

concentrations. However, the commenter does not pro-

vide any data to support the contention that using 20-

minute averages will result in more accurate ground level

concentrations. In fact, this contention is contrary to

the utility monitor data which shows that the summing

and averaging of one-hour average calculations used in

EPA’s modeling are accurately predicting maximum

three-hour ground level concentrations.

53a

5. Other Data Submitted by Commenters

The following data were submitted or referenced by

commenters in support of the allegation that use of the

P-G class A coefficients results in overpredictions. How-

ever, upon review, the agency found that the data are

either not relevant to the predictive accuracy of the P-G

class A dispersion coefficients or do not support the com-

menter’s contention.

a. TVA data.—Several commenters recommended that

EPA review monitoring and modeling data available

from three TVA power plants, Paradise, Johnsonville

and John Sevier. In addition, one commenter submitted

a document prepared by a utility consultant entitled

“Summary Report: A comparison of predicted and meas-

ured sulfur dioxide concentrations at the Paradise,

Johnsonville and John Seiver power plants.”

Review of the data for these three power plants dem-

onstrate that oniy one plant [John Sevier] had a moni-

tor located at a distance where maximum ground level

concentrations would be expected to occur during class A

meteorological conditions. Therefore, there is only one

TVA monitor from this study which provides data which

can be used to evaiuate the accuracy of dispersion models

in simulating class A ground level concentrations. Data

from that monitor has been used and it confirms the P-G

class A predictions. See agency’s evaluation of the Sevier

data in section 1.

The Agency also reviewed the utility consultant’s re-

port and found the information so incomplete as to make

evaluation of the purported results impossible. In addi-

tion, since the report does not evaluate model predictions

during class A conditions or investigate the accuracy of

dispersion coefficients, any conclusions regarding the ac-

12 This monitor was located at 1.86 km at an asimuth of 227

degrees from the plant.

54a

curacy of the P-G class A dispersion coefficients would

be inappropriate.

b. Muskingum River data.—One commenter cited

Mills’ validation study of the CRSTER model, “Improve-

ments to the Single Source Model, Volume 3: Further

Analysis of Modelling Results” (1977), as evidence that

the model overestimated concentrations for class A con-

ditions. The study was performed during 1976 at the

Muskingum River power plant. However, at the time the

study was done, the closest monitor to the Muskingum

River plant was 4.6 km away. The next closest monitor

was 5.2 km away. Both were located too far from the

source to measure high concentrations caused during

class A meteorological conditions. .EPA’s modeling for

Muskingum River demonstrated that the class A con-

centration which dictated the power plant’s emission lim-

itation occurred at a receptor located at 1.3 km from the

plant in a completely different direction from the two

monitors. Therefore the monitoring data from the 1976

study do not evaluate the predictive accuracy of the P-G

class A dispersion coefficients.

In January of 1978 the Muskingum River monitor net-

work was expanded to include a monitor at 1.7 km from

the plant. High concentrations would be expected at this

distance during class A meteorological conditions. EPA

reviewed the data from the new monitor and found that

the SO, concentrations recorded confirmed the use of the

P-G class A dispersion coefficients. See discussion of

Muskingum data in Section 1.

c. Big Bend.—One commenter recommended that EPA

review. Vaughan’s 1978 study of the Big Bend power

plant. The commenter alleges that Big Bend field data

were obtained under conditions more analogous to the

Ohio power plants than the conditions at Karlsruhe.

Based on Vaughan’s analysis of the Big Bend data, the

commenter contends that the P-G class A sigma-z values

55a

are too large and that:P-G class B or P-G class C sigma-z

values would provide a better estimation of vertical

dispersion.

However, the Big Bend power plart is located near

Tampa, Florida on the Gulf of Mexico. Review of the

study establishes that atmospheric conditions at the Big

Bend plant during the fiedd study period were signifi-

cantly effected by its location near the Gulf. A large

body of water often has a stabilizing influence on atmos-

pheric conditions near the shoreline. The unstable me-

teorological conditions associated with class A stability

were suppressed during the period of time the field study

was conducted. Therefore, most of Vaughan’s observa-

tions were for neutral or near neutral conditions. For

this reason, the field data collected in the Big Bend

study are simply not relevant to the predictive accuracy

of the P-G class A sigma-z curve.

Moreover, it is not surprising that Vaughan found that

the sigma-z values of P-G class B and P-G class C pro-

vided a good fit to the datz collected at Big Bend. Since

most of the observations were made during neutral con-

ditions, P-G class B, C or D would theoretically better

approximate these conditions. This, however, does not

indicate that the P-G class A sigma-z values are in any

way deficient.

d. East Bend.—One commenter submitted a compari-

son of theoretically- determined .sigma-z values for the

proposed site of the East Bend power plant with sigma-z

values determined by the P-G class A and class B curves.

The commenter contends that this comparison shows that

the P-G class A sigma-z values are too large and that the

P-G class B values compare favorably to the theoretically

determined values.

However, the commenter used a theoretical means of

determining dispersion coefficients which the agency is

developing and which is still in the experimental stage

56a

and unverified. The Agency’s ongoing research in the

area of dispersion coefficients was discussed in the Febru-

ary 7, 1979 notice (44 FR 7805). The Agency is re-

searching a method of determining dispersion coefficients

from on-site measured meteorological data. Accurate

application of the methodology depends on the use of

exact meteorological measurements. Contrary to specific

EPA guidance on the data required, the commenter col-

lected and used data which did not meet. these require-

ments. The commenter’s theoretical comparison employ-

ing an unverified experimental scheme, therefore, is

faulty and does not indicate that the P-G class A sigma-z

values are inaccurate.

e. Gibson data.—One commenter submitted a com-

parison of model predicted values under P-G class A

conditions with monitored values measured at the Gib-

son power plant in southwestern Indiana. The com-

menter concluded that the comparison showed predicted

concentrations were significantly higher than the actually

measured levels. Review of the Gibson data shows that

only one of the three Gibson monitors is located at a dis-

tance where maximum ground level concentrations would

be expected to occur during class A meteorological condi-

tions. This monitor is located at 1.8 km from the plant.

However, even for the relevant monitor, the commenter’s

comparison is faulty.

In making a comparison, the commenter found the

highest one-hour model predictions under class A condi-

tions and compared these with monitor measurements

recorded during the same exact tinhe periods. This com-

parison fails to identify the maximum monitored values

recorded at the Gibson monitors. The comparison, there-

fore, is biased toward showing the model overpredicting.

As recommended in EPA’s Guidance on Air Quality .

Models, an appropriate study of the predictive accuracy —

of the model would compare the highest 3-hour levels

calculated by the model with the highest 3-hour levels

57a

recorded by a monitor. at the site. Commenter’s study,

therefore, is contrary to that recommendation and does

not support a conclusion that the model is overpredicting.

There are also serious questions regarding the reli-

ability of the Gibson monitor data. According to state

officials, the Gibson monitor data has not been quality

assured and should not be relied upon. In addition, dur-

ing the period of time the data was collected, Gibson’s

operation may have been limited to keep concentrations

below the national ambient air quality standards. Ac-

cording to the operators of the Gibson plant, the moni-

toring network is used as part of a supplementary con-

trol system. Therefore, comparisons of this monitor data

with model predictions which are based on full load

parameters would be inappropriate.

f. Conesville data.—In June of 1979, Columbus and

Southern Ohio Electric Company (CSOE) set up four

SO. monitors at distances between 1.0 to 1.4 km from

its Conesville plant. On December 14, 1979, CSOE sub-

mitted data for the first three months of operation.

The agency reviewed the data and has submitted com-

ments to CSOE on the placement of the monitors. Given

the extremely limited amount of data collected under

class A conditions (three three-hour periods) and the

short period of data collection (three months), the agency

determined that at this time the data is too limited to

evaluate the predictive accuracy of the P-G class A dis-

persion coefficients: CSOE, however, did provide an ex-

tremely limited comparison of predicted concentrations

for the periods of observed class A conditions. Those

comparisons demonstrate that for the periods of meas-

ured maximum three hour concentrations, the maximum

observed 3-hour concentrations (highest and~second high-

est concentrations) were unpredicted using all of the

methods proposed by CSOE including the agency’s model-

ing methodology with the P-G class A curve.

58a

g. The remand procedures.—Commenters also crit-

icized the agency for not following the rulemaking proce-

dures set forth in Section 307(d) of the Clean Air Act.

However, the agency has determined that Section 307 (d)

is not applicable to the development of a response on the

class A remand since that activity did not constitute “the

promulgation or revision of an implementation plan”

within the meaning of section 307(b) (1) (B).

On June 29, 1978, the Court remanded the clasis A is-

sue to the agency “for further study”. The Court found

that the agency’s use of the P-G dispersion coefficients

under class A conditions was not adequately supported

by the record. The Court remanded the issue with three

options: (1) to supply the record supporting the agency’s

use of the P-G class A coefficients, (2) to modify the co-

efficients with a new supporting record, or (3) to adopt

the class B proposal suggested by the utilities. The

agency re-evaluated the use of the P-G class A disjpersion

coefficients and has determined that the P-G class A co-

efficients are appropriate for setting emission limitations.

The agency’s response to the remand, therefore, consists

of the compilation of a supplemental record to support

the originally promulgated emission limitations ffor the

four power plants at issue. The agency’s response: to the

remand has not involved either the development of new

coefficients or the substitution of different coefficiemts for

the class A coefficients, either of which would have re-

quired the promulgation of new emission limitatiions in

the implementation plan and therefore would have been

subject to the procedures of Section 307(d). Instead, the

agency has reaffirmed its original implementatiom plan

and developed a supplemental record in support of the

decision.

Commenters have not been prejudiced by the proce-

dures the agency used in responding to the Court’s re-

mand. The agency has followed a course similar to ‘section

307(d) procedures. The agency provided a detailled no-

LS NRO RNC PAGE STAI AD A SENN I AMUN a hE Re

oe

59a

tice of its proposed decision when it published its re-

consideration of the use of the P-G dispersion coefficients

for class A conditions and solicited comments. The

agency also has made available all of the data and ma-

terials it evaluated in its reconsideration and all the pub-

lic comments. Finally, the agency has considered a!l of

the comments and evaluated all of the data submitted or

referenced by commenters, including, where possible, up-

dated data submitted by commenters after the close of

_ the comment period. In short, commenters have had an

adequate opportunity to comment on the agency’s deci-

sion and to submit data for consideration by the agency.

C. Conclusion

The agency has reviewed the use of the P-G class A

dispersion coefficients to set emission limitations for the

Stuart, Cardinal, Muskingum River, and Conesville

power plants in Ohio. As part of that review the agency

has reconsidered the theoretical arguments previously

submitted by the Ohio utilities and reviewed field data

not considered by .the agency during the rulemaking.

The agency has also reviewed all comments submitted on

the agency’s February 7 notice and has evaluated new

alternatives and further data, as suggested by the

commenters.

Commenters have not provided any technical basis to

support changing the agency’s proposed findings. In fact,

utility monitor data support the agency’s determination

that it should continue to use the P-G class A dispersion

coefficients. Moreover, the data establish that all of the

alternatives proposed by commenters are contrary to ex-

perimental and monitor data. In short, the proposed al-

ternatives cannot be accepted by the agency and the

latest experimental and utility monitor data support the

continued use of the P-G class A dispersion coefficients.

Dated: June 16, 1980.

Barbara Blum,

Acting Administrator.

60a

8L Lg st g HO ‘uoqSuryse A JOATY UNSUTASNW

6 622 QT g Nu ‘Suryaey TOAlOg UYOt

(wi) eseg — (+) (ury) qequinn (aqBIg ‘AyuNOD) qU[g JeMog —

YRIS SYBIS SYBIS 10} TWO] UOT}BIO']

eA0qe wory wooly

4310H uoHIeIIG eouw SIC]

1O}IUOW 10} UO IO} UO

‘SHOBIS [BL Aly} Jo wy Z UIAIM SIOpUOW “OS YA sjuBg weM0g °Z BqUL

Sue]

ssouysnoy

wot = wT = WoT = wo T= wot = wgo'> aoej.ins

[PA9]

puno13 [PAR]

WOrZ we6T= wool= wooT= woor= ree asvelat

(q/u)

que song - g/u q/u q/u q/u que song

-uou 19081}

‘syuetd (6161) ~~ (9L6T) young ueAey Sd

IaMog ONO eynis|Iey sYyNAs[lVy -yoolg

‘SOIPNIG Ply uolssedsiq jo siojourereg jejuewledxy ‘T aquy,

6la

Table 3. Highest and second highest 3-hour SO, concentrations

observed and calculated at the closest monitor site at the

John Sevier and Muskingum River power plants.

Monitor Model

___Obeerved Calculated

Year Highest Second Year Highest Second

John Sevier

1973 17 16 1964 382 32

1974 .23 22 1970 26 18

1975 45 389 1971 24 21

1976 40 39 1972 .29 25

1977 43 31 1973 .23 16

1978 34 © 30 1974 21 Ps

Muskingum River

: 1978 39 .20 1964 43 29 "

1979 63 Al 1971 33 25

‘ 1972 40 84

. 1973.40 34

; ! 1974 36 30

1975 55 43

1976 Al B34

Table 4. Muskingum River Power Plant SO, monitors

Height

Distance Azimuth above

from from Stack

Stacks * Stacks * Base

Monitor Name (No.) (km) ot (m)

Center Bend (5) 1.7 37 74

Hackney (2) 4.6 41 74

Rich Valley (3) 8.5 36 105

Caldwell (4) 20.7 33 132

Beverly (1) 5.2 138 56

Mount Olivet (6) 6.1 240 108

* Stacks 1 and 2 are located 645m apart along a 240° azimuth.

Distances and azimuths indicated are from the centroid of the

two stacks.

62a

[Sunuiig Url pezqIWO surerserq |

099 OI-g $-0 02-8T oo" te Ts st 8L/61/TT

009T 0 8-¢ ST-TT &S Ua ve vr 6L/ST/9

00ST £0 o-$ ST-$T t¢ ve os" $T 8L/9/6

(Uy L°0Z) F A0zTUOW

00ST £-0 v-§ vI-sr Tt? ve ao sv 8L/9/6

009T 0 8-L ét-Or &o" or" ve vr 6L/S1/9

V/N or v1 S$ c0° or iv LT 6L/6/9

(UY Gg) § AOzTUOW!

099 OT-0 ot LTI-ST oo vs TS 8st 8L/61/TT

00ST £-0 r-s vI-st jt? ve oe" a 8L/9/6

00sT 0 &-S 9T-PI 6° 8a or" $9" 6L/S/¢

(UL 9h) J AOjIUOW

0g8 0 oT 9T-FT TO 80° Ts 6° 8L/S/6

OSES? O[-9 b-$ 9I-PT 10° LO oo ir 6L/22/8

00ST 0 $3 ST-&T 6T 8a" or 9° 6L/S/¢

(Wy LT) ¢ 10jTUOW

(ul) (syzuez) (Yydur) wiesU0D jo v & 3 g (1X/48q/OW)

qySIOH 1I2A09 pesdg s0pIUO ye AOPIUOP AOIUOF AQIUOF JO}IUOW! UWIBIUOD JO 10JIUOP,

Sulxiq pnojo puImM JUSWsINSvEW 78 SPUSUIOINS VE

uooulayyy aonbedo jo sinoy (uidd) yoUSIP JO soqzeqg

“OS ANoH-g pornsvey 4se43TH

Aeq Suing wonerussu0g

(AM

‘UoySuTJUNH 4@ peinsveul 4YZIey Surxrul 100d pnoyo {e4zIs-u0 peansveur peeds pulM) syusueinsveul “Og

Inoy-g oysty SuLINp S10} WOW AJBATY WNSuLysnN 7 SUOTZBAIOSGO [BOIZO[OI0EOU PUB SUOTZBIZUBDUOD “OS °“S e1qR]

63a

ENVIRONMENTAL PROTECTION AGENCY

[40 CFR Part 52]

[A-5-FRL 1061-3]

APPROVAL AND PROMULGATION OF

IMPLEMENTATION PLANS; OHIO

AGENCY: Environmental Protection Agency.

ACTION: Final rule.

SUMMARY:. This rule restarts compliance schedules

and revises attainment dates in the federally promulgated

Ohio state implementation plan for three power plants. .

Compliance with the emission limitations for the three

power plants was stayed pending the Environmental

Protection Agency’s reconsideration of its original use

of the Pasquill-Gifford (P-G) class A dispersion coeffi-

cients in setting emission limitations for these plants.

The United States Court of Appeals for the Sixth Cir-

cuit remanded the class A record to the agency on June

29, 1978. Cincinnati Gas and Electric Company v. EPA,

578 F.2d 660 (6th Cir. 1978). On June 19, 1980, EPA

published. its reconsideration of the class A issue and

affirmed its original use of the P-G class A coefficients

in setting emission limitations. for these power plants

(45 FR 41501). Thus, on June 19, 1980, EPA affirmed

the emission limitations for the power plants that were

originally promulgated by EPA on August 27, 1976.

41 FR 36324, 40 CFR 52.1881, et seq. In order to bring

the three power plants into compliance with their emis-

sion limitations, EPA is restarting the original com-

pliance schedules. The new starting date for the com-

pliance schedules is June 19, 1980. The attainment date

for these plants is June 19, 1983. EPA is also publish-

ing for comment in today’s Federal Register a proposed

alternative compliance schedule for the Conesville power

plant based on the stated intention of Columbus and

64a

Southern Ohio Electric Company to install a coal wash-

ing facility to achieve compliance. If CSOE chooses not

to install a coal washing facility to achieve compliance,

it must comply with the original compliance schedule.

EFFECTIVE DATE: June 19, 1980.

FOR FURTHER INFORMATION CONTACT: Debra

Marcantonio, U.S. Environmental Protection Agency,

Region V, Air Programs Branch, 230 South Dearborn

Street, Chicago, Illinois 60604, (312) 886-6039.

SUPPLEMENTARY INFORMATION: On June 29,

1978, the U.S. Court of Appeals for the Sixth Circuit

remanded to EPA for its further consideration the

agency’s use of the Pasquill-Gifford dispersion coefficients

in setting emission limitations for isolated rural sources

in Ohio. Cincinnati Gas and Electric Company v. EPA,

578 F.2d 660 (1978). The agency published its pre-

liminary findings on its review of the original class A

record on February 7, 1979. 44 FR 7798. The agency

stated that its preliminary review of additional scientific

data supported EPA’s original decision to use the class A

coefficients in setting emission limitations for isolated

rural power plants in Ohio. Review of the agency’s

modeling determined that only four Ohio power plants

had emission limitations which were determined by P-G

class A conditions. These power plants are Stuart (Day-

ton Power and Light Company), Conesville (Columbus

and Southern Ohio Electric Company), Cardinal and

Muskingum River (Ohio Power Company).

The Court’s remand effectively stayed compliance with

the originally promulgated emission limitations for. the

four plants pending the agency’s reconsideration of the

original class A record. 44 FR 7798 (February 7, 1979).

However, since the agency’s preliminary findings sup-

ported the original use or class A, the agency solicited

comment on whether these utilities would need more t‘me

than allowed under the current compliance schedule to

oa nis Di Sa ans ep

’

&:

y

65a

bring their sources into compliance with their original

emission limitations... 44 FR 7798. The original attain-

ment date for these four plants was June 17, 1980. 40

CFR 52.1875 (1979). EPA stated in the February 7,

1979 notice that, if necessary, it would amend the SIP

to give the utilities more time to comply.?

On June 19, 1980, EPA published its evaluation of

comments on the February 7, 1979 notice and its final

determination on use of the P-G class A coefficients.

45 FR 41501. EPA found that the scientific evidence

and utility monitor data supported the agency’s original

decision to use the P-G class A coefficients in setting

emission limitations for the four Ohio power plants.

Therefore, the agency reaffirmed the originally promul-

gated emission limitations for the four plants on June

19, 1980, and these sources are now required to come

into compliance with those emission limitations? 45 FR

41509 (June 19, 1980).

In response to EPA’s February 7, 1979 notice, only

Dayton Power and Light Company indicated that it

1 The current emission limitations for these plants are as follows:

Stuart—3.16 lbs. SO,/MBTL; Conesville—5.66 Ibs. SO,/MBTU;

Cardinal—4.76 lbs. SO,/MBTU; and Musingum River—6.48 lbs.

SO,/MBTU. See 40 CFR § 52.1881. These are some of the highest

emission limitations in the federally promulgated Ohio SO, plan.

2 Restarting compliance schedules and revising attainment dates

following a remand is proper under the Clean Air Act Section

110(a)(2)(A) 42 U.S.C. § 7410(a)(2)(A). The United States

Court of Appeals for the Sixth Circuit affirmed exactly such an ac-

tion following remand of particulate matter regulations in Ohio.

Northern Ohio Lung Association v. EPA, 572 F.2d 1143 (1978).

3 Ohio Power Company and Columbus and Southern Ohio Electric

Company filed petitions for reconsideration of the agency’s June 19,

1980 determination on class A. Both requested a stay of the deci-

sion pending the agency’s reconsideration or, if the agency denies

the reconsideration, pending judicial review. The agency will re-

spond to the petitions and requests for stay in a separate Federal

Register notice.

66a

would not need more time to comply. Dayton Power and

Light stated that its Stuart power plant is in compliance

with the originally promulgated emission limitation.

There is, therefore, no need to restart the compliance

schedule or revise the attainment date for the Stuart

power plant. However, in order to bring the Conesville,

Muskingum River and Cardinal plants into compliance,

the agency is restarting the original compliance sched-

ules for these plants. This means that these plants will

have to comply with the same interim compliance mile-

stones used for all other SO, sources in the state of Ohio.

These plants will also have the same three year period

for attainment as used for all other SO, sources in Ohio.

In addition to restarting the original compliance

schedules, the agency is proposing to promulgate an al-

ternative compliance schedule for the Conesville plant.

On January 23, 1980, Columbus and Southern Ohio Elec-

tric Company (CSOE) submitted to EPA a proposed coal

washing compliance program and schedule to meet its

original emission limitation. EPA reviewed the CSOE

schedule and determined that it is as expeditious as prac-

ticable and provides for compliance within three years of

approval. In addition, the coal washing program allows

CSOE to continue burning Ohio coal and still meet its

emission limitation. EPA is publishing for comment in

the proposed rules section of today’s Federal Register

the CSOE coal washing compliance schedule as an alter-

native to the original compliance schedule in 40 CFR

§2-1882(b). If CSOE chooses to comply by installing a

coal washing facility, it would have to meet the interim

dates in the coal washing compliance schedule. If CSOE

chooses not to install a coal washing facility, it will be

subject to the interim dates of the original compliance

schedule. EPA is proposing to make the coal washing

schedule subject to the same starting date as the original

schedule.

eras 67a

The agency has determined that the appropriate start-

ing date for the original compliance schedules for these

plants is June 19, 1980. Accordingly, the new attain-

ment date for these plants is June 19, 1983. The Court’s

remand in effect required the agency to stay compliance

for the “class A” power plants until the agency re-

sponded to the remand. The utilities, however, were on

notice that the stay would be lifted once the agency’s

responded to the remand. The utilities were also on

notice that the agency was considering reaffirming their

original emission limitations. 44 FR 7798 (February 7,

1979). In fact, the agency specifically asked the utilities

to comment on whether they would need more time to

comply with their original emission limitations. There-

fore, the utilities had fair warning that they could be

required to comply with their original emission limitz -

tions once the agency responded to the remand. Further

delay in moving these plants toward compliance with

emission limitations that were promulgated in 1976 is

unwarranted and would be contrary to the public

interest.*

The agency has determined that it is unnecessary to

provide for further comment on the original compliance

schedules. Both the utilities and the pu

This text is long and has been trimmed here. Open the source document for the complete record.

This is a copy of a public record, reproduced as it was published. It is not legal advice, and it may not be the version a court would rely on. Check the official source before you cite it.

A word about cookies

We need a few to keep you signed in and the library working. The rest help us see which pages people use and where they get stuck. They stay off unless you say yes.