Worst Case Analyses: A Continued
Requirement under the National

Environmental Policy Act?

Valerie M. Fogleman*

I. INTRODUCTION

In April 1986, the Council on Environmental Quality (the
"CEQ' or the "Council") withdrew its "worst case analysis" reg-
ulation.' The worst case analysis regulation had first appeared in
1977 as part of the CEQ's regulations for implementing the Na-
tional Environmental Policy Act (NEPA). 2

Under NEPA, federal agencies must prepare detailed state-
ments for "major federal actions significantly affecting the quality
of the human environment . .,,- The detailed statements,
known as environmental impact statements ("EISs") must include
an agency's consideration of:

(i) the environmental impacts of the proposed action,
(ii) any adverse environmental effects which cannot be

avoided should the proposal be implemented,
(iii) alternatives to the proposed action,
(iv) the relationship between local short-term uses of man's

environment and the maintenance and enhancement of long-
term productivity, and

* Visiting Instructor, University of Illinois College of Law. J.D. 1986, B.L.A. 1983,
Texas Tech University. This Article was written while the author was Natural Resources
Law Fellow, Northwestern School of Law of Lewis and Clark College. The author would
like to thank Professor William F. Funk of Northwestern School of Law for his helpful
comments on the Article, Lenair Mulford of Northwestern School of Law for her editing
and typing skills, and LetitiaJ. Mallin of Columbia University School of Law for her editing
skills and diligence.

1. Council on Environmental Quality, Final Rule, 51 Fed. Reg. 15,618 (1986); see 40
C.F.R. § 1502.22 (1985) (superseded). In this Article the term "worst case analysis regula-
tion" refers to the regulation which was withdrawn. The current regulation, which re-
placed the worst case analysis regulation in 1986, will be called "the superseding
regulation."

2. 40 C.F.R. § 1500 (1987); see 42 U.S.C. §§ 4321-4347 (1982).

3. 42 U.S.C. § 4332(2)(C) (1982).



54 COLUMBIA JOURNAL OF ENVIRONMENTAL LAw [Vol. 13:53

(v) any irreversible and irretrievable commitments of re-
sources which would be involved in the proposed action should
it be implemented.

4

The CEQs regulations provided a framework to aid federal agen-
cies in their preparation of EISs. As part of that framework, the
worst case analysis regulation addressed an important problem
occasionally faced by the agencies in preparing EISs.

The contours of the problem, as seen by the CEQ are best il-
lustrated by a synopsis of the regulation. If scientific uncertainty
or gaps in relevant information were discovered by an agency
when it was evaluating the significant adverse environmental ef-
fects of its proposed action, the uncertainty or gaps had to be
disclosed. 5 If the relevant unavailable information was essential
to a decision on alternatives to the proposed action and the over-
all costs of obtaining the information were not exorbitant, the in-
formation had to be included in the EIS.6 On the other hand, if
the costs were exorbitant, or if the means of obtaining important
information was unavailable because it was beyond the state-of-
the-art, the agency had to balance the need for the action against
the severity and risk of possible adverse environmental effects if
the action proceeded in the face of scientific uncertainty. 7 A deci-
sion to proceed with an action obligated the agency to include a
worst case analysis in its EIS, together with an indication of the
probability of the occurrence of the adverse impacts. 8 In essence,
the worst case analysis regulation addressed agency actions with
the potential for low-probability but catastrophic environmental
consequences when important information regarding such conse-
quences was unknown or conflicting. The regulation dictated
that if an agency's proposed actions involved a leap into the un-
known, the worst environmental consequences of that action were
to be analyzed.9

4. Id.
5. 40 C.F.R. § 1502.22 (1985) (superseded).
6. Id. § 1502.22(a).
7. Id. § 1502.22(b).
8. Id.
9. See Yost, Don't Gut Worst Case Analysis, 13 Envtl. L. Rep. (Envtl. L. Inst.) 10,394, 10,394

(1983).
An example of a worst case analysis was described in Sierra Club v. Sigler, 695 F.2d 957

(5th Cir. 1983). In Sigler, the Corps of Engineers proposed constructing an oil superport
in Galveston Bay, Texas. A possibility existed that construction of the superport would
result in the total cargo loss of an oil tanker in the bay. Because the environmental conse-



Worst Case Analyses

In 1986, following controversial interpretations of the worst
case analysis regulation by the Ninth Circuit,' 0 the CEQwithdrew
the regulation." The superseding regulation requires federal
agencies to disclose the existence of incomplete or unavailable
information when evaluating reasonably foreseeable significant
adverse environmental effects.' 2 If the superseding regulation is
triggered, the agency must disclose the fact that information is
incomplete or unavailable; state the relevance of such missing in-
formation; summarize "credible scientific evidence" relevant to
an evaluation of reasonably foreseeable significant adverse im-
pacts; and evaluate the impacts by the use of research methods or
theoretical approaches which are generally accepted in the scien-
tific community.iS "Reasonably foreseeable significant impacts"
include environmental effects of low probability but catastrophic
consequences if and only if an analysis of such effects "is sup-
ported by credible scientific evidence, is not based on pure con-
jecture, and is within the rule of reason."' 14

In effect, the superseding regulation shifts the focus of an
agency's inquiry into proposed actions whose environmental con-
sequences involve scientific uncertainty. Whereas the worst case
analysis regulation required an agency to analyze the worst envi-
ronmental consequences of an action where scientific uncertainty

quences were known for only the first twenty-four hours following the spill, the Fifth Cir-
cuit ordered the Corps to prepare a worst case analysis. Id. at 974-75.

The court stated that the worst case analysis could be based on the Corps' twenty-four
hour dispersion model of the oil spill's environmental consequences, as well as on existing
data about the bay's tides and currents. Id. at 974. Thus, although information on the
environmental consequences of an oil spill after its first twenty-four hours was beyond the

state-of-the art, an informative worst case analysis of the spill could be prepared that was
not unreasonably speculative. The analysis would be useful to a decisionmaker in deciding
whether to proceed with the superport's construction. Id. at 974-75.

10. See Save Our Ecosystems (SOS) v. Clark, 747 F.2d 1240 (9th Cir. 1984); Southern
Or. Citizens Against Toxic Sprays (SOCATS) v. Clark, 720 F.2d 1475 (9th Cir. 1983), cert.
denied, 469 U.S. 1028 (1984). See generally Brock, Abolishing the Worst Case Analysis, 2 NAT.
RESOURCES & ENV'T 22, 64-66 (Spring 1986) (criticizing Ninth Circuit's holding in SO-

CATS); Note, The Council on Environmental Quality's Research and iWorst Case Regulation: The
Recent Litigation, 64 OR. L. REV. 547, 558-61 (1986) (discussing SOS and SOCATS); Note,
AEPA's IVorst Case Analysis Requirement: Cornerstone or Stumbling Block, 25 NAT. RESOURCES J.
495, 507-12 (1985) (agreeing with SOCATS); Note, SOCATS: Worst CaseAnalysis in the IlVest, 6
PUB. LAND L. REV. 183, 187-90 (1985) (same).

11. Council on Environmental Quality, National Environmental Policy Act Regulations;
Incomplete or Unavailable Information, Final Rule, 51 Fed. Reg. 15,618 (1986).

12. 40 C.F.R. § 1502.22 (1987).

13. Id. § 1502.22(b).

14. Id.

1987]



56 COLUMBIA JOURNAL OF ENVIRONMENTAL LAW [Vol. 13:53

as to those consequences was involved, the superseding regula-
tion permits an agency to evaluate only the scientific evidence
that it considers to be credible. If an agency does not consider
scientific evidence concerning potential environmental conse-
quences to be credible, it can comply with the superseding regu-
lation merely by disclosing the extent of incomplete or
unavailable scientific evidence and stating the relevance of that
evidence.

Withdrawal of the worst case analysis regulation shattered the
consensus of support behind the CEQ regulations.15 While some
parties viewed the CEQs actions as an improvement to the NEPA
process, others, including members of the Senate Committee on
Environment and Public Works, viewed it as a dilution of the ben-
efits provided by NEPA to the public.' 6

This Article will examine the CEQ's withdrawal of the worst
case analysis regulation and its replacement by the superseding
regulation. Part II will summarize the background of the CEQ
regulations. Part III will discuss the origin, and Part IV, the legal-
ity, of the worst case analysis regulation. Part V will examine the
question of the CEQ's authority to withdraw the worst case analy-
sis regulation. Assuming that the CEQ has the authority to with-
draw it, Part VI will analyze the legality of the CEQ's withdrawal
of the regulation. Part VII will discuss the effect of the supersed-
ing regulation on agency decisionmaking and on judicial review in
cases involving scientific uncertainty. The Article will conclude
that the CEQ's action in withdrawing the worst case analysis regu-
lation and replacing it with the superseding regulation probably

15. Cf Yost, NEPA-The Law That Works, 3 ENVrL. F. 38, 41-42 (Jan. 1985) (describing
enthusiastic welcome accorded CEQ regulations by all interested parties). The term
"CEQ regulations" is used in this Article to mean all the CEQ regulations, not just the
worst case analysis regulation and its replacement.

16. See Letter from SenatorsJ. Randolph, R. Stafford, D. Durenberger & M. Baucus to
A. Alan Hill, Chairman, CEQ (Feb. 2, 1984); see also Department of Housing and Urban Develop-
ment, and Certain Independent Agencies Appropriations for Fiscal Year 1986, Part , Hearings on
H.R. 3629 Before a Subcomm. of the Senate Comm. on Appropriations, 99th Cong., 1st Sess. 661-
709 (1985) [hereinafter 1986 Appropriations Hearings] (comments received by CEQ in re-
sponse to Advance Notice of Proposed Rulemaking). Generally, business and industry
supported the CEQs actions while environmental organizations opposed them. See id. at
665-84 (comments by business and industry); id. at 684-92 (comments by public interest
groups including environmental organizations). Federal agencies were split in their opin-
ions of the CEQ's actions, with the majority in favor of a change to the worst case analysis
regulation. See id. at 699-709 (comments by federal agencies). A split in the opinions of
commenting state governments resulted in a wide variety of suggestions for clarifying the
regulation. See id. at 661-65 (comments by state governments).



1987] Worst Case Analyses

generated more controversy and uncertainty than it resolved. In
at least two circuits, judicial precedent requires preparation of
worst case analyses in situations involving scientific uncertainty.
Therefore, agency compliance with the superseding regulation
does not necessarily mean compliance with NEPA. Thus, in cre-
ating the superseding regulation, the CEQ may have simply ad-
ded a superfluous procedure to those already required by NEPA.

II. THE COUNCIL ON ENVIRONMENTAL QUALITY AND THE

NATIONAL ENVIRONMENTAL POLICY ACT

The CEQ was established in the Executive Office of the Presi-
dent by NEPA in 1970.17 The CEQ's three members, appointed
by the President with the advice and consent of the Senate,' 8 are
supported in their duties by a small staff.' 9 During the early
1970s, over seventy staff members supported the CEQ.2° In
1986, the CEQ's support staff numbered less than ten. 2' The
most drastic reduction occurred in 1981 when fifty-one staff mem-
bers were fired.22

The CEQ was directed by NEPA to review and appraise the ac-
tivities of federal agencies in relation to the Act.23 The small
agency was thus placed in the unenviable position of reviewing

17. 42 U.S.C. § 4342 (1982).
18. Id. The structure of the CEQ is patterned after the Council of Economic Advisors.

See Liroff, The Council on Environmental Quality, 3 Envtl. L. Rep. (Envtl. L. Inst.) 50,051,
50,052 (1973).

19. 42 U.S.C. § 4343 (1982). The support staff is provided by the Office of Environ-
mental Quality created by the Environmental Quality Improvement Act of 1970. 42
U.S.C. §§ 4371-4374 (1982). In practice, the CEQand the Office of Environmental Qual-
ity operate as one entity. Davies & Lettow, The Impact of Federal Institutional Arrangements, in
FEDERAL ENVIRONMENTAL LAw 126, 131 (Envtl. L. Inst. 1974).

20. See CEQ's Role Declines Under Carter, Reagan After Serving as Major Policy-Making Body,
16 Env't Rep. (BNA) 10 (1985).

21. Fiscal Year 1987 Budget Review: Hearings Before the Senate Comm. on Environment and Pub-
lic Works, 99th Cong., 2d Sess. 424 (1986).

22. Department of Housing and Urban Development-Independent Agencies Appropriations for
1983, Hearings Before a Subcomm. of the House Comm. on Appropriations, 97th Cong., 2d Sess.
11 (1982) (statement of Thomas Delaney, Administrative Officer, CEQJ.

23. 42 U.S.C. § 4344 (1982); see Anderson, The National Environmental Policy Act, in FED-
ERAL ENVIRONMENTAL LAw 238, 249 (Envtl. L. Inst. 1974). Other duties of the CEQ in-
clude advising the President on environmental policy, 42 U.S.C. § 4344(4) (1982),
preparing annual environmental quality reports, id. § 4344(1), and analyzing data on envi-
ronmental quality, id. § 4344(5).



58 COLUMBIA JOURNAL OF ENVIRONMENTAL LAW [Vol. 13:53

implementation of the Act by much larger federal agencies unac-
customed to having their actions directed by another agency.2 4

The CEQs relative powerlessness was compounded by the
terms of NEPA. NEPA is written broadly, 25 declaring environ-
mental policy rather than specifying procedures for the Act's im-
plementation. 2 6 It has been suggested that NEPA's supporters in
Congress expected federal agencies to begin acting in an environ-
mentally conscious manner once they had assimilated the knowl-

edge generated by the Act's "action-forcing" provisions.27

Federal agencies, however, proved unwilling to incorporate envi-
ronmental protection into their mandates. 28

Despite the federal agencies' reluctance to comply with NEPA,
the CEQ cautiously strengthened its own role in implementing
the Act. While attempting to avoid the appearance of asserting
control over federal agencies,2 9 the CEQused the authority con-

24. For example, every agency missed the CEQs deadline of October 30, 1973 for pub-
lication of proposed changes to agency regulations to reflect compliance with NEPA. See

Druley, FederalAgency NEPA Procedures, Env't Rep. (BNA) Monograph No. 23, at 1-2 (1976).

25. See City of New York v. United States, 337 F. Supp. 150, 159 (E.D.N.Y. 1972)
(describing NEPA as "broad" and "opaque"); Hanks & Hanks, An Environmental Bill of
Rights: The Citizen Suit and the National Environmental Policy Act of 1969, 24 RUTGERS L. REV.
230, 245 (1970) (describing NEPA as statutory in form but constitutional in spirit); see also
Cramton & Berg, On Leading a Horse to Water: NEPA and the Federal Bureaucracy, 71 MICH. L.
REV. 511, 512-13 (1973) (comparing NEPA to constitutional charter).

26. See R. ANDREWS, ENVIRONMENTAL POLICY AND ADMINISTRATIVE CHANGE 18 (1976);

W. RODGERS, HANDBOOK ON ENVIRONMENTAL LAW 697 (1977).

27. See Anderson, supra note 23, at 242. The major "action-forcing" provision is 42
U.S.C. § 4332(2)(C) (1982), which requires federal agencies to prepare a detailed state-
ment specifying adverse environmental impacts and including alternative actions for "ma-
jor Federal actions significantly affecting the quality of the human environment .... " But
see Caldwell, The National Environmental Policy Act: Retrospect and Prospect, 6 Envtl. L. Rep.
(Envtl. L. Inst.) 50,030, 50,031 (1976) (because NEPA's intent was reformation of adminis-

trative performance, detailed implementation of NEPA was intentionally left to administra-
tive initiative).

28. See, e.g., Calvert Cliffs' Coordinating Comm., Inc. v. AEC, 449 F.2d 1109, 1119
(D.C. Cir. 1971) (criticizing AEC's reluctance to incorporate NEPA procedures into its
review process); Rhode Island Comm. on Energy v. GSA, 397 F. Supp. 41, 58 (D.R.I.

1975) (criticizing GSA's "utter disregard of environmental concerns"). See generally Ander-
son, supra note 23, at 244 (describing agency reaction to NEPA); Lynch, The 1973 CEQ

Guidelines: Cautious Updating of the Environmental Impact Statement Process, 11 CAL. W.L. REV.
297, 300 (1975) (describing "simmering conflict" between agencies' implementation of
NEPA and CEQs interpretation of how NEPA should be implemented).

29. See Fish and Wildlife Miscellaneous-Part 5: Hearings Before the Subcomm. on Fisheries and
Wildlife Conservation and the Environment of the House Comm. on Merchant Marine and Fisheries,
98th Cong., 2d Sess. 51 (1984) [hereinafter Fish and Wildlife Hearings] (statement of R.
Cahn, former CEQ member); See Lynch, supra note 28, at 308, 321, 335 (describing CEQ
policy favoring judicial precedent wherever possible).



Worst Case Analyses

tained in NEPA supplemented by Executive Order No. 11,51430
to issue guidelines to aid federal agencies in their preparation of
EISs. 3 1 In 1971, a landmark decision by the District of Columbia
Circuit solidified the CEQ's role as a major force behind the im-
plementation of NEPA.

In Calvert Cliffs' Coordinating Committee, Inc. v. AEC, the court
used the CEQ's guidelines to reject the Atomic Energy Commis-
sion's (the "AEC") "crabbed interpretation of NEPA."32 The
court favored "[c]ompliance to the 'fullest' extent possible
.... ",33 The AEC rules for implementing NEPA, which were at
issue in the case, had permitted the agency to exclude specifi-
cally, 34 or to ignore generally, 35 environmental considerations if
the agency merely complied with the procedural requirements of
NEPA. The court rejected the AEC's interpretation of NEPA,
quoting from several CEQ guidelines3 6 and favorably contrasting
the guidelines with the AEC's rule. According to the District of
Columbia Circuit, the CEQ guidelines aimed to "'assist agencies
in implementing not only the letter, but the spirit, of the Act,' -37

while the rejected AEC rules were only in technical compliance
with NEPA.38

During much of the period since the Calvert Cliffs' decision, the
courts and the CEQ have acted almost symbiotically in develop-
ing NEPA law by means of judicial interpretation of NEPA and

30. 3 C.F.R. 902 (1970), reprinted in 42 U.S.C. § 4321 note, at 508-10 (1982), as amended
by Exec. Order 11,991, 3 C.F.R. 123 (1978). Executive Order 11,514 was drafted largely by
the CEQ. See R. ANDREWS, supra note 26, at 180 n.20; Liroff, supra note 18, at 50,052.
Originally, the OMB was to be responsible for reviewing federal agencies' compliance with
NEPA procedures. See F. ANDERSON, NEPA IN THE COURTS 11 (1973). OMB did not resist

the CEOs desire to acquire review authority for NEPA. See Liroff, supra note 18, at 50,052.
31. Council on Environmental Quality, Preparation of Environmental Impact State-

ments; Guidelines, 38 Fed. Reg. 20,550 (1973); 40 C.F.R. § 1500 (1974) (superseded).
NEPA provided the CEQ with authority to assist and advise the President in the formu-

lation of national environmental policy. 42 U.S.C. § 4344 (1982). Executive Order No.
11,514 authorized the CEQ to issue guidelines to aid federal agencies in implementing
NEPA. 3 C.F.R. 902 (1970), reprinted in 42 U.S.C. § 4321 note, at 508-10 (1982), as amended
by Exec. Order 11,991, 3 C.F.R. 123 (1978).

32. 449 F.2d 1109, 1117 (D.C. Cir. 1971).
33. Id. at 1118 (quoting 42 U.S.C. § 4332 (1982)). The court defined "to the fullest

extent possible" as setting a high standard which reviewing courts must enforce rigor-
ously. Id. at 1114.

34. See id. at 1122 (citing 10 C.F.R. § 50, app. D, at 249 (1971) (superseded)).
35. See id. at 1117-18 (citing 10 C.F.R. § 50, app. D, at 249 (1971) (superseded)).
36. See id. at 1118 n.19; id. at 1129 n.43.
37. Id. at 1118 n.19 (quoting 36 Fed. Reg. at 7724).
38. Id. at 1117.

1987]



60 COLUMBIA JOURNAL OF ENVIRONMENTAL LAW [Vol. 13:53

the CEQ guidelines3 9 and the CEQs codification of judicial deci-
sions. 40 In 1978, the CEQ's authority to implement NEPA was
further strengthened by Executive Order No. 11,991, which
granted the CEQ authority to issue regulations binding on all fed-
eral agencies.41 The earlier executive order, No. 11,514, had ena-
bled the CEQ to issue guidelines only.4 2

The CEQ regulations are unusual for two reasons. Because
they were issued eight years after NEPA's enactment, the regula-
tions incorporate case law, administrative experience, and the re-
sults of a lengthy and exhaustive rulemaking process.43 Also, the
rules apply broadly to all federal agencies, meaning that federal
agencies must adopt procedures supplementing and conforming
to the CEQ regulations. 44 Normally, agency rules govern only
the actions of the agency which promulgated them.

39. See Izaak Walton League v. Schlesinger, 337 F. Supp. 287, 294-95 (D.D.C. 1971).
See generally Yarrington, The National Environmental Policy Act, Env't Rep. (BNA) Monograph
No. 17, at 11 (1974) ("[i]t is universally accepted that the courts have shaped NEPA, and
the federal process under it, in a manner unprecedented in the history of the development
of federal programs").

40. See 4 CEQ ANN. REP. 234 (1974) (revised CEQ guidelines "incorporate[ ] much of
NEPA's legal evolution in the courts over the past 2 years"); Anderson, supra note 23, at
253 ("relationship between CEQ and the courts can be characterized as one of mutual
support and partnership"); Lynch, supra note 28, at 301 (CEQ guidelines tend to concen-
trate on noncontroversial areas supported by judicial precedent). Compare 40 C.F.R.
§ 1506.1(a) (1987) (restraining federal agencies from taking actions adversely affecting the
environment or limiting their choice of alternatives until a record of decision is issued)
with Scientists Inst. for Public Information, Inc. v. AEC, 481 F.2d 1079, 1090 (D.C. Cir.
1973) (fast breeder reactor program could not proceed to technology development stage
until AEC complied with NEPA procedures).

41. 3 C.F.R. 123 (1978) (amending Exec. Order 11,514, 3 C.F.R. 902 (1970)), reprinted
in 42 U.S.C. § 4321 note, at 508-10 (1982). Executive Order 11,991 was issued after a
congressional oversight committee, the President's Commission on Federal Paperwork,
and commentators had recommended replacement of the CEQ guidelines by binding reg-
ulations. See generally Liebesman, The Council on Environmental Quality's Regulations to Imple-
ment the National Environmental Policy Act-Will They Further NEPA's Substantive Mandate?, 10
Envtl. L. Rep. (Envtl. L. Inst.) 50,039, 50,045 (1980). The Executive Order was drafted at
the CEQ. See generally Yost, supra note 15, at 41.

42. See supra note 30 and accompanying text.
43. See 43 Fed. Reg. at 55,980 (describing background of regulations). Compare Kleppe

v. Sierra Club, 427 U.S. 390, 410-12 (1976) (criteria for cumulative actions under NEPA)
with 40 C.F.R. § 1508.24(a)(1) (1987) (CEQ criteria for cumulative actions); see also infra
note 146 (describing incorporation of scoping procedure from Massachusetts statute into
CEQ regulations).

44. 40 C.F.R. § 1507.3 (1987). The CEQregulations provided a uniform procedure for
implementing NEPA to replace the varying sets of regulations promulgated by over 70
federal agencies. 43 Fed. Reg. at 55,978.



1987] Worst Case Analyses

The basis for many of the CEQ regulations cannot be found in
NEPA or its scant legislative history. For example, the CEQregu-
lations mandate preparation of a draft and a final EIS;45 NEPA
only specifies preparation of "a detailed statement ... "46 Simi-

larly, the CEQ regulations emphasize the importance of public
participation in the NEPA process;47 NEPA is silent on the is-
sue.48  Courts have traditionally deferred to the CEQ regula-
tions,49 however, and the Supreme Court has recognized that the
executive branch made the regulations binding on federal
agencies .50

In summary, despite the disadvantages of its small size and the
vague terms of its enabling legislation, the CEQ today plays a ma-
jor role in the federal government's implementation of NEPA.
The CEO supplementing its powers by means of Executive Or-
ders Nos. 11,514 and 11,991, issued guidelines and, later, regula-
tions which provide the basic framework on which federal
agencies formulate the procedures and criteria used by them in
implementing the Act. The courts' general deference to the CEQ
regulations have further strengthened the CEQs authority to
provide the basis for federal agencies' implementation of NEPA.

III. THE WORST CASE ANALYSIS REGULATION

The worst case analysis regulation stated that:

45. 40 C.F.R. § 1502.9 (1987).
46. 42 U.S.C. § 4332(C) (1982).
47. 40 C.F.R. §§ 1503.1(a)(4), 1503.4(a) (1987).
48. NEPA merely mentions that comments of federal, state, and local agencies are to be

made available to the public through Freedom of Information Act requests. 42 U.S.C.
§ 4332(c) (1982); see 5 U.S.C. § 552 (1982). The emphasis on public participation began
in the CEQ guidelines. See generally Druley, supra note 24, at 17 ("clearest situation in
which the CEQ guidelines and the agency procedures go beyond the requirements of
NEPA is with regard to hearings"); Lynch, supra note 28 at 305 (noting appearance of
public participation requirement in CEQ guidelines).

Executive Order No. 11,514 gave the CEQ the authority to encourage public participa-
tion in the EIS process. 3 C.F.R. 902 (1970), as amended by Exec. Order 11,991, 3 C.F.R.
123 (1978), reprinted in 42 U.S.C. § 4321 note, at 508-10 (1982).

49. See, e.g., Andrus v. Sierra Club, 442 U.S. 347, 357 (1979) (granting substantial defer-
ence to CEQ regulations); Fritiofson v. Alexander, 772 F.2d 1225, 1243 (5th Cir. 1985)
(relying on CEQ regulations and commenting on regulations' consistency with judicial
precedent); Thomas v. Peterson, 753 F.2d 754, 758 (9th Cir. 1985) (relying on CEQregu-
lations and judicial precedent).

50. Andrus v. Sierra Club, 442 U.S. 347, 357 (1979) (citing Exec. Order 11,991, 3
C.F.R. 124 (1978)). The Supreme Court has reserved the issue of whether the CEQregu-
lations bind independent regulatory agencies. Baltimore Gas & Elec. Co. v. Natural Re-
sources Defense Council, Inc., 462 U.S. 87, 99 n.12 (1983).



62 COLUMBIA JOURNAL OF ENVIRONMENTAL LAW [Vol. 13:53

When an agency is evaluating significant adverse effects on
the human environment in an environmental impact statement
and there are gaps in relevant information or scientific uncer-
tainty, the agency shall always make clear that such information
is lacking or that uncertainty exists.

(a) If the information relevant to adverse impacts is essen-
tial to a reasoned choice among alternatives and is not known
and the overall costs of obtaining it are not exorbitant, the
agency shall include the information in the environmental im-
pact statement.

(b) If (1) the information relevant to adverse impacts is es-
sential to a reasoned choice among alternatives and is not
known and the overall costs of obtaining it are exorbitant or
(2) the information relevant to adverse impacts is important to
the decision and the means to obtain it are not known (e.g., the
means for obtaining it are beyond the state of the art) the
agency shall weigh the need for the action against the risk and
severity of possible adverse impacts were the action to proceed
in the face of uncertainty. If the agency proceeds, it shall in-
clude a worst case analysis and an indication of the probability
or improbability of its occurrence. 5 1

The proposed worst case analysis regulation created minimal
controversy when it first appeared in 1977. Its inclusion in the
proposed CEQ regulations was strongly commended although
concern was expressed that it might lead to undue emphasis on
improbable consequences. 5 2 In response, the CEQ amended the
proposed worst case analysis regulation to require a discussion of
the probability of the adverse environmental consequences. 53

The origin of the worst case analysis regulation is disputed.
Some courts and commentators contend that it is a codification of
NEPA case law; 54 other commentators view it as an innovation of

51. 40 C.F.R. § 1502.22 (1985) (superseded).
52. 43 Fed. Reg. at 55,984.
53. Id.; cf. Carolina Envtl. Study Group v. United States, 510 F.2d 796, 799 (D.C. Cir.

1975) (in certain cases it is necessary to consider the probability as well as the conse-
quences of agency actions).

The amendment was also responsive to earlier criticism that worst case scenarios were
included in EISs without being labeled as such. See Bardach & Pugliaresi, The Environmen-
tal-Impact Statement vs. The Real World, 49 PUBLIC INTEREST 22, 29 (Fall 1977).

54. See, e.g., Southern Or. Citizens Against Toxic Sprays, Inc. v. Clark, 720 F.2d 1475,
1478 (9th Cir. 1983), cert. denied, 469 U.S. 1028 (1984); Sierra Club v. Sigler, 695 F.2d 957,
971 (5th Cir. 1983); Bach, Worst Case Analysis: A Historical and Legal Perspective, in PROCEED-
INGS OF A SYMPOSIUM ON WORST CASE ANALYSIS 19, 20 (1985) (worst case analysis regula-
tion "was merely a formal codification of what had been long-standing court-made law");
Yost, supra note 9, at 10,395 ("worst case requirement, though not the nomenclature, ex-
isted before adoption of the CEQ NEPA regulations").



1987] Worst Case Analyses

the CEQ 55 Consensus exists, however, that a requirement to
conduct a worst case analysis was not mentioned in NEPA's legis-
lative history. 56 Congressional committees discussed worst case
scenarios such as the Santa Barbara oil spill, 57 but the possibility
that information on environmental consequences would be unob-
tainable did not seem to occur to NEPA's authors. 58 Not until
after NEPA's enactment did agencies and courts have to face the
problem of whether a project should proceed in the face of scien-
tific uncertainty regarding the project's environmental effects. 59

The worst case analysis regulation provided a means by which
agency actions could proceed when data was unavailable. 60 If sci-

entific data was available, the worst case analysis regulation was
not triggered.

6 1

55. See, e.g., Arnold, Implications of the Worst Case Analysis to Food and Fiber Production, in
PROCEEDINGS OF A SYMPOSIUM ON WORST CASE ANALYSIS 87, 89 (1985) (worst case analysis

regulation "is entirely novel and appears for the first time in [CEQ,] regulations"); Com-

ment, New Rules for the NEPA Process: CEQ Establishes Uniform Procedures to Improve Implementa-
tion, 9 Envtl. L. Rep. (Envtl. L. Inst.) 10,005, 10,008 (1979) (describing worst case analysis
regulation as "a major innovation and improvement in the EIS process").

56. See generally Note, Putting Bite in NEPA 's Bark: New Council on Environmental Quality

Regulations for the Preparation of Environmental Impact Statements, 13 U. MICH. J.L. REF. 367,
384 (1980) (NEPA "completely overlooks the possibility that potentially relevant informa-
tion may be unavailable or cost-prohibitive").

57. See S. REP. No. 296, 91st Cong., 1st Sess. 8 (1969); 3 CEQANN. REP. 222 (1973).
58. See generally Peterson, An Analysis of Title I of the National Environmental Policy Act of

1969, 1 Envtl. L. Rep. (Envtl. L. Inst.) 50,035, 50,039 (1971) (citing NEPA language at 42
U.S.C. § 4332(2)(H) (1982) (requiring federal agencies "to initiate and utilize ecological
information in the planning and development of resource oriented projects" as authoriz-
ing agencies to obtain necessary information in the event of inadequate knowledge of envi-
ronmental effects).

If necessary research is not exorbitantly costly, courts may order agencies to obtain it.

See, e.g., Brooks v. Volpe, 350 F. Supp. 269, 279 (W.D. Wash. 1972) ("NEPA requires each
agency to undertake the research needed to adequately expose environmental harms"),
aft'd, 487 F.2d 1344 (9th Cir. 1973).

59. See Environmental Defense Fund, Inc. v. Corps of Engineers, 348 F. Supp. 916, 927-
28 (N.D. Miss. 1972) (ecosystems analysis adequate for EIS if state-of-art method used),
aff'd, 492 F.2d 1123 (5th Cir. 1974). See generally Village of False Pass v. Watt, 565 F. Supp.
1123, 1149 (D. Alaska 1983) (courts adopted rule allowing agency actions to proceed in
the face of insufficient information), aff'd, 733 F.2d 605 (9th Cir. 1984); Trubek, Allocating

the Burden of Environmental Uncertainty: The NRC Interprets NEPA's Substantive Mandate, 1977
WIs. L. REv. 747, 757 (commenting on lack of direction provided to agencies on how to
proceed when they are faced with scientific uncertainty).

60. Fish and Wildlife Miscellaneous: Hearings Before the Subcomm. on Fisheries and Wildlife Con-
servation and the Environment of the House Comm. on Merchant Marine and Fisheries, Part 2, 95th
Cong., 1st & 2d Sess. 38 (1977-78) [hereinafter NEPA Oversight Hearings] (written answers
by CEQ to committee questions).

61. See Friends of Endangered Species, Inc. v.Jantzen, 589 F. Supp. 113, 116 n.l (N.D.
Cal. 1984), aff'd, 760 F.2d 976 (9th Cir. 1985).



64 COLUMBIA JOURNAL OF ENVIRONMENTAL LAw [Vol. 13:53

After its appearance in the CEQ regulations, the worst case

analysis regulation lay dormant for several years as agencies gen-
erally ignored its existence. 62 In 1983, however, the Fifth Cir-
cuit's decision in Sierra Club v. Sigler63 caused the agencies to
reexamine the need to include worst case analyses in EISs involv-
ing scientific uncertainty. Sigler involved a proposed oil superport
in Galveston Bay, Texas. The Corps of Engineers had prepared

an oil spill analysis in its EIS on the proposed project but did not
prepare an analysis of the environmental consequences of a total
cargo loss by a supertanker in the Bay. 64 The Sierra Club chal-
lenged the Corps' omission by arguing that a worst case analysis
of the total cargo loss was required under NEPA because the envi-
ronmental consequences of such an oil spill beyond the first
twenty-four hours of its existence were unknown.65

The Fifth Circuit upheld the Sierra Club's challenge because
the Club had shown that there was information on which to base
an informative and useful worst case scenario that was not unrea-
sonably speculative. 66 The court ordered the Corps to prepare
the worst case analysis because the potential total cargo loss was a
significant environmental effect of the Corps' proposed action,
and information in the analysis was important to the deci-
sionmaker in deciding whether to proceed with the proposed
action.

6 7

Later in 1983, the Ninth Circuit interpreted the worst case anal-
ysis regulation in Southern Oregon Citizens Against Toxic Sprays v.
Clark (SOC4TS).68 SOCA TS involved a proposal by the Bureau of
Land Management (the "BLM"), a division of the Department of
the Interior, to spray herbicides on some of its forests in Oregon.
The agency had prepared a programmatic EIS updated by annual
environmental assessments for a ten-year spraying program. An
environmental group sued to enjoin the program because the en-
vironmental documents did not contain a worst case analysis.
The group argued that the analysis was required because of scien-

62. See Council on Environmental Quality, Talking Points on CEQs Oversight of
Agency Compliance with the NEPA Regulations (1980) (paper prepared by CEQfor inter-
agency meetings), quoted in Liebesman, supra note 41, at 50,049.

63. 695 F.2d 957 (5th Cir. 1983).
64. Id. at 968.
65. Id. at 973.
66. Id. at 974-75.
67. Id.
68. 720 F.2d 1475 (9th Cir. 1983), cert. denied, 469 U.S. 1028 (1984).



Worst Case Analyses

tific uncertainty regarding the carcinogenicity of the herbicide
2,4-D. 69

In SOCATS, the district court had previously determined that
scientific uncertainty existed regarding the probability of harm to
human health posed by the herbicide spraying program. 70 The
BLM did not appeal the district court's finding of scientific uncer-
tainty. 7' Instead, the agency argued that a worst case analysis was
not required because it was improbable that the herbicides would
adversely affect human health. 72 The Ninth Circuit rejected this
argument, noting that the probability vel non of the herbicides'
adverse effects was the scientific uncertainty in question. 73

Despite the district court's unchallenged finding "that a seg-
ment of the credible scientific community believes that [the pro-
posed] herbicides may initiate cancer at any dosage level," 74 the
BLM proceeded to prepare a worst case analysis assuming that a
safe dosage level existed. 75 The adequacy of that worst case anal-
ysis was subsequently challenged in a second action. 76 In Save
Our Ecosystems v. Clark, the Ninth Circuit affirmed the district
court's determination that the worst case analysis was inadequate
because it did not contain an analysis of the possibility that the
herbicides were carcinogenic or mutagenic. 77 Indeed, the analysis
only discussed the herbicides' probable consequences, omitting
discussion of the worst possible consequences: namely, that the
herbicides were carcinogenic or mutagenic. 78

69. Id. at 1477.

70. Southern Or. Citizens Against Toxic Sprays, Inc. v. Watt, 13 E.L.R. 20,174, 20,176
(D. Or. 1982), aff'd as modified sub nom. Southern Or. Citizens Against Toxic Sprays, Inc. v.
Clark, 720 F.2d 1475 (9th Cir. 1983), cert. denied, 469 U.S. 1028 (1984). The district court
found that scientific uncertainty existed regarding the carcinogenicity and mutagenicity of
at least one herbicide in the spraying program (2,4-D), as well as scientific uncertainty as to
whether there was a safe level of dosage. 13 E.L.R. at 20,176.

71. 720 F.2d at 1478.

72. See id. at 1478-79.

73. Id. at 1479, 1482.
74. Save Our Ecosystems v. Watt, 13 E.L.R. 20,887, 20,888 (D. Or. 1983), aff'd in part

and rev'd in part sub nom. Save Our Ecosystems v. Clark, 747 F.2d 1240 (9th Cir. 1984); see
747 F.2d at 1479.

75. Save Our Ecosystems v. Clark, 747 F.2d 1240, 1245-46 (9th Cir. 1984).
76. Save Our Ecosystems v. Watt, 13 E.L.R. 20,887 (D. Or. 1983), aff'd in part and rev'd

in part sub nom. Save Our Ecosystems v. Clark, 747 F.2d 1240 (9th Cir. 1984).
77. 747 F.2d at 1245-46, aff'g in part and rev g in part Save Our Ecosystems v. Watt, 13

E.L.R. 20,887, 20,888 (D. Or. 1983).

78. 13 E.L.R. at 20,888.

1987]



66 COLUMBIA JOURNAL OF ENVIRONMENTAL LAW [Vol. 13:53

Instead of developing approaches to comply with the worst
case analysis regulation, the agencies reacted defensively to their
judicial defeats in Sigler, SOCA TS, and Save Our Ecosystems. 79 Thus,
in 1984 when the Supreme Court denied the Department of the
Interior's petition for certiorari in SOCA TS, 80 the agencies sought
other means to nullify the Ninth Circuit's opinions. Since they
had failed to persuade the Supreme Court to remove the judicial
precedent, the agencies attempted to remove it by administrative
process.

8 '

In 1984, the Department ofJustice, which represented the BLM
in SOCATS and Save Our Ecosystems, forwarded to the CEQ a draft
amendment of the worst case analysis regulation. 82 The CEQ
subsequently published guidance which read a threshold of rea-
sonable foreseeability of environmental effects into the regulation
in an attempt, according to the CEQ to clarify conflicting posi-
tions on construing the regulation. 83 The guidance was with-
drawn six months later, however, due to public comments
received by the CEQ. 84 Reaction to the CEQs revisions to the
worst case analysis regulation made it apparent that the worst
case regulation could not be changed without "notice and com-
ment" rulemaking procedures. 85 After a petition by the Pacific

79. See Discussion of the Worst Case Analysis, in PROCEEDINGS OF A SYMPOSIUM ON WORST
CASE ANALYSIS 170, 175 (1985) (statement by Dinah Bear, General Counsel, CEQ).

80. 720 F.2d 1475 (9th Cir. 1983), cert. denied, 469 U.S. 1028 (1984).
81. Cf Smyth, What Is the Worst Case?, I NAT. RESOURCES & ENV'T 56, 56 (Winter 1985)

(Supreme Court's denial of certiorari in SOCA TS will probably lead to "a complete over-
haul of the regulation to eliminate the effect of the Ninth Circuit's decision").

82. 40 C.F.R. 1502.22, Proposed Amendment of Regulations Implmenting NEPA, from
Roger J. Marzulla, Department of Justice to John Cooney, Jim Barnes, and Dinah Bear,
CEQ (Dec. 5, 1984); see Fish and Wildlife Hearings, supra note 29, at 13, 27 (statement of A.
Alan Hill, Chairman, CEQ).

83. Council on Environmental Quality, Notice of Proposed Guidance and Request for
Comments, 48 Fed. Reg. 36,486 (1983).

84. Council on Environmental Quality, Notice-Withdrawal of Proposed Guidance
Memorandum for Federal Agency NEPA Liaisons, 49 Fed. Reg. 4803 (1984).

85. See Department of Housing and Urban Development, and Certain Independent Agencies Appro-
priations for Fiscal Year 1985: Hearings Before a Subcomm. of the Senate Comm. on Appropriations,
Part 1, 98th Cong., 2d Sess. 278 (1984) [hereinafter 1985 Appropriations Hearings] (state-
ment of A. Alan Hill, Chairman, CEQ). Several commentators to CEQs proposed gui-
dance had suggested that the regulation be amended in lieu of the publication of
guidance. See id. at 270 (Forest Service); id. at 272 (Edison Electric Institute); cf id. at 275
(National Wildlife Federation) (guidance attempts to amend worst case analysis regulation
without proper rulemaking procedures).



1987] Worst Case Analyses

Legal Foundation to amend the regulation, 6 the CEQ began the
long process that culminated in 1986 with withdrawal of the worst
case analysis regulation in favor of a new regulation.8 7

IV. LEGALITY OF THE WORST CASE ANALYSIS REGULATION

Opponents of the worst case analysis regulation argued that
NEPA's rule of reason was violated either by the regulation,88 the
CEQ's interpretation of the regulation,8 9 or judicial interpreta-
tion of the regulation. 90 However, their arguments are flawed.

NEPA's rule of reason limits discussion of alternative agency
courses of action and environmental effects and risks to those that
are reasonable. The District of Columbia Circuit, in 1972, was
the first court to state that a rule of reason should be applied to
NEPA.

In Natural Resources Defense Council, Inc. v. Morton, the court held
that NEPA "require[d] a presentation of the environmental risks
incident to reasonable alternative courses of action." 9' If an alter-
native was merely a "remote and speculative possibilit[y], ' ' 92 envi-
ronmental effects that could not be readily ascertained did not

86. Petition to the Honorable A. Alan Hill, Chairman, Council on Environmental Qual-
ity Executive Office of the President to Amend 40 C.F.R. § 1502.22 (1983), to Reestablish
a Sound Rule of Reason Governing Worst Case Analysis in Environmental Decision Mak-
ing Mandated by National Environmental Policy Act, § 102(2)(C)(ii), 42 U.S.C.
§ 4332(2)(C)(ii) (July 6, 1984). The Pacific Legal Foundation's petition was submitted to
the CEQunder 5 U.S.C. § 553(e) (1982). See Council on Environmental Quality, Advance
Notice of Proposed Rulemaking, 49 Fed. Reg. 50,744, 50,744 (1984).

87. See Council on Environmental Quality, National Environmental Policy Act Regula-
tions; Incomplete or Unavailable Information, Final Rule, 51 Fed. Reg. 15,618 (1986).

88. See Brock, supra note 10, at 24-25.
89. See Petition to the Honorable A. Alan Hill, Chairman, Council on Environmental

Quality, Executive Office of the President, to Amend 40 C.F.R. § 1502.22 (1983), to Reest-
ablish a sound Rule of Reason Governing Worst Case Analysis in Environmental Decision-
making Mandated by National Environmental Policy Act § 102(2)(C)(ii), 42 U.S.C.
§ 4332(2)(C)(ii) at 5-11 (July 6, 1984).

90. See Arnold, supra note 55, at 89-91; see also 1986 Appropiations Hearings, supra note 16,
at 669 (comments of M. Sullivan, Industrial Forestry Association) (Ninth Circuit opinions
exceed rule of reason; agencies should not be required to speculate regarding environ-
mental effects); id. at 674 (comments of T. Choo, Dow Chemical U.S.A.) (recent court
decisions exceed rule of reason by requiring agencies to evaluate all possible adverse envi-
ronmental impacts where any uncertainty exists about those impacts).

91. 458 F.2d 827, 834 (D.C. Cir. 1972) (emphasis added).
92. Id. at 838. But cf W. RODGERS, supra note 26, at 734 n.87 (describing Teton Dam

failure as remote consequence of EIS examined in Trout Unlimited v. Morton, 509 F.2d
1276, 1283 (9th Cir. 1974)).



68 COLUMBIA JOURNAL OF ENVIRONMENTAL LAw [Vol. 13:53

need detailed discussion.9 3 In other words, the more remote
from serious consideration an alternative seemed, the less de-

tailed the discussion of environmental effects had to be. Con-
versely, if the alternative was more probable, a fuller discussion
was required. Thus, in 1978, when inclusion of the phrase "all
reasonable alternatives" in the CEQregulations9 4 was challenged
on the basis that "it was unduly broad,"9 5 the CEQ was able to
reject the challenge by arguing that courts had consistently lim-
ited the range of alternatives to exclude a full discussion of an
infinite or unreasonable number of alternatives. 9 6

As applied to EISs in which the environmental consequences
are known, the rule of reason means that agencies must fully dis-
cuss reasonably foreseeable consequences. That is, an agency
must discuss all known consequences that are not so remote so as

to be considered insignificant by a decisionmaker.9 7

The worst case analysis regulation did not violate NEPA's rule
of reason. The regulation applied to agency actions for which
sufficient knowledge existed for the action to proceed, but where
scientific knowledge regarding the action's effect on the environ-
ment was uncertain or unknown. Because a boundless number of
scenarios regarding uncertain or unknown environmental effects
could be imagined, the regulation limited the circumstances in
which the analysis was required. Improbable environmental con-
sequences were to be considered in the preparation of an EIS
only if those consequences were potentially catastrophic. The im-
probability of the environmental consequence's occurrence was
to be noted.98 Thus, remote environmental consequences were
to be considered only where it would have been unreasonable for
an agency to proceed without weighing the potentially cata-
strophic effects its action could cause.

93. 458 F.2d at 837-38; see also Environmental Defense Fund, Inc. v. Andrus, 619 F.2d
1368, 1375 (10th Cir. 1980) (applying rule of reason to discussion of alternatives).

94. 40 C.F.R. § 1502.14(a) (1987). Section 1502.14(a) requires agencies to "[rlig-
orously explore and objectively evaluate all reasonable alternatives" to a proposed action.

95. 43 Fed. Reg. at 55,983.
96. Id.

97. See San Luis Obispo Mothers for Peace v. NRC, 751 F.2d 1288, 1300 (D.C. Cir.
1984); Carolina Envtl. Study Group v. United States, 510 F.2d 796, 798 (D.C. Cir. 1975);
see also County of Suffolk v. Secretary of Interior, 562 F.2d 1368, 1383 (2d Cir. 1977)
(applying rule of reason to determine adequacy of environmental effects and alternatives).

98. 40 C.F.R. § 1502.22(b) (1987); see Council on Environmental Quality, Final Regula-
tions, 43 Fed. Reg. 55,978, 55,984 (1978).



Worst Case Analyses

The guidance that the CEQ provided to help agencies comply
with the worst case analysis regulation did not interpret the regu-
lation in such a way that it would violate the rule of reason. While
the guidance explained that worst case analyses were required so
that the fullest possible spectrum of the environmental conse-
quences of agency action would be presented,99 that guidance did
not mean that every low-probability event had to be included.
Worst case analyses, which were to be based on "available infor-
mation, using reasonable projections of the worst possible conse-
quences of a proposed action,"' 100 were to include "a low
probability/catastrophic impact event" as well as more probable
but less catastrophic events.' 0 ' Thus, the CEQs guidence identi-
fied the range of potential environmental effects to be considered
in actions involving scientific uncertainty.

Finally, judicial interpretation of the worst case analysis regula-
tion did not violate the rule of reason. Courts have determined
that it is within the rule of reason to require a worst case analysis
when: (1) information beyond the state of the art is "relevant to a
'significant' effect of a proposal"; 0 2 (2) the analysis would be "in-
formative and useful" to the decisionmaker; 0 3 and (3) "a body of
data [exists] with which a reasonable worst case analysis can be
made that is not unreasonably speculative."'' 0 4 Thus, courts have
limited the circumstances in which a worst case analysis is re-
quired.' 05 According to the courts, the rule of reason permits an
agency to initiate a project involving scientific uncertainty or data
gaps rather than requiring the agency to wait until the uncertainty

99. See Council on Environmental Quality, Forty Most Asked Questions Concerning
CEQs National Environmental Policy Act Regulations, 46 Fed. Reg. 18,026, 18,032
(1981).

100. Id.
101. Id.
102. See Sierra Club v. Sigler, 695 F.2d 957, 974 (5th Cir. 1983).
103. Id.
104. Id. at 974; see Save Our Ecosystems v. Clark, 747 F.2d 1240, 1244-45 (9th Cir.

1984); cf. 1986 Appropriations Hearings, supra note 16, at 675 (comments of Exxon Co.,
U.S.A., to CEQs Advance Notice of Proposed Rulemaking) (describing rule of reason's
application to worst case analysis regulation as twofold: (1) possibility of worst case occur-
ring and (2) scope of analysis required).

105. See Friends of Endangered Species, Inc. v.Jantzen, 589 F. Supp. 113, 116 n.l (N.D.
Cal. 1984) (worst case analysis not required when scientific data were complete for all
significant effects of agency action), aff d, 760 F.2d 976 (9th Cir. 1985); see also Village of
False Pass v. Clark, 733 F.2d 605, 616 (9th Cir. 1984) (analysis of more catastrophic worst

case not required when subsequent offshore leasing process required preparation of an
EIS).

1987]



70 COLUMBIA JOURNAL OF ENVIRONMENTAL LAw [Vol. 13:53

is resolved or the gaps filled.' 0 6 Finally, courts have limited the
scope 10 7 and depth of judicial review of the analysis.10 8 For ex-
ample, no court has found a worst case analysis to be inadequate
because the scenario analyzed was not sufficiently catastrophic. 109

In summary, neither the CEQ nor the courts have ever inter-
preted the worst case analysis regulation to violate NEPA's rule of
reason. Instead, the regulation's application has been limited to
situations in which an informative discussion of the potentially se-
vere environmental effects of an agency action involving scientific
uncertainty would be useful to a decisionmaker. By interpreting
the worst case analysis regulation in light of the rule of reason,
the CEQ and the courts have provided a framework for analyzing
worst cases scenarios that excludes the consideration of fantastic
scenarios.

V. THE CEQs AUTHORITY TO WITHDRAW THE WORST CASE
ANALYSIS REGULATION

The CEQs actions in withdrawing the worst case analysis regu-
lation may be ineffective. Although the agency possesses the au-
thority to change its own regulations as long as the resulting
policies are in accordance with NEPA, it has no authority to
change judicial precedent. The Fifth and Ninth Circuits have de-
termined that the worst case analysis regulation is a codification
of prior case law."10 No court has made a contrary finding."'

106. See Oregon Envtl. Council v. Kunzman, 614 F. Supp. 657, 663 (D. Or. 1985); see
also Save Our Ecosystems v. Clark, 747 F.2d 1240, 1249 (9th Cir. 1984) (if cost of ob-
taining relevant information is exorbitant or if means of obtaining it is beyond the state-of-
the-art, agency must prepare a worst case analysis).

107. See Sierra Club, 695 F.2d at 975 n.14 (agency "need not concern itself with phantas-
magoria hypothesized without a firm basis in evidence").

108. See id. at 976 ("preparation of a worst case analysis cannot be held to the exacting
standards imposed by the trial court because it goes beyond existing knowledge and meth-
ods of acquiring knowledge").

109. See Oregon Envtl. Council v. Kunzman, 636 F. Supp. 632, 640 (D. Or. 1986) (re-
jecting assertion that agency should have analyzed more severe impacts), aff'd, 817 F.2d
484 (9th Cir. 1987); cf note 107 supra.

110. See Southern Or. Citizens Against Toxic Sprays, Inc. v. Clark, 720 F.2d 1475, 1478
(9th Cir. 1983), cert. denied, 469 U.S. 1028 (1984); Sierra Club v. Sigler, 695 F.2d 957, 971
(5th Cir. 1983).

111. The District of Columbia Circuit, the only other circuit that interpreted the worst
case analysis regulation, was silent on the issue. However, the Fifth Circuit cited District of
Columbia Circuit cases as precedent for its finding that the regulation codified case law.
See Sierra Club v. Sigler, 695 F.2d 957, 970-71 (5th Cir. 1983) (citing Alaska v. Andrus,
580 F.2d 465, 473-74 (D.C. Cir. 1978), vacated in non-pertinent part sub nom. Western Oil &



Worst Case Analyses

Thus, because the CEQ is without authority to set aside judicial
precedent," 12 withdrawal of the worst case analysis regulation by
the CEQleaves intact the NEPA common law rule requiring worst
case analyses in EISs involving scientific uncertainty."13 In Oregon
Natural Resources Council v. Marsh, a 1987 case decided after the
CEQ had withdrawn the worst case analysis regulation, the Ninth
Circuit required the Corps of Engineers to prepare a worst case
analysis or conduct further research in a situation involving scien-
tific uncertainty." 4 The court stated that because the worst case
analysis was a codification of judicial precedent, "the rules em-
bodied in the regulation remain in effect even though the regula-
tion was rescinded." 115

If the worst case analysis is a judicially created procedure, the
CEQ's withdrawal of the regulation codifying the procedure may
have brought into issue the survival of the procedure. In Vermont
Yankee Nuclear Power Corp. v. Natural Resources Defense Council,
Inc., 116 the Supreme Court sharply criticized judicial interference
in requiring agencies to follow judicially created procedures in
rulemaking in addition to the procedures mandated by the Ad-
ministrative Procedure Act (APA) and the agency's enabling legis-
lation." 7 The Court overturned a District of Columbia Circuit

Gas Ass'n v. Alaska, 439 U.S. 922 (1978) and Scientists' Inst. for Public Information, Inc. v.
AEC, 481 F.2d 1079, 1092 (D.C. Cir. 1973)); see also Potomac Alliance v. NRC, 682 F.2d
1030, 1036-37 (D.C. Cir. 1982) (requiring agencies to consider probability and possible
environmental effects of actions taken in the face of scientific uncertainty).

112. See Oregon Natural Resources Council v. Marsh, 832 F.2d 1489, 1497 n.8 (9th Cir.
1987).

113. Circuits with no prior decisions involving the worst case analysis regulation will
have to choose between judicial and CEQ interpretation of NEPA. Although courts accord
the CEQ's interpretation of NEPA substantial deference, Andrus v. Sierra Club, 442 U.S.
347, 357 (1979), they do not necessarily accept CEQs interpretation. See Louisiana v. Lee,
758 F.2d 1081, 1083 (5th Cir. 1985) (informal advice of CEQsuch as "Forty Questions" is
not binding), cert. denied, 475 U.S. 1044 (1986); Deukmejian v. NRC, 751 F.2d 1287, 1302
n.77 (D.C. Cir. 1984) (citing Cabinet Mountains Wildemess/Scotchman's Peak Grizzly
Bears v. Peterson, 685 F.2d 678, 682 (D.C. Cir. 1982) (declining to accord substantial
deference to CEQs "Forty Questions")), vacated, 760 F.2d 1320 (D.C. Cir. 1985), cert.
denied, 107 S. Ct. 330 (1986).

114. 832 F.2d 1489, 1496-97 (9th Cir. 1987).
115. Id. at 1497 n.8; see also Methow Valley Citizens Council v. Regional Forester, 833

F.2d 810, 817 n. 1 (9th Cir. 1987) (CEQ's withdrawal of the worst case analysis regulation
"does not nullify the [worst case analysis] requirement").

116. 435 U.S. 519 (1978).
117. Id. at 546-47. The APA prescribes the minimum procedures that must be followed

by federal agencies in conducting formal and informal rulemaking, and formal adjudica-
tion. 5 U.S.C. §§ 553-557 (1982).

19871



72 COLUMBIA JOURNAL OF ENVIRONMENTAL LAW [Vol. 13:53

decision requiring the Nuclear Regulatory Commission (NRC) to
adopt additional procedures during an informal rulemaking hear-
ing.i 8 The Supreme Court held that "absent constitutional con-
straints or extremely compelling circumstances," ' "19 or "totally
unjustified departure[s] from well-settled agency procedures of
long standing,"' 20 an agency should be able to formulate its own
rulemaking procedures free of judicial intervention.' 2 ' The
Court found no support in NEPA for the additional procedural
requirements advocated by the District of Columbia Circuit. Be-
cause the APA controlled the procedures by which factual issues
were investigated during rulemaking, "NEPA [could not] serve as
the basis for a substantial revision of [the APA's] carefully con-
structed procedural specifications .... " 22 The Court remanded
the NRC rule to the District of Columbia Circuit for a review of
the adequacy of the rule's basis under the APA's arbitrary and
capricious standard, but denied the circuit court the power to de-

118. 435 U.S. at 525, revg Natural Resources Defense Council, Inc. v. NRC, 547 F.2d
633 (D.C. Cir. 1976) and Aeschliman v. NRC, 547 F.2d 622 (D.C. Cir. 1976). The cases
involved the informal rulemaking procedures used by the NRC to establish a generic rule
concerning the environmental effects of radioactive waste generated by nuclear power
plants. Id. at 528. The rule, which was to be used in individual licensing proceedings, was
published after the notice and comment procedures of section 4 of the APA. Id. at 529; see
5 U.S.C. § 553 (1982). Although the District of Columbia Circuit did not require the NRC
to follow specific procedures, it found the procedures the agency followed at its hearings
inadequate. 435 U.S. at 541-42.

119. 435 U.S. at 543.
120. Id. at 542.
121. Id. at 543-44. Vermont Yankee received mixed comments from academia. Some

commentators criticized the Court's reliance on the letter of the APA to the exclusion of
judicial interpretation of the Act by the courts since 1946. See, e.g., Davis, Administrative
Common Law and the Vermont Yankee Opinion, 1980 UTAH L. REV. 3, 7-12 (Vermont Yankee, in
forbidding lower courts to follow Supreme Court's example in creating administrative
common law, is contrary to language and legislative history of APA); Stewart, Vermont
Yankee and the Evolution of Administrative Procedure, 91 HARV. L. REV. 1804, 1814-15 (1978)
(Vermont Yankee ignores past decisions fashioning common law under the APA). Other
commentators, however, praised the decision for curtailing hybrid rulemaking by the
courts. See, e.g., Byse, Vermont Yankee and the Evolution of Administrative Procedure: A Some-
what Different View, 91 HARV. L. REV. 1823, 1823 (1978) (welcoming "Vermont Yankee as a
needed corrective to an unwholesome trend in the lower federal courts"); Scalia, Vermont
Yankee: The APA, the D.C. Circuit, and the Supreme Court, 1978 SuP. CT. REV. 345, 369 (Ver-
mont Yankee was directed to curtail the hybrid rulemaking tendencies of the District of Co-
lumbia Circuit). See generally Neely, Vermont Yankee Nuclear Power Corp. v. Natural
Resources Defense Council, Inc.: Response and Reaction in the Federal Judiciary, 14 U. BALT. L.
REV. 256, 262 n.38 (1985) (listing articles discussing Vermont Yankee).

122. 435 U.S. at 548.



19871 Worst Case Analyses

clare the rule invalid on procedural grounds.123 However, Ver-
mont Yankee may not necessarily render a CEQ-abandoned worst
case analysis requirement invalid for the following two reasons.

First, Vermont Yankee "involve[d] rulemaking procedures in their
most pristine sense,"' 124 and did not address preparation of the
EIS accompanying the rule.' 25 Preparation of an EIS is the type
of agency action that has no procedures detailed for it by the
APA. 126 Consequently, it can not be said, as it was in Vermont Yan-
kee, that the APA controls the procedures relevant to the prepara-
tion of an EIS. In spite of this distinction, several lower federal
courts have applied Vermont Yankee to agency actions in addition to
informal rulemaking, 127 including the preparation of EISs. 128 If

123. Id. at 535 n.20, 549; see 5 U.S.C. § 706(e) (1982). The District of Columbia Cir-
cuit's decision that the rule's basis was inadequate was reversed by the Supreme Court
when the Court considered a case it had remanded in light of its ruling in Vermont Yankee.
Baltimore Gas & Elec. Co. v. Natural Resources Defense Council, Inc., 462 U.S. 87 (1983),
revg Natural Resources Defense Council, Inc. v. NRC, 685 F.2d 459 (D.C. Cir. 1982); see
Baltimore Gas & Elec. Co. v. Natural Resources Defense Council, Inc., 435 U.S. 964
(1978) (mem.).

124. 435 U.S. at 524 n.l.
125. See id. at 528-30.
126. The APA contains procedures for informal rulemaking, 5 U.S.C. § 553 (1982), for-

mal rulemaking, id §§ 556-557, and formal adjudication, id. § 554. Courts review an EIS's
adequacy under one of two standards. Some circuits apply a reasonableness standard. See
Save Our Ten Acres v. Kreger, 472 F.2d 463, 466 (5th Cir. 1973); Natural Resources De-
fense Council, Inc. v. Morton, 458 F.2d 827, 834 (D.C. Cir. 1972). Other circuits apply the
APA's "arbitrary and capricious" standard. 5 U.S.C. § 706(2)(A) (1982); see Enos v.
Marsh, 769 F.2d 1363, 1372 (9th Cir. 1985); Village of False Pass v. Clark, 733 F.2d 605,
613 (9th Cir. 1984); Friends of the Earth v. Coleman, 513 F.2d 295, 298 n.3 (9th Cir.
1975). Justice White would like to resolve the conflict between the circuits over the vary-
ing standards ofjudicial review. See River Road Alliance, Inc. v. Corps of Engineers, 475
U.S. 1055, 1056 (1986) (White, J., dissenting from denial of certiorari); Gee v. Boyd, 471
U.S. 1058, 1060 (1985) (White, J., joined by Brennan and Marshall, JJ., dissenting from
denial of certiorari).

127. See Aircraft Owners & Pilots Ass'n v. FAA, 600 F.2d 965, 970 n.25 (D.C. Cir. 1979)
(Vermont Yankee applied to bar additional procedures in informal adjudication); Kenworth
Trucks of Philadelphia, Inc. v. NLRB, 580 F.2d 55, 62-63 (3d Cir. 1978) (Vermont Yankee
applied to bar additional procedures in adjudicatory proceedings despite recognition that
Vermont Yankee involved rulemaking not adjudication). See generally Neely, supra note 121, at
264-306 (discussing manner in which lower federal courts have applied Vermont Yankee).

128. See Sierra Club v. Corps of Engineers, 701 F.2d 1011, 1043 (2d Cir. 1983) (apply-
ing Vermont Yankee to find that "extremely compelling circumstances" did not exist to war-
rant court appointment of special master who proposed intrusive procedures for ensuring
adequate supplementation of EIS); North Slope Borough v. Andrus, 642 F.2d 589, 604
n.83 (D.C. Cir. 1980) (advising lower court to consider Vermont Yankee in deciding whether
to require an agency to discuss an alternative); Gloucester County Concerned Citizens v.
Goldschmidt, 533 F. Supp. 1222, 1228 (D.N.J. 1982) (applying Vermont Yankee to find re-
view of agency procedures governed by APA); National Wildlife Fed'n v. Goldschmidt, 504



74 COLUMBIA JOURNAL OF ENVIRONMENTAL LAW [Vol. 13:53

such an approach continues, it would tend to prevent the federal
courts from requiring a worst case analysis procedure in agency
rulemaking in addition to APA and individual agency require-
ments. However, Vermont Yankee does not foreclose judicial impo-
sition of the worst case analysis procedure.

Second, the worst case analysis procedure may survive because
it can be distinguished from the type of adjudicatory procedures
at issue in Vermont Yankee. 129 The worst case analysis requirement
is a quasi-procedural analytical tool.' 30 The Supreme Court has
approved the judicial imposition of such tools.131 A similar quasi-
procedural tool is the cost-benefit analysis required by the Second
Circuit in EISs. 132 Such requirements merely implement NEPA's
mandate of full disclosure 13 3 and do not impose an adjudicatory
procedure on federal agencies.

There is yet another avenue for challenging the worst case anal-
ysis concept. Although the worst case analysis procedure is
within NEPA's statutory minima, 1 4 it is not explicit in the Act.
Thus the procedure could be precluded by the Supreme Court's

F. Supp. 314, 326 n.41 (D. Conn. 1980) (applying Vermont Yankee to method by which EISs
are prepared), aft'd, 677 F.2d 259 (2d Cir. 1982); Committee Against R.R. Relocation v.
Adams, 471 F. Supp. 142, 145 (E.D. Ark. 1979) (citing Vermont Yankee as authority for not
imposing judicial procedures on agency actions).

129. See Natural Resources Defense Council, Inc. v. NRC, 547 F.2d 633, 642-43 (D.C.
Cir. 1976), rev'd sub nom. Vermont Yankee Nuclear Power Corp. v. NRC, 435 U.S. 519
(1978). The procedures suggested by Judge Bazelon included informal conferences, dis-
covery, interrogatories, and technical advisory committees. Id. at 653.

130. See Fish and Wildlife Hearings, supra note 29, at 39 (statement of N. Yost, Senior Staff
Attorney, Center for Law in the Public Interest); cf Garland, Deregulation and Judicial Re-
view, 98 HARV. L. REV. 505, 538 (1985) (describing requirement that agencies consider
alternative actions as quasi-procedural).

131. See Motor Vehicle Mfrs. Ass'n of United States, Inc. v. State Farm Mut. Auto. Ins.
Co., 463 U.S. 29, 48-51 (1983)); see aLso Garland, supra note 130, at 545 ("[t]he Court could
hardly have been more explicit in approving the judicial imposition of quasi-procedural
requirements").

132. See County of Suffolk v. Secretary of Interior, 562 F.2d 1368, 1384 (2d Cir.), cert.
denied, 434 U.S. 1064 (1977). See generally Rodgers, A Hard Look at Vermont Yankee: Envi-
ronmental Law Under Close Scrutiny, 67 GEO. L.J. 699, 715 n.120 (1979) (cost-benefit analysis
is a procedure in excess of APA requirements).

133. See Weinberger v. Catholic Action of Hawaii/Peace Education Project, 454 U.S.
139, 144 (1981) (describing one of NEPA's aims as disclosing to the public that an agency
has considered the environmental concerns of its proposed actions).

134. Sierra Club v. Sigler, 695 F.2d 957, 970 n.9 (5th Cir. 1983). The worst case analy-
sis procedure is consistent with NEPA's language. See id. at 969 (citing 42 U.S.C.
§ 4331(b) (1982) (responsibility of federal agencies to avoid unintended environmental
consequences); and 42 U.S.C. § 4332(c) (1982) (federal agencies must prepare detailed
statements which are to disclose all environmental impacts)).



Worst Case Analyses

bald statement in Vermont Yankee "that the only procedural re-
quirements imposed by NEPA are those stated in the plain lan-
guage of the Act."' 3 5 The Supreme Court cases on NEPA
generally reflect a narrow reading of the Act's required
procedures.

Nevertheless, exceptions exist. Even while it was admonishing
the District of Columbia Circuit for imposing judicially created
procedures on the NRC, the Court in Vermont Yankee cited approv-
ingly an earlier opinion in which it had agreed with the CEQ that
a "detailed statement" in NEPA meant a comprehensive rather
than a site-specific EIS in certain cases.' 3 6 NEPA contains no ex-
plicit language requiring a comprehensive EIS; this type of EIS
was devised by the CEQin a memorandum.13 7

The Court's earlier recognition that the executive branch made
the CEQ regulations binding on federal agencies 3 8 is difficult to
reconcile with its statements that the procedures NEPA imposes
are limited to those explicit in the Act.' 3 9 The Court must have
realized in that earlier case that the CEQ regulations codified
NEPA case law 140 and incorporated administrative experience.14'

135. Vermont Yankee Nuclear Power Corp. v. Natural Resources Defense Council, Inc.,
435 U.S. 519, 548 (1978) (citing Kleppe v. Sierra Club, 427 U.S. 390, 405-06 (1976)).

136. Id. at 548 (citing Kleppe v. Sierra Club, 427 U.S. 390, 409 (1976)). In K/ppe, the
Court stated that NEPA "may require a comprehensive impact statement in certain situa-
tions where several proposed actions are pending at the same time .... 427 U.S. at 409.

137. The memorandum was relied on by the District of Columbia Circuit in Scientists
Inst. for Public Information, Inc. v. AEC, 481 F.2d 1079, 1090 (D.C. Cir. 1973) (quoting
Council on Environmental Quality, Memorandum to Federal Agencies on Procedures for
Improving Environmental Impact Statements (May 16, 1972), reprinted in 3 Env't Rep.
(BNA) 82 (1972)), and was subsequently included in the CEQ guidelines. See 40 C.F.R.
§ 1500.6 (1973) (current version at 40 C.F.R. § 1502.4(a) (1987)). See generally Druley,
supra note 24, at 11.

138. Andrus v. Sierra Club, 442 U.S. 347, 357 (1979) (citing Exec. Order 11,991, 3
C.F.R. 124 (1978)).

139. See Vermont Yankee Nuclear Power Corp. v. Natural Resources Defense Council,
Inc., 435 U.S. 519, 548 (1978); Kleppe v. Sierra Club, 427 U.S. 390, 405-06 (1976).

140. For example, in Hanly v. Mitchell, 460 F.2d 640, 647-48 (2d Cir.), cert. denied, 409
U.S. 990 (1972), the Second Circuit required agencies to develop a reviewable environ-
mental record of agency actions. The Second Circuit's decision thus required notice and
comment procedures for informal agency adjudications such as planning proposed con-
struction projects. See Note, 51 TEX. L. REV. 1016, 1018 (1973) (reading Second Circuit's
decision as ensuring that a project's opponents would be given notice, and that their com-
ments would be considered by the agency). See generally 4 CEQ ANN. REP. 234 (1974)
(discussing incorporation of case law into NEPA guidelines).

141. For example, the scoping procedure was adopted from the Massachusetts mini-
NEPA. See Department of Housing and Urban Development-Independent Agencies Appropriations for
1980: Hearings Before a Subcomm. of the House Comm. on Appropriations, Part 3, 96th Cong., 1st

1987]



76 COLUMBIA JOURNAL OF ENVIRONMENTAL LAW [Vol. 13:53

Those regulations effectively established procedures which are
not apparent in the plain language of NEPA. Indeed, Justice Mar-
shall had specifically recognized that NEPA's broad statutory
mandate seemed designed to trigger development of a common
law of NEPA. 142

Whether the Court would declare the worst case analysis proce-
dure to be a judicially imposed procedure not contained in
NEPA's plain language is uncertain, therefore, in light of the
Court's earlier recognition of the validity of other procedures im-
posed by CEQ regulation and not apparent from NEPA's plain
language. If the Court does find that the worst case analysis pro-
cedure was a judicially imposed procedure not mandated by
NEPA, federal agencies would then have the option of determin-
ing whether worst case analyses were essential to their proper im-
plementation of NEPA. However, such an action by the Court
would raise the issue of the continued validity of other CEQregu-
lations requiring procedures that are arguably outside NEPA's
mandate.

Sess. 9 (1979) (statement of Nicholas Yost, General Counsel, CEQ). See generally 4 CEQ
ANN. REP. 234 (1974) (discussing incorporation of "experience gained and lessons learned
since 1971" into CEQ guidelines).

The CEQ regulations also allow agencies to create categories of activities for which EISs
are not required. See 40 C.F.R. § 1508.4 (1987). These categorical exclusions are not
mentioned in NEPA or its legislative history.

142. Kleppe v. Sierra Club, 427 U.S. 390, 421 (1976) (Marshall, J., concurring in part
and dissenting in part); see, e.g., City of West Chicago v. NRC, 701 F.2d 632, 648 n. 15 (7th
Cir. 1983) (noting that Second Circuit requires notice and comment procedures to accom-
pany the threshold decision not to prepare an EIS even though that requirement is not
specified in NEPA).

The Court could not have been endorsing the federal agencies' acceptance of proce-
dures developed under NEPA's common law when it stated that the CEQregulations were
binding on federal agencies. See Andrus v. Sierra Club, 442 U.S. 347, 357 (1979). The
agencies were reluctant to follow the CEQ regulations. In fact, some independent agen-
cies asserted that the Court's ruling did not apply to their implementation of NEPA. See,
e.g., Nuclear Regulatory Commission, Environmental Protection Regulations for Domestic
Licensing and Related Regulatory Functions and Related Conforming Amendments, 49
Fed. Reg. 9352, 9352 (1984) ("as a matter of law, the NRC as an independent regulatory
agency can be bound by CEQs NEPA regulations only insofar as those regulations are
procedural or ministerial in nature. NRC is not bound by those portions of CEQs NEPA
regulations which have a substantive impact on the way in which the Commission performs
its regulatory functions"); Federal Energy Regulatory Commission, Notice of Proposed
Rulemaking, 44 Fed. Reg. 50,052, 50,053 (1979) (Commission is not bound by CEQregu-
lations because it is an independent regulatory agency). See generally 4 CEQ ANN. REP. 236
(1974) (discussing tardiness of federal agencies in allowing public participation in NEPA
process as advocated by CEQ).



1987] Worst Case Analyses

Even though the Supreme Court's ruling in Vermont Yankee may
prevent courts from requiring a "worst case analysis" as such, the
"hard look" standard,1 43 used by courts to review the adequacy of
agency actions in implementing NEPA, 144 is still available to en-
able the courts to scrutinize agencies' actions in depth. NEPA is
an environmental full disclosure law.' 45 If a court does not have
evidence on the record that an agency considered the full spec-
trum of environmental effects, it may decide that the agency's
substantive decision was inadequate. 146 Courts are not limited to

143. See Kleppe v. Sierra Club, 427 U.S. 390,410 n.21 (1976) (citing Natural Resources
Defense Council, Inc. v. Morton, 458 F.2d 827, 838 (D.C. Cir. 1972) (court should "insure
that the agency has taken a 'hard look' at environmental consequences"). For a discussion
of the evolution of the hard look doctrine, see Rodgers, supra note 132, at 704-08.

144. Vermont Yankee was remanded to the District of Columbia Circuit to review the ade-
quacy of the rule's basis as perJudge Tamm's suggestion. 435 U.S. at 549. Judge Tamm
had recommended that the NRC take "a hard look at the waste storage issue." 547 F.2d at
658 (Tamm, J., concurring); see also Independent U.S. Tanker Owners Comm. v. Lewis,
690 F.2d 908, 922-23 (D.C. Cir. 1982) (Vermont Yankee does not preclude addition of proce-
dures necessary to ensure that courts can make a thorough and searching review of agency
action); National Indus. Sand Ass'n v. Marshall, 601 F.2d 689, 699 n.35 (3d Cir. 1979)
("[C]ourt did not specifically condemn the 'hard look' doctrine" in Vermont Yankee); Na-
tional Crushed Stone Ass'n, Inc. v. EPA, 601 F.2d 111, 116 (4th Cir. 1979) ("[a]gency
must make reasoned decisions with full articulation of the reasoning and take into account
all relevant factors"), rev'd on other grounds, 449 U.S. 64 (1980); East Texas Motor Freight
Lines, Inc. v. United States, 593 F.2d 691, 695 n.7 (5th Cir. 1979) (Vermont Yankee "did not
purport to address the principle that in order to preserve effective review, a court could
demand a reasoned decision from an administrative agency").

The hard look doctrine has been criticized as undercutting Vermont Yankee. See Natural
Resources Defense Council, Inc. v. NRC, 685 F.2d 459, 517 (D.C. Cir. 1982) (Wilkey, J.,
dissenting) (criticizing "too hard a look" doctrine adopted by majority), rev 'd sub nom. Bal-
timore Gas & Elec. Co. v. Natural Resources Defense Council, Inc., 462 U.S. 87 (1983);
Neely, supra note 121, at 302-03 (District of Columbia Circuit may not always be following
spirit of Vermont Yankee in reviewing agency actions).

145. See Columbia Basin Land Protection Ass'n v. Schlesinger, 643 F.2d 585, 594 (9th
Cir. 1981) (emphasizing NEPA's purpose as full disclosure law); Environmental Defense
Fund, Inc. v. Corps of Engineers, 325 F. Supp. 749, 759 (E.D. Ark. 1971) ("[a]t the very
least, NEPA is an environmental full disclosure law"), vacated, 342 F. Supp. 1211 (E.D.
Ark.), aft'd, 470 F.2d 289 (8th Cir. 1972).

146. See Methow Valley Citizens Council v. Regional Forester, 833 F.2d 810, 818 (9th
Cir. 1987) (agency's discussion in its EIS of relevant environmental effects was inadequate
because dicussion was based on incomplete information); cf. Metropolitan Edison Co. v.
People Against Nuclear Energy, 460 U.S. 766, 776 (1983) (quoting Vermont Yankee Nu-
clear Power Corp. v. Natural Resources Defense Council, Inc., 435 U.S. 519, 558 (1978)
("NEPA's goal [is to] 'insur[e] a fully informed and well considered decision' "); Commit-
tee for Nuclear Responsibility, Inc. v. Seaborg, 463 F.2d 783, 787 (D.C. Cir. 1971) ("full
range of responsible opinion on the environmental effects" must be included in an EIS);
Sierra Club v. Froehlke, 359 F. Supp. 1289, 1342 (S.D. Tex. 1973) (EISs "should contain
'all possible significant effects on the environment' ") (quoting Corps' NEPA regulations,
37 Fed. Reg. 2528 (1972)), rev'd on other grounds sub nom. Sierra Club v. Callaway, 499 F.2d



COLUMBIA JOURNAL OF ENVIRONMENTAL LAW [Vol. 13:53

reviewing an agency's record, but may examine evidence outside
the record to determine whether the agency ignored relevant in-
formation.147 If an agency ignored comments that its assessment
of environmental effects was based on insufficient information, its
determination based on that limited assessment may subse-
quently be viewed by a court as unreasonable. 48 Courts may re-
view opinions of members of a scientific community in order to
determine if an agency's scientific opinion had broad support or
was based on generally accepted methodology. 149 Thus, the hard
look standard enables courts to determine whether a determina-
tion by an agency was grounded on an adequate informational
base.15

0

In addition to ignoring relevant information, agencies some-
times understate negative information. Thus, the hard look stan-
dard is also applicable to agency consideration of unfavorable
environmental concerns. NEPA was designed to mandate the
preparation of EISs t5 l in order that "every significant aspect of
the environmental impact of a proposed action" be consid-
ered. 152 According to the late Judge Leventhal, hard look review

982 (5th Cir. 1974); Environmental Defense Fund, Inc. v. Corps of Engineers, 325 F.
Supp. 749, 759 (E.D. Ark. 1971) (EISs must disclose "all known possible environmental
consequences" (emphasis in original)), vacated, 342 F. Supp. 1211 (E.D. Ark.), aff'd, 470
F.2d 289 (8th Cir. 1972). See generally Axline, Contrarians, 1 NAT. RESOURCES & ENV'T 47, 48
(Spring 1985) ("on its face [the worst case analysis regulation] simply requires that agen-
cies consider the full range of possible impacts from their activities").

147. County of Suffolk v. Secretary of Interior, 562 F.2d 1368, 1384 (2d Cir.), cert. de-
nied, 434 U.S. 1064 (1977); see also National Indian Youth Council v. Andrus, 501 F. Supp.
649, 667 (D.N.M. 1980) ("information outside the administrative record may be necessary
in order for a court to fully discharge its duty under the 'rule of reason' "), aff'd sub nom.
National Indian Youth Council v. Watt, 664 F.2d 220 (10th Cir. 1981).

148. See Sierra Club v. Corps of Engineers, 701 F.2d 1011, 1031 (2d Cir. 1983).

149. See National Indian Youth Council v. Andrus, 501 F. Supp. 649, 667 (D.N.M.
1980), aff'dsub nom. National Indian Youth Council v. Watt, 664 F.2d 220 (10th Cir. 1981).

150. Cf Leventhal, Environmental Decisionmaking and the Role of the Courts, 122 U. PA. L.
REV. 509, 528 (1974).

151. See Dreyfus & Ingram, The National Environmental Policy Act: A View of Intent and Prac-

tice, 16 NAT. RESOURCESJ. 243, 255 (1976).

152. Vermont Yankee, 435 U.S. at 553; see also 115 CONG. REC. 29,059 (1969) (statement
of Sen. Church) (NEPA requires federal agencies to be aware of and concerned with "the
total environmental impact of their actions and proposed programs").

Being required to present convincing reasons why an environmental effect is insignifi-
cant prevents agencies from ignoring criticism and persistent problems involving their ac-
tions. See Steamboaters v. FERC, 759 F.2d 1382, 1393 (9th Cir. 1985); Township of Lower
Alloways Creek v. Public Serv. Elec. & Gas Co., 687 F.2d 732, 741-42 (3d Cir. 1982) (citing
Maryland Capital Park & Planning Comm'n v. United States Postal Serv., 487 F.2d 1029,
1040 (D.C. Cir. 1973)); see also Silva v. Lynn, 482 F.2d 1282, 1285 (1st Cir. 1973) (unsup-



Worst Case Analyses

"requires an analysis of the environmental consequences suffi-
cient to convince a court that they have been considered."' 53 To
prevent federal agencies from simply assuming that an action's
environmental effects are insignificant, and therefore, do not re-
quire disclosure and consideration, reviewing courts require an
agency to provide convincing reasons to support a conclusion of
insignificance. '

54

Fulfillment of the worst case analysis requirement formerly per-
mitted agencies to proceed with actions having uncertain effects,
secure in the knowledge that they had complied with NEPA. In
the future, application of the hard look standard to methodolo-
gies replacing the worst case analysis procedure may result in a
court's rejection of those methodologies, despite their general ac-
ceptance in the scientific community. 155 That is, a reviewing
court may determine that the new procedure fails to adequately
disclose potentially significant environmental effects.

The hard look standard is also used by the courts under NEPA
to review agency consideration of the environmental costs of the
proposed action. Uncertainty 156 and significant environmental
risks are costs which must be factored into a decision to pro-
ceed.' 57 For example, the risk of an accident is an effect requiring
consideration under NEPA. 158 If a worst case analysis is not pre-
pared for a proposed action involving scientific uncertainty, an

ported conclusions are insufficient in an EIS); Environmental Defense Fund v. TVA, 339 F.
Supp. 806, 809 (E.D. Tenn.) (same), aff'd, 468 F.2d 1164 (6th Cir. 1972).

153. Leventhal, supra note 150, at 528.
154. See Silva v. Lynn, 482 F.2d 1282, 1285 (1st Cir. 1973).
155. See Foundation on Economic Trends v. Heckler, 756 F.2d 143, 153-54 (D.C. Cir.

1985) (citing Vermont Yankee, 435 U.S. at 553) (uncertain environmental effects which could
occur if genetically engineered bacteria were to be dispersed must be addressed; court
rejected scientists' evaluation of the effects which had satisfied a scientific review commit-
tee); see also National Indian Youth Council v. Andrus, 501 F. Supp. 649, 667 (D.N.M.
1980) (court may look outside record to determine if agency's scientific methodology is
generally acknowledged by scientific community), af'd, 664 F.2d 220 (10th Cir. 1981).

156. Cf Sierra Club v. Sigler, 695 F.2d 957, 979 (5th Cir. 1983) ("[tlhere can be no
'hard look' at costs and benefits unless all costs are disclosed"); Alaska v. Andrus, 580 F.2d
465, 473 (D.C. Cir.) ("[o]ne of the costs that must be weighed by decisionmakers is the
cost of uncertainty"), vacated sub nom. Western Oil & Gas Ass'n v. Alaska, 439 U.S. 922
(1978).

157. See Baltimore Gas & Elec. Co. v. Natural Resources Defense Council, Inc., 462 U.S.
87, 100 (1983) ("an agency must allow all significant environmental risks to be factored
into the decision whether to undertake a proposed action").

158. Metropolitan Edison Co. v. People Against Nuclear Energy, 460 U.S. 766, 775
(1983). The risk of a catastrophe must be disclosed, but not the public's reaction to it. Id.
at 775 n.9.

1987]



80 COLUMBIA JOURNAL OF ENVIRONMENTAL LAW [Vol. 13:53

environmental cost of the action will not have been addressed.
Thus, if a reviewing court determines that an agency perfunctorily
dismissed uncertainty and risks even though it complied with the
superseding regulation, the agency's consideration of the envi-
ronmental costs of its proposed action may violate NEPA's full
disclosure mandate. 15

9

In conclusion, agencies must continue to comply with the worst
case analysis regulation in order to be sure of passing the "hard
look" standard of judicial review. The validity of judicial imposi-
tion of the procedure withdrawn by the CEQ thus does not de-
pend solely on the fact that Vermont Yankee is distinguishable from
the present situation in the following three respects: (1) the prep-
aration of an EIS is not controlled by the APA; (2) the worst case
analysis is a quasi-procedural analytical tool; and (3) some
Supreme Court precedent has recognized that the CEQ regula-
tions are valid under NEPA. Most importantly, the application of
"hard look review" will require courts to impose a procedural
equivalent to the worst case analysis regulation to ensure that the
potential environmental effects and costs of a proposed action in
cases involving scientific uncertainty have been fully considered.
Only thus can a court enforce NEPA's statutory mandate of full
disclosure.

VI. THE LEGALITY OF THE CEQS WITHDRAWAL OF THE WORST

CASE ANALYSIS REGULATION

It has been argued that the CEQ is entitled to withdraw the
worst case analysis requirement because neither Congress nor the
judiciary have mandated the preparation of worst case analyses by
federal agencies. 160 Allan Brock, who views the CEQOs actions in
withdrawing the requirement as a reinterpretation of NEPA, ar-
gues that "the question [of] whether the Council may choose to
eliminate the worst case analysis concept from its regulations is

answered by the Supreme Court's decision in Chevron."'16 1

159. See, e.g., Columbia Basin Land Protection Ass'n v. Schlesinger, 643 F.2d 585, 594
(9th Cir. 198 1) ("one of the purposes of an EIS is to ensure full disclosure of the environ-
mental consequences of a project"); Sierra Club v. Morton, 510 F.2d 813, 819 (5th Cir.
1975) (quoting Calvert Cliffs' Coordinating Comm. v. AEC, 449 F.2d 1109, 1122 (D.C.
Cir. 1971) (" '[t]he sweep of NEPA is extraordinarily broad, compelling consideration of
any and all types of environmental impact of federal action' ")); Silva v. Lynn, 482 F.2d
1282, 1285 (1st Cir. 1973) ("[NEPA] serves as an environmental full disclosure law").

160. Brock, supra note 10, at 24.
161. Id. at 23. Mr. Brock is an attorney for the Department of the Interior.



Worst Case Analyses

In Chevron, U.S.A., Inc. v. Natural Resource Defense Council, Inc., the
Supreme Court examined the Environmental Protection Agency's
(EPA) interpretation of a section of the Clean Air Act.l6 2 The Act
required permits to be issued for "the construction and operation
of new or modified major stationary sources" in areas that had
not attained national ambient air quality standards. 163 The EPA
had rejected its former definition of "source," which had required
that permits be obtained for new or modified equipment within a
plant if emissions from the new or modified equipment reached a
threshold level.' 64 The EPA then adopted a bubble concept, de-
fining a "stationary source" as an entire plant. Under the new
definition, a permit would not be required for the installation and
modification of individual pieces of equipment within a plant as
long as the total emissions from the plant were not increased. 165

In deferring to the EPA's new interpretation of "source," the
Supreme Court held that if Congress did not intend that statutory
language be given a specific interpretation, courts must defer to a
reasonable interpretation of the language by the agency to which
Congress delegated policymaking responsibilities.' 66 The Court
emphasized the complexity of the Clean Air Act,' 67 Congress's
express delegation of policymaking responsibilities to the EPA,' 68

and the EPA's flexible interpretation of statutory language in rec-
onciling the Act's conflicting policies of reducing air pollution
and encouraging economic growth. 169 Because the EPA made
the policy choice in "a detailed and reasoned fashion," the Court
deferred to the agency's interpretation of "source" as
reasonable.

70

If Brock's analogy from the Chevron precedent to the CEQ's
withdrawal of the worst case analysis regulation were valid, it
would follow that courts should defer to the CEQ's decision to

162. 467 U.S. 837 (1984); see 42 U.S.C. § 7502(b)(6) (1982).
163. 42 U.S.C. § 7502(b)(6) (1982).
164. See 467 U.S. at 857 (citing 45 Fed. Reg. 52,676, 52,697 (1980)).
165. Id. at 858-59 (citing 46 Fed. Reg. 50,766, 50,766 (1981)). The EPA thus came full

circle to a view it had originally espoused before the District of Columbia circuit had or-
dered the agency to change its prior interpretation of "source" to mean individual emis-
sions within a plant. See id. at 864; see also 15 CEQ ANN. REP. 60 (1986) (EPA has favored
the bubble concept since 1975).

166. 467 U.S. at 842-43, 865-66.
167. Id, at 848, 865.
168. Id. at 862, 865.
169. Id. at 863-64.
170. Id. at 865; see id. at 858-59, 863 (describing basis of EPA's policy change).

1987]



82 COLUMBIA JOURNAL OF ENVIRONMENTAL LAW [Vol. 13:53

withdraw. Brock's comparison of the CEQs action in withdraw-
ing the worst case analysis requirement to the EPA's action in re-
interpreting the Clean Air Act is flawed, however, for two
reasons. First, Congress did not delegate policymaking power
under NEPA to the CEQ Second, the CEQ was not reinter-
preting NEPA when it withdrew the requirement.

The CEQ derives its authority to regulate under NEPA from an
Executive Order. 17 1 Nothing in NEPA or the Executive Order
delegates policymaking authority to the CEQ. Thus, the power to
interpret NEPA in order to fill in the gaps in its broad language
belongs primarily to the courts 172 which, however, have accorded
substantial deference to the CEQ's interpretation of the Act. 173

This situation is significantly different from that in Chevron, where
the Supreme Court determined that Congress had implicitly dele-
gated controlling authority to the EPA to construe the Clean Air
Act reasonably. 174 Thus, Chevron does not apply to the CEQ's ac-
tions under NEPA because no statutory delegation of policymak-
ing power to fill in the gaps in NEPA's language has been
made. 17 5

When Congress does delegate authority to an agency to pro-
mulgate regulations, the agency may issue legislative regulations
which will be accorded controlling weight by the courts. 176 In
other words, the courts uphold the regulations "unless they are
arbitrary, capricious, or manifestly contrary to the statute."' 77

Regulations promulgated by an agency that does not have, or is

171. Exec. Order No. 11,514, § 3(h), 42 U.S.C. § 4321 note, at 509 (1982).
172. See Starr,Judicial Review in the Post-Chevron Era, 3 YALE J. ON REG. 283, 299 (1986)

(in statutes that are not complex, "the Court might find no explicit or implicit gap left by
Congress for the agency-rather than the courts-to fill"); see also Kleppe v. Sierra Club,
427 U.S. 390, 421 (1976) (Marshall, J., concurring in part and dissenting in part) (courts
have created a common law of NEPA); Anderson, supra note 23, at 242 (courts are "the
principal enforcers of NEPA").

173. Andrus v. Sierra Club, 442 U.S. 347, 357 (1979); see Anderson, supra note 23, at
342 (CEQ guidelines "to some extent fill in the gaps left by Congress").

174. 467 U.S. at 843-44; see 42 U.S.C. §§ 7411(a)(3), 7502(b)(6), 76020) (1982). See
generally Saunders, Interpretative Rules with Legislative Effect: An Analysis and a Proposalfor Public
Participation, 1986 DUKE L.J. 346, 358 (analyzing Chevron).

175. See Saunders, supra note 174, at 367 ("Chevron clearly applies only when an implicit
or explicit delegation of authority to construe has taken place .... ).

176. Batterton v. Francis, 432 U.S. 416, 425 n.9 (1977). Procedural regulations are a
subset of legislative regulations, which are promulgated by an agency to outline its own
procedures. Such regulations bind the agency legally. See B. SCHWARTZ, ADMINISTRATIVE
LAw § 59, at 159-60 (1976).

177. Chevron, 467 U.S. at 844.



Worst Case Analyses

not exercising, statutorily delegated authority are termed inter-
pretative regulations. 178 The degree of deference that such regu-
lations will be granted by the courts depends on the agency's
expertise, the consistency of its position on the issue, and the tim-
ing of the regulations.179

The CEQ regulations must be interpretative in nature because
the Council has no authority to issue legislative regulations. 80

Only Congress can delegate the authority to promulgate legisla-
tive regulations to an agency; it did not delegate that authority to
the CEQ under NEPA. In fact, Congress may have intended the
Office of Management and Budget to implement NEPA, not the
CEQ 8 1 The conclusion that the CEQregulations are merely in-
terpretative is reinforced by the CEQ's own statements upon its
promulgation of the regulations. Even though the notice and
comment procedures required for legislative regulations were fol-
lowed,' 8 2 the CEQ described its regulations as providing "formal
guidance ... on the requirements of NEPA for use by the courts
in interpreting [NEPA]."' 8 3 The CEQ also acknowledged that it
was issuing the regulations pursuant to an executive directive. 8 4

The Supreme Court's grant of deference rather than control-
ling weight to the CEQ regulations is consistent with their inter-
pretative character. Substantial deference was extended because
of the CEQs creation by NEPA, its advisory responsibilities
under the Act, and the "detailed and comprehensive process" by
which the regulations were formulated. 185 Whether the same

178. See K. DAVIS, ADMINISTRATIVE LAW TREATISE § 7.8 (1958).

179. Batterton v. Francis, 432 U.S. 416, 425 n.9 (1977). A grant of deference means
that the reviewing court defers in part to an agency's interpretation of a statute, but the
court reserves the primary authority to state what the statute means. See Saunders, supra
note 174, at 365.

180. Cf Comment, NEPA After Andrus v. Sierra Club: The Doctrine of Substantial Deference
to the Regulations of the Council on Environmental Quality, 66 VA. L. REV. 843, 844-45 & nn.6, 8
(1980) (describing CEQas presidential advisory body with no statutory authority to issue
or enforce regulations).

181. See S. REP. No. 296, 91st Cong., 1st Sess. 25 (1969). See generally F. ANDERSON, supra
note 30, at 11.

182. See 43 Fed. Reg. 55,978, 55,978 (1978). The CEQ issued draft regulations for
public review and comment according to the APA. Id.; see 5 U.S.C. § 553 (1982). Public
hearings were held and opinions of nearly 12,000 federal, state, and local agencies and
private parties were sought. 43 Fed. Reg. at 55,980.

183. 43 Fed. Reg. at 55,978.
184. Id.
185. Andrus v. Sierra Club, 442 U.S. 347, 357 (1979); see also United States v. Morton,

467 U.S. 822, 834 (1984) (because Congress explicitly delegated authority to the agency to

1987]



84 COLUMBIA JOURNAL OF ENVIRONMENTAL LAw [Vol. 13:53

high level of deference will be accorded to the regulation replac-
ing the worst case analysis regulation is doubtful because of the
circumstances surrounding its adoption. For example, the CEQ's
actions might be interpreted as an attempt to overturn judicial
precedent. 186 In addition, the superseding regulation was not
part of the comprehensive review of the entire NEPA procedure
that the Supreme Court mentioned as a reason for granting sub-
stantial deference to the original regulations. 8 7 Indeed, the su-
perseding regulation contradicts the position that the CEQ took
on scientific uncertainty during that comprehensive review.188 Fi-
nally, the superseding regulation's effect in allowing agencies to
disclose a lesser amount of negative environmental information
does not comport with NEPA's principle of full disclosure.' 8 9

The Chevron precedent does not apply to the CEQs withdrawal
of the worst case analysis requirement for a second reason. The
CEQ was not reinterpreting NEPA when it withdrew the require-
ment. Instead, withdrawal was based on the CEQ's conclusion
that "the worst case analysis requirement [was] flawed,"' 90 and
that "the new requirements provide[d] a wiser and more manage-
able approach to the evaluation of reasonably foreseeable signifi-
cant adverse impacts in the face of incomplete or unavailable
information in an EIS."'' Nowhere in its many notices in the

construe the statute, regulations are legislative and have controlling weight); Herweg v.
Ray, 455 U.S. 265, 274-75 (1982) (because Congress explicitly delegated authority to the
Secretary to interpret the statute, the interpretation has legislative effect).

186. See Methow Valley Citizens Council v. Regional Forester, 833 F.2d 810, 817 n.l I
(9th Cir. 1987); Oregon Natural Resources Council v. Marsh, 832 F.2d 1489, 1496-97 n.8
(9th Cir. 1987); see also supra text accompanying notes 79-82.

187. See Andrus v. Sierra Club, 442 U.S. 347, 357-58 (1979). The CEQregulations were
issued pursuant to an executive order by President Carter ordering the agency to create
uniform regulations to bind federal agencies in their implementation of NEPA. Id. at 357
(citing Executive Order 11,991, 3 C.F.R. 124 (1978)).

188. Cf General Elec. Co. v. Gilbert, 429 U.S. 125, 142-43 (1976). By definition, the
worst case analysis regulation required agencies to consider the worst environmental con-
sequences of agency action involving scientific uncertainty. The superseding regulation
attempts to change this requirement by omitting the language that required the prepara-
tion of a worst case analysis.

189. See 42 U.S.C. § 4321 (1982) (NEPA "promote[s] efforts which will prevent or elimi-
nate damage to the environment"); see also Columbia Basin Land Protection Ass'n v.
Schlesinger, 643 F.2d 585, 594-95 (9th Cir. 1981) (emphasizing NEPA's purpose as a full
disclosure law); cf Morton v. Ruiz, 415 U.S. 199, 237 (1974) ("[i]n order for an agency
interpretation to be granted deference, it must be consistent with the congressional pur-
pose" (citations omitted)).

190. 51 Fed. Reg. at 15,624.
191. Id. at 15,620.



Worst Case Analyses

Federal Register regarding the worst case analysis regulation does
the CEQ purport to be reinterpreting NEPA. Chevron, on the
other hand, was a case of strict statutory construction. 92

In Chevron, the EPA's re-interpretation of the Clean Air Act was
found to be reasonable. But the CEQ cannot show that its "re-
interpretation" of the worst case analysis regulation was reason-
able according to the Chevron standard, 9 3 because the action the
CEQtook was not a re-interpretation. The CEQinstead made an
administrative change, hoping to make NEPA more manageable
by federal agencies. The CEQs action, therefore, does not come
within the deferential framework of Chevron, but should be judged
by the more exacting standard of Motor Vehicle Manufacturers Associ-
ation, Inc. v. State Farm Mutual Insurance Co. 194 This standard dic-
tates that when an agency "changes its view about the wisdom of a
rule [it must] demonstrate not only that the new rule is reason-
able, but also that the agency's decision to change course is
reasonable." 95

In State Farm, the Supreme Court determined that the National
Highway Traffic Safety Administration's (NHTSA) rescission of
its passive restraints rule was arbitrary and capricious. 196 In
1977, the agency promulgated a rule requiring passive restraints
(air bags or automatic seat belts) to be installed in large cars by
model year 1982 and in all cars by model year 1984.197 Four
years later, after extending the deadline for passive restraints in
large cars because of economic difficulties in the automobile in-
dustry, the Secretary of Transportation rescinded the rule.' 98 The
NHTSA explained that rescission occurred because the agency
could no longer find that the rule would produce significant safety
benefits. 199

192. 467 U.S. at 859-62 (interpreting "source" in Clean Air Act); see Chemical Mfrs.
Ass'n v. Natural Resources Defense Council, Inc., 470 U.S. 116, 125-26 (1985) (interpret-
ing "modify" in Clean Water Act); see also Garland, supra note 130, at 550 ("Court treated
Chevron purely as a case of statutory construction").

193. See 467 U.S. at 866.

194. 463 U.S. 29 (1983).
195. Starr, supra note 172, at 298.
196. 463 U.S. at 46.
197. Id. at 37 (citing 42 Fed. Reg. 34,289, 34,290 (1977)).
198. Id. at 38 (citing 46 Fed. Reg. 53,419, 53,419 (1981)).

199. Id. In 1977, the agency had found that the rule would produce significant safety
benefits. 42 Fed. Reg. 34,289, 34,290 (1977). Since 1977, the NHTSA standard had been
modified to permit automatic seat belts to be manufactured so that they could be de-

1987]



86 COLUMBIA JOURNAL OF ENVIRONMENTAL LAW [Vol. 13:53

The Supreme Court determined that, although the NHTSA was
entitled to change its course of action, it could not do so without
providing a reasoned analysis for the change as well as for its new
course.200 The NHTSA's rescission was arbitrary and capricious,
the Court concluded, because the agency had not made a 'rational
connection between the facts and its judgment. 20 1

The facts surrounding the CEQs rescission of the worst case
analysis regulation resemble the background of the State Farm
case. The Supreme Court determined in State Farm that the Act
under which the NHTSA acted was passed "because the industry
was not sufficiently responsive to safety concerns." 20 2 NEPA was
enacted because federal agencies were not sufficiently responsive
to environmental concerns. 20 3 Also, just as the NHTSA reacted
to the problems the automobile industry encountered in comply-
ing with its passive restraints rule,20 4 the CEQreacted to the dif-
ficulties experienced by federal agencies in complying with its
worst case analysis regulation. 20 5

More importantly, State Farm should govern the CEQs actions
because of the very nature of those actions. Since the CEQ was
not reinterpreting NEPA when it replaced the worst case analysis
regulation but was changing its opinion about the wisdom of the
regulation itself, State Farm controls.20 6 The CEQ must, there-
fore, provide a reasonable explanation for the change in its
course of action as well as for the adoption of the superseding

tached. The agency determined that because many people would detach the seat belts, the
anticipated safety benefits would not be realized. See 463 U.S. at 47.

200. 463 U.S. at 42, 57.
201. Id. at 56.
202. Id. at 38.
203. See generally Fish and Wildlife Hearings, supra note 29, at 38 (statement of N. Yost,

Senior Staff Attorney, Center for Law in the Public Interest).
204. 463 U.S. at 49.
205. See 6 CEQANN. REP. 628 (1976); see also Yarrington, supra note 39, at 37 (language

and legislative history of NEPA confirm the Act's purpose in substantially changing the
decisionmaking procedures of federal agencies).

206. 463 U.S. at 42; cf 51 Fed. Reg. at 15,620 ("CEQis amending this regulation be-
cause it has concluded that the new requirements provide a wiser and more manageable
approach to the evaluation of reasonably foreseeable significant adverse impacts in the
face of incomplete or unavailable information in an EIS").



Worst Case Analyses

regulation. 207 This explanation must have an evidentiary basis in
order to withstand judicial scrutiny.20 8

The CEQ began to express its concern about the worst case
analysis regulation as early as 1983. At that time, in publishing
guidance on the worst case analysis regulation, the agency stated
its concern with "a wide variety of conflicting interpretations by
both federal agencies and reviewing courts."20 9 Two years later,
the CEQ narrowed its focus to judicial interpretation of the re-
quirement. According to the agency, "the requirement to pre-
pare a 'worst case analysis' in certain circumstances has been the
impetus for judicial decisions which require federal agencies to go
beyond the 'rule of reason' in their analysis of potentially severe
impacts." 210 The requirement allegedly "challenge[d] the agen-
cies to speculate on the 'worst' possible consequences of a pro-
posed action," 21' and was "counterproductive, because it has led
to agencies being required to devote substantial time and re-
sources to preparation of analyses which are not considered use-
ful to decisionmakers and [which] divert the EIS process from its
intended purpose."21 2 The CEQ stated that the Ninth Circuit
had ordered a federal agency to "present a discussion of a partic-
ular disastrous impact even when the agency believe[d] that no
credible scientific data ha[d] indicated that the particular impact
could be caused by the proposed action." 2 13 It can be inferred

207. See Starr, supra note 172, at 298. Judge Starr, of the District of Columbia Circuit,
stated that "if an agency changes its view about the wisdom of a rule, State Farm requires
the agency to demonstrate not only that the new rule is reasonable, but also that the
agency's decision to change course is reasonable." Id.

208. Bowen v. American Hosp. Ass'n, 476 U.S. 610, 643 (1986) (citing State Farm, 463
U.S. at 50).

209. Notice of Proposed Informational Guidance and Request for Comments, 48 Fed.
Reg. 36,486 (1983).

210. Proposed Amendment to 40 C.F.R. § 1502.22, 50 Fed. Reg. 32,234, 32,234
(1985); see also id. at 32,237 ("[t]he Council believes that the current 'worst case analysis'
requirement, as interpreted by recent judicial decisions, imposes a requirement on the
agencies which goes beyond [the] 'rule of reason' ").

211. Id. at 32,236.

212. Id.
213. Id.; see also Department of Housing and Urban Development-Inependent Agencies Appropri-

ations for 1986: Hearings Before a Subcomm. of the House Comm. on Appropriations, Part 4, 99th

Cong., Ist Sess. 61 (1985) (CEQ describing herbicide spraying cases as an impetus for
reopening rulemaking on worst case analysis regulation); Department of Housing and Urban
Development-Independent Agencies Appropriations for 1985: Hearings Before a Subcomm. of the

House Comm. on Appropriations, Part 3, 98th Cong., 2d Sess. 32 (1984) (statement of A. Alan

19871



88 COLUMBIA JOURNAL OF ENVIRONMENTAL LAW [Vol. 13:53

from these remarks that the main reason behind the CEQ's
rulemaking was the Ninth Circuit rulings.

In 1986, when the CEQ rescinded the worst case analysis re-
quirement, the agency explained that it agreed with the require-
ment's goals of disclosing the extent to which information was
incomplete or unavailable, acquiring that information if reason-
ably possible, and evaluating "reasonably foreseeable significant
adverse impacts even in the absence of all information." 21 4 How-
ever, the CEQ said that it had chosen to rescind the requirement
because it was "an unproductive and ineffective method of achiev-
ing [the requirement's] goals; one which can breed endless hy-
pothesis and speculation.- 2 1 5 The CEQ again drew on evidence
from the Ninth Circuit to support its remarks about the in-
feasibility of the worst case analysis regulation. According to the
CEQ in Save Our Ecosystems v. Clark, the Ninth Circuit had ordered
the BLM to prepare "a conjectural analysis, lacking a credible sci-
entific basis" which appeared to be an indulgence in speculation
for its own sake. 216

However, the CEQwas misreading the Ninth Circuit's opinion.
The trial court carefully weighed the competing views regarding
the credibility of the scientific base before ordering the BLM to
prepare that analysis. 2 17 That ruling was upheld by the Ninth Cir-
cuit. 21 8 Thus, the Save Our Ecosystems court did not read the worst
case analysis regulation to require "endless hypothesis and specu-
lation." Instead, the court ordered the BLM to prepare a worst
case analysis only after it had been satisfied that a credible scien-

Hill, Chairman, CEQ) (discussing "considerable litigation" involving worst case analysis
regulation "on the West Coast").

The proposed amendment of the worst case analysis regulation prepared by the Depart-
ment of Justice pinpointed one decision that, the Department stated, "even held that a
worst case analysis is required unless a federal agency can disprove an assertion that a
particular consequence is arguably possible." 40 C.F.R. 1502.22, Proposed Amendment
of Regulations Implementing NEPA, from Roger J. Marzulla, Department of Justice, to
John Cooney, Jim Barnes, and Dinah Bear, CEQ (Dec. 5, 1984). The Department ofJus-
tice represented the BLM in SOCATS and Save Our Ecosystems.

214. 51 Fed. Reg. at 15,620.
215. Id.

216. 50 Fed. Reg. at 32,236 (citing 747 F.2d 1240 (9th Cir. 1984)).
217. Save Our Ecosystems v. Watt, 13 E.L.R. 20,887, 20,888 (D. Or. 1983), aff'd in part

and rev'd in part sub nom. Save Our Ecosystems v. Clark, 747 F.2d 1240 (9th Cir. 1984).
The trial judge was "satisfied by the evidence presented at trial that some degree of

uncertainty exists in the scientific community as to the potential of some or all of these
herbicides for causing cancer and genetic mutation." Id.

218. Save Our Ecosystems v. Clark, 747 F.2d 1240, 1245-46 (9th Cir. 1984).



Worst Case Analyses

tific basis for the analysis existed. This approach is consistent
with all other judicial interpretations of the worst case analysis
requirement, none of which have required "speculation" beyond
that implicit in NEPA. 219

As with its concern about undue speculation, the CEQ's con-
cerns about conflicting interpretations by courts and federal
agencies, counterproductivity, and violations of the rule of reason
engendered by the worst case analysis regulation are unfounded.

The CEQs reference to the "wide variety of conflicting inter-
pretations [of the regulation] by both federal agencies and re-
viewing courts" 220 suggested that reviewing courts interpreted
the worst case analysis regulation inconsistently. However, the
Fifth and Ninth Circuits, the only circuits which had reviewed the
regulation, did construe the regulation consistently. In fact, the
Fifth Circuit formulated guidance to aid federal agencies in pre-
paring worst case analyses, 22' thus alleviating conflicts in inter-
pretations among agencies regarding the regulation's
requirements. The only unresolved conflict was the BLM's per-
sistent argument that it should not have been ordered by the
Ninth Circuit to prepare a worst case analysis not based on any
credible scientific evidence. 222

The CEQs assertion that the worst case analysis regulation was
counterproductive because it required agencies to devote sub-

219. See Scientists' Inst. for Public Information, Inc. v. AEC, 481 F.2d 1079, 1092 (D.C.
Cir. 1973); cf. North Slope Borough v. Andrus, 642 F.2d 589, 605 (D.C. Cir. 1980) (quot-
ing North Slope Borough v. Andrus, 486 F. Supp. 326, 347 (D.D.C. 1979): worst case
analysis in lease sale EIS " 'was a reasonable means of alerting the decisionmaker to the
dangers presented by proceeding in the face of uncertainty' "). Because NEPA requires an
agency to predict the environmental consequences of its proposed actions, it necessarily
requires an agency to speculate on what those consequences will be. See Scentists'Inst., 481
F.2d at 1092.

220. 48 Fed. Reg. at 36,486.
221. See Sierra Club v. Sigler, 695 F.2d 967, 974-75 (5th Cir. 1983).
222. In comments submitted to the CEQ during rulemaking, Interior continued to con-

tend that "[t]he record disclosed no credible, scientific evidence that 2,4-D or any of the
other proposed herbicides has any carcinogenic effect; and there was no data base upon
which any meaningful evaluation of the herbicides' supposed carcinogenicity could be
based." Comments to CEQ by Interior on the CEQs Advance Notice of Proposed
Rulemaking at 11 (Feb. 15, 1985). Interior also criticized the Ninth Circuit for "not even
consider[ing] whether a worst case analysis on the herbicides' effect(s) on human health
would reasonably affect the BLM's decision to proceed with its proposal." Id. at 10. But see
Friends of River v. FERC, 720 F.2d 93, 108 (D.C. Cir. 1983) ("Courts have low tolerance
for futility arguments pressed by parties who seek release from an obligation to engage in
serious environmental decisionmaking on the ground such an enterprise would not
change the party's mind about a planned development").

1987]



90 COLUMBIA JOURNAL OF ENVIRONMENTAL LAW [Vol. 13:53

stantial time and resources to analyses that agency deci-
sionmakers did not consider useful recalls early agency reactions
to the NEPA process. 2 23 Such an argument is an attempt to ex-
cuse non-compliance with NEPA on the basis of administrative
convenience. Similar arguments have been frequently rejected by
the courts. 2 24 The CEQalso overlooks the fact that agencies have
always had difficulty in complying with NEPA's statutory mandate
of forecasting the nature of their projects' environmental effects
in addition to complying with the worst case analysis regula-
tion. 2 25 Mere difficulty in carrying out a congressional mandate,
however, does not excuse agencies from complying with it.
Therefore, agencies may not weaken NEPA's statutory mandate
solely in order to comply more easily with it.226

Finally, the CEQs conclusion that the worst case analysis regu-
lation required agencies to go beyond the rule of reason22 7 has no
basis in fact. According to the CEQ "a specific component of the
'rule of reason' " is "that the analysis of impacts be based on
credible scientific evidence . "..."228 The Ninth Circuit decision
cited by the CEQ did not abandon the rule of reason but stated
that worst case analyses should be reasonable, 2 29 as should the

223. See generally NEPA Oversight Hearings, supra note 60, at 8 (statement of Charles War-
ren, Chairman, CEQ) (describing grudging compliance with NEPA by agencies in the early
1970's); Cortner, A Case Analysis of Policy Implementation: The National Environmental Policy Act
of 1969, 16 NAT. RESOURCES J. 323, 324-25 (1976) (describing agency complaints that
"NEPA procedures are too costly, time-consuming, inflexible, cumbersome, and
detailed").

224. See Calvert Cliffs' Coordinating Comm. v. AEC, 449 F.2d 1109, 1121 (D.C. Cir.
1971); Brooks v. Volpe, 350 F. Supp. 269, 283 (W.D. Wash. 1972); cf. Conservation Law
Foundation of New England, Inc. v. GSA, 707 F.2d 626, 633 (1st Cir. 1983) (rejecting
GSA's argument that speculation required by NEPA "is meaningless and a waste of agency
resources" because of circumstances); Prince George's County v. Holloway, 404 F. Supp.
1181, 1184 (D.D.C. 1975) (NEPA is more than "a mere formality which busy bureaucrats
can treat as an annoyance").

225. See Lynch, supra note 28, at 329. Mr. Lynch attributed the difficulty in forecasting
to bureaucratic caution about speculating. Id.

226. See Letter from SenatorsJ. Randolph, R. Stafford, D. Durenberger & M. Baucus of
the Senate Committee on Environment and Public Works to A. Alan Hill, Chairman, CEQ
(Feb. 22, 1984) (new regulation weakens environmental protection provided by NEPA).

227. 50 Fed. Reg. at 32,234.
228. 51 Fed. Reg. at 15,624.
229. Save Our Ecosystems v. Clark, 747 F.2d 1240, 1245 (9th Cir. 1984) (citing Sierra

Club v. Sigler, 695 F.2d 957, 974 (5th Cir. 1983)); see also Village of False Pass v. Watt, 565
F.2d 1123, 1152 (D. Alaska 1983) (rule of reason mandates against requiring worst case
analysis on oil spill at early stage of offshore oil and gas lease), aff'd sub nom. Village of
False Pass v. Clark, 733 F.2d 605 (9th Cir. 1984).



Worst Case Analyses

speculation and forecasting involved. 230 Indeed, the court had
'equired plaintiffs to show that a credible scientific database ex-
isted before ordering the BLM to prepare the analysis. 23 l

Surprisingly, in the preamble to the final version of the super-
seding regulation, the CEQ advocates "evaluation of reasonably
foreseeable significant adverse impacts even in the absence of all in-

formation."23 2 The CEQis thus endorsing the preparation of eval-
uations based on pure conjecture, a requirement which directly
contradicts one of its purported reasons for issuing the new
regulation.

233

The CEQs superseding regulation reads:
When an agency is evaluating reasonably foreseeable signifi-

cant adverse effects on the human environment in an environ-
mental impact statement and there is incomplete or unavailable
information, the agency shall always make clear that such infor-
mation is lacking.

(a) If the incomplete information relevant to reasonably
foreseeable significant adverse impacts is essential to a rea-
soned choice among alternatives and the overall costs of ob-
taining it are not exorbitant, the agency shall include the
information in the environmental impact statement.

(b) If the information relevant to reasonably foreseeable
significant adverse impacts cannot be obtained because the
overall costs of obtaining it are exorbitant or the means to ob-
tain it are not known, the agency shall include within the envi-
ronmental impact statement: (1) A statement that such
information is incomplete or unavailable; (2) a statement of the
relevance of the incomplete or unavailable information to eval-
uating reasonably foreseeable significant adverse impacts on
the human environment; (3) a summary of existing credible sci-
entific evidence which is relevant to evaluating the reasonably
foreseeable significant adverse impacts on the human environ-
ment, and (4) the agency's evaluation of such impacts based
upon the theoretical approaches or research methods generally
accepted in the scientific community. For the purposes of this
section, "reasonably foreseeable" includes impacts which have
catastrophic consequences, even if their probability of occur-

230. 747 F.2d at 1246 n.9 (citing Scientists' Inst. for Public Information, Inc. v. AEC,
481 F.2d 1079, 1092 (D.C. Cir. 1973)).

231. See id. at 1246. See generally Comments of the Western Natural Resources Law
Clinic to the CEQon the Proposed Amendment to the Worst Case Analysis Regulation at
2 (Sept. 27, 1985) (Professor Axline of the Western Natural Resources Law Clinic was an
Attorney for Save Our Ecosystems in the case of the same name).

232. 51 Fed. Reg. at 15,620 (emphasis added).
233. See 50 Fed. Reg. at 32,236 ("Council believes that pure conjecture... is not useful

to either the decision maker or the public") (emphasis in original).

1987]



92 COLUMBIA JOURNAL OF ENVIRONMENTAL LAW [Vol. 13:53

rence is low, provided that the analysis of the impacts is sup-
ported by credible scientific evidence, is not based on pure
conjecture, and is within the rule of reason. 234

The regulation, which was reworked numerous times before its
publication, 23 5 is confusing.236 After repeating five times that
only "reasonably foreseeable significant adverse effects" were to
be evaluated, 23 7 the regulation defines "reasonably foreseeable"
to include low-probability impacts. 23 8 Thus, the superseding reg-
ulation applies to the same situations that triggered the old
regulation. 239

The superseding regulation is also circular in its operation.
Under the regulation, an agency that cannot evaluate the poten-
tial environmental consequences of its action because of missing
scientific information must disclose that fact and state the rele-
vance of the missing information to its decision. The agency must
then summarize the "existing credible scientific evidence" 240 and
evaluate the adverse impacts using that evidence. A major por-
tion of the relevant information, however, is missing. If the
agency could evaluate the potential environmental effects of its
action with the aid of existing credible scientific evidence, it
would not have had to state that relevant information was miss-
ing, and the regulation would not have been triggered.

The CEQs rationale for adopting the superseding regulation is
an anticipated improvement in the quality of EISs and subsequent
agency decisionmaking. 24' According to the CEQ this improve-
ment will occur because the new regulation requires "more accu-
rate and relevant information about reasonably foreseeable

234. 40 C.F.R. § 1502.22 (1987). The new regulation applies to EISs prepared after
May 27, 1986. Id. § 1502.22(c).

235. In Spring 1985, nearly one year before the new regulation was published in final
form in the Federal Register, 28 drafts had been prepared. See discussion following Bear,
Worst Case Analysis: The Federal Regulation 40 C.F.R. 1502.22, in PROCEEDINGS OF A SYMPO-

SIUM ON WORST CASE ANALYSIS 3, 10 (1985) (statement of D. Bear, General Counsel,
CEQ).

236. Cf Whitney, Demise of the Council on Environmental Quality's "Worst Case"Analysis Regu-
lation, 8 GEO. MASON U.L. REV. 447, 478 (1986) (describing sentence defining "reasonably
foreseeable" in proposed regulation as "self-contradictory or incomprehensible"; defini-
tion is basically unchanged in final regulation).

237. See 40 C.F.R. § 1502.22 (1986).
238. Id. § 1502.22(b).
239. See 51 Fed. Reg. at 15,624 ("amended regulation will apply, of course, to the very

same situations to which the original regulation applies").
240. 40 C.F.R. § 1502.22(b) (1987).
241. 51 Fed. Reg. at 15,624.



Worst Case Analyses

significant adverse impacts."2 42 Because of the improved quality
of EISs, decisionmakers and the public will be better informed,2 43

and environmental protection will be strengthened.2 44 Also, the
gloss of the rule of reason will be added to the regulation. 245

However, the CEQs rationale is flawed.
The superseding regulation will not cause agencies to prepare

better EISs, thereby improving environmental quality. Under the
superseding regulation, a low probability/catastrophic conse-
quences impact must meet a three-prong test in order to be evalu-
ated by the agencies. The impact must be: (1) supported by
credible scientific evidence; (2) not based on pure conjecture; and
(3) within the rule of reason. At first glance this test bears a
strong similarity to the Fifth Circuit's test, enunciated in Sierra
Club v. Sigler, which required worst case analyses to be based on
existing data, and to be informative, useful and not unreasonably
speculative.2 46 But the new test attempts to remove the determi-
nations of "credible scientific evidence" from judicial review by
requiring an agency to consider only the scientific evidence that it
determines to be credible. The courts, however, have been the
traditional fora for determining credibility. In the NEPA context,
courts have reviewed the credibility of scientific opinions, not in
order to judge whether an agency has made the correct decision,
but to determine whether it has followed the procedures required
by law in disclosing those opinions.2 47 Apart from disagreeing
with the Ninth Circuit in Save Our Ecosystems and SOCATS, the
CEQoffered no evidence to support its contention that courts are
no longer capable of determining when scientific evidence is
credible. J

To summarize, the worst case analysis regulation was with-
drawn by the CEQ because of a change of opinion in the CEQ
regarding the wisdom of the regulation. The CEQ therefore, was

242. Id.
243. Id. at 15,620.
244. Id, at 15,624.
245. Id. at 15,622.
246. See 695 F.2d 957, 975 n.14 (5th Cir. 1983).
247. See Sierra Club v. Corps of Engineers, 772 F.2d 1043, 1053-54 (2d Cir. 1985); see

also Missouri ex rel. Ashcroft v. Corps of Engineers, 526 F. Supp. 660, 672 (W.D. Mo. 1980)
(rejecting hydrologist's opinion concerning increase of erosion caused by dam as not cred-
ible), aff'd, 672 F.2d 1297 (8th Cir. 1982); Hart & Miller Islands Area Envtl. Group, Inc. v.
Corps of Engineers, 505 F. Supp. 732, 755 (D. Md. 1980) (determining that scientific opin-
ions of environmental consequences of marine facility were credible but declining to refe-
ree conflict of scientific opinions).

19871



94 COLUMBIA JOURNAL OF ENVIRONMENTAL LAw [Vol. 13:53

required by the Supreme Court's decision in State Farm to provide

a reasonable explanation for changing its course of action as well

as for promulgating the superseding regulation. The CEQ has
failed to provide a reasonable explanation for either action.

The reasons the CEQ gave for withdrawing the worst case anal-
ysis regulation do not have an evidentiary basis. The superseding
regulation, meanwhile, will not improve the quality of EISs and
agency decisionmaking. The new regulation permits agencies to
shield their determinations of credibility from judicial review and
thus to consider only the scientific evidence that they themselves
consider to be credible. Such unsupervised determinations are
clearly open to abuse. In enabling agencies to ignore bodies of
evidence in cases where scientific evidence is necessarily scarce
and usually conflicting, the superseding regulation violates
NEPA's mandate of full disclosure.

VII. EFFECT OF THE CEQs WITHDRAWAL OF THE WORST CASE

ANALYSIS REGULATION

A. Agency Decisionmaking

The effect of the CEQs withdrawal of the worst case analysis
regulation may be negligible. The requirement for preparation of
a worst case analysis is gone but a similar analysis may be re-
quired by the courts in order to ensure that an agency has consid-
ered the full range of environmental concerns encompassed by
NEPA.

The CEO, by means of its superseding regulation, has at-
tempted to limit judicial review of agency determinations as to
which scientific evidence must be considered by agencies in order
to achieve compliance with NEPA. For example, in writing the
superseding regulation, the CEQ rejected a suggestion from the
EPA that the regulation or its preamble should state that agencies
must investigate "the available body of scientific evidence" and in-

clude the rationale used to select which evidence was credible.2 48

The superseding regulation contains no requirement that an
agency disclose the rationale behind its selection of credible
evidence.

248. Letter to Dinah Bear, General Counsel, CEQ from Allan Hirsch, Director, Office

of Federal Activities, Environmental Protection Agency, Sept. 23, 1985 (emphasis in
original).



Worst Case Analyses

The new regulation 249 seems to imply that an agency would be
able to reject, and thus to exclude from its decisionmaker's con-
sideration, the scientific opinions such as those the Ninth Circuit
ordered the BLM to consider in its worst case analysis of the her-
bicide spraying program in Save Our Ecosystems. 250 Despite the
contrary finding of the court in that case,25' the CEQ continued
to characterize the scientific opinions that the herbicides could
cause cancer at any dose as "pure conjecture.- 2 52 The CEQ's crit-
icism of this case seems to suggest that the definition of "credible
scientific evidence" may exclude "responsible opposing views,"
despite its subsequent assertion to the contrary. 253 Courts, how-
ever, have long held that responsible opposing views concerning
significant information must be disclosed in EISs,254 "even if the
responsible agency finds no merit in them whatsoever.- 255

If an agency fails to discuss an area of environmental concern
because it determines that no "credible scientific evidence" ex-
ists, it is exposing itself to a lawsuit challenging the omission.
Whereas a discussion of "responsible opposing views" resulting
in a determination that the opposing views are less than credible
is the type of agency decision that has traditionally been awarded
deference by the courts, a lack of any consideration of those views
exposes an agency to a judicial determination that the ignored
evidence was credible, and should have been discussed.

249. 40 C.F.R. § 15022.22 (1987).
250. See Save Our Ecosystems v. Watt, 13 E.L.R. 20,887, 20,888 (D. Or. 1983), aft'd in

part and rev 'd in part sub nom. Save Our Ecosystems v. Clark, 747 F.2d 1240 (9th Cir. 1984).
251. See id
252. Council on Environmental Quality, Proposed Amendment to 40 C.F.R. § 1502.22,

50 Fed. Reg. 32,234, 32,236 (1985).
253. In the comments to the new regulation, the CEQ included "responsible opposing

views" under its definition of "credible scientific evidence." See 51 Fed. Reg. at 15,623.
254. See Enos v. Marsh, 769 F.2d 1363, 1373 (9th Cir. 1985); see also Committee for

Nuclear Responsibility, Inc. v. Seaborg, 463 F.2d 783, 787 (D.C. Cir. 1971) (court's "func-
tion is only to ensure that the statement sets forth the opposing scientific views"); Silva v.
Lynn, 482 F.2d 1282, 1285 (1st Cir. 1973) (comments from responsible experts require
the response of a good faith reasoned analysis); Save the Niobrara River Ass'n v. Andrus,
483 F. Supp. 844, 866 (D. Neb. 1979) (opposing scientific views must be disclosed to-
gether with synopses of scientific views and methods of dealing with conditions asserted by
them).

255. Environmental Defense Fund v. Corps of Engineers, 325 F. Supp. 749, 759 (E.D.
Ark. 1971), vacated, 342 F. Supp. 1211 (E.D. Ark.), aft'd, 470 F.2d 289 (8th Cir. 1972).

1987]



96 COLUMBIA JOURNAL OF ENVIRONMENTAL LAw [Vol. 13:53

The CEQs statement that "[ilnformation which is unworthy of
belief should not be included in an EIS" 256 seems logical at first
glance. However, views that conflict with those held by an agency
or with an agency's goals may be more likely to be found unwor-
thy of belief.257 Courts, therefore, have required the considera-
tion of such views by agencies. 258 A decision taken in the absence
of adequate scientific data is more likely to appear to be a value
judgment than a rational and principled decision. In order to
avoid the appearance of bias, therefore, an agency should disclose
to the public as many scientific opinions regarding the action's
environmental effects as possible. Omission of conflicting views
could also limit NEPA's public participation process by restricting
debate on a project. 259 By encouraging agencies not to promote
full debate of the environmental consequences of their actions,
the CEQ frustrates the congressional purpose in enacting
NEPA.260 A full discussion of available facts rather than a con-
strained one would more effectively eliminate scientific
uncertainty.

B. Judicial Review

In reviewing a worst case analysis under the superseded regula-
tion, a court could determine that an agency acted arbitrarily by
failing to consider scientific evidence found to be credible by the
court. According to Ninth Circuit precedent, agency considera-
tion of the scientific evidence must take place whether or not the
agency considers it to be credible. 26' Although courts have tradi-
tionally deferred to agency determinations involving evidence on

256. 51 Fed. Reg. at 15,623. The CEQ derives its definition of "credible" from WEB-
STER'S DICTIONARY. Id. at 15,622-23 (quoting WEBSTER'S II NEW RIVERSIDE UNIVERSITY
DICTIONARY (1984)) (defining "credible" as "'capable of being believed' ").

257. Cf Rosenbaum, Amending CEQ's Worst Case Analysis Rule: Towards Better Decisionmak-
ing?, 15 Envtl. L. Rep. (Envtl. L. Inst.) 10,275, 10,277 n.26 (1985).

258. See Environmental Defense Fund v. Corps of Engineers, 325 F. Supp. 728, 759
(E.D. Ark. 1971) (EIS must contain opposing views), vacated, 342 F. Supp. 1211 (E.D.
Ark.), aft'd, 470 F.2d 289 (8th Cir. 1972).

259. Cf id.; Anderson, supra note 23, at 381 ("[e]ven if the agency disagrees violently
with a particular point of view, it should include it in the final statement").

260. See 42 U.S.C. § 433 1(a) (1982) (NEPA process involves "the Federal Government,
in cooperation with state and local governments, and other concerned public and private
organizations").

261. See Save Our Ecosystems v. Clark, 747 F.2d 1240, 1245-46 (9th Cir. 1984). See
generally Comment, Update: The NEPA Worst Case Analysis Regulation, 14 Envtl. L. Rep. (Envtl.
L. Inst.) 10,267, 10,271 (1984) (discussing "unusual nature of judicial review of a worst
case choice").



1987] Worst Case Analyses

the cutting edge of science, 262 the award of that deference de-
pends on the court's determination that the scientific evidence
has in fact been considered. 263 Upon a challenge to an agency's
omission, if the agency is shown to have ignored relevant scien-
tific evidence and has not explained the reasons for its decision,
courts themselves may undertake determinations of credibility.
The District of Columbia Circuit has stated that "only responsi-
ble opposing views need be included [in an EIS] and hence there
is room for discretion on the part of the officials preparing the
statement; but there is no room for an assumption that their de-
termination is conclusive. '

-
264

VIII. CONCLUSION

The worst case analysis regulation undoubtedly caused
problems for agency personnel in analyzing worst case scenarios
for their proposed projects. 265 Guidance from the CEQ on the

262. See, e.g., Baltimore Gas & Elec. Co. v. Natural Resources Defense Council, Inc., 462
U.S. 87, 103 (1983) ("[w]hen examining this kind of scientific determination, as opposed
to simple findings of fact, a reviewing court must generally be at its most deferential");
Ethyl Corp. v. EPA, 541 F.2d 1, 28 (D.C. Cir.) (en banc) ("rigorous step-by-step proof of
cause and effect" is not required when "evidence [is] difficult to come by, uncertain, or
conflicting because it is on the frontiers of scientific knowledge"), cert. denied, 426 U.S. 941
(1976); Reserve Mining Co. v. EPA, 514 F.2d 492, 519-20 (8th Cir. 1975) (deferring to
agency decision based on medical and scientific conclusions on "frontiers of scientific
knowledge"); modified, 529 F.2d 181 (8th Cir. 1976).

A less rigorous standard is, similarly, used for review of worst case analyses because the
data on which they are based are at the frontiers of science. See Sierra Club v. Sigler, 695
F.2d 957, 976 (5th Cir. 1983).

263. See Foundation on Economic Trends v. Heckler, 756 F.2d 143, 153-54 (D.C. Cir.
1985).

264. Committee for Nuclear Responsibility v. Schlesinger, 463 F.2d 783, 787 (D.C. Cir.
1971).

265. For example, the BLM prepared a worst case analysis for a proposed wildcat oil
well in mountainous terrain in the Shoshone National Forest in Wyoming. The proposed
well site was to occupy five acres accessible only by helicopter. Extensive discussions on
environmental protection for the site had taken place between all interested parties. Sci-
entific uncertainty did not exist because the environmental consequences of drilling the
well were known, as were the environmental effects of a potential blow out and concomi-
tant release of hydrogen sulfide gas. Nevertheless, the BLM prepared an EIS in which it
concluded that no significant impact on the environment existed. The agency then com-
pounded its error (an EIS is not required if the agency makes a finding of no significant
impact after it has prepared an environmental assessment, 40 C.F.R. § 1501.4(e) (1987)),
by labeling its worst case analysis as its first alternative. The worst case analyzed was a
fully developed oil field of 210 wells. The worst case analysis, which was not the environ-
mental effect of drilling the wildcat well, caused a huge public reaction. See Comments by
the Rocky Mountain Oil and Gas Association, Inc. to the CEQon the Advance Notice of
Proposed Rulemaking at 4-6 (Feb. 15, 1985).



98 COLUMBIA JOURNAL OF ENVIRONMENTAL LAw [Vol. 13:53

proper preparation of a worst case analysis could presumably
have resolved agency difficulties in determining which scenarios
to analyze, however. 266 The regulation had been in effect since
1978. By 1984, when the CEQ began considering amending the
worst case analysis regulation, the courts had their own frame-
work in place establishing when a worst case analysis should be
required. The framework required that worst case analyses be
based on existing data, that they be informative, useful and not
unreasonably speculative. 267 If scientific data were complete for
all significant impacts of an agency's action, a worst case analysis
was not required. 268

By withdrawing the regulation and imposing new requirements,
the CEQ has merely created a situation of uncertainty. Agencies
are bound to follow the superseding regulation but, at least in the
Fifth and Ninth Circuits, they are also bound to follow judicial
precedent requiring the preparation of worst case analyses in situ-
ations involving scientific uncertainty. In addition, NEPA's full
disclosure mandate arguably requires more information than
does the superseding regulation. Thus, compliance with the su-
perseding regulation may not satisfy the demands of either judi-
cial precedent or NEPA's statutory mandate.

266. The CEQ previously published guidance to aid agencies in implementing NEPA.
See, e.g., Council on Environmental Quality, Forty Most Asked Questions Concerning
CEQs National Environmental Policy Act Regulations, 46 Fed. Reg. 18,026 (1981). See
generally Comment, The Council on Environmental Quality's Guidelines and Their Influence on the
National Environmental Policy Act, 23 CATm. U.L. REV. 547, 559-70 (1974) (describing CEQ
memos).

267. Sierra Club v. Sigler, 695 F.2d 957, 974-75 (5th Cir. 1983).
268. Friends of Endangered Species, Inc. v. Jantzen, 589 F. Supp. 113, 116 n.I (N.D.

Cal. 1984), aft'd, 760 F.2d 976 (9th Cir. 1985); see also Village of False Pass v. Clark, 733
F.2d 605, 616 (9th Cir. 1984) (analysis of 100,000 barrel oil spill in St. George Basin not
required when less catastrophic worst case was analyzed and when subsequent offshore
leasing process required preparation of EIS).