¹ See, e.g., Federal Insecticide, Fungicide, and Rodenticide Act (FIFRA), 7 U.S.C. §§ 136- 136y (1988); Federal Water Pollution Control Act, 33 U.S.C. §§ 1251-1387 (1982); Clean Air Act, 42 U.S.C. §§ 7401-7642 (1983).

² “Projects” include construction, regulations, policies or recommendations undertaken by federal agencies, as well as actions of state or local governments and private businesses that involve federal funding (contracts, grants, loans) or require federal permits or licenses. 40 C.F.R. § 1508.18 (1989).

³ An Environmental Impact Statement (EIS) must describe the project's impacts, present alternatives and discuss adverse effects. See 40 C.F.R. §§ 1502.1-.25 (1989) infra note 7.

⁴ National Environmental Policy Act of 1969 (NEPA), 42 U.S.C. §§ 4321-70 (1982).

⁵ NEPA applies generically to impacts on the human environment, including effects on water, air, ecosystems, lower organisms, historical and cultural resources and economic factors. Id. at § 4331. The procedures established by NEPA are uniform across the wide range of potential effects. Impacts on human health are not treated differently under NEPA; this Note focuses on health impacts because the problems and promise of risk assessment methodologies arise in that context.

⁶ See Latin, , Good Science, Bad Regulation, and Toxic Risk Assessment, 5 YALE J. ON REG. 89, 91 (1988)Google Scholar.

⁷ 40 C.F.R. § 1503 (1989) (addresses commenting in NEPA regulations).

⁸ 42 U.S.C. § 4332 (1982) states:

⁹ Ruckelshaus, Risk, Science, and Democracy, 1 ISSUES SCI. & TECH. 19, 26 (1985). An evaluation of the various risk assessment methodologies is beyond the scope of this Note, which considers only the legal implications of their use in EISs. The literature on risk assessment is voluminous.

¹⁰ The Council on Environmental Quality (CEQ) was established by the NEPA legislation. See 42 U.S.C. §§4341-47 (1982) (describes responsibilities). CEQ issues regulations under NEPA. See 40 C.F.R. §§ 1500.1-.28 (1989).

¹¹ 40 C.F.R. § 1502.22 (1987).

¹² 42 U.S.C. §4331 (1982).

¹³ Id. at § 4332.

¹⁴ Id.. at §4332(2)(c).

¹⁵ See 40 C.F.R. § 1502.16 (1989); see also supra note 5.

¹⁶ CEQ. replaced advisory guidelines for EIS preparation with binding regulations in 1978. See infra note 30.

¹⁷ 40 C.F.R. § 1508.18 (1989).

¹⁸ See id. at § 1501.5 (addresses supervision of EIS preparation when more than one federal agency is involved).

¹⁹ Id. at § 1502.14.

²⁰ See, e.g., Scientists’ Inst, for Pub. Information, Inc. v. Atomic Energy Comm'n, 481 F.2d 1079, 1092 (D.C. Cir. 1973) (agencies must engage in reasonable forecasting of adverse impacts when data are uncertain; EIS for breeder reactor program required).

²¹ Id.

²² Alaska v. Andrus, 580 F.2d 465, 473 (D.C. Cir. 1978), vacated part II C, Western Oil & Gas Ass'n v. Alaska, 439 U.S. 922 (1978).

²³ Baltimore Gas & Elec. v. Natural Resources Defense Council, 462 U.S. 87, 97 (1983) (“Congress in enacting NEPA … did not require agencies to elevate environmental concerns over other appropriate considerations.“).

²⁴ Id.

²⁵ Id. See Administrative Procedure Act § 10(e), 5 U.S.C. § 706 (1988) (scope of review). For an analysis of agency differences in coping with NEPA, see S. TAYLOR, MAKING BUREAUCRACIES THINK 93-129 (1984).

²⁶ “NEPA merely prohibits uninformed — rather than unwise — agency action.” Robertson v. Methow Valley Citizens Council, 109 S. Ct. 1835, 1846 (1989).

²⁷ Murchison, , Does NEPA Matter? An Analysis of the Historical Development and Contemporary Significance of the National Environmental Policy Act, 18 U. RICH. L. REV. 557, 611 (1984)Google Scholar.

²⁸ Id. at 612.

²⁹ TAYLOR, supra note 25, at 321-22.

³⁰ 40 C.F.R. §§ 1500.1-.28 (1989). The regulations are unusual because they draw on both agency experience and case law. They apply to all federal agencies instead of governing only the promulgating agency. Fogelman, , Worst Case Analyses: A Continued Requirement Under the National Environmental Policy Act?, 13 COLUM. J. ENVTL. L. 53, 60 (1987)Google Scholar.

³¹ 40 C.F.R. § 1500 (1989).

³² Id. at § 1508.

³³ Id. at § 1508.8.

³⁴ Id. at § 1508.7.

³⁵ Id at § 1508.27.

³⁶ Id.

³⁸ Id. at § 1508.8.

³⁹ Id. at § 1508.27(b).

⁴⁰ The 1978 provision for incomplete or unavailable information stated:

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 uncertainty, the agency shall always make it clear that such information is lacking or that uncertainty exits.

1. (a)

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

2. (b)

   (b) If (1) the information relevant to adverse impacts is essential 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 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 include a worst case analysis and an indication of the probability or improbability of its occurrence.

40 C.F.R. § 1502.22 (1978).

⁴¹ Id. at § 1502.22(a).

⁴² The cost of obtaining information may be prohibitive because of the enormous expense of developing toxicological data. See infra note 131, at 308.

⁴³ 40 C.F.R. § 1502.22(a) (1978).

⁴⁴ When proposing the 1986 amendment, CEQ, stated that the reason for its choice of worst case analysis was not evident in the regulatory record. 50 Fed. Reg. 32,236 (1985) (proposed rule).

⁴⁵ 40 C.F.R. § 1502.22 (1981).

⁴⁶ Id.

⁴⁷ See, e.g., Scientists’ Inst, for Pub. Information, Inc. v. Atomic Energy Comm'n, 481 F.2d 1079, 1092 (D.C. Cir. 1973).

⁴⁸ See Save Our Ecosystems v. Clark, 747 F.2d 1240 (9th Cir. 1984) (Bureau of Land Management required to prepare worst case analysis prior to herbicide spraying; agencies may be required to conduct research); Southern Or. Citizens Against Toxic Sprays, Inc. v. Clark, 720 F.2d 1475 (9th Cir. 1983), cert, denied, 469 U.S. 1028 (1984) (worst case analyses required prior to herbicide spraying program despite agency's belief in its safety).

⁴⁹ For a discussion of continued judicial imposition of worst case analysis after rescission of the regulatory requirement, see infra notes 84-94 and accompanying text.

⁵⁰ Southern Or. Citizens, 720 F.2d at 1475.

⁵¹ Id. at 1479.

⁵³ Id. at 1480.

⁵⁴ Save Our Ecosystems v. Clark, 747 F.2d 1240, 1243-44 (9th Cir. 1984). See 40 C.F.R. § 1508.27(b)(5).

⁵⁵ Save Our Ecosystems, 747 F.2d at 1244 n.5.

⁵⁶ Id. at 1246.

⁵⁷ Id.

⁵⁸ Sierra Club v. Sigler, 695 F.2d 957 (5th Cir. 1983).

⁵⁹ See Save Our Ecosystems, 747 F.2d at 1248.

⁶⁰ 40 C.F.R. § 1502.22(b) (1978).

⁶¹ Save Our Ecosystems, 747 F.2d at 1248-49.

⁶² Note, Federal Agency Treatment of Uncertainty in Environmental Impact Statements Under the CEQJs Amended NEPA Regulation § 1502.22: Worst Case Analysis or Risk Threshold?, 86 MICH. L. REV. 777, 809 (1988).

⁶³ 40 C.F.R. § 1502.22 (1986).

⁶⁴ 40 C.F.R. § 1502.22 (1989). The current regulation for incomplete or unavailable information states in pertinent part:

When an agency is evaluating the reasonably foreseeable significant effects on the human environment in an environmental impact statement and there is incomplete or unavailable information, the agency shall always make clear that such information is lacking.

1. (a)

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

2. (b)

   (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 obtain it are not known, the agency shall include within the environmental impact statement:

3. (1)

   (1) a statement that such information is incomplete or unavailable;

4. (2)

   (2) a statement of the relevance of the incomplete or unavailable information to evaluating reasonably significant adverse impacts on the human environment;

5. (3)

   (3) a summary of existing credible scientific evidence which is relevant to evaluating reasonably foreseeable significant adverse impacts on the human environment; and

6. (4)

   (4) the agency's evaluation of such impacts based upon 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 occurrence is low, provided that the analysis of the impacts is supported by credible scientific evidence, is not based on pure conjecture, and is within the rule of reason.

⁶⁵ Id. at § 1502.22(a).

⁶⁶ Id. at § 1502.22.

⁶⁷ Id.

⁶⁸ Id.

⁶⁹ Id.

⁷⁰ Id.

⁷¹ 40 C.F.R. § 1502.22 (1985). See Sierra Club v. Sigler, 695 F.2d 957, 957 (5th Cir. 1983).

⁷² 40 C.F.R. § 1502.22 (1985).

⁷³ Id.

⁷⁴ Id.

⁷⁵ “The new requirement will provide more accurate and relevant information about reasonably foreseeable significant adverse impacts. To the extent that agencies were reluctant to discuss such impacts under the requirements of the original regulation, the amended regulation will not offer them an escape route.” 51 Fed. Reg. 15,624 (1986) (proposed rule). See Note, supra note 62, at 807-09.

⁷⁶ 40 C.F.R. § 1502.22 (1989).

⁷⁷ Id. 78 Id.

⁷⁹ Id.

⁸⁰ Id. CEQ supports the presentation of probabilities when reliable and relevant to the decision. Id.

⁸¹ Logically, a probabilistic inquiry is implicit in a determination of reasonable foreseeability. CEQ, may have declined to require inclusion of probabilities because of questions about their reliability and usefulness to the decisionmaker or public. Id. CEQ, may also have sought consistency with other uses of the term reasonably foreseeable, which has appeared frequently in NEPA law and regulations. Id.

⁸² Id.

⁸³ Id.

⁸⁴ See Methow Valley Citizens Council v. Regional Forester, 833 F.2d 810, 817 n.11 (9th Cir. 1987); Oregon Natural Resources Council v. Marsh, 832 F.2d 1489, 1497 n.8 (9th Cir. 1987).

⁸⁵ Id.

⁸⁶ For dicussions of the administrative law aspects of this issue and of the legality of the withdrawal by CEQ. of the worst case requirement, see Fogleman, supra note 30, at 70-94.

⁸⁷ 40 C.F.R. § 1502.22 (1986). See Note, supra note 62, at 807-09.

⁸⁸ CEQ. asserts that the Ninth Circuit cases are based on 40 C.F.R. § 1502.22 (1979) and are therefore not applicable to the amendment; CEQ's concerns about reasonableness and avoiding conjecture expressed throughout the regulation, however, would seem to belie this assertion. 51 Fed. Reg. 15,624-25 (1986).

⁸⁹ 40 C.F.R. § 1502.22 (1986). See Sierra Club v. Sigler, 695 F.2d 957, 972 (5th Cir. 1983).

⁹⁰ Robertson v. Methow Valley Citizens Council, 109 S. Ct. 1835, 1848 (1989). See Marsh v. Oregon National Resources Council, 109 S. Ct. 1851 (1989).

⁹¹ Robertson, 109 S. Ct. at 1848.

⁹² Id.

⁹³ Id. at 1849.

⁹⁴ Id.

⁹⁵ See 40 C.F.R. § 1502.22 (1989) (final sentence quoted supra note 64).

⁹⁶ See supra notes 77-83 and accompanying text.

⁹⁷ 40 C.F.R. § 1502.22 (1989).

⁹⁸ See Robertson, 109 S. Ct. at 1849; 40 C.F.R. § 1502.22(1989); Fed. Reg. 32,237 (1985).

⁹⁹ 40 C.F.R. § 1502.22 (1989).

¹⁰⁰ Fogleman, supra note 30, at 93-94.

¹⁰¹ Id. at 95. See Environmental Defense Fund v. Corps of Engineers, 325 F. Supp. 749, 759 (E.D. Ark. 1971), vacated, 342 F. Supp. 1211 (E.D. Ark. 1972), aff'd, 470 F.2d 289 (8th Cir. 1972).

¹⁰² Fogleman, supra note 30, at 79-80.

¹⁰³ NATIONAL RESEARCH COUNCIL, COMMITTEE ON THE INSTITUTIONAL MEANS FOR ASSESSMENT OF RISKS TO PUBLIC HEALTH, RISK ASSESSMENT IN THE FEDERAL GOVERNMENT: MANAGING THE PROCESS 3, 18 (1983).

¹⁰⁴ Many types of analyses serve as means to determine risks. Scientific risk assessment emphasizes quantitative factors; worst case analysis is a descriptive analysis.

¹⁰⁵ See, e.g., supra note 48.

¹⁰⁶ LAVE, L., QUANTITATIVE RISK ASSESSMENT IN REGULATION 12 (1982)Google Scholar.

¹⁰⁷ See, e.g., Leape, , Quantitative Risk Assessment in Regulation of Environmental Carcinogens, 4 HARV. ENVTL. L. REV. 86 (1980)Google Scholar.

¹⁰⁸ The policy judgment of whether a “safe” level of exposure is to be identified is especially important.

¹⁰⁹ L. LAVE, supra note 106, at 3. The characterization of uncertainty is a recognized weakness in scientific risk assessment. The source of uncertainty must be identified:

Uncertainty stemming from ignorance of the correct model or of crucial parameters must be distinguished from uncertainty due to ignorance of the outcome of a random process whose features are known…. When uncertainty is due to ignorance about structure or parameter values, a wide range of outcomes is possible. When uncertainty is due to not knowing the outcome of a random process that is understood, much more can be done to explore alternatives and to design a strategy to cope with ignorance. Even when elements of the structure are unknown, however, careful modeling can explore the range of outcomes and design strategies to minimize loss.

Id. at 53.

¹¹⁰ Commoner, , The Hazards of Risk Assessment, 14 COLUM. J. ENVTL. L. 365, 366 (1989).Google Scholar

¹¹¹ EPA Guidelines for Carcinogen Risk Assessment, 51 Fed. Reg. 33,993 (1986): Regulatory decision making involves two components: risk assessment and risk management. Risk assessment defines the adverse health consequences of exposure to toxic agents. The risk assessments will be carried out independently from considerations of the consequences of regulatory action. Risk management combines the risk assessment with the directives of regulatory legislation, together with socioeconomic, technical, political, and other considerations, to reach a decision as to whether or how much to control future exposure to the suspected toxic agents.

¹¹² Id.

¹¹³ Id.

¹¹⁴ Id. at 33,992 (“This case-by-case approach means that Agency experts review the scientific information on each agent and use the most scientifically appropriate interpretation to assess risk.“).

¹¹⁵ See EPA Guidelines for Carcinogen Risk Assessment, id. at 33,992; EPA Guidelines for Mutagenicity Risk Assessment, id. at 34,006; EPA Guidelines for the Health Risk Assessment of Chemical Mixtures, id. at 34,014; EPA Guidelines for the Health Assessment of Suspect Developmental Toxicants, id. at 34,028; EPA Guidelines for Estimating Exposures, id. at 34,042.

As an advocate and pioneer in risk assessment, the EPA developed and published these Guidelines with contributions from the larger scientific community. They are designed to ensure a consistent and reliable standard for use within an agency and to inform other agencies and the public about the procedures. Id. at 33,992. As conceived, they are designed to accomodate new methods and scientific knowledge. Id. Identification of the strengths and weaknesses of assessments because of uncertainties and assumptions is required. Id. at 33,993. The EPA explicitly recognized the need for new methodologies to characterize uncertainty, and assured revision as appropriate. For a discussion of the EPA's role in risk assessment, see Daggett, Hazen & Shaw, , Advancing Environmental Protection Through Risk Assessment, 14 COLUM. J. ENVTL. L. 315 (1989).Google Scholar

¹¹⁶ EPA Guidelines for Carcinogen Risk Assessment, 51 Fed. Reg. 33,994 (1986).

¹¹⁷ For a discussion of the components of risk assessment, see Paustenbach, , Health Risk Assessments: Opportunities and Pitfalls, 14 COLUM. J. ENVTL. L. 379, 386–409 (1989).Google Scholar

¹¹⁸ EPA Guidelines for Carcinogen Risk Assessment, 51 Fed. Reg. 33,993 (1986). Exposure assessment is especially important for environmental pollutants because the wide variability in exposure and the fate of chemicals in the environment must be considered. See Paustenbach, supra note 117, at 402-04.

¹¹⁹ EPA Guidelines for Carcinogen Risk Assessment, 51 Fed. Reg. 33,994 (1986).

¹²⁰ Id. at 33,998.

¹²¹ Latin, supra note 6, at 90.

¹²² Id. at 91.

¹²³ Id. at 93.

¹²⁴ Id. at 135.

¹²⁵ Id.

¹²⁶ For a discussion of the view that conservatism should be increased because of factors relating to linear dose-response, environmental dispersion and synergism, see Finkel, , Is Risk Assessment Really Too Conservative?: Revising the Revisionists, 14 COLUM. J. ENVTL. L. 427, 442-47 (1989)Google Scholar.

¹²⁷ Anderson, , Scientific Developments in Risk Assessment: Legal Implications, 14 COLUM. J. ENVTL. L. 411, 412-13 (1989).Google Scholar

¹²⁸ Risk assessment's predecessor in the 1970s was probabalistic risk analysis, a statistical quantification of risks that was also prone to bias and inaccuracies. See Stever, , The Use of Risk Assessment in Environmental Law, 14 COLUM. J. ENVTL. L. 329, 330 (1989)Google Scholar.

¹²⁹ Goldstein, , Risk Assessment and the Interface Between Science and Law, 14 COLUM. J. ENVTL. L. 343, 353 (1989)Google Scholar.

¹³⁰ Id. at 352.

¹³¹ Lave, , Risk Assessment and Regulatory Priorities, 14 COLUM. J. ENVTL. L. 307, 313 (1989)Google Scholar.

¹³² Finkel, supra note 126, at 439. But see generally Paustenbach, supra note 117.

¹³³ Finkel, supra note 126, at 454-55.

¹³⁴ Id. at 455.

¹³⁵ Id. at 457.

¹³⁶ Id.

¹³⁷ Latin, supra, note 6, at 139.

¹³⁸ id, at 140.

¹³⁹ Id. at 139-40.

¹⁴⁰ Id. at 141.

¹⁴¹ Id. at 142.

¹⁴² Id. at 448.

¹⁴³ Id. at 449-50.

¹⁴⁴ Id. One commentator noted that the assumption of equal susceptibility underlying risk assessment “flies in the face of a growing body of knowledge about several very common genetic, environmental, and lifestyle factors (e.g., nutritional status, stress, concurrent diseases) that can cause interindividual variations in susceptibility of several fold.” Id. at 450 (quoting FINKEL, RISK ANALYSIS 5 (1988)).

¹⁴⁵ Id. at 436.

¹⁴⁶ Lave, supra 131, at 308.

¹⁴⁷ Paustenbach, supra note 117, at 385.

¹⁴⁸ Id.

¹⁴⁹ See Lyndon, , Risk Assessment, Risk Communication and Legitimacy: An Introduction to the Symposium, 14 COLUM.J. ENVTL. L. 289, 303 (1989)Google Scholar.

¹⁵⁰ Id. at 304.

¹⁵¹ Finkel, supra note 126, at 435.

¹⁵² 40 C.F.R. § 1502.22 (1989).

¹⁵³ See supra notes 77-83 and accompanying text.

¹⁵⁴ See DeFranco, California's Toxics Initiative, 39 HASTINGS

¹⁵⁵ Risk assessment also seems appropriate in light of the NEPA definition of “significantly“: the intensity or severity of the impact is to be evaluated in part on the degree to which the impacts are highly uncertain or entail unknown or unique risks. 40 C.F.R. § 1508.27 (1989). This implies a special obligation to investigate as thoroughly as possibile; currently, risk assessment is state of the art for such circumstances.

¹⁵⁶ See 40 C.F.R. § 1502.22 (1989).

¹⁵⁷ See id. at § 1508.27.

¹⁵⁸ Cross, Beyond Benzene: Establishing Principles for a Significance Threshold on Regulatable Risks of Cancer, 35 EMORY L.J. 1, 10–11 (1986)Google Scholar.

¹⁵⁹ Industrial Union Dep't, AFL-CIO v. American Petroleum Inst., 448 U.S. 607 (1980) (plurality opinion).

¹⁶⁰ See generally Cross, supra note 158.

¹⁶¹ 40 C.F.R. § 1502.22(b)(3) (1989).

¹⁶² See Latin, supra note 6, at 124.

¹⁶³ Leape, supra note 107, at 112-13.

¹⁶⁴ Latin, supra note 6, at 145.

¹⁶⁵ Id. at 144.

¹⁶⁶ Id. at 143.

¹⁶⁷ Leape, supra note 107, at 87.

¹⁶⁸ TAYLOR, supra note 25, at 320.

¹⁶⁹ Id.

¹⁷⁰ B. FISCHHOFF, S. LICHTENSTEIN, P. SLOVIC, S. DERBY & R. KEENEY, ACCEPTABLE RISK 27 (1981) [hereinafter FISCHHOFF].

¹⁷¹ Stenzel, , A Proposal for a National Risk Assessment Clearinghouse, 14 COLUM. J. ENVTL. L. 549, 569-71 (1989)Google Scholar. The author proposes adoption of a uniform format for risk assessments and advocates that a centralized source be established for their distribution.

¹⁷² Stenzel, , The Need for a National Risk Asessment Communication Policy, 11 HARV. ENVTL. L. REV. 381, 382-83 (1987)Google Scholar. The author proposes Congressional legislation to provide and fund a public education mechanism to promote consistent characterization and comparison of risks by all federal agencies. For example, distinguishing between chronic, catastrophic and involuntary hazards avoids comparisons between risks that are perceived differently by the public. Id. at 399.

¹⁷³ Id. at 385.

¹⁷⁴ Id. at 387.

¹⁷⁵ Id. at 388. Experts can aggravate these uncertainties by omitting information and denying lingering doubts about the extent of problems. A decision can be made with the knowledge that underlying data may be inaccurate as long as this possibility is recognized. Problems arise when only the experts can detect errors. Evidence shows that both experts and laypersons are overconfident about their own judgments: “A major culprit seems to be the insensitivity to the tenuousness of the assumptions upon which beliefs are based …. The problem lies not in getting the wrong answer, but in failing to realize how great the possibility for error was.” FISCHHOFF, supra note 170, at 35.

¹⁷⁶ Id. at 144-45.

¹⁷⁷ Id.

¹⁷⁸ Fiorino, , Environmental Risk and Democratic Process: A Critical Review, 14 COLUM. J. ENVTL. L. 501, 529-30 (1989)Google Scholar.

¹⁷⁹ Id. at 528.

¹⁸⁰ See Lyndon, supra note 149, at 299.

¹⁸¹ Id. at 300.

¹⁸² See supra notes 114-20 and accompanying text.

¹⁸³ 40 C.F.R. § 1502.22 (1989).

¹⁸⁴ See generally Stenzel, supra notes 171, 172.

¹⁸⁵ Stenzel, supra note 172, at 389.

¹⁸⁵ Stenzel, supra note 172, at 389.