WILLIAM P. CLARK, SECRETARY OF THE INTERIOR, ET AL., PETITIONERS V. SOUTHERN OREGON CITIZENS AGAINST TOXIC SPRAYS, INC. No. 84-267 In the Supreme Court of the United States OCTOBER TERM, 1984 On Petition for a Writ of Certiorari to the United States Court of Appeals for the Ninth Circuit REPLY BRIEF FOR THE PETITIONERS In our petition we have explained that the court of appeals' decision requiring the Bureau of Land Management (BLM) independently to reassess the safety of herbicides employed in its land management activities that are approved for usage under the Federal Insecticide, Fungicide, and Rodenticide Act (FIFRA) effects an unprecedented and illogical expansion of federal agencies' obligations under the National Environmental Policy Act of 1969 (NEPA). We highlighted two major flaws in the analysis that impelled the court of appeals to its erroneous conclusion that Council on Environmental Quality Regulation implementing NEPA (40 C.F.R. 1502.22) require preparation of a "worst case" analysis here. First, notwithstanding the consistent teaching of the cases -- (the very cases interpreting NEPA that these regulations sought to codify) -- the court of appeals read out of the worst case analysis regulation the threshold test that governs the obligation to prepare a worst case analysis. The court simply dispensed with the requirement that it be demonstrated that the effects to be addressed in such an analysis are sufficiently probable to make their consideration "essential to a reasoned choice among alternative()" courses of action (40 C.F.R. 1502.22(b)(1)). As was acknowledged by the court of appeals in Save Our Ecosystems v. Clark, No. 83-3908 (9th Cir. Jan. 27, 1984), petition for rehearing pending, the effect of the court of appeals' instant decision is accordingly to require that worst case analyses be prepared as to "all possible long-range effects" (slip op. 24 n.9) -- regardless of whether they are of any practical importance -- unless the agency can prove that such remote events are enitrely impossible. Second, the court of appeals' decision disrupts the carefully crafted and complex framework devised by Congress in FIFRA for assessment of the health, safety and environmental effects of herbicides. There is no reason to believe that Congress intended to superimpose NEPA's general, unstructured, mandate for sensitivity to environmental consequences of federal agency action upon the detailed standards and procedures prescribed under FIFRA for addressing precisely the same concerns by requiring EPA to prepare environmental impact statements assessing the environmental effects of its own FIFRA registration decisions. Yet the court of appeals blighely insisted that Congress intended that the BLM -- an agency that proposes simply to use a registered herbicide in conformity with the limiting conditions of its registration -- undertake an environmental analysis under NEPA reconsidering the very environmental and safety issues that have been considered by EPA. The arguments of respondent and amici curiae Merrell et al. do not negate the importance of the matters presented for review. 1. The worst case analysis requirement a. Strikingly, neither respondent nor the amici endorses the court of appeals' interpretation of the worst case analysis requirement, under which federal agencies are obliged to prepare such analyses without regard to the substantiality of any element of uncertainty as to the effects of a federal action (see Pet. App. 6a). Indeed, countering our suggestion (Pet. 9-10) that the worst case analysis regulation, as applied by the court of appeals, is contrary to settled interpretations of NEPA, respondent places the regulation "squarely in the mainstream of NEPA" (Br. in Opp. 7; citations omitted; emphasis added): (Respondent) freely admits that NEPA does not require discussion of all potential impacts, no matter how remote or frivolous. However, as written * * *, neither does 40 C.F.R. Section 1502.22 require such exercises in imagination. The regulation is self-limiting in that respect. Only scientific uncertainty which is "relevant" requires disclosure. 40 C.F.R. Section 1502.22 Only if obtainable information is "essential to a reasoned choice," 40 C.F.R. Sections 1502.22(a) and 1502.22(b)(1), or is "important to the decision," must the agency publish a worst case analysis. /1/ But this, of course, is precisely our point. Both the plain language of 40 C.F.R. 1502.22 and the decisions of the courts generally interpreting NEPA apply a rule of reason that qualifies the statutory obligation to consider the environmental impacts of federal action. See Pet. 9-10. The court of appeals below, however, required a worst case analysis here -- even though it recognized "a lack of information about the probability of any adverse effect" (Pet. App. 5a). The court reasoned that, because science may not presently enable us to rule out all possibility that herbicide spraying has an effect on human health (ibid.), a worst case analysis was mandated. The court of appeals thus elided two separate elements of the "trigger" for application of the worst case analysis requirement -- i.e., that "information relevant to adverse impacts is * * * not known" and that such information is "essential to a reasoned choice among alternatives" (40 C.F.R. 1502.22(b)(1)). b. The appropriateness of further review here is confirmed by respondent's discussion (Br. in Opp. 3-6) of Sierra Club v. Sigler, 695 F.2d 957 (5th Cir. 1983), which, we have explained (Pet. 11), conflicts with the decision of the court below. The court of appeals denied that its approach conflicted with that of the Fifth Circuit, asserting (Pet. App. 5a-6a) that Sigler rejects the threshold probability standard we advocate in this case. The court of appeals erred in that respect. To be sure, in Sigler, the court of appeals rejected the Corps of Engineers' contention that the risk of a total loss of cargo by an oil supertanker was too remote to trigger the worst case analysis requirement. But the court emphasized (695 F.2d at 975 n.14): We scarcely need add that while remoteness of a possible occurrence does not permit disregarding it in circumstances as these, where a real possibility of the occurrence has been proved and a data base for evaluating its consequences established, the Corps need not concern itself with phantasmagoria hypothesized without a firm basis in evidence and actual circumstances of the contempted project, or with disasters the likelihood of which is not shown to be significantly increased by the carrying out of the project. Respondents do not defend the court of appeals' misreading of Sigler on this point. Instead they concede (Br. in Opp. 5) that the Sigler plaintiffs were property required to "'prove' something" -- i.e. that the environmental impacts that they were required to address in a worst case analysis were not "frivolous possibilities." How then do respondents seek to deny the existence of a conflict between the decision below and Sigler? Respondents resourcefully answer that, like the Sigler plaintiffs, they were actually required to make a threshold showing that an assessment of the effects to be addressed in a worst case analysis was essential to an informed decision by the responsible federal agency (Br. in Opp. 5): In Sigler, the plaintiffs proved that oil spills were to be expected. In this case, the plaintiff proved that the public would be exposed to herbicides. But this comparison is fundamentally misleading. The adverse effect that respondent seeks to have addressed in worst case analysis is not simply the possibility of human exposure to sprayed herbicides, rather it is the alleged adverse health effects of spraying on the persons so exposed. Showing that human exposure is a reasonably likely possibility is not the equivalent of establishing that the alleged human health effects are something more than "frivolous possibilities." By contrast, in Sigler, the plaintiffs had established not only that a major oil spill was a "real possibility" (695 F.2d at 975 n.14), but also that substantial adverse environmental effects were reasonably likely to be associated with such a spill (695 F.2d at 974). Thus respondent's efforts to harmonize this case with Sigler is no more successful than the court of appeals' attempt. /2/ c. Adopting a different tack than respondent, the amici opposing the petition argue (Merrell Br. 13-14) that there is credible evidence that establishes a material possibility that "adverse health effects will follow from herbicide exposure" (id. at 13; emphasis in original). Contrary to the amici's contention (id. at 24-25) that BLM was aware of this evidence but declined to consider it, BLM specifically noted the existence of the few studies cited by respondent in its annual environmental assessment of the 1982 spraying program and, after evaluation, concluded that they lacked scientific credibility. See Bureau of Land Management, U.S. Dep't of Interior, 1982 Final Vegetative Management Program: Supplemental Environmental Assessment Attch. E, a copy of which was previously lodged with the Clerk of the Court. Thus, rather than ignoring the existence of the scientific controversy surrounding the health impacts of the herbicides, the BLM disclosed the existence of the controversy, and, after weighing all the "evidence," including the fact that the EPA did not find the scientific uncertainty significant enough to warrant suspension or cancellation of the herbicides' registration under FIFRA, concluded that a probability of adverse human health effects sufficient to make resolution of scientific uncertainty essential to an informed decision had not been demonstrated. NEPA does not require more. d. Respondent's argument, in the final analysis, appears to be (Br. in Opp. 8) that a worst case analysis should be prepared here because to do so would not be burdensome or expensive, the agency being required only set forth a litany of theoretically possible effects. The goal of the CEQ regulations, however, like that of NEPA generally, is not to reward fertile imaginations with a recitation of impacts the impossibility of which has not been proven, but to assure that decisionmakers are provided with information essential to a reasoned choice among the alternative course of action open to them. Because time and resources are necessarily limited, diversion of agency attention to matters not material to the decision to be made is not simply an innocent -- albeit unauthorized -- extension of the requirements of NEPA. It is fundamentally inconsistent with Congress's plan for fostering rationality in administrative decisionmaking through NEPA. 2. The import of FIFRA registration Respondent's stark declaration (Br. in Opp. 8) that "registration of herbicides under FIFRA is irrelevant" betrays a fundamental misunderstanding of the purpose and requirements of both NEPA and FIFRA, as well as a misdirected critique of the way in which FIFRA has been implemented. Respondent (Br. in Opp. 9) and the amici (Merrell Br. 23-24) argue that FIFRA registration must be discounted because registration is not a guarantee of complete absence of adverse effects. But NEPA's "essentially procedural" mandate (Strycker's Bay Neighborhood Council, Inc. v. Karlen, 444 U.S. 223, 227 (1980)) does not require any particular level of environmental protection. By contrast, FIFRA, which establishes substantive herbicide registration criteria, including the requirement that a herbicide not engender "unreasonable adverse effects on the environment" (7 U.S.C. 136a(c)(5)(C) and (D)), provides a far greater degree of assurance against untoward human health effects, and provides the appropriate forum for concerned persons to raise any questions they may have regarding the environmental safety of herbicides. Respondent (Br. in Opp. 10) and the amici (Merrell Br. 21-23) criticize various aspects of EPA's administration of FIFRA; indeed, respondent labels the registration process "pro forma" (Br. in Opp. 11). But the court of appeals did not rest its decision in this case upon such concerns, but upon a blanket rule that FIFRA registration is wholly irrelevant to the NEPA obligations of a herbicide-using agency (see Pet. App. 7a-8a). /3/ In any event, proceedings under FIFRA itself -- rather than NEPA -- provide the appropriate vehicle for making any challenge to EPA's actions in administering the herbicide registration scheme. Given the threshold requirement -- acknowledged by respondent (see page 3, supra) -- that material uncertainty on points of sufficient importance to affect a federal agency's decision be shown, failure to consider that a herbicide has been registered leads to results that are at best peculiar. The comprehensive and detailed regulatory scheme established by Congress in FIFRA -- including the provisions for conditional registration -- reflects Congress's deliberate judgment as to the standards and procedures required to achieve an appropriate assurance of herbicide safety, and the level of uncertainty or incomplete knowledge that is acceptable. Congress could not have intended that the BLM, the National Forest Service or any other agency using herbicides employ the NEPA process to second-guess or oversee EPA's administration of FIFRA. The court of appeals' decision in this case, and its offspring -- Save Our Ecosystems -- require just such an irrational approach to herbicide regulation. That decision accordingly should not stand unreviewed. For the foregoing reasons and the reasons stated in the petition, the petition for a writ of certiorari should be granted. Respectfully submitted, REX E. LEE Solicitor General NOVEMBER 1984 /1/ Similarly, the amici acknowledge (Merrell Br. 12) that, under the worst case analysis regulation, no worst case analysis need be prepared unless "uncertainties (are) serious enough to make consideration of those uncertainties essential to a reasoned decision." /2/ The court of appeals recognized that this case is not "closely analogous" to Sigler in the respect suggested by respondent (Pet. App. 5a): In Sigler, the "worst case" was an event of low probability but catastrophic effects and the scientific uncertainty concerned those effects. In contrast, this case involves a lack of information about the probability of any adverse effect. /3/ Ironically, it was only in Save Our Ecosystems, slip op. 9, 23-25 n.9 -- in determining the scope of the independent research on herbicide health effects that was held to be required of the agency -- that the court of appeals relied on the conditional registration status of a particular herbicide.