Baltimore Gas & Electric Co. v. Natural Resources Defense Council, Inc.

*89 Justice O’Connor delivered the opinion of the Court. Section 102(2)(C) of the National Environmental Policy Act of 1969, 83 Stat. 853, 42 U. S. C. §4332(2)(C) (NEPA), requires federal agencies to consider the environmental impact of any major federal action. 1 As part of its generic rule-making proceedings to evaluate the environmental effects of the nuclear fuel cycle for nuclear powerplants, the Nuclear *90 Regulatory Commission (Commission) 2 decided that licensing boards should assume, for purposes of NEPA, that the permanent storage of certain nuclear wastes would have no significant environmental impact and thus should not affect the decision whether to license a particular nuclear power-plant. We conclude that the Commission complied with NEPA and that its decision is not arbitrary or capricious within the meaning of § 10(e) of the Administrative Procedure Act (APA), 5 U. S. C. §706. 3 The environmental impact of operating a light-water nuclear powerplant 4 includes the effects of offsite activities necessary to provide fuel for the plant (“front end” activities), and of offsite activities necessary to dispose of the highly toxic and long-lived nuclear wastes generated by the plant (“back end” activities). The dispute in these cases con *91 cerns the Commission’s adoption of a series of generic rules to evaluate the environmental effects of a nuclear power-plant’s fuel cycle. At the heart of each rule is Table S-3, a numerical compilation of the estimated resources used and effluents released by fuel cycle activities supporting a year’s operation of a typical light-water reactor. 5 The three versions of Table S-3 contained similar numerical values, although the supporting documentation has been amplified during the course of the proceedings. The Commission first adopted Table S-3 in 1974, after extensive informal rulemaking proceedings. 39 Fed. Reg. 14188 et seq. (1974). This “original” rule, as it later came to be described, declared that in environmental reports and impact statements for individual licensing proceedings the environmental costs of the fuel cycle “shall be as set forth” in Table S-3 and that “[n]o further discussion of such environmental effects shall be required.” Id., at 14191. 6 The original Table S-3 contained no numerical entry for the long-term *92 environmental effects of storing solidified transuranic and high-level wastes, 7 because the Commission staff believed that technology would be developed to isolate the wastes from the environment. The Commission and the parties have later termed this assumption of complete repository integrity as the “zero-release” assumption: the reasonableness of this assumption is at the core of the present controversy. The Natural Resources Defense Council (NRDC), a respondent in the present cases, challenged the original rule and a license issued under the rule to the Vermont Yankee Nuclear Power Corp. The Court of Appeals for the District of Columbia Circuit affirmed Table S-3’s treatment of the “front end” of the fuel cycle, but vacated and remanded the portion of the rule relating to the “back end” because of perceived inadequacies in the rulemaking procedures. Natural Resources Defense Council, Inc. v. NRC, 178 U. S. App. D. C. 336, 547 F. 2d 633 (1976). Judge Tamm disagreed that the procedures were inadequate, but concurred on the ground that the record on waste storage was inadequate to support the zero-release assumption. Id., at 361, 547 F. 2d, at 658. In Vermont Yankee Nuclear Power Corp. v. Natural Resources Defense Council, Inc., 435 U. S. 519 (1978), this Court unanimously reversed the Court of Appeals’ decision that the Commission had used inadequate procedures, finding that the Commission had done all that was required by NEPA and the APA and determining that courts generally lack the authority to impose “hybrid” procedures greater than those contemplated by the governing statutes. We remanded for review of whether the original rule was adequately supported by the administrative record, specifically *93 stating that the court was free to agree or disagree with Judge Tamm’s conclusion that the rule pertaining to the “back end” of the fuel cycle was arbitrary and capricious within the meaning of § 10(e) of the APA, 5 U. S. C. § 706. Id., at 536, n. 14. While Vermont Yankee was pending in this Court, the Commission proposed a new “interim” rulemaking proceeding to determine whether to adopt a revised Table S-3. The proposal explicitly acknowledged that the risks from long-term repository failure were uncertain, but suggested that research should resolve most of those uncertainties in the near future. 41 Fed. Reg. 45850-45851 (1976). After further proceedings, the Commission promulgated the interim rule in March 1977. Table S-3 now explicitly stated that solidified high-level and transuranic wastes would remain buried in a federal repository and therefore would have no effect on the environment. 42 Fed. Reg. 13807 (1977). Like its predecessor, the interim rule stated that “[n]o further discussion of such environmental effects shall be required.” Id., at 13806. The NRDC petitioned for review of the interim rule, challenging the zero-release assumption and faulting the Table S-3 rule for failing to consider the health, cumulative, and socioeconomic effects of the fuel cycle activities. The Court of Appeals stayed proceedings while awaiting this Court’s decision in Vermont Yankee. In April 1978, the Commission amended the interim rule to clarify that health effects were not covered by Table S-3 and could be litigated in individual licensing proceedings. 43 Fed. Reg. 15613 et seq. (1978). In 1979, following further hearings, the Commission adopted the “final” Table S-3 rule. 44 Fed. Reg. 45362 et seq. (1979). Like the amended interim rule, the final rule expressly stated that Table S-3 should be supplemented in individual proceedings by evidence about the health, socioeconomic, and cumulative aspects of fuel cycle activities. The Commission also continued to adhere to the zero-release *94 assumption that the solidified waste would not escape and harm the environment once the repository was sealed. It acknowledged that this assumption was uncertain because of the remote possibility that water might enter the repository, dissolve the radioactive materials, and transport them to the biosphere. Nevertheless, the Commission predicted that a bedded-salt repository would maintain its integrity, and found the evidence “tentative but favorable” that an appropriate site would be found. Id., at 45368. The Commission ultimately determined that any undue optimism in the assumption of appropriate selection and perfect performance of the repository is offset by the cautious assumption, reflected in other parts of the Table, that all radioactive gases in the spent fuel would escape during the initial 6- to 20-year period that the repository remained open, ibid., and thus did not significantly reduce the overall conservatism of Table S-3. Id., at 45369. The Commission rejected the option of expressing the uncertainties in Table S-3 or permitting licensing boards, in performing the NEPA analysis for individual nuclear plants, to consider those uncertainties. It saw no advantage in reassessing the significance of the uncertainties in individual licensing proceedings: “In view of the uncertainties noted regarding waste disposal, the question then arises whether these uncertainties can or should be reflected explicitly in the fuel cycle rule. The Commission has concluded that the rule should not be so modified. On the individual reactor licensing level, where the proceedings deal with fuel cycle issues only peripherally, the Commission sees no advantage in having licensing boards repeatedly weigh for themselves the effect of uncertainties on the selection of fuel cycle impacts for use in cost-benefit balancing. This is a generic question properly dealt with in the rule-making as part of choosing what impact values should go into the fuel cycle rule. The Commission concludes, hav *95 ing noted that uncertainties exist, that for the limited purpose of the fuel cycle rule it is reasonable to base impacts on the assumption which the Commission believes the probabilities favor, i. e., that bedded-salt repository sites can be found which will provide effective isolation of radioactive waste from the biosphere. ” Id., at 45369. The NRDC and respondent State of New York petitioned for review of the final rule. The Court of Appeals consolidated these petitions for all purposes with the pending challenges to the initial and interim rules. 8 By a divided panel, 9 the court concluded that the Table S-3 rules were arbitrary and capricious and inconsistent with NEPA because the Commission had not factored the consideration of uncertainties surrounding the zero-release assumption into the licensing process in such a manner that the uncertainties could potentially affect the outcome of any decision to license a particular plant. Natural Resources Defense Council, Inc. v. NRC, 222 U. S. App. D. C. 9, 685 F. 2d 459 (1982). The court first reasoned that NEPA requires an agency to consider all significant environmental risks from its proposed action. If the zero-release assumption is taken as a, finding that long-term storage poses no significant environmental *96 risk, which the court acknowledged may not have been the Commission’s intent, it found that the assumption represents a self-evident error in judgment and is thus arbitrary and capricious. As the evidence in the record reveals and the Commission itself acknowledged, the zero-release assumption is surrounded with uncertainty. Alternatively, reasoned the Court of Appeals, the zero-release assumption could be characterized as a decision-making device whereby the Commission, rather than individual licensing boards, would have sole responsibility for considering the risk that long-lived wastes will not be disposed of with complete success. The court recognized that the Commission could use generic rulemaking to evaluate environmental costs common to all licensing decisions. Indeed, the Commission could use generic rulemaking to balance generic costs and benefits to produce a generic “net value.” These generic evaluations could then be considered together with case-specific costs and benefits in individual proceedings. The key requirement of NEPA, however, is that the agency consider and disclose the actual environmental effects in a manner that will ensure that the overall process, including both the generic rulemaking and the individual proceedings, brings those effects to bear on decisions to take particular actions that significantly affect the environment. The Court of Appeals concluded that the zero-release assumption was not in accordance with this NEPA requirement because the assumption prevented the uncertainties — which were not found to be insignificant or outweighed by other generic benefits — from affecting any individual licensing decision. Alternatively, by requiring that the licensing decision ignore factors that are relevant under NEPA, the zero-release assumption is a clear error in judgment and thus arbitrary and capricious. We granted certiorari. 459 U. S. 1034 (1982). We reverse. *97 h-4 I — I We are acutely aware that the extent to which this Nation should rely on nuclear power as a source of energy is an important and sensitive issue. Much of the debate focuses on whether development of nuclear generation facilities should proceed in the face of uncertainties about their long-term effects on the environment. Resolution of these fundamental policy questions lies, however, with Congress and the agencies to which Congress has delegated authority, as well as with state legislatures and, ultimately, the populace as a whole. Congress has assigned the courts only the limited, albeit important, task of reviewing agency action to determine whether the agency conformed with controlling statutes. As we emphasized in our earlier encounter with these very proceedings, “[administrative decisions should be set aside in this context, as in every other, only for substantial procedural or substantive reasons as mandated by statute . . . , not simply because the court is unhappy with the result reached.” Vermont Yankee, 435 U. S., at 558. The controlling statute at issue here is NEPA. NEPA has twin aims. First, it “places upon an agency the obligation to consider every significant aspect of the environmental impact of a proposed action.” Vermont Yankee, supra, at 553. Second, it ensures that the agency will inform the public that it has indeed considered environmental concerns in its decisionmaking process. Weinberger v. Catholic Action of Hawaii/Peace Education Project, 454 U. S. 139, 143 (1981). Congress in enacting NEPA, however, did not require agencies to elevate environmental concerns over other appropriate considerations. See Stryckers’ Bay Neighborhood Council v. Karlen, 444 U. S. 223, 227 (1980) (per curiam). Rather, it required only that the agency take a “hard look” at the environmental consequences before taking a major action. See Kleppe v. Sierra Club, 427 U. S. 390, 410, n. 21 (1976). The role of the courts is simply to ensure that the *98 agency has adequately considered and disclosed the environmental impact of its actions and that its decision is not arbitrary or capricious. See generally Citizens to Preserve Overton Park, Inc. v. Volpe, 401 U. S. 402, 415-417 (1971). In its Table S-3 rule here, the Commission has determined that the probabilities favor the zero-release assumption, because the Nation is likely to develop methods to store the wastes with no leakage to the environment. The NRDC did not challenge and the Court of Appeals did not decide the reasonableness of this determination, 222 U. S. App. D. C., at 28, n. 96, 685 F. 2d, at 478, n. 96, and no party seriously challenges it here. The Commission recognized, however, that the geological, chemical, physical, and other data it relied on in making this prediction were based, in part, on assumptions which involve substantial uncertainties. Again, no one suggests that the uncertainties are trivial or the potential effects insignificant if time proves the zero-release assumption to have been seriously wrong. After confronting the issue, though, the Commission has determined that the uncertainties concerning the development of nuclear waste storage facilities are not sufficient to affect the outcome of any individual licensing decision. 10 It is clear that the Commission, in making this determination, has made the careful consideration and disclosure required by NEPA. The sheer volume of proceedings before the Commission is impressive. 11 Of far greater importance, *99 the Commission’s Statement of Consideration announcing the final Table S-3 rule shows that it has digested this mass of material and disclosed all substantial risks. 44 Fed. Reg. 45367-45369 (1979). The Statement summarizes the major uncertainty of long-term storage in bedded-salt repositories, which is that water could infiltrate the repository as a result of such diverse factors as geologic faulting, a meteor strike, or accidental or deliberate intrusion by man. The Commission noted that the probability of intrusion was small, and that the plasticity of salt would tend to heal some types of intrusions. The Commission also found the evidence “tentative but favorable” that an appropriate site could be found. Table S-3 refers interested persons to staff studies that discuss the uncertainties in greater detail. 12 Given this record *100 and the Commission’s statement, it simply cannot be said that the Commission ignored or failed to disclose the uncertainties surrounding its zero-release assumption. Congress did not enact NEPA, of course, so that an agency would contemplate the environmental impact of an action as an abstract exercise. Rather, Congress intended that the “hard look” be incorporated as part of the agency’s process of deciding whether to pursue a particular federal action. It was on this ground that the Court of Appeals faulted the Commission’s action, for failing to allow the uncertainties potentially to “tip the balance” in a particular licensing decision. As a general proposition, we can agree with the Court of Appeals’ determination that an agency must allow all significant environmental risks to be factored into the decision whether to undertake a proposed action. We think, however, that the Court of Appeals erred in concluding that the Commission had not complied with this standard. As Vermont Yankee made clear, NEPA does not require agencies to adopt any particular internal decisionmaking structure. Here, the agency has chosen to evaluate generi *101 cally the environmental impact of the fuel cycle and inform individual licensing boards, through the Table S-3 rule, of its evaluation. The generic method chosen by the agency is clearly an appropriate method of conducting the “hard look” required by NEPA. See Vermont Yankee, 435 U. S., at 535, n. 13. The environmental effects of much of the fuel cycle are not plant specific, for any plant, regardless of its particular attributes, will create additional wastes that must be stored in a common long-term repository. Administrative efficiency and consistency of decision are both furthered by a generic determination of these effects without needless repetition of the litigation in individual proceedings, which are subject to review by the Commission in any event. See generally Ecology Action v. AEC, 492 F. 2d 998, 1002, n. 5 (CA2 1974) (Friendly, J.) (quoting Administrative Conference Proposed Recommendation 73-6). The Court of Appeals recognized that the Commission has discretion to evaluate generically the environmental effects of the fuel cycle and require that these values be “plugged into” individual licensing decisions. The court concluded that the Commission nevertheless violated NEPA by failing to factor the uncertainty surrounding long-term storage into Table S-3 and precluding individual licensing decisionmakers from considering it. The Commission’s decision to affix a zero value to the environmental impact of long-term storage would violate NEPA, however, only if the Commission acted arbitrarily and capriciously in deciding generically that the uncertainty was insufficient to affect any individual licensing decision. In assessing whether the Commission’s decision is arbitrary and capricious, it is crucial to place the zero-release assumption in context. Three factors are particularly important. First is the Commission’s repeated emphasis that the zero-release assumption — and, indeed, all of the Table S-3 rule — was made for a limited purpose. The Commission expressly noted its intention to supplement the rule with an explanatory narra *102 tive. 13 It also emphasized that the purpose of the rule was not to evaluate or select the most effective long-term waste disposal technology or develop site selection criteria. A separate and comprehensive series of programs has been undertaken to serve these broader purposes. 14 In the proceedings before us, the Commission’s staff did not attempt to evaluate the environmental effects of all possible methods of disposing of waste. Rather, it chose to analyze intensively the most probable long-term waste disposal method — burial in a bedded-salt repository several hundred meters below ground — and then “estimate its impacts conservatively, based on the best available information and analysis.” 44 Fed. Reg. 45863 (1979). 15 The zero-release assumption cannot be evaluated in isolation. Rather, it must be assessed in relation to the limited purpose for which the Commission made the assumption. Second, the Commission emphasized that the zero-release assumption is but a single figure in an entire Table, which the *103 Commission expressly designed as a risk-averse estimate of the environmental impact of the fuel cycle. It noted that Table S-3 assumed that the fuel storage canisters and the fuel rod cladding would be corroded before a repository is closed and that all volatile materials in the fuel would escape to the environment. 16 Given that assumption, and the improbability that materials would escape after sealing, the Commission determined that the overall Table represented a conservative (i. e., inflated) statement of environmental impacts. It is not unreasonable for the Commission to counteract the uncertainties in postsealing releases by balancing them with an overestimate of presealing releases. 17 A reviewing court should not magnify a single line item beyond its significance as only part of a larger Table. Third, a reviewing court must remember that the Commission is making predictions, within its area of special expertise, at the frontiers of science. When examining this kind of scientific determination, as opposed to simple findings of fact, a reviewing court must generally be at its most deferential. See, e. g., Industrial Union Dept. v. American Petroleum Institute, 448 U. S. 607, 656 (1980) (plurality opinion); id., at 705-706 (Marshall, J., dissenting). *104 With these three guides in mind, we find the Commission’s zero-release assumption to be within the bounds of reasoned decisionmaking required by the APA. We have already noted that the Commission’s Statement of Consideration detailed several areas of uncertainty and discussed why they were insubstantial for purposes of an individual licensing decision. The Table S-3 rule also refers to the staff reports, public documents that contain a more expanded discussion of the uncertainties involved in concluding that long-term storage will have no environmental effects. These staff reports recognize that rigorous verification of long-term risks for waste repositories is not possible, but suggest that data and extrapolation of past experience allow the Commission to identify events that could produce repository failure, estimate the probability of those events, and calculate the resulting consequences. NUREG-0116, at 4-86. 18 The Commission staff also modeled the consequences of repository failure by tracing the flow of contaminated water, and found them to be insignificant. Id., at 4-89 through 4-94. Ultimately, the staff concluded that “[t]he radiotoxic hazard index analyses and the modeling studies that have been done indicate that consequences of all but the most improbable events will be small. *105 Risks (probabilities times consequences) inherent in the long term for geological disposal will therefore also be small.” Id., at 2-11. We also find significant the separate views of Commissioners Bradford and Gilinsky. These Commissioners expressed dissatisfaction with the zero-release assumption and yet emphasized the limited purpose of the assumption and the overall conservatism of Table S-3. Commissioner Bradford characterized the bedded-salt repository as a responsible working assumption for NEPA purposes and concurred in the zero-release figure because it does not appear to affect Table S-3’s overall conservatism. 44 Fed. Reg. 45372 (1979). Commissioner Gilinsky was more critical of the entire Table, stating that the Commission should confront directly whether it should license any nuclear reactors in light of the problems of waste disposal, rather than hide an affirmative conclusion to this issue behind a table of numbers. He emphasized that the “waste confidence proceeding,” see n. 14, supra, should provide the Commission an appropriate vehicle for a thorough evaluation of the problems involved in the Government’s commitment to a waste disposal solution. For the limited purpose of individual licensing proceedings, however, Commissioner Gilinsky found it “virtually inconceivable” that the Table should affect the decision whether to license, and characterized as “naive” the notion that the fuel cycle effluents could tip the balance in some cases and not in others. 44 Fed. Reg. 45374 (1979). In sum, we think that the zero-release assumption — a policy judgment concerning one line in a conservative Table designed for the limited purpose of individual licensing decisions — is within the bounds of reasoned decisionmaking. It is not our task to determine what decision we, as Commissioners, would have reached. Our only task is to determine whether the Commission has considered the relevant factors and articulated a rational connection between the facts found and the choice made. Bowman Transportation, Inc. v. *106 Arkansas-Best Freight System, Inc., 419 U. S. 281, 285-286 (1974); Citizens to Preserve Overton Park, Inc. v. Volpe, 401 U. S. 402 (1971). Under this standard, we think the Commission's zero-release assumption, within the context of Table S-3 as a whole, was not arbitrary and capricious. r-4 1 — 4 HH As we have noted, n. 5, supra, Table S-3 describes effluents and other impacts in technical terms. The Table does not convert that description into tangible effects on human health or other environmental variables. The original and interim rules declared that “the contribution of the environmental effects of. . . fuel cycle activities . . . shall be as set forth in the following Table S-3 [and] [n]o further discussion of such environmental effects shall be required.” 39 Fed. Reg. 14191 (1974); 42 Fed. Reg. 13806 (1977). Since the Table does not specifically mention health effects, socioeconomic impacts, or cumulative impacts, this declaration does not clearly require or preclude their discussion. The Commission later amended the interim rule to clarify that health effects were not covered by Table S-3 and could be litigated in individual licensing proceedings. In the final rule, the Commission expressly required licensing boards to consider the socioeconomic and cumulative effects in addition to the health effects of the releases projected in the Table. 44 Fed. Reg. 45371 (1979). 19 The Court of Appeals held that the original and interim rules violated NEPA by precluding licensing boards from considering the health, socioeconomic, and cumulative effects of the environmental impacts stated in technical terms. As does the Commission, we agree with the Court of Appeals that NEPA requires an EIS to disclose the significant health, socioeconomic, and cumulative consequences of the environ *107 mental impact of a proposed action. See Metropolitan Edison Co. v. People Against Nuclear Energy, 460 U. S. 766 (1983); Kleppe v. Sierra Club, 427 U. S., at 410; 40 CFR §§ 1508.7, 1508.8 (1982). We find no basis, however, for the Court of Appeals’ conclusion that the Commission ever precluded a licensing board from considering these effects. It is true, as the Commission pointed out in explaining why it modified the language in the earlier rules, that the original Table S-3 rule “at least initially was apparently interpreted as cutting off” discussion of the effects of effluent releases. 44 Fed. Reg. 45364 (1979). But even the notice accompanying the earlier versions stated that the Table was “to be used as a basis for evaluating the environmental effects in a cost-benefit analysis for a reactor,” 39 Fed. Reg. 14190 (1974) (emphasis added), suggesting that individual licensing boards were to assess the consequences of effluent releases. And when, operating under the initial rule, the Atomic Safety and Licensing Appeal Board suggested the desirability of discussing health effects for comparing nuclear with coal plants, In re Tennessee Valley Authority (Hartsville Nuclear Plant Units), 5 N. R. C. 92, 103, n. 52 (1977), the Commission staff was allowed to introduce evidence of public health consequences. Cf. In re Public Service Company of Indiana (Marble Hill Nuclear Generating Station), 7 N. R. C. 179, 187 (1978). Respondents have pointed to no case where evidence concerning health or other consequences of the data in Table S-3 was excluded from licensing proceedings. We think our admonition in Vermont Yankee applies with equal force here: “[WJhile it is true that NEPA places upon an agency the obligation to consider every significant aspect of the environmental impact of a proposed action, it is still incumbent upon intervenors who wish to participate to structure their participation so that it is meaningful, so that it alerts the agency to the intervenors’ position and contentions.” 435 U. S., at 553. *108 In short, we find it totally inappropriate to cast doubt on licensing proceedings simply because of a minor ambiguity in the language of the earlier rule under which the environmental impact statement was made, when there is no evidence that this ambiguity prevented any party from making as full a presentation as desired, or ever affected the decision to license the plant. IV For the foregoing reasons, the judgment of the Court of Appeals for the District of Columbia Circuit is Reversed. Justice Powell took no part in the consideration or decision of these cases. *109 APPENDIX TO THE OPINION OF THE COURT *110 Footnotes [1] Section 102(2)(C) provides: “The Congress authorizes and directs that, to the fullest extent possible ... (2) all agencies of the Federal Government shall— “(c) include in every recommendation or report on proposals for legislation and other major Federal actions significantly affecting the quality of the human environment, a detailed statement by the responsible official on— “(i) the environmental impact of the proposed action, “(ii) any adverse environmental effects which cannot be avoided should the proposal be implemented, [and] “(v) any irreversible and irretrievable commitments of resources which would be involved in the proposed action should it be implemented.” [2] The original Table S-3 rule was promulgated by the Atomic Energy Commission (AEC). Congress abolished the AEC in the Energy Reorganization Act of 1974,42 U. S. C. § 5801 et seq., and transferred its licensing and regulatory functions to the Nuclear Regulatory Commission (NRC). The interim and final rules were promulgated by the NRC. This opinion will use the term “Commission” to refer to both the NRC and the predecessor AEC. [3] Title 5 U. S. C. § 706 states in part: “The reviewing court shall— “(2) hold unlawful and set aside agency action, findings, and conclusions found to be— “(A) arbitrary, capricious, an abuse of discretion, or otherwise not in accordance with law.” [4] A light-water nuclear powerplant is one that uses ordinary water (H20), as opposed to heavy water (D20), to remove the heat generated in the nuclear core. See Van Nostrand’s Scientific Encyclopedia 1998, 2008 (D. Considine & G. Considine eds., 6th ed. 1983). The bulk of the reactors in the United States are light-water nuclear reactors. NRC Ann. Rep., Appendix 6 (1980). [5] For example, the tabulated impacts include the acres of land committed to fuel cycle activities, the amount of water discharged by such activities, fossil fuel consumption, and chemical and radiological effluents (measured in curies), all normalized to the annual fuel requirement for a model 1,000 megawatt light-water reactor. See Table S-3, reprinted in the Appendix, infra. [6] Under the Atomic Energy Act of 1954, 68 Stat. 919, as amended, 42 U. S. C. § 2011 et seq., a utility seeking to construct and operate a nuclear powerplant must obtain a separate permit or license at both the construction and the operation stage of the project. After the Commission’s staff has examined the application for a construction license, which includes a review of possible environmental effects as required by NEPA, a three-member Atomic Safety and Licensing Board conducts a public adjudicatory hearing and reaches a decision which can be appealed to the Atomic Safety and Licensing Appeal Board and, in the Commission’s discretion, to the Commission itself. The final agency decision may be appealed to the courts of appeals. A similar procedure occurs when the utility applies for an operating license, except that a hearing need be held only in contested cases. See Vermont Yankee Nuclear Power Corp. v. Natural Resources Defense Council, Inc., 435 U. S. 519, 526-527 (1978). [7] High-level wastes, which are highly radioactive, are produced in liquid form when spent fuel is reprocessed. Transuranic wastes, which are also highly toxic, are nuclides heavier than uranium that are produced in the reactor fuel. See Natural Resources Defense Council, Inc. v. NRC, 222 U. S. App. D. C. 9, 16, n. 11, 685 F. 2d, 459, 466, n. 11 (1982). [8] In Vermont Yankee, we indicated that the Court of Appeals could consider any additions made to the record by the Commission, and could consolidate review of the initial review with review of later rules. 435 U. S., at 537, n. 14. Consistent with this direction, the parties stipulated that all three versions of the rule could be reviewed on the basis of the whole record. See 222 U. S. App. D. C., at 21, n. 39, 685 F. 2d, at 471, n. 39. [9] Judge Bazelon wrote the opinion for the court. Judge Wilkey joined the section of the opinion that rejected New York’s argument that the waste-disposal technology assumed for calculation of certain effluent release values was economically infeasible. That issue is not before us. Judge Wilkey filed a dissenting opinion on the issues that are under review here. Judge Edwards of the Court of Appeals for the Sixth Circuit, sitting by designation, joined these sections of Judge Bazelon’s opinion, and also filed a separate opinion concurring in part and dissenting on the economic infeasibility issue. [10] As the Court of Appeals recognized, 222 U. S. App. D. C., at 31, n. 118, 685 F. 2d, at 481, n. 118, the Commission became increasingly candid in acknowledging the uncertainties underlying permanent waste disposal. Because all three versions of Table S-3 use the same zero-release assumption, and the parties stipulated that the entire record be used in reviewing all three versions, see n. 8, supra, we need review only the propriety of the final Table S-3 rule. We leave for another day any general concern with an agency whose initial Environmental Impact Statement (EIS) is insufficient but who later adequately supplements its consideration and disclosure of the environmental impact of its action. [11] The record includes more than 1,100 pages of prepared direct testimony, two rounds of questions by participants and several hundred pages *99 of responses, 1,200 pages of oral hearings, participants’ rebuttal testimony, concluding statements, the 137-page report of the hearing board, further written statements from participants, and oral argument before the Commission. The Commission staff has prepared three studies of the environmental effects of the fuel cycle: Environmental Survey of the Uranium Fuel Cycle, WASH-1248 (Apr. 1974); Environmental Survey of the Reprocessing and Waste Management Portions of the LWR Fuel Cycle, NUREG-0116 (Supp. 1 to WASH-1248) (Oct. 1976) (hereinafter cited as NUREG-0116); and Public Comments and Task Force Responses Regarding the Environmental Survey of the Reprocessing and Waste Management Portions of the LWR Fuel Cycle, NUREG-0216 (Supp. 2 to WASH-1248) (Mar. 1977). [12] We are reviewing here only the Table S-3 rulemaking proceedings, and do not have before us an individual EIS that incorporates Table S-3. It is clear that the Statement of Consideration supporting the Table S-3 rule adequately discloses the environmental uncertainties considered by the Commission. However, Table S-3 itself refers to other documents but gives only brief descriptions of the environmental effects it encapsulates. There is some concern with an EIS that relies too heavily on separate documents rather than addressing the concerns directly. Although we do not decide whether they have binding effect on an independent agency such as the Commission, it is worth noting that the guidelines from the Council on Environmental Quality in effect during these proceedings required that “care should be taken to ensure that the statement remains an essentially self-contained instrument, capable of being understood by the reader without the need for undue cross reference.” 38 Fed. Reg. 20554 (1973), 40 *100 CFR § 1500.8(b) (1974). The present regulations state that incorporation by reference is permissible if it will not “imped[e] agency and public review of the action. The incorporated material shall be cited in the statement and its content briefly described.” 40 CFR § 1502.21 (1982). The Court of Appeals noted that NEPA “requires an agency to do more than to scatter its evaluation of environmental damage among various public documents,” 222 U. S. App. D. C., at 34, 685 F. 2d, at 484, but declined to find that the incorporation of other documents by reference would invalidate an EIS that used Table S-3 to describe the environmental impact of the fuel cycle. The parties here do not treat this insufficient disclosure argument as a separate argument and, like the Court of Appeals, we decline to strike down the rule on this ground. We do not deny the value of an EIS that can be understood without extensive cross-reference. The staff documents referred to in Table S-3 are public documents, however, and we note that the Commission has proposed an explanatory narrative to accompany Table S-3, which would be included in an individual EIS, that may alleviate some of the concerns of incorporation. See n. 13, infra. [13] In March 1981, the Commission submitted a version of the explanatory narrative for public comment as a proposed amendment to the final fuel cycle rule. 46 Fed. Reg. 15154 (1981). The Commission has not yet adopted a final narrative. [14] In response to Minnesota v. NRC, 195 U. S. App. D. C. 234, 602 F. 2d 412 (1979), the Commission has initiated a "waste confidence” proceeding to consider the most recent evidence regarding the likelihood that nuclear waste can be safely disposed of and when that, or some other offsite storage solution, can be accomplished. 44 Fed. Reg. 61372 et seq. (1979). See id., at 45363. The recently enacted Nuclear Waste Policy Act of 1982, Pub. L. 97-425, 96 Stat. 2201, 42 U. S. C. § 10101 et seq. (1982 ed.), has set up a schedule for identifying site locations and a funding mechanism for development of permanent waste repositories. The Environmental Protection Agency has also proposed standards for future waste repositories, 47 Fed. Reg. 58196 et seq. (1982). [15] For example, Table S-3 assumes that plutonium will not be recycled. The Commission noted that, in response to a Presidential directive, it had terminated separate proceedings concerning the possibility of recyling plutonium in mixed oxide fuel. 44 Fed. Reg. 45369, n. 28 (1979). See In re Mixed Oxide Fuel, 6 N. R. C. 861 (1977); In re Mixed Oxide Fuel, 7 N. R. C. 711 (1978). [16] The Commission also increased the overall conservatism of the Table by overestimating the amount of fuel consumed by a reactor, underestimating the amount of electricity produced, and then underestimating the efficiency of filters and other protective devices. See Conclusions and Recommendations of the Hearing Board Regarding the Environmental Effects of the Uranium Fuel Cycle, Docket No. Rm 50-3, App. to Pet. for Cert, in No. 82-524, pp. 282a-293a. Additionally, Table S-3, which analyzes both a uranium-recycle and no-recycle system, conservatively lists, for each effluent, the highest of the two releases that would be expected under each cycle. 41 Fed. Reg. 45849, 45850 (1976). [17] The Court of Appeals recognized that the Commission could weigh certain generic costs and benefits of reactors against each other to produce a generic “net value” to be used in individual licensing proceedings. 222 U. S. App. D. C., at 32, 685 F. 2d, at 482. We see no reason why the Commission does not have equal discretion to evaluate certain environmental costs together to produce a generic net cost. [18] For example, using this approach the staff estimated that a meteor the size necessary to damage a repository would hit a given square kilometer of the earth’s surface only once every 50 trillion years, and that geologic faulting through the Delaware Basin in southeast New Mexico (assuming that were the site of the repository) would occur once in 25 billion years. NUREG-0116, at 4-87. The staff determined that a surface burst of a 50 megaton nuclear weapon, far larger than any currently deployed, would not breach the repository. Ibid. The staff also recognized the possibility that heat generated by the waste would damage the repository, but suggested this problem could be alleviated by decreasing the density of the stored waste. In recognition that this suggestion would increase the size of the repository, the Commission amended Table S-3 to reflect the greater acreage required under these assumptions. See 44 Fed. Reg. 45369 (1979). [19] Of course, just as the Commission has discretion to evaluate generically aspects of the environmental impact of the fuel cycle, it has discretion to have other aspects of the issue decided in individual licensing decisions.