MARSH v. OREGON NATURAL RESOURCES COUNCIL

JUSTICE STEVENS delivered the opinion of the Court.

This case is a companion to Robertson v. Methow Valley Citizens Council, ante, p. 332. It arises out of a controversial decision to construct a dam at Elk Creek in the Rogue River Basin in southwest Oregon. In addition to the question whether an Environmental Impact Statement (EIS) prepared pursuant to the National Environmental Policy Act of 1969 (NEPA), 83 Stat. 852, 42 U. S. C. § 4321 et seq., must contain a complete mitigation plan and a "worst case analysis," which we answered in Robertson, it presents the question whether information developed after the completion of the EIS requires that a supplemental EIS be prepared before construction of the dam may continue.

I

In the 1930's in response to recurring floods in the Rogue River Basin, federal and state agencies began planning a major project to control the water supply in the Basin.1 See, e. g., ch. 346, 49 Stat. 439. In 1961 a multiagency study recommended the construction of three large dams: the Lost Creek Dam on the Rogue River, the Applegate Dam on the Applegate River, and the Elk Creek Dam on the Elk Creek near its confluence with the Rogue River. See H. R. Doc. No. 566, 87th Cong., 2d Sess., 7-89 (1962). The following year, Congress authorized the Army Corps of Engineers (Corps) to construct the project in accordance with the recommendations of the 1961 study. See Flood Control Act of 1962, Pub. L. 87-874, § 203, 76 Stat. 1192-1193. The Lost Creek Dam was completed in 1977, and the Applegate Dam was completed in 1981.

Plans for the Elk Creek Dam describe a 238-foot-high concrete structure that will control the run-off from 132 square miles of the 135-square-mile Elk Creek watershed. When full, the artificial lake behind the dam will cover 1,290 acres of land, will have an 18-mile shoreline, and will hold 101,000 acre-feet of water. The dam will cost approximately $100 million to construct and will produce annual benefits of almost $5 million. It will be operated in coordination with the nearby Lost Creek Dam, where the control center for both dams will be located. Its "multiport" structure, which will permit discharge of water from any of five levels, makes it possible to regulate, within limits, the temperature, turbidity,2 and volume of the downstream flow. Although primarily designed to control flooding along the Rogue River, additional project goals include enhanced fishing, irrigation, and recreation.

In 1971, the Corps completed its EIS for the Elk Creek portion of the three-dam project and began development by acquiring 26,000 acres of land and relocating residents, a county road, and utilities. Acknowledging incomplete information, the EIS recommended that further studies concerning the project's likely effect on turbidity be developed. The results of these studies were discussed in a draft supplemental EIS completed in 1975. However, at the request of the Governor of Oregon, further work on the project was suspended, and the supplemental EIS was not filed to make it possible to analyze the actual consequences of the construction of the Lost Creek Dam, which was nearing completion, before continuing with the Elk Creek project. Following that analysis and the receipt of a statement from the Governor that he was "extremely interested in pursuing construction of the Elk Creek Dam,"3 the Corps completed and released its final Environmental Impact Statement, Supplement No. 1 (FEISS), in December 1980.

Because the Rogue River is one of the Nation's premier fishing grounds, the FEISS paid special heed to the effects the dam might have on water quality, fish production, and angling. In its chapter on the environmental effects of the proposed project, the FEISS explained that water quality studies were prepared in 1974 and in 1979 and that "[w]ater temperature and turbidity have received the most attention." FEISS 33. Using computer simulation models, the 1974 study predicted that the Elk Creek Dam might, at times, increase the temperature of the Rogue River by one to two degrees Fahrenheit and its turbidity by one to three JTU's.4 Ibid. The 1979 study took a second look at the potential effect of the Elk Creek Dam on turbidity and, by comparing the 1974 study's predictions concerning the effects of the Lost Creek Dam with actual measurements taken after that dam became operational, it "increased technical confidence in the mathematical model predictions . . . and reinforced the conclusions of the 1974 [study]." Id., at 33-34. Based on these studies, the FEISS predicted that changes in the "turbidity regime" would not have any major effect on fish production,5 but that the combined effect of the Lost Creek and Elk Creek Dams on the turbidity of the Rogue River might, on occasion, impair fishing.6

Other adverse effects described by the FEISS include the displacement of wildlife population — including 100 black-tailed deer and 17 elk — and the loss of forest land and vegetation resulting from the inundation of 1,290 acres of land with the creation of the artificial lake. Id., at 26, 38, 46. Most significantly, it is perfectly clear that the dam itself would interfere with the migration and spawning of a large number of anadromous fish,7 but this effect has been mitigated by the construction of a new hatchery.8 Id., at 35. Finally, the FEISS found that no endangered or threatened species would be affected by the project. Id., at 27.

On February 19, 1982, after reviewing the FEISS, the Corps' Division Engineer made a formal decision to proceed with construction of the Elk Creek Dam, "subject to the approval of funds by the United States Congress." App. to Pet. for Cert. 53a. In his decision, he identified the mitigation measures that had already been taken with respect to the loss of anadromous fish spawning habitat, as well as those that would "most likely" be taken to compensate for the loss of other wildlife habitat. Id., at 56a-57a. He concluded that the benefits that would be realized from the project "out-weigh the economic and environmental costs" and that completion would serve "the overall public interest." Id., at 58a. In August 1985, Congress appropriated the necessary funds.9 Act of Aug. 15, 1985, Pub. L. 99-88, 99 Stat. 314. The dam is now about one-third completed and the creek has been rechanneled through the dam.

II

In October 1985, four Oregon nonprofit corporations10 filed this action in the United States District Court for the District of Oregon seeking to enjoin construction of the Elk Creek Dam. Their principal claims were that the Corps violated NEPA by failing (1) to consider the cumulative effects of the three dams on the Rogue River Basin in a single EIS; (2) adequately to describe the environmental consequences of the project; (3) to include a "worst case analysis" of uncertain effects; and (4) to prepare a second supplemental EIS to review information developed after 1980.

After conducting a hearing on respondents' motion for a preliminary injunction, the District Judge denied relief on each of the NEPA claims.11 628 F.Supp. 1557 (1986). He first held that courts must employ a standard of "reasonableness" in reviewing an agency's compliance with NEPA. Under this standard of review, the court must " `make a pragmatic judgment whether the EIS's form, content and preparation foster both informed decision-making and informed public participation.' " Id., at 1562 (quoting California v. Block, 690 F.2d 753, 761 (CA9 1982)). Applying this standard, the District Judge concluded that the Corps had, in fact, taken a sufficiently "hard look" at the cumulative effects of the three dams and at the individual effects of the Elk Creek Dam. 628 F. Supp., at 1563-1565. He also concluded that a "worst case analysis" was not required because the Corps used state-of-the-art mathematical models, thus avoiding scientific uncertainty and the need to fill gaps in information with a worst case scenario. Id., at 1567. Finally, the District Court held that the Corps' decision not to prepare a second supplemental EIS to address new information was "reasonable."

The new information relied upon by respondents is found in two documents. The first, an internal memorandum prepared by two Oregon Department of Fish and Wildlife (ODFW) biologists based upon a draft ODFW study, suggested that the dam will adversely affect downstream fishing, and the second, a soil survey prepared by the United States Soil Conservation Service (SCS), contained information that might be taken to indicate greater downstream turbidity than did the FEISS. As to both documents, the District Judge concluded that the Corps acted reasonably in relying on the opinions of independent and Corps experts discounting the significance of the new information. Id., at 1567-1568. At the conclusion of his opinion, the District Judge directed that the motion for preliminary relief be consolidated with trial on the merits pursuant to Federal Rule of Civil Procedure 65(a)(2), and thus denied respondents' claim for a permanent injunction as well.

The Court of Appeals reversed. 832 F.2d 1489 (CA9 1987). Applying the same "reasonableness" standard of review employed by the District Court, the Court of Appeals reached a contrary conclusion, holding that the Corps had not adequately evaluated the cumulative environmental impact of the entire project. Id., at 1497. Since the Corps did not seek review of that holding, we do not discuss it. The court also held that the FEISS was defective because it did not include a complete mitigation plan and because it did not contain a "worst case analysis." Id., at 1493-1494, 1496-1497. These holdings were erroneous for the reasons stated in our opinion in Robertson v. Methow Valley Citizens Council, ante, p. 332, and will not be further discussed. With regard to the failure to prepare a second supplemental EIS, the Court of Appeals concluded that the ODFW and SCS documents brought to light "significant new information" concerning turbidity, water temperature, and epizootic12 fish disease; that this information, although "not conclusive," is "probably accurate;" and that the Corps' experts failed to evaluate the new information with sufficient care. 832 F. 2d, at 1494-1496. The court thus concluded that a second supplemental EIS should have been prepared. Judge Wallace, writing in dissent, took issue with the majority's analysis of the new information. In his view, it was reasonable for the Corps to have concluded, based on its own expert evaluation, that the information contained in the ODFW document was inaccurate, and that the information contained in the SCS document was insignificant. Id., at 1500 (opinion concurring in part and dissenting in part).

III

The subject of postdecision supplemental environmental impact statements is not expressly addressed in NEPA.13 Preparation of such statements, however, is at times necessary to satisfy the Act's "action-forcing" purpose.14 NEPA does not work by mandating that agencies achieve particular substantive environmental results. Rather, NEPA promotes its sweeping commitment to "prevent or eliminate damage to the environment and biosphere" by focusing Government and public attention on the environmental effects of proposed agency action. 42 U. S. C. § 4321. By so focusing agency attention, NEPA ensures that the agency will not act on incomplete information, only to regret its decision after it is too late to correct. See Robertson, ante, at 349. Similarly, the broad dissemination of information mandated by NEPA permits the public and other government agencies to react to the effects of a proposed action at a meaningful time. Ante, at 349-350. It would be incongruous with this approach to environmental protection, and with the Act's manifest concern with preventing uninformed action, for the blinders to adverse environmental effects, once unequivocally removed, to be restored prior to the completion of agency action simply because the relevant proposal has received initial approval. As we explained in TVA v. Hill, 437 U.S. 153, 188, n. 34 (1978), although "it would make sense to hold NEPA inapplicable at some point in the life of a project, because the agency would no longer have a meaningful opportunity to weigh the benefits of the project versus the detrimental effects on the environment," up to that point, "NEPA cases have generally required agencies to file environmental impact statements when the remaining governmental action would be environmentally `significant.' "15

This reading of the statute is supported by Council on Environmental Quality (CEQ) and Corps regulations, both of which make plain that at times supplementation is required. The CEQ regulations, which we have held are entitled to substantial deference, see Robertson, ante, at 355-356; Andrus v. Sierra Club, 442 U.S. 347, 358 (1979), impose a duty on all federal agencies to prepare supplements to either draft or final EIS's if there "are significant new circumstances or information relevant to environmental concerns and bearing on the proposed action or its impacts."16 Similarly, the Corps' own NEPA implementing regulations require the preparation of a supplemental EIS if "new significant impact information, criteria or circumstances relevant to environmental considerations impact on the recommended plan or proposed action."17

The parties are in essential agreement concerning the standard that governs an agency's decision whether to prepare a supplemental EIS. They agree that an agency should apply a "rule of reason," and the cases they cite in support of this standard explicate this rule in the same basic terms. These cases make clear that an agency need not supplement an EIS every time new information comes to light after the EIS is finalized.18 To require otherwise would render agency decisionmaking intractable, always awaiting updated information only to find the new information outdated by the time a decision is made.19 On the other hand, and as petitioners concede, NEPA does require that agencies take a "hard look" at the environmental effects of their planned action, even after a proposal has received initial approval. See Brief for Petitioners 36. Application of the "rule of reason" thus turns on the value of the new information to the still pending decisionmaking process. In this respect the decision whether to prepare a supplemental EIS is similar to the decision whether to prepare an EIS in the first instance: If there remains "major Federal actio[n]" to occur, and if the new information is sufficient to show that the remaining action will "affec[t] the quality of the human environment" in a significant manner or to a significant extent not already considered, a supplemental EIS must be prepared.20 Cf. 42 U. S. C. § 4332(2)(C).

The parties disagree, however, on the standard that should be applied by a court that is asked to review the agency's decision. Petitioners argue that the reviewing court need only decide whether the agency decision was "arbitrary and capricious," whereas respondents argue that the reviewing court must make its own determination of reasonableness to ascertain whether the agency action complied with the law. In determining the proper standard of review, we look to § 10e of the Administrative Procedure Act (APA), 5 U. S. C. § 706, which empowers federal courts to "hold unlawful and set aside agency action, findings, and conclusions" if they fail to conform with any of six specified standards.21 We conclude that review of the narrow question before us whether the Corps' determination that the FEISS need not be supplemented should be set aside is controlled by the "arbitrary and capricious" standard of § 706(2)(A).

Respondents contend that the determination whether the new information suffices to establish a "significant" effect is either a question of law or, at a minimum, a question of ultimate fact and, as such, "deserves no deference" on review. Brief for Respondents 29. Apparently, respondents maintain that the question for review centers on the legal meaning of the term "significant" or, in the alternative, the predominantly legal question whether established and uncontested historical facts presented by the administrative record satisfy this standard. Characterizing the dispute in this manner, they posit that strict review is appropriate under the "in accordance with law" clause of § 706(2)(A) or the "without observance of procedure required by law" provision of § 706 (2)(D). We disagree.

The question presented for review in this case is a classic example of a factual dispute the resolution of which implicates substantial agency expertise. Respondents' claim that the Corps' decision not to file a second supplemental EIS should be set aside primarily rests on the contentions that the new information undermines conclusions contained in the FEISS, that the conclusions contained in the ODFW memorandum and the SCS survey are accurate, and that the Corps' expert review of the new information was incomplete, inconclusive, or inaccurate. The dispute thus does not turn on the meaning of the term "significant" or on an application of this legal standard to settled facts. Rather, resolution of this dispute involves primarily issues of fact.22 Because analysis of the relevant documents "requires a high level of technical expertise," we must defer to "the informed discretion of the responsible federal agencies." Kleppe v. Sierra Club, 427 U.S. 390, 412 (1976). See also Baltimore Gas & Electric Co. v. Natural Resources Defense Council, Inc., 462 U.S. 87, 103 (1983) ("When examining this kind of scientific determination. . . a reviewing court must generally be at its most deferential"). Under these circumstances, we cannot accept respondents' supposition that review is of a legal question and that the Corps' decision "deserves no deference." Accordingly, as long as the Corps' decision not to supplement the FEISS was not "arbitrary or capricious," it should not be set aside.23

As we observed in Citizens to Preserve Overton Park, Inc. v. Volpe, 401 U.S. 402, 416 (1971), in making the factual inquiry concerning whether an agency decision was "arbitrary or capricious," the reviewing court "must consider whether the decision was based on a consideration of the relevant factors and whether there has been a clear error of judgment." This inquiry must "be searching and careful," but "the ultimate standard of review is a narrow one." Ibid. When specialists express conflicting views, an agency must have discretion to rely on the reasonable opinions of its own qualified experts even if, as an original matter, a court might find contrary views more persuasive. On the other hand, in the context of reviewing a decision not to supplement an EIS, courts should not automatically defer to the agency's express reliance on an interest in finality without carefully reviewing the record and satisfying themselves that the agency has made a reasoned decision based on its evaluation of the significance — or lack of significance — of the new information. A contrary approach would not simply render judicial review generally meaningless, but would be contrary to the demand that courts ensure that agency decisions are founded on a reasoned evaluation "of the relevant factors."

IV

Respondents' argument that significant new information required the preparation of a second supplemental EIS rests on two written documents. The first of the documents is the so-called "Cramer Memorandum," an intraoffice memorandum prepared on February 21, 1985, by two scientists employed by ODFW. See Cramer Memorandum 3a.24 The Cramer Memorandum, in turn, relied on a draft ODFW study describing the effects of the Lost Creek Dam on fish production. The second document is actually a series of maps prepared in 1982 by SCS to illustrate the composition of soil near the Elk Creek shoreline. The information was provided to the Corps for use in managing the project. Although respondents contend that the maps contained data relevant to a prediction of the dam's impact on downstream turbidity, the maps do not purport to shed any light on that subject. Nor do they purport to discuss any conditions that had changed since the FEISS was completed in 1980. The Corps responded to the claim that these documents demonstrate the need for supplementation of the FEISS by preparing a formal Supplemental Information Report, dated January 10, 1986. See U. S. Army Corps of Engineers, Portland District, Elk Creek Lake Supplemental Information Report No. 2, p. 7a (hereinafter SIR).25 The SIR explained: "While it is clear based upon our review that this information does not require additional NEPA documentation, Corps regulations provide that a Supplemental Information Report can be used to disseminate information on points of concern regarding environmental impacts set forth in the EIS."26

The significance of the Cramer Memorandum and the SCS survey is subject to some doubt. Before respondents commenced this litigation in October 1985, no one had suggested that either document constituted the kind of new information that made it necessary or appropriate to supplement the FEISS. Indeed, the record indicates that the Corps was not provided with a copy of the Cramer Memorandum until after the lawsuit was filed. Since the probative value of that document depends largely on the expert qualification of its authors, the fact that they did not see fit to promptly apprise the Corps of their concern — or to persuade ODFW to do so — tends to discount the significance of those concerns. Similarly, the absence of any pretrial expression of concern about the soil characteristics described in the 1982 SCS survey is consistent with the view that it shed little, if any, new light on the turbidity potential of the dam. Yet, even if both documents had given rise to prompt expressions of concern, there are good reasons for concluding that they did not convey significant new information requiring supplementation of the FEISS.

The Court of Appeals attached special significance to two concerns discussed in the Cramer Memorandum: the danger that an increase in water temperature downstream during fall and early winter will cause an early emergence and thus reduce survival of spring chinook fry and the danger that the dam will cause high fish mortality from an epizootic disease. Both concerns were based partly on fact and partly on speculation.

With respect to the first, the Cramer Memorandum reported that the authors of the draft ODFW study had found that warming of the Rogue River caused by the Lost Creek Dam had reduced the survival of spring chinook fry; however, the extent of that reduction was not stated, nor did the memorandum estimate the extent of warming to be expected due to closure of the Elk Creek Dam. Instead, the memorandum estimated that an increase of only one degree centigrade in river temperature in January would decrease survival of spring chinook "from by 60-80%." Cramer Memorandum 3a. The authors of the memorandum concluded that because the Elk Creek Dam is likely to increase the temperature of the Rogue River, further evaluation of this effect should be completed "before ODFW sets its final position on this project." Ibid.

The Corps' response to this concern in its SIR acknowledged that the "biological reasoning is sound and has been recognized for some time," but then explained why the concern was exaggerated. SIR 10a. The SIR stressed that because the model employed by ODFW had not been validated, its predictive capability was uncertain. Indeed, ODFW scientists subsequently recalculated the likely effect of a one-degree-centigrade increase in temperature, adjusting its estimate of a 60-to-80 percent loss downward to between 30 and 40 percent. Id., at 9a. Moreover, the SIR supplied a variable missing in the Cramer Memorandum, suggesting that the Elk Creek Dam would, in most cases, either reduce or leave unchanged the temperature of the Rogue River. Id., at 10a. Discernible increases were only found in July, August, and December of the study year, and even during those months the maximum temperature increase was only 0.6 degrees centigrade. Ibid. Finally, the SIR observed that the Cramer Memorandum failed to take into account the dam's beneficial effects, including its ability to reduce peak downstream flow during periods of egg incubation and fry rearing and its ability to reduce outflow temperature through use of the multiport structure.27 Id., at 9a-10a. Given these positive factors, the Corps concluded that any adverse effects of the 0.6-degree temperature increase can be offset. Id., at 10a.

With respect to the second concern emphasized by the Court of Appeals, the Cramer Memorandum reported the fact that "an unprecedented 76% of the fall chinook in 1979 and 32% in 1980 were estimated to have died before spawning" and then speculated that the Lost Creek Dam, which had been completed in 1977, was a contributing cause of this unusual mortality.28 Cramer Memorandum 4a. The Corps responded to this by pointing out that the absence of similar epizootics after the closure of the Applegate Dam and the evidence of prespawning mortality in the Rogue River prior to the closing of the Lost Creek Dam were inconsistent with the hypothesis suggested in the Cramer Memorandum. See SIR 10a-11a. In addition, the Corps noted that certain diseased organisms thought to have been the cause of the unusually high mortality rates were not found in the outflow from the Lost Creek Dam.29 Id., at 11a.

In thus concluding that the Cramer Memorandum did not present significant new information requiring supplementation of the FEISS, the Corps carefully scrutinized the proffered information. Moreover, in disputing the accuracy and significance of this information, the Corps did not simply rely on its own experts. Rather, two independent experts hired by the Corps to evaluate the ODFW study on which the Cramer Memorandum was premised found significant fault in the methodology and conclusions of the study.30 We also think it relevant that the Cramer Memorandum did not express the official position of ODFW. See SIR 9a. In preparing the memorandum, the authors noted that the agency had "adopted a neutral stand on Elk Creek Dam" and argued that new information raised the question whether "our agency should continue to remain neutral."31 Cramer Memorandum 3a. The concerns disclosed in the memorandum apparently were not sufficiently serious to persuade ODFW to abandon its neutral position.

The Court of Appeals also expressed concern that the SCS survey, by demonstrating that the soil content in the Elk Creek watershed is different than assumed in the FEISS, suggested a greater turbidity potential than indicated in the FEISS. 832 F. 2d, at 1495. In addition, the court observed that ODFW scientists believe that logging and road building in the Elk Creek watershed has caused increased soil disturbance resulting in higher turbidity than forecast by the FEISS. Ibid. As to this latter point, the SIR simply concluded that although turbidity may have increased in the early 1980's due to logging, "watershed recovery appears to have occurred to reduce the turbidity levels back to those of the 1970's." SIR 12a. The implications of the SCS soil survey are of even less concern. As discussed in the FEISS, water quality studies were conducted in 1974 and 1979 using computer simulation models. FEISS 33. The 1974 study indicated that turbidity in the Rogue River would increase by no more than one to three JTU's as a result of the Elk Creek Dam, and the 1979 study verified this result. Ibid. These studies used water samples taken from Elk Creek near the proposed dam site and from near the Lost Creek Dam, and thus did not simply rely on soil composition maps in drawing their conclusions. Id., at 18-19, 21-22, 33-34. Although the SIR did not expressly comment on the SCS survey, in light of the in-depth 1974 and 1979 studies, its conclusion that "the turbidity effects are not expected to differ from those described in the 1980 EISS" surely provided a legitimate reason for not preparing a supplemental FEISS to discuss the subject of turbidity. SIR 12a.

There is little doubt that if all of the information contained in the Cramer Memorandum and SCS survey was both new and accurate, the Corps would have been required to prepare a second supplemental EIS. It is also clear that, regardless of its eventual assessment of the significance of this information, the Corps had a duty to take a hard look at the proffered evidence. However, having done so and having determined based on careful scientific analysis that the new information was of exaggerated importance, the Corps acted within the dictates of NEPA in concluding that supplementation was unnecessary. Even if another decisionmaker might have reached a contrary result, it was surely not "a clear error of judgment" for the Corps to have found that the new and accurate information contained in the documents was not significant and that the significant information was not new and accurate. As the SIR demonstrates, the Corps conducted a reasoned evaluation of the relevant information and reached a decision that, although perhaps disputable, was not "arbitrary or capricious."

The judgment of the Court of Appeals is accordingly reversed, and the case is remanded for further proceedings consistent with this opinion.

It is so ordered.