This case involves the adequacy of an Environmental Impact Statement (EIS), as supplemented, prepared by the Army Corps of Engineers for the construction of Warm Springs Dam in Northern California. The central issue is whether the Corps was required to supplement its EIS further when it became aware of a U.S. Geological Survey (USGS) study indicating that the Maacama Fault, which at its closest point is six miles from the dam, might be capable of generating an earthquake of greater magnitude than the dam was designed to withstand.

The case is on appeal from the district court's denial of a permanent injunction against further construction on the project. Appellant is the Warm Springs Dam Task Force (Task Force). We affirm.

The Warm Springs Dam is designed to be a 319-foot earth-fill dam across Dry Creek, a major tributary of the Russian River in Sonoma County, California. The dam would create Lake Sonoma, with a surface of 3,600 acres and a storage capacity of 381,000 acre-feet of water, and would serve the combined purposes of flood control, water conservation and recreation. The dam site is located on the inactive Dry Creek earthquake fault. Several active faults, including the San Andreas Fault, the Healdsburg Fault and the Maacama Fault, pass within several miles of the dam site.

Congress authorized construction of the Warm Springs Dam by the Army Corps of Engineers in the Flood Control Act of 1962, Pub.L.No. 87-874, § 203, 76 Stat. 1173, 1192 (1962). Following the enactment of NEPA, which took effect in 1970, the Corps prepared an EIS for the dam project. The EIS was released in final form in 1973.

In 1974, the Task Force attacked the adequacy of the EIS on several grounds and brought an action in the district court to enjoin further construction. The district court found that the EIS fully complied with NEPA and denied the motion for the injunction. Warm Springs Dam Task Force v. Gribble, 378 F. Supp. 240 (N.D.Cal.1974). A request for an injunction pending appeal was denied by the district court and by this court, but an injunction was granted by Mr. Justice Douglas, sitting as a Circuit Justice. Warm Springs Dam Task Force v. Gribble, 417 U.S. 1301, 94 S. Ct. 2542, 41 L. Ed. 2d 654 (Douglas, Circuit Justice, 1974), motion to vacate stay denied, 418 U.S. 910, 94 S. Ct. 3202, 41 L. Ed. 2d 1156 (1974). On appeal, we remanded for reconsideration of the issues of seismic safety and water quality, in an unpublished order, on August 18, 1975. The injunction preventing further construction remained in effect pending the district court hearing.

Prior to the new hearing in the district court, the Corps prepared a Supplement to the Final Environmental Impact Statement (S-EIS), addressing the problems of seismic safety and water quality. In the new hearing, the district court concluded that the S-EIS complied with NEPA and the court therefore denied the Task Force's motion for a permanent injunction. Warm Springs Dam Task Force v. Gribble, 431 F. Supp. 320 (N.D.Cal.1977). It is from that judgment that the Task Force now appeals. The Task Force in this appeal limits its objections to the questions of seismic safety and does not persist in the contentions regarding water quality.

The district court and this court denied a motion brought by the Task Force for an injunction pending appeal, Warm Springs Dam Task Force v. Gribble, 565 F.2d 549 (9th Cir. 1977), and this court has since denied the Task Force's motion for a stay pending disposition on the merits.

The Task Force challenges the adequacy of the S-EIS on several grounds, raising the following issues:1. Did the Corps have a duty to obtain the USGS's written comments prior to filing its final S-EIS?

2. Must the S-EIS be revised in light of new evidence developed by the USGS concerning the seismic safety of the dam?

3. Does the S-EIS adequately discuss the consequences of surface displacement on the Dry Creek Fault?

4. Must the S-EIS include a discussion of the consequences of catastrophic failure of the dam?

In addition to raising the procedural issues relating to the adequacy of the S-EIS, the Task Force attacks the substantive decision of the Corps to proceed with the project. That challenge raises the question of whether the decision to build the dam should be set aside as arbitrary, capricious or an abuse of discretion.

Following the 1974 litigation and the remand by this court for consideration of the issues of seismic safety and water quality, the Corps commenced preparation of the supplement to the EIS. The Corps hired the private firm of Dames and Moore to conduct a "dynamic analysis" of the dam to test the seismic safety of the design. The process involved constructing a model of the dam in a laboratory and subjecting it to simulated earthquakes that would represent "the maximum credible earthquakes" the dam would be required to withstand.

The tests conducted by Dames and Moore were based on the assumption that the faults that could generate earthquakes having the greatest destructive force on the dam were the San Andreas Fault, which is nineteen miles from the dam and which could generate an earthquake of a magnitude of 8.3 on the Richter Scale, and the Healdsburg Fault, which is two miles from the dam and which could generate an earthquake of a magnitude of 7.0 on the Richter Scale. Dames and Moore did not consider the Maacama Fault, which is six miles from the dam, because that fault was believed to extend no more than twenty-six miles and was thus capable of generating a maximum credible earthquake of only 6.6 on the Richter Scale; the resulting destructive force of such a quake would be less than that which could be generated by quakes on either the San Andreas or the Healdsburg Faults. Dames and Moore also determined that the maximum credible earthquake of 7.0 on the Healdsburg Fault would cause less destructive force on the dam than would the maximum credible earthquake of 8.3 on the San Andreas Fault. The maximum destructive force that the dam must be engineered to withstand was thus determined to be an earthquake of 8.3 on the San Andreas Fault, with the reservoir filled to the maximum water level behind the dam.

The Corps circulated its draft of the S-EIS in May 1976, requesting comments from interested parties and agencies. The USGS was one of the agencies formally requested to submit comments, but it filed no written response. The Corps distributed its final draft of the S-EIS, incorporating all comments received, in September 1976.

During the Corps's preparation of the S-EIS, Dr. Darrell G. Herd, a geologist with the USGS, was mapping the faults of the Northwest San Francisco Bay Region, including the dam site. He testified at trial that he had submitted his conclusions in draft form for review within the USGS in April 1976, one month before publication of the draft S-EIS and five months before distribution of the final S-EIS.

Dr. Herd further testified that his study revealed that the Maacama Fault, which previously had been thought to extend a distance of about 40 kilometers (25 miles),1 appeared to extend at least 110 kilometers (68 miles) and could extend 230 kilometers (143 miles). Dr. Herd also suggested that it was possible, although highly unlikely, that the Maacama Fault could be part of a great en echelon fault system that could extend 470 kilometers (292 miles) and could rupture as one continuous fault system.

Dr. Herd admitted that his mapping did not extend far enough to the north to indicate fully the length of the Maacama Fault. He stated that for a more accurate assessment of the earthquake hazard presented by the Maacama Fault, geologic mapping must be undertaken to determine the extent of the northward continuation of the Maacama Fault.

Section 102(2)(C) of NEPA, 42 U.S.C. § 4332(2)(C), provides that prior to making the environmental impact statement,

the responsible Federal official shall consult with and obtain the comments of any Federal agency which has jurisdiction by law or special expertise with respect to any environmental impact involved.

(emphasis added). Appellant claims the Corps violated NEPA by failing to obtain written comments from the USGS prior to publication of the S-EIS. If the Corps had obtained these comments, appellant contends, it would have been made aware of Dr. Herd's studies and might have halted construction of the dam.

The Corps concedes that the USGS is an agency having special expertise in geology and seismic activity within the meaning of the statute and that the Corps did not obtain written comments from the USGS.2 However, it contends the Corps fulfilled its statutory duty because: (1) it consulted informally with USGS personnel in preparing the S-EIS; and (2) it sent a copy of its draft S-EIS to the USGS, specifically soliciting the agency's comments.

Section 4332(2)(C) is designed to attract knowledgeable agency comment on the environmental issues raised by proposed federal projects. By requiring the proposing agency to consult with and obtain comments from agencies possessing special experience and expertise in those issues, Congress sought to further the statutory purpose of encouraging widespread discussion and consideration of the environmental risks and remedies associated with the pending project.

Informal inter-agency consultation furthers these statutory policies; in fact, effective implementation of statutory policies requires that agency officials be encouraged to consult freely with knowledgeable individuals outside the agency. But informal consultation alone is not sufficient compliance with the statute. Section 4332(2)(C) requires each agency possessing special expertise to comment in writing on its official view of the environmental consequences of the proposed action. This requirement is essential to maintain the integrity and regularity of the decisionmaking process. By requiring the commenting agency to take an official position, even it it be "no comment," Congress encourages the agency to direct the draft EIS for study to those personnel within its organizational structure most likely to bring to light any additional facts that should be considered or to raise any reasoned disagreement with the draft's conclusions. Regular review procedures are thereby established. If the proposing agency could comply with the statutory requirement merely by selecting individuals within the commenting agency to serve as consultants, as the Corps suggests, there would be too great a risk that the only individuals contacted would be those the proposing agency considered most likely to support its proposal. Some official consideration by the independent "expert" agency is clearly called for.

The duty to obtain official written comments is established by the statutory language of section 4332(2)(C). This section does not merely require the proposing agency to consult with and obtain comments from independent agencies possessing special expertise. Section 4332(2)(C) also provides:

Copies of such statement and the comments and views of the appropriate Federal, State, and local agencies, which are authorized to develop and enforce environmental standards, shall be made available to the President, the Council on Environmental Quality and to the public as provided by section 552 of title 5, and shall accompany the proposal through the existing agency review processes; . . . .

(emphasis added). By requiring agency comment to be submitted in a form that will allow "Copies . . . (to) be made available (and to) accompany the proposal," the statute imposes on the proposing agency a duty to obtain written comments. There would be no "copies" of oral comments to be made available or to accompany the proposal otherwise. Moreover, oral comments alone would not promote the congressional goal of guaranteeing a widespread public airing of environmental issues. Only if the commenting agency takes an official position in writing can there be full compliance with section 4332(2) (C).

This interpretation is confirmed by the guidelines promulgated by the Council on Environmental Quality (CEQ). Not only do these guidelines require the proposing agency to "obtain the comments" of any federal agency having jurisdiction by law or expertise with respect to any environmental impact, but they also place on the commenting agencies an affirmative duty to respond. By contrast to the treatment afforded other prospective commenting entities, whose silence in the face of a request for comments can be construed as a response of "no comment," 40 C.F.R. § 1500.9(f), the guidelines now state that federal agencies having jurisdiction by law or expertise "shall comment on statements within their jurisdiction, expertise or authority." 40 C.F.R. § 1503.2 (emphasis added).3 While these agencies are permitted to respond that they have no comment, the guidelines clearly require that some written response be made.

Notwithstanding our conclusion that the Corps violated NEPA by not obtaining the written official comments of USGS, we decline to hold that the district court should be reversed and an injunction issued on this ground. Relief under NEPA should be remedial rather than punitive. See Realty Income Trust v. Eckerd, 564 F.2d 447, 456 (D.C.Cir.1977). The Corps made a good faith effort to comply with NEPA as it reasonably understood it, and no prejudice resulted from its failure to obtain USGS's written comments.

When the Corps circulated its draft S-EIS, it reasonably believed NEPA required only that agencies with special expertise be requested to submit comments. The Corps's NEPA regulations provided that:

Reporting officers should establish a time limit of not less than 45 days for reply after date of publication of draft environmental statement by CEQ in the FEDERAL REGISTER. In the absence of a specific request for an extension of time, a lack of response may be presumed to indicate that the agency consulted has no comment to make.

33 C.F.R. § 209.410(k)(1) (emphasis added). Similarly, CEQ guidelines provided:

Agencies seeking comments shall establish time limits of not less than forty-five (45) days for reply, after which it may be presumed, unless the agency or party consulted requests a specified extension of time, that the agency or party consulted has no comment to make.

40 C.F.R. § 1500.9(f) (emphasis added). Not having received a response from the USGS within the 45 day period the Corps, pursuant to its regulations, could reasonably have presumed the USGS simply had no comment to make.

The Corps's good faith is also reflected in the efforts it undertook to obtain comments. The Corps sent the draft of the S-EIS and a request for comments to nearly 20 federal agencies and 200 other entities, including state, county and local agencies, public officials, conservation groups, educational institutions, newspapers and private individuals. Moreover, after waiting the prescribed period the Corps held a hearing to receive public comment on its draft. All the comments received, numbering about 80, were published in a 387-page appendix to the S-EIS.

Appellant suggests only one prejudicial consequence of the Corps's good faith failure to obtain written comments from USGS. It contends that if the USGS had been required to take a formal position with respect to the draft S-EIS it might have made reference to Dr. Herd's draft study challenging the Corps's seismic assumptions. Thus alerted, the Corps (which was in fact unaware of the study) might have confronted the possibility that Maacama was the controlling fault for purposes of dam design. Except for speculation by Dr. Herd in his testimony at trial, however, there is no indication in the record that the Corps would have learned of Dr. Herd's study had it obtained written comments from the USGS.

Moreover, even if we assume the USGS would have referred to Dr. Herd's report in its written comments, no prejudice resulted from the omission. As will be discussed in greater detail below, not only did the Corps subsequently consider Dr. Herd's report, but in response to the report the Corps commissioned a series of seismic studies that eventually showed Dr. Herd's concerns to be unwarranted. It would be fruitless now to order the Corps to obtain written comments from the USGS in the expectation these comments would lead the Corps to institute a study it has already completed and fully considered.

Dr. Herd's study revealed that the Maacama Fault extended considerably farther north than earlier believed. While he stated that further mapping would be necessary to determine the length of the fault, Dr. Herd observed that there was good geomorphic evidence the fault extended 110 kilometers to Willits (with a capability of generating an earthquake of 7.3 magnitude), and that it was possible the fault extended 230 kilometers to Cape Mendocino (with a capability of generating an earthquake of 8.1 magnitude only six miles from the dam). Dr. Valera of Dames and Moore testified that an earthquake of 8.1 magnitude on the Maacama Fault would be the controlling earthquake because it would exert a considerably more destructive force on the dam than an 8.3 earthquake on the San Andreas Fault. The S-EIS does not discuss whether the dam, as designed, could withstand such a force. Dr. Valera testified he could not state whether the dam could withstand such an earthquake.

The question we must resolve is whether this information, which came to the attention of the Corps after the publication of its S-EIS, requires a further supplementation of the EIS.

We start with the premise that a federal agency has a continuing duty to gather and evaluate new information relevant to the environmental impact of its actions. See 42 U.S.C. § 4332(2)(A), (B); Essex County Preservation Ass'n v. Campbell, 536 F.2d 956, 960-61 (1st Cir. 1976); Society for Animal Rights, Inc. v. Schlesinger, 512 F.2d 915, 917-18 (D.C.Cir.1975).

This does not mean, however, that supplementation is required whenever new information becomes available. The Corps's regulations, for example, require a supplement to be filed if the EIS "becomes deficient because certain environmental effects of the project were not discussed" but not if "it is necessary only to clarify or amplify a point of concern".4 The CEQ guidelines, which at the time of the hearing created no mandatory duty to supplement an EIS in light of new information,5 now provide at 40 C.F.R. § 1502.9(c) that agencies

(1) Shall prepare supplements to either draft or final environmental impact statements if:

(ii) There are significant new circumstances or information relevant to environmental concerns and bearing on the proposed action or its impacts. (emphasis added).

These regulations do not in themselves provide a suitable standard for reviewing an agency's decision not to supplement an EIS in light of new information. However, the standard applied in reviewing an agency's decision not to file an EIS in the first instance is appropriate here as well; the decision will be upheld if it was reasonable. See City of Davis v. Coleman, 521 F.2d 661, 673 (9th Cir. 1975). When new information comes to light the agency must consider it, evaluate it, and make a reasoned determination whether it is of such significance as to require implementation of formal NEPA filing procedures. Reasonableness depends on such factors as the environmental significance of the new information, the probable accuracy of the information, the degree of care with which the agency considered the information and evaluated its impact, and the degree to which the agency supported its decision not to supplement with a statement of explanation or additional data. Cf. Maryland-National Capitol Park & Planning Comm'n v. United States Postal Serv., 487 F.2d 1029, 1040 (D.C.Cir.1973).

Dr. Herd's report threatened to undermine the major assumption underlying the S-EIS, that an earthquake on the San Andreas Fault would pose a greater potential threat to dam safety than an earthquake on any other fault. By suggesting that the Maacama Fault might extend 230 km and be capable of generating an 8.1 earthquake only six miles from the dam, Dr. Herd raised the possibility that Maacama, and not San Andreas, would be the controlling earthquake for the purposes of the dam.

The accuracy of Dr. Herd's data was far from settled. His report admittedly dealt in possibilities and speculation. It was more significant for the questions it raised than for the answers it gave. Describing his findings to California's Division of Mines and Geology, Dr. Herd wrote:

A primary difficulty in assessing the earthquake magnitude for the Maacama fault is the present lack of data. The Maacama fault length as shown in MF-818 is 40 km. However . . . there is good geomorphic evidence that the fault system continues (110 km) to Willits. However, it is possible that the Maacama fault system continues northward as a continuous fault system (230 km) into the Cape Mendocino area.

As this language reflects, the 230 km estimate was an unverified "possibility" that could be tested only through further mapping.

While not so definitive as to compel initiation of the formal supplementation process, Dr. Herd's study raised sufficient environmental concerns to require the Corps to take another hard look at the issues. At the time of trial this had not yet been done. The Corps merely argued that the environmental considerations raised by Dr. Herd had been adequately explored in the EIS and its supplement. But this was not true. Neither the EIS or S-EIS dealt with the fact that the Maacama Fault might be the controlling earthquake for design purposes. The Corps' response to the new information contained in the Herd report did not satisfy NEPA. On the basis of the information then available, the Corps' decision not to file a further supplement to the S-EIS was not reasonable.

Although the Corps' actions prior to trial did not comport with NEPA, however, the deficiency has since been cured. In February 1977, a month after the start of trial, the Corps launched an extensive ten-month study of the Maacama Fault and its possible impact on the dam. This study was prepared in conjunction with state and federal agencies, including the USGS and the California Division of Mines and Geology. It was based on prior studies, including the work of Dames and Moore and Dr. Herd, and extensive aerial photographic mapping. Published in January 1978, the study concluded the Maacama Fault extended 150 km and was capable of generating a maximum credible earthquake of 7.5, a less destructive force on the dam than the 8.3 on the San Andreas that the dam was designed to withstand.6 The study therefore resolved the concerns voiced by Dr. Herd and reaffirmed that the controlling earthquake at Warm Springs was the one upon which the S-EIS was based.

The Corps went even further. After receiving the ten-month study the Corps commissioned two University of California at Berkeley professors, one a professor of Civil Engineering, one a professor of Seismology, to review it. Their report concluded that the ten-month study's finding that Maacama could generate only a 7.5 earthquake was an "extremely conservative evaluation," and that

a magnitude 71/2 event on the Maacama Fault would be no more damaging in its effects on the Warm Springs Dam than a magnitude 81/4 event on the San Andreas Fault. Accordingly if the design has adequate stability against the latter event it will have adequate stability against any earthquake on the Maacama Fault and no further studies for earthquakes on the Maacama fault need be conducted for this dam.

This conclusion was reaffirmed by the Corps's Independent Board of Consultants and by California's Secretary of Resources in February 1978.

On the basis of these studies the Corps could reasonably conclude, as it did, that the potentially adverse environmental effects disclosed by the Herd report were not significant and therefore did not require the preparation and circulation of a formal supplement to the S-EIS. These supervening events, although properly before us,7 were not before the district court. In such circumstances "the preferred procedure is to remand to give the district court an opportunity to pass on the changed circumstances." Korn v. Franchard Corp., 456 F.2d 1206, 1208 (2d Cir., 1972). Remand is not required, however, where what has occurred since the judgment was entered is undisputed and "demands one result only." Id. The district court could not order the Corps to conduct studies already completed to answer questions the Corps has already answered on a basis that could not be successfully challenged.

The Task Force raises the contention that the EIS fails to discuss adequately that the dam is being constructed on the inactive Dry Creek Fault, where activity could be induced by earthquakes on nearby faults, or by the accumulation of water in the reservoir behind the dam, and noted that the USGS has a new study on the effect the impoundment of water in reservoirs has in triggering earthquakes. The district court stated:

As to the question of the effect of the impoundment of water on the triggering of earthquakes, the subject is, in fact, discussed in the EIS. In its discussion, the EIS cites as authority for its conclusions many of the same authorities relied upon by the U.S.G.S. in its new report. And while the new report has some statistical data not contained in the EIS that might lead one to different conclusions than those contained in the EIS as to the possible effect of water impoundment on seismicity in the Warm Springs project area, the subject matter itself is adequately covered.

431 F.Supp. at 323. We agree with the district court that this new report would not require the preparation of a revised EIS.

The Task Force complains that the EIS contains no discussion of the consequences of total failure of the dam in the wake of a catastrophic seismic event. We hold that such discussion is unnecessary.

An impact statement must be particularly thorough when the environmental consequences of federal action are great. See Iowa Citizens for Environmental Quality, Inc. v. Volpe, 487 F.2d 849, 852 (8th Cir. 1973). However, an impact statement need not discuss remote and highly speculative consequences. Trout Unlimited v. Morton, 509 F.2d 1276, 1283 (9th Cir. 1974); Environmental Defense Fund, Inc. v. Hoffman, 566 F.2d 1060, 1067 (8th Cir. 1977). Any substantial risk that the dam could fail would be intolerable; and, if the agency were to proceed in the face of that risk, that would constitute an abuse of agency discretion. Everyone recognizes the catastrophic results of the failure of a dam; to detail these results would serve no useful purpose.

E. Substantive Review of the Decision to Proceed

The Task Force urges us to find that the Corps's substantive decision to proceed with the dam project was unlawful. The substantive review of agency action is governed by the Administrative Procedure Act, 5 U.S.C. § 706(2)(A). Lathan v. Brinegar, 506 F.2d 677, 692-93 (9th Cir. 1974) (en banc ). That subsection authorizes the reviewing court to set aside agency action that is found to be "arbitrary, capricious, an abuse of discretion, or otherwise not in accordance with law."

In Citizens to Preserve Overton Park, Inc. v. Volpe, 401 U.S. 402, 91 S. Ct. 814, 28 L. Ed. 2d 136 (1971), the Supreme Court described the limits of judicial review under § 706(2)(A):

(T)he 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. (citations omitted) Although this inquiry into the facts is to be searching and careful, the ultimate standard of review is a narrow one. The court is not empowered to substitute its judgment for that of the agency.

Id. at 416, 91 S.Ct. at 823-24. Agency action may not be set aside as arbitrary or capricious unless there is no rational basis for the action. Bowman Transp., Inc. v. Arkansas-Best Freight System, Inc., 419 U.S. 281, 290, 95 S. Ct. 438, 444, 42 L. Ed. 2d 447 (1974); Short Haul Survival Committee v. United States, 572 F.2d 240, 244 (9th Cir. 1978).

Were the Corps to have proceeded with the project without further investigation of the length of the Maacama Fault after learning of Dr. Herd's study, a persuasive argument for appellant's position could be made. However, after learning of Dr. Herd's reports, the Corps made a thorough study of the effect of the fault and reasonably concluded that no substantial adverse environmental effects were presented.

We cannot say that the Corps's decision to proceed with the Warm Springs Dam project was arbitrary, capricious or an abuse of discretion.

AFFIRMED.

The adequacy of an agency's response to "significant new information" is a point on which authority is scarce, and the question will assume ever greater importance as the life cycle of projects requiring NEPA compliance lengthens. Vermont Yankee emphasized that an EIS must be finished at a fixed time, and yet the CEQ regulations reasonably recognize that in some instances new evidence may be revolutionary enough to require a new or supplemental EIS. Our present interpretation of 40 C.F.R. § 1502.9(c)(1)(ii) finds support in an analogous section on "Incomplete or Unavailable Information." See 40 C.F.R. § 1502.22(b)(2); Gelpe & Tarlock, The Uses of Scientific Information in Environmental Decisionmaking, 48 S.Cal.L.Rev. 371, 392-96, 419-27 (1974).

Not every item of new evidence requires a formal EIS or supplement. There are other and less cumbersome ways for an agency to evaluate new information and explain why it finds no need for a new EIS. I agree with the court that the Maacama Fault Study performed this needed function here. I concur in the judgment.