Sierra Club v. James M. Sigler, Etc., Pelican Terminal Company and Galveston Wharves, Intervenors-Appellees

GEE, Circuit Judge: We review a trial court opinion upholding a Final Environmental Impact Statement, and the Army Corps of Engineers decision to issue permits based on it, authorizing the private construction of a multipurpose deepwater port and crude oil distribution system at Galveston, Texas. The trial court’s judgment is affirmed in part, reversed in part, and the case remanded. I. FACTS AND DISPOSITION BELOW We sketch a factual overview of the case, reserving more detail for our analysis of the legal issues presented. Galveston Bay, separated from the Gulf of Mexico by Galveston Island and the Bolivar Peninsula, is part of a system of estuarine bays along the Texas coast. The Bay is Texas’ largest estuary and serves as a nursery and habitat for vast numbers of wildlife, including fish and migratory birds. For example, 98% of Texas’ commercial fishing industry is based on fish that spend part of their life cycle in the Bay. However, the Bay also has served for many years as a commercial waterway for the ports of Galveston, Houston, and Texas City. Oil tankers averaging 50,000 deadweight tons (dwt) make over 2,000 vessel trips annually through its congested channels, sharing those channels with many other types of vessels. The nation’s largest concentration of refineries and petrochemical plants is found on the western shores of the Texas Gulf Coast stretching from Free-port in the south to Beaumont in the north. Between these two points lies the City of Galveston, which is located on the Galveston Island flanking the main channel into Galveston Bay. In 1974, Galveston Wharves (“Wharves”), a utility of the City of Galveston, sought to extend and deepen the channel into the Bay to facilitate the traffic in large ships. That proposal was revived in February 1978, when permit applications were filed with the Army Corps of Engineers. In September of the same year, the project was revised by the Wharves and the Pelican Terminal Company (“PELCO”), who plan to build the project without federal participa *962 tion. 1 The Wharves and PELCO proposal contemplated building an oil terminal to offload very large crude carriers (VLCC’s or “supertankers”) 2 of up to 320,000 dwt and to deepen and lengthen the channels into the Bay and the Port of Galveston to accommodate them. The proposed location of this “superport” oil terminal is Pelican Island, which is inside the Bay and adjacent to Galveston City. The 1978 proposal included a crude oil tank farm and pipeline distribution system to carry the oil to other areas of the Gulf Coast. Deepening the channels also would permit other large vessels to use the Bay, including bulk commodities carriers. In fact, the proposal would extend the deepened Port of Galveston channel one and one-half miles beyond the oil terminal specifically to accommodate such carriers. Oil tankers and their inherent dangers are not new; in fact, they have operated in Galveston Bay for years. However, this project will be the first in the United States to permit oil tankers to operate in a wildlife estuary. The primary focus of the controversy, therefore, has been on the effects of a major oil spill in a waterway which serves as a breeding ground for much of Texas’ wildlife population. In 1978, pursuant to permit applications filed on behalf of the superport project, work began on the Environmental Impact Statement (“EIS”) required by the National Environmental Policy Act of 1969 (NEPA), 42 U.S.C. § 4321 et seq. As a result of coordinated efforts by the Corps, several state and federal agencies, and many private entities, a draft EIS (“DEIS”) was produced in April of 1979. After comments were received on the draft, including some from the plaintiffs-appellants, and a public hearing was held in Galveston, a final EIS (“FEIS”) was issued in September of 1979. 3 *963 Based upon the FEIS and his authority under federal law, on July 8, 1980, the Galveston District Engineer, Colonel James M. Sigler of the U.S. Army Corps of Engineers, issued five permits authorizing the deepening of the channel and construction of the oil terminal, tank farm, and pipeline system. These permits were based on findings of fact made by Colonel Sigler in conformance with Corps regulations. On May 19, 1981, the Sierra Club and four other plaintiffs 4 brought suit against the Corps and other federal defendants 5 challenging the adequacy of the FEIS and *964 the issuance of the permits. Under Section 10(e)(2) of the Administrative Procedure Act (APA), 5 U.S.C. § 706(2), they asserted that the Corps’ decision making process violated, inter alia, NEPA, the Fish and Wildlife Coordination Act (FWCA), 16 U.S.C. § 661 et seq., section 404 of the Federal Water Pollution Control Act (FWPCA), 33 U.S.C. § 1344, section 10 of the Rivers and Harbors Appropriation Act of 1899, 33 U.S.C. § 403, and various federal regulations governing Corps activities. The Wharves, on behalf of the City of Galveston, and PELCO intervened as defendants. The trial court in Galveston expedited the matter and held seven days of hearings beginning October 21, 1981. After post-trial briefs and final arguments, the court denied all challenges to the FEIS and upheld the issuance of the permits on February 3, 1982. 532 F.Supp. 1222 (S.D.Tex. 1982). The plaintiffs appeal, advancing only three of the contentions they made below: (1) the FEIS failed to perform a “worst case” oil spill analysis, (2) the FEIS failed to analyze the environmental costs of bulk cargo activities, and (3) the Corps failed to consider an alternative offshore port. Appellees sought and were granted an expedited appeal. We are aware of no construction on the project during the pendency of this suit. II. ISSUES ON APPEAL A. STANDARDS OF REVIEW As a preliminary matter, we set out the proper standards of review of the adequacy of the FEIS, of the permit-issuing decision of the Corps, and of the opinion of the trial court. 1. Preparation of the FEIS and Decision to Issue the Permit The umbrella standard governing judicial review of the adequacy of the FEIS and the Corps’ decision to issue the permits is whether those actions were “arbitrary, capricious, an abuse of discretion, or otherwise not according to law,” or “without observance of procedure required by law.” Administrative Procedures Act, 5 U.S.C. § 706(2)(A) and (D). The Supreme Court has explained that: To make this finding the court must consider whether the decision was based on a consideration of relevant factors and whether there has been a clear error of judgment.... 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. Citizens to Preserve Overton Park v. Volpe, 401 U.S. 402, 416, 91 S.Ct. 814, 823, 28 L.Ed.2d 136 (1971). An agency decision is entitled to a “presumption of regularity.... But that presumption is not to shield [the agency’s] action from a thorough, probing, in-depth review.” Id. at 415, 91 S.Ct. at 823. This circuit has followed Overton Park in its review of agency decisions under the APA. See, e.g., Suntex Dairy v. Block, 666 F.2d 158, 162 (5th Cir. 1982)) cert. denied,--U.S.-, 103 S.Ct. 59, 74 L.Ed.2d 62 (1982). FEIS Preparation. Preparation of the FEIS was governed by NEPA and by the regulations enforcing it promulgated by the Council on Environmental Quality (CEQ) because CEQ regulations implementing NEPA are binding on all federal agencies. See Andrus v. Sierra Club, 442 U.S. 347, 356-58, 99 S.Ct. 2335, 2340, 60 L.Ed.2d 943 (1979). Section 102(C) of NEPA, 42 U.S.C. § 4332(C) is an “action-forcing” 6 provision which requires federal agencies such as the Corps to prepare a “detailed statement” when they propose “major Federal actions significantly affecting the quality of the *965 human environment.” This policy is intended to ensure that those agencies consider the environmental effects of their actions during their decisionmaking process. See Kleppe v. Sierra Club, 427 U.S. 390, 410 & n. 21, 96 S.Ct. 2718, 2730 & n. 21, 49 L.Ed.2d 576 (1976). The Corps’ decision to issue permits for the superport project is subject to NEPA. See, e.g., Bankers Life and Casualty Co. v. Village of North Palm Beach, Florida, 469 F.2d 994 (5th Cir.1972), cert. denied, 411 U.S. 916, 93 S.Ct. 1543, 36 L.Ed.2d 307 (1973). Therefore, the Corps was required to prepare and use an EIS in its decisionmaking process. NEPA is a short statute with broad goals and imprecise methods of accomplishing them. As Justice Marshall has stated, “this vaguely worded statute seems designed to serve as no more than a catalyst for development of a ‘common law’ of NEPA.... [Cjourts have responded in just that manner and have created such a ‘common law.’ ” Kleppe, 427 U.S. at 421, 96 S.Ct. at 2735 (Marshall, J., concurring in part and dissenting in part). This circuit, in the evolution of its NEPA “common law,” has announced three criteria for determining the adequacy of an EIS: (1) whether the agency in good faith objectively has taken a hard look at the environmental consequences of a proposed action and alternatives; (2) whether the EIS provides detail sufficient to allow those who did not participate in its preparation to understand and consider the pertinent environmental influences involved; and (3) whether the EIS explanation of alternatives is sufficient to permit a reasoned choice among different courses of action. Isle of Hope Historical Association, Inc. v. United States Army Corps of Engineers, 646 F.2d 215, 220 (5th Cir.1981) (per curiam) (adopting opinion of district court). In applying these criteria, we are not to concern ourselves with the merits of the agency’s decision; our concern instead is with the integrity of NEPA-EIS process used to make that decision. See Richland Park Homeowners Association, Inc. v. Pierce, 671 F.2d 935, 941 (5th Cir.1982) (citing Strycker’s Bay Neighborhood Council, Inc. v. Karlen, 444 U.S. 223, 227, 100 S.Ct. 497, 499, 62 L.Ed.2d 433 (1980) (per curiam)); see also Vermont Yankee Nuclear Power Corp. v. NRDC, 435 U.S. 519, 557-58, 98 S.Ct. 1197, 1218, 55 L.Ed.2d 460 (1978). Our duty is to determine whether the plaintiff has proven “by a preponderance of the evidence, rather than by a prima facie showing of deficiencies that the EIS . .. was inadequate.” Sierra Club v. Morton, 510 F.2d 813, 818 (5th Cir.1975) (footnote omitted). We must assess the agency’s compliance with NEPA’s procedural requirements under a “rule of reason,” id. at 819, and not “fly speck” the EIS, Citizens for Mass Transit, Inc. v. Adams, 630 F.2d 309, 313 (5th Cir.1980) (citing Lathan v. Brinegar, 506 F.2d 677, 693 (9th Cir.1974)). The CEQ regulations governing our review are a bone of contention among the parties. CEQ advisory guidelines were in force when preparation of the DEIS began. However, a new set of mandatory CEQ regulations was promulgated in November, 1978, see 43 Fed.Reg. 55,977-56,007 (1978), several months before the April 1979 filing of the DEIS, but with an effective date of July 30, 1979, see id. at 56,002; 40 C.F.R. § 1506.12. The new regulations exempted from their coverage any EIS whose draft was filed before they became effective, but they encouraged voluntary compliance by such exempt EIS’s. See 40 C.F.R. § 1506.-12(a). Therefore, at the time of filing, the Corps had the choice of proceeding under the old guidelines or the new regulations, and it expressly chose without reservation or exception the new regulations, see DEIS at iii; FEIS, Vol. I at iii. Despite this voluntary adherence, however, the Corps now claims that judicial review of the adequacy of the FEIS should proceed under the old advisory guidelines. The Sierra Club argues the contrary. Authority on this issue is divided. In the first case on point, NRDC v. Callaway, 524 F.2d 79 (2d Cir. 1974), the NRDC alleged that the Corps had issued a dredge disposal permit in inland waters in violation of Environmental Protection Agency guidelines. Those guide- *966 lines were not applicable to the Corps permit since they covered only ocean disposal of dredged material, not disposal in inland waters. See id. at 84. However, the Corps’ decision specifically referred to the ocean disposal criteria and, the court held, “relied at least in part upon [them] to support” issuance of the permit. Id. at 84-85. Therefore, the court refused to accept the Corps’ arguments that the criteria were “irrelevant” to the case. The court declared that “[b]y its own use of the standards, it has made them applicable to this case.” Id. at 85; see also Atchison, Topeka & Santa Fe Ry. v. Alexander, 480 F.Supp. 980, 993 (D.D.C.1979) (citing the Callaway holding with approval), aff’d in part, rev’d in part on other grounds sub nom. Izaak Walton League of America v. Marsh, 655 F.2d 346 (D.C.Cir.1981), cert. denied sub nom. Atchison, Topeka, & Santa Fe Ry. v. Marsh, 454 U.S. 1092, 102 S.Ct. 657, 70 L.Ed.2d 630 (1981). Despite its citation in Alexander approving Callaway, the District Court of the District of Columbia did not consider Callaway in a subsequent case similar to the one before us today. In North Slope Borough v. Andrus, 486 F.Supp. 332, 346-47, (D.D.C. 1979), aff’d in part and rev’d in part on other grounds, 642 F.2d 589 (D.C.Cir.1980), the defendant Department of the Interior (DOI) had prepared an EIS for an oil and gas lease sale in the Beaufort Sea. DOI was in the same position as the Corps here: the new CEQ regulations did not apply, yet the agency chose to proceed under them. See id. at 346. The court held that while the EIS was faulty under the CEQ regulations, the agency would not be held to them since they did not apply. 7 See id. at 346 & n. 27. Faced with the choice of following Callaway or North Slope, we follow Callaway because it is the more logical rule and it better serves the goals of NEPA. The purpose of judicial review under NEPA is to ensure the procedural integrity of the agency’s consideration of environmental factors in the EIS and in its decision to issue permits. If the agency follows a particular procedure, it is only logical to review the agency’s adherence to that procedure, not to some altogether different one that was not used. The rule of North Slope permits an agency to rely on a procedure to justify a decision yet escape judicial examination of whether it honestly and objectively employed that procedure. Such judicial abdication of NEPA responsibilities invites abuse by agencies attempting to insulate themselves from the judicial oversight mandated by the APA and NEPA. Thus the rule of Callaway better implements the intent of NEPA. When an agency voluntarily announces its adherence to a particular regulation, it is the adherence to that regulation which should be examined by a reviewing court. Permit Decision. Although Colonel Sigler relied on the FEIS in reaching his decision to issue the permits, NEPA alone does not establish our standard in reviewing that decision. *967 NEPA imposes two duties on federal agencies: the preparation of the EIS, 42 U.S.C. § 4332(C), and adherence to the CEQ regulations, 42 U.S.C. § 4332(B). We already have discussed the standard of review applied in assessing the adequacy of an agency EIS. The second duty which arises out of 42 U.S.C. § 4332(B) and the CEQ regulations requires federal agencies to “identify and develop methods and procedures .. . which will insure that presently unquantified environmental amenities and values may be given appropriate consideration in decisionmaking along with economic and technical considerations.” Section 4332(B) also requires the agency to use the EIS and other environmentally conscious procedures in its decisionmaking process. See 115 Cong.Rec. 29,055 (1969). The mandate to develop and use environmentally conscious decisionmaking procedures found in Section 4332(B) is supplementary to existing policy mandates of federal agencies, see 42 U.S.C. § 4335, and each agency must mesh the requirements of NEPA with its own governing statute as far as possible, id. § 4333. See Calvert Cliffs Coordinating Committee v. United States Atomic Energy Commission, 449 F.2d 1109, 1115 & n. 12 (D.C.Cir.1971). In this case, section 404 of the Federal Water Pollution Control Act, 33 U.S.C. § 1344, section 103 of the Marine Protection, Research, and Sanctuaries Act of 1972, 33 U.S.C. § 1413, and section 10 of the Rivers and Harbors Appropriations Act of 1899, 33 U.S.C. § 403, are the statutory commands the Corps must integrate with the requirements of NEPA. These laws prohibit construction, dredging, filling, or disposing of material in United States waters without the approval of the Army. The Corps ensures compliance with these laws by issuing permits for these activities when it determines that they are in the “public interest.” The Corps has established a policy governing its consideration of permit applications under these statutes and NEPA. See 33 C.F.R. § 320 et seq. The permit decision is based on a multitude of factors and criteria, environmental and non-environmental, which are balanced to determine whether the permit is in the “public interest.” See, e.g., 33 C.F.R. § 320.4(a)(1), (2). NEPA’s requirements are subsumed in these factors and criteria, see id. §§ 209.-410, 320.3(d), 425.4, as they must be, see 42 U.S.C. § 4333; 40 C.F.R. § 1500.2. Thus, under the laws and regulations governing the Corps, the EIS and other NEPA requirements are but factors in the decision-making process; the EIS is not itself a decision document. The decision document is a written “findings of fact” which contains the “public interest” analysis and is prepared by the District Engineer. See 33 C.F.R. § 325.2(a)(1). It is this document and the evidence supporting its analysis that must be reviewed when the decision to issue a permit is challenged. And it must be reviewed for adherence to Corps regulations, not only to NEPA, for if a court were to review it only under NEPA, the numerous non-NEPA factors which the Corps must consider would be ignored. See Environmental Defense Fund v. Marsh, 651 F.2d 983, 1002-03 (5th Cir.1981) (citing Nader v. NRC, 513 F.2d 1045 (D.C.Cir.1975); Zabel v. Tabb, 430 F.2d 199, 214 (5th Cir.1970), cert. denied, 401 U.S. 910, 91 S.Ct. 873, 27 L.Ed.2d 808 (1971); Pacific Molasses Co. v. FTC, 356 F.2d 386, 389-90 (5th Cir.1966). 2. Opinion of the Trial Court In reviewing the legal rulings of the district court, the court of appeals is free to examine them and reach its own conclusions. See, e.g., United States v. Grayson County State Bank, 656 F.2d 1070, 1075 (5th Cir.1981), cert. denied, 455 U.S. 920, 102 S.Ct. 1276, 71 L.Ed.2d 460 (1982). Factual findings, of course, are governed by the “clearly erroneous” standard of Fed.R. Civ.P. 52(a). See Morton, 510 F.2d at 818; see also Pullman-Standard v. Swint, 456 U.S. 273, 284-88, 102 S.Ct. 1781, 1788-89, 72 L.Ed.2d 66 (1982). The case before us is based largely, as administrative law cases usually are, on a documentary record. Therefore, we will observe this circuit’s longstanding rule that while the “clearly *968 erroneous” standard always applies to a district court’s factfinding, if that factfinding is based on documentary evidence alone, “[t]he appellant’s burden, under Fed.R. Civ.P. 52(a), of showing that the trial judge’s findings are ‘clearly erroneous’ is not as heavy ... as it would be if the case had turned on the credibility of witnesses appearing before the trial judge.” Sicula Oceanica, S.A. v. Wilmar Marine Eng. & Sales Corp., 413 F.2d 1332, 1333 (5th Cir. 1969); see also, e.g., Robinson v. Vollert, 602 F.2d 87, 92 n. 8 (5th Cir.1979); Nash v. Estelle, 597 F.2d 513, 518 (5th Cir.) (en banc), cert. denied, 444 U.S. 981, 100 S.Ct. 485, 62 L.Ed.2d 409 (1979); Cooper v. Department of Navy of the United States, 594 F.2d 484, 486 (5th Cir.1979) (on petition for rehearing). With the standards of review established, we turn to the issues on appeal. B. “WORST CASE” OIL SPILL ANALYSIS IN THE FEIS The principal environmental concern of the Sierra Club has been the environmental damage which could result from a total cargo loss by a supertanker in the Bay. The Sierra Club has consistently sought a study of this “worst case” possibility. Without such a study, the Sierra Club claims that a reasoned decision which fully and fairly considers all the relevant environmental factors is not possible. 1. Facts The FEIS oil spill analysis has three parts: (1) a probability analysis, (2) a dispersion model, and (3) an environmental impact analysis. From these three analyses, the FEIS concludes that the project will not cause a significantly greater probability of an oil spill than presently exists and the likely environmental harm of such spill will not be significantly greater than a spill under present conditions. Summarizing the FEIS probability analysis, the trial court concluded that the use of supertankers will reduce the number of tankers needed to transport a certain quantity of oil. However, all of this reduction will occur in the areas of the Bay traversed by tankers going to Houston and Texas City. There will be an increase in tanker traffic in the channel leading into the Port of Galveston since it presently has no tanker traffic. The probability of an oil spill in some parts of the Bay will drop, but that in the channel leading to the Port of Galveston will rise. 8 See 532 F.Supp. at 1230. The probability analysis also examined the correlation between spill size and tanker size, data on spill frequency, and other data to conclude that the project will not increase the probability of a major oil spill in the Bay. See id. The dispersion model studied the effect of an average spill over only a 24-hour period. Yet from this model the FEIS concludes: The effects of any spill which might occur in the inshore or offshore area were analyzed based on the use of a model to predict the fate of the oil and on existing literature concerning the effects on different habitats. It is predicted that although any substantial spill might have a severe local impact,- the impact within Galveston Bay or along the Texas Gulf Coast would be minor and that impact would be reduced by prompt and adequate clean-up measures. FEIS Vol. I § 4.2.10.2, at 4-143 (emphasis added). The trial court noted that this “quite limited” dispersion model did not support this conclusion about “any spill.” See 532 F.Supp. at 1231. The trial court, despite the shortcomings of the FEIS in these analyses, nonetheless upheld the FEIS against all challenges. See id. at 1232. The Sierra Club does not repeat these challenges on appeal. *969 The Sierra Club also alleged that the FEIS failed to include a “worst case” analysis required by CEQ regulations. However, the court held that such an analysis would add nothing to the FEIS oil spill analysis and therefore was not required. See id. at 1234. It is this holding that the Sierra Club challenges on appeal. 2. The CEQ Worst Case Regulation We already have held that the Corps’ voluntary compliance with the new CEQ EIS-preparation regulations requires that this court review the Corps’ adherence to these regulations. One of those regulations requires that under certain conditions an EIS contain a “worst case” analysis. When an agency is evaluating significant adverse effects on the human environment in an environmental impact statement and there are gaps in relevant information or scientific uncertainty, the agency shall always make clear that such information is lacking or that uncertainty exists. (a) If the information relevant to adverse impacts is essential to a reasoned choice among alternatives and is not known and the overall costs of obtaining it are not exorbitant, the agency shall include the information in the environmental impact statement. (b) If (1) the information relevant to adverse impacts is essential to a reasoned choice among alternatives and is not known and the overall costs of obtaining it are exorbitant or (2) the information is relevant to adverse impacts, is important to the decision and the means to obtain it are not known (e.g., the means for obtaining it are beyond the state of the art) the agency shall weigh the need for the action against the risk and severity of possible adverse impacts were the action to proceed in the face of uncertainty. If the agency proceeds, it shall include a worst case analysis and an indication of the probability or improbability of its occurrence. 40 C.F.R. § 1502.22 (1981). The trial court held that the information available from a worst case analysis was not essential in this case and therefore beyond the “statutory minima” of NEPA. See 532 F.Supp. at 1234 (worst case analysis would not have “meaningfully illuminated the dangers or materially added to the decisionmaker’s awareness of the spectrum of consequences involved”). The Sierra Club attacks this holding on appeal. The parties make several arguments over the scope and applicability of this regulation. NEPA and the Worst Case Regulation. The Corps first argues that the catastrophic “worst case” oil spill analysis urged by the Sierra Club — involving a total cargo loss by a supertanker in the Bay — is beyond the statutory minima of NEPA and therefore need not be performed. To resolve this challenge, we turn first to the language of NEPA itself because “ ‘the starting point in every case involving construction of a statute is the language itself.’ ” Watt v. Alaska, 451 U.S. 259, 265, 101 S.Ct. 1673, 1677, 68 L.Ed.2d 80 (1981) (quoting Blue Chip Stamps v. Manor Drug Stores, 421 U.S. 723, 756, 95 S.Ct. 1917, 1935, 44 L.Ed.2d 539 (1975) (Powell, J., concurring)). We perceive some language in NEPA which may be said to endorse generally the concept of a worst case analysis. See 42 U.S.C. § 4331(b)(3) (responsibility of federal government to avoid “unintended” consequences of environmental use), 4332(C) (EIS to disclose all environmental impacts). However, because as noted earlier NEPA is a very general statute, see Kleppe, 427 U.S. at 421, 96 S.Ct. at 2735 (Marshall, J., concurring in part and dissenting in part), its literal language does not require a worst case analysis. Failing to find a justification for or delineation of the worst case analysis on the face of NEPA does not end the inquiry however. “This is because the plain meaning rule is ‘rather an axiom of experience than a rule of law, and does not preclude consideration of persuasive evidence if it exists.’ ” Watt, 451 U.S. at 266, 101 S.Ct. at 1677 (quoting Boston Sand Co. v. United States, 278 U.S. 41, 48, 49 S.Ct. 52, 53, 73 L.Ed.2d 170 (1928) (Holmes, J.)). Persuasive evidence that a *970 worst case analysis as envisioned by the Sierra Club is required by NEPA can be found in the case law interpreting NEPA as well as the history and interpretation of the CEQ worst case regulation. 9 Because NEPA is silent on the problem of uncertainty resulting from missing information, the courts have been forced to grapple with this issue case by case and have established a “rule of reason” approach. [0]ne of the functions of a NEPA statement is to indicate the extent to which environmental effects are essentially unknown. It must be remembered that the basic thrust of an agency’s responsibilities under NEPA is to predict the environmental effects of proposed action before the action is taken and those effects fully known. Reasonable forecasting and speculation is thus implicit in NEPA, and we must reject any attempt by agencies to shirk their responsibilities under NEPA by labeling any and all discussion of future environmental effects as “crystal ball inquiry.” “The statute must be construed in the light of reason if it is not to demand what is, fairly speaking, not meaningfully possible * * But implicit in this rule of reason is the overriding statutory duty of compliance with impact statement procedures to the “fullest extent possible.” Scientists' Institute for Public Information, Inc. v. Atomic Energy Commission, 481 F.2d 1079, 1092 (D.C.Cir.1973) (footnotes omitted). The EIS also must consider the probabilities of the occurrence of any environmental effects it discusses when it evaluates their environmental impact. E.g., Carolina Environmental Study Group v. United States, 510 F.2d 796, 799 (D.C.Cir.1975). Cf. Ethyl Corp. v. EPA, 541 F.2d 1, 18 (D.C. Cir.) (en banc), cert. denied, 426 U.S. 941, 96 S.Ct. 2662, 49 L.Ed.2d 394 (1976) (EPA must consider magnitude of risk of harm in rule-making); Reserve Mining Co. v. EPA, 514 F.2d 492, 519-20 (8th Cir.1975) (en banc) (magnitude of risk of harm must be considered when injunction abating a health hazard is sought). Notably, the unavailability of information, even if it hinders NEPA’s “full disclosure” requirement, should not be permitted to halt all government action. See, e.g., County of Suffolk v. Department of Interior, 562 F.2d 1368, 1378-79 (2d Cir. 1977), cert. denied, 434 U.S. 1064, 98 S.Ct. 1238, 55 L.Ed.2d 764 (1978); Jicarilla Apache Tribe of Indians v. Morton, 471 F.2d 1275, 1280 & n. 11 (9th Cir.1973). This is particularly true when information may become available at a later time and can still be used to influence the agency’s decision. See, e.g., Sierra Club v. Morton, 510 F.2d 813, 827-28 (5th Cir.1975). The evolution of NEPA’s “common law” on the issue of uncertainty reached the following state just prior to the CEQ’s issuance of its worst case analysis regulation: One of the costs that must be weighed by decisionmakers is the cost of uncertainty —i.e., the costs of proceeding without more and better information. Where that cost has been considered, and where the responsible decisionmaker has decided that it is outweighed by the benefits of proceeding with the project without further delay, the courts may not substitute their judgment for that of the decision-maker and insist that the project be delayed while more information is sought. Kleppe v. Sierra Club, 427 U.S. 390,410 n. *971 21, 96 S.Ct. 2718 [2730 n. 21], 49 L.Ed.2d 576 (1976). State of Alaska v. Andrus, 580 F.2d 465, 473-74 (D.C.Cir.), vacated in non-pertinent part sub nom. Western Oil & Gas Association v. Alaska, 439 U.S. 922, 99 S.Ct. 303, 58 L.Ed.2d 315 (1978) (emphasis original, footnote omitted). The CEQ’s worst case analysis regulation merely codifies these judicially created principles. The CEQ has declared, in accord with the requirement of Scientists Institute, that “the purpose of the analysis is to carry out NEPA’s mandate for full disclosure to the public of the potential consequences of agency decisions, and to cause agencies to consider these potential consequences when acting on the basis of scientific uncertainties or gaps in available information.” Forty Most Asked Questions Concerning CEQ’s NEPA Regulations, 46 Fed.Reg. 18,026,18,-032 (1981) (answer to question 20b). The CEQ regulation permits an agency to proceed after it weighs the costs of uncertainty. Alaska v. Andrus, 580 F.2d 465; County of Suffolk, 562 F.2d 1368; Jicarilla, 471 F.2d 1275. The probability analysis is intended to indicate the likelihood that the worst case will occur and to avoid undue emphasis on the disclosure of the worst case consequences. See Fed.Reg. 55,984 (1978). The probability requirement enunciated in Carolina and followed in other environmental law cases is incorporated in the CEQ probability analysis. Because the CEQ worst case regulation is in accord with the language and legislative history of NEPA and closely tracks NEPA case law, we conclude that it is not beyond the statutory minima of NEPA. Scope of the Worst Case Regulation. The CEQ regards the worst case analysis provision as very important and has resisted suggestions to weaken it. In its response to comments on the draft of section 1502.22, the CEQ said: Comments on § 1502.22: Incomplete or unavailable information. [Draft] [Section 1502.22 provided, among other things, that agencies prepare a worst case analysis of the risk and severity of possible adverse environmental impacts when it proceeds with a proposal in the face of uncertainty. This provision received strong support from many commenters. Several commenters expressed concern that this requirement would place undue emphasis on the possible occurrence of adverse environmental consequences regardless of how remote the possibility might be. In response, the Council added a phrase designed to ensure that the improbability as well as the probability of adverse environmental consequences would be discussed in worst case analyses prepared under this section. Section 1502.22 stated that if information is essential to a reasoned choice among alternatives and is not known and the costs of obtaining it are not exorbitant, the agency shall include the information in the environmental impact statement. Some commenters inquired into the meaning of the term “costs.” The Council intends for this word to be interpreted as including financial and other costs and adopted the phrase “overall costs” to convey this meaning. 43 Fed.Reg. 55,984 (1978), (emphasis original). 10 *972 The CEQ’s interpretation of its worst case regulation makes it quite clear that the Sierra Club’s catastrophic worst case is precisely what the CEQ intended: NEPA requires that impact statements, at a minimum, contain information to alert the public and Congress to all known possible environmental consequences of agency action. Thus, one of the federal government’s most important obligations is to present to the fullest extent possible the spectrum of consequences that may result from agency decisions and the details of their potential consequences for the human environment. .. . For example, if there are scientific uncertainty and gaps in the available information concerning the numbers of juvenile fish that would be entrained in a cooling water facility, the responsible agency must disclose and consider the possibility of the loss of the commercial or sport fishery. In addition to an analysis of a low probability/catastrophic impact event, the worst case analysis should also include a spectrum of events of higher probability but less drastic impact. Forty Most Asked Questions Concerning CEQ’s NEPA Regulations, 46 Fed.Reg. 18,-026, 18,032 (1981) (part of answers to ques-. tions 20a and 20b) (third emphasized word original). The CEQ’s use of the term “catastrophic” makes it clear that the CEQ worst case analysis regulation, if it applies to this case, requires the EIS for this project to discuss the “catastrophic impact” of a total cargo loss by a supertanker in the Bay. The Corps then must consider that impact and the probability of its occurrence in making its decision. It is well-established that the CEQ’s regulations and its interpretation of them are binding on this court and on agencies preparing EIS’s. “[Regulations promulgated by administrative bodies ... are usually given the force and effect of statutory law.” 3 C.D. Sands, Sutherland Statutory Construction § 65.05, at 174 (1974); see also K. Davis, 2 Administrative Law Treatise § 7.13 (1979). A unanimous Supreme Court in Andrus declared that “CEQ’s interpretation of NEPA is entitled to substantial deference.” 442 U.S. at 358, 99 S.Ct. at 2341. 11 By this the Court has meant that “ ‘the construction of a statute by those charged with its execution should be followed unless there are compelling indications that it is wrong.’ ” FCC v. WNCN Listeners Guild, 450 U.S. 582, 598, 101 S.Ct. 1266, 1276, 67 L.Ed.2d 521 (1981), (quoting Red Lion Broadcasting Co. v. FCC, 395 U.S. 367, 381, 89 S.Ct. 1794, 1801, 23 L.Ed.2d 371 (1969)). Therefore, if the CEQ interprets NEPA as mandating a particular worst case analysis, the CEQ’s interpretation is binding on this court and on agencies preparing EIS’s unless it is shown that the interpretation conflicts with the language or legislative intent of NEPA or the teachings of the Supreme Court. The Corps has presented no proof that the CEQ’s worst case analysis regulation as the CEQ has interpreted it conflicts with any of these authorities. Indeed, there is ample support for the regulation in the statute, its legislative history, and case law. The innovation the worst case analysis represents is precisely the type the Supreme Court unanimously approved in Andrus. While prior CEQ guidelines did not require a worst case analysis, the Court in Andrus held that a change in interpretation by the CEQ is no reason to decline to defer to its new interpretation. This innovation “occurred during the detailed and comprehensive process, ordered by the President, of transforming advisory guidelines into mandatory regulations applicable to all federal agencies.” 12 442 U.S. at 358, 99 *973 S.Ct. at 2341. During the CEQ’s comprehensive effort to review ten years of NEPA experience, it enacted the worst case analysis requirement. That regulation is sensible and addresses a widely recognized abuse by agencies — who previously had cloaked themselves in ignorance 13 — without preventing agency action. Uncertainty as to environmental consequences need not bar action as long as the uncertainty is forthrightly considered in the decisionmaking process and disclosed in the EIS. Our holding does not bar the Corps from issuing these permits, it only requires them to obey the regulation by considering and disclosing all the possible environmental impacts of their issuance. The Worst Case Regulation Applied. Having established that the Sierra Club’s worst case scenario — that of a total cargo loss by a supertanker in the Bay — is the one required by the worst case analysis regulation, we turn to the question of whether the regulation is triggered here. As with statutes, see, e.g., Waff, 451 U.S. at 265, 101 S.Ct. at 1677, interpretation of an administrative regulation must begin with its text. This regulation is quite straightforward. There are two ways in which it triggers the preparation of a worst case analysis: (1) “[i]f the information relevant to [significant] adverse impacts is essential to a reasoned choice among alternatives and is not known and the overall costs of obtaining it are exorbitant” or (2) if “the information relevant to [significant] adverse impacts is important to the decision and the means to obtain it are not known (e.g. the means for obtaining it are beyond the state of the art).” 40 C.F.R. § 1502.-22(b)(1), (2). The trial court apparently regarded this case as falling under section 1502.22(b)(1), “essential” unknown information, and held the missing information not essential. See 532 F.Supp. at 1233-34. Whether that holding is clearly erroneous we need not evaluate since it is apparent to us that this case qualifies under section 1502.22(b)(2), “important” unknown information “beyond the state of the art.” A total cargo loss by a supertanker is undoubtedly a significant adverse impact. No party can seriously question the importance of the analysis of such an oil spill to this permit decision. Indeed, the probabilities and consequences of oil spills are at the heart of this controversy. And all parties acknowledge that an analysis of a supertanker oil spill involving a total cargo loss beyond 24 hours after it occurs is beyond the state of the art. Thus it is clear that subsection (b)(2) is invoked in this case. Nonetheless, the trial court held and the Corps argues that a worst case analysis is not required for remote consequences. Here the trial court’s legal analysis and the Corps’ argument are incorrect. In the environmental law field, as in any field of administrative law, precedent must be carefully weighed because statutes and regulations may overrule judicial precedent. The worst case regulation is very recent in origin and is owed substantial deference by the courts. See Andrus, 442 U.S. at 356-58, 99 S.Ct. at 2340-41. Therefore the starting point for the legal analysis of this regulation is not precedent which predates it, but the language of the regulation itself. Of course, the regulation may not exceed the scope of its authorizing statute, but beyond this limitation a court must carefully inquire as to whether a precedential case has been overruled or limited by the regulation. *974 In its analysis of the worst case regulation, the trial court began, not with its language, but with principles drawn from case law which predated the regulation. There was no inquiry into the regulation’s language and purpose at all, nor was there any reference to the “substantial deference” ruling of the Supreme Court in Andrus. Thus the trial court began its legal analysis with incorrect premises. This led to numerous errors. As to remoteness, the triggering provisions do not use it as a criterion. The remoteness problem is instead addressed by mandating the preparation of a worst case analysis and indicating to .the decisionmaker “the probability or improbability of its occurrence.” 40 C.F.R. § 1502.22(b); see also 43 Fed.Reg. 55,984 (1978) (CEQ adds probability analysis to worst case regulation to avoid undue emphasis on remote possibilities). Thus the fact that the possibility of a total cargo loss by a supertanker is remote does not obviate the requirement of a worst case analysis in the FEIS. However, the decisionmaker may (and should) consider remoteness when using the worst case analysis in his decision-making process. The trial court and the Corps also misunderstand the role of “speculation” in preparing a worst case analysis. The trial court acknowledged that a worst case analysis is “somewhat speculative” by nature. See 532 F.Supp. at 1233; see also 40 C.F.R. § 1502.22(b) (analysis to be done when relevant information is not known). Yet, despite the fact that NEPA permits, even demands, “reasonable speculation,” see Scientists’ Institute, 481 F.2d at 1092, the trial court proceeded to require that the plaintiffs demonstrate a method for performing an “accurate” analysis that can support a “detailed discussion” of alternatives. See id. at 1234. Based on this demand, the court labeled the worst case analysis mere “guesswork” based on “uninformed speculation and conjecture.” See id. at 1233-34. This interpretation of the regulation and the plaintiff’s burden under it reads the regulation out of existence. The Sierra Club presented evidence demonstrating that it is possible to create an informative and useful worst case scenario that reasonably limits speculation. See, e.g., Trial Transcript Vol. VII at 455-59, 483-84. Thus, for instance, although the 24-hour dispersion model represents the state of the art in scientific methods, a worst case analysis could go beyond that state of the art based on known information about tides and currents in the Bay. See id. The preparation of a worst case analysis is not intended to and therefore cannot be held to the exacting standards imposed by the trial court because it goes beyond existing knowledge and methods of acquiring knowledge. There must, of course, be a base of information upon which to project past these limits, but the projections themselves cannot be subjected to the same rigorous scrutiny other information in the EIS must endure. The record reveals that the Sierra Club in this case adequately established the base of information upon which a worst case analysis could be premised. As the Sierra Club claims, this case presents precisely the type of situation for which the worst case regulation was designed. All parties agree that a total cargo loss could occur and could wreak catastrophic environmental damage in the Bay. While this damage is a “significant adverse effect,” there is considerable uncertainty about its likelihood, scope, and consequences; information on it is certainly important, if not essential, to the Corps’ decision, yet that information is beyond the state of the art. However, there is a body of data with which a reasonable worst case analysis can be made that is not unreasonably speculative. Remoteness does not bar a worst case analysis so founded and should instead be weighed by the Corps when it applies the worst case analysis in its decisionmaking process. The Corps must at least consider information relevant to a “significant” effect of a proposal, if that information is both “important” to that de *975 cisión and not based on “unreasonable speculation.” 14 The Corps, however, may not hide behind its ignorance of the worst case consequences and avoid confronting the costs of proceeding in the face of uncertainty. If the Corps decides to proceed with the issuance of these permits despite the uncertainty caused by this lack of information, the FEIS must be rewritten by the Corps to include a worst case analysis. C. FEIS ANALYSIS OF THE ENVIRONMENTAL IMPACT OF BULK CARGO ACTIVITIES The Sierra Club repeats its allegation made to the trial court that the FEIS failed to consider the environmental impacts of the bulk commodities activities associated with the deepening of the channels while attributing substantial benefits to those activities. See 532 F.Supp. at 1235-36. It argues that this approach skewed the FEIS’ and the Corps’ analysis of the proposed Galveston project, “presenting a biased and misleading picture of [its] value” in violation of NEPA and the Corps’ permit decisionmaking regulations. The trial court recognized this imbalance, see id. at 1235, yet held that the Corps was not legally required to evaluate these environmental impacts in the FEIS, see id. at 1236. Alternatively, the trial court held that the Sierra Club failed to show how this skewed cost-benefit analysis in the FEIS sufficiently disturbed the overall balance of costs and benefits as determined by the Corps to require reconsideration of the permits by the Corps. See id. at 1239-40. 1. Arguments of the Parties The Sierra Club advances three arguments in support of its claim that the failure of the FEIS to analyze the environmental impacts of the bulk commodities activities renders it deficient. Relying on Chelsea Neighborhood Association v. U.S. Postal Service, 516 F.2d 378 (2d Cir.1975), the Sierra Club first argues that since the FEIS relied on the benefits of the bulk cargo activities, the FEIS must also evaluate the adverse effects of those activities. Second, the FEIS should have assessed those adverse effects as cumulative effects of related proposals. Third, the adverse effects should have been examined as secondary or indirect effects of the multipurpose deep-water port. The Sierra Club also disagrees with the trial court’s alternative holding that this skewed cost-benefit analysis is insufficient to require a remand to the Corps for reconsideration of the permit decision. The Corps counters the first argument by asserting that Chelsea is distinguishable and has been limited by Kleppe. Therefore, it need not have evaluated in the FEIS any activity not part of the proposal being considered and, in any event, the discussion of adverse impacts was adequate. Answering the second attack, the Corps contends that the bulk terminals were not related proposals at all under Kleppe, but remote contingencies. In response to the third assertion, the Corps claims that the discussion of secondary impacts was legally and factually sufficient. The Corps supports the trial court’s holding that even if the FEIS cost-benefit analysis was skewed, it was irrelevant to the ultimate permit decision. Before addressing these arguments, it is necessary to briefly recite the facts necessary for an informed discussion. 2. Facts The FEIS contemplates that the deepened channel into the Port of Galveston will extend some one and one-half miles past the PELCO terminal solely to serve bulk commodities carriers. However, it addresses almost exclusively only the adverse environmental impacts of deepening the chan *976 nel and constructing the oil terminal and related facilities while virtually ignoring the adverse environmental impact of, for instance, more bulk carriers, expansion of existing bulk facilities, or the construction of new bulk facilities. Aggravating this asymmetry in the cost analysis is the consistent assumption throughout the FEIS that the modification and/or construction of bulk facilities will occur and that commodities tonnage handled as a result will grow dramatically. See, e.g., Vol. I §§ 4.1.-8.3.3 to 4.I.8.3.5. Based on this assumption, the FEIS cites many benefits that will flow from these terminals. Summarizing the benefits of these anticipated projects, the FEIS paints a rosy picture indeed: growth in commodity shipments aiding U.S. agriculture; greater efficiency in handling commodities; a reduced U.S. trade deficit; higher employment during construction and operation of these projects with minimal population growth and little impact on local services, housing, and the infrastructure; income and taxes from the projects beneficial to the local economy; the number of bulk container ships will actually drop initially due to their larger size and there will be no net increase in bulk container ship traffic despite a larger volume of commodities; and industrial development will be stimulated without interfering with local land use plans or violating the air pollution laws. Against these benefits are arrayed a few environmental and economic costs. The first claimed cost is that of dredging the channel and disposing of the dredged material. However, this includes the cost of the channel only, not the cost of the anticipated bulk terminals. Another cost is the reduction of bulk commodity traffic at other Texas ports, but only by 1.5% according to ■the FEIS, a minimal cost. The FEIS also recognizes that building these terminals will increase vehicular traffic on Pelican Island and its causeway. Finally, the FEIS acknowledges that the use of the channel by bulk commodity carriers could affect the oil spill probability analysis, but it declines to quantify that effect because the number of such ships is unknown. The Sierra Club has correctly pointed out several major deficiencies in this cost-benefit analysis. First, when the FEIS promoted the benefits of the projects, it used data on the quantity of commodities that would be shipped in the channel, yet when it came to determining whether the ships carrying these commodities could increase the risk of a collision and oil spill, the FEIS pleaded a lack of data. See supra note 8 and accompanying text. It seems to us quite obvious that since the FEIS discusses the size of commodity carriers expected to use the deepened channels and the anticipated quantity of commodities moving through the port, simple arithmetic should reveal the number of commodity ships that will share the channel with supertankers. Therefore, the costs of commodity transportation should include an analysis of its effect on oil spill probabilities. Second, while the pollution from other industrial development is mentioned briefly, there is no discussion of the air (particularly dust), water, and soil (through leaching) pollution costs of these commodities terminals. Third, absent also is data on the fire and explosion hazards of such terminals. These are not unlikely events if history is a guide, as Colonel Sigler himself acknowledged at trial. See Trial Transcript Vol. VI at 361-62. Fourth, the socio-economic costs are given only cursory and conclusory discussion. We agree with the conclusion of the district court that “it is assumed throughout the FEIS that these [commodities] projects will reach fruition, resulting in substantial economic benefits attributed directly to the instant project.... Curiously, however, the FEIS declines to consider the potential [adverse] environmental effects of this expansion . . . . ” 532 F.Supp. at 1235. With this factual background, we examine the arguments of the parties. 8. FEIS Cost-Benefit Analysis Introduction. In its various forms, 15 cost-benefit analysis is one of several useful *977 decisionmaking tools, 16 but like any tool it can be misused. For instance, the Corps itself has received criticism for its misuse of cost-benefit analysis in assessing its water projects. 17 In the 1960’s such criticism helped lead to the enactment of NEPA and at least two commentators have concluded that the Corps has responded to this criticism and NEPA by improving its incorporation of environmental values into its decisionmaking process. 18 In issuing permits for dredging and landfills, the subject of this lawsuit, the Corps’ historical record is much better. In 1968, before NEPA, the Corps revised its permit regulations declaring that in considering permit applications, it would weigh all relevant factors, including the effect of the proposed work on navigation, fish and wildlife, conservation, pollution, esthetics, ecology, and the general public interest. 33 C.F.R. § 209.120(d)(1) (1968). A subsequent congressional report praised the Corps for this policy and strongly encouraged its expansion and broader application. See House Committee on Government Operations, Our Waters and Wetlands: How the Corps of Engineers Can Help Prevent Their Destruction and Pollution, H.R.Rep. No. 917, 91st Cong., 2d Sess. (1970). 19 Aware of the value of cost-benefit analysis, but also cognizant of its limitations and particularly its misuses, we turn to the specifics of cost-benefit analysis under NEPA. NEPA. NEPA does not require an agency to take the action that is the most compatible with environment, nor does it permit a court to substitute its judgment for that of the agency on the wisdom of the action taken by the agency. See Strycker’s Bay, 444 U.S. at 227, 100 S.Ct. at 499 (citing, e.g., Kleppe); Vermont Yankee Nuclear Power Corp. v. NRDC, 435 U.S. 519, 98 S.Ct. 1197, 55 L.Ed.2d 460 (1978). “The primary purpose of the impact statement is to compel federal agencies to give serious weight to environmental factors in *978 making discretionary choices.” Monroe County Conservation Council, Inc. v. Volpe, 472 F.2d 693, 697 (2d Cir.1972); accord Kleppe, 427 U.S. at 490-10 & nn. 19, 21, 96 S.Ct. at 2730 & nn. 19, 21. NEPA therefore mandates at least a broad, informal cost-benefit analysis by federal agencies of the economic, technical, and environmental costs and benefits of a particular action; a formal monetary analysis is not required. 20 See South Louisiana Environmental Council, Inc. v. Sand, 629 F.2d 1005, 1011-12 (5th Cir.1980) (hereinafter cited as “SLEC”); 42 U.S.C. § 4332(B); 40 C.F.R. § 1502.23. The District of Columbia Circuit has explained: “Environmental amenities” will often be in conflict with “economic and technical considerations.” To “consider” the former “along with” the latter must involve a balancing process. In some instances environmental costs may outweigh economic and technical benefits and in other instances they may not. But NEPA mandates a rather finely tuned and “systematic” balancing analysis in each instance. [footnote from original:] Senator Jackson [NEPA’s principal sponsor] specifically recognized the requirement of a balancing judgment. He said on the floor of the Senate: “Subsection 102(b) requires the development of procedures designed to insure that all relevant environmental values and amenities are considered in the calculus of project development and decisionmaking. Subsection 102(c) establishes a procedure designed to insure that in instances where a proposed major Federal action would have a significant impact on the environment that the impact has in fact been considered, that any adverse effects which cannot be avoided are justified by some other stated consideration of national policy, that short-term uses are consistent with long-term productivity, and that any irreversible and irretrievable commitments of resources are warranted.” 115 Cong.Rec. (Part 21) 29055 (1969). Calvert Cliffs’ Coordinating Committee v. United States Atomic Energy Commission, 449 F.2d 1109, 1113 & n. 9 (D.C.Cir.1971). In order for an agency to carry out this broad systematic cost-benefit analysis, it is vitally important that the FEIS relied on by the agency fully and accurately disclose the environmental, economic, and technical costs associated with the project. The extent of that disclosure is the issue in this case. Cost-Benefit Analysis in this Case. The Second Circuit in Chelsea confronted an issue very similar to the one we face here. In Chelsea, the Post Office had acquired land in New York City for use in building a new facility. Airspace rights above the facility were granted to the City for use in building public housing. 516 F.2d at 381. The FEIS prepared by the Post Office was challenged, inter alia, on the grounds that the Post Office used the benefits of the housing project as a “selling point” and a principal reason to go forward with the facility while ignoring many of the disadvantages that would flow from it. Id. at 387. The court held the FEIS deficient on this basis. Id. at 388. The skewed cost-benefit analysis in Chelsea is remarkably similar to the skewed analysis found here by the trial judge, 532 F.Supp. at 1235, and substantiated by the record. The trial court in the instant case found the FEIS deficient due to its skewed cost-benefit analysis, and agreed with the holding of Chelsea as a “general proposition,” 532 F.Supp. at 1236, but held the legal rule of Chelsea had been limited to “actual proposals” due to the Supreme Court’s ruling *979 in Kleppe. Id. Therefore, the court,held that Chelsea did not compel disclosure of the costs associated with these bulk commodities activities because these activities were speculative possibilities, not actual proposals. It is this legal holding of the district court that the Sierra Club challenges on appeal. It being a legal holding, we make our own determination of its validity. See supra Part II.A.2. Kleppe addressed the question of whether the Department of the Interior had to prepare an EIS for a broad program to lease federal coal reserves in a four-state area in the western United States. The Court held that the program did not invoke NEPA as it was not a “proposal.” Only the program’s parts (issuing a lease, granting a permit) were proposals subject to NEPA. In so holding, the Court interpreted NEPA in the following way: The statute [NEPA], however, speaks solely in terms of proposed actions; it does not require an agency to consider the possible environmental impacts of less imminent actions when preparing the impact statement on proposed actions. Should contemplated actions later reach the state of actual proposals, impact statements on them will take into account the effect of their approval upon the existing environment; and the condition of that environment presumably will reflect earlier proposed actions and their effects. 427 U.S. at 410 n. 20, 96 S.Ct. at 2730 n. 20; see Environmental Defense Fund v. Marsh, 651 F.2d 983, 999 (5th Cir.1981). The Corps’ and the trial court’s reliance on Kleppe is misplaced. In Kleppe, the Court was concerned with the point at which NEPA’s requirements are triggered by a definitive proposal. In analyzing that issue the Court held that environmental impacts need not be evaluated for actions that are not imminent. Under Kleppe then, the Corps need not have prepared an EIS on the bulk commodities terminals if they were not imminent. However, once the Corps chose to trumpet the benefits of bulk cargo activities in the EIS as a “selling point” for the oil project, it rendered a decision that these activities were imminent. NEPA therefore requires full disclosure and analysis of their costs. See Chelsea, supra. The Corps cannot tip the scales of an EIS by promoting possible benefits while ignoring their costs. Simple logic, fairness, and the premises of cost-benefit analysis, let alone NEPA, demand that'a cost-benefit analysis be carried out objectively. There can be no “hard look” at costs and benefits unless all costs are disclosed. If an agency were permitted to cite possible benefits in order to promote a project, as the Corps has done here, yet avoid citation of accompanying costs by hiding behind Kleppe, the cost-benefit analysis in the EIS would be reduced to a sham: such a “cost-benefit analysis” would always be tipped in favor of benefits. Kleppe cannot be used to defend a skewed cost-benefit analysis; it was concerned solely with determining when an EIS with its informal cost-benefit analysis must be prepared. Once that threshold is crossed, the analysis must be objective. This case is beyond the threshold at issue in Kleppe, since the Corps had to prepare an EIS on the superport project. The issue here is one not discussed in Kleppe: once an EIS is required, can the costs of any claimed benefits be ignored? This issue was correctly resolved in Chelsea, and as NEPA required it must, Chelsea answered it negatively. We therefore hold that the district court reached an erroneous legal conclusion in failing to apply the principle of Chelsea to this skewed cost-benefit analysis. We follow Chelsea, apply it to the trial court’s well-substantiated factual conclusion that the FEIS is skewed, and hold the FEIS deficient under NEPA. 21 *980 4. Reconsideration of the Permit Decision by the Corps Although we hold that the FEIS is deficient under NEPA and Chelsea, that holding does not necessarily require a reversal of the Corps’ decision to issue the permits and a remand for reconsideration. The adequacy of the FEIS and the procedural integrity of the Corps’ permit decision are two different, although related, issues. The trial court held that even if the FEIS was deficient, it did not sufficiently taint the Corps’ permit decision as to require a decision to reverse and remand. See 532 F.Supp. at 1239-40. We now turn to this holding of the trial court. We note briefly that a different standard of review is applied to this issue than was applied when analyzing the adequacy of the FEIS. See supra Part II.A.l. We review the Corps’ decision to issue the permits for its conformance to Corps regulations containing both NEPA and non-NEPA criteria. See id.; Marsh, 651 F.2d at 1002; 33 C.F.R. § 320 et seq. In addition, it should be noted that there are two parts to the trial court’s alternative holding: the first is the applicable legal standard, a determination which in no way binds us, and the second is the application of that standard to the facts. The trial court relied on SLEC, 629 F.2d 1005, for its legal standard when it held that the Sierra Club had not shown that the economic benefits of the commodity activities touted by the FEIS sufficiently disturbed the Corps’ overall decision to issue the permits. See 532 F.Supp. at 1239 (citing 629 F.2d at 1013-14). SLEC, however, is not the correct legal standard against which to assess the effect of the deficient FEIS on the Corps’ permit decision. SLEC involved an EIS prepared by the Corps, for one of its navigation projects that had already been authorized and funded by Congress. The plaintiffs attacked the FEIS on several grounds, one of which was the use of certain data to calculate economic benefits outlined in the EIS. See id. at 1013-15. This court affirmed the holding of the trial court that the selection of such data to calculate economic non-environmental benefits is left to agency discretion subject to “extremely limited” judicial review since it is “extremely tangential to the concerns expressed in NEPA.” Id. at 1014. This discretion was not abused in SLEC because the plaintiffs had not shown that the data affected the environmental analysis in the EIS. See id. at 1013-14. An important additional justification for this limited judicial role was the fact that Congress had already authorized and funded the project. See id. at 1015. Omnipresent separation of powers considerations therefore prohibited the court from reviewing the project’s economic justification unless that justification was “so flawed” as to “grossly distort[ ] the environmental considerations.” Id.; see also id. at 1011 (“assessment of purely economic costs and benefits is . . traditionally .. . the province of Congress”); 1013 (“only if Congress was mislead by the inclusion [in the EIS] of . . . erroneous benefits in its consideration of environmental consequence may [the court order reworking of the EIS]”). This case presents substantially different facts and legal standards from those found in SLEC. First, here the Sierra Club has not challenged the Corps’ economic benefit calculations, the issue in SLEC, for bulk commodities activities. Instead it attacks the exclusion of the environmental costs of those activities from the FEIS. This exclusion is hardly “tangential” to the concerns of NEPA as was the economic data selection in SLEC and therefore it is not defensible as an exercise of agency discretion. The FEIS is not deficient because it includes economic benefits, but because it excludes environmental costs. That exclusion strikes at the heart of NEPA. See Chelsea, supra. Further, in SLEC, separation of powers considerations precluded judicial review of the congressional decision to build the *981 project; the court could order only the reworking of the FEIS if it was found deficient. In the instant case, however, private permits are at stake and the Corps, not Congress, is the ultimate decisionmaker. That decision is governed by NEPA and Corps regulations. In sharp contrast to the situation in SLEC, the Corps’ adherence to those standards is clearly reviewable by this court under the APA. Therefore, if the deficient FEIS led to a permit decision in violation of NEPA or Corps regulations, that decision must be reconsidered in light of a corrected FEIS. We thus reach an issue not reached in SLEC since it was foreclosed by the separation of powers principle: did the exclusion sufficiently taint the permit decision so as to require its consideration by the Corps? The correct legal standard for measuring the impact of the deficient FEIS on the permit decision is found not in SLEC but in the Corps regulations on issuing permits. The decision to be analyzed is Colonel Sigler’s written “findings of fact” which contains the required “public interest” analysis justifying the permit decision. At this point, we face a fork in the road. Generally, a remand to the trial court for reconsideration of the evidence in light of the correct legal standard would be appropriate here. The alternative is for this court to apply the correct legal standard to the record evidence for a determination of whether the Corps permit decision must be reversed and remanded for reconsideration. This alternative is a limited one, as the Supreme Court has explained: [Wjhere findings [by the district court] are infirm because of an erroneous view of the law, a remand is the proper course unless the record permits only one resolution of the factual issue. Pullman-Standard, 456 U.S. 273, 292, 102 S.Ct. 1781, 1792, 72 L.Ed.2d 66 (1982) (citing Kelley v. Southern Pacific Co., 419 U.S. 318, 331-32, 95 S.Ct. 472, 479, 42 L.Ed.2d 498 (1974)). We choose the alternative path because, as our discussion infra demonstrates, our close review of the record reveals that there is “only one possible resolution of the factual issue” of whether the deficient FEIS tainted the Corps’ permit decision. The choice of the alternative path is bolstered by three additional reasons— one peculiar to environmental litigation generally and two peculiar to this case. First; the most important, this and other courts have noted that protracted litigation in environmental cases can kill projects by delay. See, e.g., SLEC, 629 F.2d at 1014 n. 9; see also Strycker's Bay, 444 U.S. at 223, 100 S.Ct. at 221 (noting protracted nature of environmental litigation); Vermont Yankee, 435 U.S. at 557-58, 98 S.Ct. at 1218 (noting same). If we were to remand to the trial court for application of the correct legal standard, an appeal to us, and perhaps beyond, would undoubtedly follow, entailing at least several more months of litigation. As the appellees have repeatedly told us, construction has been delayed pending the resolution of this litigation and further delay could kill the project. These considerations weigh strongly in favor of resolution of this issue by this court at this time. Second, we already have ordered a remand to the Corps for the performance of a worst case oil spill analysis. For the Corps to proceed with that obligation only to get the FEIS possibly returned again later on this issue, or for the Corps to delay that undertaking until this issue is resolved by the district court, and by us on a likely appeal, is a waste of time and resources. Such piecemeal resolution of these issues should be avoided. Our decision to apply the correct legal standards to the evidence ourselves is supported and eased by a third consideration— the record evidence on this issue is almost exclusively documentary. A thorough review of the record reveals that very little oral testimony was given on the role of the bulk commodities projects in the permit decision. And what little there was supports the conclusions we reach from the documentary evidence. Therefore, the expertise of the trial court in weighing the testimony of live witnesses is not crucial to the resolution of this issue. Cf. Sicula Oceánica, 413 F.2d at 1333 (factfinding of trial court reviewed *982 under “clearly erroneous” standard; however, appellant’s burden of proving findings “clearly erroneous” reduced when factfinding is based solely on documentary evidence). The Corps regulations on the “public interest” analysis are quite specific as to the procedure that must be followed in issuing a permit: The decision whether to issue a permit will be based on an evaluation of the probable impact of the proposed activity and its intended use on the public interest. Evaluation of the probable impact which the proposed activity may have on the public interest requires a careful weighing of all those factors which become relevant in each particular case. The benefit which reasonably may be expected to accrue from the proposal must be balanced against its reasonably foreseeable detriments. The decision whether to authorize a proposal, and if so, the conditions under which it will be allowed to occur, are therefore determined by the outcome of the general balancing process .... That decision should reflect the national concern for both protection and utilization of important resources. All factors which may be relevant to' the proposal must be considered; among those are conservation, economics, aesthetics, general environmental concerns, historic values, fish and wildlife values, flood damage prevention, land use, navigation, recreation, water supply, water quality, energy needs, safety, food production, and, in general, the needs and welfare of the people. No permit will be granted unless its issuance is found to be in the public interest. 33 C.F.R. § 320.4(a)(1) (emphasis added). Seeking to discover whether the deficient FEIS prevented the Corps from adhering to these standards, we begin our “searching and careful” inquiry into the facts, as required by Overton Park, see 401 U.S. at 416, 91 S.Ct. at 823, with the most important piece of documentary evidence demonstrating the effect or lack of effect of the skewed FEIS cost-benefit analysis on the permit decision, Colonel Sigler’s “findings of fact.” The Colonel first summarized five “criteria and specific policy considerations regarding the public interest” that played a role in his decision. 22 These considerations were quite broad and overlap. Of these considerations, number 5 was given the most attention in the Colonel’s summary with a list of “minor” environmental concerns covering two full pages in a twenty-five-page document. See Plaintiffs’ Exhibit 1 at 2-4. Consideration 2 on alternatives to the superport was affected by the skewed cost-benefit analysis since the failure to disclose all of the superport’s environmental costs made it appear more' attractive vis-a-vis the alternatives than it actually was. Considerations 4 and 5 involved evaluation of environmental costs that also must have been underestimated due to reliance on the skewed FEIS. From the Colonel’s summary of his decisionmaking process, we easily conclude that environmental issues played a substantial role in the decision. *983 This conclusion is reinforced by Part IV of the document, “Relevant Major Issues.” See Plaintiffs’ Exhibit 1 at 10-22. In this section, the Colonel listed five such issues in the following order: environmental issues; public health, welfare, and safety; socioeconomic effects; need for facility; and alternatives. Id. at 10-22. The skewed FEIS analysis prevented discussion of commodities activities costs in each of these areas. None of the many environmental, health, safety, or socioeconomic costs of bulk facilities activities are discussed, yet the benefits of bulk commodities activities are cited. See, e.g., id. at 19, 22. This is particularly clear in the discussion of alternatives. There, the Colonel relied on a Department of Transportation study which concluded that while both projects were in the national interest, the onshore project had an edge due to its “crude oil/dry bulk cargo” flexibility. See id. at 22. Thus this section of the document also mirrored the flaws of the FEIS. Another piece of relevant documentary evidence is the text of Colonel Sigler’s press conference announcing the permit decision. See Plaintiffs’ Exhibit 2. This seven-page document was distilled from the findings of fact and, according to Colonel Sigler’s testimony at trial, “points to most of the more important reasons [for granting the permit] that were stated in the findings of fact.” Trial transcript Vol. VI at 164; see also id. at 324-25, 331. The press conference document covered five “major considerations which have become identified as the more significant aspects of the proposed project.” Plaintiffs’ Exhibit 2 at 4-7. The Colonel listed them as health and safety, oil spills, other environmental effects, and two considerations labelled “benefits,” supplying United States energy needs, and more efficient handling and increased volume of bulk commodities. Here again the effects of the skewed EIS are patent. “Health and safety” and “other environmental effects” should include those created by bulk commodities activities, yet none are mentioned while one of the two benefits cited is created by such activities. If more evidence need be documented, there are other references in the record which reveal that this project had a dual-purpose — aid of both the oil import and bulk commodities trade. See, e.g., FEIS App. I §§ 2.0-2.4, at ii, A-6 to -22 (project has two purposes — “oil needs and bulk commodities trade”). One purpose, oil, may have been more important than the other, bulk commodities, but the latter purpose was nonetheless quite significant. Overall then, the inclusion of dry bulk benefits and the exclusion of countervailing environmental costs played a large and significant role in Colonel Sigler’s decision, just as they did in the FEIS. The record overwhelmingly refutes the Corps’ claim that increased bulk commodities traffic “plays a minor role in the FEIS” and in the Corps’ final decision. Brief of the Corps at 38-39. It is apparent to us that the skewed FEIS tainted the Colonel’s permit decision-making process preventing the “careful weighing of ail [relevant] factors” necessary in the “general balancing process” required by Corps regulations. Important and significant environmental costs were omitted from the FEIS and therefore were not considered by the Colonel in his permit decision. Since the Corps’ decision was not made “according to law,” it must be reversed. The FEIS must be redone to consider the costs of bulk commodities activities it extols, and the permit decision by the Corps is reversed and remanded to the Corps for reconsideration in light of the corrected FEIS. We, of course, may not and do not express an opinion on the merits of the permits. The Corps may reach the same decision, but only after considering an adequate FEIS. If that decision is procedurally adequate under NEPA and Corps regulations, it will pass judicial muster. See Chelsea, 516 F.2d at 389. D. CORPS’ CONSIDERATION OF OFFSHORE PORT ALTERNATIVES. While conceding that the discussion of offshore port alternatives in the FEIS “met the statutory minima of NEPA,” the *984 appellants press us to conclude that Colonel Sigler did not adequately consider such non-environmental factors in his final decision. Specifically, they urge that he failed to factor into his decision the possibility of a smaller, single-point offshore terminal proposed after the collapse of an earlier, more expansive, project that he did consider. The testimony of Colonel Sigler convinces us, however, that he adequately considered a range of such alternatives and that we cannot, consonant with the standard of review enunciated in Part II.A.l. above, interfere with his decision on this head. In so concluding, we are mindful of the Supreme Court’s caution, voiced in a similar case, that ... If upon the coming down of the order litigants might demand rehearings as a matter of law because some new circumstance has arisen, some new trend has been observed, or some new fact discovered, there would be little hope that the administrative process could ever be consummated in an order that would not be subject to reopening. Vermont Yankee Nuclear Power Corp. v. N.R.D.C., 435 U.S. 519, 554-5, 98 S.Ct. 1197, 1217, 55 L.Ed.2d 460 (1978), quoting ICC v. Jersey City, 322 U.S. 503, 514, 64 S.Ct. 1129, 1134, 88 L.Ed. 1420 (1944). III. CONCLUSION The decision of the trial court upholding the FEIS and the Corps permit decision is affirmed in part, reversed in part, and the ease is remanded. The Corps must rework the FEIS to correct the deficiencies noted in this decision and must reconsider its permit decision in light of the rewritten FEIS. AFFIRMED IN PART, REVERSED IN PART, AND REMANDED. Footnotes [1] Nevertheless, under federal law, the Department of the Army is charged with regulating construction, dredging, filling, or disposing of material in United States waters. See, e.g., 33 U.S.C. §§ 403, 1344, 1413. Therefore, permits from the Corps of Engineers were required before any dredging or construction could begin. [2] These enormous ships are described in N. Mostert, Supership (1974). [3] The manner of preparation of the EIS as documented by the record leads us to question its objectivity, the extent to which it represents the Corps’ independent evaluation of the relevant information, and the reasonableness of the Corps’ reliance on it in reaching its decision to issue the permits. Under NEPA, the federal official responsible for a major federal action, such as issuing these permits, must prepare the EIS. See 42 U.S.C. § 4332(C). Corps regulations require the District Engineer (DE) processing the permit applications to obtain from the applicant information the DE considers necessary to prepare the EIS. See 33 C.F.R. § 325.4(b)(3). In addition, the Corps in this case voluntarily adhered to the new mandatory Council on Environmental Quality (“CEQ”) regulations on EIS preparation. We hold infra that this voluntary adherence requires that the Corps’ preparation of the EIS be scrutinized under those regulations which insist that an “agency shall independently evaluate the information submitted and shall be responsible for its accuracy.” 33 C.F.R. § 1506.5(a). The intent of this regulation is that “acceptable work not be redone, but that it be certified by the agency.” Id. In February 1978, the applicants informed the Corps that they had hired a private consulting firm to design the deepened channel and to prepare an environmental assessment report (EAR) for the Corps’ use in preparing the EIS. See Admin. Record Vol. II at 245-46, 250-51. Cooperation ensued between the Corps, the consulting firm, and the applicant’s law firm, with meetings and exchanges of correspondence. See, e.g., id. at 136-37, 148-58, 234-38, 241-43. In fact, it appears from the record that the draft EAR (“DEAR”) produced by the consulting firm was intended to do more than merely provide the Corps with the relevant environmental information; more accurately, the consulting firm essentially prepared the EIS. The Corps had only brief opportunities to review the parts of the DEAR it received before it issued the final EAR (“FEAR”). See, e.g., id. at 164 j[ 2. The last of a series of DEARs was submitted to the Corps on November 20, 1978; meetings were held on the project in mid-December; and the FEAR was issued by the Corps in January of 1979. Following issuance of the DEAR, the applicant’s law firm and consulting firm continued to influence the process by leading negotiations with various agencies concerning special conditions attached to the permits. See, e.g., id. Vol. IV at 225; Vol. V at 155; Vol. XII at 156. The record also suggests that the consulting firm, not the Corps, may have prepared the DEIS and the FEIS. See id. Vol. V at 16, 21. *963 Although there was a designated EIS preparer in the Corps’ Galveston office, see id. Vol. VI at 5; see also id. Vol. VII at 248-49, it appears that his role was one of coordinating and overseeing work actually done primarily, if not exclusively, by the consultant hired by the applicants and by the various other consultants hired as subcontractors. The DEIS and FEIS call this person a “Study Manager” and list five other Corps employees as “Reviewers.” See FEIS Vol. II at 6-1; DEIS at 6-1. Both also declare that the FEAR was the “primary source of information” used in their preparation of the DEIS and FEIS. See FEIS Vol. II at 6-1; DEIS at 6-1. Finally, both documents have a “List of Preparers” which contains the names of the consultant’s employees who worked on the DEAR, FEAR, DEIS and FEIS. See FEIS Vol. II at 6-3 to -5; DEIS at 6-3 to -5. This record leaves us with the distinct impression that most, if not all, of the preparation of the DEIS and FEIS was done by the private consulting firm hired by the applicants. While the Corps ostensibly was supervising, we are concerned by the relatively brief time the Corps was given to review and comment on the documents, and the degree of thoroughness of the Corps’ review and supervision. The role of the private firm in the preparation of the DEIS and the FEIS is particularly troubling in this case because the consulting firm also had a stake in the project which it was evaluating. Although the conflict of inter-. est itself may not have been illegal under the old CEQ advisory guidelines on EIS preparation, Corps rubberstamping of a consultant-prepared FEIS is. See Sierra Club v. Lynn, 502 F.2d 43, 58-59 (5th Cir.1974), cert. denied, 421 U.S. 994, 95 S.Ct. 2001, 44 L.Ed.2d 484; 422 U.S. 1049, 95 S.Ct. 2668, 45 L.Ed.2d 701 (1975) (permitting a “financially interested private contractor” to participate in EIS preparation, but barring agency abdication of its duties by “reflexively rubberstamping a statement prepared by others”). In addition, the new man-" datory CEQ regulations, are especially disapproving of the use of contractor-prepared statements when the contractor may have a conflict of interest. In fact, the regulations require contractors involved in the preparation of a statement to execute a “disclosure statement ... specifying that they have no financial interest in the outcome of the project.” 33 C.F.R. § 1506.5(c). This regulation, as set forth by the CEQ, “is designed ... to minimize the conflict of interest inherent in the situation of those outside the government coming to the government for money, leases or permits while attempting impartially to analyze the environmental consequences of their getting it.” 43 Fed.Reg. 55, 987 (1978). We realize that the preparation of an FEIS is a mammoth task and that CEQ and Corps regulations permit the participation of private consultants. These regulations recognize the reality that private consultants play an important, if sometimes troubling, role in modern government. See, e.g., D. Guttman & B. Willner, The Shadow Government (1976). And, of course, administrative agencies have discretion in the performance of their duties. See, e.g., South Louisiana Environmental Council, Inc. v. Sand, 629 F.2d 1005, 1014 (5th Cir.1980). Nonetheless, an agency may not delegate its public duties to private entities, see Lynn, 502 F.2d at 59, particularly private entities whose objectivity may be questioned on grounds of conflict of interest. We have taken a hard look at the preparation of the EIS, and the problems we have noted lead us to question its objectivity and whether it represents the independent judgment of the Corps. We express no legal opinion on whether the Corps improperly rubberstamped an FEIS prepared by someone else or whether a violation of 33 C.F.R. § 1506.5 occurred since those issues were not specifically presented for our review. See 5 U.S.C. § 706. However we do take note of these problems and counsel caution in the future in the broader context of our mandate to review the record to ensure that the Corps “in good faith objectively [took] a hard look at the environmental consequences of the project.” Isle of Hope Historical Association, Inc. v. United States Army Corps of Engineers, 646 F.2d 215, 220 (5th Cir.1981) (per curiam) (adopting opinion of district court). [4] The other plaintiffs are the Environmental Defense Fund, Inc. (EDF), Galveston Bay Conservation and Preservation Association, Gulf Coast Fishermen’s Environmental Defense Fund, and Texas Environmental Coalition. For purposes of our discussion, the phrase “Sierra Club” encompasses all the plaintiff-appellants. [5] The original federal defendants were James M. Sigler, District Engineer, United States Army Corps of Engineers, Galveston District; Joseph K. Bratton, Chief of Engineers, Department of the Army; James G. Watt, Secretary of the Interior; F. Eugene Hester, Director, LTnited States Fish and Wildlife Service (FWS); and *964 David M. Harris, Field Supervisor, Galveston Field Office, FWS. For the purposes of our discussion, the term “Corps” includes all defendant-appellees. [6] See 115 Cong.Rec. 40,416, 40,419 (1969). [7] The Corps also urges upon us the holding in National Center for Preservation Law v. Landrieu, 496 F.Supp. 716 (D.S.C.), aff’d per curiam, 635 F.2d 324 (4th Cir.1980), in support of its argument. However, Landrieu is distinguishable and inapposite. In Landrieu, the agency, as with the Corps here and the agency in North Slope, had filed its DEIS before the effective date of the new regulations and was therefore exempt from them. Id. at 737. There the similarity between our case and North Slope on one hand, and Landrieu on the other, ends because the agency in Landrieu apparently did not voluntarily adhere to the new CEQ regulations. See id. Therefore the court evaluated the agency’s duty to adhere to them under the “fullest extent practicable” standard found in 33 C.F.R. § 1506.12(a) and held that the determination of whether it was practicable to adhere was a question of administrative discretion, apparently not reviewable by a court. See id. The court reinforced its decision by erroneously stating that the CEQ regulations were merely advisory and not mandatory. See id. The Supreme Court and the regulations themselves declare otherwise. See Andrus, 442 U.S. at 356-58, 99 S.Ct. at 2340; 40 C.F.R. § 1500.3; see also Exec. Order No. 11,991, 42 Fed.Reg. 26,967 (1977). The Corps here declared without reservation that it was following the new CEQ regulations, thus not invoking the “fullest extent practicable” standard. Landrieu is inapposite because it did not reach the issue in our case: the effect of voluntary adherence to CEQ regulations on judicial review of agency actions. [8] This analysis does not include increased ship traffic in other commodities spurred by the deeper channel, such as bulk carriers of coal and grain. Because of this deficiency, the probability analysis underestimates the risk of a collision which in turn underestimates the risk of an oil spill. We hold infra that the Corps’ plea of lack of data on this increased traffic is unacceptable. [9] There is also some support for a worst case analysis in NEPA’s legislative history which illustrates congressional awareness of man’s limited understanding of the environmental consequences of his actions. See Senate Comm, on Interior and Insular Affairs, National Environmental Policy Act of 1969, S.Rep. No. 296, 91st Cong., 1st Sess. 9-10 (1969). Despite this awareness, Congress mandated full disclosure of environmental consequences, apparently leaving to the courts and the CEQ the determination of the extent of the mandate when information is missing and expensive, or impossible, to obtain. There has long been concern that federal agencies fail to take uncertainty into account in their decisions, particularly in the environmental area. See, e.g., F. Anderson, NEPA in the Courts: A Legal Analysis of the National Environmental Policy Act 214-15 (1973); J. Krutilla & A. Fisher, The Economics of Natural Environments 39 (1975). [10] The provision has been highly praised: The imposition of an affirmative obligation to go beyond the limits of currently available information when necessary for an informed decision is a major innovation and improvement in the EIS process. Current ecological knowledge is still in many instances quite limited, and a duty to develop new information will serve both to expand the frontiers of environmental knowledge and prevent agencies from hiding behind this ignorance of a project’s true environmental ramifications. Comment, New Rules for the NEPA Process: CEQ Establishes Uniform Procedures to Improve Implementation, 9 Envt’l L.Rep. (ELI) 10,005, 10,008 (1979). Other commentators have echoed this praise. See, e.g., 3 Harv.Envt’l L.Rev. 347, 374 (1980) (worst case analysis has two purposes: “to force a systematic balancing of the conflicting needs for action and complete information, and to remove the potential for ignoring unfavorable information — a loophole otherwise availa *972 ble to agencies that want to proceed with undesirable courses of action”). [11] Thus overruling our previous declaration that the “CEQ does not have the authority to prescribe regulations governing compliance with NEPA.” Hiram Clarke Civic Club, Inc. v. Lynn, 476 F.2d 421, 424 (5th Cir.1973). [12] For a description of this process, see 43 Fed.Reg. 55,980431 (1978); Yost, Streamlining *973 NEPA —An Environmental Success Story, 9 B.C. Envt’l Aff.L.Rev. 507, 508-09 (1981-82). [13] The seriousness of this problem is revealed in the short history of this regulation. In 1980, the CEQ did a preliminary review of agency compliance with its new regulations. See CEQ, Talking Points on CEQ’s Oversight of Agency Compliance with NEPA Regulations (1980). This review led to the following conclusion on the worst case analysis regulation: “EIS’s rarely even address this requirement. The need to address this and include a worst case analysis is especially critical for many new energy development projects where considerable information is not available.” Quoted in Liebesman, The Council on Environmental Quality’s Regulations to Implement the National Environmental Policy Act — Will They Further NEPA’s Substantive Mandate?, 10 Envt’l L.Rep. (ELI) 50,039, at 50,049 (1980). [14] We scarcely need add that while remoteness of a possible occurrence does not permit disregarding it in such circumstances as these, where a real possibility of the occurrence has been proved and a data base for evaluating its consequences established, the Corps need not concern itself with phantasmagoria hypothesized without a firm basis in evidence and the actual circumstances of the contemplated project, or with disasters the likelihood of which is not shown to be significantly increased by the carrying out of the project. [15] “Cost-benefit analysis” can take many forms. It varies from a formal analysis in *977 which all costs and benefits are quantified in an identical unit of measurement, usually dollars, and compared, to an informal analysis where costs and benefits are identified, quantified if possible, and balanced. See Kasper, Cost-Benefit Analysis in Environmental Decisionmaking, Geo. Wash. L.Rev. 1013, 1013-15 (1970). Under the Council on Environmental Quality’s regulations governing the implementation of NEPA, an EIS need not include a formal cost-benefit analysis and in fact a more informal analysis is preferred “when there are important qualitative considerations.” 40 C.F.R. § 1502.-23. At a minimum, the EIS “should at least indicate those considerations, including factors not related to environmental quality, which are likely to be relevant and important to a decision.” Id. [16] Apparently, the first appearance of cost-benefit analysis in a federal statute was in the Flood Control Act of 1936, ch. 688, § 1, 49 Stat. 1570 (1936) (codified at 33 U.S.C. § 701a (1976)). Since 1976, it has become a common tool in legislative and administrative decision-making and, not unexpectedly, a source of legal controversy. See, e.g., American Textile Mfgrs. Inst., Inc. v. Donovan, 452 U.S. 490, 101 S.Ct. 2478, 69 L.Ed.2d 185 (1981) (Secretary of Labor not required to perform cost-benefit analysis for health and safety standards issued under § 6(b)(5) of the Occupational Health and Safety Act, 29 U.S.C. § 655(b)(5), Congress balanced costs and benefits); Exec. Order No. 12,291, 46 Fed.Reg. 13,193 (1981) (cost-benefit analysis required for all regulatory action). Its growth in popularity has paralleled the rise of the jurisprudential school endorsing economic analysis of law. See, e.g., R. Posner, Economic Analysis of Law (2d Ed. 1977). [17] See, e.g., U.S. Comptroller General, General Accounting Office, An Overview of Benefit-Cost Analysis for Water Resources Projects— Improvements Still Needed (1978) (CED 78 127); Note, Environmental Impact Assessment for Water Resource Projects: The Army Corps of Engineers, 45 Geo.Wash.L.Rev. 1095 (1977). [18] See S. Mazmanian & J. Nienaber, Can Organizations Change?, 2-3 (1979). [19] Despite whatever progress has been made in ending the misuses of cost-benefit analysis, particularly in environmental decisionmaking, there remains a sizable group of scholars who question its very applicability to environmental decisionmaking. See, e.g., Baram, Cost-Benefit Analysis, An Inadequate Basis for Health, Safety and Environmental Decisionmaking, 8 Ecology L.Q. 473 (1980); Jaffe, Benefit-Cost Analysis and Multi-Objective Evaluation of Federal Water Projects, 4 Harv.Envt’l L.Rev. 85 (1980); Rodgers, Benefits, Costs, and Risks: Oversight of Health and Environmental Decisionmaking, 4 Harv.Envt’l L.Rev. 191 (1980). [20] A leading, early case described NEPA as “[a]t the very least, ‘.. . an environmental full disclosure law.’ ’’ Monroe County, 472 F.2d at 697, (quoting Environmental Defense Fund, Inc. v. Corps of Engineers, 325 F.Supp. 749, 759 (E.D.Ark.1971), injunction vacated and case dism’d, 342 F.Supp. 1211 (E.D.Ark.), aff’d, 470 F.2d 289 (8th Cir.), injunction denied, 412 U.S. 931, 93 S.Ct. 2749, 37 L.Ed.2d 160 (1972)). Subsequent judicial opinions have declared that full disclosure is but one of many NEPA requirements. See, e.g., Public Serv. Co. v. NRS, 582 F.2d 77, 85 n. 12 (1st Cir.), cert. denied, 439 U.S. 1046, 99 S.Ct. 721, 58 L.Ed.2d 705 (1978). [21] Because we hold that the Corps must include the cost of bulk cargo activities in the FEIS under Chelsea, we need not reach and express no opinion on the other two arguments for inclusion of these costs in the FEIS advanced by the Sierra Club: (1) as cumulative effects of related proposals and (2) as a secondary or indirect effect of the port. We therefore proceed to analyze the trial court’s alternative holding that even if the FEIS was deficient *980 under NEPA, that deficiency did not sufficiently disturb the Corps’ “public interest” analysis as to require a remand to the Corps Jor correction of the FEÍS and reconsideration of the permit decision. See 532 F.Supp. at 1239-40. [22] The five factors influencing the Colonel’s decision were, as listed by the Colonel: 1. The relative extent of the public and private need for the proposed structure or work. 2. The desirability of using appropriate alternatives, locations, and methods of accomplishing the objectives of the proposed activity3. The extent and permanence of the beneficial and/or detrimental effects which the proposed structure or work may have on public or private uses to which the area is suited. 4. The probable impact of the proposal in relation to the cumulative effect created by existing and other anticipated structures or work in the general area. 5. The public interest review encompasses a wide range of environmental concerns mandated by specific legislation and Executive Orders. In accordance with these directives I have identified these specific policy considerations and carefully evaluated the impacts of the proposed activity on each of these factors. Several of these considerations developed as major issues during the permit process and are discussed in Section IV. The other policy considerations in 33 CFR 320.-4(b)-(e), 323.5(a), and 324.2(b) were of relatively minor significance and are discussed below. Plaintiffs’ Exhibit 1 at 2.