The FAA also took into account the Port Authority's reasons for wanting a cargo hub in Toledo. In recent years, more than fifty major companies have left the Toledo metropolitan area, and with them, over seven thousand jobs. The Port Authority expects the cargo hub at Toledo Express to create immediately more than two hundred permanent and six hundred part-time jobs with a total payroll value of more than $10 million. After three years, according to the Port Authority, the hub should create directly more than one thousand permanent jobs at the airport and one hundred and fifty other, airport-related jobs. The University of Toledo estimates that the new Toledo Express will contribute at least $42 million to the local economy after one full year of operation and nearly $68 million per year after three. In addition, the Port Authority expects the expanded airport, and Burlington's presence there, to attract other companies to Toledo. All of those factors, the Port Authority hopes, will lead to a renaissance in the Toledo metropolitan region.

Having thought hard about these appropriate factors, the FAA defined the goal for its action as helping to launch a new cargo hub in Toledo and thereby helping to fuel the Toledo economy. The agency then eliminated from detailed discussion the alternatives that would not accomplish this goal. Each of the different geometric configurations would mean technological problems and extravagant costs. So would plans to route traffic differently at Toledo Express, or to build a hub at one of the other airports in the city of Toledo. None of the airports outside of the Toledo area would serve the purpose of the agency's action. The FAA thus evaluated the environmental impacts of the only proposal that might reasonably accomplish that goal -- approving the construction and operation of a cargo hub at Toledo Express. It did so with the thoroughness required by law. See 40 C.F.R. 1502.16. [FN 7]

[FN 7] Judge Buckley maintains that the FAA, having decided to discuss the socioeconomic impacts of inaction in Toledo on Toledo, was obliged then in its "No Action" section to discuss the socioeconomic impacts of inaction in Toledo on Fort Wayne. See post at 209-10. As Judge Buckley's dissent reveals, see id., information concerning Fort Wayne's economy is already available for consumption elsewhere in the EIS. See Tongass Conservation Soc'y v. Cheney, 924 F.2d 1137, 1142-43 & n. 5 (D.C.Cir.1991). In any event, the FAA also discussed the (beneficial) environmental effects of inaction in Toledo on Toledo, so one can infer that it should have discussed the (presumably negative) environmental effects of inaction in Toledo on Fort Wayne. Because Toledo's loss would be many other cities' potential gain, moreover, there would be no reason to limit the FAA's discussion to Fort Wayne: Indeed, one can infer that the FAA should have discussed the socioeconomic and environmental impacts of inaction in Toledo on Peoria, Akron, Detroit, and every other site assessed by Burlington's consultant. Cf. post at 207-10.

The EIS demonstrates that the discussion of the socioeconomic and environmental impacts of inaction is the flip side of the discussion of the impacts of action. If, for example, the FAA were to approve the Port Authority's application, Toledo would lose environmentally but gain socioeconomically, and Peoria and Fort Wayne and the other cities would lose socioeconomically but gain environmentally. If the FAA were to reject the Port Authority's application, Toledo would remain somewhat quieter but lose some jobs, and either Peoria (or another city) might gain noise along with jobs or Fort Wayne might retain some of both. But the FAA was not obliged to discuss the environmental or socioeconomic impacts of approving airport expansions in Peoria or Fort Wayne or any of the other cities: None, as we have explained, would have fulfilled the goal of the agency's action, and all were therefore unreasonable and beyond the scope of the FAA's responsibilities. The upshot of Judge Buckley's approach, it seems to us, would be to force an agency to discuss the socioeconomic and environmental impacts of even unreasonable alternatives -- to do the very thing in the section on the do-nothing alternative that the agency need not do in the statement's main body.

Citizens agrees that the FAA need only discuss reasonable, not all, alternatives to Toledo Express. Relying on Van Abbema v. Fornell, 807 F.2d 633 (7th Cir.1986), however, Citizens argues that "the evaluation of 'alternatives' mandated by NEPA is to be an evaluation of alternative means to accomplish the general goal of an action; it is not an evaluation of the alternative means by which a particular applicant can reach his goals." Id. at 638 (construing NEPA s. 102(2)(E), 42 U.S.C. 4332(2)(E)). According to Citizens, the "general goal" of the Port Authority's proposal is to build a permanent cargo hub for Burlington. Since, in Citizens' view, Fort Wayne (and perhaps Peoria) will accomplish this general goal just as well as Toledo, if not better, Baer Field is a reasonable alternative to Toledo Express, and the FAA should have discussed it in depth. Since it did not, this court should force the FAA to prepare a new (or supplemental) environmental impact statement.

We see two critical flaws in Van Abbema, and therefore in Citizens' argument. The first is that the Van Abbema court misconstrued the language of NEPA. Van Abbema involved a private businessman who had applied to the Army Corps of Engineers for permission to build a place to "transload" coal from trucks to barges. See 807 F.2d at 635. The panel decided that the Corps had to survey "feasible alternatives ... to the applicant's proposal," or alternative ways of accomplishing "the general goal [of] deliver[ing] coal from mine to utility." Id. at 638; see also Trout Unlimited v. Morton, 509 F.2d 1276, 1286 (9th Cir.1974). In commanding agencies to discuss "alternatives to the proposed action," however, NEPA plainly refers to alternatives to the "major Federal actions significantly affecting the quality of the human environment," and not to alternatives to the applicant's proposal. NEPA s, 102(2)(C), 42 U.S.C. 4332(2)(C) (emphasis added). An agency cannot redefine the goals of the proposal that arouses the call for action; it must evaluate alternative ways of achieving its goals, shaped by the application at issue and by the function that the agency plays in the decisional process. Congress did expect agencies to consider an applicant's wants when the agency formulates the goals of its own proposed action. Congress did not expect agencies to determine for the applicant what the goals of the applicant's proposal should be.

The second problem with Van Abbema lies in the court's assertion that an agency must evaluate "alternative means to accomplish the general goal of an action," 807 F.2d at 638 (emphasis deleted) -- a statement that troubles us even if we assume that the panel was alluding to the general goals of the federal action instead of to the goals of the private proposal. Left unanswered in Van Abbema and Citizens' brief (and at oral argument) is why and how to distinguish general goals from specific ones and just who does the distinguishing. Someone has to define the purpose of the agency action. Implicit in Van Abbema is that the body responsible is the reviewing court. As we explained, however, NEPA and binding case law provide otherwise.

In chiding this court for having overreached in construing NEPA, a unanimous Supreme Court once wrote that Congress enacted NEPA "to ensure a fully informed and well-considered decision, not necessarily a decision the judges of the Court of Appeals or of this Court would have reached had they been members of the decisionmaking unit of the agency." Vermont Yankee, 435 U.S. at 558, 98 S.Ct. at 1219. We are forbidden from taking sides in the debate over the merits of developing the Toledo Express Airport; we are required instead only to confirm that the FAA has fulfilled its statutory obligations. Events may someday vindicate Citizens' belief that the FAA's judgment was unwise. See id. at 557-58, 98 S.Ct. at 1218-19. All that this court decides today is that the judgment was not uninformed. See Methow Valley, 490 U.S. at 351, 109 S.Ct. at 1846.

The regulations of the Council on Environmental Quality provide that an environmental impact statement "shall [contain] a full and fair discussion of significant environmental impacts" and that "[i]mpacts shall be discussed in proportion to their significance." 40 C.F.R. 1502.1, 1502.2(b); se NEPA s 102(2)(C)(i), (ii), 42 U.S.C. 4332(2)(C)(i), (ii). The EIS in this case discusses more than twenty impacts that the expanded Toledo Express would have on the environment, including the airport's effects on people's homes and neighborhoods; on the quality of the air, the water, and the earth; on architectural, archeological, and cultural resources; on sewage disposal; on traffic patterns; on swamps, marshes, bogs, and rivers; and on bats, butterflies, grass, flowers, and trees. [FN 8] The EIS also states flatly that "[a]ircraft sound emissions" -- noise, in a word -- are "often the most noticeable environmental effect[s] an airport will produce on the surrounding community." In all, the FAA devotes about half of its discussion on environmental consequences to the effects of an increase in noise. Although Citizens does not argue that the FAA failed to discuss the impacts of noise in rough proportion to the effects' importance, it does argue that the discussion is incomplete and unfair. We disagree.

[FN 8] Interestingly enough, birds and deer show no signs of being affected by the noise or exhaust fumes at the Toledo Express Airport, and officials of the Ohio Department of Natural Resources have also seen there some endangered animals, such as the spotted turtle (Clemmys guttata). State officials report that endangered plants, such as the cross-leaved milkwort (Polygala cruciata) and twisted yellow-eyed grass (Xyris torta ), are even thriving right beside the main runways.

Citizens concedes, if only implicitly, that the rule of reason guiding the FAA necessarily covers the agency's discussion of particular environmental impacts. See Natural Resources Defense Council v. Morton, 458 F.2d at 834. Relying on Davison v. Department of Defense, 560 F.Supp. 1019 (S.D.Ohio 1982), however, Citizens contends that in discussing the impacts of noise, a reasonable agency would at least estimate the number of people whom an expanded airport would keep awake. Citizens points out, moreover, that the EPA criticized the FAA's original choice of methods, and that in response to the EPA's comments the FAA agreed to modify its analysis in future cases. For these reasons, Citizens argues, we should find that the FAA's discussion was inadequate.

We think that Davison provides only weak support for Citizens' argument. In Davison, the Air Force decided to sell part of an old base to a firm that planned to use it for a cargo hub. Reviewing the adequacy of the resulting EIS, the court held that the Air Force had unreasonably failed to quantify with some precision the people whom the hub activity would keep up at night, had unreasonably neglected to discuss whether local residents would become accustomed to the noise, and had unreasonably overlooked the physiological effects of long-term sleep disturbance. See 560 F.Supp. at 1036-37. Here, in contrast,the FAA did all but the first. On remand in Davison, moreover, the Air Force then stated in a supplement to the final EIS that "from one hundred to one thousand people may be awakened from sleep, possibly repeatedly, for up to four hours per night, approximately 250 nights per year." To the extent that the logic of Davison would impose a similar requirement on the FAA -- and the Air Force's estimate in Davison was not quite the paradigm of precision that Citizens demands here -- we think it inconsistent with circuit precedent.

In examining the impacts of noise on the environment, the FAA relies on wisdom and experience peculiar to the agency and alien to the judges on this court. We have thus held consistently that the rule of reason guides every aspect of the FAA's approach, including its choice of scientific method. See, e.g., Sierra Club v. Department of Transp., 753 F.2d 120, 128 (D.C.Cir.1985); see also Valley Citizens for a Safe Environment v. Aldridge, 886 F.2d 458, 469 (1st Cir.1989). Employing here a method that we have previously endorsed, see Sierra Club v. Department of Transp., 753 F.2d at 128, the FAA proceeded to mold a body of data, dissect it, and display it in comprehensible forms. The agency's choice of method was obviously not capricious. Nor were the factual conclusions that followed. See Marsh v. Oregon Natural Resources Council, 490 U.S. 360, 377, 109 S.Ct. 1851, 1860-61, 104 L.Ed.2d 377 (1989) ("Because analysis of the relevant documents 'requires a high level of technical expertise,' we must defer to 'the informed discretion of the responsible federal agencies.'" (citations omitted)); Valley Citizens, 886 F.2d at 467-69.

The EPA's criticisms of the FAA, and the agencies' subsequent deal, do not change our view of the FAA's findings. Congress wants the EPA to participate when other agencies prepare environmental impact statements. See 42 U.S.C. 7609(a). The EPA participated here. But the FAA, not the EPA, bore the ultimate statutory responsibility for actually preparing the environmental impact statement, and under the rule of reason, a lead agency does not have to follow the EPA's comments slavishly -- it just has to take them seriously. See Alaska v. Andrus, 580 F.2d at 474. The FAA considered the EPA's criticisms in this case and decided that enough had been done. That the FAA sensibly resolved to avoid any interagency disputes in the future does not make its decision in this case unreasonable. We uphold the FAA's discussion of the impacts of increased noise.

The regulations of the Council on Environmental Quality require that an environmental impact statement "be prepared directly by or by a contractor selected by the lead agency." 40 C.F.R. 1506.5(c). If the agency decides to contract out the work on the EIS, the agency must choose the contractor "to avoid a conflict of interest," and the contractor must "execute a disclosure statement prepared by the lead agency ... specifying that [it has] no financial or other interest in the outcome of the project." Id. Citizens argues that the FAA violated the regulations by publishing an EIS prepared for the most part by a contractor (Coffman Associates) that the agency did not itself select and that did not in any event fill out the necessary disclosure forms. The FAA maintains that it (the FAA), and not Coffman, prepared the EIS, that even if Coffman did prepare the EIS, it (the FAA), and not the Port Authority, selected Coffman, and that even though Coffman did not fill out the disclosure statement, its (Coffman's) failure to do so was harmless error. We reject each of the FAA's contentions. Offered the choice of preparing the environmental impact statement in-house, the FAA chose the other permissible option and hired consultants, including Coffman. The FAA then wrote the consultants' names and qualifications, including Coffman's, in a chapter of the EIS entitled "List of Preparers," see id. s 1502.17, a gesture that undermines the agency's current litigating position -- that Coffman did not prepare the EIS, but that the FAA did instead, mostly by commenting actively on Coffman's drafts. Ultimately, however, the agency's theory founders on the plain meaning of the regulations. Although the CEQ regulations do not define the word "prepare," the dictionary does; in context, it means here "to put into written form: draw up ... [directed the commission to prepare proposals....]." Webster's Third New Int'l Dictionary 1790 (unabridged ed.1981); see Sierra Club v. Marsh, 714 F.Supp. 539, 550-51 (D.Me.1989). That is just what Coffman did, as the agency freely admits. We need not decide whether the FAA's active editing of Coffman's drafts -- behavior consistent with the agency's obligation to "furnish guidance" to consultants and "participate in the preparation [of] and ... independently evaluate the statement prior to its approval," 40 C.F.R. 1506.5(c) -- made it, too, a preparer of the EIS. We are certain, however, that Coffman's initial drafts and responses to the FAA's comments made Coffman more than the agency's amanuensis.

Once the FAA decided not to prepare the environmental impact statement directly, it was obliged to pick a contractor itself, and not to delegate the responsibility. See id. The EIS states that the Port Authority, not the agency, chose Coffman to work on the environmental assessment, and later, on the environmental impact statement. The EIS also states that the agency "concurred" in Coffman's selection. The FAA argues that its concurrence in the Port Authority's choice satisfied its duty under the regulations. We need not page through the dictionary at length to decide that concurring in someone else's choice of consultant is not the same as choosing a consultant of one's own.

By failing to select the consultant that prepared the environmental impact statement, the FAA violated CEQ regulations. Citizens urges us to remedy this breach by invalidating the EIS. We see no reason to do so, however, at least not solely on the ground that the FAA neglected to search on its own for a competent contractor. This particular error did not compromise the "objectivity and integrity of the [Nepa] process." Forty Questions, 46 Fed.Reg. at 18,031; see Sierra Club v. Sigler, 695 F.2d 957, 963 n. 3 (5th Cir.1983) (CEQ regulations are " '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.'" (quoting 43 Fed.Reg. 55,987 (1978))); cf. 40 C.F.R. 1500.3 ("[I]t is the [CEQ's] intention that any trivial violation of these regulations not give rise to any independent cause of action.").

The more serious infraction, in our view, was Coffman's failure to fill out the disclosure form exacted of consultants that prepare environmental impact statements. See id. s 1506.5(c). Citizens points out that Coffman (in addition to having prepared the EIS) has started to prepare the Port Authority's Part 150 study, and that the scope of the study will vary directly with the status of the airport (since the Port Authority is relying on the study to fine tune its mitigation plans). The FAA argues that Coffman had no reason to know while preparing the EIS that the agency would want it to expand the Part 150 study. The FAA may well be correct, but neither the petitioners nor this court can know for certain in the absence of a completed disclosure form. Moreover, the CEQ regulations prohibit broadly any "financial or other interest in the outcome of the project." Id. (emphasis added); see Forty Questions, 46 Fed.Reg. at 18,031 (interpreting "conflict of interest" to mean "any known benefits other than general enhancement of professional reputation"). The FAA promised the petitioners in a letter that "Coffman does not have an undisclosed stake in the project that would potentially disqualify it." That ipse dixit does not reassure us. We therefore order the FAA to have Coffman execute an appropriate disclosure statement, see 40 C.F.R. 1506.5(c), and, should the agency find that a conflict exists, to decide -- promptly -- on the measures to take in response.

Under section 4(f) of the Department of Transportation Act of 1966, the Secretary of Transportation may not approve a project requiring the use of a park unless he determines, first, that there is no "prudent and feasible alternative" to using the land, and second, that the project includes "all possible planning to minimize harm to the park ... resulting from the use." Transportation Act s. 4(f), 49 U.S.C. 303(c). [FN 9] The FAA (which is part of the Department of Transportation, see 49 U.S.C. 106(a)) acknowledged that the proposed expansion of Toledo Express would constructively "use" the Springbrook campground since flights from the airport would subject the camp to nighttime noise of up to L subdn 75 decibels, about 10 to 15 decibels more than now. Cf. Allison v. Department of Transp., 908 F.2d 1024, 1030 (D.C.Cir.1990) (no section 4(f) use when a park is subjected to only minor increases in airplane noise). The agency nonetheless decided that while there might be a feasible alternative to using the campground, cf. Citizens to Preserve Overton Park, Inc. v. Volpe, 401 U.S. 402, 411, 91 S.Ct. 814, 821, 28 L.Ed.2d 136 (1971) (alternative is "feasible" unless "as a matter of sound engineering" it should not be built), there existed no prudent one, and that the project would ease the harm to Springbrook by moving it elsewhere inside the park, but outside the reach of L subdn 65 decibels. Citizens argues that a feasible and prudent alternative to using the campground did exist: leaving the airport in Toledo alone and expanding the airport in Fort Wayne instead. Citizens also argues that the project does not adequately diminish the harm to Springbrook because the FAA did not consider, among other ideas, fining the owners of planes that are noisy, and because the FAA has not said where exactly in Oak Openings it plans to put the new campground.

[FN 9] The Secretary [of Transportation] may approve a transportation program or project ... requiring the use of publicly owned land of a public park, recreation area, or wildlife and waterfowl refuge of national, State, or local significance, or land of an historic site of national, State, or local significance ... only if --

(1) there is no prudent and feasible alternative to using that land; and

(2) the program or project includes all possible planning to minimize harm to the park, recreation area, wildlife and waterfowl refuge, or historic site resulting from the use.

Transportation Act s 4(f), 49 U.S.C. s 303(c).

Reasoning by analogy to NEPA, the FAA argues that an alternative must be imprudent under section 4(f)(1) if it fails to accomplish a proposal's objectives. According to the FAA, since a hub in Fort Wayne would do nothing for Toledo, and since the health of the Toledo economy was a primary reason for the Port Authority's application, Fort Wayne was an imprudent alternative and the FAA did not act arbitrarily in approving the use of Springbrook. In effect, the FAA's argument would mean that anytime an alternative is unreasonable under NEPA (and thus would not have to be discussed in detail in the environmental impact statement), the alternative would also be imprudent within the meaning of section 4(f)(1) of the Transportation Act (and thus would not block approval of a transportation project).

Although an agency's analysis under NEPA and the Transportation Act might proceed in similartracks, the two statutes are not precisely the same. The Transportation Act differs from NEPA in at least two ways. First, the Transportation Act requires the agency to evaluate "prudent ... alternatives to using th[e] land " -- alternatives to the project, that is -- not alternatives to the federal action. Second, contrary to the FAA's argument, the case law uniformly holds that an alternative is imprudent under section 4(f)(1) if it does not meet the transportation needs of a project. See Hickory Neighborhood Defense League v. Skinner, 910 F.2d 159, 164 (4th Cir.1990); Druid Hills Civic Ass'n v. Federal Highway Admin., 772 F.2d 700, 715 (11th Cir.1985); Arizona Past & Future Found. v. Dole, 722 F.2d 1423, 1428-29 (9th Cir.1983). The Transportation Act is similar to NEPA in that the agency bears the responsibility for defining at the outset the transportation goals for a project and for determining which alternatives would reasonably fulfill those goals.

Having focused on the statutes' apparent similarities, and disregarded their differences, the FAA never quite specified with ideal coherence the transportation goals of the project at Toledo Express. In future cases, the agency should bear in mind the differences between NEPA and the Transportation Act, and the agency's section 4(f) documentation package should reflect the concerns specific to the latter statute. Still, in approving in this case the use of the park in Toledo, the FAA reasonably defined the transportation goals of the project as providing the Toledo area with a modern, effective cargo hub. Given this definition of the project's aims, the FAA need not have examined in detail the relative flaws of Baer Field, including its antiquated condition, its distance from Burlington's main markets (Detroit and Chicago), Fort Wayne's limited pool of labor, and the city's failure to come up with the necessary financing. It was enough for the agency to find that a hub in Baer Field would not fulfill the transportation goals of the project at Toledo Express and that Fort Wayne was therefore less than a prudent alternative to using Toledo. Because its conclusion was reasonable, the FAA did not violate section 4(f)(1).