Davison v. Department of Defense

OPINION AND ORDER DUNCAN, District Judge. This action is brought under the National Environmental Policy Act (NEPA), 42 U.S.C. §§ 4321 eí seq., and regulations promulgated thereunder, to challenge the sufficiency of an environmental impact statement (EIS) assessing the proposed use of Rickenbacker Air National Guard Base (Rickenbacker) ' near Groveport, Ohio. Plaintiffs oppose the proposed addition of civilian air cargo operations at the base, and *1022 they allege that the EIS prepared by the Department of the Air Force does not paint an accurate picture of the noise impact that an air cargo facility would have on the communities surrounding Rickenbacker. By establishing the inadequacy of the EIS, plaintiffs hope to force defendant, the Secretary of the Air Force (the Secretary), to reconsider his approval of the proposed plan. At issue is whether the data presented by the EIS was sufficient to allow the Secretary to make an intelligent choice between the available alternative uses for the base. For the reasons that follow, the Court finds that further study is needed. I Prior to 1979, Rickenbacker was a fully operational Strategic Air Command base which served as headquarters for the 301st Air Refueling Wing. It employed approximately 1800 military personnel and 1200 civilians, and it housed at least 82 military aircraft, many of which were KC-135 refueling tankers. On March 29, 1979, the Secretary announced that Air Force activities at Rickenbacker would be sharply reduced as part of a plan to consolidate and streamline air refueling resources. Beginning in July of that year, SAC activities at the base ceased, manpower was gradually withdrawn, and most of the KC-135 aircraft were relocated. The 301st Air Refueling Wing was inactivated, and the base was turned over to the Ohio National Guard. When the phaseout is complete, the Air Force expects that only 148 military personnel and 768 civilians will remain there. As a result of this reduction in forces, much of the land and many of the buildings at Rickenbacker are no longer needed by the Air Force. Procedures for disposing of the excess property were commenced, and the City of Columbus and several of its surrounding communities expressed interest in acquiring the facilities for use as an “industrial air park.” To this end defendant Rickenbacker Port Authority (RPA) was created to receive the excess lands and to negotiate an agreement with the Air Force for joint use of the property retained by it. In order to comply with the requirements of NEPA, the Secretary commissioned a private consulting firm to produce an EIS on the proposed use of Rickenbacker as an air cargo facility. A draft was released for review and comment on July 24, 1981, and public hearings were held in Groveport on September 2, 1981. In response to many expressions of concern over the potential noise impact of the proposed air cargo facility, an expanded study of noise effects on sleep and education was undertaken, and its results are included in the final draft of the EIS. 1 In October 1981, the EIS was submitted to the Secretary. The final EIS surveys several potential environmental effects of flight operations at the proposed air cargo facility, but it concludes that the only significant impact resulting from air cargo operations would be on noise levels in the area. The executive summary appearing at the front of the EIS notes that: The introduction of regular air cargo flights at Rickenbacker ANGB will expand the area presently influenced by aircraft noise and will produce new noise impacts because of the introduction of a significant number of night and early morning flights .... The night and early morning flying activity, if unmitigated, is expected to awaken 800 and 485 Groveport inhabitants from sleep during summer and winter, respectively. This disturbance will occur on the average of one and one-half days per week. Unmitigated cargo related noise may adversely influence the teaching environment in five Groveport schools, for an average of three hours per week. A few lines later the summary also notes that: The introduction of mitigation measures will significantly reduce, but not eliminate the noise impacts described above. These include “head-to-head” landing and *1023 takeoff procedures (simultaneous landings from and takeoff to the south, away from populated areas) and the installation of an instrument landing system on a southern approach. The Air Force is committed to insuring that these mitigation measures are included in the proposed action. 1 Department of the Air Force, Final Environmental Impact Statement for the Proposed Joint Use of Rickenbacker ANGB (hereinafter EIS) at ii-iii (October 1981). After reviewing these findings, Tidal W. McCoy, an Assistant Secretary of the Air Force, issued a written decision approving the proposed plan. McCoy noted that the joint use proposal was an ideal solution from the Secretary’s point of view because it allowed the government to preserve an airfield complex that could be quickly reconverted to a military base if necessary, and it enabled the Air Force to maintain reserve forces at the base at a relatively low cost. With respect to the adverse environmental' effects of air cargo operations, McCoy concluded that: The potential noise impact is the item of most concern to the communities surrounding Rickenbacker ANGB. This issue was most evident during public hearings held in Groveport, Ohio, on September 2,1981. The major concern expressed by the public was the impact from noise anticipated from RPA’s introduction of a regular air cargo operation. (Note: Public comments are contained in the EIS.) Accordingly, mitigation measures identified in the EIS can be implemented to help reduce the noise impact of additional flight activity to the surrounding community. Record of Decision on Joint Use of Rickenbacker ANGB, at 2 (December 14, 1981). Pursuant to this decision, Air Force representatives were authorized to enter into negotiations with RPA concerning lease and joint use agreements. Plaintiffs began this action well before publication of the Secretary’s final decision, but it was not until sometime thereafter that the suit began to assume its present form. The case was initially filed in October 1980 by a group of retired Air Force personnel and concerned citizens calling themselves the Military Defense Action Group (MDAG). Several of the retirees used Rickenbacker facilities to obtain needed goods and medical treatment at a substantial discount, and they hoped to preserve this service. In addition, the group objected generally to the reduction of forces at a base which they viewed as vital to the national defense. The original complaint alleged violations of NEPA, but it made no express reference to the noise impact of the proposed air cargo operations; instead it asked for a restraining order on grounds that the Secretary had not adequately considered the economic ramifications of curtailing military operations at Rickenbacker, particularly the effect on retired military personnel who depend on the facilities for medical supplies. The Court held a hearing on plaintiffs’ motion for injunctive relief, but it denied the request on October 27, 1980. Approximately six months later, plaintiffs amended their complaint to include a claim under NEPA contesting the sufficiency of the Secretary’s environmental assessment. 2 Shortly after publication of the final EIS, plaintiffs again moved for a TRO enjoining the Air Force and RPA from executing or implementing a joint use agreement. Plaintiffs charged, inter alia, that the final EIS failed to inform the Secretary of a “massive sleep disturbance affecting over 1,000 people.” Upon meeting with counsel, the Court denied the motion but set the case down for trial on the merits in early February 1982. In the interim plaintiffs filed a second amended complaint which added as plaintiffs Citizens Opposing Noise, a group composed of residents of communities nearby *1024 Rickenbacker. The new complaint also included detailed objections to the analysis of noise impact contained in the EIS. Subsequently the Village of Groveport, the Village of Lockbourne, Ohio, and the Grove-port-Madison School District were joined as additional parties plaintiff. A few days before the trial was to begin, the Secretary and other federal defendants moved for a continuance. They notified the Court that a number of errors had been made in compiling the noise-impact data upon which the EIS was based. The continuance was granted, and the parties entered into a consent order whereby it was agreed that while the data was being re-evaluated, neither the Air Force nor RPA could proceed further with implementation of the joint use plan. A few weeks later, the federal defendants informed the Court that the erroneous data in the EIS had been “reaccomplished.” A rather detailed document which purported to correct the final EIS was submitted, and defendants moved for a new trial date. Beginning on April 14, 1982, the Court heard four days of testimony regarding the sufficiency of the final EIS. Plaintiffs produced several exhibits plus the testimony of a number of noise experts and citizens who allegedly will be affected by the proposed air industrial park. Defendants relied primarily on the EIS itself, as well as the testimony of several experts who, directly or indirectly, played a role in its preparation. The parties were then given an opportunity to submit post-trial memoranda, and final arguments were heard on April 30. Pending a decision on the merits, the Court entered an order barring defendants from commencing air cargo operations or transferring title to the excessed lands. Plaintiffs’ complaint states essentially that the October 1980 EIS does not comply with the requirements of NEPA in that it fails to measure and identify properly the nature and degree of noise impact on the human environment. The complaint also charges that the excessed lands cannot legally be transferred to RPA without consideration, and that the proposed air industrial park would compete with Port Columbus International, a nearby civil airport, and thus violate 49 U.S.C. §§ 1507 and 1508. No evidence or argument was presented by plaintiffs on any of these latter theories, so the Court must find in favor of defendants on them. The only issue properly before the Court is whether the final EIS is adequate as a matter of law with respect to its evaluation of the noise which could result from the proposed air cargo operation. II It should be emphasized from the outset that the final decision on how best to use the facilities at Rickenbacker rests exclusively with the Secretary of the Air Force, not this Court. Although NEPA establishes “significant substantive goals for the Nation,” Vermont Yankee Nuclear Power Corp. v. NRDC, 435 U.S. 519, 558, 98 S.Ct. 1197, 1219, 55 L.Ed.2d 460 (1978), it does not wrest from administrative agencies their traditional decisionmaking power. The statute requires only that: to the fullest extent possible ... (2) all agencies of the Federal Government shall— (C) include in every recommendation or report on proposals for legislation and other major Federal actions significantly affecting the quality of the human environment, a detailed statement of the responsible official on— (i) the environmental impact of the proposed action, (ii) any adverse environmental effects which cannot be avoided should the proposal be implemented, (iii) alternatives to the proposed action, (iv) the relationship between local short-term uses of man’s environment and the maintenance and enhancement of long-term productivity, and (v) any irreversible and irretrievable commitments of resources which would *1025 be involved in the proposed action should it be implemented. 42 U.S.C. § 4332. These requirements are largely procedural, and once the agency in question has complied with them, the role of the courts is limited to determining whether or not the responsible administrators have taken a “hard look” at the environmental consequences of the proposed action. Kleppe v. Sierra Club, 427 U.S. 390, 410 n. 21, 96 S.Ct. 2718, 2730 n. 21, 49 L.Ed.2d 576 quoting Natural Resources Defense Council, Inc. v. Morton, 458 F.2d 827, 838 (D.C.Cir.1972); Evans v. Train, 460 F.Supp. 237, 241 (S.D.Ohio 1978); Ohio, ex rel. Brown v. U.S.E. P.A., 460 F.Supp. 248, 250 (S.D.Ohio 1978). As the Supreme Court has observed: NEPA was designed to “insure a fully informed and well-considered decision,” but 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.” Strycker’s Bay Neighborhood Council v. Karlen, 444 U.S. 223, 227, 100 S.Ct. 497, 499, 62 L.Ed.2d 433 (1980) (per curiam), quoting Vermont Yankee, supra, 435 U.S. at 558, 98 S.Ct. at 1219. See also, Citizens to Preserve Overton Park v. Volpe, 401 U.S. 402, 416, 91 S.Ct. 814, 823, 28 L.Ed.2d 136 (1971); and Kleppe, supra, 427 U.S. at 410, 96 S.Ct. at 2730. Clearly, then, this Court cannot substitute its own judgment for that of the Secretary of the Air Force. With respect to environmental impact statements, the scope of review is correspondingly narrow. The “detailed statement” requirement of NEPA was designed not to give federal judges a means by which to second guess agency decisionmakers, but rather to allow the courts to assure that the agency in question has made a good faith effort to explore the concerns voiced within the statute. Minnesota Public Interest Research Group v. Butz, 541 F.2d 1292, 1299 (8th Cir.1976), cert. denied, 430 U.S. 922, 97 S.Ct. 1340, 51 L.Ed.2d 601 (1977). In assessing the sufficiency of an EIS, a reviewing court’s inquiry is thus limited to determining: (1) whether the EIS discusses all five of the factors listed in 42 U.S.C. § 4332(C) and otherwise meets the procedural requirements set forth therein; (2) whether the EIS as a whole evidences a good faith effort on the part of the preparer to comply with the demands of NEPA; and (3) whether the EIS contains sufficient information to enable the agency to make a reasonable and independent choice among the available alternatives. Id., 541 F.2d at 1298-1301; Save Our Invaluable Land (Soil), Inc. v. Needham, 542 F.2d 539, 542-543 (10th Cir.1976), cert. denied, 430 U.S. 945, 97 S.Ct. 1580, 51 L.Ed.2d 792 (1972); Sierra Club v. Morton, 510 F.2d 813, 819-820 (5th Cir.1975); National Helium Corp. v. Morton, 486 F.2d 995, 1001-1003 (10th Cir.1973), cert. denied, 416 U.S. 993, 94 S.Ct. 2405, 40 L.Ed.2d 772 (1974); and Evans v. Train, supra, 460 F.Supp. at 241-242. The touchstone of this test is reasonableness. Minnesota Public Interest Research Group v. Butz, 541 F.2d at 1300. NEPA does not require perfection, and an EIS will not be rejected simply because it fails to address all possible contingencies or omits the views of a particular expert. Id.; see also Environmental Defense Fund v. Tennessee Valley Auth., 492 F.2d 466, 468, n. 1 (6th Cir.1974) (per curiam); and Kentucky, ex rel. Beshear v. Alexander, 655 F.2d 714, 718 (6th Cir.1981). Against NEPA’s command that potential environmental problems be considered “to the fullest extent possible,” the Court must balance notions of feasibility and practicality. Id. Consequently the Court will not reject an EIS in part or in whole unless it is convinced that the alleged defect has fundamentally impaired the decisionmaker’s ability to take a “hard look” at the relevant environmental issues. Here it is undisputed that the final EIS on Rickenbacker deals, with each of the five topics enumerated in § 4332(C), and plaintiffs have not pointed to any failure by the Secretary to comply with other procedural prerequisites of the statute. Likewise, no evidence has been produced that would in any way indicate a lack of good faith on the *1026 part of the Secretary or the preparers of the EIS. A great deal of evidence challenging specific data within the EIS was adduced at trial, however, and the Court must review it in some detail to determine whether or not the Secretary was able to give a “hard look” to potential noise problems created by proposed air cargo operations. Ill The final EIS is divided into two volumes, each of which contains over 150 pages. The first volume reviews the proposed plan, the viable alternatives to it, and the data compiled on potential environmental effects of an air cargo operation. This is the “meat” of the EIS, and a significant portion of it is devoted to noise and its potential impact on the communities near the base. The second volume catalogs the public comments received on the proposed use. It includes letters from members of the public to officials connected with the project, including President Reagan, and it also contains a transcript of the public hearings held in September 1981. The bulk of these comments are from citizens residing in and around the Rickenbacker area, and most express some degree of concern over the increased noise that may result if Rickenbacker is ultimately used as an air industrial park. The EIS deals with the noise problem from several different angles. Among other things, it presents a graphic comparison of noise levels at Rickenbacker before and after the proposed operations have begun. On several maps of the surrounding area, “noise contours” are plotted both for existing military operations and proposed civilian operations. These contours were “reaccomplished” prior to trial, and they were the basis for many of plaintiffs’ objections to the EIS. The report also includes a noise analysis of feasible steps that could be taken by private air carriers to mitigate the noise created by the new facility. In addition the EIS contains somewhat detailed estimates of potential classroom disruption and sleep disturbance which might be caused by private air cargo flights of the type proposed. On its face the report appears to discuss rather thoroughly both the magnitude of the noise increase resulting from the proposed facility and the harms that can reasonably be expected to flow from it. Plaintiffs, however, have raised many specific and rather technical objections to the document. A. Average Noise Levels . Plaintiffs take issue first with the noise level data in general. This data purports, by way of noise contour maps, inter alia, to show which additional areas in the Rickenbacker vicinity will be exposed to a significantly higher Day-Night Average Sound Level (DNL) 3 as a result of planned air cargo operations. Using a DNL of 65 decibels (dB) as the threshold figure for significant noise exposure, 4 the EIS Executive Summary concludes that 67% of Groveport dwellings (or about 2100 Groveport residents) will fall into contours greater than 65 dB under the proposed plan, as opposed to 22% under current operations. Plaintiffs claim that these figures are misleading because they tend to overestimate the noise levels produced by current operations and underestimate the noise which would be generated by air cargo flights. Defendants *1027 recognize that some flaws do exist in the reported data, but on the basis of newly-computed figures they claim that, if anything, the EIS noise contours actually overstate the probable noise increase. Plaintiffs point out correctly that the greatest change brought about by the proposed air freight operation would be the introduction of nighttime flights between the hours of 11:00 p.m. and 7:00 a.m. They argue that use of 24-hour DNL figures artificially blurs the obvious differences between nighttime versus daytime air traffic. The Secretary notes in response that DNL figures in the EIS were further broken down into 12-hour intervals, and that the reaccomplished data submitted to the Court even includes Equivalent Noise Level (Leq) figures for the standard 15-hour daytime period and the 9-hour nighttime period. 5 Much of plaintiffs’ criticism with respect to the use of such data is well taken. Notwithstanding the 10 dB “penalty” that is built into DNL figures for nighttime operations, use of 12- and 24-hour average noise levels can be misleading to the reader, since the contour maps produced therefrom do not indicate peak flight times or noise levels within the measured period. Where, as here, the peak times under the new plan may differ considerably from those under the old, this omission can be crucial. For example, the contours shown for current operations at Rickenbacker between 7:00 p.m. and 7:00 a.m. 6 are produced almost exclusively by military flights between 7:00 p.m. and 11:00 p.m. — before most people go to sleep. In contrast, the contours shown for the same hours under the proposed air cargo usage 7 are produced primarily by flights which will occur between 12:00 midnight and 6:00 a.m. When the maps showing these two sets of contours are placed side-by-side, the difference in DNL between the current use and the proposed plan seems relatively small. To the sleeping public, however, the difference in overall noise impact between the two operations may well be dramatic. 8 In spite of these difficulties, the Court finds that the use of 12- and 24-hour DNL noise contour figures does not fatally flaw the EIS. The DNL maps are somewhat useful in providing the decisionmaker with a rough idea of the areas which will be most affected by the proposed use. Coupled with the tables in the EIS which estimate the acreage and number of residences falling into the 65 dB contour under both the current and proposed plans, 9 the maps do inform the reader that a significant number of people in the vicinity may be exposed to unacceptable noise levels for the first time as a result of civilian air cargo traffic. Because the data is expressed in terms of noise level averages, it tends to obscure the noise-related peaks and valleys that may be experienced by area inhabitants, but the limited value of this type of measuring device should be readily apparent to the decisionmaker upon reading the entire EIS. Although DNL averages and contour maps are heavily emphasized, a *1028 number of other noise measurement methodologies are discussed and employed in the EIS. See EIS at IV-1 through IV-8. 10 The use of these other measurement techniques makes it relatively clear that noise level averaging is but one way to evaluate overall noise impact. Plaintiffs next attack the accuracy of the data used to produce noise contour configurations and other noise-level indicators mentioned above. They charge that the EIS understates the potential noise levels produced by the proposed air cargo facility in that (1) it does not take into account operations on Rickenbacker’s “inside runway”; (2) it uses inaccurate baseline data for civilian cargo aircraft; and (3) it assumes unrealistically that 1985 FAA guidelines for engine noise reduction will be timely met by most civilian aircraft. The Court finds these contentions largely unavailing. The preparers of noise contour data in the EIS apparently assumed that all current operations now occur, and will continue to occur, on Runway 05R-23L. This is the “outside” runway, which has traditionally been used in the vast majority of military operations at Rickenbacker. Parallel to this, however, is Runway 05L-23R, the “inside runway”; its flight line takes aircraft approximately 1000 feet closer to Groveport than does the outside strip. Plaintiffs maintain that failure to consider operations which may occur on the inside runway impermissibly skews the noise contour data, at least with respect to Grove-port. The Court, though, has serious doubts as to whether the effect of this 1000 foot shift in contour lines would be statistically significant. 11 Even if it is, the testimony of several defense witnesses, including the executive director of RPA, established that the inside runway is not now suitable for sustained air cargo use, and that no plans are in the offing to make it so. The inside runway is temporarily being used while the outside strip is being renovated, and some reference to both runways has been made in presentations to prospective civilian tenants at the base, but it appears that long-term use of the inner runway would be impossible without major rebuilding efforts, none of which are contemplated in the reasonably foreseeable future. The Court therefore finds that the failure to calculate noise contours from the flightline of the inside runway was not a material omission. The baseline data used to estimate noise levels produced by the proposed air industrial park was adduced from a rather sophisticated computer program that is regularly used in Air Force noise estimation projects. Plaintiffs produced accoustical expert Dr. Angelo Campanella, who, on the basis of his own studies, disputed the validity of these baseline figures. He testified that he had taken actual noise level readings at an air cargo facility like the one proposed at Rickenbacker and had concluded that the projected figures in the EIS seriously underestimates the actual Single-Event Noise Levels (SEL), and hence DNL, that would result from industrial air cargo flights. 12 Dr. Campanella admitted on cross-examination, however, that his measurements were taken for only 24 to 48 hours at each location tested, and that he was unable to match specific readings with particular aircraft. In sum, he was unable to show conclusively that there was statistically significant error in the government baseline data. Defendants, by contrast, produced several highly qualified experts who defended the Air Force noise map data, and who questioned Dr. Campanella’s methodology. Given the weight of the tes *1029 timony, the Court must conclude that plaintiffs have failed to meet their burden on this claim. The same is true with respect to the charge of likely non-compliance with FAA 1985 engine-noise requirements. Although plaintiffs introduced some evidence to show that engine-noise deadlines had been extended in the past, defendants produced, among others, Dr. William J. Galloway, a private noise consultant who testified that a great number of civilian cargo aircraft were already in compliance with the 1985 guidelines. He testified further that he expected full compliance with the new guidelines by 1985. The Court has no reason to doubt this testimony. In addition to the abovegoing assaults on the noise estimates of future operations, plaintiffs also challenge the accuracy of the data used to estimate the noise impact of current operations at the base. The complaint charges that the Air Force utilized flight operation figures for current annual operations which are approximately three times greater than the actual amount. This, of course, would have the effect of minimizing the overall change in noise levels accruing from the proposed use. In support plaintiffs offered some evidence to show that Rickenbacker “Air Traffic Control Logs,” which presumably were used in preparation of the EIS, showed higher flight totals than the corresponding “Daily Traffic Counts” kept by air traffic controllers at the base. An air traffic controller from the base admitted, though, that the discrepancy might be attributable in part to the fact that the latter records often do not include “touch-and-go” takeoffs and landings commonly used in training. Even if this is not the case, plaintiffs failed to make clear whether the Logs did in fact provide the data upon which the EIS is based, or whether the discrepancy in flight totals between the two sets of records is statistically significant in relation to average noise levels. Consequently the Court has no choice but to find in favor of defendants on this claim as well. The Court does not mean to imply that the noise level estimates used in the EIS are in any sense unflawed. They are estimates at best, and, with more time and money, many additional steps could have been taken to increase their accuracy, both in terms of the raw data base and the methodology used to present such data. Viewed in their totality, though, the noise contour maps and other noise level indicators appear adequately to apprise the decisionmaker of the significant impact of air cargo operations on average noise levels in the Rickenbacker area. As suggested above, the usefulness of average noise level contours is quite limited where the major change wrought by the proposed use will be a shift in peak flight times, but this limitation should be evident to the average reader upon reviewing the whole EIS. Hence the Court declines to find that use of this data in the EIS left the decisionmaker without adequate information upon which to make a reasoned choice between alternative uses. B. Mitigation Measures The second main prong of plaintiffs’ attack focuses on ways in which the EIS proposes to mitigate noise level increases. Generally plaintiffs allege that the EIS overemphasizes the noise attenuation that will result from these mitigation measures. Plaintiffs particularly question the feasibility the proposed 20° right-hand turn for aircraft taking off to the northeast over Groveport, and the proposed “head-to-head” flight operations at the air cargo facility. The EIS states that on days when civilian cargo aircraft must, because of the prevailing winds, take off toward Groveport, a 20° right turn away from the village will be taken as soon as departing aircraft reach a safe altitude. At trial plaintiffs presented some testimony that this turn could not be accomplished at a safe altitude near enough to Rickenbacker to have any appreciable effect on noise levels in Groveport. Defendants rebutted this testimony by producing a number of pilots and air traffic controllers who established not only that the turn was feasible fairly close to the base, *1030 but also that it was currently a standard operating procedure for military flights at Rickenbacker. In fact, the current commander of the Air National Guard in Ohio testified that the noisiest and most cumbersome military aircraft now in use at Rickenbacker, the KC-135, routinely makes a 40° noise abatement turn upon taking off to the northeast. It thus appears that a 20° turn shortly after takeoff is entirely feasible, although the precise quantity of noise abatement attained by the maneuver is unclear. The EIS does not estimate the decibel reduction in DNL or SEL, if any, that may be expected from the proposed noise abatement turn, nor does it make clear whether the current and proposed noise level averages already reflect this noise abatement measure. If so, the executive summary describing the effect of mitigation procedures may be slightly misleading, because it implies that the 20° turn will decrease the proposed noise levels even further. Despite these rather minor deficiencies, however, the Court finds plaintiffs’ objections concerning this part of the EIS to be without merit. Equally unpersuasive are plaintiffs’ protestations regarding the implementation of “head-to-head” operations. The EIS states that: The Air Force can institute procedures for “head-to-head” operations, particularly for late-night civil operations. Given calm wind conditions and adequate control, aircraft can land from the southwest on Runway 05 and take off to the southwest on Runway 23. EIS, at IV-21. 13 This less-than-lucid description of the proposed mitigation measure created a great deal of confusion at trial, and plaintiffs introduced a large amount of testimony pointing out the rather obvious safety problems involved in requiring various aircraft to take off in, and land from, the same direction at the same time. Testimony by various defense witnesses made it relatively clear, however, that the term “head-to-head” describes a procedure whereby all takeoffs and all landings occur in the same direction, but at well staggered time intervals. One of Rickenbacker’s potential civilian tenants has proposed, for example, that almost all incoming air cargo flights would land at the base between 12:00 noon and 2:00 p.m. or 12:00 midnight and 2:00 a.m., that all outgoing flights would take off between 4:00 p.m. and 6:00 p.m. or 4:00 a.m. and 6:00 a.m. Under the proper weather conditions, this type of operation is routine. Plaintiffs argue nevertheless that the EIS overstates the percentage of time during which head-to-head operations are possible. According to the testimony developed at trial, the feasibility of such operations depends largely on wind direction and velocity. 14 Based on long-term government observations of the prevailing winds and historical runway designations at Rickenbacker, the EIS estimates that head-to-head operations can take place up to 70% of the time. A former Air Force meteorologist, testifying for plaintiffs, disputed this figure. He noted that planes attempting to take off from Rickenbacker to the southwest would experience a favorable headwind approximately 72% of the time. Conversely, aircraft attempting to land from a southwesterly direction would experience favorable headwinds only 28% of the time. Since both landings and takeoffs are ordinarily conducted into the wind, he conclud *1031 ed that head-to-head flying would be possible much less often than indicated in the EIS. He admitted on cross-examination, however, that his calculations did not take into account the percentage of time that winds at Rickenbacker would be so calm as to make takeoff and landing directions optional. He candidly acknowledged that the amount of time in which winds were calm could have a significant effect on the viability of head-to-head operations. The historical wind data included in the EIS, as well as the testimony of several defense experts, suggests that this omission on the part of plaintiffs’ meteorologist may explain at least part of the discrepancy between his estimates and those of the government. At any rate, in this respect plaintiffs have failed to show that the EIS estimate is seriously flawed. Here again this portion of the EIS is hardly free of weaknesses. It does not set forth in any precise manner the way in which estimates of feasible head-to-head flying time were calculated, and it does not explain which mitigation measures, if any, have already been figured into noise data on the proposed use. The Court cannot say, though, that these deficiencies make the entire EIS discussion of mitigation misleading as a matter of law. A better job could clearly have been done, but the job that was done is sufficient to allow the decisionmaker to take a “hard look” at potential mitigation techniques. C. Health Impact Although the Rickenbacker EIS does not contain a discrete section advising the Secretary about the potential health hazards of increased noise, it does identify a DNL of 65 dB or above as the level at which noise becomes “normally unacceptable.” At that level, according to EIS Table B-3, 15% of the population will report being highly annoyed, and adverse community reaction will be significant. A note appended to this table states: Research implicates noise as a factor producing stress-related health effects such as heart disease, high-blood pressure and stroke, ulcers and other digestive disorders. The relationships between noise and these effects, however, have not as yet been quantified. In addition, the EIS states elsewhere that while some studies have identified correlations between noise and physiological abnormalities in school children, such effects are minimal. In response to these statements, plaintiffs produced Dr. Karl D. Kryter, a recognized expert in the field of noise impact on health and human perception. 15 Dr. Kryter criticized the use of a 65 dB/DNL threshold, claiming that significant adverse health effects are observable at DNL levels well below that. He cited as proof a series of 1977 research studies done in Amsterdam by a well-known noise expert which indicated that cardiovascular disease, drug usage, and visits to doctors all began to increase at DNL levels as low as 55 dB. 16 Kryter contended that these studies should at the very least have been mentioned in the EIS. In rebuttal defendants produced a noise expert with equally impressive credentials, Dr. David Krantz. He criticized the methodology of the studies relied upon by Dr. Kryter, and maintained generally that health effects resulting from DNL levels below 65 dB were as yet unquantifiable. With all due respect to these two highly qualified doctors, this dispute, *1032 though important, boils down to little more than a difference of opinion between experts. In a dispute over the sufficiency of an EIS, the Court has neither the expertise nor the authority to resolve such a conflict. Kentucky ex rel. Beshear v. Alexander, supra, 655 F.2d at 718; Environmental Defense Fund v. Tennessee Valley Authority, supra, 492 F.2d at 468 n. 1; Mid-Shiawassee County Concerned Citizens v. Train, 408 F.Supp. 650, 656 (E.D.Mich.1976), aff’d 559 F.2d 1220 (6th Cir.1977). As the Ninth Circuit has observed: we note that disagreement among experts will not serve to invalidate an EIS. Indeed, “further studies, evaluation and analyses by experts are almost certain to reveal inadequacies or deficiencies.” Environmental Defense Fund v. Corps of Engineers, 342 F.Supp. 1211, 1217 (E.D. Ark.1971), aff’d., 470 F.2d 289 (8th Cir. 1972). The purpose of an EIS is to inform the decision makers of the environmental ramifications of the proposed action .... The statement need not achieve scientific unanimity on the desirability of proceeding with the proposed action. (Citations omitted); Life of the Land v. Brinegar, 485 F.2d 460, 472-473 (9th Cir. 1973), cert. denied, 416 U.S. 961, 94 S.Ct. 1979, 40 L.Ed.2d 312 (1974). Certainly the Rickenbacker EIS would have benefited from the addition of the Amsterdam studies and Dr. Kryter’s input, but the report cannot reasonably be expected to include references to each and every relevant scientific study or opinion. It is enough that the current state of scientific knowledge on a particular subject be represented fairly and in a reasonably balanced manner. See Mid-Shiawassee County Concerned Citizens v. Train, 408 F.Supp. at 656. The Court finds that this was done here, at least with respect to effects of noise on human health and perception. D. Classroom Disruption According to the EIS, three elementary and two secondary schools in the Rickenbacker area will fall into DNL noise contours above 65 dB upon implementation of the proposed air cargo plan. 17 The report states further: The principal impact on these schools will result from the estimated maximum number of thirty-one air cargo landings to take place in the early afternoon. Under present operational conditions landings would be from the north, over the affected schools, on approximately 30 percent of the days (one and one-half days per week, two hours per day for an average of three hours per week) .... ... This number [31 air cargo landings] is expected to be somewhat less than the number of military landings currently taking place during school hours under similar conditions. EIS at IV-17. The EIS concludes that since single-event peak levels for civilian and military aircraft are roughly equivalent, the increase in DNL levels for schools affected by air cargo traffic will be negligible. Table IV-7 shows that three of the affected schools will suffer a total increase in 12-hour daytime DNL of only one decibel; the other two schools will be unaffected in terms of DNL. 18 As indicated above, the averaging of noise level figures in this context is somewhat inappropriate; the Court suggests that the better measure of noise impact on schools is probably the number of peak level noise events that will be loud enough to disrupt classroom activities within a given time period. This figure does not appear anywhere in the EIS. Even assum *1033 ing that the number of classroom disruptions is 31 on the worst days when all air cargo landings are from the north, however, there is no evidence to show that this number is significantly greater than that already experienced by Rickenbacker area schools. In fact, a number of defense witnesses suggested that the opposite may be true. They opined that most classroom disruptions under current operations result from KC-135’s, and that almost all air cargo vehicles envisioned for use at Rickenbacker are significantly quieter than a KC-135. The plain implication is that total classroom disruptions under the proposed plan may be less than .the number experienced presently or in the recent past. The Court has a great deal of sympathy for those teachers from the Groveport-Madison system who testified at trial regarding the professional frustration they experience during noisy overflights. Unfortunately none of these witnesses could provide any solid evidence that the problem would be exacerbated by the introduction of air cargo flights. Without this type of evidence, the government’s failure to document the precise number of classroom disruptions resulting from the proposed use does not amount to a material omission. E. Sleep Disturbance The greatest single environmental impact occasioned by the proposed air park will unquestionably be on sleep. With the proposed introduction of approximately 31 air cargo landings between midnight and 2:00 a.m., and a corresponding number of takeoffs between 4:00 a.m. and 6:00 a.m. five nights a week, Groveport residents are understandably concerned about whether or not they will be awakened by the inevitable overflights and, if so, how often. A sizeable portion of the EIS’ is devoted to potential sleep disturbance, but plaintiffs claim that it does not even begin to give the reader a true feel for the magnitude of the problem. The Court agrees. In order to place the sleep disturbance dilemma in its proper perspective, the decisionmaker must be made aware of the fact that regular late night aviation is a heretofore unknown phenomenon in the Rickenbacker area. Numerous Groveport residents testified that military aircraft at the base have almost never flown after 11:00 p.m., at least not for the last two or three years. Whether or not significant late night flying took place prior to that is not entirely clear from the record, but the testimony of several older Groveport residents suggests that the Air Force has almost always curtailed its flight operations between 11:00 p.m. and 6:00 a.m. A few of the residents indicated that they had adjusted their sleep habits to coincide with this pattern. It thus became rather evident at trial that the proposed nighttime air cargo operations will, unlike daytime flights, impose an entirely new environmental impact on those within earshot of the base. Rather surprisingly, this fact does not appear anywhere in the text of the EIS. 19 In fact, a number of tables included within the report suggest just the contrary. In the chapter which describes the environment of the affected area generally, Table III — 3 purports to summarize current peri *1034 ods of military operation. 20 The table shows KC-135 landings between 10:00 and midnight, and between 1:00 a.m. and 4:00 a.m. five nights a week. Nothing in the table explains the frequency of these flights, so the unenlightened reader is naturally led to believe that nighttime military operations are regularly conducted at Rickenbacker. This conclusion is backed up by Table IV-3 which deals with the percentage *1035 of people awakened by overflights. 21 The table estimates the outdoor SEL levels for all aircraft at Rickenbacker, both military and civilian, and from this it projects the percentage of Groveport residents awakened by a particular overflight. The table properly implies that KC-135 overflights will generally cause more sleep disruption than those of most civilian air cargo planes, but it also supports the inference that military overflights during normal sleeping hours are common. *1036 Taken together, these tables could easily mislead the decisionmaker into thinking that the environmental impact of the proposed nighttime cargo operations is not significantly greater than that already experienced by area inhabitants. This is clearly not the case, and the remainder of the EIS does not sufficiently disabuse the reader of these notions. To some extent the increase in DNL shown in the noise contour data, coupled with knowledge of the nighttime “penalty” in DNL calculations, could indicate to the decisionmaker that more flights would occur during normal sleep periods under the proposed use than under existing operations, but the great magnitude of this difference should have been made plain in the EIS. The Court cannot fathom why this most important detail appears to have been buried under mountains of less relevant data. The EIS also lacks precision in its reporting of the frequency with which sleep disturbance may occur. The report notes that: From 10 to 45 percent of those people living in the noisiest Groveport areas directly under the approach flight path will be disturbed by overflights. The problem is significantly more severe in the summer when windows are open. The fraction of people disturbed drops by approximately five percent for every 1000 feet added to the distance from the runway centerline. Thus, there is little effect at. the western edge of Groveport. Under the summer conditions with windows open, approximately 25 percent of the population of Groveport, or approximately 800 people, may be expected to be disturbed from sleep on those days when there are landings from the north, i.e., one and one-half days per week on the average. In the winter this percentage will drop to approximately 15 percent or 485 people disturbed. Takeoff disturbance would be less because of noise abatement flight procedures. EIS at IV-5 and IV-7. Table IV-3 breaks these estimates down further, showing the percentage of sleep awakenings caused by each type of aircraft. 22 What neither the text nor the table makes abundantly clear, however, is that the percentage calculations contained therein refer to the amount of sleep disturbance resulting from each overflight. With up to 31 landings and 31 takeoffs each night, this presumably means that on nights when all takeoffs or all landings must be made over the northeast, some Groveport residents face the possibility of being awakened 31 times in the span of two hours. Since head-to-head operations will ordinarily be impossible one and one-half nights a week, this massive disruption may occur on the average of six nights per month. In addition, the figures themselves are somewhat misleading. One of the defense experts noted at trial that although each summertime landing may awaken approximately 800 people in Groveport, the individuals within each 800-person group may differ from flight to flight. In other words, 31 overflights in a two-hour period will not necessarily awaken 800 individuals 31 times; different flights will awaken different people, so the number of persons who are awakened at least once in a 31-over-flight span may be well in excess of 800. None of this is explained in the EIS. Instead the document informs the decision-maker that the maximum amount of time during which the outdoor noise level will normally exceed 85 dB(A) 23 is only three to *1037 five minutes in a two-hour period. EIS at IV — 7. This statistic is meaningless since the peak levels described above would be short in duration but spread out over the entire two-hour span. In other words, some individuals in Groveport would apparently be treated in the middle of the night to a 10-second noise event above 85 dB(A) on the average of once every four minutes for a two-hour period. If 31 B-747’s landed during this time, the effect on residents living near the eastern edge of Groveport would be somewhat akin to having a power mower started up on the lawn outside their bedroom window 31 times between midnight and 2:00 a.m. Compare EIS Tables III — 2 with IV-1. The reader thus cannot gain any real appreciation of the potential disruption simply by being told the number of minutes that aircraft noise will occur when all of the overflight peak level events are strung together. With the aid of four days of expert testimony and a number of reaccomplished figures, the Court has to some extent been able to draw rather tortured inferences from various diverse parts of the report, but it is doubtful that the decision-maker could have done the same simply by reading and rereading the final EIS. The EIS treatment of potential sleep disturbance is more notable for what it leaves out than what it includes. The decision-maker is never informed of the quantity of night flights that have traditionally landed or taken off from Rickenbacker. He is not told the number of times per night that the average Groveport resident can expect to be awakened by northerly overflights when such overflights are necessary. He cannot ascertain from the data presented how many people in Groveport or elsewhere will be awakened at least once on these “worst case” nights. Finally, the EIS fails to tell him whether sleep disturbance will diminish over time as residents become accustomed to periodic overflights, and whether this diminution, if any, is significant, or whether there are important physiological effects that can be expected from long-term exposure to noise-induced sleep disturbance. 24 Answers to at least some of these questions are vital if the decisionmaker is ever to be made fully aware of the sleep disturbance problem. Without them he cannot even begin to take a “hard look” at the environmental consequences of air cargo operations. In light of NEPA’s command that the relevant agencies explore the unavoidable environmental consequences of their proposed actions “to the fullest extent possible,” the Court is hard-pressed to believe that the Air Force has fulfilled its duty here. The Court is well aware that there are flaws in almost any environmental study, but it believes nonetheless that this particular portion of the EIS has so much room for improvement that it is inadequate as a matter of law. There is simply not enough relevant information concerning sleep interruption in the EIS to allow the decisionmaker to weigh fairly the environmental pros and cons of the air cargo proposal. F. The EIS as a Whole Viewed in its entirety, the EIS evidences a good faith effort on the part of the Air Force to identify and evaluate the environmental impact of the proposed air industrial park at Rickenbacker. With one exception, the report deals sufficiently with the adverse consequences of increased noise in the surrounding area. Because the one exception involves such a profound potential environmental problem, however, the Court is *1038 unwilling to approve in full either the document or the final decision based thereupon until the question of sleep disturbance is reviewed more thoroughly. IV Having held that a portion of the Rickenbacker EIS is inadequate to satisfy the provisions of NEPA, the Court must next turn to the question of an appropriate remedy. Where a federal agency has failed to give due consideration to the environmental consequences of a proposed action, the courts as a first step generally remand the proceedings to the agency for further study and publication. City of Romulus v. County of Wayne, 392 F.Supp. 578 (E.D. Mich.1975), vacated as moot, 634 F.2d 347 (6th Cir.1980); City of New York v. United States, 337 F.Supp. 150 (E.D.N.Y.1972), supplemented, 344 F.Supp. 929 (1972); see also, Red Line Alert v. Adams, 10 Environ.L. Reptr. 20314 (D.Mass.1980). It should be relatively clear from the abovegoing discussion that such a procedure is called for in the present case. The Court has found numerous flaws in various parts of the final EIS, and so apparently has the Air Force. The fact that much of the data in the document has had to be “reaccomplished” casts some doubt on the accuracy of the EIS as a whole. Ideally, then, the Secretary should reconsider the entire report, including the revised figures and noise contours contained in the addendum compiled just prior to trial. Only a portion of the EIS has been found insufficient as a matter of law, though, and the Court would be overstepping its bounds to require the document to be completely rewritten. Defendants need only compile a supplemental EIS regarding the impact of the proposed air industrial park on the sleeping habits of those individuals how living in the Rickenbacker area. The Court will leave to the preparers the precise details of production and distribution, but it is expected that the new supplement will in every respect comply with the procedural prerequisites of NEPA and its supporting regulations, including those pertaining to comment by the public. The document should include the pertinent reaccomplished statistics, and should also inform the Secretary that significant recalculations have been made in areas other than sleep disturbance. The Court of course cannot tell the federal defendants how best to study the problem, but hopefully the supplemental statement will address the issues raised above, and will include a discussion of the potential disruption of sleep in areas other than just Groveport. Once this has been done, defendants are free to submit the supplement, along with public comments thereon, to the Secretary for a final decision on the proposed use. It was urged by the government at closing arguments that a remand would be largely meaningless, since reconsideration of the noise impact of the air cargo facility is hardly likely to change the Secretary’s final decision. The Court recognizes that the remedy it can offer here is limited, but as Judge Friendly has cogently observed in a similar context: To permit' an agency to ignore its duties under NEPA with impunity because we have serious doubts that its ultimate decision will be affected by compliance would subvert the very purpose of the Act and encourage further administrative laxity in this area. The systematic investigation ... which NEPA requires may well reveal substantial environmental consequences and ... may possibly provide an economic alternative to [the proposed project, or] compel further consideration of its propriety and necessity. City of New York, supra, 377 F.Supp. at 160. With respect to air cargo activity in the interim, the Court believes that it would be highly inequitable to allow defendants to proceed unfettered in implementing the proposed air cargo operations pending com *1039 pletion of the EIS supplement. In deciding whether to stay further agency action until the requirements of NEPA have been met, the courts are guided by traditional principles of equity and injunctive relief. See City of Romulus, supra, 392 F.2d at 594. Here those principles militate in favor of an injunction, since: the more time and resources [defendants] are allowed to invest in this project, the greater becomes the likelihood that compliance with section 102 of NEPA, and the reconsideration of the project in light of the provisions of section 101, will prove merely to be an empty gesture .... Accordingly, “unless the plaintiffs receive now whatever relief they are entitled to, there is danger that it will be of little or no value to them or to anyone else when finally obtained.” (Citations omitted); Environmental Defense Fund v. Tennessee Valley Authority, 468 F.2d 1164, 1183 (6th Cir.1972), cert. denied, 414 U.S. 1036, 94 S.Ct. 535, 38 L.Ed.2d 327 (1973), quoting Latham v. Volpe, 455 F.2d 1111, 1117 (9th Cir.1971). In addition, since relatively few resources have been invested in the project to date, the harm to defendants in granting an injunction is relatively slight. The delay caused by this litigation could possibly result in the loss of rental revenue or even the loss of a prospective tenant, but this harm, if it exists, can easily be mitigated by speedy compliance with the terms of NEPA and its regulations. On the other hand, the only air cargo operations directly affected by future revisions in the EIS are those which are scheduled to occur at night. The Court sees no reason why daytime operations must also be enjoined. The commencement of daytime flying would not represent the type of investment of resources that would be likely to prejudice the Secretary’s ultimate decision, since any civilian air carrier who wished to commence daytime flights would have to do so with the express understanding that nighttime operations could be ruled out permanently, and that the Secretary could select another alternative for use of the base. If RPA is able to find a tenant willing to accept these contingencies, then it may begin to implement daytime air cargo operations immediately. Defendants should not be able to ignore the mandate of NEPA and still proceed with the proposed agency action, but neither should they be unduly penalized where the infraction pertains to only a portion of the proposal. Conclusion For the foregoing reasons, the Court finds that the Department of the Air Force has violated Sections 102(2)(C)(i) and (ii) of NEPA, 42 U.S.C. § 4332(2)(C)(i) and (ii), in that it has not adequately explored the impact of air cargo operations on human sleep cycles in the area surrounding Rickenbacker ANGB. In all other respects the final EIS prepared by the Air Force is legally sufficient. Until such time as a supplemental EIS has been prepared in accordance with the guidelines set forth herein, and a final decision on the basis of the supplement has been rendered by the Secretary, the parties should, with some exceptions, be bound by the substance of the interim order to which defendants agreed at the close of the trial. Accordingly, defendants shall not cause or permit air cargo operations involving major structural alterations to take place at Rickenbacker, nor shall they allow title to the excess lands or facilities at the base to be transferred. Defendants also shall not introduce any air cargo operations to Rickenbacker pursuant to the joint use agreement or interim lease between the hours of 10:00 p.m. and 7:00 ,a.m. Defendants may, however, allow air cargo operations to commence during all other hours, provided that such operations do not require major structural alterations or irrevocable commitments of fixed resources at the base. To the extent that they are inconsistent with this opinion and order, all previous orders of *1040 the Court restricting implementation of the joint use proposal at Rickenbacker are hereby rescinded. So ORDERED. Footnotes [1] The October draft is hereinafter referred to as the “final EIS.” It does not include the “reaccomplished” data prepared in February 1982. See infra. [2] This amended complaint was filed even before the Secretary issued the draft EIS or held public hearings on it. [3] The EIS defines “DNL” as “the 24-hour average sound level, in decibels (dB), obtained after the addition of a 10 dB penalty for noise occurring between 10:00 p.m. and 7:00 a.m.” EIS at III — 5. DNL values for airports are calculated by estimating the noise contributions from each significant aircraft operation and summing them over a 24-hour period. See EIS Appendix A. [4] In choosing a threshold DNL figure, the preparers of the EIS relied upon a HUD publication entitled “Guidelines for Considering Noise in Land Use Planning and Control,” reprinted at Appendix B of the EIS. According to Table B-l contained therein, HUD, DOT, and EPA recognize a DNL of 55 dB as a goal for acceptable outdoor noise exposure. DNL above 65 dB is deemed to be significant and “normally unacceptable.” Plaintiffs charge that this 65 dB threshold is itself too high because significant adverse health effects allegedly begin at a DNL of 55 dB or below. See text at 1031-1032. [5] Equivalent Sound Level (Leq) contours are used to describe the average noise level over a specified time period. Leq is measured in decibels, and is roughly equivalent to DNL without the 10 dB nighttime penalty described in note 3. See Department of the Air Force, Reaccomplished Noise Data and Maps Used in the Final EIS for the Proposed Joint Use of Rickenbacker ANGB (hereinafter Reacc. Data) at III — 14 (Feb. 1982). [6] EIS, Fig. IV-2; see also Reacc. Data, Atch. 5. [7] EIS, Fig. IV-3; see also Reacc. Data, Atch. 8. [8] The 9- and 15-hour Leq contours included in the reaccomplished data are a bit more helpful, since the time periods measured therein more closely approximate those of the greatest concem here. They are still potentially misleading, however, because the contours shown for existing nighttime operations are probably produced by flights on either end of the 9-hour continuum (between 10:00 p.m. and 11:00 p.m. or 6:00 a.m. and 7:00 a.m.), whereas the contours shown under the proposed plan would be generated by flights in the middle of the 9-hour period (i.e., between 12:00 midnight and 6:00 a.m.). At any rate, the reaccomplished figures were never formally submitted to the Secretary, and they played no part in his decision. [9] See Tables IV-4 through IV-6. The reaccomplished figures show significantly less probable impact. See Reacc. Data, Atch. 15, 17-19. [10] Among these alternate methodologies is a means by which the decisionmaker can determine the peak noise level of particular overflights. See note 12, infra. [11] See EIS at IV-5 and IV-7; see also discussion infra, at 1036. [12] Single Event Noise Level (SEL) is the term used to describe the noise exposure contribution of a single short-duration noise event. It is expressed in A-weighted decibels, or dB(A)’s, as if the entire event had taken place in the span of one second. As such, it is useful in expressing the comparative peak noise levels produced by different aircraft. See EIS, Appendix A. [13] For purposes of takeoffs and landings, the runway is named according to the direction in which the plane using it is headed. Each runway therefore has two distinct designations corresponding to opposite points on the compass. When aircraft using Rickenbacker are proceeding toward or away on a heading of 050°, they are said to be using Runway 05. Conversely, when they are proceeding on a heading of 230°, they are said to be using Runway 23, even though both designations refer to the same slab of concrete. [14] Safety dictates that most takeoffs and landings be conducted into a headwind. Tailwind operations are possible, but only if the tailwind component on a given runway is minimal, usually less than ten knots. [15] This apparently is the same Dr. Kryter whose testimony was relied upon by the court in City of Romulus v. County of Wayne, 392 F.Supp. 578, 592-594 (E.D.Mich.1975), vacated as moot, 634 F.2d 347 (6th Cir.1980). [16] Knipschild, Medical Effects of Aircrañ Noise: Community Cardiovascular Survey, 40 Int. Archives of Occupational & Environmental Health (hereinafter Archives) 185 (1977); Knipschild, Medical Effects of Aircraft Noise: General Practice Survey, 40 Archives 191 (1977); Knipschild & Oudshoorn, Medical Effects of Aircrañ Noise: Drug Survey, 40. Archives 197 (1977). [17] The reaccomplished figures include three other schools which will probably fall into this range upon implementation of air cargo operations. None of these three schools will apparently experience a DNL above 66 dB, however. See Reacc. Data at III — 9; see also Atch. 18. [18] See also, Reacc. Data at Atch. 18. The reaccomplished figures show even less overall DNL increase. [19] As noted earlier, the EIS Executive Summary refers to “the introduction of a significant number of night and early morning flights.” To one with extensive outside knowledge of present and proposed operations at Rickenbacker, this could be taken as an indication that regular nighttime flying is a novel feature of the proposed use. The language is ambiguous at best, however, and it would be difficult, if not impossible, for the average lay reader to conclude solely on the basis of this passage that nighttime operations at the base were heretofore rare. Defendants argue that this fact can be inferred from remarks in the public comment section of the EIS. They have not pointed to any specific examples therein, though, and even if they had, the Court believes that the unique nature of the proposed nighttime flying is so important that it should have been emphasized in the main body of the report. [20] This table is reproduced below. [21] This table is reproduced below. [22] See note 21, supra. [23] According to the testimony of Dr. Peter Daley, who supervised the preparation of the EIS, noise intensity figures can be presented either in straight decibels (dB) or in A-weighted decibels (dB(A)). The former unit measures absolute energy output, while the latter is weighted to account for differences in human audiometric perception. The units are used almost interchangeably throughout the EIS, and the Court’s use of a dB(A) figure here is consistent with its use in the report. [24] As noted above, the precise link between high DNL exposure and physiological disorders is far from clear; only now have scientists begun to quantify the connection. This may not be true of the correlation between health and chronic sleep disturbance, though. There is nothing on the record to indicate that the preparers of the EIS ever tried to locate research on this subject. The Court believes that such research, if it exists, would be highly relevant to the issues at hand.