State of Wis. v. Weinberger, 578 F. Supp. 1327 (W.D. Wis. 1984)

U.S. District Court for the Western District of Wisconsin - 578 F. Supp. 1327 (W.D. Wis. 1984)
January 31, 1984

578 F. Supp. 1327 (1984)

STATE OF WISCONSIN, Plaintiff,
and
County of Marquette Michigan, Intervening Plaintiff,
v.
Caspar W. WEINBERGER, individually and as Secretary of the Department of Defense, the United States Department of Defense, John F. Lehman, Jr., individually and as Secretary of the Department of the Navy, and the United States Department of the Navy, Defendant.

No. 83-C-672-C.

United States District Court, W.D. Wisconsin.

January 31, 1984.

*1328 *1329 *1330 *1331 Shari Eggleson, Asst. Atty. Gen., Madison, Wis., for the State of Wis.

Patricia Micklow, Chief Civ. Counsel, Marquette County, Marquette, Mich., for intervening plaintiff County of Marquette, Michigan.

John Byrnes, U.S. Atty., Madison, Wis., for defendant.

 
            TABLE OF CONTENTS
  I. POSTURE OF THE CASE ........................ 1332
 II. SCOPE OF JUDICIAL REVIEW ................... 1333
III. FACTS ...................................... 1334
     A. Administrative Record ................... 1334
        1. Background of controversy ............ 1334
           a. Navy ELF proposals prior
              to 1981 ........................... 1334
           b. Environmental documentation
              for pre-1981 ELF proposals ........ 1335
              1) 1972 environmental impact
                 statement ...................... 1336
              2) 1975 supplemental environmental
                 impact statement ............... 1337
              3) National Academy of
                 Sciences study ................. 1337
              4) 1977 environmental impact
                 statement ...................... 1338
        2. 1981 reactivation of ELF communication
           proposal ............................. 1340
           a. Decision .......................... 1340
           b. Scope of current proposal ......... 1340
           c. Environmental impact assessment
              for the current proposal .......... 1341
        3. Navy-supported research in
           biological effects ................... 1341
           a. Pensacola primate experiments ..... 1341
           b. UCLA monkey behavior
              study ............................. 1342
           c. Continued studies ................. 1342
              1) bird migration studies ......... 1342
              2) multigenerational mice
                 studies ........................ 1342
              3) slime mold studies ............. 1343
        4. IIT Research Institute activities .... 1343
        5. Ecological information ............... 1343
           a. Forest Service wildlife
              surveys ........................... 1343
           b. Ecological monitoring
              program ........................... 1344
           c. Electromagnetic field
              measurements ...................... 1344
        6. Scientific literature ................ 1344
           a. Wertheimer-Leeper articles ........ 1345
           b. Rhode Island epidemiology
              study ............................. 1345
           c. Stockholm epidemiology
              study ............................. 1345
           d. Delgado study ..................... 1345
           e. Letters to the Editor ............. 1346
           f. Other publications ................ 1347
              1) World Health Organization
                 report ......................... 1347
              2) Montana Department of
                 Natural Resources and
                 Conservation report ............ 1347
        7. Navy review of scientific
           information .......................... 1348
           a. Environmental Review
              Committee ......................... 1348
 
*1332
           b. IIT Research Institute
              literature review ................. 1349
              1) Wertheimer-Leeper
                 articles ....................... 1349
              2) Rhode Island epidemiology
                 study .......................... 1349
              3) Stockholm epidemiology
                 study .......................... 1350
              4) Delgado study .................. 1350
              5) Letters to the Editor .......... 1350
           c. Other reviews ..................... 1350
           d. IIT Research Institute public
              interest review ................... 1350
        8. Biological effects information in
           administrative record in fall, 1981 .. 1350
     B. Trial Testimony ......................... 1351
        1. Navy review of scientific
           information in conjunction with
           its 1981 decision to reactivate
           Project ELF .......................... 1351
        2. Post-1977 scientific research ........ 1352
           a. Research studies not included
              in the administrative record ...... 1352
              1) Blackman and Adey window
                 research ....................... 1352
                 a) the study ................... 1352
                 b) expert assessment ........... 1353
              2) Batelle Laboratories
                 studies ........................ 1353
                 a) the study ................... 1353
                 b) expert assessment ........... 1353
              3) Slime mold studies ............. 1353
                 a) the study ................... 1353
                 b) expert assessment ........... 1354
              4) Additional research ............ 1354
                 a) the studies ................. 1354
                 b) expert assessment ........... 1354
           b. Post-1977 scientific literature
              in the administrative record ...... 1354
              1) epidemiology studies ........... 1354
              2) Delgado study .................. 1355
              3) Letters to the Editor .......... 1355
 IV. OPINION .................................... 1355
     A. Scope of the National Environmental
        Policy Act .............................. 1355
        1. Exemption of presidential
           decision ............................. 1355
        2. Exemption of activities begun
           before January 1, 1970 ............... 1355
     B. Standard of Review of an Agency
        Decision not to Prepare a Supplemental
        Environmental Impact Statement .......... 1356
     C. Navy's Obligation to File a Supplemental
        Environmental Impact
        Statement ............................... 1357
        1. Substantiality of changes in the
           proposed action ...................... 1357
           a. Maintenance and modernization
              activities ........................ 1358
           b. Reduction of environmental
              impact ............................ 1358
           c. Conversion to a fully operational
              system ............................ 1359
        2. Significance of new information
           of biological effects of electromagnetic
           radiation ............................ 1360
           a. Environmental significance of
              the new information ............... 1361
           b. Degree of care with which
              the Navy considered the new
              information ....................... 1363
           c. Sufficiency of Navy's explanation
              of its decision not to file
              a supplemental environmental
              impact statement .................. 1364
     D. Remedy .................................. 1364
  V. ORDER ...................................... 1365
 
I. POSTURE OF THE CASE

CRABB, Chief Judge.

This is a civil action in which plaintiff State of Wisconsin seeks a preliminary and permanent injunction of any additional work by defendants on Project ELF submarine communication facilities in northern Wisconsin and the upper peninsula of Michigan until defendants have prepared a supplemental environmental impact statement. Plaintiff contends a supplemental environmental impact statement is required by the National Environmental Policy Act of 1969, 42 U.S.C. §§ 4321-4335 and implementing regulations.

Marquette County, Michigan has been granted leave to intervene permissively, with its participation limited to the issue whether a supplemental environmental impact statement is necessary for the proposed ELF Project in northern Wisconsin and the upper peninsula of Michigan because of significant new information concerning biological effects of extremely low frequency electromagnetic radiation.

*1333 A hearing on plaintiffs' motion for a preliminary injunction was consolidated with a trial on the merits of the case. At the conclusion of the trial, I denied plaintiffs' motion for a preliminary injunction. I held that the question of ultimate success on the merits was a close one and plaintiffs had not shown a reasonable likelihood of prevailing; that the main thrust of plaintiffs' case was directed to the operation of the ELF facility and the biological consequences of that operation and plaintiffs had not shown that defendants' continued work on the construction aspects of the Wisconsin and upper Michigan ELF facilities constituted irreparable harm; that the balance of harms was approximately even; and that the public interest would not be served or disserved by the issuance of an injunction at that time. I stated that construction work such as the preparation of the sites, erection of buildings, upgrading of existing buildings, and installation of antennae could go forward if defendants were willing to assume the risk that they would subsequently be enjoined from completing the work and putting the facilities into operation.

 
II. SCOPE OF JUDICIAL REVIEW

As a threshold matter, I must consider defendants' motion to limit judicial review of the Navy's compliance with the National Environmental Policy Act to a review of the administrative record.

There are two exceptions to the ordinary rule that judicial review of agency actions is confined to a review of the record that was before the agency at the time of the decision: where the administrative record does not provide an adequate explanation of the administrative actions, and in a complex, technical case where it is necessary to look beyond the record to determine whether the agency considered all factors relevant to its decision.

In the first instance, the reviewing court may receive evidence to clarify the administrative record. Camp v. Pitts, 411 U.S. 138, 142-43, 93 S. Ct. 1241, 1244, 36 L. Ed. 2d 106 (1973) (per curiam); Citizens to Preserve Overton Park, Inc. v. Volpe, 401 U.S. 402, 420, 91 S. Ct. 814, 825, 28 L. Ed. 2d 136 (1971); Environmental Defense Fund, Inc. v. Costle, 657 F.2d 275, 285 (D.C.Cir.1981); ASARCO, Inc. v. United States Environmental Protection Agency, 616 F.2d 1153, 1159-60 (9th Cir. 1980). It is well established that an agency's decision not to issue an environmental impact statement must be supported by a record sufficiently developed to permit judicial review. City of West Chicago, Illinois v. United States Nuclear Regulatory Comm'n, 701 F.2d 632, 651 (7th Cir.1983). If the basis of the agency decision cannot be discerned from the administrative record, the reviewing court has the option of remanding to the agency for clarification of the agency action or eliciting explanatory information from agency officials in the form of testimony or affidavits. Environmental Defense Fund, Inc. v. Costle, 657 F.2d at 285; City Federal Savings & Loan Ass'n v. Federal Home Loan Bank Board, 600 F.2d 681, 689 (7th Cir.1979).

In the second instance, where the case is complex, the court may be unable to evaluate the adequacy of the information considered by the agency unless it looks beyond the administrative record to determine what the agency ignored. ASARCO, Inc. v. United States Environmental Protection Agency, 616 F.2d at 1160.

This case implicates both exceptions to the ordinary rule confining judicial review to the administrative record. Because the basis for the Navy's decision not to prepare a supplemental environmental impact statement is not readily apparent from the administrative record, I will consider trial testimony from Navy officials explaining the basis for their decision. In addition, because this case requires a determination whether the Navy considered relevant new information in deciding not to prepare a supplemental environmental impact statement, I will review information outside the administrative record to decide whether new information became available after preparation of the 1977 environmental *1334 impact statement and whether that new information is so significant as to require the preparation of a supplemental environmental impact statement.

 
III. FACTS 
A. Administrative Record

The administrative record in this case consists of five volumes of information collected throughout the history of the Navy's ELF communication proposals. In addition to the Navy's environmental impact statements for its ELF proposals and a 1977 National Academy of Sciences study, which make up almost half the administrative record, the record includes 1) information about the current proposal; 2) reports from Navy-supported research; 3) reports from the IIT Research Institute with which the Navy has contracted for environmental monitoring; 4) information about ecological effects of Project ELF; 5) scientific literature in the area of electromagnetic field effects; and 6) minutes from the Navy's Environmental Review Committee.

I find that the administrative record contains evidence of the following:

 
1. Background of controversy 
a. Navy ELF proposals prior to 1981

Since World War I, the Navy has used very low frequency radio signals to communicate with its submarines from fixed shore-based transmitters.[1] Although very low frequency signals can penetrate seawater, they can do so only to a depth of 30-40 feet. To receive a very low frequency transmission, a submarine must either operate at a shallow depth or send a buoyed antenna to receiving range while operating at a low speed. In either case, the submarine becomes more vulnerable to detection. To reduce this vulnerability, the Navy has been looking for alternative communication systems.

At least since 1958, the Navy has been studying and developing military communication systems that utilize "extremely low frequency" electromagnetic radiation. "Electromagnetic radiation" refers to the waves emitted from an electric charge in motion, which has both electric and magnetic components. (Any energy that moves through space at the speed of light is a form of electromagnetic radiation.) Extremely low frequency electromagnetic radiation refers to nonionizing electrical waves in the frequency range from 30 to 300 Hz with wavelengths from 6200 to 620 miles long. These very long wavelengths can penetrate seawater to depths of approximately 300 to 400 feet.

The primary uses of extremely low frequency electromagnetic radiation are commercial power systems and the naval communication system that is the subject of this lawsuit. Commercial power systems operate at a frequency of 60 Hz in the United States and Canada and at frequencies of 40-50 Hz in much of Europe. Initially, the Navy's ELF communication system was intended to operate at 45 to 75 Hz. It was changed to a frequency range of 72 to 80 Hz.

Throughout the rest of this opinion, I will refer to electromagnetic radiation in the extremely low frequency range by writing it out: extremely low frequency. To identify the Navy's communication system, I will use the abbreviation for extremely low frequency: ELF.

On July 1, 1968, the Navy announced plans to construct an ELF communication test facility at Clam Lake in northern Wisconsin. This facility, which includes 28 miles of overhead antenna, became operational *1335 in 1969, achieving full capacity in 1970.

From 1969 through 1972, the Navy conducted research in conceptual designs for ELF communication systems and in environmental effects. The Navy then proposed Project Sanguine: a three-phase plan for research and development at the Wisconsin test facility for the period from 1972-1976, consisting of a concept formulation phase, a validation phase, and a full-scale development phase. As of 1972, the Navy planned to construct two transmitters and a 6300 square mile grid of buried antenna and to have the system operational by late 1976.

During 1973 and the first quarter of 1974, the Navy conducted environmental studies, focusing on interference mitigation activities, biological experiments, ecological surveys, and development of environmental site evaluation techniques, all as part of the concept formulation phase. In 1975, the Navy entered the validation phase of Project Sanguine. The Navy never went on to the full-scale development phase of Project Sanguine. Instead, in 1977, it proposed Project Seafarer, an experimental ELF communication system with two test facilities, the one at Clam Lake and an additional test facility with a transmitter station, an adjacent transmitter control center, and 130 miles of buried antenna cables, to be located in the upper peninsula of Michigan. As part of an effort to reduce the size of the fully operational system, the Navy planned experimental operations in the northern Wisconsin and upper Michigan test facilities to explore the system's capabilities. At the time, the Navy was proposing that when Seafarer was fully operational, it would have five surface transmitter stations and 2400 miles of buried antenna cables dispersed in a 4000 square mile area.

The Navy chose upper Michigan as the location for the new Project Seafarer test facility because of the transmitting qualities of the dry granite bedrock in northern parts of Wisconsin and Michigan, because of lower costs compared with other sites, and because the location would make it possible to operate synchronously with the Wisconsin test facility. Synchronous operation means that a signal is issued from both sites at the same instant, increasing the strength of the signal. A synchronous signal is necessary to reach distant submarines.[2]

For fiscal year 1978, President Carter supported the allocation of $20.1 million for Project Seafarer research and development with the understanding that none of this money would be used for work on a Michigan site. At that time, President Carter planned to review personally any future funding requests for deploying a Seafarer system in Michigan.

On February 16, 1978, President Carter decided to terminate Project Seafarer for an indefinite period. He recommended that the Navy study possible alternate sites and inform the public officials and residents of Michigan and Wisconsin of the need for an ELF submarine communication facility.

Following the President's lead, Congress did not fund Project ELF, as it came to be known, and the project entered a dormant phase, lasting from approximately April, 1979, to March, 1981.

 
b. Environmental documentation for pre-1981 ELF proposals.

From the time the Navy first announced its interest in an ELF communication system, public officials and other interested persons have voiced concerns about the implications of continuously exposing humans, animals, and plants to extremely low frequency electromagnetic radiation, a subject that until recently has received little, if any, scientific study.[3] These concerns are *1336 threefold, focusing on the potential effects of 1) the frequencies to be used by the Navy; 2) the power intensity of the electric fields produced by operation of the ELF antennae; and 3) the intensity of the magnetic fields generated by the electric current passing through the antennae.[4] Acknowledging the public's interest in the subject of biological effects, the Navy addressed it in its 1972 environmental impact statement, its 1975 supplemental environmental impact statement, and its 1977 environmental impact statement to which it attached a 1977 National Academy of Sciences study it had commissioned. These reports are included in full in the administrative record and are summarized below.

 
1) 1972 environmental impact statement

In the April, 1972 final environmental impact statement for the first three phases of Project Sanguine, the Navy concluded that there would be no significant adverse environmental impact from any of the three developmental phases of the project or from the construction and deployment of a fully-operational system in late 1976 at the Wisconsin test facility.

From biological studies it had funded, the Navy was satisfied that the only biological effects that had been observed had occurred at electromagnetic field intensities much higher than the maximum electromagnetic fields associated with the Sanguine system. Further, from its assessment of the available literature in the areas of effects on chromosomes, fertility of rats, growth and development, behavior, biorhythms, soil microbiology, biochemical systems, and electric safety, the Navy concluded that there was only a small chance that similar Sanguine fields would have any deleterious biological effect.

The Navy noted the history of electric power lines: "Long experience with low-level ELF fields such as those produced by existing commercial power lines, and which would be produced by a Sanguine System, has shown thus far no untoward biological effects are to be expected." The Navy recognized the importance of research concerning biological effects at frequencies far different from those produced by an ELF antenna: "Its value lies in the identification of potential mechanisms by which biological reactions might take place. Such knowledge is helpful in establishing hypotheses for effects at ELF."

The Navy stated that it would proceed with its research on biological effects, with "re-assessment on a continuing basis from additional research during the anticipated 4-year validation and development phases of the project."

Among the public comments submitted to the Navy in response to the 1972 environmental impact statement were those of the Bock Committee, an ad hoc committee formed in 1969 by the governor of Wisconsin and headed by Robert Bock, dean of the *1337 graduate school of the University of Wisconsin-Madison. This committee expressed concern that the ecological studies at the Wisconsin test facility were of little value, because there was no environmental data from the preconstruction period to serve as a baseline. It contended that the results of studies conducted at the test facility were not informative, because the test facility antennae were operating only for short periods and only intermittently, creating conditions quite different from those that might be expected of a fully operational system. The committee suggested research directed at subtle long-term effects, synergistic conditions, and multiple environmental stresses. Finally, the committee emphasized the importance of acknowledging the difficulties of extrapolating from laboratory to real-world conditions. The committee concluded that, because of the inadequacy of the existing research, the Navy had no basis for determining whether the system could operate without major, long-term environmental impact.

 
2) 1975 supplemental environmental impact statement

In 1975, before deciding to proceed with the validation phase of Project Sanguine, the Navy released a supplement to the 1972 environmental impact statement. In this supplement, the Navy updated its evaluation of the biological and ecological effects of extremely low frequency electromagnetic radiation and concluded:

 
The results of two additional years of testing and trends observed in continuing experiments provide additional factual data upon which the risk of biological and ecological reactions from a Sanguine System can be assessed. That risk continues to appear acceptably small.
 
3) National Academy of Sciences study

In 1975, the Navy requested the National Academy of Sciences to study the possibility of harm to plants, animals, and human beings from the electromagnetic fields produced by the ELF communication system. The committee published its report in August, 1977. The Navy incorporated this report in its 1977 final environmental impact statement.

The committee identified two electric shock hazards, one of which could be alleviated by altering the design of ground terminals, the second of which could be minimized by developing a rapid and reliable detection system.

The committee evaluated reports from Navy-sponsored experiments and the published literature on biological effects of electromagnetic radiation. It found no likelihood of genetic hazards from ELF fields, no ground for concern about the possible effects of electromagnetic radiation on fertility, growth, and development, little ground for concern about adverse effects on cell growth and division, no effects on human triglyceride concentrations, and no reason for concern about possible effects on circadian rhythms.[5]

In several areas, the committee recognized that some biological effects other than electric shock might occur. Noting the limited nature of the data on biological effects and inconclusive research results, the committee recommended further research in many areas, including effects of extremely low frequency electromagnetic radiation on 1) behavior and adaptation of fish sensitive to weak electric fields; 2) orientation and comb-building behavior of bees; 3) magnetotactic bacteria (bacteria that move toward the earth's magnetic north pole); 4) bird navigation and migration patterns; and 5) mammalian neurophysiological behavioral, and neurochemical effects. The committee concluded that, if an ELF communication system were to be built and operated, continued research on *1338 biological effects of ELF electromagnetic radiation was essential.

As part of its study, the National Academy of Sciences commissioned Ross Adey to do a chapter on the anatomy and biophysics in brain cells in weak extremely low frequency fields. In this chapter, Adey noted his recent observation of amplitude and frequency windows in ELF modulation of VHF (147 MHz) and UHF (450 MHz) radio frequency fields and cited a study of his own on this subject (Adey & Bawin, 1977). Adey stated that the observations were suggestive. He did not expand further on either the concept or significance of "windowing."

The committee reviewed epidemiology studies performed in Eastern Europe and the U.S.S.R. that showed a correlation between biological effects and electromagnetic fields associated with electric power installations. Although it characterized these studies as difficult to evaluate because of the absence of descriptions of study methods, of discussions of methods of selecting controls and characteristics of the controls, and of quantified data, the committee acknowledged the potential value of epidemiological research for predicting the effects of electromagnetic radiation.

With regard to biological effects, the committee concluded that there was no good evidence of any significant effects:

 
The Committee's considered opinion is that such fields will not cause significant and adverse biologic disturbance.... Although the available data are extremely limited, some effects not related to shock may occur.... Thus, like many other perturbations introduced by man, this involves some uncertainty and calls for continuing research.
 
4) 1977 environmental impact statement

In conjunction with its 1977 Project Seafarer proposal the Navy released a draft environmental impact statement for comment in February, 1977, and filed a final statement in December, 1977. The Navy later sent the National Academy of Sciences findings to the Council on Environmental Quality and distributed them as a supplement to the organizations and agencies that had received the draft environmental impact statements.

In the final environmental impact statement, the Navy summarized the state of knowledge concerning biological effects of electromagnetic radiation: "Totally acceptable scientific knowledge in this regard is not available now to either prove or disprove many important issues, and is unlikely to be in the very near future despite constantly increasing interest and research."

From what it characterized as casual observations, conclusions drawn from experience with existing ELF electromagnetic fields, and analysis of scientific experiments in the area, the Navy reached the overall conclusion in the 1977 environmental impact statement that "No adverse electromagnetic influences are expected on humans, animals or plants due to Seafarer operations." It stated that "There are research results suggesting the existence of some rather diverse and subtle effects from particular types of ELF exposures. However, no adverse biological effect has been completely substantiated or seems highly probable from a Seafarer-like operation."

Throughout this environmental impact statement, as in its predecessors, the Navy continued to rely on the hazard-free experience with electric power systems.

 
There is nearly one hundred years of experience living with 50 and 60 Hz electrical fields. No generally-accepted adverse human or ecological effects due to such fields have been observed in homes, workplaces, communities or in the immediate vicinity of electric power transmission corridors.

In the 1977 environmental impact statement, the Navy reviewed the overall results of research in several specific areas of biological and ecological effects. Relying on research it had funded on growth in bacteria, onion roots, bean roots, sunflowers, tadpoles, mice, and chickens, the Navy *1339 concluded that there was no evidence of adverse effects.

From studies that had shown slower rates of growth for slime molds and rats exposed to electromagnetic radiation, the Navy acknowledged the potential biological effects of electromagnetic radiation on early growth in certain species, but minimized the importance of this possibility:

 
Experimental evidence to date indicates normal offspring of plants, animals, and humans are produced in ELF fields comparable to and even larger than Seafarer fields. No study has shown teratogenic effects (embryological defects) produced by ELF fields.... The biological risk of ELF effects on growth of selective species does not appear to lead to any important ecological ramifications.

The Navy concluded that Seafarer electromagnetic fields would not affect human health because all of the research indicating acute neurophysiological and behavioral effects on humans exposed to electromagnetic radiation had been undertaken in the frequency range from 2 to 20 Hz (rather than in the 72 to 80 Hz range at which ELF would operate). In addition, Navy medical surveillance at the Wisconsin test facility from 1971 to 1975 had revealed no disease prevalence or physiological alterations: "In spite of scant research and a limited theoretical foundation, some subtle effects of a biochemical nature have been reported. However, none of the evidence for biochemical effects occurs at field strengths or frequencies typical of Seafarer, and therefore is not evidence that such an antenna would affect biochemical processes." The Navy also stated, "Experimental evidence that ELF fields affect physiological processes is not conclusive, but is sufficiently important to warrant careful examination."

In the 1977 environmental impact statement, the Navy reiterated its pledge that "if any deleterious biologic effects were identified, the operation of the system would be discontinued." The Navy planned ongoing research and monitoring efforts: "The risk of electromagnetic effects will be evaluated through a continuing environmental monitoring program.... The results of the program would be reviewed periodically by the Department of Defense and the state to determine whether changes in Seafarer test operations are necessary to reduce environmental risks to mutually acceptable levels."

In response to the National Academy of Sciences study, the Navy proposed to support biological and ecological research in many of the areas recommended by the National Academy of Sciences, such as bird migration patterns and electromagnetic field effects on electrosensitive fish and magnetotactic bacteria; to continue two studies on fertility, growth, and development of mice and rats, one study of triglyceride concentration studies with rats and monkeys, and primate studies on complex nervous system effects; and to evaluate the Electric Power Research Institute studies on honey bee behavior under commercial power lines.[6] To investigate indications of potential biological effects, the Navy proposed to include field studies of soil organisms and plants in its ecological monitoring plan for the selected site area in Michigan. However, the Navy planned no further research in the areas of genetics, circadian rhythms, and human serum triglyceride concentrations, because it concurred in the National Academy of Sciences committee's finding of no likely hazard.

When the Navy distributed the September, 1977 Seafarer environmental impact statement the public expressed concern over the size of the proposed fully operational system. Citizen groups voiced the fear that the Navy would expand the test facility into a large fully operational system without any subsequent environmental documentation.

*1340 In accordance with its duties under the National Environmental Policy Act and § 309 of the Clean Air Act, the United States Environmental Protection Agency reviewed the draft environmental impact statement and the National Academy of Sciences study. In a letter of November 11, 1977, the Environmental Protection Agency suggested that further environmental impact documentation would be necessary before any decision were made to operate a full-scale ELF system at one or both of the test facilities. The Navy attached this letter to the comments section of the final environmental impact statement and indicated its agreement in the margin.

 
2. 1981 reactivation of ELF communication proposal 
a. Decision

In April, 1981, after President Reagan agreed with defendant Weinberger's recommendation that the President decide whether to reactivate the Navy's ELF communication system, the Navy began planning the implementation of the system and determining the communication capability necessary to reach United States strategic submarines throughout the world. Defendant Weinberger submitted an ELF proposal to the President on August 13, 1981.[7] On October 8, 1981, President Reagan approved the recommended proposal and advised Congress "that we have decided to proceed with the ELF Communications System deployment."

 
b. Scope of current proposal

The system recommended by defendant Weinberger and approved by President Reagan requires upgrading the existing facility in Wisconsin, constructing a new facility of comparable size in upper Michigan with 56 miles of above-ground antennae, and supplying submarines with appropriate ELF receivers.

According to a report distributed in December, 1981 by the Naval Electronic Systems Command, the upgrading of the Wisconsin facility is to include installing new ELF power amplifiers and ancillary electronic equipment, expanding the existing ELF transmitter building by about 8620 square feet to accommodate the new electronic equipment, installing standby electric power units in a separate 2800 square foot building, replacing the present security personnel building with a larger gate house, burying tanks for storing diesel fuel for the standby power units, installing a fault detection system on the present 28 mile antenna system, and improving the four ground terminals to enhance electrical safety. The new construction is to be entirely within the confines of the existing site with installation of the new equipment scheduled for 1984. All present operations and programs are to be continued throughout 1982-1985, including ongoing ecological monitoring, wildlife surveys, and interference mitigation activities.

Construction in upper Michigan is to include installation of four ground terminals, a 56 mile network of above-ground transmitting antennae, and a transmitter located on about two acres of land. The Michigan facility is to be operated synchronously with the Wisconsin facility to produce the signal strengths needed to reach the most distant United States submarine stations. Suitable sites for the facility and the antennae corridors were to be selected in 1982 with the assistance of Michigan state agencies. Construction work was to be initiated in 1983 and completed in 1984. Installation of equipment is to be completed in 1985 and synchronous operations with the Wisconsin facility are to begin in 1986.

 
*1341 c. Environmental impact assessment for the current proposal

On June 7, 1983, the Navy issued, but did not circulate to the public or to interested federal agencies, an environmental impact assessment of the upgrading of the Wisconsin facility.[8] In this assessment, the Navy concluded that upgrading the Wisconsin ELF facility would produce neither permanent nor temporary significant environmental impact. The Navy noted that the 1977 National Academy of Sciences study and the 1977 environmental impact statement had determined that ELF would produce no adverse biological effects. The Navy explained that the upgrading work would not increase the electromagnetic field intensities which would remain below the safety standard recommended by the World Health Organization in 1982.[9]

 
3. Navy-supported research in biological effects

In response both to public criticism of its ELF proposals and to the general lack of information about biological effects of extremely low frequency electromagnetic radiation, the Navy has sponsored several research studies on this subject. Reports of those studies that appear in the record and were written after 1977 are summarized below.

 
a. Pensacola primate experiments

Since 1969, the Navy has conducted primate experiments at the Pensacola Naval Aerospace Medical Research Laboratory. Up until 1977, researchers were unable to find any physiological differences between monkeys exposed to electromagnetic radiation and control groups of monkeys. Following the publication of the National Academy of Sciences study in August, 1977, the researchers analyzed some of the experimental data on the basis of the sex of the monkeys. In November, 1977, the researchers noted that exposed males gained weight at a faster rate than the control group males so that by the end of one year the exposed males were approximately 11% heavier than the control group males.

In March, 1978, a National Academy of Sciences panel evaluated the data at the Navy's request. The panel found small but statistically significant differences between exposed and control groups in several areas. The exposed group had greater body weight, lower blood glucose content, lower blood urea nitrogen, and lower blood glutamyltranspeptidase. The panel noted substantial evidence of a real effect of exposure on male animals, but found that the actual values of the exposed group males were within normal ranges, despite the quantitative differences between exposed and control group males. The panel did not recommend any changes in the Seafarer decision, based on the Pensacola findings. However, it recommended continuing research to examine the differences that had been identified. As of September, 1978, the Navy planned to continue the primate experiments at Pensacola.

Early in 1983, the Pensacola Naval Aerospace Medical Research Laboratory submitted *1342 to the Navy an annual report of the primate experiments for fiscal year 1982. In four paragraphs, the laboratory reported that preliminary results of the experiments disclosed no significant growth rate effects but did disclose some inconclusive endocrine and circadian rhythm changes.

 
b. UCLA monkey behavior study

In June, 1980, a group of scientists from the University of California at Los Angeles, submitted a final report of their Navy-supported research begun in 1969 on effects of electromagnetic radiation on monkey behavior. Titled "An Evaluation of Possible Effects of Modulated 76 Hz Electric Fields on Behavior and EEG of Monkeys. Phase 2: Frequency Modulation," the UCLA report arrived at four conclusions: 1) The scientists found that the monkeys' behavioral patterns were frequency specific with a low threshold of 7 Hz, that is, their behavioral changes occurred in response to specific frequencies, rather than simply in response to electromagnetic radiation. This phenomenon was analogous to a phenomenon of frequency-specific calcium efflux that had been observed in in vitro neonatal chicken brains. 2) The data suggested some degree of dose dependent results because results at 100 volts per meter were inconclusive, while results at lower intensities showed an effect, suggesting a voltage "window" similar to that found in certain calcium efflux studies (which are not further identified and which are not part of the administrative record) and by Kalmijn (a scientist who found an effect upon the electric sense of sharks within a small range of low voltage stimulation but not at substantially higher voltage stimulation).[10] 3) Relatively long periods of exposure contributed to the systematic array of results. 4) The data suggested a hypothesis that the animals were forced to pay attention to their own internal mileau because of the absence of regulating cues in the form of sound or light.

 
c. Continued studies 
1) bird migration studies

Between 1974 and 1979, the Navy contracted with a group of scientists to study the effects of ELF fields on birds flying over the Wisconsin test facility. Early observations suggested that the mean direction of migration shifted when the north-south leg of the antenna network was activated. When both legs or only the east-west leg were activated, no shifts in migration could be confirmed.

In an August, 1979 report, scientists noted that further analysis had confirmed earlier findings of some shifts in migration patterns. However, from their findings of systematic differences in the birds' flight when flying over areas known to be magnetically anomalous, the scientists concluded that temporary, highly localized changes in migratory behavior in the vicinity of the test facility when the antennae were activated would have minimal effects on the migratory goals of birds.

 
2) multigenerational mice studies

An early multigenerational study of mice exposed to modulated signals of 76 Hz (within the 72 to 80 Hz range used by the ELF system) disclosed differences between the exposed and control groups in the sixth generation in terms of litter size and fertility. However, because the entire experimental litter died over a weekend and the sixth generation males had been killed to save space, no valid results could be determined. Repeated exploration with the sixth generation females disclosed some differences. The study was repeated during 1977. In a final report apparently dated January, 1980, the scientists found no *1343 significant differences between exposed and control populations.

 
3) slime mold studies

In Navy-sponsored research experiments conducted under laboratory conditions, Dr. Goodman of the University of Wisconsin-Parkside found that exposure to weak electromagnetic fields affected the growth of slime mold. Subsequently, Goodman conducted similar tests at the Clam Lake ELF site to determine whether the same physiological effects could be observed under exposure to actual electromagnetic fields. The record does not contain any of the results of these later tests.[11]

 
4. IIT Research Institute activities

Since 1968, except for the "dormant period" from April, 1979 to March, 1981, when Project ELF was not funded, the Navy has contracted with IIT Research Institute for the provision of technical services for the Navy's Environmental Compatibility Assurance Program. (IIT Research Institute is a not-for-profit institute located on the campus of the Illinois Institute of Technology.) The contract establishing work requirements for the 1976-1978 fiscal years provided for "electromagnetic environments surveys and analysis."

IIT Research Institute prepared all the plans and reports for the Navy's ecological monitoring program. In addition, it provided the Navy with occasional summaries of legal developments concerning National Environmental Policy Act requirements, with electric and magnetic field measurements in and around the Wisconsin test facility, with several summaries of professional conferences and discussions with scientists about current research, and with evaluations and copies of some articles about biological effects of electromagnetic radiation, which are discussed infra at III, A, 6; III, A, 7, b.

Although the contract provided for IIT Research Institute to submit monthly reports to the Navy summarizing its activities in the previous month, the administrative record contains only seven IIT Research Institute reports for the months August, 1981, October, November, December, 1982, February, April, June, and August, 1983. In the reports for August, 1981, October, 1982, November, 1982, and December, 1982, IIT Research Institute reported little information on biological effects of electromagnetic radiation. IIT Research Institute did report discussions between IIT Research Institute personnel and five scientists who were conducting experiments on biological and ecological effects of electromagnetic radiation, but there are no reports of the experiments themselves in the administrative record.[12]

 
5. Ecological information

Much of the information in the administrative record pertains to the ecological, rather than the biological, effects of Project ELF. This ecological information is summarized below.

 
a. Forest Service wildlife surveys

Since July, 1971, the Navy has tracked animal populations at the Clam Lake site *1344 by contracting with the United States Forest Service for annual studies of small mammal populations, deer herds, eagle and osprey nesting populations, and ruffed grouse populations in the area of the Wisconsin ELF test facility. The record includes the 1976, 1978, and 1982 annual survey data submitted by the Forest Service.[13]

 
b. Ecological monitoring program

The Navy established the ecological monitoring program to ensure that the electromagnetic fields produced by operating an ELF communication system in Marquette County, Michigan would have no adverse effect on vegetation and wildlife located near the project area.

The administrative record contains a compilation of 1982 reports of the Navy's ecological monitoring program, prepared by IIT Research Institute and covering the period from July, 1982 through October, 1982. One of the emphases of the ecological monitoring program in upper Michigan was the collection of base data of the sort that had not been collected before the Wisconsin test facility had been constructed. During the first four months of the studies, twelve of the thirteen researchers[14] collected base information for further work in the areas of plant cover, tree studies, litter decomposition, microflora, soil amoebae, soil arthropods, earthworms, small mammals, nesting birds, aquatic ecosystems, and bees.

The record also contains an August, 1983 IIT Research Institute report of the 1982 ecological monitoring program and the plans for continuation of ongoing studies of terrestrial vegetation, soil amoebae, slime mold, soil arthropods and earthworms, native bees, small mammals and nesting birds, migrating birds, wetland biota, and aquatic biota.

 
c. Electromagnetic field measurements

In April, 1978, January, 1982, and February, 1983, pursuant to its contract with the Navy, IIT Research Institute measured and reported the electric and magnetic fields near the ELF antennae and the test facility. Those reports show that magnetic field measurements in the thirty mile area around the Wisconsin test facility have ranged from 0.000001 gauss to 0.189 gauss, but generally are about 0.001 gauss in the vicinity of the antenna. According to 1983 measurements, electric field intensities in the area of the Wisconsin test facility range from 0.00005 volts per meter to 0.16 volts per meter at 76 Hz. However, previous measurements recorded one unusually high electric field measurement of 157 volts per meter at 75 Hz.

 
6. Scientific literature

Until the summer of 1983, the only scientific literature in the administrative record on electromagnetic field effects consisted of the reports from Navy-supported research, none of which had ever been published in a scientific journal.

In April, 1983, IIT Research Institute reported that it had utilized four search services to discover papers published since 1978 pertaining to biological effects of extremely low frequency electromagnetic radiation and that the searches disclosed 567 citations, 134 of which seemed relevant to Project ELF. In May, 1983, IIT Research Institute reported having used six search services to identify papers pertaining to biological effects of extremely low frequency electromagnetic fields, obtaining a list of 1249 citations. In its June, 1983, monthly report, IIT Research Institute provided the Navy with copies and critiques of five articles and five Letters to the Editor which are now part of the administrative record. These are summarized below.

 
*1345 a. Wertheimer-Leeper articles

In these two articles, Dr. Nancy Wertheimer and Ed Leeper discussed two epidemiological studies they conducted in which they found a correlation between high-current wire configurations and the incidence of childhood and adult cancer. From this correlation, the authors concluded that magnetic fields may produce a condition or environment that facilitates the onset of cancer. They did not contend that they had established the existence of a causal relationship.

In the first article, published in a peer review journal in 1979,[15] the authors reported locating residences of children in the greater Denver area who had died of cancer within a certain time period, to determine the proximity of those residences to transformers or to electric wires carrying current of a high magnitude. The authors used the term "high-current configurations" to refer to a residence less than forty meters from primary wires (13 kv) built to carry high currents, less than 20 meters from 3 to 5 smaller primary wires (50-230 kv), or less than 15 meters from still smaller wires that issue directly from transformers without yet losing current through a service drop. All other configurations were considered low-current configurations. The authors found a statistically significant correlation between the incidence of childhood cancer and the proximity to the children's residences of high-current configurations. (Electric power lines produce magnetic fields in the range of 0.001 to 0.1 gauss. These magnetic field measurements are comparable to those in the vicinity of the ELF antennae.)

In the 1982 article published in a peer review journal, Wertheimer and Leeper discuss the results of a study in which they found a statistically significant correlation between the incidence of adult cancer and the proximity of high-current configurations to the patients' residences.

 
b. Rhode Island epidemiology study

In a 1980 article, "Electric-Wiring Configurations and Childhood Leukemia in Rhode Island," in the American Journal of Epidemiology, a peer review journal, Fulton and others reported on their attempt to duplicate in Rhode Island the Wertheimer-Leeper procedures relating leukemia to power line configurations. The researchers concluded from their data that if such a relationship exists, it is very small. It was their opinion that the findings in the Denver study must be the result of an interaction between magnetic fields and some other factor that is stronger in Denver than in Rhode Island.

The journal published two Letters to the Editor responding to this article. In one, the writer challenged the statistics used in the study. In the other, Wertheimer and Leeper criticized the Rhode Island study for having an urban bias in the control group. They reanalyzed the Rhode Island data using a different control group and found a small correlation of the kind identified in their Denver studies.

 
c. Stockholm epidemiology study

In an article entitled "Electrical Constructions and 50 Hz Magnetic Field at the Dwellings of Tumor Cases (0-18 years of age) in the County of Stockholm," the authors found a correlation between the incidence of cancer and residency in a dwelling located within 150 meters of a visible 200 kilovolt high tension wire assumed to have a magnetic field of 0.003 gauss or more (as compared to the magnetic field averaging about 0.001 gauss and going to a high of 0.189 gauss in the vicinity of the ELF facility).

 
d. Delgado study

In a 1982 article published in a peer review journal and entitled "Embryological Changes Induced by Weak, Extremely Low Frequency Electromagnetic Fields," Delgado *1346 and others described the exposure of fertilized chicken eggs during the first stages of development to electromagnetic fields of 10 Hz, 100 Hz, and 1000 Hz. The researchers found that frequencies of 100 Hz had a significantly more powerful effect on chicken embryogenesis than did frequencies of 10 Hz or 1000 Hz. Such exposure arrested development at a very early stage, limiting development to the three primitive layers at which there are no signs of the neural tube, brain vesicles, the auditory pit, foregut, the heart, vessels, or somites. Magnetic field intensities of 0.012 gauss had a greater effect than exposures to 0.12 gauss or 0.0012 gauss.[16]

The researchers concluded that electromagnetic field effects occur in both frequency and power windows.[17] They referred to similar findings in other studies, such as: a 1976 study by Bawin and Adey which found a major decrease in calcium efflux at frequencies of 10 and 56 volts per meter but only a small decrease at fields at 5 and 100 volts per meter;[18] a 1975 study by Bawin, Kaczmarek, and Adey finding an effect at 6-20 Hz which disappeared at 35 Hz; a 1978 study by Bawin, Sheppard, and Adey finding a power window above and below which no significant effect could be observed; and a 1979 Adey study discussing possible mechanisms of such window effects.

A report of Delgado's work appeared in MicroWave News in March, 1983. The author stated that most scientists interviewed thought that effects of this magnitude should be subjected to further testing for reliability.

 
e. Letters to the Editor

On July 22, 1982, The New England Journal of Medicine published a letter from a Dr. Milham of the Washington Department of Social Services entitled "Mortality from Leukemia in Workers Exposed to Electrical and Magnetic Fields," in which Milham reported evidence of more deaths from leukemia among men working in electric or magnetic field occupations. From this correlation, Milham concluded that a causal relationship might exist.

The New England Journal of Medicine published a response to the Milham letter on November 25, 1982 which was written by a Dr. Liburdy and entitled "Carcinogenesis and Exposure to Electrical and Magnetic Fields." Liburdy took issue with Milham's conclusion that there is a causal association between leukemia and electric or magnetic fields. He summarized existing research in the area of biological effects of electromagnetic radiation and referred to a 1978 National Institute for Occupational Safety and Health report citing more than 60 studies which had identified various mutagenic, teratogenic, developmental, hematologic, and neurologic effects of electromagnetic radiation. None of the scientists whose studies were cited in the report had found any relationship between electromagnetic radiation and cancer.

In a November 20, 1982 Letter to the Editor of The Lancet entitled "Leukemia in Workers Exposed to Electrical Magnetic Fields," Wright, Peters, and Mack reported finding a higher incidence of leukemia in white males in electrical occupations in Los Angeles County from 1972 to 1979.

In a January 29, 1983 Letter to the Editor of The Lancet, McDowall reported a higher than average leukemia mortality rate among electrical workers in England and Wales.

In an April 30, 1983 Letter to the Editor of The Lancet, Coleman, Bell, and Skeet reported the results of their investigation of the proportional incidence of cancer in ten electrical occupations in Southeast England *1347 from 1961 to 1979, which disclosed a slightly increased incidence of leukemia among electrical workers.

A 1982 Letter to the Editor published in Physics in Medicine and Biology entitled "Magnetic Fields Affect the Lac Operon System" did not use epidemiological techniques. The authors examined the enzymatic lac operon system in E. coli to look at the effect of alternating magnetic fields on the growth rate of bacteria, applying pulse signals of 0 to 0.007 gauss to the bacteria.[19] The researchers found the rate of beta galactosidase synthesis, a process controlled by a repressor protein on the DNA chain, to be dependent on magnetic field strength. As the strength of the magnetic field increased the rate of synthesis decreased at first, reaching a low of less than one-third the normal rate at an application of 0.0027 to 0.0030 gauss, then returning to the control value at an applied field strength of 0.0032 gauss. As the application increased from 0.0051 to 0.0056 gauss, the rate of synthesis increased rapidly, but fell to the normal rate when a 0.0058 gauss strength field was applied. (Magnetic field measurements at the ELF facility are generally about 0.001 gauss in the vicinity of the antennae but have been as high as 0.189 gauss.)

 
f. Other publications

Also a part of the administrative record are a chapter of a World Health Organization publication devoted to nonionizing electromagnetic radiation research and a Montana Department of Natural Resources Conservation Report, both of which were submitted to the Navy by IIT Research Institute in the summer of 1983.

 
1) World Health Organization report

In 1982, the World Health Organization published a Regional Publication devoted to nonionizing radiation protection. Although published by the World Health Organization, the book is not intended to represent the views of the organization. Rather, each chapter represents the views of its author.

The particular chapter on biological effects of electromagnetic radiation is a summary of data and research in the area, including a synopsis of exposures found to be safe in experiments. The chapter covers research published before 1979, with only three sources dated 1978 or later.

In his summaries and conclusions, the chapter's author did not attempt to recommend a safe exposure standard. The author found adverse effects associated with electromagnetic fields on nervous systems. He also found evidence of cardiovascular disorders, blood changes, changes in reaction time, increases in body temperature, changes in cholesterol and triglyceride levels, and stress symptoms. He concluded that many effects could be attributed to secondary factors, that animal studies could not be used to make findings about humans, and that reported physical effects appeared to be transient functional disorders rather than organic changes or permanent injuries. Finally, he concluded that there is no danger to human health from exposures to electric fields of 20 kilovolts per meter or to magnetic fields of 0.003 gauss.[20]

 
2) Montana Department of Natural Resources and Conservation report

In February, 1983, the Montana Department of Natural Resources and Conservation completed a report prepared by a Dr. Sheppard who had been commissioned by the department to evaluate a 500 kilowatt transmission line proposed for installation in Montana.

Sheppard reviewed human health effects of chronic exposure to electromagnetic radiation from high-voltage transmission lines. In his report, Sheppard noted both the inadequacy of existing information for reaching unequivocal conclusions about health effects and the difficulties in evaluating *1348 certain studies, such as those using epidemiological techniques. He defined safety in the context of a cost-benefit analysis, weighing human health and environmental risks against benefits to be gained from the project.[21]

Although noting that many studies disclosed biological effects from electromagnetic radiation, Sheppard discounted these effects in his cost-benefit analysis because of the uncertainty of the effects, the nature of the effects identified, and the benefits of the project.

 
7. Navy review of scientific information 
a. Environmental Review Committee

In February, 1982, the Navy established an Environmental Review Committee. The committee's function was to assist in formulating and planning implementation of environmental protection plans and to review results of environmental activities and studies. It was not assigned the task of performing a comprehensive review of the scientific literature on biological effects of electromagnetic radiation or evaluating the potential risks and benefits of Project ELF. Composed primarily of Navy personnel, the committee was responsible for monitoring and evaluating ongoing contracts and studies for environmental effects. Three non-voting observers attended the meetings: one from the United States Forest Service, one from the Wisconsin governor's office, and one from the Michigan governor's office.

At its May 6, 1982 meeting, the committee did review some human health studies which are described in general terms, but not identified in the minutes of the meeting. The committee concluded that field intensities associated with the ELF communication system would have no undesirable effect on occupational or public health. The committee stated that it found inapplicable the health survey results it reviewed (because ELF fields are of lower field intensities than those encountered by the electrical workers that were the subjects of the surveys) and the epidemiology studies (because causation was not proven, all other potential causes were not disposed of, and the studies concerned electric transmission lines). The committee stated that no definitive conclusions could be determined from any laboratory research to date and that the Navy's primate experiments must continue for at least another year before reliable analysis would be possible. In the minutes, the committee discussed the literature on biological effects in 1½ pages.

At its June 2, 1983 meeting, the committee discussed the public opposition to Project ELF arising from the fear of potentially adverse health effects. The committee recommended that the Navy develop a public response, using the following arguments:

 
1) The position of the National Academy of Sciences, the preeminent group of scientific expertise, which found that the concerns were invalid or unwarranted, is ignored by critics; 2) The Michigan Department of Public Health, after a year of study, has not developed a scientifically acceptable proposal to submit to the Navy for public health studies; 3) There are no scientifically-designed studies reported in the literature suggesting that ELF electromagnetic fields affect health; 4) A scientific study requires design inputs from engineers and scientists, and such teaming has not been evident among those contending to have found a relationship between human health and electromagnetic fields; 5) Critics contend that 60 Hz and 76 Hz fields are somehow "different," but nevertheless use questionable *1349 60 Hz reports to show that 76 Hz ELF fields are dangerous.

Other than the minutes of the two meetings, the record contains no reports by this committee that address the subject of biological effects of extremely low frequency electromagnetic radiation.

 
b. IIT Research Institute literature review

As noted earlier, IIT Research Institute supplied the Navy with copies of five articles and five Letters to the Editor in June, 1983. In addition to that literature, IIT Research Institute supplied its evaluations of these published writings and a May, 1981 report of public and scientific interest in electromagnetic field effects. The evaluations and the report are part of the record and are summarized below.

 
1) Wertheimer-Leeper articles

IIT Research Institute prepared an eighteen-page evaluation of the Wertheimer-Leeper childhood cancer article, criticizing the study for employing oversimplified explanations of electric power systems, questionable assumptions, and debatable interpretations of data, in order to reach seemingly unjustified conclusions. IIT Research Institute highlighted other defects in the study, such as the absence of any measurements of magnetic field intensities inside the homes, the authors' apparent limited understanding of power systems, the selection of references to support a preconceived hypothesis that is not representative of the state of knowledge, the lack of complete address histories, the absence of engineering analysis of the data, and the inadequate analysis of social class and sex factors.

In a two-page evaluation of Wertheimer and Leeper's second article (on the incidence of adult cancer), IIT Research Institute stated that this study was highly speculative and provided no useful information about possible relationships between power systems and cancer.

The record contains evidence of two efforts by IIT Research Institute to obtain additional information on the Wertheimer-Leeper studies. The first is its April, 1983 report of a meeting with Dr. Jim Stebbins of Argonne National Laboratory. The meeting centered on epidemiology studies associating electromagnetic fields with human health effects, with particular reference to the Wertheimer-Leeper studies. Stebbins, an epidemiologist and consultant to the New York Public Service Commission, provided personal background information about Wertheimer and Leeper. Stebbins characterized the Wertheimer-Leeper methodology as unconventional because of the manner in which they selected control groups and the fact that they confined their research to only those magnetic fields produced at 60 Hz.

The second effort is a report by a Dr. Crocetti, forwarded to the Navy by IIT Research Institute in the summer of 1983. Crocetti, a scientist associated with the New York Public Service Commission, had evaluated the Wertheimer-Leeper studies and recommended that they be repeated with certain modifications. She recommended that Wertheimer repeat the study because of her existing data base, her expertise, and the working relationships she had established. Crocetti noted the need for subsequent independent replication. For this purpose, she recommended that anyone undertaking a new study start with actual measurements of the magnetic fields, including measurements inside the homes, and that the researcher collect exposure data and addresses for the total time between birth and death, remove data on survivors in the adult study, and obtain from the electric company the actual ages of specific wiring configurations.

 
2) Rhode Island epidemiology study

In a two-page evaluation, IIT Research Institute concluded that the Rhode Island electrical wiring-childhood leukemia article was seriously flawed because it lacked adequate data and controls. IIT Research Institute recommended that the report not be used "to support the view that no relationship exists or should be expected to exist *1350 between exposure to ELF electromagnetic fields and cancer in children."

 
3) Stockholm epidemiology study

In its three-page evaluation of the Stockholm epidemiology study, IIT Research Institute concluded that the study was so incomplete as to be inconclusive or of little use.

 
4) Delgado study

In its three-page evaluation of the Delgado study, IIT Research Institute criticized the methodology used to report the magnetic fields, the significance of the study for Project ELF because of the use of pulse signals, and the statistical methods used by the researchers. IIT Research Institute concluded that the study was not definitive because of peer skepticism, lack of independent verification, and inadequacy of the description of the engineering system.

 
5) Letters to the Editor

In five one-page statements, IIT Research Institute evaluated the five Letters to the Editors which it sent to the Navy in the summer of 1983.

In its evaluation of the Milham letter, IIT Research Institute concluded that Milham's views were highly speculative.

With respect to the letter from Wright and others referring to findings among white males in electrical occupations in Los Angeles County, IIT Research Institute concluded that the study was highly speculative and that the absence of data other than a table and the possibility that other causes might account for the correlation detracted from any value this study might have.

IIT Research Institute advised the Navy that because of the limited data, no conclusions could be drawn from the McDowall and Coleman Letters to the Editor of The Lancet on the incidence of cancer in electrical occupations in Great Britain.

IIT Research Institute concluded that the results of a study on the effects of magnetic fields on the enzymatic lac operon system were not germane to Project ELF because the researchers used pulse signals and magnetic field exposures different from those to be expected near an ELF antenna.

 
c. Other reviews

The administrative record contains the minutes of the Navy's own Environmental Review Committee, but no evidence of any outside reviews of post-1977 scientific information sought by the Navy other than the 1978 National Academy of Sciences review of the primate experiment data described supra at 1341-1342, and IIT Research Institute's brief 1983 evaluations of some of the post-1977 published literature. Neither of these reviews is comprehensive.

 
d. IIT Research Institute public interest review

In May, 1981, Martin Abromavage, IIT Research Institute's engineering advisor for the ELF communication system contract, prepared a report for the Naval Electronic Systems Command describing the public, political, scientific, and professional interest in the biological and ecological effects of extremely low frequency electromagnetic fields from 1969 through 1978 and the Navy's research in this area. It was designed merely to provide information for the Navy to use in deciding whether to re-initiate biological and ecological research and contained no conclusions or recommendations. The report contained an extensive bibliography with the sources on biological effects limited to the years before and including 1978.

 
8. Biological effects information in administrative record in fall, 1981

In August, 1981 when defendant Weinberger submitted his ELF proposal to the President, and in the fall of 1981 when the Navy decided to proceed with the reactivation of the project without filing a supplemental environmental impact statement, the administrative record contained the following information on biological effects of electromagnetic radiation:

*1351 1) the 1978 National Academy of Sciences report of the Pensacola primate data showing effects on growth;

2) the 1978 one-page letter summarizing findings from the GTE-Sylvania medical department of two years of medical surveillance of workers at the Wisconsin ELF site;

3) a 1979 report from the bird migration studies in Wisconsin confirming earlier findings of a migration effect;

4) a 1980 report on the UCLA study finding "windows" and some effects on monkey behavior;

5) a 1980 report of the multigenerational mice study disclosing no effects;

6) an IIT Research Institute report on the public interest in the biological and ecological effects of ELF electromagnetic fields from 1969 through 1978; and

7) an IIT Research Institute monthly report dated August, 1981, in which the institute advised that it had conducted a search of published biological studies by one scientist.

None of the scientific information on biological effects described above had been published in scientific journals.

As of the end of 1981, the administrative record on which the Navy based its decision to proceed with Project ELF did not contain any copies of or references to published scientific literature on biological effects of electromagnetic radiation or any summaries or analyses of such information, any independent or in-house analyses of any Navy-supported research other than the National Academy of Sciences review of the Pensacola primate data, or any comprehensive review or evaluation of the scientific information that was part of the record.

 
B. Trial Testimony

From the testimony at trial, I find the following facts.

 
1. Navy review of scientific information in conjunction with its 1981 decision to reactivate Project ELF

In August, 1981, when the Navy submitted its recommendation to the President, the Navy had not reviewed previous environmental impact statements or the environmental consequences of the proposed action. In forwarding the recommendation, the Navy made no determination whether the National Environmental Policy Act required any environmental documentation supplementing its previous environmental impact statements.

In the fall of 1981 after President Reagan's decision to revive Project ELF, Admiral William D. Smith, then Director of Naval Communications in the Office of Chief of Naval Operations, decided that the Navy would not prepare another environmental impact statement in conjunction with the reactivation of Project ELF. He based his decision on two factors: 1) legal advice from Navy legal counsel that the National Environmental Policy Act did not require a supplemental environmental impact statement, and 2) his conclusion that the current proposal was essentially the same proposal considered in the 1977 environmental impact statement with modifications that only reduced the size and magnitude of the project and thereby lessened the impact on the environment. He did not base his decision on a review of the available scientific information on biological effects of electromagnetic radiation.[22]

The Navy relied on the IIT Research Institute for information about published literature on biological effects of electromagnetic radiation. To keep abreast of current literature, IIT Research Institute reviewed monthly technical newsletters, four professional government search services, and information obtained through professional conferences and professional contacts. Although IIT Research Institute *1352 has prepared environmental impact statements for the Navy, it has never been charged with the responsibility of deciding whether an environmental impact statement should be filed.

 
2. Post-1977 scientific research

For the purpose of assessing the significance of new information on biological effects of electromagnetic radiation, plaintiffs called Dr. Nancy Wertheimer, Dr. John Moulder, and Dr. Robert Becker as expert witnesses. Defendants called Dr. Karl Straub and Dr. Don Justesen.

Dr. Wertheimer is a clinical associate professor in the Department of Preventive Medicine in Biometrics, the University of Colorado Medical Center. She has a Ph.D. in experimental psychology and has done postdoctoral work in advanced biochemistry and in epidemiology. She is a member of the American College of Epidemiology and the Bioelectromagnetic Society. She is the co-author of the two articles on childhood and adult cancer which have been described previously.

Dr. Moulder is an associate professor of radiology and radiation oncology at the Medical College of Wisconsin. He has a doctorate in cell biology and has done work in biophysics. He is a member of the Nonionizing Radiation Subcommittee of the Wisconsin Radiation Protection Council.

Dr. Becker is a board-certified orthopedic surgeon and professor of orthopedic surgery at the State University of New York, Upstate Medical Center. He has had published between thirty to forty scientific papers dealing with the biological effects of extremely low frequency electromagnetic radiation and is a co-author of a book, Electromagnetism and Life, published in 1982. In 1973, he served at the request of the Navy as a member of a committee evaluating the biological effects of Project Seafarer.

Dr. Straub is a physician and scientist with a joint appointment as Associate Chief of Staff of Research at the Veterans Administration Hospital in Little Rock, Arkansas and as Professor of Medicine and Professor of Biochemistry at the University of Arkansas Medical Sciences campus. He worked in the area of nonionizing radiation for two and one-half years while he was in the Navy Medical Corps and has published three or four articles on the subject. He serves on an advisory committee on biological effects of electromagnetic radiation for the International Electrical & Electronic Engineers Society.

Dr. Justesen is a Professor of Psychiatry at the University of Kansas School of Medicine and is Director of the Behavioral Radiology Laboratories of the Veterans Administration Medical Center in Kansas City, Missouri. His doctorate is in experimental psychology. He is president-elect of the Bioelectromagnetic Society and past chairman of the International Electrical & Electronic Engineers Society's Committee on Man and Radiation. He has done research in the area of extremely low frequency nonionizing electromagnetic radiation.

 
a. Research studies not included in the administrative record

Of the new information on the human health effects of extremely low frequency electromagnetic radiation that has been generated since 1977, the most important studies in addition to those contained in the administrative record are the Electrical Power Research Institute studies at the Batelle Laboratories, Blackman and Adey's frequency and power intensity window research, and Goodman's slime mold research. A summary of each of these projects is set out below, together with the experts' assessments of the significance of the work.

 
1) Blackman and Adey window research 
a) the study

Since 1977, two scientists, Blackman and Adey, have performed a number of cellular function experiments in which they have observed frequency and power intensity window phenomena. In four studies, two published in 1979 and 1981 and two in 1982, Blackman explored the effects of extremely low frequency electromagnetic fields on *1353 the permeability of chicken brain cells and found power intensity windows and frequency windows at 15 Hz and at 75 Hz. (ELF is to operate at a range of 72-80 Hz.) In 1980, Adey published in a peer review journal an article entitled "Frequency and Power Windowing in Tissue Interactions with Weak Electromagnetic Fields," in which he described his observation of two different frequency and power intensity windows in chemical, physiological, and behavioral responses to weak electromagnetic fields: one at 0.00007 volts per meter for extremely low frequency tissue gradients and one at 10 volts per meter for amplitude modified radio frequencies and microwave gradients. In his article, Adey attributed these window effects to amplification mechanisms that are important in immunological, endocrine, and neurobiological excitation reactions.

 
b) expert assessment

With the exception of Becker, who did not comment on it, all of the expert witnesses agreed that the Blackman-Adey research is of importance because it establishes that the dose-response curve (the pattern of occurrences of effects in response to exposure to different frequencies) is not linear, but unusual and complex. Also, the research demonstrates the existence of windows at 15 Hz and at 75 Hz.

Although Adey had done work on the phenomenon of intensity windows before 1977, it is only since then that he has confirmed the phenomenon through additional studies and that his results have been replicated by Blackman.

Straub is of the opinion that the Adey-Blackman research is not directly relevant to the ELF facility because of the differences in electric field intensities.

 
2) Batelle Laboratories studies 
a) the study

In a series of studies of minipigs (Hanford miniature swine) conducted in 1978, 1979, and 1982 at the Batelle Laboratories, and sponsored by the Electrical Power Research Institute, researchers have found teratogenic effects (birth defects) from electromagnetic radiation at 60 Hz, a higher intensity than that proposed for Project ELF. Although most of the Batelle studies have produced negative results, several behavior tests, neurophysiologic tests involving synaptic function, and tests of the rate of bone fracture repair have been positive for effects.

 
b) expert assessment

Moulder, Straub, and Justesen commented on the Batelle research, and praised it. Moulder noted the inherent credibility of results that are contrary to what the sponsor would have wanted. Justesen termed the research excellent, noting that it uses pigs which have glabrous skin (unencumbered by hair) as do humans. Straub believes that the Batelle studies are likely to produce significant results. However, both he and Justesen consider the results irrelevant to Project ELF because of the differences in the intensities of the electric and magnetic fields anticipated at the ELF facilities and those generated by high voltage power lines which are the subject of the Batelle studies.

 
3) slime mold studies 
a) the study

In 1979 and 1983, Goodman and others at the University of Wisconsin-Parkside conducting Navy-supported research with slime mold, published articles in which they reported finding alterations in basic cell functions and oxygen consumption in response to electromagnetic radiation at higher field strengths than those proposed for Project ELF.[23] (Although the research was sponsored by the Navy, neither the 1979 nor the 1983 article is a part of the Navy's administrative record.)

 
*1354 b) expert assessment

Wertheimer, Moulder, Straub, and Justesen commented on the Goodman studies, noting that these studies have disclosed responses to electromagnetic radiation. (Straub was a member of the Navy committee that chose Goodman's studies as a project to be funded by the Navy in the early 1970's). Straub believes that the studies are probably not of much relevance to the question of potential human health hazards from Project ELF because of the possible difference in the field strengths used in the laboratory and those that will be present at the ELF facilities. In Justesen's opinion Goodman's slime mold studies have definite and meritorious implications for the ELF facility, although the effective strengths would be considerably higher than what are projected for Project ELF.

 
4) Additional research 
a) the studies

Two epidemiology studies found a correlation between electromagnetic fields associated with overhead power lines and suicide (Perry, 1981, and Reichmanis, date unknown). In addition, one questionnaire study reported reproductive hazards among electrical workers (Nordstrom 1981).

 
b) expert assessment

All of the experts except Becker commented on the striking results noted in the Nordstrom study.

Justesen was a member of the board of the peer review journal Bioelectromagnetics when it published Nordstrom's study, "Reproductive Hazards among Workers at High Voltage Substations." He explained that publication of the study indicates that a competent scientist had read it in advance of publication and found it competently drawn and carefully qualified. Justesen finds it not relevant to ELF because of the high field strengths involved in Nordstrom's study.

Straub believes that the Nordstrom study of reproductive hazards among electrical workers is of little validity because it is a questionnaire study and that it is of little relevance to Project ELF because of the differences in the fields in the study and those anticipated at the ELF sites.

Moulder, Straub, and Justesen commented on the Perry and Reichmanis studies. Straub noted that he has doubts about the merits of the Perry and Reichmanis studies, but does not consider himself knowledgeable enough about the statistical methods used in the studies to form an opinion of their validity. Justesen had a graduate student check Perry's study. The student found an inverse relationship between the incidence of suicide and higher field strengths, simply by applying the conventional means of calculating a correlation between two variables.

 
b. Post-1977 scientific literature in the administrative record 
1) epidemiology studies

Moulder noted that the Wertheimer-Leeper studies have prompted other scientists to try to replicate the results suggested in the studies. In Rhode Island, the researchers failed to find a correlation; in Stockholm, the researchers found data supporting the Wertheimer-Leeper findings.

Becker observed that the number of epidemiology studies that have been done since 1977 establish a corpus of work in this area.

According to Straub, the Wertheimer-Leeper studies suffer from a lack of longitudinal measurements of magnetic field intensities (measuring total exposure over a period of time) and from the authors' inability to exclude other environmental factors from the study.

Justesen questions the validity of the Stockholm epidemiology study, because, he said, it relies on a questionnaire.[24] He agrees with Wertheimer's description of her epidemiology studies as exploratory and suggests that further steps be taken to *1355 determine whether the indicant (the high-current configurations) can be verified.

 
2) Delgado study

All of the expert witnesses commented on Delgado's research, noting the importance of the findings of a window effect and of embryological changes in response to exposure to electromagnetic radiation.

Straub believes that the research has certain weaknesses: the use of a pulsed field, having many frequencies, rather than a sinusoidal, or standard sign wave, field; the absence of any indication that Delgado had guarded against vibration; the possibility that the ends of the eggs were deprived of proper humidity; and Delgado's failure to measure the magnetic fields inside the apparatus in which the experiment was conducted.

Justesen characterizes Delgado's work as excellent but irrelevant to the character of the fields at Clam Lake or in Michigan because of the differences in field intensities.

 
3) Letters to the Editor

Moulder, Straub, and Becker commented on the Letters to the Editor relating studies of workers in electrical occupations.

Straub places little weight on the studies described in the Letters to the Editor of The Lancet by Coleman, McDowall, and Wright reporting an increase in the incidence of leukemia among workers in electrical occupations, because the studies are retrospective in nature and do not exclude the possibility that other environmental factors could be causative. Justesen views the Coleman, Milham, and McDowall letters as having only anecdotal value.

 
IV. OPINION 
A. Scope of the National Environmental Policy Act 
1. Exemption of presidential decision

Defendants' first contention is that the decision to proceed with Project ELF is not one that is subject to the requirements of the National Environmental Policy Act. They assert that 1) it was the President of the United States rather than the Navy who made the decision to reactivate the project; and 2) presidential decisions are not subject to the Act.

I need not decide whether presidential decisions are exempt under the Act. It was the Navy that prepared the reactivation recommendations for the approval of the President and it is the Navy that plaintiffs are suing, not the President of the United States. The Navy is an agency under the Act. Weinberger v. Catholic Action of Hawaii/Peace Education Project, 454 U.S. 139, 102 S. Ct. 197, 70 L. Ed. 2d 298 (1981). Agency recommendations proposing programmatic actions are subject to National Environmental Policy Act environmental documentation requirements. Andrus v. Sierra Club, 442 U.S. 347, 363-64, 99 S. Ct. 2335, 2343-2344, 60 L. Ed. 2d 943 (1979).

I conclude that, in preparing its recommendation to the President to reactivate Project ELF and in proceeding with that project, the Navy is subject to the requirements of the National Environmental Policy Act. See California v. Block, 690 F.2d 753 (9th Cir.1982); Marquez-Colon v. Reagan, 668 F.2d 611, 615 n. 3 (1st Cir.1981). See also Friends of the Earth, Inc. v. Weinberger, 562 F. Supp. 265 (D.D.C.1983), appeal dismissed, 725 F.2d 125 (D.C.Cir. 1983) (district court construed amendment exempting the presidential MX missile deployment proposal from National Environmental Policy Act requirements as rendering moot alleged violations associated with earlier proposals and reports, holding that the Act's requirements would apply to agency actions taken after the presidential report was issued, such as site selection or other implementation actions).

 
2. Exemption of activities begun before January 1, 1970

Defendants' second contention is that the current Project ELF proposal does not constitute a major, federal action significantly affecting the quality of the human environment subject to the National Environmental *1356 Policy Act, because it merely continues operations that began before January 1, 1970, the effective date of the Act.[25] Accordingly, defendants argue, the 1981 Project ELF proposal is exempt from the Act's environmental impact statement requirements.

Activities commenced before the effective date of the Act are covered by the Act if the action remaining to be completed after January 1, 1970 is substantial. Westside Property Owners v. Schlesinger, 597 F.2d 1214, 1223-24 (9th Cir.1979); Scherr v. Volpe, 466 F.2d 1027, 1034 (7th Cir.1972). Accord Swain v. Brinegar, 517 F.2d 766, 773-74 (7th Cir.1975), rev'd and remanded en banc on other grounds, 542 F.2d 364 (7th Cir.1976); Barta v. Brinegar, 358 F. Supp. 1025, 1028 (W.D.Wis.1973).

As of 1970, when the Act took effect, the ELF system consisted of no more than a test facility in northern Wisconsin. Since then, in a series of proposals, the Navy has proceeded with plans to make the system a fully operational, permanent one, transmitting synchronously with a coordinate system in Michigan. This change in form and focus is substantial enough to make the project subject to the provisions of the Act.

Moreover, the primary issue in this case is the Navy's decision not to prepare a supplemental environmental impact statement regarding potential biological effects of ELF electromagnetic radiation. Even if the construction of the revised ELF system did not represent substantial federal action covered by the Act, the Navy has a continuing duty to gather and evaluate information about the biological effects of electromagnetic radiation. Environmental Defense Fund v. Tennessee Valley Authority, 468 F.2d 1164, 1176 (6th Cir.1972).

I conclude that the current Project ELF proposal represents a substantial change from the system that was in operation before 1970 and, as such, it is subject to the requirements of the National Environmental Policy Act.

 
B. Standard of Review of an Agency Decision not to Prepare a Supplemental Environmental Impact Statement

The National Environmental Policy Act of 1969 imposes essentially procedural duties on administrative agencies considering actions that may affect the environment. 42 U.S.C. § 4332; Vermont Yankee Nuclear Power Corporation v. Natural Resources Defense Council, Inc., 435 U.S. 519, 558, 98 S. Ct. 1197, 1219, 55 L. Ed. 2d 460 (1978). The Act does not prescribe the outcome of substantive decisions; it merely requires agency decisionmakers to take into account the potential environmental ramifications of their actions. 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 National Environmental Policy Act has two objectives: 1) to ensure "that environmental concerns [are] integrated into the very process of agency decisionmaking," Andrus v. Sierra Club, 442 U.S. at 350, 99 S. Ct. at 2337, and 2) "to inform the public that the agency has considered environmental concerns in its decisionmaking process." Weinberger v. Catholic Action of Hawaii/Peace Education Project, 454 U.S. at 142-43, 102 S. Ct. at 201. These objectives are met through the requirement that environmental impact statements be prepared (thereby ensuring that the agency makes environmental considerations a part of the decisionmaking process) and disclosed publicly (thereby informing the public that the agency has considered environmental consequences in the decisionmaking process). Id. at 143, 102 S. Ct. at 201.

When an agency decides not to prepare an environmental impact statement, a court reviewing the legality of that decision must refrain from judging the merits of the proposed decision and determine *1357 only whether the agency followed the procedural mandates of the National Environmental Policy Act. Strycker's Bay Neighborhood Council, Inc. v. Karlen, 444 U.S. 223, 227-28, 100 S. Ct. 497, 499-500, 62 L. Ed. 2d 433 (1980) (per curiam). Neither the statute nor its legislative history contemplates that a court should substitute its judgment for that of the agency on the merits of the particular decision or on the environmental consequences of the proposed action. Kleppe v. Sierra Club, 427 U.S. at 410 n. 21, 96 S. Ct. at 2730 n. 21.

This circuit employs the "arbitrary and capricious" standard in reviewing an agency's decision; that is, the reviewing court must defer to the agency's determination of the need for an environmental impact statement unless the court finds that the determination was arbitrary or capricious, an abuse of discretion, or otherwise not in accordance with law. City of West Chicago, Illinois v. United States Nuclear Regulatory Commission, 701 F.2d 632, 651 (7th Cir.1983); Assure Competitive Transportation, Inc. v. United States, 635 F.2d 1301, 1308 (7th Cir.1980).

 
C. Navy's Obligation to File a Supplemental Environmental Impact Statement

Council on Environmental Quality regulations require that "[a]gencies ... [s]hall prepare supplements to either draft or final environmental impact statements if ... [t]he agency makes substantial changes in the proposed action that are relevant to environmental concerns; or ... [t]here are significant new circumstances or information relevant to environmental concerns and bearing on the proposed action or its impacts." 40 C.F.R. § 1502.9(c) (1) (i) and (ii) (emphasis added). By its own regulations, the Navy must comply with Department of Defense and Council on Environmental Quality regulations pertaining to assessments of the environmental consequences of proposed actions. 32 C.F.R. § 775.3(b) (1). Department of Defense regulations require supplementation of environmental impact statements in accordance with Council on Environmental Quality regulation 40 C.F.R. § 1502.9(c). 32 C.F.R. § 214.6(D) (4).

Plaintiffs contend that in this case the proposed changes in the project are so substantial and the new information relevant to environmental concerns so significant that the Navy must file a supplemental environmental impact statement. Plaintiffs base this contention chiefly on the new information concerning biological effects and the alleged inadequacy of the Navy's consideration of that information, but I will address as well the issue of the substantiality of the changes as they relate to the changes in the proposed project since 1977.[26]

 
1. Substantiality of changes in the proposed action

In determining whether a particular change is substantial, the reviewing court must evaluate the proposed modifications in relation to the previous environmental documentation. See, e.g., California v. Block, 690 F.2d 753 (9th Cir.1982); Environmental Defense Fund v. Marsh, 651 F.2d 983 (5th Cir.1981). If previous environmental documentation for a proposed action included discussion of an alternative that is finally selected, supplementation of the environmental impact statement is unnecessary. Natural Resources Defense Council, Inc. v. City of New York, 672 F.2d 292, 298 (2d Cir.), cert. dismissed, 456 U.S. 920, 102 S. Ct. 1963, 72 L. Ed. 2d 462 (1982). However, if neither the environmental consequences nor the overall feasibility of the changes in the proposed action can be discerned from the previous environmental documentation, a supplemental environmental impact statement is required. See Conservation Law Foundation v. *1358 General Services Administration, 707 F.2d 626, 632-34 (1st Cir.1983).

In this case, the Navy's 1977 environmental impact statement evaluated the proposed synchronous operation of the Wisconsin and Michigan test facilities and the proposed construction of a second test facility and antenna network in upper Michigan. Supplementary environmental documentation would be necessary for the 1981 proposal only to the extent the 1977 environmental impact statement did not address significant aspects of the current proposal.

 
a. Maintenance and modernization activities

Many of the proposed changes in the 1981 Project ELF proposal can be described as minimal. For example, equipping the Clam Lake facility with improved power amplifiers and electronic equipment is not a substantial change in the proposed action. Expansion of the existing transmitter building to accommodate the new electronic equipment, installation of standby electric power units, and replacement of the present security building with a larger one are maintenance and modernization activities which would have been expected as part of the design and system development activities in earlier phases of the ELF project. They are insubstantial changes which do not require supplementation of the 1977 environmental impact statement. Township of Springfield v. Lewis, 702 F.2d 426, 436-38 (3d Cir.1983).

 
b. Reduction of environmental impact

The 1981 ELF proposal incorporates many of the recommendations of the 1977 National Academy of Sciences study. Installing a fault detection system will lessen the risk of electric shock hazards from cable faults, and altering the design of the four ground terminals will minimize electric shock from current in the ground terminals, thus lessening the environmental impact of the project.

Both the 1977 Project Seafarer proposal and the 1981 Project ELF proposal envisioned synchronous operation of the existing Wisconsin ELF facility and the ELF facility to be constructed in upper Michigan. However, the 1981 Project ELF proposal changes the dimensions and nature of the proposed facility in upper Michigan. In the 1977 proposal, the Michigan facility was to consist of a transmitter station, an adjacent transmitter control center, and 130 miles of buried antenna cables. The 1981 Project ELF plan proposes construction in Michigan of four ground terminals, a transmitter building located on about two acres of land, and a reduced network of above-ground antennae of 56 miles. The 1981 proposal reduces the size of the antenna network and lessens environmental disruption from construction, maintenance, and repairs, because the antennae will be above ground. This will also serve to speed the diffusion of electric field intensities since air is a poorer conductor of electricity than is soil.

Generally, supplementation of environmental impact statements is not required when changes in the proposed action lessen the environmental impact of the action. Township of Springfield v. Lewis, 702 F.2d at 437-38; National Indian Youth Council v. Watt, 664 F.2d 220, 225 (10th Cir.1981); Concerned Citizens on I-190 v. Secretary of Transportation, 641 F.2d 1, 6 (1st Cir.1981). Contra National Wildlife Federation v. Marsh, 721 F.2d 767, 782-83 (11th Cir.1983); Environmental Defense Fund v. Marsh, 651 F.2d at 996-97. Of course, when there is a change in the expected benefits as well as in the environmental impact of a proposed action, the agency may need to reevaluate the balance between the environmental costs and economic benefits of the project, but that is not the situation in the present case. Massachusetts v. Watt, 716 F.2d 946, 950-51 (1st Cir.1983).

The changes in ELF that the Navy has made to reduce adverse environmental impacts and to improve mitigation techniques reflect efforts to comply with the National Environmental Policy Act which *1359 should not be deterred by the imposition of overly stringent supplementation requirements. Township of Springfield v. Lewis, 702 F.2d at 438; Concerned Citizens on I-190 v. Secretary of Transportation, 641 F.2d at 6. The proposed reduction in the size of the Michigan antenna network, the proposed lessening of electric field intensities away from the immediate vicinity of the antennae, and implementation of the National Academy of Sciences study's recommendations to reduce electric shock hazards demonstrate concern for environmental and human factors. These proposed changes reduce adverse environmental impact and do not require a supplemental environmental impact statement.

Although the construction of additional ground terminals in Michigan might increase the environmental impact, those construction activities are not a subject of this lawsuit.

 
c. Conversion to a fully operational system

Project Seafarer proposed the synchronous operation of two test facilities. Project ELF envisions a permanent, fully operational ELF communication system. If this conversion is a substantial change from the 1977 proposal, the Navy would be required to supplement the 1977 environmental impact statement.

Fashioning a test for determining whether a supplemental environmental impact statement is necessary, the Court of Appeals for the Ninth Circuit identified two factors to be considered: 1) whether the alternative finally selected is within the range of alternatives the public reasonably could have anticipated as being under consideration in earlier environmental documentation, and 2) whether public comments on earlier environmental impact statement proposals were adequate to inform the agency of the public's attitudes toward the chosen alternative. California v. Block, 690 F.2d at 772.

In this case, the Navy's 1977 environmental impact statement addressed the concept of synchronous operation of the two test facilities and the construction of the new facility. From the 1977 environmental impact statement, the public would have anticipated synchronous operation of Michigan and Wisconsin ELF test facilities and construction of an ELF facility in upper Michigan with an antenna network about double the size of that currently proposed.[27] However, the 1977 environmental impact statement did not refer to a permanent, fully operational ELF communication system which was to fulfill the communication needs of the entire United States submarine force. The question is whether the public would have anticipated such a permanent, fully operational system from the 1977 environmental impact statement and whether public comments informed the Navy of public sentiment toward the 1981 Project ELF.[28]

From the information in the 1977 environmental impact statement that the 1977 test operation was designed in part to find ways to reduce the system to a less intrusive size, the public would have anticipated the likelihood of a permanent ELF system. It is clear that the public could have anticipated that the Navy would not expend the *1360 considerable sums necessary to construct the ELF test facilities unless it intended to expand the system and operate it for a significant period of time. Some of the public comments submitted in response to the 1977 draft environmental impact statement focused on precisely that point, expressing concerns about the long-term operation of an ELF communication system in northern Wisconsin and upper Michigan.

Under the 1981 proposal, the ELF facilities would be operated twenty-four hours a day. This continuous operation of the test facilities could have been expected in 1977. In fact, the Bock Committee criticized the Navy for operating the Wisconsin test facility under conditions different from those proposed for the fully operational system, finding fault with the Navy's intermittent activation of the system. From such comments the public was on notice that the Navy would be likely to operate the Project Seafarer ELF system continuously.

Therefore, I conclude that the Project ELF proposal for a permanent, fully operational system changes the previously proposed Project Seafarer only incrementally and does not require supplementation of the 1977 environmental impact statement.[29]

I turn then to the issue that is at the heart of plaintiffs' case: whether significant new information on the biological effects of electromagnetic radiation has been generated since the Navy filed its 1977 environmental impact statement and, if so, whether the Navy gave adequate consideration to that information when it determined to proceed with Project ELF without filing a supplemental environmental impact statement.

 
2. Significance of new information of biological effects of electromagnetic radiation

Even after it files an environmental impact statement, a federal agency has a continuing duty to gather and evaluate new information relevant to the environmental impact of its proposed actions. California v. Watt, 683 F.2d 1253, 1267 (9th Cir.1982), rev'd on other grounds sub nom. Secretary of the Interior v. California, ___ U.S. ____, 104 S. Ct. 656, 78 L. Ed. 2d 496 (1984); Warm Springs Dam Task Force v. Gribble, 621 F.2d 1017, 1023 (9th Cir.1980) (per curiam); Society for Animal Rights, Inc. v. Schlesinger, 512 F.2d 915, 917-18 (D.C.Cir.1975); Stop H-3 Ass'n v. Lewis, 538 F. Supp. 149, 168 (D.Hawaii 1982). The agency must be particularly diligent when the proposed action includes a number of phases, to be implemented over a substantial period of time. Southern Oregon Citizens Against Toxic Sprays, Inc. v. Clark, 720 F.2d 1475, 1480 (9th Cir.1983). The Council on Environmental Quality has recommended that if a proposal has not yet been implemented or if an environmental impact statement concerns an ongoing program, the agency should re-examine any environmental impact statement that is more than five years old to determine whether 40 C.F.R. § 1502.9(c) requires preparation of a supplemental environmental impact statement. Council on Environmental Quality, Forty Most Asked Questions Concerning CEQ's National Environmental Policy Act Regulations, 46 Fed.Reg. 18,026, 18,036 (1981).

An agency is not required to file a supplemental environmental impact statement each time new information comes to light. However, when the new information is "significant," the agency must take care to evaluate the information and its potential *1361 impact. If, after conducting a thorough review, the agency decides not to file an environmental impact statement, it must support that decision with a statement of explanation or additional data. Warm Springs Dam Task Force v. Gribble, 621 F.2d at 1024.

The scope and nature of this statement would seem to depend upon the environmental importance of the new information. The more significant the information, the more substantial the written explanation must be.

However, the duty of careful evaluation does not change with the increasing or decreasing environmental importance of the new information. The duty remains constant: to conduct a thorough and comprehensive review of the new information and its potential consequences.

In evaluating the Navy's decision not to file a supplemental environmental impact statement, I have reviewed the administrative record to determine whether the record is sufficient to support the finding that the Navy acted reasonably, and not arbitrarily and capriciously in making that decision. Assure Competitive Transportation, Inc. v. United States, 635 F.2d at 1308 n. 7.

 
a. Environmental significance of the new information

I begin with the threshold question whether the information on biological effects generated since 1977 is significant enough to impose upon the Navy the duties of evaluation and explanation. Neither the Act nor the implementing requirements provide much help in defining significance. The regulations note that the information must be new and must be relevant to environmental concerns, 40 C.F.R. § 1502.9(c) (1) (ii), but provide no additional explanation.

In the absence of any clear directive from Congress or the Council on Environmental Quality, I have evaluated the significance of the new information in this case by considering whether the information is new (that is, whether it is more than a mere affirmation of earlier-known information); whether the information is meritorious (that is, whether experts in the field assess it as valid and important or worthy of further investigation); whether the information is accessible (that is, whether it was published or is otherwise available to the Navy); and finally, whether it is relevant to the environmental consequences of Project ELF (that is, whether it is the kind of information a rational decisionmaker would want to consider in making the decision to proceed or not to proceed with Project ELF).

In 1977, there was little or no evidence to contradict the findings and conclusions of the 1977 National Academy of Sciences study and the Navy's 1977 environmental impact statement that extremely low frequency electromagnetic radiation had no effect upon animal fertility, growth and development, or behavior. Scientific studies undertaken since then have produced results that raise questions about the validity of prior assumptions of the safety of extremely low frequency electromagnetic radiation.

In the areas of cellular function, enzymatic function, animal behavior, and growth and development, researchers such as Delgado, those at UCLA, and Adey and Blackman have found evidence that the biological effects of extremely low frequency electromagnetic radiation occur in a nonlinear dose-response relationship. These researchers and others have identified both the phenomenon of frequency and power intensity windows and the probable existence of two such windows at 15 Hz and 75 Hz.

Researchers at Batelle Laboratories have found teratogenic effects, as well as effects on animal behavior and on neurophysiology from exposure to electromagnetic fields at 60 Hz. Goodman at University of Wisconsin-Parkside has observed alternations in basic cell functions and in oxygen consumption in slime mold exposed to extremely low frequency electromagnetic radiation.

*1362 Delgado has found embryological effects from exposure to electromagnetic radiation.

The primate studies at Pensacola and at UCLA reveal effects upon growth rate and behavior from exposure to electromagnetic radiation.

In addition, epidemiology studies raise the possibility that there may be a correlation between the incidence of cancer and the magnetic fields associated with electric power lines. Although they are only suggestive at this time, the 1977 epidemiology studies, such as those by Wertheimer-Leeper, provide quantified data and descriptions of control group characteristics and selection criteria sufficient to allow outside evaluation of their merit, in contrast to the earlier epidemiology studies from Eastern Europe and the U.S.S.R.

These studies provide new information that was not available to the Navy or to the public in 1977. Although the research results do not prove that biological effects will result from exposure to extremely low frequency electromagnetic radiation, neither are they mere reconfirmations of the Navy's 1977 assessment of the negligibility of any potential environmental effects.

The merit of the new information has been attested to by the expert witnesses. At trial, expert witnesses for both plaintiffs and defendants identified post-1977 research either as new information worth further inquiry or as important research.

The new information was accessible to the Navy. It derives entirely from Navy-sponsored studies or from published articles.

The new information raises questions about long-term exposure to ELF electromagnetic radiation that should be taken into consideration by the decisionmaker. For example, the window studies cast doubt on the Navy's reliance on long-term hazard free experience with electric power lines as well as the Navy's discounting of research performed at frequencies and intensities different from those of the ELF facility. The studies demonstrate that the nature of the relationship between exposure and effect is not yet sufficiently understood to permit extrapolation from the presence or absence of effects at any one frequency. Moreover, the observation of a frequency window effect within the range at which the ELF antennae will operate is information that is highly relevant to an assessment of the environmental impact of the project.

The 1977 environmental impact statement is no longer adequate as a source of information necessary to a rational decision on the relative risks and benefits of Project ELF, because it includes none of the scientific work done since 1977 or any evaluation of that work as it might affect the Navy's 1977 assessment of the safety risk of the project. The post-1977 scientific research warrants a new look at the environmental consequences of the Navy's proposed ELF communication system which the 1977 environmental impact statement cannot provide.

I conclude that the scientific information on biological effects generated since 1977 is significant enough to require careful review by the Navy. I refrain expressly from finding that any one of the studies, or all of them taken together, invalidate the Navy's previous assertion that long-term exposure to ELF electromagnetic radiation will produce no adverse biological effects. The research results remain equivocal. The scientists do not agree on the import of that research, and I am not prepared to evaluate the substance of the individual studies or to weigh the relative merits of the various scientific opinions in this complex area. See Stop H-3 Ass'n v. Lewis, 538 F. Supp. at 168.

Despite the continuing uncertainty over the potential biological effects of electromagnetic radiation, those effects must be taken into account by the Navy because they represent significant new information relevant to the environmental consequences of the proposed action. This does not mean that the Navy must postpone operation of Project ELF until all uncertainty has been resolved. It does mean *1363 that the Navy must undertake a considered review of all that is known to date and evaluate the relative risks and benefits before proceeding with Project ELF.

 
b. Degree of care with which the Navy considered the new information

As I have found, when the Navy made its reactivation recommendation to the President and proceeded with implementation of the project, it did so without any review of the scientific information generated since 1977. At no time since those decisions were made or at any time since the filing of its final 1977 environmental impact statement together with the National Academy of Sciences study, has the Navy reviewed the scientific information on biological effects of electromagnetic radiation in any comprehensive manner.[30]

Defendants contend that the Navy's establishment of an Environmental Review Committee and an ecological monitoring program, the various research projects it funded, and its contract with IIT Research Institute for literature reviews and other work, demonstrate a high degree of concern for keeping abreast of the relevant developments in the field of electromagnetic radiation effects and a careful consideration of those developments.

However, my view of the Navy's efforts is that they are of relatively little use in assessing the potential danger of electromagnetic radiation on animal life. By themselves, the activities do not show a high degree of care and, in any event, they are not a substitute for a thorough review of the relevant information and a careful weighing of the risks and benefits of Project ELF.

For example, the Environmental Review Committee focused primarily on the impact of Project ELF on the physical environment. Although the committee devoted parts of two meetings to discussions of research on biological effects of electromagnetic radiation and public opposition to Project ELF based on the potential danger of adverse biological effects, the committee's evaluation of these subjects was cursory at best.

Similarly, the Navy's ecological monitoring program was devoted essentially to the environmental consequences of Project ELF. Only one of the thirteen research projects that made up the program was concerned with potential biological effects of extremely low frequency electromagnetic radiation and that was the Goodman slime mold field study at Clam Lake, for which there are no results as yet.

In addition to the slime mold study, the Navy has sponsored several animal studies. Of these, the Pensacola primate growth study and the UCLA monkey behavior study appear to be the most substantive. The Pensacola study was reviewed by a National Academy of Sciences panel and the UCLA study has been referred to frequently in other literature on biological effects of extremely low frequency electromagnetic radiation. However, other Navy studies such as the bird migration study, the multigenerational mice study, and the review of the physical condition of Navy workers at the Clam Lake site, produced inconclusive results or merely confirmed earlier findings. Moreover, the Navy never subjected any studies other than the Pensacola primate study to peer review and it never attempted a comprehensive review of the pieces of information generated by the studies it had sponsored.

With respect to the contract with IIT Research Institute for literature reviews, the evidence shows that IIT Research Institute evaluated articles and provided the Navy with studies and summaries of articles only on a sporadic basis and only after the decision had already been made not to supplement the 1977 environmental impact statement. It was not until June, 1983 that IIT Research Institute provided the Navy *1364 with any post-1977 literature or analysis of such literature. At that time, it forwarded to the Navy five articles and five Letters to the Editor. Later in the summer (about the time this lawsuit was filed), IIT Research Institute forwarded the World Health Organization report, the Montana Department of Natural Resources and Conservation report, and the Crocetti evaluation of the Wertheimer-Leeper studies.

IIT Research Institute did not evaluate either the state of knowledge in the field of biological effects or attempt to assess the cumulative findings of the scientific literature that had been published since 1977. If this is a failing on the part of IIT Research Institute, it does not relieve the Navy of any of its obligations under the National Environmental Policy Act. It is not IIT Research Institute, but the Navy that bears the responsibility for determining whether cumulative findings warrant further evaluation or whether a supplemental environmental impact statement must be prepared.

I conclude that the record does not demonstrate that the Navy fulfilled its duty of conducting a thorough and comprehensive review of the new scientific information on the biological effects of extremely low frequency electromagnetic radiation and the significance of that information.

 
c. Sufficiency of Navy's explanation of its decision not to file a supplemental environmental impact statement.

At the time it made the recommendation to the President to reactivate Project ELF, the Navy did not provide a statement of explanation or data to support its decision not to supplement the 1977 environmental impact statement. When Admiral Smith made the decision that it was not necessary to file a supplemental environmental impact statement, he did so without providing any written explanation for his decision. At no time since then has the Navy given an explanation of its decision to continue with the construction of Project ELF without filing a supplemental environmental impact statement.

Neither the June, 1983 nor the October, 1983 environmental assessment supplies the requisite explanation of the Navy's earlier decision not to prepare a supplemental environmental impact statement. These assessments were directed primarily to construction activities. In the June, 1983 environmental assessment on the upgrading of the Wisconsin facility, the Navy relied on the 1977 National Academy of Sciences study and the 1977 environmental impact statement for evaluations of electromagnetic field effects. In the October, 1983 environmental assessment of the construction of the Michigan facility, the Navy did not discuss effects of electromagnetic radiation (because the subject had been thoroughly covered elsewhere) but merely reiterated the conclusion of the 1977 environmental impact statement that there was no credible evidence of adverse human health or environmental effects. These assessments do not demonstrate the reasonableness of the Navy's decision not to prepare a supplemental environmental impact statement.

I find that the record does not demonstrate that the Navy provided an explanation or additional data for its decision not to prepare and file a supplemental environmental impact statement.

In summary, I find and conclude that in proceeding with the reactivation of Project ELF without undertaking a thorough and comprehensive review of the significant new information on biological effects of electromagnetic radiation that has been generated since 1977 the Navy abused its discretion. In so proceeding, the Navy acted in violation of the National Environmental Policy Act.

 
D. Remedy

Remedies for violations of the National Environmental Policy Act should be consistent with the statutory objectives and the broader public interest. Because the procedures of the National Environmental Policy Act are designed to ensure that environmental information is available *1365 to public officials and the public before decisions are made, injunctive relief is proper if an agency commits itself to a course of action without complying with the mandated procedures. Massachusetts v. Watt, 716 F.2d at 952; People Against Nuclear Energy v. United States Nuclear Regulatory Commission, 678 F.2d 222, 234-35 (D.C.Cir.1982), rev'd on other grounds sub nom. Metropolitan Edison Co. v. People Against Nuclear Energy, ___ U.S. ___, 103 S. Ct. 1556, 75 L. Ed. 2d 534 (1983). The issuance of an injunction preserves the status quo so that there remains an opportunity for decisionmakers and the public to choose among alternative courses of action. Furthermore, it provides an incentive for rapid compliance with the National Environmental Policy Act. Environmental Defense Fund v. Marsh, 651 F.2d at 1005.

Defendants argue that an injunction is an inappropriate remedy in this case, citing Weinberger v. Romero-Barcelo, 456 U.S. 305, 102 S. Ct. 1798, 72 L. Ed. 2d 91 (1982). However, that case addressed violations of the Federal Water Pollution Control Act, a statute imposing substantive environmental obligations. In contrast, the National Environmental Policy Act imposes procedural obligations on federal agencies. When an agency makes a decision without complying with the National Environmental Policy Act, the harm the Act seeks to prevent has been suffered and injunctive relief is appropriate. Massachusetts v. Watt, 716 F.2d at 952.

In this case, I have found there is significant new information which the Navy must evaluate in a thorough and comprehensive manner. However, the Navy's duty does not end with an evaluation. The Navy has the further duty of filing either a supplemental environmental impact statement or an adequate explanation of its decision not to file such a statement.

I am not in a position to determine the full significance of the new information on biological effects. It may be that the significance of the new information is such that the Navy could have fulfilled its obligation of explanation in some way other than by filing a supplemental environmental impact statement. However, at this time it would not be in the public's interest or in the Navy's to permit the Navy to go forward with Project ELF without requiring it to file a supplemental environmental impact statement. The public has displayed an interest in this case and in the human health implications of Project ELF. A supplemental environmental impact statement will allow the public to address the issue of biological effects, as well as serving to advise the public that the Navy has taken the issue into consideration. The Navy has an interest in avoiding future litigation over the sufficiency of any explanation short of a supplemental environmental impact statement.

Moreover, the entire thrust of the National Environmental Policy Act is to encourage the filing of environmental impact statements to ensure that agencies make environmental considerations a part of their decisionmaking process and to inform the public that the agencies have considered environmental consequences in the decisionmaking process.

 
V. ORDER

IT IS ORDERED that plaintiffs' motion for a permanent injunction is GRANTED and defendants are enjoined from taking any further action in respect to construction of the new ELF facility in Marquette County, Michigan, to upgrading the existing ELF facility in Wisconsin, or to supplying submarines with ELF receivers until they have prepared and filed a supplemental environmental impact statement in compliance with the requirements of the National Environmental Policy Act.

NOTES

[1] Very low frequency radio waves are those in the frequency range from 3 to 30 kilohertz (kHz) with corresponding wavelengths from 62 to 6.2 miles long.

Electricity can flow as direct current (constantly flowing in one direction through a circuit) or as an alternating current (with the direction of the flow changing). The rate at which electric current alternates is referred to as its frequency which is measured in cycles. A cycle encompasses initial flow in one direction, then the other direction, and then back to the first direction. A hertz is the unit of cycles per second; that is, one cycle per second is the equivalent of 1 hertz (Hz).

[2] When operational, the Michigan and Wisconsin test facilities will be able to transmit independently of each other only if operated at different frequencies. The current proposal envisions separate operation only for maintenance or other backup purposes.

[3] In contrast to the well-studied subject of ionizing radiation, nonionizing radiation such as extremely low frequency electromagnetic radiation is a relatively recent subject of scientific inquiry. (Ionizing radiation is electromagnetic radiation that produces ionization in otherwise neutral media; above a certain level of energy, the photon carries sufficient energy to break bonds and alter the structure of a molecule. Examples of ionizing radiation include x-rays, gamma-rays, and high-speed charged particles.) Although many organizations have devoted themselves to the study of ionizing radiation since World War II, only three organizations, all formed after 1975, Bioelectromagnetism, Bioelectric Repair and Growth Society, and International Society of Bioelectricity, study nonionizing radiation and its effects.

[4] The energy radiated by an extremely low frequency antenna produces a nonionizing electromagnetic field which consists of a magnetic field and an electric field. An electric current passing through a wire creates a magnetic field, the intensity of which is expressed in gausses. Because a magnetic field is produced by the flow of electric current, its intensity is directly proportional to the concentration of the current flow through an antenna. (For reference, the earth's natural magnetic field is relatively constant at 0.5 gausses.)

Electric fields are described in terms of voltage, the pressure that pushes current through a circuit over a given distance. The typical measurement is expressed in terms of volts per meter. Electric fields exist in a variety of media, including for the purposes of electromagnetic radiation, air and soil. Electric fields in soil are expressed in terms of step potential, the voltage that would exist between a person's feet when taking a step equal to one meter in length. Electric field intensity depends directly on antenna current and frequency and is also affected by the near-surface electric conductivity of the earth.

[5] According to the committee, it derived its conclusions from evaluations of pertinent studies and literature searches in which the committee disregarded poorly designed studies, unreliable studies (those that had not yielded the same results on repetition), results at exposures different from those proposed for the ELF communication system, and preliminary studies which had not been confirmed independently by other studies or by independent investigation of the results.

[6] From the administrative record, it appears that the Navy continued a bird migration study, one study of mice growth, and two primate studies: one on growth and one on behavior. Although the Navy never evaluated the Electric Power Research Institute's bee studies, it did include bees as a subject of study in the ecological monitoring program.

[7] Although this August, 1981 recommendation is referred to throughout the administrative record and in the trial testimony of Admiral William D. Smith, Director of Naval Communications in the Office of the Chief of Naval Operations from July, 1980, to July, 1983, the recommendation itself is not part of the administrative record which is before the court for review. Also missing from the administrative record is the recommendation referred to in President Reagan's April, 1981 memorandum.

[8] In a separate environmental impact assessment which was published in the Federal Register on October 21, 1983, after the trial in this case, and is not a part of the administrative record, the Navy evaluated the activities associated with the Michigan portion of Project ELF: the construction of the transmitter and ancillary support buildings, installation of the antenna lines, and other aspects of the construction and installation project. The Navy concluded that construction and installation of the ELF facility in upper Michigan would cause no significant impact on the environment. It did not consider biological or environmental effects from operating the communication system

as that subject has been thoroughly addressed in previous environmental documentation. The Navy has, continuously since the publication in 1977 of its Environmental Impact Statement, reviewed new scientific information on extremely low frequency propagation with the conclusion that there is no credible evidence of adverse effect to human health or the environment.

48 Fed.Reg. at 48,862.

[9] Although the Navy referred to this safety standard, a close review of the World Health Organization publication, which was supposed to have set a standard, reveals no such standard. See infra discussion at 1347.

[10] The term "window" seems to be used by these researchers to describe that point or points on the spectrum of frequencies or power intensities at which an effect of exposure is observed. In order situations, such as rocket launching, for example, "window" is used to describe that interval in time or that particular set of circumstances under which an event can occur. In a somewhat similar sense, a power intensity or frequency window would be that point at which there is an opening or an opportunity for an effect to occur from a dose or an exposure.

[11] The Navy initiated an additional, short-lived research effort. In 1970, it began a study focusing on biological evaluations of the 24 workers at the Wisconsin test facility with a control group established at another Navy facility. Although it abandoned the study in 1978 without reaching any significant conclusions, the facility's contractor, GTE-Sylvania, conducted initial quarterly and subsequent semi-annual physicals of employees at the Wisconsin ELF site as a matter of corporate policy, beginning in April, 1976. These examinations included electrocardiograms, serum chemistry profiles, blood counts, and serum triglyceride and blood pressure determinations. In August, 1978, the GTE-Sylvania medical department summarized the findings from the medical surveillance in a one-page letter, concluding that there were no abnormal trends.

[12] In late October, 1978, IIT Research Institute prepared a summary of a workshop on the blood brain barrier in experimental animals exposed to microwave energy. It is difficult to understand why this report is included in the administrative record since the workshop dealt with microwaves, which are very short waves, and with very low frequency radiowaves, rather than with the long waves of extremely low frequency electromagnetic radiation. There is no definition in the record of a blood brain barrier or any explanation of its significance.

[13] This data is of little use, because of changes in the methods of collecting the data, the absence of any analysis of the data, and the gaps in reporting.

[14] The thirteenth research project was the Goodman slime mold study. Because that study concerned biological rather than ecological effects, I have summarized it as part of the discussion of the Navy-sponsored research in biological effects, above at 1342-1343.

[15] Articles published in peer review journals are screened in advance of publication by other in the same fields.

[16] In this study, the magnetic field was measured in microTesla (mT). One microTesla equals 0.01 gauss. I have converted the measurements into gauss for ease of comparison. The greater effect was observed at 1.2 mT, while exposures to 12 mT and 0.12 mT had less of an effect.

[17] See supra note 10 at 1342.

[18] The term calcium efflux is not defined in the record.

[19] The study measured the magnetic field intensities in microTesla.

[20] In the chapter, the author identified the magnetic field intensity in microTesla, 0.3 mT, which translates to 0.003 gauss.

[21] Dr. Sheppard cited 253 sources for information on biological effects of electromagnetic radiation. Of these, 101 were issued before or during 1977, and 152 were issued after 1977. Dr. Sheppard discussed 31 pre-1977 sources that disclosed relationships between electromagnetic radiation and biological effects, and 29 reports that found no such relationship. With regard to the post-1977 research, 16 sources found no effects, while 68 sources found effects. Of the sources finding no effects, nine were issued after 1977 but before 1981, five were issued in 1981, and two were issued after 1981. Of those finding an effect, 24 were issued after 1977 but before 1981, 16 were issued in 1981, and 28 were issued after 1981.

[22] At trial, Admiral Smith testified that he decided that the scientific literature regarding biological effects of electromagnetic radiation did not warrant diverting resources to evaluate such studies. At the time he made this decision, the administrative record did not contain any scientific literature. Thus, I find that when Admiral Smith made this decision he did so without ever having reviewed the scientific literature he characterized as not significant.

[23] As of July, 1982, the Navy began funding slime mold research at the Clam Lake facility as part of its ecological monitoring program. It is unclear from the administrative record whether it was the Navy that sponsored any of the post-1977 laboratory slime mold research including that reported in the 1979 and 1983 articles.

[24] Justesen is in error on this point.

[25] In their answer defendants raised a third defense, that of laches. From their failure to refer to this defense in their briefs, I assume that defendants have abandoned it. Taylor v. Fee, 233 F.2d 251, 259 (7th Cir.1956).

[26] In Section IV A. 2, supra, the issue was the substantiality of the changes in the ELF communication facility proposal since January 1, 1970. In this section the issue is the narrower one of the substantiality of the changes that have been made in the proposal since the last environmental impact statement was filed in 1977.

[27] At the time of the 1977 proposal, the Navy conceived of a fully operational system consisting of five transmitter stations and 2400 miles of buried antenna cables dispersed in a 4000 square mile area. In contrast, under the 1981 proposal, the dimensions of the fully operational Michigan facility approximate those of the 1977 proposed test facility for upper Michigan and the dimensions of the fully operational Wisconsin facility will be no larger than the original test facility.

[28] From the 1977 environmental impact statement, the public would not have expected the ELF communication system to fulfill the communication needs of the entire United States submarine force. However, this programmatic change has little, if any, environmental significance. Although plaintiffs have argued that Project ELF and perhaps this 1981 proposed change would make northern Wisconsin and upper Michigan more likely to be a target in a nuclear attack, this sort of harm is not cognizable under the National Environmental Policy Act. See Metropolitan Edison Co. v. People Against Nuclear Energy, ___ U.S. ___, 103 S. Ct. 1556, 75 L. Ed. 2d 534 (1983).

[29] An argument could be made that, given the Navy's history of preparing environmental impact statements at various phases of the ELF communication system, the public would have assumed that further environmental documentation would be forthcoming before the Navy would decide to convert the test system into a fully operational communication system. In fact, the Navy assured the public and the Environmental Protection Agency of its intention to do so. However, these promises were made in the face of the expansive full-scale system envisioned in the earlier Project Seafarer proposal. The 1981 proposal does not approach the scope of the previously envisioned system. Even though the system will be permanent and fully operational, it is not a substantial change from the 1977 proposal.

[30] At trial, Drs. Straub and Justesen devoted much of their testimony to critical analyses of recent biological effects research. Such analyses, in a more thorough and comprehensive form, are the kinds of evaluation that the Navy might have obtained before it made its reactivation recommendation.