Natural Resources Defense Council, Inc. v. Evans

OPINION AND ORDER GRANTING PLAINTIFFS’ MOTION FOR A PRELIMINARY INJUNCTION LAPORTE, United States Magistrate Judge. TABLE OF CONTENTS Page I. INTRODUCTION.1012 II. LIKELIHOOD OF PREVAILING ON THE MERITS.1013 A. Scientific Background, Including Basis for 180 dB Threshold.1014 B. Marine Mammal Protection Act.1017 1. Specified Geographic Region.1019 2. Small Numbers.1023 a. Statute of Limitations.1024 b. Whether NMFS Acted Outside the Scope of its Authority.1024 3. The Final Rule’s Definition of “Harassment”.1027 a. Potential to Disturb.1028 b. Significance Requirement.1028 c. Impact on Individual Mammals.1030 4. Negligible Impact .1032 5. Mitigation and Monitoring.1033 C. National Environmental Policy Act.1037 1. Reasonable Alternatives Analysis.1038 2. Consideration of Reasonably Foreseeable Environmental Impacts.1041 3. The Navy’s Refusal to Supplement the EIS.1042 4. Reliance on Unpublished White Paper not Subject to Public Comment-1044 D. Endangered Species Act.1045 1. Regulatory Definition of Adverse Modification.1046 2. Incidental Take Statements.1048 III. INJUNCTIVE RELIEF.1051 I. INTRODUCTION Plaintiffs, various environmental organizations and a concerned individual, seek a preliminary injunction against federal officials to prevent the United States Navy’s peacetime use of a low frequency sonar system for training, testing and routine operations. 1 This new technology, Surveillance Towed Array Sensor System (“SUR-TASS”) Low Frequency Active Sonar (“LFA”), sends out intense sonar pulses at low frequencies that travel hundreds of miles in order to timely detect increasingly quiet enemy submarines. Plaintiffs charge that the National Marine Fisheries Service *1013 (“NMFS”) improperly approved use of SURTASS LFA in as much as 75 percent of the world’s oceans in violation of the Marine Mammal - Protection Act (“MMPA”), the Endangered Species Act (“ESA”), the National Environmental Policy Act (“NEPA”), and the Administrative Procedure Act (“APA”). Plaintiffs claim that these violations will cause irreparable injury by harassing, injuring and killing marine mammals with sensitive hearing and other sea creatures, many of them rare and endangered, including whales, dolphins, seals, sea turtles and salmon. Defendants counter that they have fully complied with the applicable laws. Defendants argue further that enjoining the peacetime use of LFA sonar would harm national security, even though they would still be free to use it during wartime or periods of heightened threat, because training and testing is necessary for military readiness. As explained below, the Court recognizes the importance of this new sonar technology to national security. The Court also commends defendants’ sponsorship of independent scientific research to advance our limited understanding of the effects of low frequency sound on marine mammals. Plaintiffs have shown, however, that they are likely to prevail on a number of issues. These include the likelihood of establishing that the authorization of harassment of up to 12 percent of marine mammals violates the “small numbers” limitation and that NMFS has im-permissibly narrowed the definition of harassment, in violation of the MMPA; that NMFS acted arbitrarily in postponing the designation of additional “off limits” areas within the ocean where marine mammals and endangered species are likely to be particularly abundant, and did not sufficiently analyze reasonable alternatives, in violation of NEPA; and that, by relying on an illegal regulatory definition of adverse modification and not including proper incidental take statements in its two biological opinions, NMFS violated the ESA. Plaintiffs have also raised serious questions on the merits on the issues of whether NMFS acted arbitrarily and capriciously in choosing the specified geographic regions identified in the Final Rule, and whether the taking authorized will have more than a negligible impact on marine mammals. However, the Court is not predicating any injunctive relief upon these issues. Defendants are likely to prevail on the remaining issues. The Court concludes that a preliminary injunction should issue. Plaintiffs have shown the likelihood of irreparable injury and of a future violation of the ESA. At the same time, the Court must consider the public interests both in national security and in protecting marine mammals and endangered species. Accordingly, the Court concludes that a carefully tailored preliminary injunction should issue, which permits the use of LFA sonar for testing and training in a variety of ocean conditions, but provides additional safeguards to reduce the risk to marine mammals and endangered species. II. LIKELIHOOD OF PREVAILING ON THE MERITS The Court reviews challenges under the MMPA, ESA, NEPA, and APA to ensure that the agency has not acted in a manner that is “arbitrary, capricious, an abuse of discretion, or otherwise not in accordance with law.” Okanogan Highlands Alliance v. Williams, 236 F.3d 468, 471 (9th Cir.2000); 5.U.S.C. § 706. To obtain a preliminary injunction, the moving party must show either (1) a combination of probable success on the merits and the possibility of irreparable injury or (2) that serious questions are raised and the balance of hardships tips in its favor. These two formulations *1014 represent two points on a sliding scale in which the required degree of irreparable harm increases as the probability of success decreases. Roe v. Anderson, 134 F.3d 1400, 1402 (9th Cir.1998). “In exercising their sound discretion, courts of equity should pay particular regard for the public consequences in employing the extraordinary remedy of injunction.” Weinberger v. Romero-Barcelo, 456 U.S. 305, 312-13, 102 S.Ct. 1798, 72 L.Ed.2d 91 (1982) (citing Railroad Comm’n. v. Pullman Co., 312 U.S. 496, 500, 61 S.Ct. 643, 85 L.Ed. 971 (1941)). A. Scientific Background, Including Basis for 180 dB Threshold One important scientific dispute between the parties is whether the standard of 180 decibels (“dB”) adopted by NMFS as the threshold for probable injury to marine life is arbitrary and capricious. Decibels measure sound intensity or loudness on a logarithmic scale; for example, a sound measuring 180 dB is approximately ten times more intense than a 170 dB sound. LFA sonar uses low frequency sound waves which travel farther distances in the ocean with less loss of intensity than higher frequency sound waves. Plaintiffs contend that substantial scientific evidence indicates that injury to marine mammals occurs at much lower levels than 180 dB. Defendants respond that an independent scientific team arrived at the 180 dB threshold based on a review of the relevant literature, the results of a specially designed Scientific Research Program (“SRP”), and underwater acoustical modeling. (2nd Johnson Dec. at ¶ 7, 9-11; EIS p. 4.201.) Under the SRP, independent scientists designed controlled scientific studies of the impact of LFA sonar on marine mammals at sound levels between 120 dB and approximately 155 dB, which they conducted over a one-year period using an LFA-equipped ship provided by the Navy. The studies tested the effect of LFA sonar on four species of endangered baleen whales, which specialize in hearing sounds in the low frequency range in which LFA sonar operates, and thus were expected to be most sensitive to LFA sonar. The results surprised the scientists: Prior to the LFS SRP, the expectation was that whales would begin to show avoidance responses at RLs [Received Levels] of 120 dB. Immediately obvious avoidance responses were expected for levels > 149 dB. The LFS SRP experiments detected some short-term behavioral responses at estimated RLs between 120-155 dB. In the Phase II research, avoidance responses were sometimes obvious in the field when the LF source was in the gray whale migration path. Although several behavioral responses were revealed through later statistical analysis, there was no significant change in a biologically important behavior detected in any of the three phases. Most animals that did respond returned to normal baseline behavior within a few tens of minutes. (EIS at p. ES-16.) Dr. Tyack, Senior Scientist at Woods Hole Oceanographic Institution, originally became concerned about the potential impact of the LFA sonar system on whales as a volunteer science advisor to plaintiff NRDC. (Tyack Decl. ¶ 6.) He was subsequently retained by the Navy as one of two principal investigators of SRP. He explained: The SRP was designed to study exposure ranges from 120-160 dB, the range in which we expected to see significant responses. The experiments were carefully designed to start at the low end of this exposure range and slowly work up, stopping at the level at which significant responses were observed. The results *1015 from the SRP show minor enough responses that most scientific reviewers have urged further study of higher exposure levels, at least from 160-180 dB. (Tyack Dec. at ¶¶ 29-30.) Similarly, Dr. Clark, the other principal investigator of the SRP concluded: The SRP results support the conclusion that the received level at which behavioral responses occur is around 140 dB, not 120 dB as expected based on the earlier gray whale research. This result cannot be extended to arctic species such as the bowhead and beluga whales, which are known to be extremely sensitive to noises from human activities. However, since the Navy has stated that LFA will not be used in the Arctic, these sensitive species are not placed at risk. For lower latitude areas, the SRP results reduce the scale of potential impact by as much as several orders of magnitude. The results showing that responses last for only tens of minutes and involve modest changes in behavior does not mean that animals are not responsive to LFA sounds. It means that their response levels are much less than those expected based on the best evidence available prior to the SRP. (Clark Dec. at ¶ 22; see also Fristrup Dec. at ¶ 11 (the brevity and subtlety of the behavioral responses observed “are strong indicators that LFA exposures at received levels up to 155 dB could not affect surviv-orship or reproduction.”)). The SRP did not test responses of marine mammals to LFA sonar at received levels above 155 dB, but instead used modeling to extrapolate from a presumed 95 percent risk at a received level of 180 dB. According to Dr. Kurt Fristrup, Having attached the Risk Continuum to 95% risk at 180 dB, the remaining choice involved determining how rapidly risk declined with decreasing received level. The OEIS/EIS incorporates a plausible, worst-case assumption that biologically significant behavioral reactions could begin to appear at received levels just above the received levels we achieved in the experiments (155 dB). As a result, the Risk Continuum provides an upper bound for the plausible impact of LFA signals in the range of received levels for which no experimental evidence is available. For example, a 50% risk value is assigned for exposure to one LFA signal at a received level of 165 dB. Thus, we expect that there is less than a 50% chance that exposure to a single LFA broadcast at 165 dB could result in a biologically significant response. (Fristrup Déc. at ¶¶ 16-18.) In addition to results from the SRP, defendants relied on extrapolations from levels of sound that cause injury to other species, including humans and guinea pigs. 67 Fed.Reg. 46779. Mr. Johnson, Technical Director for the Chief of Naval Operations, explains that “[hjearing loss due to sound exposure is well studied in humans and other land animals, but data for marine mammals are sparse. These data gaps ... prompted the use of models and extrapolations, in order to provide a rational basis for the assessment of risk potential.” (2nd Johnson Dec. at ¶ 8.) Plaintiffs argue that the mass stranding of marine mammals, primarily beaked whales, in March 2000 in the Bahamas, which the Navy’s and NMFS’s own task force linked to military exercises involving the use of underwater mid-frequency sonar, demonstrates that the injury threshold of 180 dB is too high. The task force report indicated that the injured whales were likely exposed to levels of 165 dB. The whales sustained hemorrhages in the inner ear, in some tissues adjacent to the ear, and in the fluid spaces surrounding the brain, as well as clotting in the cerebral ventricles, although their deaths ap *1016 parently resulted most immediately from protracted exposure upon beaching. (Ket-ten Dec. at ¶¶ 16-17.) Rescuers returned some whales to the sea, but those whales have not been seen again. (Balcomb Dec. at ¶ 10.) Plaintiffs’ expert contends that they either died at sea or were largely driven to abandon their habitat. (Balcomb Dec. at ¶ 11.) Defendants’ expert argues that there is not enough data on resighting rates to support this conclusion. (Ketten Dec. at ¶ 38.) Plaintiffs’ expert Balcomb, however, testified at the hearing that he and his scientific team had studied beaked whales in the area prior to the mass stranding and identified and photographed 35 as frequent visitors. Yet, in the two years since the stranding, his team has seen only one of the previously identified whales return to the area. The experts on both sides agree that the mechanism of injury in the Bahamas strandings is unknown. (Potter Dec. at ¶ 15; Cudahy Dec. at ¶ 17.) What they dispute is the implication to be drawn from this lack of scientific knowledge; in particular, whether LFA, which operates at much lower frequencies than the sonar involved in the Bahamas strandings, is likely to cause similar injuries. Plaintiffs expert Dr. Potter, Research Associate Professor of the Tropical Marine Science Institute, opines that “[sjince the mechanism is unknown, it is not scientifically justifiable to assume anything about its frequency-dependence. The mechanism may well also apply at low frequencies, such as those used by the SURTASS LFA sonar.” (Potter Dec. at ¶ 15.) Defendant’s experts Dr. Cudahy and Dr. Fristrup counter that it is speculative to assume that the unknown mechanism involved in the strandings related to mid-frequency sonar will apply when low frequency sonar is employed. For example, Dr. Cudahy states that: [while] there is little data on the non-auditory physiological impact of mid-frequency underwater sound on animals or humans.... there is an extensive data set on non-auditory and auditory injury due to low frequency underwater sound, collected on over 500 animals and over 100 humans. The conclusions drawn in the EIS regarding tissue damage at low frequencies were based in part on these data. Currently, there is no established mechanism for the tissue damage observed in the marine mammals stranded in the Bahamas in March 2000 and very little data to bring to bear on what happened. Nor is there a data set collected on other animals exposed to mid-frequency underwater sound that addresses non-auditory damage. This makes extrapolation from mid-frequency data to low-frequency data very problematic. In order to make such an extrapolation, clear physiological data on a large sample set (tens or hundreds) of animals exposed to mid-frequency underwater sound is needed. In the absence of such data all that exists are hypotheses. Again, the best extrapolations will be made from data collected in the same frequency region and for comparable organ systems. Thus, it is incorrect to draw the kind of correlation that the plaintiffs in this case assert between the stranding of beaked whales in the Bahamas when exposed to mid-frequency sonar and possible impacts to marine mammals associated with the operation of the low frequency SURTASS LFA. (Cudahy Dec. at ¶ 17-18.) Similarly, Dr. Fristrup points out that natural sources of loud low frequency sound, such as earthquakes and lightning strikes, are common in the ocean, so marine mammals likely adapted to such loud low frequency sounds in their evolution. (Fristrup Dec. at ¶ 25.) Dr. Ketten concludes: *1017 We are logically compelled to infer that the traumas observed in the Bahamian strandings, whatever the mechanism, is species or taxa specific and is not common to whales much less marine mammals per se. For this reason alone it is highly inappropriate to construe this event as an indicator of similar events in other whales from the same or other sonars or acoustic devices. . . (Ketten Dec. at ¶ 33.) The EIS states that “[e]urrent evidence would suggest that[,] while beaked whales may be sensitive to frequencies above SURTASS LFA sonar, there is little evidence that they are more sensitive to LFA sounds than the species selected as subjects for the LFS SRP.” (Stafford Dec. Ex. 1 at 3.2^17) (as corrected). Plaintiffs respond that defendants themselves have extrapolated from mid- and even high-frequency sound to low frequency sound, and across species, even though they chastise plaintiffs’ experts for doing so with respect to the mass strandings. (Navaro Dec. Ex. 14 at 1-23 to 1-28, 4.2-21 to 4.2-23; 67 Fed.Reg. 46737, 46740-46741; Navaro Dec. Ex. 3 at 104-06). Furthermore, there have been prior strandings correlated with the use of mid-frequency military sonar, although these events were less well studied. (Stafford Dec. Ex. 20.) For example, another mass stranding of beaked whales occurred along the west coast of Greece in 1996, which was correlated with the movements of an active sonar system operated by NATO in both low and mid-frequency bands. (Stafford Dec. Ex. 20.) It appears to the Court that both plaintiffs’ and defendants’ experts make reasonable points about the possible implications of the strandings, but that both sets of experts must, of necessity, engage in some speculation, given the current state of scientific uncertainty. The possibility that the stranding in the Bahamas, and other strandings, could foretell similar injuries from LFA sonar is very troubling. It would be more protective of marine mammals to adopt the plaintiffs’ experts’ more conservative approach to uncertainty and not deploy LFA sonar unless and until further scientific research rules out a similar impact from LFA sonar. The law is clear, however, that when qualified experts on both sides -reach carefully reásoned but different conclusions, the Court must defer to the agency’s experts: “When specialists express conflicting views, an agency must have discretion to rely on the reasonable opinions of its own qualified experts even if, as an original matter, a court might find contrary views more persuasive.” Marsh v. Oregon Natural Resources Council, 490 U.S. 360, 378, 109 S.Ct. 1851, 104 L.Ed.2d 377 (1989) (quoting Citizens to Preserve Overton Park Inc. v. Volpe, 401 U.S. 402, 416, 91 S.Ct. 814, 28 L.Ed.2d 136 (1971)). In conclusion, while the Court is concerned about potential dangers of LFA operated at levels below 180 dB, on this record, plaintiffs have not shown that they are likely to prevail on their argument that adoption of the 180 dB threshold was arbitrary and capricious, even in light of the stranding in the Bahamas and other mass strandings. Qualified experts, many of whom are not Navy employees but researchers at major scientific institutions such as Woods Hole, adopted the 180 dB threshold based on reasonable original research and review of the literature. While plaintiffs’ qualified scientists reach different conclusions in an area of scientific uncertainty and legitimate disagreement among experts, the Court is not empowered to adopt their views in lieu of the reasonable views of defendants’ qualified experts. B. Marine Mammal Protection Act The Marine Mammal Protection Act (“MMPA”) was enacted in 1972 to prevent *1018 the extinction or depletion of marine mammal stocks as a result of man’s activities. 16 U.S.C. § 1361(1). “[S]ueh species and population stocks should not be permitted to diminish beyond the point at which they cease to be a significant functioning element in the ecosystem of which they are a part, and, consistent with this major objective, they should not be permitted to diminish below their optimum sustainable population.” 16 U.S.C. § 1362(2). The MMPA generally prohibits the taking of marine mammals, with certain statutory exceptions. 16 U.S.C. § 1371(a)(3). “Take” is defined as “to harass, hunt, capture, collect, or kill, or attempt to harass, hunt, capture, collect or kill, any marine mammal.” 50 C.F.R. § 216.3; 16 U.S.C. § 1362(13). The definition of “take” includes any negligent or intentional act which results in disturbing or molesting a marine mammal. 50 C.F.R. § 216.3. The MMPA defines “harassment” as “any act of pursuit, torment or annoyance” that: (i) has the potential to injure a marine mammal or marine mammal stock in the wild; or (ii) has the potential to disturb a marine mammal or marine mammal stock in the wild by causing disruption of behavioral patterns, including but not limited to, migration, breathing, nursing, breeding, feeding, or sheltering. 16 U.S.C. § 1362(18)(A). Harassment as defined in subsection (I) is referred to as Level A harassment. 16 U.S.C. § 1362(18)(B). Harassment as defined in subsection (ii) is referred to as Level B harassment. 16 U.S.C. § 1362(18)(C). Citizens of the United States who engage in a specified activity other than commercial fishing within a specified geographical region may petition the Secretary to authorize the incidental, but not intentional, taking of small numbers of marine mammals within that region. 16 U.S.C. § 1371(a)(5)(A). Such authorization is limited to a period of not more than five consecutive years. Id. The Secretary “shall allow” the incidental taking if the Secretary finds that “the total of such taking during each five-year (or less) period concerned will have a negligible impact on such species or stock and will not have an unmitigable adverse impact on the availability of such species of stock for taking for subsistence uses .... ” Id. If the Secretary allows the incidental taking, the Secretary also must prescribe regulations setting forth: (I) permissible methods of taking pursuant to such activity, and other means of effecting the least practicable adverse impact on such species or stock and its habitat, paying particular attention to rookeries, mating grounds, and areas of similar significance, and on the availability of such species or stock for subsistence uses; and (ii) requirements pertaining to the monitoring and reporting of such taking. Id. Thus, to receive a “small take” authorization, an activity must: (I) be limited to a “specified geographical region,” (ii) result in the incidental take of only “small numbers of marine mammals of a species or population stock,” and (iii) have no more than a “negligible impact” on species and stocks. In addition, in issuing an authorization, the Secretary must: (iv) provide for the monitoring and reporting of such takings, and (v) prescribe methods and means of effecting the “least practicable adverse impact” on species and stock and their habitat. 16 U.S.C. § 1371(a)(5)(A). There is no private right of action under the MMPA. Hawaii County Green Party v. Clinton, 124 F.Supp.2d 1173, 1190 (D.Haw.2000) (citing Didrickson v. U.S. Dep’t of Interior, 982 F.2d 1332, 1338 (9th *1019 Cir.1992)). Citizens challenging actions done under the MMPA must sue under the APA. Id. Therefore, actions challenged under the MMPA are reviewed under the APA “arbitrary and capricious” standard. Plaintiffs argue that the Final Rule 2 violates the MMPA in four ways. First, they contend that the Final Rule is not limited to a specified geographical region. Second, they argue that the Final Rule uses an improper definition of “small numbers.” Third, they claim that the Final Rule uses an improper definition of “harassment.” Finally, plaintiffs argue that the Final Rule will have more than a negligible impact on marine mammals. 1. Specified Geographical Region The Final Rule authorizes incidental taking by Level A and Level B harassment of mysticete whales (whales without teeth), odontocete whales (whales with teeth), and pinnipeds (seals, sea lions, fur seals, and walruses) in 15 different biomes, divided into numerous provinces and subprovinces. 50 C.F.R. § 216.180. Plaintiffs argue that the “provinces” identified by NMFS are gargantuan in scale and far too large to meet the MMPA’s requirement of a “specific geographical region.” 16 U.S.C. § 1371(a)(5)(A). Defendants argue, on the other hand, that there is no requirement in either the statute or the regulations that the specified geographic regions must be small, as long as they are no larger than necessary to accomplish the specified activity. In reviewing the NMFS’ interpretation of the MMPA, the Court must first determine whether Congress has directly spoken to the precise question at issue. Chevron, U.S.A., Inc. v. Natural Resources Defense Council, Inc., 467 U.S. 837, 842, 104 S.Ct. 2778, 81 L.Ed.2d 694 (1984). “If the intent of Congress is clear, that is the end of the matter; for the court, as well as the agency, must give effect to the unambiguously expressed intent of Congress.” Id. at 842-43, 104 S.Ct. 2778. The Court “must reject administrative constructions which are contrary to clear Congressional intent.” Id. at 843 n. 9, 104 S.Ct. 2778. If Congress has not directly addressed the precise question at issue, the court may not simply impose its own construction of the statute, but must determine whether the agency’s answer is based on a permissible construction of the statute. Id. at 843, 104 S.Ct. 2778. “The court need not conclude that the agency construction was the only one it permissibly could have adopted to uphold the construction, or even the reading the court would have reached if the question initially had arisen in a judicial proceeding.” Id. at 843 n. 11, 104 S.Ct. 2778. If Congress has expressly delegated authority to elucidate a specific provision of the statute by regulation, those regulations are given controlling weight unless they are arbitrary, capricious, or manifestly contrary to the statute. Id. at 843-44, 104 S.Ct. 2778. If the legislative delegation to an agency on a particular, question is implicit rather than explicit, a court may not substitute its own construction of a statutory provision for a reasonable interpretation made by the administrator of an agency. Id. at 844, 104 S.Ct. 2778. The only language in the legislative history that addresses the “specified geographic region” requirement provides: It is the intention of the Committee that both the specified activity and the specified region referred to in section *1020 101(a)(5) be narrowly identified so that the anticipated effects will be substantially similar. Thus, for example, it would not be appropriate for the Secretary to specify an activity as broad and diverse as outer continental shelf oil and gas development. Rather, the particular elements of that activity should be separately specified as, for example, seismic exploration or core drilling. Similarly, the specified geographical region should not be larger than is necessary to accomplish the specified activity, and should be drawn in such a way that the effects on marine mammals in the region are substantially the same. Thus, for example, it would be inappropriate to identify the entire Pacific Coast of the North American continent as a specified geographical region, but it may be appropriate to identify particular segments of that coast having similar characteristics, both biological and otherwise, as specified geographical regions. H.R.Rep. No. 97-228 (1981), reprinted in 1981 U.S.C.C.A.N. 1458, 1981 WL 21352 at **1469-70. The Code of Federal Regulations defines “specified geographical region” as “an area within which a specified activity is conducted and which has similar biogeographic characteristics.” 50 C.F.R. § 216.103. Initially, NMFS’ proposed rule divided the world’s oceans into sixteen regions. See 66 Fed.Reg. 15390 (2001) (proposed 50 C.F.R. § 216.180). At that time, NMFS explained that: NMFS believes that the regions described in this proposed rule are in keeping with Congress’ legislative intent in enacting this provision. Although SURTASS LFA sonar requires fairly large geographic regions because of the Navy’s need to deploy the system on a world-wide basis, these areas have been selected so as to retain similar biological characteristics within each region. As a result, NMFS believes that these areas are large enough to accomplish the specified activity without being so large that the effects on marine mammals will not be substantially the same. It should be noted that the regions described in this proposed rule differ from those contained in the Navy’s original application and described in the ANPR. Based on a suggestion made by NMFS in the ANPR, the U.S. Navy revised its original proposal for 10 regions to one that proposes to adopt, with modification, the United Nation Food and Agriculture Organization’s (FAO) division of the world’s oceans into 16 distinct areas 66 Fed.Reg. 15378. NMFS then received objections that this division of the world’s oceans into sixteen regions did not meet the requirement of the MMPA for a “specified geographical region.” 67 Fed.Reg. 46768 (2002). NMFS agreed that the use of those 16 regions violated its own definition of “specified geographical region” as “an area within which a specified activity is conducted and which has certain biogeo-graphic characteristics.” Id. (citing 50 C.F.R. § 216.103.) NMFS agreed that “the 16 areas designed in the proposed rule document were not based on biogeographic characteristics as specified in the definition, but were based on other considerations by the- U.N. Food and Agricultural Organization.” Id. NMFS then adopted its current approach of dividing the oceans into 15 biomes, and 54 provinces within those biomes, as designed by Longhurst (1998). Id. NMFS stated that it believed that this approach met the statutory definition because “a biome is the most likely geographic region to contain the majority of a specific marine mammal stock, especially those that are migratory.” Id. While admittedly, the Longhurst schematic was designed for plankton, it is *1021 the best scientific application available for designating-specified geographic regions because no biogeographic concept has been designed for marine mammals and, in general, the distribution of marine organisms at higher trophic levels resembles the general geographic patterns of primary productivity, with the largest aggregations concentrated in coastal areas and zones of upswelling. (Longhurst, 1998). Id. at 46768-69. “These provinces and biomes effectively delineate the area wherein discrete population units reside thereby allowing NMFS to analyze impacts from SURTASS LFA sonar on a species and/or stock basis.” Id. at 46769. Plaintiffs object that the biomes and provinces identified by NMFS are still far too large. Plaintiffs have provided a map, attached as Exhibit A to their motion, showing the very, large size of some of these provinces. According to plaintiffs, Province 60 is larger than the continental United States and encompasses six million square miles of open ocean. The Court notes that Province 66 covers the entire Pacific coast from roughly Cabo San Lucas at the southern tip of Baja California to the Canadian border. Plaintiffs argue that if “it would be inappropriate to identify the entire Pacific coast of the North American Continent as a specified geographical region,” H.R.Rep. No. 97-228 (1981), reprinted in 1981 U.S.C.C.A.N. 1458, 1981 WL 21352 at **1469-70, then surely an area twice the size of the United States violates the MMPA. Defendants argue that the specified regions need not be small, but they should not be larger than necessary to accomplish the specified activity. (Hollingshead Decl. ¶ 19.) Here, fairly large areas were needed in order for a SURTASS LFA sonar mission to remain within one, or at most two specified geographic regions. (Id.) NMFS felt that it had three choices: (1) deny the requested authorization because Naval operations, could not, and should not, be confined to a single biogeographic area; (2) issue up to 54 sets of regulations so that each set of regulations would be effective in only one area; or (3) issue a single set of regulations and then issue a Letter of Authorization designating which areas a single ship would operate within in any single year. Id. NMFS chose the third alternative as an efficient way to comply with the requirements of the MMPA. Id. Choosing smaller regions would not have worked because NMFS had to make the geographic regions big enough to accomplish the specified activity. Id. ¶ 20. LFA can be heard at very large distances from the vessel; plaintiffs acknowledge that the LFA sonar has a sound pressure level of approximately 140dB more than 400 miles from the vessel. Id. In addition, “LFA sonar in most cases does not transmit equitably throughout the ocean but has a narrow ray path that tends from its origin towards the ocean bottom (below the habitation zone of marine mammals) and reflects back towards the surface and back down again with its second and third reflection at the surface upwards of 100km (54nm) and 150 km (81nm), respectively, from the vessel.” Id. “Therefore, smaller geographic regions would be functionally inappropriate, as sounds could easily transmit across a number of them.” Id. Defendants also point out that smaller geographic units are not necessarily geographically stable; some, for instance, may change during an El Niño period. (Holl-ingshead Decl. ¶ 22.) Plaintiffs also argue that Congress intended that “specified geographic region ... should be drawn in such a way that the effects on marine mammals in the region are substantially the same.” H.R.Rep. No. 97-228 (1981), reprinted in 1981 U.S.C.C.A.N. 1458, 1981 WL 21352 at *1469. The Code of Federal Regulations *1022 similarly defines “specified geographical region” as “an area within which a specified activity is conducted and which has similar biogeographic characteristics.” 50 C.F.R. § 216.103. Plaintiffs interpret this language to require that the abundance and distribution of particular marine mammals must be relatively uniform within any given specific geographical area, but the language does not address distribution of mammalian populations throughout the area. Instead, it requires that the effects on marine mammals be substantially the same throughout the region, and that there be similar biogeographic characteristics throughout the region. Plaintiffs’ expert Rodney M. Fujita, who has a Ph.D. in marine ecology, attests that the Longhurst biomes are not particularly useful for estimating biological impacts on specific populations of marine mammals or other organisms. (Fujita Decl. ¶ 7.) According to Fujita, the provinces identified by NMFS are so large that each one contains many diverse habitats, species assemblages, and levels of productivity. (Id. ¶ 8.) “Even if NMFS’ purpose in creating very large biogeographieal provinces was to ensure that they contain whole stocks of migratory marine mammals, the boundaries are somewhat biologically arbitrary, failing to correspond to population distributions of gray whales, blue whales, and other species.” (Id. ¶ 12.) Fujita particularly notes that: to ensure that marine mammal impacts are substantially the same in a biogeo-graphic area, it is necessary to consider the effects of LFA on smaller areas where marine mammals congregate to feed, breed, and rear their young. Disruption of communication by LFA signals may be especially harmful in such areas. (Id.) Fujita concludes that: The biogeographic biomes and provinces defined by NMFS do not have homogeneous ecological or biogeographieal characteristics. Each province contains many distinctive habitats and biogeo-graphic subdivisions, some of which may be vitally important to marine mammals and others less important. Thus, these biomes and provinces are not consistent with the intent of the MMPA to limit permitted activities to distinct biogeo-graphic areas with similar characteristics. (Id. ¶ 13.) NMFS acknowledges in the Final Rule that the biomes and provinces were not chosen because of their specific relevance to marine mammals. 67 Fed. Reg. at 46768-69. NMFS stated, however, that “it is the best scientific application available for designating specified geographic regions because no biogeographic concept has been designed for marine mammals ....” Id. at 46769; see -also Hollingshead Decl. IT 21. Fujita does not dispute this point. Defendants contend that to the extent scientific information permits, the specified geographical regions have been drawn so that the effects on marine mammals in a specified region are substantially the same. (Hollingshead Decl. ¶24.) Anticipated effects on marine mammals from LFA sonar noise will be based primarily on their hearing anatomy and on water mass characteristics (such as water temperature), which influence the way in which sonar sound propagates. Id. Defendants fail to explain how the enormous provinces set forth in the Final Rule have similar biogeographic characteristics, however. Even water temperature obviously will be dramatically different within provinces that stretch for thousands of miles. Plaintiffs assert that because the Final Rule contains no limitation on how many provinces may be involved in any given deployment of the LFA system, the Final *1023 Rule in fact imposes no specific geographical limitation on LFA’s deployment at all. NMFS has conceded that “no world-wide authorizations have previously been granted.” 66 Fed.Reg. 15378. NMFS. acknowledges in the Final Rule that “[t]he total area that would be available for SURTASS LFA sonar to operate includes about 70-75 percent of the world’s oceans.” 67 Fed. Reg. 46761. NMFS noted, however, that “this in no way equates to affecting 70-75 percent of the world’s ocean area. The current authorization is for only two SUR-TASS LFA sonar vessels — normally one if the Atlantic Ocean/Mediterranean Sea and the other in the Pacific/Indian ocean.” Id. The Navy is “required to notify NMFS annually as to which provinces or subpro-vinces it intends to operate SURTASS LFA sonar system in the upcoming year, and the extent of the take (by harassment) it expects to encounter during the mission.” 67 Fed.Reg. 46769. See also id. at 46788 (50 C.F.R. § 216.187). Thus, each year, the Navy will be limited to operating in certain specified geographical regions. Given the enormous scope of the SUR-FASS LFA system, the geographic areas need to be quite large. It is troublesome that NMFS has chosen large areas that undisputedly do not have homogeneous ecological or biogeographical characteristics. Plaintiffs have established serious issues with respect to whether NMFS violated the MMPA by choosing such undifferentiated geographical areas, particularly in light of the failure to carve out sufficient areas of special biological importance for feeding, breeding, and the like that lie within these large areas and make them less homogenous. See Section II.B.5 below. Plaintiffs have not presented any evidence, however, disputing NMFS’ conclusion that no alternative biogeographical scheme currently exists for marine mammals that can readily be applied here. Thus, plaintiffs have not shown a likelihood of success on their claim that NMFS acted in an arbitrary and capricious manner in choosing the specified geographical regions identified in the Final Rule. Although the NFMS’ choices may be flawed, on this record they do not appear to be so flawed that the Court will likely invalidate them as arbitrary and capricious. At most, plaintiffs have raised a serious question on the merits. 2. Small Numbers Plaintiffs also argue that NMFS is violating the MMPA by using an erroneous definition of “small numbers” that conflicts with the plain language of the MMPA. Under the MMPA, the Secretary can authorize the incidental taking of small numbers of marine mammals if the Secretary finds that the total amount of such taking will have a negligible impact on those species or stock of marine mammals. 16 U.S.C. § 1371(a)(5)(A). The MMPA does not define “small numbers,” but NMFS has promulgated a regulation which provides that “[s]mall numbers means a portion of a marine mammal species or stock whose taking would have a negligible impact on that species or stock.” 50 C.F.R. § 216.103. Plaintiffs contend that this definition dilutes the stringent protections for marine mammals imposed by Congress by improperly merging two separate statutory requirements. Under the MMPA, the Secretary can only authorize the taking of “small numbers” of marine mammals and must ensure that the total amount of the taking has only a “negligible impact” on any species or stock of marine mammals. In other words, plaintiffs argue that even if a particular species has a large population and thus it would require a fairly large number of takes to have a greater than negligible impact on that species, the Secretary is still limited, to authorizing incidental takes of only a small number of such marine mammals. *1024 a. Statute of Limitations Defendants’ first argument is that plaintiffs’ challenge to the regulation is time-barred. Civil actions against the United States are subject to a six-year statute of limitations, except in certain circumstances not relevant here. 28 U.S.C. § 2401(a). The regulation at issue, 50 C.F.R. § 216.103, was promulgated in final form on May 18, 1982, more than twenty years ago. 47 Fed.Reg. 21255 (1982). The Ninth Circuit has held that a challenge to a mere procedural violation in the adoption of a regulation or other agency action must be brought within six years of the decision. Wind River Mining Corp. v. United States, 946 F.2d 710, 715 (9th Cir.1991). Similarly, policy-based facial challenges to the government’s decision must also be brought within six years of the decision. Id. If, however, a challenger contests the substance of an agency decision as exceeding constitutional or statutory authority, the challenger may do so later than six years follow the decision by filing a complaint for review of the adverse application of the decision to the particular challenger.... The government should not be permitted to avoid all challenges to its actions, even if ultra vires, simply because the agency took the action long before anyone discovered the true state of affairs.... [Thus,] a substantive challenge to an agency decision alleging lack of agency authority may be brought within six years of the agency’s application of that decision to the specific challenger. Id. at 715-16 (emphasis in original). Here, plaintiffs challenge the definition of “small numbers” that is contained in 50 C.F.R. § 216.103 on the ground that it is ultra vires because it flatly contradicts the statutory language of the MMPA. Under Wind River, plaintiffs are time-barred from challenging the regulation itself, but are not time-barred from challenging the application of that regulation to them, unless it was first applied to them more than six years ago. Wind River, 946 F.2d at 715-16 (noting in particular the discussion of Oppenheim v. Campbell, 571 F.2d 660 (D.C.Cir.1978)). Defendants argue that at least two of the plaintiffs challenged NMFS’ issuance of a small take authorization to the Navy under the MMPA in 1994, more than six years ago, without raising a challenge to the definition of “small take,” citing NRDC v. United States Dept. of the Navy, 857 F.Supp. 734 (C.D.Cal.1994) (vacated by consent decree). The Court need not consider whether those particular plaintiffs are time-barred from challenging the application of the “small numbers” definition here, because defendants make no argument that the remainder of the plaintiffs also filed similar lawsuits more than six years ago without raising the issue. Thus, even if certain plaintiffs are time-barred from making this argument, the remainder of the plaintiffs are not. Accordingly, the Court finds that plaintiffs’ challenge to the application of the definition of “small numbers” to the Final Rule is not time-barred. b. Whether NMFS Acted Outside the Scope of its Authority The MMPA specifically authorizes the Secretary to prescribe regulations for the taking of marine mammals “as he deems necessary and appropriate to insure that such taking will not be to the disadvantage of those species and population stocks and will be consistent with the purposes and policies set forth in section 1361 of this title.” 16 U.S.C. § 1373. Section 1361 provides, in relevant part: The Congress finds that — ■ (1) certain species and population stocks of marine mammals are, or may *1025 be, in danger of extinction or depletion as a result of man’s activities; (2) such species and population stocks should not be permitted to diminish beyond the point at which they cease to be a significant functioning element in the ecosystem of which they are a part, and, consistent with this major objective, they should not be permitted to diminish below their optimum sustainable population. í-j }¡« ‡ sf: (6) marine mammals have proven themselves to be resources of great international significance, esthetic and recreational as well as economic, and it is the sense of the Congress that they should be protected and encouraged to develop to the greatest extent feasible commensurate with sound policies of resource management and that the primary objective of their management should be to maintain the health and stability of the marine ¿cosystem. Whenever consistent with this primary objective, it should be the goal to obtain an optimum sustainable population keeping in mind the carrying capacity of the habitat. 16 U.S.C. § 1361. Section 1371(a)(5)(A) of the MMPA permits the Secretary to authorize the incidental take of “small numbers of marine mammals of a species or population” if the Secretary finds “that the total of such taking during each five-year (or less) period concerned will have a negligible impact on such species or stock_” 16 U.S.C. § 1371(a)(5)(A). The plain language indicates that “small numbers” is a separate requirement from “negligible impact.” To treat them as identical would appear to render the reference to “small numbers” mere surplusage. Furthermore, Congress made its intent clear when it added this section to the MMPA in 1981. The legislative history demonstrates that Congress intended that “small numbers” and “negligible impact” serve as two separate standards. The- legislative history provides: The taking authorized under these new provisions is the taking of small numbers of marine mammals. The Committee recognizes the imprecision of the term ‘small numbers’, but was unable to offer a more precise formulation because the concept is not capable of being expressed in absolute numerical limits. The Committee intends that these provisions be available for persons whose'taking of marine mammals is infrequent, unavoidable, or accidental. It should also be noted that these new provisions of the Act provide an additional and separate safeguard in that the Secretary must determine that the incidental takings of small numbers of marine mammals have a ‘negligible’ impact upon the species from which such takings occur. This additional test is meant to serve as a separate standard restricting the authority of the' Secretary. The term ‘negligible’ is intended to mean an impact which is able to be disregarded. In this regard, the Committee notes that Webster’s dictionary defines the term ‘negligible’ to mean ‘so small or unimportant or of so little consequence as to warrant little or no attention.’ Unless a particular activity takes only small numbers of marine mammals, and that taking has a negligible impact on the species, the new provisions of sections 101(a)(4) and (5) are not applicable to that activity. H.R.Rep. No. 97-228 (1981), reprinted in 1981 U.S.C.C.A.N. 1458, 1981 WL 21352 at *1469 (emphases added). By defining “small numbers” to mean “a portion of a marine mammal species or stock whose taking would have a negligible impact on that species or stock,” NMFS *1026 has improperly collapsed two standards, which Congress expressly intended to be separate, into a single one. In so doing, NMFS eliminated the possibility that the two standards will serve as separate safeguards restricting the extent of takes. Plaintiffs’ argument that these two standards have been conflated was raised by others at the time the definition was proposed in 1981, and again in the comments to the Final Rule. In response, NMFS stated: NMFS does not believe that the term can be expressed as an absolute number or percentage or be defined in any absolute terms. However, NMFS feels that by defining “small numbers” to mean a portion of a marine mammal species or stock whose taking would have a negligible impact, an upper limit is placed on the term, and the phrase effectively implements the Congressional intent .... 67 Fed.Reg. 46764. By conflating the two terms, however, NMFS has eliminated the ability of the two terms to act, as intended, as separate checks on the Secretary’s authority. For example, where populations of marine mammals are large, the number of mammals taken before there is a greater than negligible impact on the population may also be large. .The statute, however, expressly requires that the number of marine mammals that may be taken incidentally must be small. NMFS’ contention that the “greater than negligible impact” threshold is an upper limit fails to recognize that by defining “small numbers” that way, the regulation permits the Secretary to allow incidental takes that are quite large in number. For example, in the Final Rule, one comment expressed concern that the takings permitted are not “small” and that more than 16 percent of the blue whales in the eastern North Atlantic, more than 10 percent of the beaked whales in the Mediterranean Sea, and more than 12 percent of the elephant seals in the eastern North Pacific will be affected. 67 Fed.Reg. 46764. In response, NMFS did not deny this possibility. Id. Instead, it noted that this was the worst case scenario, not the situation that will most likely take place, due to the Navy’s likely voluntary avoidance of certain areas in certain seasons where marine mammals are likely to be particularly abundant. Id. NMFS noted that 12.4 percent of the elephant seals will be affected only if SURTASS LFA sonar operated in both offshore central California for one mission and offshore Washington on another mission. Id. In fact, under another scenario, NMFS acknowledged that as many as 18.6 percent of elephant seals could be affected. Id. NMFS also stated that a more realistic estimate is that 1 to 2 percent of stocks would be affected during a single 20-day mission. Id. at 46765. Later in the Final Rule, NMFS states: Short-term incidental harassment levels between 1 and 12 percent and below are considered by NMFS to comply with the MMPA as Level B harassment at this level is unlikely to result in significant effects on any species’ or stock’s reproduction or survival. Therefore, in order for incidental takings by SURTASS LFA sonar under this regulation to be negligible, takings by SURTASS LFA sonar operations during the effective time period (1 year) of any LOA issued for. such Navy operations must not exceed 12 percent of any marine mammal stock. 67 Fed.Reg. 46780. NMFS then went on to say that “this 12 percent level should not be interpreted to mean that the Navy will take up to 12 percent of all affected marine mammal stocks.” Id. “In most cases, with carefully planned SURTASS LFA sonar missions (e.g., to avoid certain biogeographic provinces during seasons of *1027 increased marine mammal abundance), the total annual Level B takes are expected to be significantly less than this level.” Id. Nothing in the Final Rule, however, requires the Navy to ensure that takes of marine mammals are at the low end of this wide range of up to 12 percent. In order to obtain a Letter of Authorization, the Navy must provide an estimate of the “percentage of marine mammal species/stocks potentially affected in each specified geographic region for the 12-month period of effectiveness of the Letter of Authorization.” 67 Fed.Reg. 46788 (50 C.F.R. § 216.187(c)(4)). The Final Rule provides that issuance of each Letter of Authorization will be based on a determination that the number of marine mammals taken by the activity will be small, and will have no more than a negligible impact on the species of stock of affected marine mammals. 67 Fed.Reg. 46788 (50 C.F.R. § 216.188(c)). Since these two requirements are improperly defined to mean the same thing, however, there is no independent requirement that the take be small, as required by Congress. ■ The default provision of the MMPA is that “no permit may be issued for the taking of any marine mammal.” 16 U.S.C. § 1371(a) (emphases added). The intent of Congress is that the taking of even a single marine mammal is to be avoided. Incidental takes permitted under section 1371(a)(5)(A) must be small and have a negligible impact on the affected species or stock of marine mammals. 16 U.S.C. § 1371(a)(5)(A). A definition of “small number” that permits the potential taking of as much as 12 percent of the population of a species is plainly against Congress’ intent. Accordingly, plaintiffs are likely to prevail on their contention that NMFS acted outside the scope of its authority in applying the definition of “small numbers” that appears in 50 C.F.R. § 216.103 to the Final Rule. 3. The Final Rule’s Definition of “Harassment” Plaintiffs argue that the Final Rule also uses an illegal definition of “harassment.” The MMPA generally prohibits the taking of marine mammals, with certain statutory exceptions. 16 U.S.C. § 1371(a)(3). The MMPA and the regulations promulgated thereunder define “take” as “to harass, hunt, capture, collect, or kill, or attempt to harass, hunt, capture, collect or kill, any marine mammal.” 50 C.F.R. § 216.3; 16 U.S.C. § 1362(13). The definition of “take” includes any negligent or intentional act which results in disturbing or molesting a marine mammal. 50 C.F.R. § 216.3. The MMPA defines “harassment” as “any act of pursuit, torment or annoyance” that: (I) has the potential to injure a marine mammal or marine mammal stock in the wild; or (ii) has the potential to disturb a marine mammal or marine mammal stock in the wild by causing disruption of behavioral patterns, including but not limited to, migration, breathing, nursing, breeding, feeding, or sheltering. 16 U.S.C. § 1362(18)(A). Harassment as defined in subsection (I) is referred to as Level A harassment. 16 U.S.C. § 1362(18)(B). Harassment as defined in subsection (ii) is referred to as Level B harassment. 16 U.S.C. § 1362(18)(C). Plaintiffs complain that the Final Rule uses a different definition for Level B harassment. The Final Rule provides that “[f]or Level B incidental harassment takings, NMFS will determine whether takings by harassment are occurring based on whether there is a significant behavioral change in a biologically important activity, such as feeding, breeding, migration or sheltering.” 67 Fed.Reg. 46721-22. The *1028 Final Rule also provides that “for small take authorizations (as opposed to- intentional takings), a Level B harassment taking occurs if the marine .mammal has a significant behavioral response in a biologically important behavior or activity.” 67 Fed.Éeg. 46740. 3 Plaintiffs argue that this definition changes the statutory definition in two important respects. First, it requires that there be an actual disruption of behavioral patterns, rather than merely a potential for disruption, as'required by the statute. Second, it requires that the disruption be significant, although the statute contains no such limitation. Plaintiffs also complain that defendants have applied this erroneous definition in a way that excludes1 harassment to individual members of a marine mammal population, in violation of the MMPA’s definition of “harassment” to include potential effects on individuals. a. Potential to Disturb Plaintiffs argue that, whereas the MMPA defines Level B harassment, as any act that has “the. potential to disturb”, a marine mammal “by causing disruption of behavioral patterns,” the Final Rule defines Level B harassment as an action that actually causes a significant biological change in a biologically important behavior or activity. See 67 Fed.Reg. 46721-22, 46740. Thus plaintiffs argue that NMFS has re-written the definition of “harassment” from an activity that has the potential to disturb to an activity that actually causes such a disturbance. One of the comments t.o the Final Rule made this same, argument. 67 Fed.Reg. 46762. In response, NMFS cited the actual text of the MMPA’s definition of Level B harassment, which it acknowledged defined harassment as “potential to disturb,” but nonetheless stated that “NMFS considers a Level B harassment to have occurred if the marine mammal has a significant behavioral response in a biologically important' activity.” 67 Fed.Reg. 46763. The Final Rule provides no explanation as to why NMFS believes it appropriate to ignore Congress’ definition of Level B harassment, which considers an act to be harassing if it “has the potential to disturb a marine mammal or marine mammal stock in the wild by causing disruption of behavioral patterns” (16 U.S.C. § 1362(18)(A)) (emphasis added), even if the disruption does not actually occur. NMFS did consider potential harassment at length in the Final Rule, however. 67 Fed.Reg. 46780. Thus, although NMFS used an erroneous definition of harassment, it does not appear that this erroneous definition caused any particular harm. Accordingly, although plaintiffs may prevail on their claim that NMFS acted arbitrarily and capriciously by. ignoring Congress’ express definition of harassment in the MMPA, they have not shown any irreparable injury from NMFS’ use of the wrong definition in the Final Rule. b. Significance Requirement Plaintiffs also argue that NMFS has inappropriately inserted the requirement that the disruption be significant, when the MMPA’s definition of “harassment” requires only that there be “the potential to disturb a marine mammal or marine mammal stock in the wild by causing disruption of behavioral patterns, including but not limited, to, migration, *1029 breathing, nursing, breeding, feeding, or sheltering.” 16 U.S.C. § 1362(18)(A). One of the comments to the Final Rule made this same argument. 67 Fed.Reg. 46762. NMFS responded: Under an interpretation of “harassment,” as broad as some have suggested the MMPA requires, an incidental taking could be presumed to occur for even a single pinniped lifting or turning its head to look at a passing pedestrian, offshore watercraft, aircraft or dolphins riding a boat’s bow wave. For those takings that are clearly incidental to an otherwise lawful activity, NMFS believes that such a strict interpretation was not intended by Congress, when it amended the MMPA in 1994 and added a definition for harassment. ... [T]o disrupt a behavioral pattern, the activity would need to disrupt an animal’s normal pattern of biological traits or behavior, not just cause a momentary reaction on the part of a marine mammal. Furthermore, if the only reaction to an activity on the part of the marine mammal is within the normal repertoire of actions that are required to carry out the behavioral pattern for that species of marine mammal, NMFS considers the activity not to have caused an incidental disruption of the behavioral pattern, provided the animal’s reaction is not otherwise significant enough to be considered disruptive due to length or severity. For example, if there is a short-term change in breathing rates or a somewhat shortened or lengthened diving sequence that is within the animal’s normal range of breathing patterns and diving cycles but there is not a disruption to the animal’s overall behavioral pattern (i.e., the changes are not biologically significant), then these responses do not rise to a level requiring a small take authorization or, if under a small take authorization, does not constitute an incidental take. 67 Fed.Reg. 46763. At oral argument, plaintiffs argued that the plain language of Congress’ definition of harassment indicates that any potential disruption to behavioral patterns is significant. For example, it appears to be undisputed that LFA can and does actually disrupt the singing of humpback whales. See 67 Fed.Reg. 46732 (“Study results in TR1 indicate that 6 cases of humpback song cessation were considered possible responses to SURTASS LFA sonar transmissions.”) Plaintiffs argued that any disruption of humpback whale singing is a disruption of a behavioral pattern, and thus falls within the MMPA’s definition of harassment. This argument was also raised in response to the Final Rule. One commenter argued that because the humpback whale singing is related to mating behavior, any change is likely to be significant to the limited gene pool of the endangered humpback whale. 67 Fed.Reg. 46734. The NMFS responded that many of the whales continued to sing and interact during the LFA transmission, and “[m]ost of the whales that did respond resumed activities normal for the breeding area within an hour.” 67 Fed.Reg. 46735. Because the cessation of singing was not widespread, the NMFS concluded that it is unlikely that SURTASS LFA would have a significant or widespread impact to this biologically important behavior. Id. While NMFS may be correct that overall mating will not suffer appreciably, this analysis ignores the fact that Congress considered disruptions to the behavioral patterns of single animals to be important, as described further in the following section. Nonetheless, the Court concurs with defendants that NMFS’ requirement that there be a potential for significant disrup *1030 tion is a reasonable attempt to distinguish between mere responses bymarine mammals to the specified activity, and the type of disruptions to behavioral patterns that Congress was expressly concerned about, as indicated in the MMPA’s definition of “harassment.” 4 Plaintiffs have raised a serious question, however, whether NMFS, by focusing on the significance of the disruption to the entire stock, rather than the significance to, individual marine mammals, and by focusing on actual rather than potential disruption, has impermissi-bly narrowed the definition of harassment under the MMPA. c. Impact on Individual Mammals Plaintiffs’ final argument with respect to NMFS’ definition of harassment is that NMFS improperly excludes effects on single members of a marine mammal population in violation of the MMPA’s definition of “harassment.” The MMPA’s definition of “harassment” expressly applies to acts that affect “a marine mammal or marine mammal stock in the wild.” 16 U.S.C. § 1362(18)(A). Plaintiffs complain that at one point in the Final Rule, NMFS states: Examples of significantly disrupted behavior would be where pinnipeds flee a haulout beach or rookery en masse due to a disturbance, or animals either leave' an area of habitation for a period of time, or diverge significantly from their migratory path to avoid either an acoustic or a visual interference. Non-significant behavioral responses would be when only a few pinnipeds leave the haulout or mill-about, but many pin-nipeds are alert to the disruption; or when marine mammals make minor course corrections that are not discerna-ble either to observers or directional plotting, and which requires statistical manipulation in order to determine that a course correction has taken place. 67 Fed.Reg. 46763. It is not an unreasonable reading of this paragraph to conclude that NMFS does not consider significant disruptions to the behavioral patterns of a single marine mammal to constitute harassment under the MMPA. If NMFS defines' disruptions to behavioral patterns as harassment only if they affect an entire stock of marine mammals, then that violates the MMPA. The MMPA is expressly concerned with harassment to “a marine mammal” as well as harassment of a “marine mammal stock.” 16 U.S.C. § 1362(18)(A). ' Defendants state, however, that NMFS does not contend that a single marine mammal cannot be harassed: *1031 It is true that Level B harassment can occur with an individual marine mammal. The example was only meant to illustrate that, in the context of the non-reaction of the majority of the sea lions present, the one or few sea lions that leave the haulout beach would not be deemed to have had a disturbance to their behavioral patterns, even if they did so in response to the subject stimulus. This behavior would be considered within the normal range of the animal’s or species’ behavioral pattern. However, each situation must be analyzed on a case by case basis. (Hollingshead Decl. ¶ 17.) Defendants put forward a somewhat different argument in their brief. There, unlike in the Hollings-head declaration, they do not contend that there is no harassment if a single animal flees a beach or rookery in response to the specified activity. Instead, defendants argue that: Rather, the point of the sea lion example is that when a few sea lions leave a beach or rookery, this behavior is entirely consistent with the species’ natural behavior pattern. A single sea lion, or a few, may retreat to the water to cool off on a hot day, to scratch an itch, to avoid jostling of other sea lions on the beach, or in response to a sea gull alarm call. However, an en masse flight of all the sea lions from a haulout beach or rookery is not consistent with the species’ repertoire of behavioral responses and would be considered a disruption of a behavioral pattern for that species. (Opposition brief at 16.) This argument appears to go to how one would determine whether the marine mammals were in fact responding to the specified activity. Defendants appear to be arguing that if only a few marine mammals leave the beach or rookery, those animals are unlikely to be responding to the specified activity. Both the defendants and the Hollings-head declaration, however, appear to be viewing disruptions to behavioral patterns on a species level, rather than on an individual level. In expressing concern about harassment to “a marine mammal,” Congress was concerned about harassment to individual animals. Thus, if an individual marine mammal in a rookery flees that rookery in response to the specified activity, and does not return, or fails to return in the usual period of time, that animal has been harassed within the meaning of the MMPA, even if other animals in the group did not leave in response to the specified activity. It may well be, however, that when a marine biologist sees a single marine mammal leave the beach, while others of its type remain undisturbed, it is very difficult, if not impossible, to determine whether the animal is leaving of its own accord, or whether it is a particularly sensitive animal that is fleeing in response to the specified activity. The definition of harassment, however, encompasses potential harassment to single individuals, even if other individuals of that species in the same location do not appear to be harassed by the same activity. In fact, by focusing on potential harassment, the statute appears to consider all of the animals in a population to be harassed if there is the potential for the act to disrupt the behavioral patterns of the most sensitive individual in the group. Defendants acknowledge that single animals can be harassed under the MMPA. Indeed, the Final Rule was issued on the basis of projections of potential effects to marine mammals. 67 Fed.Reg. 46780. Thus, plaintiffs have raised a serious question as to whether defendants may have measured harassment in a way that excludes harassment of individual marine mammals, but it is not clear whether, or to what extent, this affected the final determination. *1032 4. Negligible Impact The MMPA permits NMFS to issue a small take permit orily if it can first find that the taking authorized by the permit will have a “negligible impact” on 'marine mammals. 16 USC §§ 1371(a)(5)(A), (D). Plaintiffs argue that the drafters of the MM'PA’s “small take” provision 'cited Webster’s dictionary to indicate-what they intended by “negligible impact”: an impact that is “so small or unimportant or of so little consequence as to warrant little or no attention.” H. Rept. No. 228, 97, Cong., 1st Sess. 19 (1981). Defendants counter that plaintiffs rely on an outdated -Congressional definition of “negligible impact,” because in 1986 Congress changed the definition to “an impact resulting from an activity that cannot reasonably be expected to, and is not reasonably likely to, adversely affect the species or stock through effects on annual rates of recruitment, or survival.”. 54 Fed.Reg. 40340; 50 C.F.R. § 216.103. ' In 1989, the U.S. Fish and Wildlife Service explained that, ' while sympathetic with the concerns expressed by the ' Commenters, [the Service] believes that the clear congressional intent behind the 1986 Amendments was to alter the standard for determining negligible impact.... To capture the intent of the amendment, the Service has adopted, substantially- without change, the definition of negligible impact set out in the Senate’s ‘Section-by-Section Analysis.’ ■ 54‘ Fed.Reg. 40340 (citing 132 Cong. Rec. 16305 (Oct. 15, 1986)). The Senate explained in their Section-by-Section Analysis that Section 411: amends the [MMPA] and makes- conforming amendments to the [ESA] to allow incidental taking of depleted as well as non-depleted specifies of .marine mammals under certain conditions.... The term ‘negligible impact’ as. applied to populations means an "impact that cannot reasonably be expected to, and is not reasonably likely to affect adversely the overall population through effects on annual rates of recruitment or survival. 132 Cong. Rec. 16305 (Oct. 15, 1986.) Thus, it seems that the Senate has amended other ■ sections of the MMPA and, in doing so, arguably clarified the definition of negligible impact without formally amending it.- In any case, plaintiffs have not shown that they are likely to establish that the agency’s definition is arbitrary and capricious or contrary to law. Plaintiffs argue that NMFS arbitrarily adopted 180 dB as the minimum exposure level necessary to cause injury and downplayed the severity or extent of likely impacts. As discussed above, plaintiffs are not likely to prevail on this record on their claim that NMFS’ adoption of 180 dB as the minimum exposure level necessary to cause injury was arbitrary or capricious. To the extent that plaintiffs’ qualified experts disagree with defendants’ qualified experts in an area of scientific uncertainty, where there is room for legitimate disagreement) the Court must defer to the Agency’s reliance- on its experts’ reasonable opinions. Marsh, 490 U.S. at 378, 109 S.Ct. 1851. Defendants contend that the following factors support" a finding of negligible impact: (1) findings from -the SRP demonstrating little impact On marine mammals from exposure to sound- levels up to 155 dB; (2) the small number of SURTASS LFA sonar systems that would be operating world-wide; (3) the relatively low-.-duty-cycle, the short run of missions, and the location of operations off-shore (where marine mammal abundance is lower than in coastal waters); (4) the fact- that typically, although not always, the sonar’s path is deflected below a water depth inhabited by marine mammals for approximately 75 *1033 percent of the distance between the source and the first convergence zone (CZ) as well as between the first CZ and the second CZ, depending on ocean conditions; (5) the fact that during CZ propagation, the narrow width of the ray path and the 1,000-fold decrease in sound intensity immediately outside the ray path further limit potential for exposure; (6) the fact that the vessel must be underway (continuously moving) while transmitting, limiting exposure of marine mammals to those few minutes when the LFA sound is moving through the portion of the water column where the animal is swimming; and (7) the implementation of highly effective mitigation measures that make it unlikely that a marine mammal will enter the 180 dB sound field during sonar transmission without being detected and the signal given to shut down the system. Yet plaintiffs have raised some serious issues about the efficiency of the mitigation measures adopted and whether harassment of as much as 12 percent of marine mammal stocks will in fact be negligible. While “negligible impact” is a more qualitative and relative concept than the issue of “small numbers,” and involves harm to reproduction and survival, not merely harassment, some of the same concerns are implicated. As discussed above, under the Final Rule, the Navy retains discretion to operate in biologically rich zones of the ocean during sensitive time periods such as mating and migration seasons. While defendants may exercise their discretion to avoid these areas and time periods, nothing in the Final Rule requires them to do so. Furthermore, while the mitigation measures adopted will help reduce harm to marine mammals and are very commendable as far as they go, the evidence shows, as explained below, that the planned mitigation is not likely to be as effective as defendants contend, especially for certain marine mammals. At this stage, plaintiffs have at least raised a serious issue on the merits regarding “negligible impact.” 5. Mitigation and Monitoring In issuing a small take permit, the MMPA requires the Secretary to provide for the monitoring and reporting of such takings, and prescribe methods and means of effecting the “least practicable adverse impact” on species and stock in their habitat. 16 U.S.C. § 1371(a)(5)(A). Plaintiffs argue that the Final Rule fails to set forth sufficient measures for mitigation, monitoring, and reporting to ensure a negligible impact on species, and fails to adequately assess the actual impact on marine mammals. The purpose of the monitoring requirement is to assure that the take allowed under the permit is, in fact, small, and also has only a negligible impact on affected species. H. Rpt. No. 228, 97th Cong., 1st Sess. 18-20 (1981). The LFA mitigation zone is the area within the 180 dB isopleth 5 of the SUR-TASS LFA sonar source sound field. (Na-varo Dec. Ex. 3 at 6.) The distance from the SURTASS LFA sonar source, where signals are broadcast at 215 dB (Navaro Dec. Ex. 14 at 2-3), to the 180 dB isopleth is approximately 1 km, which the Navy initially proposed-as the mitigation zone. 67 Fed.Reg. 48147. NMFS required an ádditional 1-kñometer. buffer zone, within which sound levels are expected to drop to about 173 dB. (Navaro Dec. Ex. 3 at 6.) NMFS required certain mitigation measures within the resulting 2 km (about 1.2 mile) zone to try to avoid exposure of sea animals to levels of 173-180 dB. At 2 km, the Sound Pressure Level (“SPL”) from *1034 the SURTASS LFA sonar will be approximately 173 dB'. 67 Fed.Reg. 46781. Plaintiffs suggest that all mammals within 40 nautical miles of the source will be exposed to levels of sound of 165 dB or more. Because the sound waves do not radiate uniformly outward from the source, but usually travel in a zig-zag path, however, only patches of sea within the 40 mile radius will have such high levels of sound. Defendants’ expert, Dr. Ellison, explained: “[T]here are two regions in which sound levels of 165 dB or higher can be found. The first region can be described as a contiguous path from the source and extends in a narrow beam a few hundred feet in width that extends down and out to a point about 8 [nautical miles] from the source and at a depth of 3000m. The second region consists of several small isolated patches at a range near 20 [nautical miles]. To assert that LFA can create sound levels as high as 165 dB at ranges out to 20 or 40 nmi, may on the face of it be correct, however it belies the obvious discontinuous nature of the sound field in the region.” (Ellison Dec. at 7:3-10.) 6 Plaintiffs do not rebut this explanation. Plaintiffs argue that, under the terms of the LFA permit, virtually all mitigation and monitoring will be limited to the 2 km perimeter around the LFA source despite the fact that this zone consists of - only about five percent of the area in which LFA sonar propagates. Plaintiffs contend that therefore the mitigation measures will not apply to 95 percent of the marine animals exposed. This calculation, however, ignores the fact that LFA sonar operates in particular paths and not in the entire area within a 40 nautical mile radius. Moreover, many of the exposures will be at a much lower decibel level. Defendants’ Risk Continuum presumes a 95 percent risk of a biologically significant response such as avoidance or harm at a received level of 180 dB for a single broadcast of the LFA signal, declining to a 70-80 percent risk at 173 dB and a 50 percent risk at 165 dB. (Fristrup Dec. at ¶¶ 16-18.) In the Final Rule, NMFS establishes a tripartite program to limit harm to marine mammals: a high frequency active sonar system (“HF/M3”) to monitor for marine mammals within 2 kilometers of the LFA source and suspend operation during the period that they are detected in that zone; visual observation for marine mammals and endangered sea turtles from the deck of the source ship; and passive acoustic monitoring. 67 Fed.Reg. 46787. [The Final Rule] adopts, with modification, the Navy’s proposal to use visual, passive acoustic, and active acoustic monitoring of the area surrounding the SURTASS LFA sonar array to prevent the incidental injury of marine mammals that might enter the 180 dB SURTASS LFA mitigation _ zone.... If a marine mammal (or ESA-listed sea turtle) is detected within the 180-dB SUTRASS LFA sonar mitigation zone, SURTASS LFA sonar transmissions will be immediately delayed or suspended. Transmissions may commence/resume 15 minutes after the marine mammal/sea turtle has left the area of the 180 dB sound field or there is no further detection of the animal within the 180 dB sound field.... NMFS has concluded that the 180-dB safety zone needs to be augmented to ensure to the greatest extent practicable that marine mammals are not subject to potential injury. In that regard, as an added safety measure, NMFS has established an interim “buffer zone” extending an additional 1 km (0.54 nm) beyond the 180-dB LFA mitigation zone. Therefore, as soon as a marine mammal (or ESA-listed sea tur- *1035 tie) is detected by the HF/M3 sonar, the SURTASS LFA sonar will either be turned off or not turned on. This is a feasible mitigation measure since recent testing of the HF/M3 sonar indicates effective levels of detection up to 2 km (1.1 nm). At 2 km (1.1 nm), the SPL from the SURTASS LFA sonar will be approximately 173 dB. SURTASS LFA sonar operators would be required to estimate SPLs prior to and during each operation to provide the information necessary to modify the operation, including delay or suspension of transmissions, in order not to exceed the mitigation sound field criteria. 67 Fed.Reg. 46781. These measures are laudable as far as they go. Realistically, however, they will not detect all marine mammals and endangered species within the 2 km zone. Visual monitoring is not very effective even in the best of conditions, particularly for smaller animals who spend long periods under water, much less in rough seas or in the dark. Passive sonar also misses quieter animals. While the active sonar is fairly effective in detecting large whales, it is much less effective in detecting smaller animals, such as fast moving dolphins and certain sea turtles. For example, in a test of bottlenose dolphins, only 55 percent were detected. (Stafford Ex. 1 at 2-20 to 2-21.) Defense witness Joseph Johnson acknowledged at the hearing that smaller animals such as sea turtles are even more likely to escape detection. Furthermore, none of these measures are designed to detect marine mammals beyond 2 km (1.2 miles) from the LFA source. Defendants claim that by collecting data, including ship position, marine mammals observations, and times of transmission, NMFS and the Navy will be able to compile information about the effects of SURTASS LFA beyond the 2 km safety zone. (Defendants’ opposition at 25:13— 15.) Yet, by definition, animals that go undetected, even injured ones, will not be counted. Further, plaintiffs point out that this after-the-fact information, resulting in a report on the effects of current operations five years from now, is too little too late. The parties dispute whether monitoring of animals that may be exposed beyond’the 2 km zone is feasible. NMFS acknowledges that “the commenter[s][are] correct that behavioral modifications can be expected at lower SPLs [below 180 dB].” According to NMFS, however: the proposed monitoring (visual, passive acoustic and active acoustic), is not likely to be as effective at the greater distances where these impacts are likely to occur. As a result, NMFS prefers to require the Navy to concentrate monitoring in an area wherein marine mammals are more likely to incur an injury, than at distances wherein the incidental taking will be limited to short-term behavioral modifications. Since monitoring is less likely to be effective at distances much greater than the 180-dB isopleth, and because the Navy has requested a small take authorization for the incidental harassment of marine mammals, NMFS has preliminarily determined that the establishment of a safety zone at the 180 dB isopleth is the distance that is most practicable for reducing potential impacts on marine mammals to the lowest level. 66 Fed. Reg. 15,380-81. Plaintiffs propose requiring pre-operation surveys of local areas of operations. (Stafford Dec. Ex. 29 at 26-27.) NMFS reasonably responded to this comment in the Final Rule, however, that “[p]re-operation surveys are not practical since the SURTASS LFA sonar vessel normally operates independent from the fleet and too^ distant from shore to make aerial surveys- practical.” 67 Fed. *1036 Reg. 46750. On the record so far, plaintiffs have not shown that monitoring of a larger zone would be feasible. Given the practical inability of current technology to detect and avoid exposure to marine mammals outside the 2 km zone, avoiding intense sonar in key habitat for feeding, mating, breeding and migration becomes critical to effective mitigation. In other words, excluding these biologically rich zones from LFA operations is the only way to ensure that takes are limited to small numbers with a negligible impact on survival and reproduction. While generally the choice of mitigation measures may be within defendants’ discretion, as set forth above, NMFS may not authorize more than small numbers of takes with a negligible impact. Defendants have established certain geographic exclusion zones, but the parties dispute whether they are adequate. Specifically, defendants established geographic exclusion zones that prevent LFA sonar from exposing marine mammals to signals at 180 dB or above in:' coastal waters within a 12' nautical mile zone from shore (as well as 145 dB within known human dive sites); the Arctic and' Antarctic; and three “Offshore Biologically Important Areas (OBIAs).” These OBIAs are the 200 meter isobath of the North American Eastern Coast, year round; the Costa Rico Dome, year round; and the Atlantic Convergence Zone, October through March. (Navarro Dec. Ex. 14 at 2-12 — 2—13.) Yet the Navy is still free to deploy LFA sonar -in some 70 to 75 percent of the world’s oceans. The parties agree that coastal waters support especially rich concentrations of marine mammals. Studies show-that sea life generally flourishes from the shore to the continental shelf, which roughly parallels the shore line but generally extends out well over 12 nautical miles. ■ LFA sonar is effective in detecting submarines in most (but no.t all) conditions at distances of 40 to 200 miles. Defendants plainly do not act arbitrarily in refusing to bar operating closer to shore where conditions make that necessary for effective detection. Yet it appears that such coastal areas are a minority. Plaintiffs’ argument that most offshore areas could receive greater protection than 12 nautical miles appears persuasive. Coastal waters are not the only areas of rich concentrations of marine mammals, as defendants recognized in creating the three OBIAs. Marine mammals (and other endangered species) migrate and feed in areas far from shore. Yet, defendants postponed adding other OBIAs, despite their own experts’ recognition that other areas probably should be designated. For example, both Plaintiffs expert Fujita and defendant’s expert Hollingshead agree that the Oyashio/Kuroshio area off Kamchatka is an example of an area that qualifies for future nomination as an OBIA. (Fujita Dec. at ¶ 20; Hollingshead Dec. at ¶ 22.) Defendants also acknowledge that boundaries between provinces may in the future merit special attention as areas to avoid or as possible candidates for OBIA. (Defendants’ Opp. at p.22 n.8.) Indeed, defendants’ argument that the Navy is likely to voluntarily avoid areas and times of high concentration of marine mammals, thereby reducing “takes” to 1-2 percent rather than 12 percent, underscores their ability to delineate those areas and avoid them. 67 Fed.Reg. 46780. Nonetheless, the Navy is currently free to deploy LFA in these sensitive areas. Instead of timely protecting these areas, NMFS instead postponed consideration of additional exclusion zones to sometime in the future. ■ Moreover, it set up a process by which members of the public would largely bear the burden of proving that additional exclusion zones are warranted. NMFS noted that: *1037 Commenters recommended additional mitigation measures, such as geographical restrictions above and beyond those proposed by the Navy, including an extension of the coastal exclusion zone beyond the limits of the U.S. territorial sea and the territorial seas of other countries, expansion of the Southern Ocean whale sanctuary, and the addition of biologically significant offshore areas; and a timely, transparent, and publicly accountable procedure for supplementing the initial list of restrictions. 66 Fed.Reg. 15,381. NMFS nonetheless refused to consider adding more OBIAs during the rulemaking process: NMFS does not consider it necessary to expand the list of OBIAs prior to its making the required determinations under section 101(a)(5)(A) of the MMPA. While some of the areas mentioned in the comment would qualify for nomination as an OBIA, a delay in the rule-making process to implement additional OBIAs is not warranted, especially considering the high level of effectiveness of the tripartite monitoring system.... NMFS considers a public review and comment period a necessary step in establishing new OBIAs. Once this final rule is implemented, NMFS will accept petitions for OBIAs in accordance with 50 CFR 216.191 promulgated in this final rule. However, as stated in the preamble to the proposed rule, petitions will not affect authorizations for taking marine mammals within those areas until an OBIA is final (if that is the determination). It should be recognized that NMFS may also nominate areas as OBIAs, but does not believe that it should be the sole proponent for nominating areas and that was the reason for allowing it to be a public process following standard rulemaking practice. 67 Fed.Reg. 46747-48. See also 50 C.F.R. § 216.191 (emphasis added). Plaintiffs are likely to prevail in showing that defendants acted arbitrarily in postponing the designation of additional OBIA merely to avoid delay, despite the fact that some of the proposed areas appear to be good candidates. In sum, the various measures adopted by defendants are commendable and will help reduce the risk to marine mammals. Further, plaintiffs have not shown that monitoring could be effective at distances beyond the zone chosen by defendants of 2 km from the LFA source. Yet these measures leave open the possibility that up to 12 percent of marine mammals will be harassed or worse. While defendants have designated some exclusion zones, they have arbitrarily postponed designating additional exclusion zones that defendants’ own experts acknowledge are likely appropriate. Yet both sides agree that avoiding certain areas of the ocean during seasons when marine mammals are abundant in those areas is important to reducing their potential exposure. Indeed, defendants’ argument that the Navy intends to avoid doing exercises in areas and seasons where marine mammals are concentrated to reduce the number of animals taken below 12 percent shows the feasibility of imposing greater restrictions without compromising the sonar’s effectiveness. C. National Environmental Policy Act Plaintiffs also argue that defendants failed to comply with NEPA. The Court must determine whether the Environmental Impact Statement (“EIS”) was “arbitrary and capricious, an abuse of discretion, or otherwise not in accordance with law.” Okanogan Highlands Alliance v. Williams, 236 F.3d 468, 471 (9th Cir.2000); City of Carmel-By-The-Sea v. U.S. Dep’t of Transportation, 123 F.3d 1142, 1150 (9th Cir.1997); Western Radio Servs. Co., Inc. v. Espy, 79 F.3d 896, 900 (9th Cir.1996). *1038 Courts apply a “rule of reason” standard, which assesses “whether an EIS contains a reasonably thorough discussion of the significant aspects of the probable environmental consequences.” Churchill County v. Norton, 276 F.3d 1060, 1071 (9th Cir.2001) (quoting Trout Unlimited v. Morton, 509 F.2d 1276, 1283 (9th Cir.1974)); see also City of Carmel, 123 F.3d at 1150-51 (“the National Environmental Policy Act requires a ‘reasonably thorough’ discussion of. the environmental consequences in question, not unanimity of opinion, expert or otherwise.”) in making this determination, a court must make a “ ‘pragmatic judgment whether the EIS’s form, content, and preparation foster both informed decision-making and informed public participation.’ ” Churchill County, 276 F.3d at 1071; City of Carmel, 123 F.3d at 1150-51. “‘Once satisfied that a proposing agency has taken a “hard look” át a decision’s ■ environmental consequences, [our] review is at an end.’ City of Carmel, 123 F.3d at 1151 (quoting Idaho Conservation League v. Mumma, 956 F.2d 1508, 1519 (9th Cir.1992)). Plaintiffs assert that the EIS 'is arbitrary and capricious in four respects: (1) failure to consider reasonable alternatives; (2) failure to consider reasonably foreseeable environmental impacts; (3) failure to supplement the EIS in light of the Bahamas stranding; and (4) reliance on a “white paper” not subject to public comment. 1. Reasonable Alternatives Analysis An EIS must discuss “reasonable alternatives” to the proposed "action. 42 U.S.C.-§ 4332(2)(C)(iii); City of Carmel, 123 F.3d at 1155. In this section, agencies shall “[r]igorously- explore and'objectively evaluate all reasonable alternatives, and for alternatives which were eliminated from detailed study, briefly discuss the reasons for their having been eliminated.” 40 C.F.R. § 1502.14(a). The “rule of reason” guides .the, choice of alternatives and the extent to which the EIS must discuss each alternative. City of Carmel, 123 F.3d at 1155 (citing Citizens Against Burlington v. Busey, 938 F.2d 190, 195 (D.C.Cir.1991)). “The [EIS] need not consider an infinite range of alternatives, only reasonable and feasible ones.” City of Carmel, 123 F.3d at 1155; see also Laguna Greenbelt, Inc. v. U.S. Dep’t of Transportation, 42 F.3d 517, 524 (9th Cir.1994); Seattle Audubon Society v. Moseley, 80 F.3d 1401, 1404 (9th Cir.1996); 40 C.F.R. § 1502.14(a)-(c). The rangé of alternatives that is deemed reasonable depends upon “the underlying purpose and need to which the agency is responding in proposing the alternatives including the proposed action.” 40 C.F.R. § 1502.13; see also City of Carmel, 123 F.3d at 1155 (“The stated goal of a project necessarily dictates the range of reasonable alternatives and_ an agency cannot define its objectives in unreasonably narrow terms”). A court should uphold “an agency’s definition of objectives so long as the objectives that the agency, chooses are reasonable, and we uphold its discussion of alternatives so long as the alternatives are reasonable and the agency discusses them in reasonable detail.” Citizens Against Burlington, 938 F.2d at 195. ÑRDC contends that the EIS failed to properly consider all reasonable alternatives that would allow deployment of LFA in a manner consistent with NEPA, in effect only considering one preordained alternative — the one . chosen. The Navy’s EIS set forth three alternatives: (1) a No Action alternative; (2) full deployment with no mitigation or monitoring; (3) and the Navy’s preferred alternative. Plaintiffs contend, however, that the first alternative was rejected out of hand because it plainly did not meet the goal, while the second alternative, was a “straw *1039 man.” Plaintiffs further contend that the Navy failed to consider other reasonable alternatives that would meet the Navy’s stated purpose, while limiting the damage inflicted by LFA sonar on the marine ecosystem. The EIS rejected the No Action alternative as not meeting project objectives: Under the No Action Alternative, which would foreclose employment of LFA sonar technology, the U.S. Navy’s ability to locate and defend against enemy submarines would be greatly impaired. The lack of long-range submarine detection capability would make it possible for potentially hostile submarines to clandestinely place themselves into position to threaten U.S. Fleet units and land-based targets. Without this long-range surveillance capability, the reaction times to submarines would be greatly reduced and the effectiveness of close-in, tactical systems to neutralize threats would be seriously, if not fatally, compromised. (Stafford Dec. Ex. 1 at 2.3.1.) The No Action alternative is plainly inconsistent with the project’s overarching purpose and need “to improve U.S. detection of quieter and hard-to-find submarines at long range,” (Stafford Dec. Ex. 1 at 2.3.1) and thus the Navy “did not act unreasonably in rejecting the no-action alternative on the ground that it would not meet the purpose and need of the proposed project.” Friends of Southeast’s Future v. Morrison, 153 F.3d 1059, 1067 (9th Cir.1998). Moreover, “[t]he fact that the description of the no-action alternative is shorter than those of the other proposals does not necessarily indicate that the no-action alternative was not considered seriously.” Oregon Natural Resources Council v. Lyng, 882 F.2d 1417, 1423 n. 5 (9th Cir.1989). Plaintiffs argue persuasively that the second alternative, full deployment with no mitigation or monitoring, is per se illegal under the MMPA and is therefore a phantom option. Defendants assert that alternative two “would [only] be inconsistent with the MMPA and the ESA if the action would result in the taking of marine mammals and/or species listed as threatened or endangered and taking authority had not been obtained.” (Navaro Dec. Ex. 14 at 2-23, Ch. 2.3.3.) However, the “small take” exception to the MMPA provides that the Secretary shall allow such a take if the Secretary prescribes regulations setting forth requirements pertaining to the monitoring and reporting of such taking. 16 U.S.C. § 1371(a)(5)(A). While defendants are correct that the NMFS may issue authorizations under the MMPA even where there is no practical mitigation, that is not the case here. The chosen alternative includes mitigation measures. The real question, however, is not the absolute number of alternatives considered, but rather whether the agency considered all alternatives reasonably related to the purposes of the project, however few. Laguna Greenbelt, Inc. v. U.S. Dept. of Transportation, 42 F.3d 517, 524 (9th Cir.1994). None of the cases addressing this issue are really on point, however. Rather, this case appears to fall somewhere in between those cases that have upheld the considerations of only a few alternatives as adequate and those that have found the consideration of only one or two alternatives inadequate. In Laguna Greenbelt, the plaintiffs argued that the U.S. Department of Transportation violated NEPA by considering only three alternatives for a tollroad project: the proposed route; the no action alternative; and a second alternative route that differed from the preferred alternative only in method of operation and manner of connection to an interstate. Id. at 523-24. The EIS also included six alternatives that were briefly considered but *1040 rejected. Id. at 524. Laguna Greenbelt held that the EIS adequately addressed alternatives: “The EIS -discusses in detail all thé alternatives that were feasible and briefly discusses the reasons others were eliminated. This is all NEPA requires — : there is no minimum number of alternatives that must be discussed.” Id. Plaintiffs correctly point out that, in Laguna Greenbelt, the agency examined two bona fide alternatives, in addition to the No Action alternative, whereas here the Navy only considered one action alternative. At the same time, the case teaches that the number of alternatives alone is not determinative. 7 Id. Plaintiffs rely on California v. Block, 690 F.2d 753, 768 (9th Cir.1982). There, the court rejected an analysis that only considered those alternatives leading to the agency’s desired end result. In Block, however, the agency offered no explanation at all for its consideration of only those alternatives that' allocated more acreage to nonwilderness than to wilderness, and failed to realistically assess the relevant resource trade-offs. Id. Plaintiffs also rely on Natural Resources Defense Council v. United States Dept. of the Navy, 857 F.Supp. 734, 739-40 (C.D.Cal.1994) (the “Shipshock ” case), where the court rejected the Navy’s decision to proceed with “ship shock” testing of its vessels in the California Channel Islands because of an inadequate alternatives analysis. 8 There, NMFS contended that the' MMPA did not require it to consider the possibility of alternative sites beyond those prepared by the Navy. Id. at 737. The court rejected this contention, stating that “[t]he- Final Rule’s promulgation was premised on an impermissible determination that alternatives outside the [Outer Sea Test Range] did not have to be considered.” Id. at 740. Thus, Shipshock appears to be a more egregious case of failing to consider obvious alternatives. Here, on the one hand, plaintiffs argue that the Navy should have considered the alternative of training only in areas of low marine mammal abundance and biological productivity. NMFS explained persuasively in the Final Rule, however, that “this mitigation measure is not practicable since the Navy needs to operate in areas with different water characteristics.... This would not be available to the Navy if it were limited to biologically unproductive areas.” 67 Fed.Reg. 46750. NMFS also reasonably explained why plaintiffs’ suggestion to reduce source level, duty cycle, and annual transmission hours would not fulfill the Navy’s long-range submarine detection objectives, based on extensive design work and testing. (Navaro Dec. Ex. Í4 at 10-51, 10-55); 67 Fed.Reg. 46750. NMFS’ rejection of these proposed alternatives for failure to fulfill the goal of effective use of LFA technology was not arbitrary or capricious. On the other hand, as discussed above, defendants failed to consider an action alternative that restricted LFA operations in areas that even its experts agree should probably be designated as OBIAs or which *1041 extended the coastal exclusion zone beyond 12 nm to the biologically rich area of the continental shelf in the many parts of the world where LFA sonar could operate effectively at such an offshore distance. Defendants’ explanation of why it did not consider such an alternative boils down to wanting to retain broad discretion to operate in all coastal waters beyond 12 nautical miles, even though the Navy does not intend to actually do so, and not wanting to take the time to designate more OBIAs: The geographic restriction is for the SURTASS LFA sound field of 180 dB, not the location of the vessel. While the U.S. Navy plans to operate mainly in waters significantly greater than 12 nm (22 km) offshore, it should not be precluded from operating in waters near 12 nm (22 km) from shore, provided the SPL does not exceed 180-dB at a distance of 12 nm (22 km) from any coastline. For this reason, NMFS has not implemented the recommended restriction on SURTASS LFA sonar operations. However, because SURTASS LFA sonar transmissions will be restricted to SPLs below 145 dB within known dive sites, the LFA vessel will remain at distances greater than 12 nm (22 km) from shore in most situations .... While some of the areas mentioned in the comment would qualify for nomination as an OBIA, a delay in the rulemaking process to implement additional OBIAs is not warranted. 67 Fed.Reg. 46746-48; see also Navaro Dec. Ex. 14 at 10-149—10-152. In conclusion, plaintiffs have shown a likelihood of establishing that defendants acted arbitrarily in only considering in effect one alternative — the chosen one — and not considering a feasible alternative excluding more, but not all, areas of high marine mammal concentration, while preserving the ability to train in a variety of conditions. 2. Consideration of Reasonably Foreseeable Environmental Impacts When an agency is evaluating reasonably foreseeable significant adverse effects on the human environment in an [EIS] and there is incomplete or unavailable information, the agency shall always make clear that such information is lacking.... If the information relevant to reasonably foreseeable significant adverse impacts cannot be obtained because the overall costs of obtaining it are exorbitant or the means to obtain it are not known, the agency shall include within the [EIS]: (1) A statement that such information is incomplete or unavailable; (2) a statement of the relevance of the incomplete or unavailable information to evaluating reasonably foreseeable significant adverse impacts on the human environment; (3) a summary of existing credible scientific evidence which is relevant to evaluating the reasonably foreseeable significant adverse impacts on the human environment, and (4) the agency’s evaluation of such impacts based upon theoretical approaches or research methods generally accepted in the scientific community. For the purposes of this section, ‘reasonably foreseeable’ includes impacts which have catastrophic consequences, even if their probability of occurrence is low, provided that the analysis of the impacts is supported by credible scientific evidence, is not based on pure conjecture, and is within a rule of reason. 40 C.F.R. § 1502.22. According to plaintiffs, the EIS reveals that the Navy knows next to nothing about the hearing sensitivity, acoustic ecology, and behavioral responses of many of the species that may be affected by LFA’s use. Plaintiffs contend that the Navy has not considered all “reasonably foreseeable significant adverse impacts,” and failed to use “theoretical ap *1042 proaches or research methods generally accepted in the scientific community'.” Defendants counter that their scientific methodology represents a careful and conservative approach reached by • a consensus of qualified scientists. “The rationale for using representative species to study the potential effects of LF sound on marine animals emerged' from an extensive review in several workshops by a broad group of interested parties: academic scientists, federal regulators, and representatives of environmental and animal welfare groups.” (Navaro Dec. Ex. 14-at 1-20); 67 Fed.Reg. 46732. “Since it was impossible to study all species and all settings, the goal of the SRP was to select the most sensitive studies for the most vulnerable animals. The precautionary use of this approach was to apply these results even to species known tó be much less sensitive, even species that could barely hear the signals.” (Tyack Dec. at ¶ 13.) Plaintiffs have not shown that they are likely to prevail on this issue. While more research is desirable, the agency did not act arbitrarily and capriciously. It developed new scientific information using generally accepted scientific methods, while acknowledging the remaining gaps ■ in scientific knowledge. Thus, the Court cannot say that defendants acted unreasonably. “[A] reviewing court must remember that [when] the Commission is making predictions, within its area of special expertise, at the frontiers of science ... [the] reviewing court must be at its most deferential.” Baltimore Gas and Electric Co. v. NRDC, 462 U.S. 87, 103, 103 .S.Ct. 2246, 76 L.Ed.2d 437 (1983). 3. The Navy’s Refusal to Supplement the EIS A Supplemental EIS (“SEIS”) is required when there “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)(l)(ii). However, an agency need not supplement an EIS every time new information comes to light after the EIS is finalized. To require otherwise would render agency decisionmaking intractable, always awaiting updated information only to find the new information outdated by the time a decision is made. On the other hand ... NEPA does require that agencies take a “hard look” at the environmental .effects of their planned action, even after a.proposal has received initial approval. .Application of the “rule of reason” thus turns on the value of the new information to the still pending de-cisionmaking process. In this respect the decision whether to prepare a supplemental EIS is similar to the decision whether to prepare an EIS in the first instance: If there remains “major federal action” to occur, and if the new information is sufficient to show that the remaining action will “affect the quantity of the human environment” in a significant manner or to a significant extent not already considered, a supplemental EIS must be prepared. Marsh v. Oregon Natural Resources Council, 490 U.S. 360, 373-74, 109 S.Ct. 1851, 104 L.Ed.2d 377 (1989) (quoting 42 U.S.C. § 4332(2)(C)). Once again, the standard of review of an agency determination that an SEIS is not required is whether the decision was arbitrary or capricious. Id. As noted above, in March 2000, whales from at least three different species beached themselves in the Bahamas. In December 2001, the Navy published a Joint Interim Report with NMFS on the stranding. Plaintiffs argue that the new information included in the Joint Interim Report should have triggered an SEIS. *1043 In Warm Springs Dam Task Force v. Gribble, 621 F.2d 1017, 1025 (9th Cir.1980), the new information threatened to undermine the major assumption underlying the current EIS. The court held that, while the new information was “not so definitive as to compel initiation, of the formal supplementation process, the [study] raised sufficient environmental concerns to require the [defendant] to take another hard look at the issues.” Id. Here, defendants argue that they did take a “hard look.” They addressed the Bahamas mass stranding briefly in the January 2001 EIS and in the July 16, 2002 Final Rule, and more extensively in the ROD. The Final Rule explains that [s]onars differ in their operating characteristics, and marine mammal species differ greatly in the sounds to which they are susceptible. This is often overlooked by the public. The scientific investigation regarding the Bahamian beaked whale stranding found that the tactical mid-range frequency sonars that were in use aboard'U.S. Navy and allied ships during the March 15-16, 2000, Bahamas sonar exercise were the most plausible source of acoustic or impulse trauma to six beaked whales. 67 Fed.Reg. 46725. 9 Plaintiffs contend that the ROD does not provide the “hard look” that NEPA requires. The ROD never mentions the Navy’s and NMFS’ own Joint Interim Report on the Bahamas strandings, which found that the strandings may have been caused in part by sound pressure levels significantly less than 180 dB. (Stafford Dec. Ex. 10 at v.) However, the ROD gives three reasons that the Navy did not find it necessary to provide an SEIS regarding the Bahamas stranding: First ... SURTASS LFA sonar was not' involved in the Bahamas stranding, and it has been confirmed that SURTASS LFA sonar has never been associated with any strandings. Second, the LFS SRP made systematic evaluations of the animals most likely to be potentially affected by low frequency sound. While beaked whales, the primary species that-stranded in the Bahamas, may be sensitive to frequencies above that employed by SURTASS LFA sonar, the available evidence does not show that they are more sensitive to low frequency sounds than the species selected as subjects for the LFS SRP (baleen whales). Finally, the interim report on the Bahamas stranding concluded that the cause of this stranding was the confluence of the Navy mid-range frequency sonar and contributory factors including the presence of a strong surface duct, unusual underwater bathymetry, constricted channel with limited egress, intensive active use of multiple sonar units over an extended period of time, and the presence of beaked whálés that appear *1044 to be sensitive to the frequencies produced by these sonars. 67 Fed. Reg. 48152. As discussed above, plaintiffs’ experts raise troubling questions about whether unknown mechanisms involved in the mid-frequency sonar contributing to the Bahamas strandings may also apply to low-frequency sonar. However, “[b]ecause analysis of these factual issues requires a high level of technical expertise, [the court] must defer to the informed discretion of the responsible federal agencies.” Laguna Greenbelt, Inc. v. U.S. Dept. of Transportation, 42 F.3d at 530. In La-guna Greenbelt, the court held that the decision not to prepare an SEIS was not arbitrary or capricious because “the FHA relied on substantial technical expertise possessed by two federal agencies charged with responsibility for the respective sectors of the affected environment.” Id. at 529-30. Similarly, here, the Navy and NMFS relied on qualified experts in determining that the information in the Joint Interim Report and the evidence from the Bahamas strandings did not rise to the level of significance that requires an SEIS. While reasonable minds may differ on the significance of the Bahamas stranding to the use of LFA sonar, plaintiffs have not shown that they are likely to prevail in establishing that defendants’ decision not to prepare an SEIS was arbitrary or capricious. 4. Reliance on Unpublished White Paper not Subject to Public Comment After receiving numerous comments, in the wake of the Bahamas stranding, that LFA sonar would cause non-auditory or tissue damage, NMFS asked the Navy to prepare a paper analyzing the existing data. The Navy employed Dr. William Edison and Dr. Edward Cudahy to write a white paper that defendants relied on but that was not disclosed and subjected to public comment. Plaintiffs argue that by relying on an undisclosed scientific document prepared for the Navy, NMFS violated the APA. The APA [Administrative Procedure Act] requires an agency to publish a general notice of rule making in the Federal Register, “give interested persons an opportunity to participate in the rule making through submission of written data, views, or arguments with or without opportunity for oral presentation,” and “after consideration of the relevant matter presented,” “incorporate in the rules adopted a concise general statement of their basis and purpose.” [However,] “[a]n agency can add material to the administrative record after the close of a public comment period when that material is a response to the comments .... An agency may use ‘supplementary data,’ unavailable during the notice and comment period, that ‘expands on and confirms’ information contained in the proposed rulemaking and addresses ‘alleged deficiencies’ in the pre-existing data, so long as no prejudice is shown.” Idaho Farm Bureau Federation v. Babbitt, 58 F.3d 1392, 1401-02 (9th Cir.1995) (quoting 5 U.S.C. § 553(b)-(e); Solite Corp. v. EPA, 952 F.2d 473, 484 (D.C.Cir.1991)). In Babbitt, the court concluded that the agency violated this provision, because “the USGS study did not merely supplement or confirm existing data.... [Rather, the] study provided unique information that was not duplicated in other reports.” Id. at 1402-03. Moreover, the study was critical to the agency’s decision, yet its accuracy was questionable. Id. Here, NMFS did rely heavily on the white paper for the principle that “the potential for in vivo damage to marine mammals from exposure to underwater LF sound will not occur at a level less than *1045 180 to 190 dB.” 67 Fed.Reg. 46727 (citing Cudahy and Ellison, 2002). Defendants point out correctly, however, that all of the literature analyzed in the white paper, except for one reference, was either from previously published peer-reviewed scientific papers or from papers submitted by commenters to the proposed rule. The only new study reviewed in the white paper expands upon the testing of mice and guinea pigs that occurred under the auspices of the Navy during the SRP. While the data on mice'did'not change, the new data on guinea pigs- indicated onset of injury above 180 dB, compared to the prior threshold of 174 dB. According to plaintiffs, this new finding was critical to the Navy’s conclusion that each of the studies based on underwater measurements of terrestrial mammals supports a damage threshold on the order of 180 to . 190 dB. However, Dr. Cudahy explains that he and Dr. Ellison did not rely heavily on the new data for two reasons. “First, these data simply confirmed and expanded upon -previous data.... Second .. .. the number of guinea pigs tested [less than 10] was low.” (Cudahy Dec. at ¶ 15.) The white paper itself explains: For non-shallow-water lung resonance cases, the best estimate is from the mouse data (over 400 animals tested). Those data indicate that lung damage would start at about 181 dB. Lower, thresholds were originally estimated for a small number of guinea pigs (approximately 174 dB). Additional mice and guinea pigs have been tested since that report. These data did not significantly change the resonant frequencies or damage risk thresholds for mice cited in that report but did result in small changes to the damage risk threshold for guinea pigs. These data indicate that the inception of damage threshold is 184 dB for the mice and above 180 dB for the guinea pigs. The total number of mice tested for lung damage is over 400, but the number of guinea pigs is less than 10, thus we have much more confidence in the values based on the mice data. (Navaro Dec. Ex. 10 at 4-5.) The white paper thus "appears to have merely supplemented existing data. Therefore, plaintiffs have not shown a likelihood of prevailing on their argument that failure to publish the white paper for comment was arbitrary and capricious. D. Endangered Species Act The ESA prohibits any person from “taking” species listed as endangered and empowers the United States Fish and Wildlife Service (“FWS”) and NMFS to promulgate regulations prohibiting the taking of any species listed as threatened. 16 U.S.C. §§ 1533, 1538(a)(1)(A)-(B), (G). Actions challenged under the ESA are also reviewed under the APA “arbitrary and capricious” standard. Village of False Pass v. Clark, 733 F.2d 605, 609-10 (9th Cir.1984). Section 7 of the ESA requires each federal agency, through consultation with NMFS or FWS? to: insure that any action authorized, funded, or carried out by [the] agency ... is not likely .to jeopardize the continued existence of, any endangered species or threatened species or. result in the destruction or adverse modification of habitat of such speeies which is determined by the Secretary [of the Interior or of Commerce] ... to be critical. 16 U-S.C. § 1536(a)(2). To- ensure compliance with this requirement, the ESA sets out a three-step consultation process in which the agency with jurisdiction over .the species (here, NMFS) evaluates the nature and extent of jeopar-; dy to the species. Under this process, the Navy, as the agency proposing to take an action, first inquires of-NMFS whether any threatened or endangered species are present in the area of the proposed action. *1046 Thomas v. Peterson, 753 F.2d 754, 763 (9th Cir.1985); 16 U.S.C. § 1536(c)(1). Next, if the answer is affirmative, the Navy then prepares a biological assessment to determine whether the species is likely to be affected by the action. Thomas, 753 F.2d at 763; 16 U.S.C. § 1536(c)(1). Third, if NMFS determines, based on the biological assessment, that the action the Navy proposes to take is likely to affect a threatened or endangered species, the two agencies must engage in formal consultation. Alternatively, if NMFS determines that the action the Navy proposed to take would not likely adversely affect a protected species, NMFS could attempt informal consultation. Formal consultation results in a biological opinion from NMFS which states a conclusion as to whether the proposed action is likely to jeopardize the continued existence of a listed species or result in destruction or adverse modification of critical habitat. 50 C.F.R. § 402.14. If the biological opinion concludes that the proposed action would jeopardize the species or adversely affect critical habitat, then the proposed action may not go forward unless NMFS can suggest an alternative to avoid the adverse impact. Id.; 16 U.S.C. § 1536(b)(3)(A). If the biological opinion concludes that the proposed action will not violate the Act, NMFS may still require mitigation measures. Thomas, 753 F.2d at 763; 16 U.S.C. § 1536(b)(4)(ii)-(iii). Here, given the scope of the proposed activity and the number of species potentially impacted, the Navy and NMFS agreed upon the need for formal consultation on both the deployment of SURTASS LFA and NMFS’ issuance of the Final Rule authorizing take. Accordingly, NMFS issued a biological opinion, dated May 30, 2002, that addressed the effects on threatened and endangered species globally over the next five years. (Navarro Decl. Ex 3.) NMFS subsequently issued a supplemental biological opinion, dated August 16, 2002, which addressed the effects within specified regions of the Pacific Ocean where the Navy proposed to operate during the next year. (Navaro Dec. Ex. 4.) Plaintiffs argue that the biological opinions violate the ESA in three respects: (1) the first biological opinion concludes that the LFA system will not cause the destruction or adverse modification of critical habitat based upon an illegal regulation; (2) contrary to NMFS’ own regulations, the first biological opinion fails to estimate the amount or extent of endangered and threatened species that will be taken by the deployment of the LFA, while the supplemental biological opinion does so only for some of the affected species; and (3) the first biological opinion’s conclusion that the use of the LFA system will not jeopardize the continued existence of any endangered or threatened species, or result in the destruction or adverse modification of designated critical habitat, is arbitrary and capricious and contrary to the best available scientific evidence. 1. Regulatory Definition of Adverse Modification The ESA does not expressly define either “jeopardy” or “adverse modification.” By regulation, defendants define the phrase “to jeopardize the continued existence of’ any species as “to engage in an action that reasonably would be expected, directly or indirectly, to reduce appreciably the likelihood of both the survival and the recovery of a listed species.” 50 C.F.R. § 402.02 (emphasis added). The regulation defines “destruction or adverse modification” of critical habitat as “a direct or indirect alteration that appreciably diminishes the value of critical habitat for both the survival and recovery of a listed *1047 species.” 50 C.F.R. § 402.02 (emphasis added). The Fifth' Circuit has carefully examined the statutes and determined that this regulatory definition of adverse modification violates the ESA by “requiring consultation only where an action affects the value of critical habitat to both the recovery and survival of a species.’’ Sierra Club v. U.S. Fish and Wildlife Service, 245 F.3d 434, 442 (5th Cir.2001) (emphasis in original). Instead, the ESA requires consultation where an action affects species’ recovery through alteration of critical habitat. Id. at 441. • -. The Sierra Club court Reasoned that conflating the two separate requirements of the effect on survival and the effect on recovery diluted the protections that Congress set forth in'-the statute, because a decline in critical habitat that might not threaten the survival of a species could nonetheless threaten the species’ ability to recover to healthy population levels. Id. While this out-of-circuit decision is not binding, this Court finds its reasoning cogent and persuasive. Other courts, have also found it persuasive. See Natural Resources Defense Council v. U.S. Dept. of the Interior, CV-99-5246 SVW (Ctx) (C.D.Cal. July 12, 2002)(reproduced in Stafford Dec. Ex. 31) (following Sierra Club); New Mexico Cattle Growers Ass’n v. U.S. Fish and Wildlife Service, 248 F.3d 1277, 1283 (2001) (citing Sierra Club favorably). ' '■ Plaintiffs argue that NMFS necessarily relied on this illegal regulatory definition of adverse modification in concluding in its supplemental biological opinion that the LEA would not result in the adverse modification of any critical habitat. Defendants respond that, because NMFS concluded that the deployment of the SURTASS LFA was unlikely to adversely affect critical habitat for the Stellar sea lion, Hawaiian monk seal, and northern right whale, it did not reach a decision on whether the action would adversely modify critical habitat. . ■ • ■ Defendants rely on an inapplicable provision of the ESA regarding informal consultation: “If during informal consultation it is determined by the agency, with the written concurrence of the Service, that the action is.,not likely to affect listed species or critical habitat, the consultation process is .terminated, and no further action is necessary.” 50 C.F.R. § 402.13. 10 Here, however, defendants engaged in formal consultation, resulting in the biological opinion, which must state .a, conclusion “as to whether the action, taken together with cumulative effects, is likely to jeopardize the continued existence of listed species or result in the destruction or adverse modification of, critical habitat.” 50- C.F.R. § 402.14(g)(4). Indeed, the cover letter for the biological opinion states “[t]he [biological opinion] concludes that the SUR-TASS LFA sonar system is not likely to jeopardize the' continued existence- of threatened or endangered species under NMFS’ jurisdiction or destroy or adversely modify critical habitat that has .been designated .for those species.” (Navaro De.c. Ex. 3.). Moreover, the supplemental biological opinion states that “[t]his assessment also determines if it is reasonable to expect the Navy’s SURTASS.LFA sonar will result in the destruction -or. adverse modification of critical habitat.” (Stafford Dec. Ex. 26 at 25.) Accordingly, defendants' failed to examine whether LFA 'sonar is likely to adversely affect the recovery of these species, even if it would not affect their survival. Therefore, plaintiffs are likely to prevail in showing that defendants 'ar *1048 bitrarily and capriciously terminated consideration of critical habitat prematurely. 2. Incidental Take Statements The regulations promulgated under the ESA require that when NMFS concludes that an action and the resultant incidental take of listed species will not violate section 7(a)(2) of the ESA, and, in the case of marine mammals, where the taking is authorized pursuant to section 101(a)(5) of the MMPA, NMFS “will provide with the biological opinion a statement concerning incidental take that (I) specifies the impact, i.e., the amount or extent, or such incidental taking on the species[.]” 50 C.F.R. § 402.14(0(1). If the amount or extent of taking specified in this incidental take statement (“ITS”) is exceeded, reinitiation of formal consultation is required. 50 C.F.R. § 402.16. Plaintiffs argue that defendants are in violation of the ESA because the May 80, 2002 biological opinion and the August 16, 2002 supplemental biological opinion do not include an ITS that specifies the amount or extent of the incidental take of endangered and threatened species. The letter accompanying the May 30, 2002 biological opinion states: Please note that this biological opinion does not include an incidental take statement because the programmatic nature of the proposed actions would not allow NMFS to estimate the amount or extent of threatened or endangered species that would be “taken” incidental to the employment of SURTASS LFA sonar. For example, the species that could be taken incidental to the SURTASS LFA sonar system will vary from ocean to ocean, the particular region of an ocean, and timing. Consequently, NMFS will amend this biological opinion to include incidental take statements when NMFS’ Marine Mammal Conservation Division prepares letters of authorization for the SURTASS LFA sonar systems that identify more specific employment schedules. The information in those letters of authorization would allow us to estimate the amount or extent of incidental take for particular threatened or endangered species. (Navarro Deck, Ex. 3 at 1.) The “Incidental Take Statement” included in the May 30, 2002 biological opinion similarly states: Because of the geographic scope and scale of this programmatic biological opinion NMFS cannot estimate the amount or extent of incidental take of threatened or endangered species by the proposed employment of SURTASS LFA sonar. Consequently, NMFS will identify the amount or extent of take that would be associated with the employment of SURTASS LFA when we review the annual letters of authorization for compliance with section 7 of the Endangered Species Act of 1973, as amended. (Id. at 148.) The biological opinion also recognizes, however, that reinitiation of formal consultation is required if the amount or extent of incidental take is exceeded. (Id. at 150.) Plaintiffs complain that reinitiation can never be triggered because the biological opinion contains no attempt to estimate the amount or extent of the take. The Ninth Circuit has held that “[i]n general, Incidental Take Statements set forth a ‘trigger’ that, when reached, results in an unacceptable level of incidental take, invalidating the safe harbor provision, and requiring the parties to reinitiate consultation.” Arizona Cattle Growers’ Ass’n v. United States Fish and Wildlife, 273 F.3d 1229, 1249 (9th Cir.2001). “Ideally, this ‘trigger’ should be a specific number.” Id. A numerical limit is not required where infeasible, however, and the Ninth Circuit has upheld ITS’s that used a combination *1049 of numbers and estimates. Id. Congress itself, in the legislative history, only required that “[w]here possible, the impact should be specified in terms of a numerical limitation.” Id. at 1250 (quoting H.R.Rep. No. 97-567 at 27 (1982), reprinted in 1982 U.S.C.C.A.N. at 2827). In the absence of a specific numerical value, however, the defendant must establish that no such numerical value could be practically obtained. Id. The terms of an Incidental Take Statement do not operate in a vacuum. To the contrary, they are integral parts of the statutory scheme, determining, among other things, when consultation must be reinitiated. Id at 1251.' , The May 30, 2002 biological opinion makes no attempt at all to estimate the incidental take, of threatened or endangered species, and seeks to defer estimating the incidental take until it reviews the annual letters of authorization. Defendants concede in their opposition brief that the May 30, 2002 biological opinion does not include an ITS. Defendants point to no statutory authority for this omission. The biological opinion itself is required to contain an ITS that “[s]pecifies the impact, i.e., the amount or extent, of such incidental taking on the species.” 50 C.F.R. § 402.14(i)(l). Defendants argue that “[without more specific information about the times and locations of planned operations it was not appropriate for NMFS to issue an ITS.” (Opposition brief at 43.) Defendants must establish, however, that no numerical value could be practically obtained. Arizona Cattle Growers’ Ass’n, 273 F.3d 1229, 1250. Defendants have not presented any evidence that it would have been impractical to have included an ITS specifying the amount of the extent of the incidental take. Moreover, even where numerical values are improper, the ITS still must contain some surrogate for defining the amount or extent of incidental take. Id. As the May 30, 2002 biological opinion does not contain an ITS, it violates the ESA. On August 16, 2002, defendants issued a supplemental biological opinion addressing the proposed letter of authorization for the period August 16, 2002 through August 15, 2003. (Navarro Decl, Ex. 4.) During that period of time, the Navy plans to operate one ship in provinces 52, 53, 56, 60 and 64, which cover large areas of the Pacific Ocean. (Id. at 3, 5.) NMFS determined that the Navy’s action in thes.e provinces may affect the Steller sea lion, Hawaiian monk seal, seven types of endangered whales, six types of endangered or threatened sea turtles, and numerous populations of endangered or threatened chinook salmon, chum salmon, coho salmon, and steelhead. (Id. at 5-6.) . The supplemental biological opinion does provide an ITS, which arguably might have cured the deficiency of the original biological opinion, except that it is incomplete. The supplemental biological opinion estimates the numbers of Steller,sea lions,, blue whales, fin whales, humpback whales, right whales, sei whales, and sperm whales, but does not attempt to estimate the numbers of Hawaiian monk seals; Pacific gray whales, sea turtles or salmon that might be taken. (Id. at 41^12.) For these species, the supplemental biological opinion provides that: The extent of take will be limited to the LFA mitigation zone and the additional buffer zone required by the letter of authorization. Take will have been exceeded if (I) individual members of these species are harmed, injured, or killed Within this area as result of exposure to LFA sonar transmissions, (ii) individual members of these species exhibit biologically-significant responses to LFA sonar transmissions within the buffer zone, *1050 or (iii) individual members of these species enter the LFA mitigation zone during an LFA sonar transmission and exhibit biologically-significant responses following a transmission. (Id. at 42.) The supplemental biological opinion also provides: NMFS does not have an estimate of the number of threatened or endangered species that would be “taken” (through harassment) by the proposed action. However, the numbers are expected to consist of a small number of individual animals. (Id. at 46.) 11 The only explanation provided for failing to provide estimates for these species is that the Navy “did not conduct Acoustic Integration Model simulations for these species[.]” Id. at 42. Plaintiffs complain that this statement is unsupported by any evidence that the Navy was asked to estimate the take of these species, but could not. See Arizona Cattle Growers’ Ass’n, 273 F.3d at 1250 (“In the absence of a specific numerical value, ... [defendants] must establish that no such numerical value could be practically obtained.”) Plaintiffs are correct. Defendants have provided no evidence that it was impractical to obtain estimates of the incidental take for Hawaiian monk seals, Pacific gray whales, sea turtles or salmon, which collectively represent some 20 endangered species. Thus, they have not established that no such numerical value could be .practically obtained. Defendants instead skip to the next step, and argue that they provided an adequate surrogate for estimating the incidental take for those species. “[T]he use of ecological conditions as a surrogate for defining the amount or extent of incidental take is reasonable so long as these conditions are finked to the take of the protected species.” Id. at 1250. Arizona Cattle Growers Ass’n explained: Take can be expressed also as a change in habitat characteristics affecting the species (e.g., for an aquatic species, changes in water temperature or chemistry, flows, or sediment loads) where data or information exists which finks such changes to the take of the fisted species.... [I]f a sufficient causal fink is demonstrated (i.e., the number of burrows affected or a quantitative loss of cover, food, water quality, or symbionts), then this can establish a measure of the impact on the species of its habitat and provide the yardstick for reinitiation. Id. (quoting Final ESA Section 7 Consultation Handbook, March 1998 at 4^17 to 4-48). The court nonetheless rejected the following ITS language as inadequate: The service concludes that incidental take of loach minnow from the proposed action will be considered to be exceeded if any of the following conditions are met: [Condition 1] Ecological conditions do not improve under the proposed livestock management. Improving conditions can be defined through improvements in watershed, soil condition, trend and condition of rangelands (e.g. vegetative fitter, plant vigor, and native species diversity), riparian conditions (e.g., vegetative and geomorphologic: bank, terrace, and flood plain conditions) (e.g., *1051 channel profile, embeddedness, water temperature, and base flow) within the natural capabilities of the landscape in all pastures on the allotment within the Blue River watershed. Id. at 1249. The Ninth Circuit held that this language did not sufficiently discuss the causal connection between Condition 1 and the taking of the species at issue: “Based upon the lack of an articulated, rational connection between Condition 1 and.-the taking of species, as well as the vagueness of the condition itself, we hold that its implementation was arbitrary and capricious.” Id. at 1250. Here, unlike in the Ninth Circuit case just cited, defendants do not attempt to link changes in environmental conditions to the taking of endangered species. Instead of estimating the incidental take, defendants essentially state that a taking of any individual of that' species within the LFA mitigation zone and buffer zone will be considered to be too much. (Navaro Dec. Ex. 4 at 42.) The purpose of the ITS is to provide a “‘trigger,’ that when reached, results in an unacceptable level of incidental take, invalidating the safe harbor provision, and requiring' the parties to reinitiate consultation.” Arizona Cattle Growers Ass’n, 273 F.3d at 1249. At first blush, it may appear that by setting that trigger at one animal, defendants satisfied the purpose of the ITS, even without attempting to provide an actual estimate of the likely amount of the incidental take. On closer inspection, however; defendants’ limitation of the trigger to an animal taken within the 2 km mitigation and buffer zone defeats the purpose of the ITS and lacks a rational causal connection because it excludes most of the takes that will occur. Just the other side of the 2 km border, the received level of LFA sonar is 173 dB, at which ■ about 70-75 percent of exposed animals would be taken. And at forty miles away, the received level is still as high as 165 dB, at which 50 percent of exposed animals would be taken. Yet the causal connection to these takes is ignored under defendants’ ITS for these, species. Instead, defendants’ approach focuses solely on where the take occurs, not whether it was caused by LFA sonar. For example, if Pacific gray whales beach outside the LFA mitigation zone and buffer zone, the issue should be “were they injured by SURTASS LFA?” Under this standard, the issue will become “where were they'injured?” For the reasons set forth above, plaintiffs are likely to prevail on their argument that defendants violated the ESA by failing to include an ITS in the May 30, 2002 biological opinion. Plaintiffs are also likely to prevail on their argument that defendants violated the ESA by failing to include an ITS in the August 16, 2002 supplement biological opinion for Hawaiian monk seals, Pacific gray whales, sea turtles or salmon, because defendants have not established that no such numerical values could be practically obtained, and in any case chose an arbitrary proxy. III. INJUNCTIVE RELIEF The legal standard governing issuance .of a preliminary injunction requires the moving party to show either: (1) a combination of probable success on the merits and the possibility of irreparable injury or (2) that serious questions are raised and the balance of hardships .tips in its favor. These two formulations represent two points on a sliding scale in which the required degree of irreparable harm increases as the probability of success decreases. Roe v. Anderson, 134 F.3d 1400, 1402 (9th Cir.1998). Thus, to determine whether in-junctive relief is appropriate in the context of environmental litigation, courts apply a *1052 “traditional balance of the harms analysis.” National Parks & Conservation Ass’n v. Babbitt, 241 F.3d 722, 737 (9th Cir.2001); Forest Conservation Council v. United States Forest Serv., 66 F.3d 1489, 1496 (9th Cir.1995); Amoco Prod. Co. v. Village of Cambell, 480 U.S. 531, 541, 107 S.Ct. 1396, 94 L.Ed.2d 542 (1987). The bases for injunctive relief are irreparable injury and inadequacy of legal remedies. See Amoco, 480 U.S. at 542, 107 S.Ct. 1396; Alaska Wilderness Recreation & Tourism Ass’n v. Morrison, 67 F.3d 723, 732 (9th Cir.1995). “In each case, a court must balance the competing claims of injury and must consider the effect on each party of the granting or withholding of the requested relief ....” Amoco, 480 U.S. at 542, 107 S.Ct. 1396. “Environmental injury, by its nature, can seldom be adequately remedied by money damages and is often permanent or at least of long duration, i.e., irreparable.” Id. at 545, 107 S.Ct. 1396; Sierra Club v. United States Forest Serv., 843 F.2d 1190, 1195 (9th Cir.1988). “If such injury is sufficiently likely, therefore, the balance of the harms will usually favor the issuance of an injunction to protect the environment.” Amoco, 480 U.S. at 545, 107 S.Ct. 1396; Sierra Club, 843 F.2d at 1195; Singleton, 75 F.Supp.2d at 1141. In the NEPA context, irreparable injury flows from a failure to evaluate the environmental impact of a major federal action. See Thomas v. Peterson, 753 F.2d 754, 764 (9th Cir.1985); American Motorcyclist Ass’n v. Watt, 714 F.2d 962, 966 (9th Cir.1983) (“The premise for relaxing the equitable tests in NEPA cases is that irreparable damage may be implied from the failure of responsible authorities to evaluate thoroughly the environmental impact of a proposed federal action.”) The harm at stake when the government fails to comply with the NEPA procedures “is a harm to the environment, but the harms consists of the added risk to the environment that takes place when governmental decisionmakers make up their minds without having before them an analysis (with prior public comment) of the likely effects of their decision upon the environment.” Sierra Club v. Marsh, 872 F.2d 497, 500 (1st Cir.1989) (emphasis in original); National Parks & Conservation Ass’n, 241 F.3d at 723 n. 8 (finding issuance of a preliminary injunction for a NEPA violation was justified under Marsh). Nonetheless, in “ ‘unusual circumstances’ an injunction may be withheld, or, more likely, limited in scope.” National Parks & Conservation Ass’n, 241 F.3d at 737 n. 18. In determining whether to issue an injunction, courts also consider the public interest. See Amoco, 480 U.S. at 542, 107 S.Ct. 1396; Singleton, 75 F.Supp.2d at 1141. “[W]here an injunction is asked which will adversely affect a public interest for whose impairment, even temporarily, an injunction bond cannot compensate, the court may in the public interest withhold relief until a final determination of the rights of the parties, though postponement may be burdensome to the plaintiff.” Weinberger v. Romero-Barcelo, 456 U.S. 305, 312-13, 102 S.Ct. 1798, 72 L.Ed.2d 91 (1982) (quoting Yakus v. United States, 321 U.S. 414, 440, 64 S.Ct. 660, 88 L.Ed. 834 (1944)). In Wein-berger, the Supreme Court upheld the denial of a preliminary injunction because the merely technical violations at issue were not harming the environment, whereas granting injunctive relief would seriously harm not only the Navy, but also the general welfare. Id. at 310, 102 S.Ct. 1798. The Supreme Court distinguished the strong presumption in favor of injunc-tive relief under ESA, however, from this traditional test, recognizing that under the ESA “Congress ... had foreclosed the traditional discretion possessed by an eq *1053 uity court and had required the District Court to [issue an injunction],.. to preserve ... an endangered species.”1 Amoco, 480 U.S. at 543 n. 9, 107 S.Ct. 1396 (citing Romero-Barcelo’s distinction of TVA v. Hill, 437 U.S. 153, 98 S.Ct. 2279, 57 L.Ed.2d 117 (1978)). As the Ninth Circuit explained: The traditional test for preliminary injunctions ... is not the test for injunctions under the [ESA]. In cases involving the ESA, Congress removed from the courts their traditional equitable discretion in injunction proceedings of balancing the parties’ competing interests.. The language, history, and structure of the ESA demonstrates Congress’ determination that the balance of hardships and the public interest tips heavily , in favor of protected species. Nevertheless, these cases do. not -stand for the .proposition that-courts no longer must look at the likelihood of future harm before deciding whether to grant an injunction under, the ESA. Federal courts are not obligated to .grant an injunction for every violation of the law. .The plaintiff must make a showing that a violation of the ESA is at least likely in the future. National Wildlife Federation v. Burlington Northern Railroad, 23 F.3d 1508, 1510-11 (9th Cir.1994) (citing Marsh, 816 F.2d at 1383; see also TVA v. Hill, 437 U.S. 153, 173-74, 98 S-Ct. 2279, 57 L.Ed.2d 117 (1978); Friends of the Earth v. United States Navy, 841 F.2d 927, 933 (9th Cir.1988)). Plaintiffs have satisfied this standard by showing that an ESA violation due to an inadequate ITS is likely. As set forth above, plaintiffs have shown that they are1 likely to prevail- on establishing violations of the MMPA, NEPA, the ESA and the APA. They have also shown the possibility, indeed probability, of irreparable injury, particularly under the liberal standard applicable under these statutes. It is undisputed that marine mammals, many of whom depend on sensitive hearing for essential activities like finding food and mates and avoiding predators, and some of whom are endangered species, will at a minimum be harassed by the extremely loud and far traveling LFA sonar. For example, the important reproductive behavior of singing by the endangered humpback whale is affected at levels well'below 180 dB, although how significantly is debated. Other endangered species, such as sea turtles, will also be in LFA sonar’s path. Plaintiffs have also shown the possibility of irreparable harm to the marine environment that supports the existence of these species." For example, as the Ninth Circuit recognized in reversing the denial of a preliminary injunction, “acoustic environment appears to be very important to humpback whales.” National Parks & Conservation Ass’n, 241 F.3d at 727; at 737. Such injury cannot be remedied by money and is likely to be long lasting. Thus, plaintiffs have satisfied the first formulation of the test for issuance of a preliminary injunction: a' combination 'of probable success on the merits and a possibility of irreparable injury. The-Court has also balanced the hardships and considered the public interest. Here, as plaintiffs point out, a preliminary injunction will likely only be in place for under a year and-only affect use of LFA sonar in peacetime. The Navy has already delayed deployment for years by its own failure originally to timely initiate the required environmental processes of an EIS and consultation. The public interest in -the survival and flourishing of marine mammals and endangered species, as well as a healthy marine environment, is extremely strong. Indeed, Congress enacted the MMPA and -ESA in recognition of this compelling public interest, not only *1054 to the American public but to the international community, and not only to present generations but to future generations to come. For example, Congress found that “marine mammals have proven themselves to be resources of great international significance, esthetic and recreational as well as economic....” 16 U.S.C. § 1361. Stewardship of the world’s precious oceans and the marine life within them is undoubtedly of utmost importance. At the same time, the Navy has shown that a total ban on use of LFA sonar for training and testing would pose a hardship. More broadly, the public has a compelling interest in protecting national security by ensuring military preparedness and the safety of those serving in the military from attacks by hostile submarines. Defendants have made a showing that use of LFA sonar is likely to significantly increase our ability to timely detect very quiet submarines. The Deputy Chief of Naval Operations ranks the threat posed by modern quiet submarines at the “very top of those facing the U.S. Navy,” presenting “a clear and present danger in crucial parts of the world including the Persian Gulf, along the Korean Peninsula, and in the Taiwan Strait.” (Nathman Dec. at ¶ 8; see also Padgett Dec. at ¶¶ 8-13.) Vulnerability to this threat could inhibit the Navy’s ability to conduct “the full spectrum of operations from combat, to support for peacekeeping, to non-combatant evacuation, to peacetime presence.” (Nathman Dec. at ¶ 9; see also Padgett Dec. at ¶ 11; 67 Fed.Reg. 46716-17.) Plaintiffs are correct that the public interest at issue here in use of LFA sonar during peacetime is not as compelling as it would be in wartime or in a time of a declared heightened threat. Plaintiffs correctly point out that a preliminary injunction would not interfere with the Navy’s ability to use LFA sonar during war or in response to imminent threat: War, combat, and heightened threat conditions are determined by the Congress or the National Command Authorities (NCA), not the U.S. Navy.... Since these determinations are not made by the Navy, both the small take application and the Navy’s Draft and Final EISs are specifically limited to employment of the SURTASS LFA sonar during training, testing, and routine military operations and will not cover use of the SURTASS LFA system in self-defense, in times of war, combat or heightened threat conditions. 67 Fed.Reg. 46717. Nevertheless, the complexity of how sound travels through the ocean under varying conditions makes it important to train personnel to operate the system in advance — before the sonar’s reliable use becomes critical. “No operational commander can employ a system, of any type, with confidence that it is effective in combat unless the personnel using the system have already trained to use it and have used it, in a variety of realistic situations.” (Nathman Dec. at ¶ 10.) Balancing the harms and weighing the public interest, the Court concludes that a preliminary injunction should issue, but that it should not impose a complete ban on peacetime use of LFA sonar. Rather, the- preliminary injunction should be carefully tailored to reduce the risk to marine mammals and endangered species by restricting the sonar’s use in additional areas that are particularly rich in marine life, while still allowing the Navy to use this technology for testing and training in a variety of oceanic conditions. Cf. National Parks & Conservation Ass’n v. Babbitt, 241 F.3d at 737 n. 18 (even in environmental cases where unusual circumstances are present, an injunction most likely should be limited in scope, rather than withheld altogether). In particular, the preliminary injunction will extend the coastal buffer zone beyond 12 nautical miles in those *1055 coastal areas where LFA sonar can effectively operate at. that distance, and will include additional, interim Offshore Biologically Important.Areas that are reasonable candidates for permanent status, such as the Oyashio/Kuroshio area off Kamchatka. Defendants have acknowledged in the rule-making record and in declarations to the Court that they can restrict operations in certain parts of the ocean, during particular seasons, where LFA-equipped vessels, are more likely to encounter marine mammals and endangered species. A tailored injunction will help ensure that they do so in compliance with the statutory mandates, including MMPA’s mandate that LFA sonar have only a negligible impact on small numbers of marine mammals. Accordingly, the parties are ordered to meet and confer on the precise terms.of a preliminary injunction consistent with this opinion. The Court will hold a case management conference on November 7, 2002 at 1:30 p.m. to address finalizing the preliminary injunction and setting further dates. Meanwhile, defendants should not deploy LFA sonar. The parties should plan for a hearing on the merits in June, 2003, with briefing completed a month before the hearing.' The parties should file (and fax to chambers) a brief case management statement on November 5, 2002. IT IS SO ORDERED. Footnotes [1] Plaintiffs are Natural Resources Defense Council, Inc.; The Humane Society of the United States; Cetacean Society International; League for Coastal Protection; Ocean Futures Society; and Jean-Michel Cousteau. Defendants are Donald L. Evans, Secretary of the United States Department of Commerce; the National Marine Fisheries Service ("NMFS”); William Hogarth, Assistant Administrator for Fisheries of the National Oceanographic & Atmospheric Administration; Conrad C. Lautenbacher, Jr., Vice Admiral, Administrator of the National Oceanographic & Atmospheric Administration; the United States Department of the Navy; Gordon R. England, Secretary of the United States Department of the Navy; and Vern Clark, Admiral, Chief of Naval Operations. [2] The MMPA requires that NMFS give notice and an opportunity for public comment for public comment when processing a small take request. This process culminated in the Final Rule for issuance of a one-year Letter of Authorization for SURTASS LFA operations. [3] Plaintiffs' motion cites to a summary' of this definition in one of the comments to the Final Rule. 67 Fed.Reg. 46762. In their opposition brief,, defendants disingenuously claim that plaintiffs have seized upon a comment and have erroneously attributed the comment to NMFS. In fact, as pointed out above, the comment accurately summarizes NMFS’ position as set forth in other places in the Final Rule. [4] Plaintiffs argue in their reply'brief that'Congress- explicitly considered, but rejected, the term “significant'' in earlier drafts of the "harassment” definition. Plaintiffs cite H.RRep. No. 103-349 at § 15 (1994), but that section, arid indeed that entire House Report, makes no mention of an earlier draft of the Level B "harassment” definition that included the word "significant.” It, did contain a different definition of Level A harassment, under which harassment was "potential to harm,” while- "harm” was separately defined as "an act which is likely to kill to injure a marine mammal, significantly reduce its reproductive potential, or result in habitat modification or degradation that is likely to significantly impair .essential behavioral patterns.” H.R. No. 103-349 at § 15 (1994). That definition was later changed to "potential to injure” and the definition of "harm” was dropped entirely. Presumably, plaintiffs are trying to draw conclusions from the fact that Congress at one point considered defining "harm” to include acts that result in habitat modification that is likely to significantly impair essential behavior patterns, at the same time that Congress was defining Level B harassment as an act that has the potential to distprb a marine mammal or marine mammal stock in, the wild by causing disruption of behavioral patterns, without using the word "significant” before "disruption.” The Court finds this argument convoluted and unpersuasive. [5] An isopleth is a line drawn on a map connecting points having the same numerical value of some variable, here the area within the perimeter around the ship transmitting the LFA sonar at which the received level is 180 dB. [6] A nautical mile is 1.852 kilometers or 1.1508 miles. [7] In addition, here, the Navy and NMFS also originally considered alternative methods of submarine detection’ that did not involve SURTASS LFA sonar or other active acoustic technologies. (Navaro Dec. Ex. 14 at 1-8 — 1-9, Table 1-1, Ch. 2.3.4.) “The Navy legitimately concluded and properly explained that these alternative technologies cannot reliably provide U.S. forces with long-range detection (hundreds of nautical miles) and longer reaction times due to a number of critical factors.” (Navaro Dec. Ex. 14 at 1-8 — 1-9.) This discussion was adequate. [8] Defendants distinguish Shipshock on the basis that the Navy did not conduct an EIS in that case. However, the Shipshock court specifically does not address the Navy’s failure to issue an EIS. Id. at 739. [9] Both sides rely on cases that are not applicable. Defendants cite a Seventh Circuit case, State of Wisconsin v. Weinberger, 745 F.2d 412, 420 (7th Cir.1984), which held that the Court’s "task is the limited one of determining whether or not the new information presents a seriously different picture of the likely environmental consequences of the proposed action not adequately envisioned by the original EIS.” This case is not binding on this Court, however, and indeed conflicts with the Ninth Circuit's decision in Warm Springs. Id. at 429. Marsh adopted an "arbitrary and capricious” standard, rather than the "reasonableness” standard used earlier in Warm Springs. This change does not, however, make State of Wisconsin good law in this circuit. Plaintiffs cite Idaho Sporting Congress, Inc. v. Alexander, 222 F.3d 562, 566 (9th Cir.2000) for the proposition that, as a matter of law, the discussion in the Navy’s ROD cannot substitute for an SEIS. Idaho Sporting Congress explained that supplemental information reports cannot serve as a substitute for an SEIS. Id. Here, however, the Navy’s ROD did not substitute for an SEIS, but rather explained' why the Navy concluded that an SEIS was not necessary. 67 Fed.Reg. 48152. [10] Defendants misquote the language as “may affect but is not likely to 'adversely affect.” [11] The Court agrees with plaintiffs that defendants’ failure or inability to provide an estimate of the incidental take makes it difficult to understand how defendants have any basis for quantifying the take as “small.” In addition, defendants’ failure to provide any estimate of the incidental take of Hawaiian monk seals and Pacific gray whales, which are marine mammals, also casts doubt on their assertion in the Final Rule that only "small numbers” of marine mammals are likely to be taken.