In a July 15, 2016 decision, the Ninth Circuit Federal District Court ruled the National Marine Fisheries Service 2012 Final Rule regarding peace time use by the Navy of LFA sonar was inadequate to protect marine mammals. The Court stated: NMFS was required to analyze whether its proposed mitigation measures reduce the effects of LFA sonar to the “least practicable adverse impact.” The agency failed to do so.
According to the Court documents, the U.S. Navy uses Low Frequency Active (LFA) sonar vessels around the world to detect potentially hostile submarines. LFA sonar uses a set of transmitting projectors that are suspended by a cable from an ocean surveillance ship. The projectors produce low-frequency sound pulses at an intensity of approximately 215 decibels (dB), in sequences that last 60 seconds on average. While beneficial to national defense, LFA sonar can harm many marine mammal species, particularly “low-frequency hearing specialists” such as baleen whales, but also sperm whales and pinnipeds such as seals and walruses.
LFA sonar disrupts the hearing of these animals and can cause physical injury at sound levels greater than 180 dB. Effects from exposures below 180 dB can cause short-term disruption or abandonment of natural behavior patterns. These behavioral disruptions can cause affected marine mammals to stop communicating with each other, to flee or avoid an ensonified area, to cease foraging for food, to separate from their calves, and to interrupt mating.
LFA sonar can also cause heightened stress responses from marine mammals. Such behavioral disruptions can force marine mammals to make trade-offs like delaying migration, delaying reproduction, reducing growth, or migrating with reduced energy reserves.
Court's Summary of the Decision
The panel reversed the district court’s grant of summary judgment to federal defendants in a case relating to the proper scope under the Marine Mammal Protection Act (“MMPA”) of mitigation measures required to protect marine mammals when the responsible federal agency, the National Marine Fisheries Service, sought to approve incidental “take” relating to military readiness activities, namely, the Navy’s peacetime use of Surveillance Towed Array Sensor System Low Frequency Active sonar.
The Fisheries Service most recently authorized incidental take of marine mammals from Low Frequency Active sonar use for five years beginning in 2012 in a Final Rule.
The panel held that the 2012 Final Rule did not establish means of “effecting the least practicable adverse impact on” marine mammal species, stock and habitat, as was specifically required by the MMPA. The panel further held that the Fisheries Service impermissibly conflated the “least practicable adverse impact” standard with the “negligible impact” finding; and concluded that to authorize incidental take, the Fisheries Service must achieve the “least practicable adverse impact” standard in addition to finding a negligible impact.
The panel held that the Fisheries Service did not give adequate protection to areas of the world’s oceans flagged by its own experts as biologically important, based on the present lack of data sufficient to meet the Fisheries Service’s designation criteria. The panel remanded for further proceedings.
In the 2012 Final rule mitigation included shutdown of the sonar upon detection of a marine mammal and a coastal exclusion zone. The third mitigation involved designation of Offshore Biologically Important Area(s) (OBIAs). The Court concluded the NMFS erred in its designation of OBIAs.
FROM THE COURT DOCUMENTS:
The Navy’s plans for use of LFA sonar, as approved by NMFS, have gone through several iterations, resulting in increased protection for marine mammals. We have every reason to believe that the Navy has been deliberate and thoughtful in its plans to follow NMFS guidelines and limit unnecessary harassment and harm to marine mammals. But the question is whether NMFS has satisfied the Congressional mandate that mitigation measures ensure the “least practicable adverse impact” on marine mammals...
The facts before NMFS do not support its unexplained conclusion that the Final Rule’s mitigation measures achieve the “least practicable adverse impact” on marine mammal species, stock, and habitat.
We consider the practical impacts of mitigation measures adopted by NMFS and the likely practical impact of mitigation measures that were not adopted. Plaintiffs, while acknowledging that the three chosen mitigation measures reduce incidental take to some extent, contend that the measures are not sufficient to effect the least practicable adverse impact on marine mammal species, stock, and habitat, as the MMPA requires. Plaintiffs do not challenge the shutdown protocol and the coastal exclusion zone, two of the three mitigation measures discussed in Part II above. They do, however, contend that these measures taken as a whole are inadequate to make up for deficiencies in the third mitigation measure, the designation of OBIAs (Offshore Biologically Important Area).
The parties’ dispute over OBIA designation dates to the beginning of the Navy’s LFA sonar program. In the two previous iterations of LFA sonar rulemaking, many of the same Plaintiffs in this case challenged NMFS’s OBIA designations as underinclusive. Plaintiffs first challenged the 2002 Final Rule’s designation of only three OBIAs. The district court concluded that NMFS acted arbitrarily and capriciously in refusing to designate more OBIAs despite knowing of potentially sensitive areas, and that NMFS improperly shifted the burden to members of the public to prove that more OBIAs were necessary.
Five years later, the 2007 Final Rule prescribed 10 OBIAs, which Plaintiffs again challenged, and which the district court again concluded were inadequate to meet the MMPA’s least practicable adverse impact standard. Nat. Res. Def. Council v. Gutierrez, No. C-07-04771 EDL, 2008 WL 360852 at *10 (N.D. Cal. Feb. 6, 2008).
For its 2012 rulemaking, the most recent in the series, NMFS flagged 73 candidate OBIAs by consulting prior designated OBIAs, the World Database on Protected Areas, the First and Second Editions of Marine Protected Areas for Whales, Dolphins, and Porpoises by Dr. Erich Hoyt, and senior NMFS scientists identified as “subject matter experts.”
Four subject matter experts, all senior NMFS scientists, raised concerns about OBIA selection to NMFS’s Office of Protected Resources in a 2010 White Paper titled Identifying Areas of Biological Importance to Cetaceans in Data-Poor Regions. The White Paper authors were concerned that identifying OBIAs based only on known data would be difficult because in many instances, “relevant cetacean data are lacking for the appropriate region or spatial scale.” The White Paper recommended against equating data-poor regions with “zero population density” or “no biological importance.” Stated another way, the White Paper cautioned that NMFS should not assume that no or minimal data meant there were no or minimal cetacean populations in those areas.
The subject matter experts concluded that “proven ecological principles” suggest a precautionary approach that protects three types of areas as OBIAs:
(1) continental shelf waters and waters within 100 km of the continental slope;
(2) 100 km around all islands and seamounts that rise within 500 meters of the ocean surface; and
(3) regions of high primary productivity, known to correspond to higher sperm whale presence, as explained in a 2009 monograph that the White Paper cites. Michael A. Huston & Steve Wolverton, The Global Distribution of Net Primary Production: Resolving the Paradox, 79 Ecological Monographs 343 (2009).
Separate from the White Paper, but consistent with its recommendations, the Marine Mammal Commission urged that it was “not possible” to “ensure adequate protection of marine mammals” “if candidate areas are rejected simply because of insufficient information.” Again, no data does not mean no cetaceans.
None of the subject matter experts who wrote the White Paper was involved in drafting the Final Rule. The White Paper appears to have played no role in the drafting of the Final Rule until less than two months before the Final Rule was finalized, when one of the rule drafters told the Navy that she had “unearthed” the subject matter experts’ “guidelines for selecting OBIAs in data-poor areas.”
Even still, NMFS only responded to the White Paper in the Final Rule preamble’s response-to-comments section, which referred to the White Paper’s authors as “several other commenters” rather than as NMFS subject matter experts specifically convened to provide their expertise to the selection process.
NMFS’s chosen OBIA designation criteria differ significantly from the White Paper’s recommendations. The agency employed a multiple-step designation process that required for designation presence of one or more of the following attributes:
high densities of animals,
known breeding/calving grounds,
migration routes, or
small distinct populations with limited distributions.
NMFS evaluated these criteria based on what it termed the “best available information.” This designation method resulted in NMFS cutting nearly 70% of the candidate OBIAs.
NMFS’s stated reason for cutting so many potential OBIAs was that there were insufficient data proving at least one of the chosen criteria above, even though such data do not exist for most of the world’s oceans. NMFS also cut some areas that its subject matter experts had nominated for protection based on their judgment, regional expertise, or non-peer-reviewed literature, stating that those areas “require[d] more justification.”
As the district court observed in its review of the 2007 Final Rule’s OBIA selection, the current list of OBIAs once again shows what the Marine Mammal Commission deemed a “bias toward U.S. waters.” Only one OBIA was designated in each of the Caribbean Sea, Mediterranean Sea, Antarctic Convergence Zone, Southeast Atlantic, northwest Pacific, and southeast Pacific, and no area was designated on the Pacific Coast of South America. By contrast, NMFS designated four OBIAs each in the northwest Atlantic and the Northeast Pacific.
Plaintiffs contend that the resulting list of 22 OBIAs was an arbitrary and capricious policy choice. Defendants respond that NMFS considered the White Paper’s recommendations for data-poor regions but properly chose a different approach, to which this court must defer.
First, Defendants contend that NMFS was under no obligation to follow the White Paper’s guidelines, because the White Paper itself acknowledged the existence of a policy choice between a “precautionary” approach that “minimize[s] the chances of overlooking biologically important areas,” or a “pure” approach that “minimize[s] the chances of nominating sites that are of marginal biological importance and, therefore risk[s] overlooking biologically important areas.” Defendants contend that NMFS’s resulting policy choice is entitled to deference, in essence that NMFS’s explaining its decision not to adopt the White Paper’s recommendations is all that the APA requires.
The district court agreed with Defendants.
First, the district court found that the White Paper acknowledged that the “precautionary” approach that it advanced “risked designating OBIAs in areas of ‘marginal biological importance’ that did not meet NMFS’s criteria.” The district court concluded that because the White Paper refrained from choosing between the “precautionary” and “pure” approaches, it was inappropriate for the district court to substitute its judgment for that of the agency. The district court was satisfied that “NMFS chose the pure approach and explained its decision in the record, including reference to the White Paper and reasons for choosing a different approach.”
Second, the district court found it probative that the White Paper did not recommend specific OBIAs, but only provided guidelines for inferring biological significance.
Although review under the APA is deferential, here we evaluate the agency’s choices in the context not just of the APA, but also of the MMPA’s least practicable adverse impact requirement, which sets a “stringent standard.” We conclude that NMFS erred because the measures adopted do not result in the “least practicable adverse impact” on marine mammal species, stock, and habitat.
OBIAs are a central component of the Final Rule’s mitigation measures. The White Paper recommended a “precautionary” approach toward OBIA designation. The subject matter experts made clear that given the state of the science, particularly the many data-poor areas of the world’s oceans, NMFS faced a choice whether to protect areas likely to have biological importance based on “proven ecological principles,” or instead to “minimize the chances of nominating sites that are of marginal biological importance and, therefore, risk overlooking biologically important areas.” These competing options would either risk overprotection, or risk underprotection.
NMFS chose the latter option without evaluating whether its choice satisfied the least practicable adverse impact standard. It should have considered whether the precautionary approach would give more protection to marine mammals, and then whether that protection would impede military training to a degree making that mitigation not practicable.
For areas of potentially high biological importance, NMFS’s protocol made non-designation the default, and required specific data to overturn that conclusion.
This default is directly adverse to the subject matter experts’ recommended principle that shelf and slope areas should be protected absent “specific data to the contrary.” NMFS identified no science to support its conclusion that protecting data-poor areas of potential importance as OBIAs would not reduce adverse impacts on marine mammal species and habitat. In fact, Defendants themselves, as well as the district court, seem to agree with Plaintiffs that the Final Rule chose to forego some protections that would have further reduced the impact on marine mammals.
For example, Defendants accept that there may be alternative OBIA criteria or mitigation measures that NMFS could have reasonably selected, but argue that the agency’s choice should still be given deference. Yet the MMPA requires the “least practicable adverse impact,” and the agency has offered no explanation why it meets that standard—in fact, as explained above, it instead argues that it does not have to meet that standard.
Nor did NMFS consider more protected areas only to conclude that more protection was not practicable. Although Defendants argue that national security would be threatened without the long-range detection capacity that LFA sonar provides, NMFS’s decision to cut the list of potential OBIAs did not rely on practical considerations that particular additional OBIAs would impede military readiness training, aside from one decision regarding the Southern California Bight, which Plaintiffs did not challenge. Otherwise, military practicability played no role in NMFS’s decision; in fact, NMFS deemed it “immaterial.” Similarly, although counsel for Defendants claimed at oral argument that the Final Rule intended to balance the equities between military readiness and conservation, we hold that such balancing must be made explicit in rulemaking.
We owe “[o]ur highest deference” to the agency’s “technical analyses and judgments within its area of expertise.” League of Wilderness Defs. Blue Mountains Biodiversity Proj. v. Allen, 615 F.3d 1122, 1131 (9th Cir. 2010). But we do not “rubber-stamp . . . administrative decisions that [we] deem inconsistent with a statutory mandate or that frustrate the congressional policy underlying a statute.” Ocean Advocates v. U.S. Army Corps of Engineers, 402 F.3d 846, 859 (9th Cir. 2005) (alterations in original) (citation omitted). The agency decision here conflicts with the statutory mandate requiring mitigation at levels that yield the least practicable adverse impact. An agency conclusion that is in “direct conflict with the conclusion of its own experts,”—here, the agency’s drastic reduction of OBIAs by eliminating candidate OBIAs in datapoor waters against the recommendations of its subject matter experts—is arbitrary and capricious.
Defendants also urge us to defer to the agency’s chosen OBIA selection criteria, which differed from the White Paper’s, on the ground that we should not second-guess an agency’s reasonable treatment of scientific data. Defendants rely on this court’s decision in San Luis & Delta-Mendota Water Authority v. Jewell, 747 F.3d 581 (9th Cir. 2014), in which we stated that we will “reject an agency’s choice of a scientific model ‘only when the model bears no rational relationship to the characteristics of the data to which it is applied.’” Id. at 621 (first quoting Nat’l Wildlife Fed’n v. EPA, 286 F.3d 554, 565 (D.C. Cir. 2002); then quoting Appalachian Power Co. v. EPA, 135 F.3d 791, 802 (D.C. Cir. 1998)). In San Luis, experts differed on which of two methods was best suited to assess the effect of certain water projects on the endangered delta smelt. Id. The U.S. Fish and Wildlife Service (FWS) chose the more conservative method, and the court found that the choice was supported by the record and within FWS’s discretion. Id. at 610. But here, NMFS did not choose between competing methods designed to answer the same question; it made a policy choice not to protect areas—composing most of the world’s oceans—for which little scientific data exist.
This policy choice is underprotective compared to the alternative proposed by the agency’s subject matter experts. Although NMFS considered the White Paper’s discussion of data-poor regions, the record does not show that NMFS critiqued the White Paper’s scientific analysis or concluded that its proposed guidelines were unsound. NMFS instead offered two reasons for refusing to designate additional OBIAs based on the White Paper’s principles.
First, NMFS said that areas identified using the White Paper’s ecological principles did “not meet the criteria we established” for designating such areas. But this distinction is tautological. The White Paper’s criteria were different than those NMFS ultimately chose, but the difference itself does not explain why NMFS’s criteria were equally or more capable of meeting the statutory standard, particularly in areas where site-specific data do not exist.
Defendants contend that NMFS need not consider mitigation measures that are not supported by the best available information because federal regulations provide that the Final Rule and its mitigation measures are to be based on the “best available information.” Moreover, Defendants contend, “[t]he determination of what constitutes the ‘best scientific data available’ belongs to the agency’s ‘special expertise.’”
Defendants make a case that NMFS used the best available science in deciding if a potential OBIA met its screening criteria. But the screening criteria themselves required a definitive showing of biological significance even though NMFS’s own experts concluded that “[t]he task of identifying OBIAs for cetaceans is particularly difficult for regions in which data on cetacean distribution or population density are limited or lacking entirely, which includes the majority of the world’s oceans.”
This selection of screening criteria, in our view, was a policy choice, not a scientific determination. We review it according to the standard voiced by the Supreme Court in State Farm, and hold that NMFS “failed to consider an important aspect of the problem,” namely the underprotection that accompanies making conclusive data an indispensable component of OBIA designation. This systematic underprotection of marine mammals cannot be consistent with the requirement that mitigation measures result in the “least practicable adverse impact” on marine mammals.
Second, in response to the White Paper’s conclusion that it is “not acceptable to proceed in the decision making process as if the ‘no data’ scenario were equivalent to . . . ‘no biological importance,’” NMFS reasoned that OBIAs are but one component of a “suite” of mitigation measures. However, the other two mitigation measures, the shutdown zone and coastal exclusion zone, apply regardless of whether an area is considered potentially biologically important or not. Relative to these mitigation measures, the only heightened protection possible under the agency’s plan is designation as an OBIA.
Defendants repeatedly emphasize that NMFS’s decision not to designate an area as an OBIA did not mean that the agency assumed the area was biologically unimportant. But this is exactly how NMFS treated data-poor areas when it categorically barred their designation as OBIAs.
Furthermore, the MMPA’s mitigation requirement applies to marine mammal “species or stock and its habitat,” and NMFS must “pay particular attention to rookeries, mating grounds, and areas of similar significance[.]” 16 U.S.C. § 1371(a)(5)(A)(i)(II)(aa) (emphasis added). This statutory guidance means that protecting marine mammal habitat from the effects of LFA sonar is of paramount importance under the MMPA. Defendants’ reference to their “suite” of mitigation measures does not repair the under inclusive OBIA designation protocol, particularly in light of the subject matter experts’ conclusion that “the other forms of mitigation [are] considerably less effective than specifying OBIAs.” The result is that a meaningful proportion of the world’s marine mammal habitat is underprotected.
Relying on the other two mitigation measures—shutdown upon detection of a marine mammal and a coastal exclusion zone—in all areas of the ocean not designated as OBIAs ignores that OBIAs are one of only two mitigation measures capable of measurably reducing Level B harassment. The shutdown zone around LFA sonar vessels is not large enough to protect marine mammals from Level B effects between 165 and 175 dB. As a result, unless an area is designated as an OBIA or lies within 22 km of the coast, there is minimal mitigation of Level B harassment.
Although Defendants emphasize that OBIAs are not the core component of the Final Rule’s “suite” of mitigation measures, the record does not show that the other mitigation measures achieve the least practicable adverse impact.
The complete decision can be viewed on the Court's website.