UNITED STATES DISTRICT COURT FOR THE DISTRICT OF COLUMBIA
____________________________________
)
OCEANA, INC., )
)
Plaintiff, )
)
v. )
)
WILBUR L. ROSS, United States )
1
Secretary of Commerce, et al., ) Civil Action No. 08-1881 (PLF)
)
Defendants, )
)
and )
)
FISHERIES-SURVIVAL FUND, )
)
Defendant-Intervenor. )
____________________________________)
OPINION
This matter comes before the Court on defendants’ notice that the National Marine Fisheries Service (“NMFS”) had revised its Incidental Take Statement, thus completing its remand in response to the Court’s December 17, 2014 Opinion and Order. Plaintiff Oceana, Inc. filed a response to the notice, challenging the adequacy of the agency’s revisions on remand, and the parties proceeded to brief the matter. Upon consideration of the revised Incidental Take
1 Pursuant to Rule 25(d) of the Federal Rules of Civil Procedure, the Court substitutes as defendant the current Secretary of Commerce, Wilbur L. Ross, for former Secretary Penny Pritzker.
Statement, the parties’ briefs, the relevant legal authorities, and the entire record in this case, the Court will remand to the agency for a limited purpose. 2
I. FACTUAL AND PROCEDURAL BACKGROUND This case involves a challenge to the 2012 Biological Opinion (“BiOp”) issued by NMFS that pertains to the operation of the Atlantic Sea Scallop Fishery and its impact on a threatened species under the Endangered Species Act – the Northwest Atlantic population segment of loggerhead sea turtles. The Court previously issued an opinion granting in part and denying in part the parties’ cross-motions for summary judgment and, in turn, remanding the case to the agency for the limited purpose of addressing two deficiencies in the 2012 BiOp. See Oceana, Inc. v. Pritzker, 75 F. Supp. 3d 469 (D.D.C. 2014). In its prior opinion, the Court described the relevant statutory and regulatory framework and recounted the factual and procedural history of this case. The Court thus recites here only those matters relevant to resolving the parties’ instant dispute.
The Endangered Species Act (“ESA”) of 1973, as amended, 16 U.S.C. § 1531 et seq., created a comprehensive legislative and regulatory scheme that seeks to preserve and protect species of animals facing man-made threats to their continued existence. See Lujan v.
2 In reaching its decision, the Court has reviewed the following filings, including the exhibits attached thereto: First Amended Complaint for Declaratory and Injunctive Relief (“Am. Compl.”) [Dkt. No. 80]; Notice of Proposed Schedule for Remand (“Proposed Sched.”) [Dkt. No. 111]; Federal Defendants’ Notice of Completion of Remand (“Notice of Completion”) [Dkt. No. 113]; Federal Defendants’ Notice of Filing Supplemental Administrative Record (“Suppl. R.”) [Dkt. No. 114]; Plaintiff’s Response to Notice of Completion of Remand (“Pl. Resp.”) [Dkt. No. 116] and Plaintiff’s Errata thereto (“Errata”) [Dkt. No. 117]; Federal Defendants’ Response in Support of Notice of Completion of Remand (“Defs. Resp.”) [Dkt. No. 127]; Defendant-Intervenor Fisheries-Survival Fund’s Response in Opposition to Plaintiff’s Response to Notice of Completion of Remand (“Def.-Intervenor Resp.”) [Dkt. No. 128]; and Plaintiff’s Reply to Notice of Completion of Remand (“Pl. Reply”) [Dkt. No. 129].
3
Defs. of Wildlife, 504 U.S. 555, 558 (1992); Tenn. Valley Auth. v. Hill, 437 U.S. 153, 180 (1978). As part of this scheme, Section 7 of the ESA sets forth “the steps that federal agencies must take to ensure that their actions do not jeopardize endangered wildlife and flora.” See Nat’l Ass’n of Home Builders v. Defs. of Wildlife, 551 U.S. 644, 652 (2007). In particular, Section 7(a)(2) requires that each federal agency, “in consultation with and with the assistance of [NMFS or the U.S. Fish and Wildlife Service (“FWS”)], insure that any action authorized, funded, or carried out by such 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 species . . . .” See 16 U.S.C. § 1536(a)(2). 3 The Section 7 consultation process culminates in the issuance of a Biological Opinion, or BiOp, in which the consulting agency sets forth its “opinion, and a summary of the information on which the opinion is based, detailing how the agency action affects the species or its critical habitat.” See 16 U.S.C. § 1536(b)(3)(A); see also 50 C.F.R. § 402.14(h). Where the consulting agency concludes that the agency action is not likely to jeopardize the continued existence of the species but is nonetheless likely to result in some “incidental take,” the BiOp must include an Incidental Take Statement (“ITS”) specifying the permissible extent of this impact on the species. See 16 U.S.C. § 1536(b)(4); 50 C.F.R. § 402.14(i). The ITS must set
3 FWS and NMFS jointly administer the ESA. See 50 C.F.R. § 402.01(b). FWS administers the statute with respect to species under the jurisdiction of the Secretary of the Interior, while NMFS covers those species under the jurisdiction of the Secretary of Commerce. See Nat’l Ass’n of Home Builders v. Defs. of Wildlife, 551 U.S. at 651. The agency whose action is at issue is known as the “action agency,” while either FWS or NMFS serves as the “consulting agency.” See Nat’l Wildlife Fed’n v. Nat’l Marine Fisheries Serv., 524 F.3d 917, 924 (9th Cir. 2008). In this case, NMFS serves as both the “action agency” and the “consulting agency”: NMFS’ Sustainable Fisheries Division of its Northeast Regional Office administers the fisheries management program that governs the Atlantic Sea Scallop Fishery, making it the action agency here, while the Protected Resources Division of the same Regional Office has served as the consulting agency. See Am. Compl. ¶¶ 18-19.
4
forth conditions that include “reasonable and prudent measures” considered “necessary or appropriate to minimize” the extent of any incidental takings. See 50 C.F.R. § 402.14(i)(1)(ii). 4 And if the amount or extent of incidental taking ever exceeds that specified in the ITS, the action agency must reinitiate Section 7 consultation “immediately.” See 50 C.F.R. § 402.14(i)(4); see also 50 C.F.R. § 402.16(a). As a result, incidental take monitoring is a key component of any ITS – without the ability to monitor incidental takes, these regulatory requirements become meaningless.
In its earlier opinion, the Court reviewed NMFS’ 2012 BiOp, in which the agency determined that the operation of the Atlantic Sea Scallop Fishery would not jeopardize the continued existence of the Northwest Atlantic population segment of loggerhead sea turtles. See Oceana, Inc. v. Pritzker, 75 F. Supp. 3d at 473. The Court found that the 2012 BiOp survived Oceana’s challenges in large part, but remanded to the agency for two discrete purposes. See id. at 499.
First, the Court remanded to allow the agency to better explain its reliance on a monitoring surrogate to measure loggerhead turtle takes caused by dredge fishing. See Oceana, Inc. v. Pritzker, 75 F. Supp. 3d at 494-97. Because technologies meant to benefit loggerheads have also made direct observations of takes more difficult, NMFS had chosen to measure takes by a surrogate – namely, by “dredge hour,” or the number of hours spent dredge fishing. See id. at 494-95. Specifically, the agency had considered a variety of monitoring mechanisms, but ultimately decided to monitor takes by using the number of hours spent dredge fishing in
4
As defined by the ESA, to “take” means “to harass, harm, pursue, hunt, shoot, wound, kill, trap, capture, or collect, or to attempt to engage in any such conduct.” See 16 U.S.C. § 1532(19). Although Section 9 of the ESA generally prohibits any taking of a listed species, see 16 U.S.C. § 1538(a)(1), incidental takes are permissible if they occur in accordance with the conditions set forth in an ITS, see 50 C.F.R. § 402.14(i)(5).
5
Mid-Atlantic waters from May through November as a surrogate for actual takes. See id. at 496. The Court, however, found that “the 2012 BiOp fail[ed] to sufficiently explain how the specific number of dredge hours that NMFS ha[d] chosen as a monitoring surrogate adequately serve[d] as a proxy for the numerical take limit of 161 loggerheads.” See id. at 497. And because NMFS had not adequately explained whether and how 252,323 hours spent dredge fishing equated to 161 takes, it had not shown that the dredge hour surrogate would serve as an adequate “trigger” for requiring reinitiation of consultation when actual takes surpassed the take limit. See id. at 496-97. The Court thus remanded the ITS in order for NMFS to “more clearly explain the connection,” or, if unable to do so, to choose a monitoring mechanism that “does align with the numerical take limit.” See id. at 497. In doing so, the Court directed the agency to “address Oceana’s valid concern regarding the effectiveness of linking an hour-based surrogate to a numerical take limit, in the context of a Fishery where conditions change on a continuous basis.” See id.
Second, the Court remanded with regard to the agency’s decision to evaluate loggerhead takes resulting from trawl gear fishing only once every five years. See Oceana, Inc. v. Pritzker, 75 F. Supp. 3d at 497-99. In reaching this decision, the Court acknowledged that “[t]he five-year timetable may reflect very real limitations on NMFS’ data collection capabilities,” thus rendering the NMFS’ proposed monitoring system for trawl takes “the best available option for measuring trawl takes in the Fishery.” See id. at 498. But because NMFS had given only “terse treatment” to trawl take monitoring in the 2012 BiOp, the Court remanded in order to allow the agency to “either provide a more thorough explanation of its choice, or, if unable to do so, reach a different conclusion.” See id. at 499.
6
NMFS thereafter revised the ITS for the 2012 BiOp and now contends that it has completed its required tasks on remand by more thoroughly explaining its chosen monitoring methods. See Notice of Completion. Oceana filed a response arguing that the ITS remains defective and, as a result, NMFS still has failed to demonstrate that its monitoring methods are not arbitrary and capricious. See Pl. Resp. In support of this position, Oceana attached to its response the declaration of professional statistician George Weaver, Ph.D. NMFS moved to strike Dr. Weaver’s declaration, as well as those portions of Oceana’s response that relied on his declaration. The Court denied this motion, ruling that Oceana would be permitted to rely on Dr. Weaver’s expertise and that NMFS would be permitted to proffer a rebuttal expert. The parties thereafter submitted briefs, with supplemental expert declarations, arguing the merits of whether the agency had in fact completed its remand in accordance with the Court’s December 17, 2014 Opinion and Order.
II. LEGAL STANDARD
Under the Administrative Procedures Act, a reviewing court shall “hold unlawful and set aside agency action, findings, and conclusions found to be . . . arbitrary, capricious, an abuse of discretion, or otherwise not in accordance with law.” See 5 U.S.C. § 706(2)(A). This “arbitrary and capricious” standard is a deferential one – “it requires that agency action simply be ‘reasonable and reasonably explained.’” See Cmtys. for a Better Env’t v. Envtl. Prot. Agency, 748 F.3d 333, 335 (D.C. Cir. 2014) (quoting Nat’l Tel. Coop. Ass’n v. Fed. Commc’ns Comm’n, 563 F.3d 536, 540 (D.C. Cir. 2009)); see also Kennecott Greens Creek Mining Co. v. Mine Safety & Health Admin., 476 F.3d 946, 954 (D.C. Cir. 2007) (“[The] standard of review under the arbitrary and capricious test is only reasonableness, not perfection.”). As a result, a reviewing court will uphold an agency action so long as the agency has “articulated a satisfactory
7
explanation for its action including a rational connection between the facts found and the choice made.” See FirstEnergy Serv. Co. v. FERC, 758 F.3d 346, 352 (D.C. Cir. 2014) (quoting Motor Vehicle Mfrs. Ass’n v. State Farm Mut. Auto. Ins. Co., 463 U.S. 29, 43 (1983)).
Furthermore, courts “give an extreme degree of deference to the agency when it is evaluating scientific data within its technical expertise.” See Cmtys. for a Better Env’t v. Envtl. Prot. Agency, 748 F.3d at 336 (quoting City of Waukesha v. Envtl. Prot. Agency, 320 F.3d 228, 247 (D.C. Cir. 2003)). A court must remain mindful that it reviews an agency’s scientific judgments “not as the chemist, biologist, or statistician that [the court is] qualified neither by training nor experience to be,” and thus it may exercise only the “narrowly defined duty of holding agencies to certain minimal standards of rationality.” See Troy Corp. v. Browner, 120 F.3d 277, 283 (D.C. Cir. 1997) (quoting Ethyl Corp. v. Envtl. Prot. Agency, 541 F.2d 1, 36 (D.C. Cir. 1976) (en banc)).
Nonetheless, the Court’s review must be “searching and careful.” See Colo.
River Cutthroat Trout v. Salazar, 898 F. Supp. 2d 191, 199 (D.D.C. 2012) (quoting Nat’l Envtl. Dev. Ass’n’s Clean Air Project v. Envtl. Prot. Agency, 686 F.3d 803, 810 (D.C. Cir. 2012)). A decision may be deemed arbitrary and capricious where an agency “relied on factors which Congress has not intended it to consider, entirely failed to consider an important aspect of the problem, [or] offered an explanation for its decision that runs counter to the evidence before the agency[] or is so implausible that it could not be ascribed to a difference in view or the product of agency expertise.” See Cablevision Sys. Corp. v. Fed. Commc’ns Comm’n, 649 F.3d 695, 714 (D.C. Cir. 2011) (quoting Motor Vehicle Mfrs. Ass’n v. State Farm Mut. Auto. Ins. Co., 463 U.S. at 43); accord Agape Church, Inc. v. Fed. Commc’ns Comm’n, 738 F.3d 397, 410 (D.C. Cir. 2013). Therefore, just as the Court may not “substitute [its] judgment for that of the
8
agency,” see Rural Cellular Ass’n v. Fed. Commc’ns Comm’n, 588 F.3d 1095, 1105 (D.C. Cir. 2009), it also generally may not “affirm an agency decision on a ground other than that relied upon by the agency,” see Manin v. Nat’l Transp. Safety Bd., 627 F.3d 1239, 1243 (D.C. Cir. 2011).
In addition, where an administrative agency has been ordered to reconsider or explain an earlier decision on remand, as is the case here, the agency has an “affirmative duty to respond to the specific issues remanded” by the Court. See Defs. of Wildlife v. Kempthorne, No. 04-1230, 2006 WL 2844232, at *12 (D.D.C. Sept. 29, 2006) (first citing Tex Tin Corp. v. Envtl. Prot. Agency, 992 F.2d 353, 355 (D.C. Cir. 1993); then citing Ass’n of Civilian Technicians v. Fed. Labor Relations Auth., 370 F.3d 1214, 1223 (D.C. Cir. 2004)). The agency “retains some discretion to determine how it ‘may best proceed to develop the needed evidence and how its prior decision should be modified in light of such evidence as develops.’” See id. at *11 (quoting Fed. Power Comm’n v. Transcon. Gas Pipe Line Corp., 423 U.S. 326, 333-34 (1976)). And the remanding court “may not dictate to the agency the ‘methods, procedures, or time dimension,’ for its reconsideration.” See id. (quoting SEC v. Chenery Corp., 318 U.S. 194, 196 (1947)). Nor may the court demand that an agency reach any particular result. See id. (citations omitted). But the Court retains jurisdiction to enforce the terms of its remand order where the agency does fail to adequately respond. See id. at *12 (citing Int’l Union, United Auto., Aerospace & Agric. Implement Workers of Am., UAW v. Occupational Safety & Health Admin., 976 F.2d 749, 751 (D.C. Cir. 1992)).
III. ANALYSIS
Oceana challenges the revised ITS submitted by NMFS, arguing that the agency has not adequately complied with the requirements of the Court’s remand. Oceana asserts that
9
the revised ITS still fails to justify both the dredge hour surrogate for dredge fishing takes and the five-year timetable for trawl fishing takes and, as a result, NMFS’ reliance on these monitoring mechanisms is arbitrary and capricious. In addition, Oceana argues that the agency disregarded the findings of a recent relevant study, in violation of the requirement that it use the “best scientific and commercial data available” in developing the revised ITS. See 16 U.S.C. § 1536(a)(2); 50 C.F.R. § 402.14(g)(8).
A. Use of the Dredge Hour Surrogate for Monitoring Dredge Takes On remand, the Court directed NMFS to more clearly explain its reliance on the dredge hour surrogate for monitoring takes. In response, the agency issued a revised ITS which explains that the use of underwater video monitoring remains infeasible, see Notice of Completion Ex. 1 at 8, and that the use of dredge hours as a proxy or surrogate for the number of takes is still the best alternative in the short term, see id. at 9-12. It includes a more thorough explanation of the benefits of a monitoring surrogate for dredge fishing takes generally, and the benefits of using dredge hours as a surrogate in particular. The agency explains that “effort data,” such as dredge hours, “is well-documented and available by location and season.” See id. at 10. In addition, “[t]o clarify the link between the number of sea turtles taken and the dredge hour surrogate” – a primary concern raised by the Court in its earlier opinion – the agency includes a scatterplot graph of the relationship between biennial averages of estimated takes of hard-shelled sea turtles and biennial averages of vessel trip report dredge hours in the Mid-Atlantic scallop fishery from May through November for the years 2001 to 2008. See id. at 10-11. The revised ITS explains that this scatterplot includes a fitted regression line with an R-squared value of 0.9164 and illustrates “the strong positive linear relationship” that exists between dredge effort and hard-shelled sea turtle takes. See id. at 10 & n.5. And because of this
10
strong positive linear relationship, the agency maintains that the scatterplot demonstrates the reasonableness of using a dredge hour surrogate to determine whether the incidental take level for sea turtles has been exceeded. See id. at 10.
Oceana asserts that the revised ITS continues to fall short. Oceana argues that the revised ITS still does not demonstrate that the selected number of dredge hours serves as a reliable proxy for monitoring whether and when the take limit has been exceeded. See Pl. Resp. at 8-13. Oceana criticizes the agency’s reliance on a scatterplot that does not include the necessary data and metrics to allow independent scientific assessment and verification, as is customarily included with similar regression models. See id. at 9-12; see also Pl. Resp. Ex. 1 at ¶¶ 14-23. As presented, Oceana argues, it is impossible to determine whether the agency’s scatterplot actually represents a linear relationship or, instead, represents two clusters of data that reflect operational changes to the fishery. See Pl. Resp. at 9-10; see also Pl. Resp. Ex. 1 at ¶ 15. Thus, Oceana maintains, the revised ITS fails to adequately explain how an hour of dredging equals a specific number of takes in a fishery “where conditions change on a continuous basis.” See Pl. Resp. at 7, 10 n.2 (quoting Oceana, Inc. v. Pritzker, 75 F. Supp. 3d at 497).
In response, the agency asserts that it has adequately explained the connection between dredge hours and takes. See Defs. Resp. at 4-9. In doing so, however, the agency explicitly disclaims any reliance on a linear regression model. See id. at 7-9; see also Defs. Resp. Ex. 1 at ¶¶ 5-6; Defs. Resp. Ex. 2 at ¶¶ 3-5. The scatterplot, the agency explains, is meant to illustrate only “the positive relationship between commercial dredge hours and estimated sea turtle takes.” See Defs. Resp. at 8; see also Defs. Resp. Ex. 1 at ¶¶ 5-6; Defs. Resp. Ex. 2 at ¶ 3. Despite the implications of the plain language of the revised ITS, see Notice of Completion Ex. 1 at 10-11; see also Pl. Resp. Ex. 1 at ¶¶13, 15; Pl. Reply Ex 1 at ¶¶ 5-8, the agency states that the
11
scatterplot is not a linear regression model and is “not intended to be a predictive model whereby sea turtle takes can be predicted based on a certain level of dredge effort.” See Defs. Resp. Ex. 1 at ¶ 5; see also Defs. Resp. at 8. And because NMFS is not even attempting to make predictions, the agency explains, the statistical practices discussed by Oceana in its critique of the revised ITS do not apply. See Defs. Resp. at 8; see also Defs. Resp. Ex. 1 at ¶ 5.
Of course, the agency’s disclaimers beg the question: What good is a surrogate based on a model that lacks any predictive value? And why describe a scatterplot as illustrating a “strong positive linear relationship” with a linear formula of “y = 0.0014x – 337.05” and a fitted regression line with an R-squared value of .9164, see Notice of Completion Ex. 1 at 10-11, if the scatterplot does not in fact reflect a linear relationship?
When it first remanded this case, the Court directed NMFS to better explain the rationale underlying its reliance on the dredge hour surrogate. See Oceana, Inc. v. Pritzker, 75 F. Supp. 3d at 497. And the agency responded by publishing a revised ITS that explicitly illustrated the relationship between dredge hours and takes as a “strong positive linear relationship.” See Notice of Completion Ex. 1 at 10-11. But when Oceana challenged this interpretation of the data, NMFS did an about-face and expressly disclaimed any reliance on a predictive, linear relationship. See Defs. Resp. at 7-9; Defs. Resp. Ex. 1 at ¶¶ 5-6; Defs. Resp. Ex. 2 at ¶¶ 3-5. In doing so, the agency has now contradicted the plain language of its own revised ITS. See Notice of Completion Ex. 1 at 10-11 & n.5 (describing the scatterplot as illustrating a “strong positive linear relationship” modeled by a regression line with an R-squared value of 0.9164 and a linear formula of “y = 0.0014x – 337.05”). NMFS thus has left things precisely where they were prior to remand – without a sufficient basis to believe that 359,757 dredge hours will serve as an adequate proxy for 161 takes and thus provide an appropriate mechanism for determining
12
whether and when the take limit is exceeded. 5 As the Court has stated before, “[a] surrogate is only as useful as its fit with the actual object of study for which the surrogate is substituting.” See Oceana v. Pritzker, 75 F. Supp. 3d at 497 (citing Wild Fish Conservancy v. Salazar, 628 F.3d at 531). And here, the agency has offered directly conflicting explanations for its reliance on 359,757 dredge hours as a surrogate with an adequate “fit” for 161 takes. 6 The Court acknowledges that pragmatic, real world constraints necessarily limit the abilities of agencies to collect perfect data and apply it in perfect ways. This is particularly so considering that monitoring of takes has become more difficult precisely because more protective equipment is now being used in the industry. See Oceana, Inc. v. Pritzker, 75 F. Supp. 3d at 494-95. It very well may be that the use of a dredge hour surrogate now is the best method reasonably available to monitor dredge fishing takes, and the specific dredge hour number selected is the agency’s best estimate in light of real world variations and uncertainty. But if that
5 The Court notes that, in revising the ITS, NMFS has adjusted the total number of dredge hours selected to serve as a surrogate for 161 takes, from 252,323 hours to 359,757 hours. See Notice of Completion Ex. 1 at 11 n.6. The agency explains that the adjusted total number of hours was calculated after the agency “performed additional quality control checks on the dredge hour data.” See id. In challenging the revised ITS, Oceana does not appear to take issue with this adjusted total amount of hours in particular.
6 Similarly, the Court finds perplexing the defendants’ assertion in their response brief that the dredge hour measure is intended to serve neither as a “surrogate” nor as a “proxy” for an anticipated number of takes. See Defs. Resp. at 6 n.3. Despite what defendants may now claim, the revised ITS unequivocally states that the agency “will use fishing effort (i.e., dredge hours) as a surrogate measure for monitoring actual takes of loggerhead . . . .” See Notice of Completion Ex. 1 at 9 (emphasis added); see also Notice of Completion at 9 (“The dredge hour proxy functions as a trigger because if any two-year running average of dredge hours exceeds the average number of hours from 2007 and 2008 then NMFS will consider the incidental take level to have been exceeded . . . .” (emphasis added)); Notice of Completion Ex. 1 at 3 (“The monitoring plan also includes a ‘dredge hour’ (i.e., effort) monitoring surrogate for the dredge fishery which will be used to determine whether the specified numerical incidental take level has been exceeded. A surrogate is necessary because gear modifications (e.g., chain mats and turtle deflector dredges), implemented to avoid serious injury and mortality to sea turtles, prevent turtles from entering the dredge bag.”).
13
is the case, the agency simply must provide a reasonable explanation for that conclusion – the Court cannot permit the agency instead to cloak its decisions with the imprimatur of scientific and statistical methods it either cannot or will not substantiate.
Simply put, the revised ITS advances a rationale now expressly disclaimed by the agency. And as a result, the revised ITS “fails to explain how [359,757] dredge hours equals 161 takes.” See Oceana, Inc. v. Pritzker, 75 F. Supp. 3d at 497. Because the revised ITS “suffer[s] from the same underlying defect” that warranted remand in the first instance, the Court will again remand the revised ITS to the agency. See Defs. of Wildlife v. Norton, No. 99-0927, 2003 WL 24122459, at *6 (D.D.C. Jan. 7, 2003); see also Defs. of Wildlife v. Kempthorne, 2006 WL 2844232 at *13. On remand, the Court directs NMFS either to more clearly explain its reliance on the dredge hour surrogate and its selection of 359,757 dredge hours as a surrogate for 161 takes or, if unable to do so, to select a more appropriate surrogate or other mechanism for monitoring loggerhead takes resulting from dredge fishing.
B. Use of the Five-Year Timetable for Monitoring Trawl Takes In its prior opinion, the Court expressed concerns regarding the agency’s decision to evaluate takes resulting from trawl gear fishing and assess whether the annual take limit had been exceeded only once every five years. See Oceana, Inc. v. Pritzker, 75 F. Supp. 3d at 497-99. The Court acknowledged that “[t]he five-year timetable may reflect very real limitations on NMFS’ data collection capabilities,” but the Court found that the “single paragraph” NMFS had devoted to explaining monitoring of trawl takes was insufficient to allay the Court’s concerns. See id. at 498-99.
On remand, the agency revised the ITS to include a much more fulsome discussion of trawl take monitoring, explaining why no better monitoring alternatives exist than
14
the agency’s method for evaluating five-year averages. Oceana argues, however, that the agency’s explanations remain inadequate. See Pl. Resp. at 13-16. It maintains that the revised ITS merely reiterates the same terse explanation the Court found insufficient initially, without explaining why the agency cannot use annual estimates, as it does for other fisheries, or why it cannot increase observer coverage. See id.
The Court disagrees. The revised ITS explains that observed loggerhead interactions with trawls are rare and depend on a wide range of both human and natural factors that vary greatly over a short time period. See Notice of Completion Ex. 1 at 13. Even a very simple model requires at least twenty to thirty bycatch events, but it is uncommon to have this many observed loggerhead interactions in a single year. See id. Thus, the agency often must pool data across years to have sufficient information to produce a robust, model-based estimate of total interactions with reliable confidence intervals. See id. at 12-13. In addition, it normally takes a year to process, clean, and analyze the collected data. See id. at 13. The agency did consider whether annual estimates might be preferable. See id. But in light of the practical constraints on data collection and the existing scholarship on how best to employ sparse data in a reliable manner, the agency decided to undertake “[l]ess frequent but more comprehensive assessments, which explicitly address uncertainty.” See id. at 13 (quoting Suppl. R. Ex. 1 at 41); see also Suppl. R. Ex. 1 at 26-27. Thus, the agency determined that, for loggerheads, the model of re-estimating takes in the scallop trawl fishery “approximately every five years” amounts to the best available monitoring option. See Notice of Completion Ex. 1 at 13. 7
7 Compared to this determinations for loggerhead turtles, the agency explains that the raw annual numbers of observed takes for leatherback, Kemp’s ridley, and green sea turtles are the “best available scientific information” and “reviewing the raw annual numbers of observed takes is the only available method for monitoring the incidental take level in trawl gear” for these three species. See Notice of Completion Ex. 1 at 13. Oceana critiques the
15
The revised ITS appears to reflect the agency’s careful judgment in light of the best scientific information available and the practical constraints on its monitoring capacities. The agency has thus accomplished what the Court asked it to do – namely, explain why it has concluded that the five-year monitoring system for trawl takes represents the best available method. See Oceana, Inc. v. Pritzker, 75 F. Supp. 3d at 498-99. The agency further explained the source of the lack of data – the rarity and variability of loggerhead observations – making clear that real-world conditions and practical considerations create “very real limitations on NMFS’ data collection capabilities.” See id. at 488. The Court will not second-guess the agency’s reasoned application of its technical expertise to reach these conclusions. See Cmtys. for a Better Env’t v. Envtl. Prot. Agency, 748 F.3d at 336.
Furthermore, the Court does not find it necessary to remand this matter in order for the agency to more extensively address Oceana’s proposal that NMFS increase observer coverage of the trawl fishery. The revised ITS makes clear that observer coverage was only one of many factors precluding the production of a reliable short-term take estimate. See Notice of Completion Ex. 1 at 13. The agency explained that it could not produce reliable short-term estimates because observed turtle takes are rare events “dependent on a wide range of both human and natural factors that vary greatly over short time periods.” See id. (emphasis added). For example, observed turtle takes are dependent not only on observer coverage, but on the number of vessels and time spent fishing, regulatory regimes and market forces, water
agency’s use of annual monitoring mechanisms for other species and in other contexts as evidence that annual monitoring would be feasible for loggerhead turtles in this case. See Pl. Resp. at 14-15. But to the extent these other documents are reviewable in this case, compare Defs. Resp. at 12 n.4, with Pl. Reply at 6 n.4, these differences appear to indicate instead that the agency has carefully considered the unique circumstances of each species and setting on a case-by-case basis. See also Defs. Resp. at 12 n.4.
16
temperature and weather conditions, distribution of prey, and shifting distributions and abundance of loggerhead themselves. See id. And what is implicit in this discussion is explicitly confirmed in defendants’ brief – the possibility of increased observer coverage, even if a feasible monitoring option, would not have altered the outcome of the decision to adopt a five-year monitoring timetable in light of these many other variables. See Defs. Resp. at 12; see also Notice of Completion Ex.1 at 7, 13; Notice of Completion Ex. 2 at 4-5.
The Court also notes that the agency’s determinations regarding observer coverage do not occur in a silo. Observer coverage has been an extensively litigated issue under the Magnuson-Stevens Fishery Conservation and Management Act of 1976 (“MSA”), 16 U.S.C. § 1801 et seq. Under the MSA, Congress created eight Regional Fishery Management Councils that monitor and oversee multiple fisheries in each region’s waters and develop and maintain fishery management plans, which must be approved by NMFS. See Oceana, Inc. v. Ross, 275 F. Supp. 3d 270, 275-81 (D.D.C. 2017). Pursuant to this authority, NMFS has approved a comprehensive standard bycatch reporting methodology (“SBRM”) that addresses fishery observer coverage across the Northeast Region. See id. at 277-81. In separate litigation, Oceana already has challenged the amended 2015 SBRM, and in that case, Judge Ellen Segal Huvelle denied Oceana’s motion for summary judgment and granted the government’s cross-motion for summary judgment, rejecting Oceana’s arguments that the 2015 SBRM violated the MSA, the Administrative Procedure Act, and the National Environmental Policy Act. See id. at 297. 8 In light of this existing comprehensive scheme for allocating observer coverage – which was upheld by Judge Huvelle and responds to numerous competing priorities and mandates requiring
8 The Court notes that Oceana filed its notice of appeal from that decision on October 17, 2017, to the United States Court of Appeals for the District of Columbia Circuit. The appeal is pending in that court, with case number 17-5247.
17
observer resources – the Court is not inclined to second-guess the agency’s determinations here. Rather, the Court will defer to the agency’s reasonable decision that increased observer coverage would not alter its broader take monitoring capabilities and will uphold its adoption of a five-year timetable for monitoring trawl takes.
C. Use of the Best Scientific and Commercial Data Available Finally, Oceana argues that NMFS failed to use the “best scientific and commercial data available” in developing the revised ITS because it “did not consider the important findings of a recent scientific study related to setting ITS levels.” See Pl. Resp. at 16. The Warden (2015) study describes how fisheries may have differential effects on species populations depending on the ages of the animals incidentally taken, explaining that “[f]or long-lived, slow-maturing species, like marine mammals, sea turtles, and sharks, even small increases in mortality in adult and sub-adult age classes can have disproportionately large population-level effects.” See Suppl. R. Ex. 1 at 30. The Warden (2015) study thus examines the impact of federally authorized incidental fishing mortality on a simulated Northwest Atlantic loggerhead sea turtle population. Having analyzed these simulations, the Warden (2015) concludes that “[i]t is important . . . for incidental takes to be specified and then monitored by life stage or by adult equivalents whenever possible,” because this results in a more reliable method for monitoring population-level impacts. See id. at 40. Oceana argues that NMFS irrationally failed to consider specifying incidental take by life stage or adult equivalents, despite this recent study being the best available science. See Pl. Resp. at 17.
In formulating a Biological Opinion, NMFS is required by statute to “use the best scientific and commercial data available.” See 16 U.S.C. § 1536(a)(2); see also 50 C.F.R.
18
§ 402.14(g)(8). The purpose of this provision is two-fold. First, the requirement that each agency use the best data available is meant “to ensure that the ESA not be implemented haphazardly, on the basis of speculation or surmise.” See Bennett v. Spear, 520 U.S. 154, 176 (1997). Second, the requirement also serves “to avoid needless economic dislocation produced by agency officials zealously but unintelligently pursuing their environmental objectives.” See id. at 176-77.
The requirement that NMFS “use the best scientific and commercial data available” precludes the agency from disregarding “scientifically superior evidence” or unreasonably relying on certain sources to the exclusion of other, better scientific evidence. See Las Vegas v. Lujan, 891 F.2d 927, 933 (D.C. Cir. 1989) (interpreting analogous language in 16 U.S.C. § 1533(b)(1)(A)). This requires that, even where data may be inconclusive, an agency must rely on the best available scientific information. See Sw. Ctr. for Biological Diversity v. Babbitt, 215 F.3d 58, 60 (D.C. Cir. 2000) (citing Las Vegas v. Lujan, 891 F.2d at 933) (interpreting analogous language in 16 U.S.C. § 1533(b)(1)(A)).
It is important to note, however, that the plain language of the provision requires NMFS only to use the best data available, not the best data possible. See also Am. Wildlands v. Kempthorne, 530 F.3d 991, 998-99 (D.C. Cir. 2008) (interpreting analogous language in 16 U.S.C. § 1533(b)(1)(A)); Sw. Ctr. for Biological Diversity v. Babbitt, 215 F.3d at 60-61; Colo. River Cutthroat Trout v. Salazar, 898 F. Supp. 2d at 208 (also interpreting analogous language in 16 U.S.C. § 1533(b)(1)(A)). Accordingly, under the “best available data” requirement, an agency has no obligation to conduct independent studies and tests to acquire the best possible data. See Am. Wildlands v. Kempthorne, 530 F.3d at 998; Sw. Ctr. for Biological Diversity v. Babbitt, 215 F.3d at 60. “Rather, [the ‘best available data’] provision ‘merely prohibits the
19
[agency] from disregarding available scientific evidence that is in some way better than the evidence [the agency] relies on.’” See Am. Wildlands v. Kempthorne, 530 F.3d at 998 (quoting Sw. Ctr. for Biological Diversity v. Babbitt, 215 F.3d at 60). And, of course, courts still must give great deference where an agency exercises its scientific expertise in interpreting and applying the best data available. See Cmtys. for a Better Env’t v. Envtl. Prot. Agency, 748 F.3d at 335-36; Colo. River Cutthroat Trout v. Salazar, 898 F. Supp. 2d at 208-09.
Here, the agency did not fail to consider or otherwise disregard the best available science. To the contrary, the agency explicitly acknowledged that the Northeast Fisheries Science Center (“NEFSC”) had recently released a peer-reviewed scientific publication concluding that monitoring of population-level impacts is more reliable if incidental takes are specified and monitored by life stage or by adult equivalents. See Notice of Completion Ex. 2 at 6. And NMFS made clear that it “do[es] not yet have sufficient information to apply that model to the scallop fishery,” although the agency is “currently working with the NEFSC on a methodology that may allow [it] to do so” in the future. See id.
This determination is not out-of-line with the agency’s statutory and regulatory obligations under the ESA. Although the Warden (2015) study may provide insights for developing future best practices, the agency adequately considered the study and reasonably determined that it would be premature to apply its general model to the specific circumstances of the scallop fishery. In fact, the Warden (2015) study itself explicitly cautions that its simulations of loggerheads are limited in several respects and the study “did not evaluate the probability that simulations were likely to mimic the true population.” See Suppl. R. Ex. 1 at 39-41. The agency’s assessment thus cannot be characterized as merely “disregarding” better, available scientific evidence. See Am. Wildlands v. Kempthorne, 530 F.3d at 998. To the contrary,
20
despite the fact that the agency has no obligation to conduct independent studies and tests to acquire new data, see id.; Sw. Ctr. for Biological Diversity v. Babbitt, 215 F.3d at 60, NMFS has noted that it is currently working to develop a methodology based on the Warden (2015) study for potential application to the scallop fishery, see Notice of Completion Ex. 2 at 6.
Finally, it was not arbitrary and capricious for the agency to decide not to delay issuance of its amended ITS until it could develop such a scallop fishery model based on the Warden (2015) study. The parties had agreed that the remand would be completed by May 1, 2015, less than six months after the Court’s opinion and order remanding the case, see Proposed Sched., and the Court accordingly issued an order approving this deadline. In light of this quick turn-around and the apparent benefits to both parties of promptly responding to the Court’s concerns, it was not arbitrary or capricious for the agency to issue the amended ITS while simultaneously working to analyze and potentially apply the model of the Warden (2015) study to the scallop fishery. And in doing so, the agency made clear that if it “determine[s] that a methodology exists that will provide . . . sufficient data to attribute the authorized take to a specific life stage or by adult equivalents, [the agency] will consider whether this provides a basis to reinitiate consultation . . . .” See Notice of Completion Ex. 2 at 6. Of course, because the agency will be required to amend the revised ITS with regard to the issue of dredge hours in any event, see supra Part III(A), NMFS may now find it appropriate for NMFS to incorporate the Warden (2015) study’s model and evaluate takes based on life stage. Or perhaps not. At this time, this matter is best left to the agency’s appropriate exercise of its expertise.
21
IV. CONCLUSION
For the foregoing reasons, the Court will remand to NMFS for a limited purpose of revising the ITS as it pertains to the use of dredge hours as a surrogate for loggerhead takes caused by dredge fishing. An order consistent with this opinion shall issue this same day.
SO ORDERED.
__________/s/____________ PAUL L. FRIEDMAN
United States District Judge DATE: August 17, 2018