UNITED STATES DISTRICT COURT FOR THE DISTRICT OF COLUMBIA
November 25, 2009
OCEANA, INC., PLAINTIFF,
GARY F. LOCKE, ET AL., DEFENDANTS.
The opinion of the court was delivered by: Ellen Segal Huvelle United States District Judge
MEMORANDUM OPINION AND ORDER
Plaintiff Oceana, Inc. ("Oceana") has filed for summary judgment on its claim that defendants Secretary of Commerce Gary F. Locke, the National Oceanic and Atmospheric Administration, and the National Marine Fisheries Service ("NMFS") (collectively, "the agency") violated the Magnuson-Stevens Fishery Conservation and Management Act, 16 U.S.C. §§ 1801-1884; the National Environmental Policy Act ("NEPA"), 42 U.S.C. §§ 4321-4370f; and the Administrative Procedure Act ("APA"), 5 U.S.C. § 706, through the issuance of a final rule regarding an amendment to the agency's standardized bycatch reporting methodology ("SBRM" or "the SBRM Amendment") for the thirteen federal fisheries in the northeastern United States. Exhibit 3 to plaintiff's summary judgment motion consists of a declaration and October 8, 2008 report from Dr. Murdoch McAllister. Defendants have moved to strike plaintiff's motion for summary judgment on the grounds that the materials comprising Exhibit 3 are not part of the administrative record and therefore are not properly before the Court. For the reasons set forth below, the Court grants defendants' motion.
In June 2007, defendants prepared a final draft of the proposed SBRM amendment. (See Administrative Record ["AR"] 2346-2983.)*fn1 The draft's Executive Summary explains that "[g]enerally, an SBRM can be viewed as the combination of sampling design, data collection procedures, and analyses used to estimate bycatch in multiple fisheries." (AR 2351.) The primary sources of information about fishery discards are "at-sea fishery observers" (i.e., scientists who board commercial fishing vessels to observe and record discards occurring on the trip), surveys of recreational fisheries, and reports from fishing vessel trips, which can be supplemented with information from other sources. (Id.; see also Pl.'s Mot. for Summ. J. ["Pl.'s SJ Mot."] at 7.) This information can be used to assess fishery stock and give scientific advice to fishery managers. (AR 2351.)
The SBRM was developed to "evaluat[e] the effectiveness of the allocation of fisheries observer effort across multiple fisheries to monitor a large number of species." (AR 2351.) The SBRM consists of seven principal components: mechanisms for collecting discard data; techniques for analyzing that data in order to allocate discard observers; a performance standard, known as a coefficient of variation ("CV"), for measuring the precision of the bycatch estimates; and four means of reviewing and refining the SBRM's effectiveness. (AR 2352, 2487 n.28.)
On September 24, 2007, the final day for public comment on both the proposed SBRM amendment and the proposed implementing regulations, the Lenfest Ocean Program submitted a report by Dr. McAllister that "focused primarily on the problem of bias in the SBRM's sampling design and statistical method."*fn2 (Pl.'s SJ Mot. at 9; see AR 3120-61 ("2007 McAllister Report").) McAllister's report expressed "serious reservations about the apparent low degree of scientific rigor in the determination of the SBRM." (AR 3124.) That same day, plaintiff submitted similar comments and also included a copy of the McAllister report. (AR 3186-3200 (Oceana Comments); see also AR 3188 & n.9.)
On October 18, 2007, Patricia Kurkul, the regional administrator of NMFS, received an analysis by Dr. Nancy Thompson, head of the NMFS's New England Fisheries Science Center ("Science Center"), of the 2007 McAllister Report. (See generally AR 3882-85.) Dr. Thompson concluded that although McAllister raised important issues, his criticisms did not "provide a sufficient basis to disapprove the SBRM Amendment." (AR 3882.) Thompson also stated that the Science Center would provide "a more detailed response" to the McAllister report "for inclusion in the response to public comments in the preamble to the final rule." (AR 3885.)
On October 22, 2007, the NMFS approved the SBRM amendment. (AR 3916; see also 3239-3880 (final SBRM amendment).) On December 20, Thompson presented Kurkul with her further analysis of McAllister's concerns. (See AR 3919-28.) Thompson's analysis provided "additional technical justification for the approaches" taken by the agency, summarized the additional work conducted since the first analysis, and reached similar conclusions as before. (AR 3919-20.) Thompson observed that "[i]n addition to the methods described in the SBRM" for measuring the potential biases in observer data, the agency's preferred "combined ratio" method of estimating discards was "validated" by a working paper authored by Paul Rago (and others) that was presented for peer review at an October 2007 Groundfish Assessment Review Meeting ("GARM"). (AR 3924-25; see AR 4713-49 (Rago paper).) Thompson also noted that the agency had conducted simulation studies which supported the SBRM's estimation methods, and she singled out a working paper by Chris Legault that was also presented for peer review at the GARM. (AR 3925-26; see AR 4750-61 (Legault paper).) The GARM committee characterized Dr. Legault's paper as concluding that the agency's combined ratio estimator was more "pragmatic" than the "direct estimator" advocated by McAllister, because McAllister's method relied upon data that is often unknown and whose estimates are not as "reliable" as the agency's preferred data estimates. (AR 3926.)
On January 28, 2008, the agency issued the final rule implementing the SBRM amendment. See 73 Fed. Reg. 4,736 (Jan. 28, 2008) (codified at 50 C.F.R. pt. 648). (See also AR 3952-74 (final rule).) The rule's preamble presented the agency's responses to a variety of public comments, including the 2007 McAllister report. (AR 3962-66.) On February 25, plaintiff initiated this lawsuit. On October 8, McAllister issued a report to plaintiff's counsel that reviewed the final SBRM rule, entitled "Follow-up review of the NMFS' Standardized Bycatch Reporting Methodology: is something fundamentally wrong with the SBRM approach which is likely to result in serious error?" (See Pl.'s SJ Mot., Ex. 3 at 34-76 ("2008 McAllister Report").)
On January 5, 2009, plaintiff moved to compel the inclusion of certain documents (but not the 2008 McAllister Report) in the administrative record. See generally Oceana, Inc. v. Locke ("Oceana III"), 634 F. Supp. 2d 49 (D.D.C. 2009) (affirming magistrate judge's denial of plaintiff's motion to compel inclusion), aff'g No. 08-CV-318, 2009 WL 1491516 (D.D.C. May 28, 2009). On September 25, plaintiff filed its summary judgment motion and several supporting exhibits. Exhibit 3 to that motion consisted of the 2008 McAllister Report, an eleven-page declaration by McAllister, and his curriculum vitae (collectively, "the McAllister testimony"). (See Pl.'s SJ Mot., Ex. 3 at 1-11 ("McAllister Declaration").) In a footnote, plaintiff's motion argues that the Court may consider the McAllister testimony under Esch v. Yeutter, 876 F.2d 976 (D.C. Cir. 1989). (See Pl.'s SJ Mot. at 12 n.6.) In Esch, the D.C. Circuit observed the general rule that "judicial review of agency action is normally to be confined to the administrative record," but also observed that in eight situations, courts have departed from the general rule and permitted the introduction of extra-record information. 876 F.2d at 991 & n.166. Plaintiff contends that the McAllister testimony falls under three of the eight situations described in Esch, because (1) "it provides background information to help the court understand the complex issues before it" relating to the precision of the agency's discard estimation methods; (2) "it addresses relevant factors that the agency failed to consider" relating to statistical bias in the agency's discard estimation methods; and (3) "it highlights environmental consequences and reasonable alternatives neglected in the agency's [environmental assessment] . . . ." (Pl.'s SJ Mot. at 12 n.6 (citing Esch, 876 F.2d at 991).)
On October 15, 2009, defendant moved to strike the McAllister testimony, arguing that even if plaintiff's footnote were the proper vehicle for seeking its admission, (1) Esch's discussion of exceptions to the administrative record rule is not controlling law, and (2) even if it were, plaintiff "'must first establish that the agency acted in bad faith or otherwise behaved improperly, or that the record is so bare that it prevents effective judicial review.'" (Defs.'s Mot. to Strike at 5 (quoting County of San Miguel v. Kempthorne, 587 F. Supp. 2d 64, 79 (D.D.C. 2008)) (emphasis omitted).) Plaintiff responds that Esch correctly describes the state of the law in this Circuit, and that no showing is required before it may submit McAllister's testimony. In the alternative, plaintiff contends that the agency committed "procedural irregularities" in promulgating the SBRM rule which constitute "bad faith or improper behavior" that would justify admission of the testimony. (See Pl.'s Opp'n to Mot. ["Pl.'s Opp'n"] at 10-11 (citing Fund for Animals v. Williams, 391 F. Supp. 2d 191, 199 (D.D.C. 2005).)
I. APPLICABLE STANDARD
Judicial review of agency actions "is confined to the full administrative record before the agency at the time the decision was made." Camden County Council on Econ. Opportunity v. U.S. Dep't of Health & Human Servs., 563 F. Supp. 2d 262, 265 (D.D.C. 2008) (citing Envtl. Def. Fund, Inc. v. Costle, 657 F.2d 275, 284 (D.C. Cir. 1981)). "When reviewing that record, the Court considers 'whether the agency has considered the relevant factors and articulated a rational connection between the facts found and the choice made.'" Id. (quoting Jifry v. FAA, 370 F.3d 1174, 1180 (D.C. Cir. 2004)).
The parties dispute the legal principles governing the Court's potential consideration of the McAllister testimony. As recently explained by Chief Judge Lamberth in The Cape Hatteras Access Preservation Alliance v. U.S. Department of Interior, some of that confusion is attributable to the case law's discussion of "supplementing" an administrative record in two analytically distinct situations. See No. 09-CV-236, 2009 WL 3683120, at *2 (D.D.C. Nov. 4, 2009); see also Pac. Shores Subdivision v. Army Corps of Eng'rs, 448 F. Supp. 2d 1, 6 (D.D.C. 2006). One situation involves a claim "that some information that should have properly been included in the administrative record was not," while the other involves a request for judicial consideration of evidence that exists apart from the administrative record, on the theory that if the Court did not consider that evidence, "reviewing agency action would be unnecessarily difficult." Cape Hatteras, 2009 WL 3683120, at *2; see also Pac. Shores, 448 F. Supp. 2d at 6 (distinguishing "seek[ing] to include evidence in the record" from "moving the Court to review evidence outside of or in addition to the administrative record"). Plaintiff's submission of the McAllister testimony falls into the latter category of "extra-record" evidence, because those documents were not before the agency when it issued the final rule.
This Court has previously observed that Esch's discussion of eight exceptions to the general rule regarding consideration of extra-record evidence was dicta. See Oceana II, 384 F. Supp. 2d at 218 n.17. Upon further consideration of Chief Judge Lamberth's opinion in Cape Hatteras, see 2009 WL 3683120, at *3-*4, this Court is persuaded that Esch should be read narrowly. This result is consistent with the D.C. Circuit's decision in IMS, P.C. v. Alvarez, which only identified four "accepted exceptions to the principle that the court cannot consider information that falls outside the agency record." 129 F.3d 618, 624 (D.C. Cir. 1997); see Cape Hatteras, 2009 WL 3683120, at *4 (same). There, the Court of Appeals rejected the proposed submission of extra-record affidavits because the plaintiff had neither "made the 'strong showing of bad faith or improper behavior' required to justify" considering extra-record evidence, IMS, 129 F.3d at 624 (quoting Citizens to Preserve Overton Park v. Volpe, 401 U.S. 402, 420 (1971)), nor shown that "the agency failed 'to explain administrative action [so] as to frustrate effective judicial review'" by failing "to examine all relevant factors" or "to adequately explain its grounds for decision." Id. (quoting Camp v. Pitts, 411 U.S. 138, 142-43 (1973)). Accordingly, the Court will only permit the submission of the McAllister testimony if plaintiff has shown that the agency (1) acted in bad faith in reaching its decision, (2) engaged in improper behavior in reaching its decision, (3) failed to examine all relevant factors, or (4) failed to adequately explain its grounds for decision. See id.
II. THE MCALLISTER TESTIMONY WILL BE STRICKEN
A. Plaintiff Has Not Demonstrated the Agency's Bad Faith or Improper Behavior
As an initial matter, the Court is not persuaded by plaintiff's contention that defendants engaged in "procedural irregularities" that constitute evidence of "bad faith or improper behavior" sufficient to justify admission of the McAllister testimony. (See Pl.'s Opp'n at 10.)
Plaintiff suggests that the agency's procedural decision not to prepare an Environmental Impact Statement ("EIS") demonstrates defendants' bad faith. (Pl.'s Opp'n at 10-11.) The Court rejects this argument. While plaintiff is free to challenge the legal merits of that decision, the mere fact of that decision, without any specific allegations of impropriety, does not show bad faith. Cf. Fund for Animals, 391 F. Supp. 2d at 198 (permitting supplementation of record where plaintiffs made prima facie showing that agency knowingly excluded relevant and adverse information from record).
Plaintiff also argues that the agency "chiefly relie[d]" upon the Rago and Legault working papers, which were neither completed nor made public "until after the close of the comment period and after the agency had made its final decision," thereby depriving Oceana of an "opportunity to review or respond to these studies on the record." (Pl.'s Opp'n at 10; see also Pl.'s SJ Mot. at 29-30.) The Court disagrees. Based upon its review of the record, the Court concludes that neither the SBRM amendment nor its implementing rule "chiefly rel[y]" on the Rago and Legault papers.
First, the SBRM amendment does not cite either the Rago or Legault working paper.*fn3 Nor could it have done so. The agency approved the amendment on October 22, 2007, before those two papers were completed on October 24 and later presented for peer review at the GARM. (See Def.'s Notice of Filing Am. Supplement to Admin. R., Ex. 1 (Supplemental AR Index) at 2 (listing October 24, 2007 date for both papers).)
Second, the agency's final rule, issued in January 2008, cites fourteen peer-reviewed publications by name when responding to McAllister's concerns, but not the Rago or Legault working papers, which are only discussed indirectly by reference to their consideration during the October 2007 GARM. (See AR 3962-66 & nn.1-15.) Rago's findings are cited merely as "another" way to validate the SBRM's methods, "[i]n addition to the methods described in the SBRM" (AR 3965), and Legault's simulation study of six discard estimation methods is referenced as one of multiple simulation studies. (See AR 3965 ("[S]imulation tests of alternative estimators have been conducted for several species. . . . NMFS has also conducted studies to estimate total landings from the observed sample data and have found good agreement for the methods used in the SBRM." (emphasis added).)
Accordingly, the Court concludes that plaintiff has not made the "strong showing" of bad faith or improper behavior required to justify admission of the McAllister testimony.
B. McAllister's Recent Criticisms of the Agency's Quantitative Analyses Are Inadmissible
1. "Background Information" on "Complex Issues"
Plaintiff argues that the McAllister testimony provides helpful background information regarding complex quantitative issues relating to the precision of the agency's discard estimation methods and the Rago and Legault papers. The Court construes this argument as one that the agency failed "to adequately explain its grounds for decision." IMS,129 F.3d at 624. Nonetheless, plaintiff has not shown that the agency's rationale cannot be understood because of the technical issues involved. On the contrary, there are a number of instructive discussions in the SBRM amendment, the 2007 McAllister Report, and the final SBRM rule itself. If, at a later date, the Court concludes that the record does not contain sufficient explanations, it may then be appropriate for the Court to seek clarification from the agency, "'either through affidavits or testimony, such additional explanations of the reasons for the agency decision as may prove necessary.'" Envtl. Def. Fund, 657 F.2d at 285 (quoting Pitts, 411 U.S. at 143). However, these "new materials should be merely explanatory of the original record and should contain no new rationalizations" for the agency's decision. Id. "If the agency action, once explained by the proper agency official, is not sustainable on the record itself, the proper judicial approach has been to vacate the action and to remand the matter back to the agency for further consideration." Id.
In the alternative, "[a]lthough the subject matter of this case does have a highly technical aspect," the McAllister testimony is not "primarily explanatory in nature." Corel Corp. v. United States, 165 F. Supp. 2d 12, 31 (D.D.C. 2001). It is primarily "argumentative," and thus inappropriate for the Court to consider. Id. The title of McAllister's 2008 report "quite accurately encapsulates the purpose and nature of his submissions," id., since it asks: "[I]s something fundamentally wrong with the SBRM approach which is likely to result in serious error?" (2008 McAllister Report at 1 (italics added).) The report itself "attack[s] the merits" of the agency's analytical choices and conclusions and then "conduct[s] [its] own . . . analysis," from which McAllister concludes that the agency should have chosen different inputs when developing the SBRM. Corel, 165 F. Supp. 2d at 31. (See, e.g., 2008 McAllister Report at 12 ("The NMFS' replies were examined, classified according to the specific points in the [2007 McAllister Report] that they address and are summarized below. We evaluate their substance in the context of the SBRM and of the terms of reference provided by Oceana." (emphasis added)).)*fn4
The mere fact that McAllister disputes the agency's responses to his initial comments and "reiterate[s] scientific criticisms" already in the record (Pl.'s Opp'n at 3) does not transform the reasons for his disagreement into relevant "background information." Nor is this Court's prior opinion in Oceana II to the contrary. (See id. at 12.) There, the Court denied a motion to strike plaintiff's submission of a post-decisional extra-record letter by a scientist who was one of the "original developers" of the analytical model used by the agency in its challenged action and who criticized the agency's reliance upon her work. 384 F. Supp. 2d at 217-18 & n.17. Where a scientist challenges the manner in which an agency has relied upon her own research, her unique familiarity with the meaning of that research can constitute "particularly relevant" background information about the basis for the agency's decision. See id. (internal quotation marks omitted); cf. Carlton v. Babbitt, 26 F. Supp. 2d 102, 108, 111 (D.D.C. 1998) (ruling that agency should have considered scientist's declaration, which was submitted prior to agency's decision, because agency ultimately relied solely upon that scientist's research article for certain conclusions, yet his declaration "explicitly disclaim[ed] the [agency's] optimistic reading of his article," making "his understanding of his own article . . . particularly relevant"). But that is not the case here; McAllister's testimony about the agency's discard estimation method is "offered primarily to attack the propriety of the challenged agency action, [and therefore, it] will be stricken." Corel, 165 F. Supp. 2d at 32.
2. Agency's Failure to Consider "Relevant Factors"
Plaintiff suggests that the McAllister testimony "addresses relevant factors that the agency failed to consider" relating to the accuracy of the agency's discard estimation methods. (Pl.'s SJ Mot. at 12 n.6.) Specifically, McAllister asserts that the agency failed to consider how systematic biases in discard estimates and their variance could significantly skew the number and allocation of bycatch observers that the agency deems necessary to achieve its chosen 30 percent CV measure of performance. (McAllister Decl. ¶ 23.) However, bias is hardly a "factor not considered" by the agency. (Id.) To the contrary, Chapter 5 of the draft SBRM amendment, entitled "Sampling Design and Estimation of Precision and Accuracy" (see AR 3381-3436), thoroughly discusses issues of bias. (See, e.g., AR 3383 (citing "non-representative sampling" and "statistical properties of the consistency of the estimators" as "two primary potential sources of bias in a sampling program such as the at-sea observer program"); AR 3411 ("Several analytical tests were conducted to evaluate the potential sources of bias in the 2004 observer data.").) McAllister's chief criticism actually appears to be that the agency concluded "that neither the data nor the statistical estimators showed significant bias . . . ." (2008 McAllister Report at 25.) He argues that the evidence which the agency invoked "to support the hypothesis of no bias" was "far from . . . conclusive" and "could actually be used to reject that same hypothesis." (Id.) Thus, McAllister's testimony as to bias "does not add factors that [the agency] failed to consider as much as it questions the manner in which [the agency] went about considering the factors it did." Corel, 165 F. Supp. 2d at 31-32.
Plaintiff also argues that the McAllister testimony should be considered because it purportedly raises issues that the agency "neglected" during its NEPA analysis (Pl.'s SJ Mot. at 12 n.6), which the Court will construe as a claim that the agency failed "to examine all relevant factors." IMS,129 F.3d at 624.*fn5 Specifically, plaintiff claims that the agency failed to account for "cumulative impacts" of the SBRM amendment "related to bias and error over time . . . ." (Pl.'s Opp'n at 16; see also Pl.'s SJ Mot. at 31.) "Cumulative impact" is one of the factors to be considered in determining whether a proposed action will significantly affect the environment and is defined as "the impact on the environment which results from the incremental impact of the action when added to other past, present, and reasonably foreseeable future actions regardless of what agency (Federal or non-Federal) or person undertakes such other actions." 40 C.F.R. § 1508.7 (emphasis added). The McAllister Declaration contends that the SBRM's 30 percent CV performance standard is too imprecise to accurately detect changes in discard levels, and that this imprecision could prevent the agency from adequately managing discards over time. (McAllister Decl. ¶¶ 27-29.)
Plaintiff mischaracterizes the agency's treatment of the cumulative impact issue. The final SBRM rule notes that Section 7.3 of the SBRM amendment "explicitly provides a discussion of the expected cumulative effects associated with the action." (AR 3960.) Section 7.3 explains that the SBRM amendment is an administrative measure focusing on processes for collecting and analyzing existing levels of bycatch, and that the establishment of an SBRM performance standard was a change from the status quo, which lacked such a standard. (AR 3514.) The agency concluded that because the performance standard was a "purely administrative feature," it would not presently affect the environment any differently than the status quo's baseline. (AR 3514-15.) Nor would the "CV-based performance standard," on its own, have any cumulative impact over time, because the SBRM amendment would not implement any changes to fishing operations. (See AR 3515.) Such operational changes would only be implemented if the SBRM yielded evidence that discard levels were too high and the agency subsequently took "a new management action" to reduce bycatch. (AR 3516.) The agency observed that such a management action was not "reasonably foreseeable," and that it was therefore "not practicable to conduct a NEPA analysis on these potential impacts at this time." (Id.) See also Izaak Walton League, 655 F.2d at 377 ("NEPA does not require federal agencies to examine every possible environmental consequence. Detailed analysis is required only where impacts are likely.").
The agency considered the issue of cumulative impact and concluded that no cumulative impact was posed by the SBRM amendment and its performance standard. McAllister's attack upon the supposedly insufficient precision of the 30 percent CV merely re-argues plaintiff's original criticism that "the less precise the methodology, the greater the risk to the environment." (AR 3198 (Oceana Comments).) Thus, the testimony does not offer "proof of an environmental factor inappropriately excluded from consideration by the assessing agency." The Humane Soc. of U.S. v. Dep't of Commerce, 432 F. Supp. 2d 4, 15 (D.D.C. 2006).*fn6
For the aforementioned reasons, the Court grants defendants' motion to strike [Dkt. 36] plaintiff's motion for summary judgment. Plaintiff may re-file its motion for summary judgment on or before December 15, 2009, but it may not include the 2008 McAllister Report or the McAllister Declaration.