UNITED STATES DISTRICT COURT FOR THE DISTRICT OF COLUMBIA

NATURAL RESOURCES DEFENSE COUNCIL, INC., et al.,

Plaintiffs,

v. Case No. 20-cv-1150 (CRC)

PAUL DOREMUS, in his official capacity as Acting Assistant Administrator for Fisheries, et al.,

Defendants.

MEMORANDUM OPINION

Pursuant to its duties under the Endangered Species Act (“ESA”) and a previous court

order, the National Marine Fisheries Service (“NMFS”) recently undertook a status review of

two species of river herring and determined that listing the species as endangered or threatened

was not warranted. Natural Resources Defense Council, Inc. and other plaintiffs brought this

action under the Administrative Procedure Act (“APA”) and the ESA, challenging that listing

decision. To facilitate the Court’s review, NMFS compiled an administrative record consisting

of materials the agency considered in deciding not to list the species.

Plaintiffs now move to compel the NMFS to add four categories of documents to the

administrative record. The Court agrees with plaintiffs that one document—the Supporting

Information section of a scientific paper by Dr. Jonathan Hare and others—should be included in

the administrative record because it was substantively cited in the agency’s status review report,

a key document on which the listing decision relied. As to the other documents at issue,

plaintiffs have not carried their burden to compel completion or supplementation of the

administrative record. I. Background

In 2013, NMFS decided not to list two fish species, the blueback herring and the alewife,

as threatened or endangered. Endangered and Threatened Wildlife and Plants,

(Aug. 12, 2013). Another court in this district vacated and remanded that decision in

2017. Nat. Res. Def. Council, Inc. v. Rauch,

(D.D.C. 2017).

Following the remand, the NMFS re-initiated a status review of the two fish species.

, 38,673 (Aug. 15, 2017). To do so, NMFS formed a status review team

(“SRT”) composed of scientific experts from NMFS and other federal and state agencies. See

Endangered Species Act Listing Determination for Alewife and Blueback Herring,

, 28,631 (June 19, 2019) (“2019 Listing Determination”). The SRT “considered a variety

of scientific information from the literature, unpublished documents, and direct communications

with researchers working on alewife and blueback herring, as well as technical information

submitted to NMFS,” and prepared a status review report based on its work.

. The

status review report underwent independent review and was then finalized.

NMFS then

considered the status review report as part of its decision-making process and concluded that

listing either fish species was not warranted. See

.

Plaintiffs filed this lawsuit challenging the listing decision in May 2020. 1 In August

2020, the government filed the index to the administrative record and served a copy of the

administrative record on plaintiffs. See Notice of Filing Certified List of the Administrative

1 The first-named defendant in the Complaint is Chris Oliver, who was then serving as Assistant Administrator for Fisheries at NMFS. Paul Doremus, the current Acting Assistant Administrator for Fisheries, is automatically substituted for former Assistant Administrator Oliver pursuant to Federal Rule of Civil Procedure 25(d).

2 Record, ECF No. 20; Index to Administrative Record, ECF No. 20-1. Plaintiffs followed with

the present motion, which is now fully briefed.

II. Legal Standard

When reviewing agency action under the APA, the Court must review “the whole

record.”

. The whole record consists of “all documents and materials that the

agency directly or indirectly considered, no more and no less.” Oceana, Inc. v. Ross,

(D.D.C. 2018) (“Oceana I”) (Cooper, J.) (quoting Maritel, Inc. v. Collins,

(D.D.C. 2006)). The agency is responsible for compiling the administrative

record and is entitled to a “strong presumption of regularity in having done so.” Marcum v.

Salazar,

(D.D.C. 2010).

That said, an agency “may not skew the record in its favor by excluding pertinent but

unfavorable information.” Fund for Animals v. Williams,

(D.D.C.

2005). Nor may an agency exclude information that it considered on the grounds that the

information was not relied upon in its final decision. Ad Hoc Metals Coal. v. Whitman,

(D.D.C. 2002). “A complete administrative record should include all

materials that might have influenced the agency’s decision, and not merely those on which the

agency relied in its final decision.” Amfac Resorts, L.L.C. v. U.S. Dep’t of the Interior,

(D.D.C. 2001) (internal quotation marks omitted). “[I]f the agency decisionmaker

based his decision on the work and recommendations of subordinates, those materials should be

included as well.” Id.; see also Miami Nation of Indians of Ind. v. Babbitt,

(N.D. Ind. 1996) (“[A] document need not literally pass before the eyes of the final agency

decision maker to be considered part of the administrative record.” (quoting Clairton

Sportsmen’s Club v. Pa. Turnpike Comm’n,

(W.D. Pa. 1995)).

3 “Courts in this District have long held that materials that fall within the scope of the

deliberative process privilege are not part of the administrative record.” Oceana I,

. Whether the deliberative process privilege is asserted in an APA case or in the

Freedom of Information Act context, the test for the privilege’s applicability is the same.

. To qualify for the privilege, a document must be both “predecisional” and “deliberative.”

(quoting In re Sealed Case,

(D.C. Cir. 1997)).

A party seeking to supplement the administrative record may do so in two ways. First, a

party may seek to include “evidence that should have been properly a part of the administrative

record but was excluded by the agency.” Oceana I,

(quoting Univ. of

Colo. Health at Mem’l Hosp. v. Burwell,

(D.D.C. 2015)). This pathway is

available regardless of whether the agency initially excluded the materials “by design or

accident.” Univ. of Colo. Health v. Azar,

, 200 (D.D.C. 2020) (quoting

Marcum,

). To overcome the presumption of regularity, a plaintiff “must

only ‘put forth concrete evidence’ and ‘identify reasonable, non-speculative grounds for [her]

belief that the documents were considered by the agency and not included in the record.’”

Oceana I,

-79 (quoting Charleston Area Med. Ctr. v. Burwell,

(D.D.C. 2016)).

Second, a party may seek to include “‘extra-judicial evidence that was not initially before

the agency’ but that the plaintiff ‘believes should nonetheless be included in the administrative

record.’”

(quoting Univ. of Colo. Health at Mem’l Hosp., 151. F. Supp. 3d at 13). To

supplement the record in this way, the plaintiff must “‘demonstrate unusual circumstances

justifying a departure from th[e] general rule’ against considering extra-record evidence.”

at

77 (quoting City of Dania Beach v. F.A.A.,

(D.C. Cir. 2010)). The D.C.

4 Circuit has recognized three such unusual circumstances: “(1) if the agency ‘deliberately or

negligently excluded documents that may have been adverse to its decision, (2) if background

information [is] needed ‘to determine whether the agency considered all the relevant factors,’ or

(3) if the ‘agency failed to explain administrative review so as to frustrate judicial review.’”

at 77-78 (quoting City of Dania Beach,

).

III. Analysis

Plaintiffs ask the Court to order NMFS to complete the record by including four

categories of documents: (1) the complete version of a scientific paper by Dr. Jonathan Hare and

others, including the paper’s Supporting Information section; (2) certain materials created by the

SRT in preparation of the status review report, including scoresheets prepared by individual SRT

members; (3) three summaries of telephone conferences attended by SRT members and members

of the Atlantic States Marine Fisheries Commission; and (4) external emails between NMFS

staff and other parties relevant to the listing decision. The Court will grant plaintiffs’ motion as

to the first category and deny it as to the rest.

A. Supporting Information from the Hare et al. Scientific Article

Plaintiffs request that the Court compel the inclusion of the “Supporting Information”

materials for the scientific paper by Hare et al. See Mem. in Supp. of Mot. to Compel at 4-7.

Although the main text of the paper is already in the record, see ECF No. 22-4, plaintiffs contend

that the Supporting Information was also considered by the agency and therefore must be

included as well. See Reply at 8. The Court agrees.

The Court considered a similar situation in Oceana I. See

. In that

case, the plaintiff challenged NMFS’s adoption of an amendment to an NMFS fishery

management plan for a species of shark.

. The plaintiff sought to supplement the

5 administrative record with studies and reports that were cited in the final agency environmental

impact statement (“EIS”) for the amendment.

. The Court held that documents that

were cited substantively in the EIS were considered by the agency and thus belonged in the

administrative record.

Documents that are “included in the list of references appended to the

end of each EIS chapter” also belonged in the record, while sources that were cited merely as a

source of further information did not need to be included.

.

The status review report in this case, like the EIS in Oceana I, is an agency-generated

scientific document that constitutes part of the basis for the ultimate agency decision. Thus,

Oceana I’s analysis of documents based on their “treatment[] in the EIS,”

, provides a

useful framework for deciding whether the Hare paper’s Supporting Information should be

included in the administrative record based on its treatment in the status review report.

The status review report discusses the Hare paper in detail and cites the paper

substantively on multiple occasions. See, e.g., Status Review Report at 18-20, 104, 128, ECF

No. 25-2. Accordingly, there is no dispute that the main text of the Hare paper belongs in the

record, and indeed, it is already included. See ECF No. 22-4. Only the Supporting Information

is in dispute.

The status review report cites the Hare paper simply as “Hare et al. 2016a,” without

pinpoint citations to clarify whether the report is referring to the paper’s main text or its

Supporting Information. Substantively, however, some of the citations refer to information that

comes from the Supporting Information. In particular, the report cites the Hare paper for its

discussion of “the importance of the alewife’s and blueback herring’s high spawning site

fidelity.” Status Review Report at 19. This information cannot be found in the main text of the

paper. See ECF No. 22-4 (lacking any discussion of spawning site fidelity). It is, however,

6 included in the Supporting Information. See Supporting Information at 5, ECF No. 22-5

(“Alewife have a relatively high degree of spawning site fidelity[.]”);

(“Blueback

Herring have a relatively high degree of spawning site fidelity[.]”).

Thus, the plaintiffs have successfully put forth concrete, non-speculative evidence for

their belief that the Supporting Information—not only the main text—was “referred to,

considered by, or used by [the agency] before it issued its final rule.” Ad Hoc Metals Coal.,

. The Supporting Information was “use[d] to justify an assertion or

proposition,” which “indicates that the [agency] consulted and thought about—and therefore

considered—that document.” Oceana I, 290 F. Supp. at 80. Accordingly, the Court will order

NMFS to include it in the administrative record.

B. Individual Scoresheets Created by the Status Review Team

Plaintiffs next seek inclusion of SRT members’ individual scoresheets, as well as written

comments explaining those scoresheets. These documents were properly excluded from the

administrative record as deliberative.

As part of the process leading to the listing decision, SRT members created documents

that showed the scores each member assigned to various risks facing the two fish species and

documented the rationales behind the members’ scoring decisions. Anderson Decl. ¶¶ 14-15,

ECF No. 25-1. Plaintiffs contend that these materials are “a key part of the basis of NMFS’s

ultimate determinations” and thus should be included in the record. Mem. in Supp. of Mot. to

Compel at 7. 2

2 Plaintiffs argue only that the scoresheets were before the agency but not included in the record; they make no argument to admit the scoresheets through the second method of supplementation. See Mem. in Supp. of Mot. to Compel at 7-10; Reply at 9-15.

7 In response, the government emphasizes that the SRT members’ individual scores were

aggregated and were considered as a mean score for the SRT’s scientific opinion and, ultimately,

for the listing decision. See Status Review Report at 119 (“Based upon the available information

summarized here and the SRT’s individual expert opinions, considered as a mean score, alewife

have a low risk of extinction[.]” (emphasis added)); id. at 144 (same for blueback herring);

Anderson Decl. ¶ 23 (“The aggregate team score is what NMFS considered in making its listing

determination.”). Thus, the government argues, the listing decision was not based on the

individual scores but was instead based on the mean score alone. Opp’n at 11.

The Court need not determine whether the individual scoresheets and comments were

considered by the agency, because even if they were, they are protected by the deliberative

process privilege. To be covered by that privilege, a document must be both “predecisional” and

“deliberative.” Oceana I,

(quoting In re Sealed Case,

).

At the threshold, it is clear—and undisputed—that the individual scoresheets and

comments are predecisional. See Reply at 13 (conceding that the scoresheets are predecisional).

The Court is also persuaded that the scoresheets and comments are deliberative. Based

on unrebutted evidence provided by the government, these materials would reveal the range of

views expressed by SRT members in their deliberations over what conclusions the team should

draw regarding threats to the river herring. See Anderson Decl. ¶¶ 15, 18, 21-22.

Plaintiffs argue that the materials are not deliberative, relying in part on Petroleum

Information Corp. v. U.S. Dep’t of the Interior,

(D.C. Cir. 1992). In Petroleum

Information Corp., plaintiffs sought records from the Bureau of Land Management’s computer

data bank.

. The D.C. Circuit held that those records were not protected by

the deliberative process privilege, which “is centrally concerned with protecting the process by

8 which policy is formulated.”

(emphasis in original). The Circuit emphasized three

factors that weighed against applying the deliberative process to the BLM records:

(1) the information within the data bank was all publicly available; (2) the information within the documents was not associated with a significant policy decision since the Bureau’s mission was ‘essentially technical and facilitative’ in that its ‘task [was] to organize public records in a more manageable form’; and (3) the materials were unlikely to diminish official’s candor or injure the quality of agency decisions because the materials did not ‘embody agency judgments’—it was a record-keeping task.

Urb. Air Initiative, Inc. v. Env’t Prot. Agency,

(D.D.C. 2017) (quoting

Petroleum Info. Corp.,

).

Plaintiffs offer Petroleum Information Corp. for the proposition that the deliberative

process privilege “does not cover non-policy discussions regarding science and facts.” Reply at

3. However, the application of the privilege does not depend on a “fact/opinion” dichotomy.

Oceana I,

. Even when material is “purely factual in nature,” it may be

privileged if “the selection or organization of facts is part of an agency’s deliberative process.”

(quoting Ancient Coin Collectors Guild v. U.S. Dep’t of State,

(D.C. Cir.

2011)).

Nor does Petroleum Information Corp. imply that materials produced in the process of

making a scientific determination cannot be deliberative. Recent cases from this district have

rejected that overbroad reading of Petroleum Information Corp. For example, in Center for

Biological Diversity v. U.S. Environmental Protection Agency, the EPA asserted the deliberative

process privilege to justify its withholding of records it had generated during its effort to

determine, pursuant to Section 7(a)(2) of the ESA, whether a pesticide was likely to jeopardize

endangered species or their habitats. See

(D.D.C. 2019). The court

upheld this withholding, explaining that although “EPA’s decision under Section 7(a)(2) . . .

9 depends on a scientific assessment[,] this does not divest the agency’s decisionmaking process of

eligibility for [deliberative process] protection.”

. The court further noted that this result

was “consistent with well-established law in this Circuit that the deliberative process privilege

operates to shield from disclosure agency decision-making reflecting the collection, culling and

assessment of factual information or scientific data.”

(collecting cases).

Similarly, in Urban Air Initiative, Inc., the EPA conducted a scientific study to help

produce an updated fuel emissions model. See

. Plaintiffs in that case,

relying on Petroleum Information Corp., argued for the disclosure of predecisional records about

the EPA study.

. The court found that the documents were privileged, in part because

none of the three factors that motivated Petroleum Information Corp. applied to the documents at

issue.

An analysis of the individual scoresheets and comments in this case comes out the same

way. First, it cannot be said that the information contained in the individual scoresheets and

comments is all “publicly available.” Petroleum Info. Corp.,

. To the contrary,

the scoresheets and comments were generated solely to inform the status review, and nothing

suggests they were disseminated beyond the SRT. Second, the SRT’s work was not merely an

organizational effort but was associated with the agency’s policy decision from the beginning.

The SRT was convened for the purpose of providing a scientific basis upon which the agency

could make its listing decision, see 2019 Listing Determination, 84 Fed. Reg. at 28,631, and it

did far more than “organize public records in a more manageable form,” Petroleum Info. Corp.,

. Third, release of the individual scoresheets could inhibit scientists from

speaking freely in future endeavors similar to the SRT. SRT Members were assured their

opinions would remain confidential, an assurance the NMFS deemed necessary because some

10 SRT members were concerned that their professional judgments might be inconsistent with the

official positions of their employers. See Anderson Decl. ¶ 25.

All told, the individual scoresheets and comments reflect the “collection, culling, and

assessment of factual information or scientific data” in the process of drawing scientific

conclusions for policymaking purposes. Ctr. for Biological Diversity,

.

The deliberative process privilege therefore applies.

C. AMFSC Call Summaries

Plaintiffs next seek to supplement the record by including three summaries created by the

Atlantic States Marine Fisheries Commission (“ASMFC”) of conference calls attended by

NMFS Status Review Team members. Mem. in Supp. of Mot. to Compel at 10-14. “The

ASMFC is an interstate compact organization that manages fishing on certain Atlantic coast fish

populations including alewife and blueback herring.”

at 10 (citing ASMFC, Shad & River

Herring, http://www.asmfc.org/species/shad-river-herring). The ASMFC’s 2017 River Herring

Stock Assessment Update was considered by NMFS in its listing decision and is already

included in the record. See 2017 Update Vol. I, ECF No. 25-6; 2017 Update Vol. II, ECF No.

25-7. Several of the SRT members were also members of the ASMFC subcommittee that

produced the 2017 Update. Compare Status Review Report at iii (listing SRT members) with

2017 Update Vol. I (listing subcommittee members in front matter).

Plaintiffs contend that the call summaries must also be part of the administrative record

because they were in the possession of SRT members and “give context to” the report’s findings.

Mem. in Supp. of Mot. to Compel at 12-13. However, plaintiffs have not set forth concrete, non-

speculative evidence that these call summaries were considered by the agency. Instead, plaintiffs

argue that these documents must have been “at least indirectly considered” because some SRT

11 members participated in the ASMFC 2017 Update process, attended the conference calls, and

had access to the call summaries. See

.

This bare assertion cannot support a finding of indirect consideration. “Documents and

materials indirectly considered by agency decision-makers are those that . . . were so heavily

relied on in the recommendation that the decisionmaker constructively considered them.”

GeorgiaCarry.org v. U.S. Army Corp of Eng’rs,

(N.D. Ga. 2016)

(citation omitted). Plaintiffs point to no specific instance in the record where the ASMFC calls

are referenced or appear to be relied on, let alone the heavy reliance needed to find indirect

consideration.

In the alternative, plaintiffs argue the call summaries should be added to the record via

the second method of supplementation. Plaintiffs argue the summaries are “needed to determine

whether the agency considered all the relevant factors.” See Mem. in Supp. of Mot. to Compel at

13 (quoting Oceana I,

). 3 But to add a document to the record on that

basis, the Court must find that the document does “more than raise ‘nuanced points’ about a

particular issue; it must point out an ‘entirely new’ general subject matter that the defendant

agency failed to consider.” Oceana, Inc. v. Ross,

, 70 (D.D.C. 2020)

(“Oceana II”) (Cooper, J.) (quoting Pinnacle Armor, Inc. v. United States,

(E.D. Cal. 2013)). In cases involving complicated, scientific analyses, “intermediary

3 Plaintiffs also suggest that the call summaries were “negligently excluded” from the record as “documents that may be adverse to [NMFS’s] decision.”

(quoting Oceana I,

). This argument fails because plaintiffs have not alleged bad faith on the agency’s part. See Dist. Hosp. Partners, L.P. v. Burwell,

(D.C. Cir 2015) (showing that information “was deliberately or negligently excluded” requires “a strong showing of agency bad faith” (internal quotation marks and brackets omitted)); Oceana I,

.

12 evidentiary factors which lead to the ultimate conclusion” can constitute relevant factors that

justify supplementing the record “if ignored.”

Plaintiffs fail to point to any “intermediary evidentiary factors,” Oceana II, 454 F. Supp.

at 70, that appear in the call summaries but may have been ignored by the agency. Plaintiffs

argue that the October 2, 2016 call summary “indicates that current blueback herring populations

in New Hampshire are in poor condition, including specifically in the blueback-dominated

Oyster River.” Reply at 17-18. But the administrative record already notes the diminished

population levels in the Oyster River, in addition to low harvest levels. See, e.g., 2017 Update

Vol. II at 49 (“The numbers of river herring returning to the Oyster River fishway have been

decreasing since the mid 1990’s.”); id. at 53 (“There is typically very little river herring harvest

that occurs in the Oyster River, usually less than 800 per year . . . In response to diminishing

returns of river herring to the Oyster River fish ladder the river was closed to the taking of river

herring by any method beginning in 2012.”).

Plaintiffs also argue that the April 14, 2017 call summary should be included because it

“states that river herring runs in ‘some’ New Hampshire rivers have been ‘extirpated’”; because

it “notes that data sets in rivers with extirpated populations need careful treatment in the Stock

Assessment Update”; and because it “describes methodological issues to and caveats with the

Update, including the lack of an ‘objective, quantitative framework to making status

determination.’” Reply at 18. Again, this information is not sufficiently distinct from what the

administrative record already reflects. The ASMFC’s 2017 Update, which is in the record,

contains a lengthy description of the state of river herring populations in rivers in New

Hampshire, accompanied by detailed data. See 2017 Update Vol. II at 47-116. The 2017 Update

also discusses its own methodology and the limitations of the data on which it relies. See 2017

13 Update Vol. I at 4-8. The April 14, 2017 call summary adds only “nuanced points” to the

information contained in the 2017 Update. Oceana II, 454 F. Supp. at 70.

Similarly, the June 19, 2017 call summary does not include any ignored evidentiary

factors or entirely new subject matter. As plaintiffs note, this call summary contains a “critique

of the . . . terminology” used in the 2017 Update to characterize the alewife population in a

specific river. Reply at 18 (citing ECF No. 22-13 at 2). This distinction between the call

summary and the information in the existing administrative record is more semantic than

substantive.

Accordingly, the Court will deny the motion with respect to each of the ASMFC call

summaries.

D. External Emails

Finally, plaintiffs ask the Court to order NMFS to search for “external emails relevant to

the Listing Decision and to include them in the record.” Mem. in Supp. of Mot. to Compel at 19.

Plaintiffs note that the record contains only four email files, compared to over 1,600 emails that

were included in the record for the previous listing decision from 2013. Id. at 18; Reply at 19.

The government responds that plaintiffs have not overcome the presumption of regularity

to supplement the record. Opp’n at 19-20. According to the government, the relative lack of

emails in the record is due to the “centralized” decision-making process for the 2019 listing

decision, as compared to the more diffuse processes that led to the 2013 listing decision. Id. at

20. Specifically, the more recent internal procedures were intended to funnel all relevant

information (presumably including emails) to the lead author of the status review report, Tara

Trinko Lake. “Any information exchanged related to team business between team members or

from team members to external parties would have ultimately been provided to Ms. Trinko Lake

14 in some form, if that information was considered by the status review team.” Anderson Decl.

¶ 8. Therefore, the government searched Ms. Trinko Lake’s email in preparing the

administrative record but did not search the emails of other SRT members. Id. ¶¶ 8, 27.

Plaintiffs offer no concrete, non-speculative grounds to believe that other emails were

considered, or that emails were excluded in bad faith. They merely argue that the notion of a

“nearly email-free” rulemaking process “stretches the presumption of reality well beyond its

breaking point.” Reply at 20-21.

Of course, courts are “not required to exhibit a naivete from which ordinary citizens are

free,” and this Court should look carefully at the government’s proffered reasons for a record

nearly devoid of emails. Dep’t of Com. v. New York,

(2019) (citation

omitted). Given the small number of emails in the record, plaintiffs are understandably skeptical

of the government’s assertion that the agency established (and consistently followed) a procedure

to funnel all relevant emails to Ms. Trinko Lake’s inbox. However, this representation is not

facially implausible, and the Court must accept it absent concrete, non-speculative evidence to

the contrary.

***

15 IV. Conclusion

For the foregoing reasons, the Court will grant plaintiffs’ motion to compel

supplementation of the administrative record with respect to the Hare et al. Supporting

Information and deny it in all other respects. A separate Order shall accompany this

Memorandum Opinion.

CHRISTOPHER R. COOPER United States District Judge

Date: June 7, 2021

16