UNITED STATES DISTRICT COURT FOR THE DISTRICT OF COLUMBIA

A.P. BELL FISH COMPANY, INC. et al.,

Plaintiffs,

v.

GINA RAIMONDO, in her official capacity as Secretary of Commerce, et al., Civil Action No. 22-1260 (TJK) Defendants,

and

COASTAL CONSERVATION ASSOCIATION et al.,

Defendant-Intervenors.

MEMORANDUM OPINION

This case is about the allocation of red-grouper-fishing privileges in the Gulf of Mexico

among recreational and commercial fishermen. Plaintiffs—commercial fishermen and a trade as-

sociation—launch several legal challenges at the most recent allocation. In general, they argue

that the relevant agency based the allocation on bad data, applied those data arbitrarily, disregarded

pro-conservation mandates and other legal requirements, and committed itself to a new policy

without weighing suitable alternatives. But because the administrative record does not support

those claims, the Court will grant summary judgment for Defendants.

Background

Plaintiffs sue mainly under the framework established by the Magnuson-Stevens Conser-

vation and Management Act of 1976,

Stat. 331, 331–61 (codified as amended at

et seq.) (“MSA”). The MSA established eight regional fishery coun-

cils.

(a)(1). Those councils recommend fishery-management measures to the

Secretary of Commerce (“the Secretary”).

§ 1852(h)(1). The councils sometimes must act, in

turn, on the recommendations of a scientific committee created to assist with the more technical

aspects of fishery management. See id. § 1852(g)(1), (h)(6)–(7).

When the Secretary receives a recommendation from one of the councils, she must inde-

pendently review its legality and initiate the notice-and-comment process.

(a)(1).

She must then approve or disapprove the plan within 30 days after the comment period.

§ 1854(a)(3). In practice, the Secretary acts through the National Marine Fisheries Service (“the

Service”), an office of the National Oceanic and Atmospheric Administration inside the Depart-

ment of Commerce. See AR 7948.

One of the councils’ functions is to recommend to the Service “annual catch limits for each

of [their] managed fisheries.”

(h)(6). That process is based on a council’s as-

sessment of a fishery’s stock. See, e.g., AR 7177. A council’s scientific committee reviews the

results of a stock assessment and recommends values for inputs that are used to create annual catch

limits. See AR 7979. A council reviews those recommendations and produces its own recom-

mendations to be incorporated in a fishery management plan. See AR 7995, 8000–01. The Service

then uses those recommendations and the notice-and-comment process to generate a final rule.

See

(a)(1)(B), (2)–(3).

Plaintiffs oppose a final rule known as Amendment 53 to the Fishery Management Plan for

the Reef Fish Resources of the Gulf of Mexico (“A53”). AR 7946. A53 reduced the amount of

red grouper that may be fished from the Gulf of Mexico. See AR 7987–95. It also altered the

2 proportions of allowable catch that may be fished by recreational and commercial fishermen. See

The latter aspect of A53 is the focus of Plaintiff’s challenge.

A. The MSA

Chief among the MSA’s substantive legal requirements are the ten “national standards for

fishery conservation and management.”

(a). Each national standard directs the

Secretary to ensure that a management plan “comports with” “a specific and essential policy ob-

jective.” C & W Fish Co. v. Fox, Jr.,

(D.C. Cir. 1991). Essential though they

are, the national standards sometimes present “competing goals.” N.C. Fisheries Ass’n v.

Gutierrez,

(D.D.C. 2007). Thus, one of the Service’s critical roles is to

“strike the appropriate balance” among them.

That decision must receive a “a high degree of

deference” because it implicates the agency’s technical expertise. Nat. Res. Def. Council v. Nat’l

Marine Fisheries Serv.,

(D.D.C. 2014).

But the Service’s discretion is bounded. The national standards express justiciable “limi-

tations . . . on the Service’s authority.” Groundfish F. v. Ross,

(D.D.C.

2019); see also, e.g., Guindon v. Pritzker (Guindon I),

, 195–97 (D.D.C. 2014);

Guindon v. Pritzker (Guindon II),

, 194–95 (D.D.C. 2017); see also generally

(f)(1). Courts must ensure the Service weighs the requirements of the national

standards and engages in “reasoned decision making” that is “supported in the record.” Oceana,

Inc. v. Raimondo (Oceana II),

, 29 (D.D.C. 2021).

One way the Service interprets the national standards is by promulgating regulations. Such

regulations lack “the force and effect of law.”

(b). So they do not qualify for

Chevron deference. United States v. Mead Corp.,

, 226–27 (2001). Still, courts have

tended to extend them “considerable deference” under the Skidmore framework because of “their

3 thoroughness, the agency’s expertise, and the administrative formalities involved in their promul-

gation.” Guindon I,

(quotation omitted). The Service’s regulations inter-

preting the national standards are codified at

.305–600.355.

The MSA contains other legal requirements too. Section 1853(a), for example, codifies

fifteen elements that “[a]ny fishery management plan” must contain. Those include the directive

to incorporate “conservation and management measures” that will “prevent overfishing and rebuild

overfished stocks [and] protect, restore, and promote the long-term health and stability of the fish-

ery.”

(a)(1). Unlike the national standards, those elements are not competing

policy objectives that must be balanced—they are discrete legal requirements that, if unsatisfied,

may warrant judicial vacatur of a management plan. See, e.g., N.C. Fisheries Ass’n,

, 103–04 (declining to order vacatur despite finding such a deficiency).

B. Factual Background

A53 altered a management plan that has governed reef fishing in the Gulf since 1984. See

AR 7974; Reef Fish Fishery of the Gulf of Mexico,

, 39548–58 (Oct. 9, 1984)

(“FMP”). To provide context for A53, the Court briefly traces that scheme’s evolution.

1. Red-Grouper Management Before 2009

The original management plan did not limit the volume of red grouper that could be caught.

See FMP, 49 Fed. Reg. at 39548. Later data-collection efforts, though, raised concerns about the

long-term health of the red-grouper stock. See Amendment Number 1 to the Fishery Management

Plan for the Reef Fish Resources of the Gulf of Mexico,

, 2078–79 (Jan. 22,

1990) (A1). The Service responded with a minimum-size limit and quotas for both the recreational

and commercial fishing sectors.

. The Service explained that it had allocated allowable

catch among the sectors according to its estimates of the “historical percentages harvested by each

user group during the base period of 1979–87.” Proposed Amendment Number 1 to the Reef Fish

4 Fishery Management Plan,

, 41307 (Oct. 6, 1989) (PA1); see also A1, 55 Fed.

Reg. at 2085 (approving that proposal).

In other words, the Service planned to set overall catch limits based on the need to prevent

overfishing and then to allocate the total amount of catch authorized among the sectors based on

estimates of their historical participation in the red-grouper fishery. See PA1, 54 Fed. Reg. at

41307. The Service’s catch-estimation efforts have thus undergirded its management plans. As

Plaintiffs observe, the commercial sector is subject to “[s]trict monitoring,” which makes measures

of that sector’s fishing efforts more exact. ECF No. 20 at 17–19. But estimating catch for the

recreational sector is an imprecise science at best.

The Service estimates recreational catch in two steps. First, it approximates the catch rate,

the number of fish caught per angler, per trip. See AR 18891. Second, it gauges fishing effort, the

number of fishing trips. See id. By multiplying those numbers together, the Service derives its

estimate of the total catch by species. Id. Each step is much more complex than that simple

description betrays, but that is the basic idea.

Those two constituent estimations can be performed in different ways. When the Service

first allocated allowable catch, it used estimation methods the Court will call collectively

“CHTS.” 1 Under the CHTS system, the Service calculated the catch rate through a survey based

on dockside intercepts of fishing vessels. See AR 18988–89. And it appraised fishing effort by

randomly sampling coastal households by calling landline telephones and asking respondents “to

provide details about all fishing trips that occurred during the prior 60 days.” AR 18923–25.

1 CHTS stands for “Coastal Household Telephone Survey.” AR 18888. The survey to which that name refers formed only a part of the Service’s method for estimating fishing effort, which is itself only a component of the estimation process for total catch. See AR 18886–92. But greater terminological precision is unnecessary for this case.

5 The initial red-grouper allocation did not abate concerns about the stock’s vitality. In fact,

CHTS-based assessments in 1999 and 2002 concluded that the stock level needed improvement.

Notice of Rebuilding Plan for Red Grouper in the Gulf of Mexico,

, 1278 (Jan.

8, 2004). The Service’s solution was a “10-year red grouper rebuilding plan” that aimed for “a

reduction . . . in overall red grouper harvest.” Proposed Rebuilding Plan for Red Grouper in the

Gulf of Mexico,

, 7899 (Feb. 20, 2004). Meanwhile, the Service observed that

its estimates of relative historical fishery participation between the recreational and commercial

sectors had been mostly constant. Rebuilding Plan for Red Grouper in the Gulf of Mexico,

, 33317 (June 15, 2004). So it applied “the same percentage reductions to each

sector,” effectively preserving the extant allocation.

The same year the Service implemented the rebuilding plan, “recreational red grouper land-

ings . . . significantly exceed[ed] the target catch level.” Proposed Gulf of Mexico Recreational

Grouper Fishery Management Measures,

, 16275 (Mar. 31, 2006). So the

Service, concluding that excessive recreational fishing was hampering its efforts, reduced the per-

person, per-day catch limit, which it predicted would reduce recreational red-grouper landings

substantially. Gulf of Mexico Recreational Grouper Fishery Management Measures,

, 34534 (Mar. 31, 2006).

Some experts warned that the poor quality of data measuring recreational fishing might be

contributing to the problem of managing the red-grouper stock, and they recommended a “major

overhaul” of the CHTS system. See Committee on the Review of Recreational Fisheries Survey

Methods, National Research Council, Executive Summary of Review of Recreational Fisheries

Survey Methods 3–4 (2006), https://sedarweb.org/documents/s24rd36-executive-summary-re-

6 view-of-recreational-fisheries-survey-methods/. Congress responded with a law directing the Sec-

retary to “improve the quality and accuracy of information generated” by the Service’s estimation

methods. Magnuson-Stevens Fishery Conservation and Management Reauthorization Act of

2006, § 201,

Stat. 3575, 3611 (codified at

(g)(3)(A)).

The “goal,” Congress explained, was data with “acceptable accuracy and utility for each individual

fishery.”

Thus, the Service began phasing out CHTS in 2008, although it did not implement

its most substantive changes until 2013. AR 18890, 18895, 18921–22.

2. Amendment 30B

The pre-A53 allocation was set in 2009 by Amendment 30B (“A30B”). By that time, new

data showed that the red-grouper stock had recovered. Proposed A30B,

, 68391

(Nov. 18, 2008). So the Service increased allowable catch, which also prompted it to reconsider

how catch was allocated among the sectors. See

A30B continued the practice of allocating “based on historical percentages harvested by

each [sector].” AR 10479. But using that approach required the Service first to decide which

years’ data would be used as the “base period.” See

The Service recognized “danger” in setting

a base period of too few years or relying on years in which external factors may have influenced

catch numbers. See AR 10480. So it considered that red-grouper-specific quotas had recently

been introduced and that other regulatory changes may have impacted the total catch in some years

for which there was data. See AR 10480, 10483–44. It ultimately selected a period “based on all

available years during which grouper were identified by species”—that is, 1986 to 2005—the

“longest and most robust time series.” AR 10483. That base period, the Service determined,

would minimize the influence of external factors and thus best capture each sector’s historical

participation in the fishery. See AR 10484.

7 That allocation method gave 76 percent of the total allowable catch to the commercial sec-

tor and 24 percent to the recreational sector. AR 10482. And because then-recent MSA amend-

ments had directed the Service to set catch limits annually, AR 10485, 2 future sector quotas would

be set by multiplying the sector’s allocation by a given year’s catch limit. Cf. AR 10494. For this

reason, the allocation set by A30B played a key role in setting sector catch limits in the years just

before A53.

3. A New Estimation Method for Recreational Catch

The Service continued its attempts to improve its estimations of recreational catch. In

2013, it debuted a new method for estimating the catch rate per angler, per trip. AR 18988. Alt-

hough that change was significant, the technical details are peripheral to this case. See AR 18893–

18913. Then, in 2015, the Service certified a new method for gauging fishing effort. AR 18990.

That change is directly relevant to the parties’ dispute, so it warrants further explanation.

The new fishing-effort-estimation method relies on a mail survey instead of a telephone

survey. AR 18915–16. Survey participants are drawn from all households within surveyed states,

but the sample is “optimize[d]” by including data that classifies households as coastal and notes

whether an address corresponds to a database of fishing licenses. AR 18914–15. The survey asks

participants to report “the total number of . . . fishing trips taken during the [period] by each house-

hold member.” AR 18916. Various statistical techniques are then applied to “estimate total [fish-

ing] effort.” AR 18917.

The Service has offered several rationales for this change. To begin, CHTS was based on

a technology—landline phones—the use of which was evaporating. AR 12513. So response rates

2 See also Magnuson-Stevens Fishery Conservation and Management Reauthorization Act of 2006 § 103(c)(3), 120 Stat. at 3581.

8 plummeted over time, particularly after 2000. AR 12492. Similarly, the reliance on landlines

biased the survey demographic towards older participants. AR 19471. And those who lived more

than 50 miles from the coast were excluded entirely. See AR 18888. What’s more, landline-

survey participants may not have had accurate information about fishing effort from their house-

holds. AR 19472. And even if they did at one time, the short duration of a phone call may have

contributed to participants’ failure to recall that information. See AR 12498. Thus, the purpose

of the new survey was to reduce “undercoverage, nonresponse, and measurement errors.” Id.

Despite those methodological changes, the basic estimation process remained the same.

The Service combines its estimates of the catch rate with those of fishing effort to calculate total

catch. See AR 18941–57. The Court will refer to this new process that uses the mail survey and

the altered catch-rate survey as “FES.” 3

When the Service deployed FES, it found that its catch estimates “were, on average, several

times larger than the CHTS estimates.” AR 18971. That was unsurprising to some extent because

the point of FES was to correct CHTS’s undercount of fishing effort. AR 12498. But it also posed

a problem: The Service needed a consistent “recreational fishing time series dating back to 1981”

in part because it uses the historical estimates of recreational catch to set annual catch limits.

AR 18971. If the new method to measure fishing effort yielded systematically higher recreational

catch estimates than the old method used to set the limits—even though fishing effort had not

changed—the policy objective of the annual catch limit would be frustrated. See AR 17420 (deny-

ing that there was a “sudden increase in fishing effort”).

3 Again, “FES” is a terminological simplification. “FES” stands for “Fishing Effort Sur- vey” and thus properly refers only to the mailed survey, which is only one component of estimating total recreational catch. AR 18914.

9 To solve that problem, the Service developed what it called a “calibration” model.

AR 18971. The idea was to convert “CHTS estimates to FES estimates and vice versa.”

AR 18972. In other words, the model was a tool for answering the question, “What would [esti-

mates of] historical recreational fishing efforts look like had FES been implemented [sooner]?”

AR 19472. The model was built first by running “both CHTS and FES . . . concurrently for three

years.” Id. Then the Service used statistical techniques, accounting for the traits of the constituent

years, to build the model. See AR 18971–86. The result, the Service hoped, was a model that

“capture[d]” historical changes in CHTS estimation because of decreased landline use, resulting

in a single, consistent estimated time series. AR 18974.

A panel of independent scientists reviewed the calibration model. It “expressed concern

on several topics” but considered none “sufficient to preclude implementation of the model.”

AR 18693. For one thing, it concluded, there was no way to be sure that the revised estimates of

historical fishing effort were more accurate—that is, closer to true fishing effort—than those pro-

duced by CHTS. AR 18696. That is because in both cases the data is self-reported, and so not

externally verifiable. AR 18699. But the panel still concluded that, from a mathematical perspec-

tive, the model was “reasonable and scientifically defensible.” AR 18693. The model “relies on

standard, well-known, and highly regarded methodology,” the panel explained, and it also includes

“several innovations [that help] achieve analytical consistency.” Id. Thus, the panel acknowl-

edged that using the model to convert pre-FES “estimates of recreational fishing effort” to FES-

compatible units was “a well-suited and statistically-appropriate approach.” AR 18698.

FES “officially replaced . . . CHTS in January of 2018.” AR 18922. The Service’s most

recent assessment of the red-grouper stock, completed in 2019, relied on the updated estimates.

AR 7538. It thus had to make sense of fishing-effort calculations that were “nearly three times

10 higher” for part of the recreational sector. AR 7225–26. So the calibration model played an im-

portant role in gauging the health of the stock. See id. FES is also used to monitor recreational

fishing against annual quotas. AR 8124.

4. An Emergency Reallocation

Meanwhile, there were signs that the red-grouper stock in the Gulf of Mexico was strug-

gling. The 2017 fishing season was unproductive, and 2018 brought a “severe red tide,” which

likely killed many red grouper. See AR 7965, 7976. In 2019, the Service responded with an

emergency rule that reduced catch limits. AR 7976–77. That reduction preserved the allocation

of 76 percent to the commercial sector and 24 percent to the recreational sector. See AR 7977.

Or, more accurately, it preserved the extant allocation at least as expressed one way.

The 2019 catch limits were set by using what the Service calls “CHTS units.” AR 8135.

In other words, the limits—which are denominated in pounds of grouper—relied on the assump-

tion that CHTS would be used to track recreational catch. But that was no longer true. AR 8124.

And because FES-based estimates are much higher, proceeding on that assumption would create

the impression that the recreational sector was fishing its quota much faster than it had in the past.

That impression, as far as anyone can tell, would have been wrong; recreational fishing effort has

not suddenly spiked. AR 17420. The Service’s solution was to convert the quotas from “CHTS

units” into “FES units” with its calibration model. See AR 8135.

Recall, though, that FES is a method of estimating recreational catch. So figures related

to commercial catch, which is tracked instead of estimated, require no such conversion. See

AR 8135. The result is that the 2019 emergency-rule quota, when expressed in “FES units” instead

of “CHTS units,” states a higher value for the recreational catch limit but the same value for the

commercial catch limit. See id. Thus, comparing the two quotas in “FES” units results in an

11 allocation of 60.08 percent to the commercial sector and 39.92 percent to the recreational sector.

See AR 7967.

Expressed this way, then, under the 2019 emergency rule, the Service de facto reallocated

total catch. See ECF No. 20 at 48–50. The rule, after all, says nothing about CHTS or FES units.

See Temporary Rule to Establish Management Measures for Red Grouper in the Gulf of Mexico,

, 22390 (May 17, 2019). Using CHTS units, the Service set the annual recre-

ational catch limit at “1.00 million lb.”

But then the Service, using FES—the estimation

method it considered to be the best available—allowed the recreational sector to go above that and

catch up to 2.1 million pounds. AR 7967.

However the figures are expressed, at least on paper, they ostensibly allow the same

amount of recreational fishing activity. See AR 7967. Even though the estimates of recreational

anglers’ progress toward their quota appears much faster under FES, the same increased clip would

have been part of the estimates of red-grouper stock that were used to generate catch limits in the

first place if FES had been used beforehand.

5. Amendment 53

A53, among other things, sought to complete the transition from CHTS to FES. Formally,

A53 implemented two related actions. AR 7966. For this case, it is enough to say that A53 altered

the annual catch limits based on the results of the 2019 stock assessment. See

And, for the first

time, it set the recreational sector’s limit explicitly in “FES units.” AR 7967.

The Service said its purpose was two-fold. First, it wished to “revise the red grouper allo-

cation between the commercial and recreational sectors using the best scientific information avail-

able.” AR 7966. Second, it sought to change the total “allowable harvest of red grouper based on

the results of the recent stock assessment” and the recommendation of the relevant scientific com-

mittee.

It also explained the regulation was needed to “use the best scientific information

12 available to establish Gulf red grouper sector allocations, . . . ensur[e] that the historical participa-

tion by the recreational and commercial sectors is accurately reflected by the sector [quotas], and

that [the] recreational [quota] is consistent with [FES] data.”

The Service’s regulations establish a structured process for setting catch limits. 4 First, a

scientific committee uses a stock assessment to predict the amount of catch that would jeopardize

a stock’s long-term productive capacity. It then reduces that figure to account for inherent estima-

tion uncertainty. From there, a regional council takes over, subject to the Service’s (and ultimately

the Secretary’s) approval. It sets an aggregate catch limit, which cannot exceed the figure gener-

ated by the scientific committee.

(h)(6). It may divide that catch limit by sector,

as it has done here. That figure is then further reduced to protect against the effects of management

uncertainty. Thus, setting catch limits combines scientific and policy decisions.

Although the sector allocation appears to follow earlier scientific determinations, the causal

arrow runs both ways. That is, the allocation itself affects the final calculation of the figure that

would jeopardize a stock’s long-term productive capacity. To see why, it is important first to

understand some key differences between the sectors’ fishing behavior.

The two sectors fish under different regulatory requirements. See generally AR 8176–80.

One consequence of that regime disparity is that the fish caught by recreational anglers are, on

average, younger and smaller than those caught by commercial anglers. AR 9692. Another is that

the proportion of their catch that recreational anglers retain is much lower than the comparable

figure for commercial fishermen. See AR 4409–10. That second attribute will require further

explanation, but the upshot for now is that a pound of fish brought to shore (“landings”) by the

4 To avoid a hailstorm of initialisms, the Court will describe that process only in general terms. Further detail is available at

(e)–(f).

13 recreational sector has different consequences for the fishery than a pound of fish landed by the

commercial sector. Specifically, as the recreational sector catches more fish, the fishing mortality

rate rises, and the stock’s fecundity falls.

Thus, when more of the total allowable catch is allocated to the recreational sector, the total

allowable catch allowed for both sectors must also decrease. AR 7967, 8070. Otherwise, the

consequences for the stock’s long-term productive capacity would be more severe than first pre-

dicted. So when the Service gives more of the total allowable catch to recreational anglers, the

impact on commercial anglers takes two forms. First, obviously, commercial anglers’ slice of the

pie shrinks as that of recreational anglers grows. And second, the whole pie shrinks. That phe-

nomenon was among the factors used to set the total allowable red grouper catch in A53. See

AR 7967, 8070.

Two other factors drove that calculation. First was the assumption that total catch needed

to decrease, partly because of expected mortality caused by the recent red tide. AR 7965–66.

Second was the higher historical catch estimates produced by the FES system.

FES estimates

helped produce an assessment that the minimum stock size need not be as large as previously

thought. See AR 4470. That assessment, other things equal, increased the total amount of catch

that could be permitted under A53. See AR 7965. The net effect of those factors was a range of

total catch limits much lower than those in A30B—but not as low as it would have been without

the introduction of FES. Compare AR 7967 with AR 10482.

As for the allocation decision, there are many relevant factors, but only one requires more

exposition. Earlier the Court noted that the recreational sector retains fewer of its caught fish than

does the commercial sector. Another way to put that fact is that recreational fishing creates more

14 “bycatch” than does commercial fishing. 5 And concepts related to bycatch underlie several of the

legal challenges before the Court.

One important source of bycatch is “the discard of whole fish at sea or elsewhere.”

(c)(1). And the bycatch disparity between the sectors exists because “the rec-

reational sector discards an order of magnitude more fish . . . than [does] the commercial sector.”

AR 8070. In fact, that gap is now thought to be larger than previously was known. AR 4409–11.

But simply measuring total discards does not tell the whole story of what is happening to

a fishery. Another aspect of the problem is discard mortality—the proportion of discarded fish

that die from their encounter with an angler. By that metric, the commercial sector’s footprint is

much larger. See AR 8169–80. Still, that effect is not so large that it vitiates the much higher

recreational discard rate. On net, recreational fishing kills more discarded fish than does commer-

cial fishing. AR 8070.

Based on those effects and many others, the Service’s task was to allocate total catch among

the sectors. To that end, it weighed the consequences of six alternative allocations. AR 7967. The

first alternative was to do nothing: leave the current limits and allocation in place.

The second

alternative was to update the total catch limit given the new stock assessment based on FES but to

leave the 76-24 allocation in place.

The third, fourth, and fifth allocations would all have

updated the total catch limit, and each would have used FES data to set an allocation based on

relative fishery participation during some historical period.

The sixth allocation would have

updated the total catch limit, held the commercial quota constant, and allocated the rest of allow-

able catch to the recreational sector. See

5 “The term ‘bycatch’ means fish that are harvested in a fishery, but that are not sold or kept for personal use.”

(c). That term includes fish that are killed by fishing gear even if those fish are not captured.

§ 600.350(c)(1).

15 The Service selected the third alternative. That allocation was based on historical catch

data—converted to “FES units” for the recreational sector—from 1986 to 2005. AR 7967. In

other words, it was based on relative catch during the same years as the allocation from A30B.

AR 7967, 10482. The Service selected that alternative after concluding that allocation would pre-

serve the basic policy of A30B while updating the limits based on the best scientific information

available, apply the necessary reduction in total catch roughly equally among the sectors, and cause

“the greatest net economic benefits” among the possible allocations. AR 7995.

C. Procedural History

Plaintiffs are commercial fishing companies and a trade organization that represents such

companies. ECF No. 1 at 6–8. They sued seeking to vacate A53. Id. at 43. In their view, A53

violates the MSA, the Administrative Procedure Act (“APA”), and the National Environmental

Policy Act (“NEPA”). Their position is endorsed by an amicus that, despite being an industry

organization that represents recreational anglers, believes A53’s allocation will ultimately harm

charter fishermen. ECF No. 23 at 5–6. The federal-government defendants—the Secretary, the

Service, and the National Oceanic and Atmospheric Administration—are joined by two defendant-

intervenors, the State of Louisiana and the Coastal Conservation Association, an organization of

recreational anglers. See Minute Order of July 26, 2022; Minute Order of Sept. 1, 2022. Each of

the parties has moved for summary judgment. See ECF Nos. 20 (Plaintiffs’ motion), 31 (Defend-

ant-Intervenor Louisiana’s cross-motion), 33 (Defendant-Intervenor Coastal Conservation Asso-

ciation’s cross-motion), and 35 (Defendants’ cross-motion). The Court held a hearing on those

motions on October 18, 2022.

Legal Standard

Although this case comes before the Court in a summary-judgment posture, the ordinary

standard for deciding such motions does not apply. Plaintiffs brought this case under the APA, so

16 the Court “sits as an appellate tribunal.” Groundfish,

(quoting Am. Biosci-

ence, Inc. v. Thompson,

(D.C. Cir. 2001)). The Court has no factfinding role;

instead it reaches legal conclusions about whether the agency action is “supported by the admin-

istrative record and otherwise consistent with the APA standard of review.”

(quoting Citizens

for Resp. & Ethics in Wash. v. S.E.C.,

(D.D.C. 2013)).

The MSA limits the scope of APA review. Under

(f), the Court may

vacate A53 only under the grounds described in

(2)(A)–(D). That is, the Court must

vacate A53 if it is “arbitrary, capricious, an abuse of discretion, or otherwise not in accordance

with law” or “in excess of statutory jurisdiction, authority, or limitations, or short of statutory

right.” Groundfish,

(quoting

(2)(A), (C)). And the statutory

authority to which that standard refers comes not from the APA itself, but from the MSA. See id.;

Navab-Safavi v. Broad. Bd. of Governors,

(D.D.C. 2009).

The arbitrary-and-capricious-review standard is familiar. Courts must ensure that an

agency has “examine[d] the relevant data and articulate[d] a satisfactory explanation for its ac-

tion.” F.C.C. v. Fox Television Stations,

(2009) (quotation omitted). That

standard does not empower a court to “substitute its judgment for that of the agency.”

(quota-

tion omitted). In other words, the agency’s “policy choices” are not up for debate—only the “ex-

planation it has given.”

. That explanation must include a “rational connection between

the facts found and the choice made.” Motor Vehicle Mfrs. Ass’n of U.S., Inc. v. State Farm Mut.

Auto Ins. Co.,

(1983) (quotation omitted). It must also consider all “important

aspect[s] of the problem”—that is, those Congress considered relevant.

The Service also invokes deference under Chevron U.S.A., Inc. v. N.R.D.C.,

(1984). ECF No. 40 at 14–15. To decide whether that framework applies, the Court must first ask

17 whether the agency has exercised congressionally delegated lawmaking authority, a necessary

predicate of Chevron deference. See United States v. Harmon,

, 57 (D.D.C.

2020); see also Mead, 533 U.S. at 226–27. If it has, the Court should proceed to “exhaust the

traditional tools of statutory construction to determine whether Congress has spoken to the precise

question at issue.” N.R.D.C. v. Daley,

, 752 (D.C. Cir. 2000) (quotations omitted).

If so, the question is answered, and no deference is owed. Id. If not, the Court must extend def-

erence to any “permissible agency interpretation of the statute.” Id. (quotation omitted).

Analysis

Plaintiffs’ challenges to A53 fall into four categories. First, Plaintiffs contend the Service

failed to satisfy the national standards. ECF No. 20 at 30–39, 44–46. Second, they say A53 dis-

regarded other MSA requirements. Id. at 39–44. Third, they bring miscellaneous claims under

the arbitrary-and-capricious standard. Id. at 46–53. Fourth, they maintain the Service did not

adequately weigh alternatives as required by NEPA. Id. at 54–56.

The Court also notes that at times, Plaintiffs’ briefing veers into a general attack on fishery

management in the Gulf of Mexico. For instance, they complain that the commercial sector is

disadvantaged by “a more stringent and effective regulatory regime that is lacking in the recrea-

tional sector.” ECF No. 20 at 36. They also suggest the change to FES was arbitrary and without

scientific foundation. See ECF No. 38 at 8–10. Those sorts of arguments can be relevant insofar

as they bear on whether A53 was rational and supported by the administrative record. But the

Court must carefully define the scope of its review. Judicial review of MSA-authorized regulations

is limited to those challenged within 30 days of promulgation.

(f)(1). Here, that

means A53—and only A53. See ECF No. 1 at 4. That limitation is jurisdictional. Nat’l Mining

Ass’n v. U.S. Dep’t of the Interior,

(D.C. Cir. 1995).

18 Plaintiffs reply that A53 reconsiders and applies these other aspects of fishery management,

placing them within A53’s scope. ECF No. 38 at 22–23. True, an agency’s reconsideration of a

policy—a genuine reevaluation of the appropriate course—permits new challenges. See Pub. Cit-

izen v. Nuclear Regul. Comm’n,

, 150–51 (D.C. Cir. 1990). And “a party against

whom a rule is applied may, at the time of application, pursue substantive objections to the rule”

notwithstanding a time limit for a procedural challenge. Indep. Cmty. Bankers of Am. v. Bd. of

Governors of the Fed. Rsrv. Sys.,

(D.C. Cir. 1999).

But neither of those principles applies here. Mechanisms for managing the recreational

sector, including the method for estimating its catch, are not “applied” against the commercial

sector. Those can indirectly affect the catch limits the Service applies to commercial anglers, but

Plaintiffs can challenge that aspect of A53 anyway. And the Court simply disagrees that in prom-

ulgating A53, the Service genuinely reconsidered its broader management approach to the recrea-

tional sector. Plaintiffs say, for example, that the Service responded to one of its comments about

recreational-fishing management. ECF No. 38 at 22. But a response to a comment already made

is not an “invitation to comment,” which is sufficient to reopen an issue. Pub. Citizen,

. 6 And the example Plaintiffs cite simply describes, in past tense, the results of a prior review.

See AR 17425. In short, after reviewing the administrative record, the Court concludes that A53

“considered” only the actions it actually took: setting catch restrictions on red grouper in the Gulf

of Mexico.

Thus, some aspects of fishery management are outside the scope of this case. Prominent

among those is the implementation of FES to monitor recreational catch and assess the stock. That

6 Plaintiffs, in other words, may not “comment on matters other than those actually at issue, goad an agency into a reply, and then sue on the grounds that the agency has re-opened the issue.” Pub. Citizen,

(quotation omitted).

19 transition predated A53, and its effects are irreversible for purposes of this case—CHTS no longer

exists, at Congress’s direction. If the Service is right about the relationship between CHTS esti-

mates and FES estimates, the now-immutable use of FES for estimating recreational catch means

that preserving the prior nominal allocation would actually have reduced real-world recreational

fishing opportunities dramatically. See AR 7967 (showing that alternative 2 would have reduced

the recreational quota in FES units from 2.1 million to 1.18 million pounds); AR 7994 (explaining

that alternative 2 could have resulted in the Service’s closing the recreational fishing season as

early as June 13).

But the present immutability of estimating recreational catch with FES does not decide the

case. Although the Service understood A53 to be a translation of the status quo into new “units,”

that does not mean A53 lacks reviewable policy content, as some parties have implied. See ECF

Nos. 31-1 at 14–15, 33-1 at 19. In A53, the Service considered which historical base period to

use, AR 7987 (alternatives 3–5), and it even considered an alternative that would have jettisoned

that approach altogether,

(alternative 6). So the policy decision to base its allocation on esti-

mated effort between 1986 and 2005 was freshly selected, and it must be reviewed for consistency

with the MSA, APA, and NEPA. Still, the unchangeable fact that FES will be used in the imme-

diate future to estimate recreational catch bears heavily on whether the Service’s decision was

rational and whether its explanation is logical and supported by the record.

With that scope of review in mind, the Court turns to Plaintiffs’ specific claims.

A. The National Standards

1. National Standard 4

National standard 4 applies when the Service finds it necessary to “allocate . . . fishing

privileges.”

(a)(4). Allocations must be “fair and equitable to all . . . fishermen”

and “reasonably calculated to promote conservation.”

Plaintiffs attack A53 on both flanks.

20 Start with the requirement to promote conservation. The MSA does not define “conserva-

tion,” but it defines “conservation and management.”

(5). That term

refers to all of the . . . measures (A) . . . useful in rebuilding, restoring, or maintain- ing, any fishery resource . . . ; and (B) which are designed to assure that (i) a supply of food and other products may be taken, and that recreational benefits may be obtained, on a continuing basis; (ii) irreversible or long-term adverse effects on fishery resources and the marine environment are avoided; and (iii) there will be a multiplicity of options available with respect to future uses of these resources.

Although the term “conservation” has not been given a precise definition, one court in this

district has recognized that a measure that likely would “cause overharvesting” does not “promote

conservation.” Guindon I,

(quoting

(a)(4)).

Plaintiffs contend the definition of “conservation and management” cannot be used to un-

derstand the meaning of “conservation” in national standard 4. ECF No. 38 at 15. If that is so, the

Court must look to the ordinary meaning of “conservation.” See Sebelius v. Cloer,

(2013). But it is not clear that method would produce any different understanding of the term.

The D.C. Circuit has explained that the ordinary meaning of “conservation” includes preventing

damage to a natural resource. See Copper Valley Mach. Works, Inc. v. Andrus,

, 601

& n.7 (1981). Plaintiffs do not suggest otherwise. 7 And the MSA definition encompasses

measures designed to preserve the availability of natural resources on a “continuing basis” and to

avoid “long-term adverse effects.”

(5)(i)–(ii). So whether under the statutory

definition or the ordinary meaning, the promote-conservation requirement of national standard 4

contemplates protecting fisheries from damage to their long-term vitality, such as from “overhar-

vesting.” Guindon I,

.

7 In fact, their argument is that A53 fails to promote conservation precisely because it will damage the red-grouper stock. ECF No. 20 at 30–31.

21 Moreover, as Plaintiffs observe, the standard demands that A53 promote conservation.

ECF No. 20 at 30. In Groundfish, this Court explained that to “promote” means to “advance or

further” conservation. 375 F. Supp. 2d at 89. Neutrality, in other words, is not enough. Id.

So how does A53 advance the vitality of the red-grouper stock? A simple answer is that it

reduces the total number of grouper that may be caught. And that reduction was based on the

scientific committee’s assessment of the number that could be caught without jeopardizing the

stock’s long-term vitality, considering the bycatch effects of the allocation decision. On its face,

that reduction seems “reasonably calculated to promote conservation.”

(a)(4).

Plaintiffs, though, say the allocative aspect of A53 must independently promote conserva-

tion. ECF No. 20 at 30–32; ECF No. 38 at 14. That is, they reason the Service must distribute the

total catch in a way that will “‘contribute to the growth, enlargement, or prosperity of’ . . . the red-

grouper stock.” ECF No. 38 at 14 (quoting Groundfish,

). They also contend

that A53’s allocation does not qualify because, having increased the recreational share of the total

catch limit, the Service necessarily increased bycatch, reduced the fecundity of the stock, and “in-

creased management uncertainty” relative to other possible allocations.

Plaintiffs read too much into Groundfish. They emphasize language from that case ex-

plaining that the requirement to promote conservation applies “specifically to the allocation of

fishing privileges.” Groundfish,

(quoting

(a)(4)) (altera-

tions adopted); ECF No. 38 at 14. That quote eliminates essential context. The opinion explains

that the promote-conservation requirement applies “specifically to the allocation of fishing privi-

leges, not to the [fishery management plan] as a whole.” Groundfish,

(cleaned up and emphasis added).

22 Placing that statement in context reveals its meaning. In Groundfish, the Service had

amended a management plan to “reserve a portion” of allowable catch for anglers who would

process those fish at particular plants. Groundfish,

. It did so because it

hoped to preserve processing plants to benefit the communities in which those plants operated.

at 79–80. But that entire agency action left the total allowable catch unchanged. See

For that

reason, the Service claimed that what mattered was that the whole fishery management plan still

promoted conservation, even though the amendment was merely conservation neutral.

.

The Court rejected that position, explaining that an amendment that allocates fishing privileges

must itself further conservation—not simply “the [fishery management plan] as a whole.”

A53 is nothing like the amendment in Groundfish. The equivalent argument here would

be that A53 need not further conservation because the 1984 plan combined with other amendments

already does. But the Service has not made and need not make that argument because A53 furthers

conservation by reducing the number of red grouper that may be caught. The idea that the alloca-

tive aspect of A53 must further conservation apart from the rest of the agency action is unsupported

by Groundfish.

Nor can that idea find support in national standard 4. No doubt, the requirements of that

standard apply to an “allocation.”

(a)(4). But there is no reason to read that term

as narrowly as Plaintiffs do. Courts have recognized that an agency action can be an “allocation”

even if it does not explicitly distribute fishing privileges. 8 That can be true where, for example,

the action closes a fishing zone to one sector and not the other. United Cook,

,

at *13–14. If the second-order effects of an action can render it an allocation, it follows that the

8 See, e.g., United Cook Inlet Drift Ass’n v. Nat’l Marine Fisheries Serv., No. 3:21-CV-247 (JMK),

, at *13–14 (D. Alaska June 21, 2022) (citing Nat’l Coal. for Marine Conservation v. Evans,

(D.D.C. 2002)).

23 term “allocation” in national standard 4 refers to an agency action that “result[s] in [a] direct dis-

tribution[ ] of fishing privileges”—not simply to the distributive aspect of the action.

(c)(1).

Only that reading can make sense of the term “conservation.”

(a)(4). As

the Court has explained, to conserve is to protect the benefits of the red-grouper stock so they are

available “on a continuing basis.” See

§ 1802(5)(i). The continued availability of red grouper

is affected by the number and the characteristics of the fish that die, not by the identity of those

who catch them. It is true that each fish caught by anglers of different sectors has disparate con-

servational consequences, but that is why the aggregate catch limit is adjusted to account for the

allocation. When all was said and done, each allocation alternative was projected to have near-

identical consequences for conservation. AR 8125; see also AR 7972 (explaining how the final

catch limits account for the increased management uncertainty caused by a larger allocation to the

recreational sector). Thus, it is unclear how the distributive decision itself could ever “promote

conservation.” What matters is the total catch limit—so long as that limit accounts for factors like

bycatch, effects on the average age of a stock, and management uncertainty, as this one does.

In summary, an action that “result[s] in [a] direct distribution[ ] of fishing privileges” is an

allocation that must comply with national standard 4.

(c)(1). A53 distributes

fishing privileges, so it must, among other things, be “reasonably calculated to promote conserva-

tion.”

(a)(5). A53 is reasonably calculated to promote conservation because it

reduces the total catch limit for red grouper to a level that the relevant scientific committee deter-

mined would preserve the stock so that its benefits will be available “on a continuing basis.”

§ 1802(5)(i). And Plaintiffs have provided no reason to think that committee erred. Thus, the

24 Service’s explanation for the action is “rational and supported by the record.” C & W Fish Co.,

.

On to the fair-and-equitable requirement. Courts do not “second-guess” the Service’s al-

location decision “simply because [it has] a greater impact upon one group or type of fishermen.”

N.C. Fisheries Ass’n,

(quotation omitted). Of course, every allocation that

helps one sector hurts the other.

So harm to one sector does not violate the standard where the

harm is “outweighed by the total benefits received by another group or groups.” C & W Fish Co.,

(quoting the regulation now codified at

(c)(3)(i)(B)). But an

allocation that “permanent[ly] disadvantage[s]” one sector while failing to consider an important

aspect of the problem is not fair and equitable and does not comply with national standard 4. See

Guindon II,

(emphasis omitted).

Plaintiffs say A53 is unfair because it doubly diminishes the commercial catch limit. ECF

No. 20 at 33–34. That is so because it first reduces the commercial limit in allocating more of the

total catch limit to recreational anglers, causing a second reduction in the commercial limit via the

required changes to the total catch limit to account for increased bycatch.

The second effect,

Plaintiffs claim, is a forced “subsid[y]” of recreational fishing by the commercial sector. Id. at 34.

Moreover, because the Service rejected that characterization, they say the Service’s explanation

for the allocation was irrational. Id.

The Court perceives neither unfairness nor irrationality in A53. As the Court has ex-

plained, the supposed double penalty Plaintiffs decry results from an adjustment to the total catch

limit that accounts for the nature of recreational fishing and the rules that apply to it, which are

outside the scope of this case. Their position thus equates to saying that any increase in the allo-

25 cation to the recreational sector is unfair. And that position is untenable because the MSA explic-

itly protects recreational fishing. 9 And the Service explained that the MSA requires it to preserve

recreational opportunities and that A53’s allocation “provides the greatest overall benefit to the

Nation with respect to both food production and recreational opportunities”—that is, that the rec-

reational benefits of the allocation outweigh the commercial harms. AR 17423. The Court has no

basis to question that judgment. N.C. Fisheries Ass’n,

; C & W Fish Co.,

.

Plaintiffs also see unfairness in the fact that the allocation is based on sector data that “are

not comparable in reliability or accuracy.” ECF No. 38 at 17. Specifically, the data from the

commercial sector are more accurate because, unlike those from the recreational sector, they are

tracked, not estimated. ECF No. 20 at 35. That fact, they say, “penalizes the commercial sector.”

And they analogize that disparity to Guindon II, in which the court held that a management

plan was not fair and equitable because it structurally disadvantaged the commercial sector—alt-

hough Plaintiffs concede the analogy is imperfect.

at 35–36.

The analogy is less than imperfect; it is inapposite. The problem in Guindon II was that a

reallocation based on revised estimates of prior recreational catch included data from years in

which recreational anglers had exceeded their quota.

. The commercial

sector, meanwhile, could not exceed its quota because of the strict tracking regime in place.

Thus, only the recreational sector conceivably could get its allocation increased.

Plaintiffs cannot identify a similar structural disparity. Here, that the commercial monitor-

ing regime is stricter than that for the recreational sector does not necessarily advantage either

9 See, e.g.,

(33)(A) (defining the “optimum” “yield from a fishery” to include consideration of both “food production and recreational opportunities”).

26 sector—at least when it comes to reallocation—because estimation errors run in both directions.

So it appears equally possible that the Service will underestimate recreational catch and allocate

to recreational anglers fewer grouper than it otherwise would have, disadvantaging that sector. In

any event, the Court cannot conclude that A53’s allocation structurally disfavored either party, and

so cannot hold for Plaintiffs on this claim.

Finally, Plaintiffs assert that A53 is unfair because it “was not designed to promote any

particular [fishery-management-plan] objective.” ECF No. 20 at 36–37. They point to the Ser-

vice’s regulation interpreting national standard 4, which explains that “[t]he motive for making a

particular allocation should be justified in terms of the objectives of the [fishery management

plan].”

(c)(3)(i)(A). Although that regulation does not have the force of law,

the APA requires an agency to “comply with its own regulations.” Nat’l Env’t Dev. Ass’n’s Clean

Air Project v. E.P.A.,

(D.C. Cir. 2014) (quotation omitted).

One obstacle to this claim is that A53 purports to promote specific management objectives.

In the final rule, the Service explained the allocation was justified by the goals of “preventing

overfishing and promoting stability in the fishery by allowing for enhanced fisher flexibility and

increasing fishing opportunities to the extent practicable.” AR 8264. Plaintiffs protest that this

explanation was added by the Service after the council finalized the recommendation the Service

adopted. ECF No. 38 at 17–18. This fact, they say, “proves” that those “objectives[ ] . . . were an

afterthought,” not the “‘motivation’ for the amendment.”

at 17–18.

The Court will not read so much into the timing of that language’s inclusion in the rule.

A53 may have been based on the council’s and the scientific committee’s recommendations, but

it was ultimately the Service’s decision whether to promulgate it.

(a)(1)(B),

27 (2)–(3). So the relevant “motive” belonged to the Service. And both its explanation and the ad-

ministrative record as a whole show a “rational connection” between A53 and “legitimate” man-

agement objectives.

(c)(3)(i)(A).

At bottom, Plaintiffs offer nothing that can overcome the “high degree of deference” the

Court owes to the Service’s assessment of the fairness of A53. Nat. Res. Def. Council,

. Plaintiffs, then, cannot prevail on their claims under national standard 4.

2. National Standard 9

National standard 9 directs that a management plan must, “to the extent practica-

ble, . . . minimize bycatch and . . . to the extent bycatch cannot be avoided, minimize the mortality

of such bycatch.”

(a)(9). That directive is qualified by the phrases “to the extent

practicable” and “minimize.” Thus, the Service must balance minimizing bycatch with “compet-

ing environmental and economic considerations.” The Ocean Conservancy v. Gutierrez,

(D.D.C. 2005). The Court’s review is limited to ensuring the Service rationally

“evaluated” whether A53 satisfied this mandate. See Flaherty v. Bryson,

(D.D.C. 2012).

Plaintiffs recognize that national standard 9 is a qualified directive. ECF No. 20 at 37.

Still, they say, A53 “presents a novel question” because it increases bycatch without countervailing

“biological benefit[s].”

. They expect A53 to “reduce yields, increase the risk of over-

fishing, and increase bycatch.”

(citations omitted). So, the argument goes, A53 does not “re-

flect a rational balancing of the MSA’s objectives.”

Plaintiffs’ position conflicts with the record. True, for the reasons the Court has already

explained, allocating more fish to recreational anglers will reduce the yield the stock can bear

without depleting it. AR 9686. But the record disclaims a rise in the risk of overfishing. The

28 Service found that effect conceivable, but explained that it was “unlikely that any of the [alterna-

tives at issue] will result in any significant increase in [the risk of overfishing].” AR 8082. The

Service so found because the effective quota was adjusted to account for the allocation to the

recreational sector.

By the same token, A53 is not estimated to increase bycatch. Bycatch is

higher than it would be for a plan with the same total catch limit and a higher allocation to the

commercial sector, see AR 17306, but such a plan is not the relevant benchmark. Again, the ef-

fective quota was adjusted to account for the allocation, and anyway, A53 reduced the total catch

limit by 19 percent. See AR 7967.

Given that reduction and its attendant decrease in bycatch, the Court cannot agree that A53

lacks countervailing “biological benefit[s].” ECF No. 20 at 38. The Service’s explanation of why

A53 satisfies national standard 9 is rational and supported by the record.

3. National Standard 2

National standard 2 provides that a management plan must be “based upon the best scien-

tific information available.”

(a)(2). The plans, in other words, must be “dili-

gently researched and based on sound science.” The Ocean Conservancy,

.

The Service’s decisions on scientific matters deserve special deference because they “implicate

substantial agency expertise.”

So “[a]bsent some indication that superior or contrary data was

available and that the agency ignored such information, a challenge to the agency’s collection of

and reliance on scientific information will fail.” N.C. Fisheries Ass’n,

.

Plaintiffs make just such a claim. The Service justified A53 partly based on an estimate of

its economic consequences. See AR 8071–77. One relevant input is the estimated value that an

average recreational angler assigns to the right to keep a second red grouper. See AR 8074. The

Service based its estimate of that value on a 2012 paper.

Plaintiffs observe the authors of the

29 2012 paper “published a new paper in 2021” that estimated a lower value to recreational fishermen.

ECF No. 20 at 44–46. But the Service did not incorporate that analysis in its projections.

The Service gave two reasons why it omitted that paper. First, it said the new paper “was

not available” when the economic analysis underlying A53 was done. AR 17431–32. Second, it

explained that the 2021 paper and the 2012 paper answer different questions. The former estimates

the “economic value associated with different bag limits”—i.e., the number of fish a recreational

angler may catch in a given day; the latter approximates the “economic value per fish given a

change in expected harvest.”

In its judgment, A53 “require[d] an estimate of the latter.”

So that was what it used.

Plaintiffs attack the first explanation without acknowledging the second. See ECF No. 20

at 45–46; ECF No. 38 at 24–25. This issue, they say, “largely boils down to whether” the 2021

paper was “available” during the drafting of A53’s economic analysis. ECF No. 38 at 24. Not so.

The Service’s second explanation satisfies national standard 2, so the Court need not reach

the first. The Service “may choose between conflicting facts and opinions, so long as it justifies

the choice.” Ctr. for Biological Diversity v. Blank,

(D.D.C. 2013) (quo-

tations omitted and alteration adopted). A determination that one method more precisely answers

the relevant question is such a justification. See, e.g., Fishermen’s Finest, Inc. v. Locke,

, 896–97 (9th Cir. 2010). Especially under the greater deference owed to that technical assess-

ment, the Court has no reason to find fault with that explanation. See Nat’l Fisheries Inst. v.

Mosbacher,

, 223–24 (D.D.C. 1990). So it cannot sustain this challenge to A53.

B. Other MSA Requirements

Plaintiffs also bring claims based on two miscellaneous MSA requirements. First, they say

the Service has failed to ensure accountability for dead discards. ECF No. 20 at 39–42 (citing 16

(a)(15)). Second, they contend that A53 does not include a “standardized bycatch

reporting methodology.”

at 42–44 (citing

(a)(11)). Neither claim succeeds.

1. Accountability Measures

Management plans must “establish a mechanism for specifying annual catch limits in the

plan . . . , including measures to ensure accountability.”

(a)(15). Such measures

“are management controls to prevent [catch limits] from being exceeded, and to correct or mitigate

overages of [catch limits] if they occur.” Flaherty,

(quoting

(g)(1)) (alteration adopted). Effective controls require the Service “accurately [to] mon-

itor catch [and bycatch] during the fishing season.” Oceana, Inc. v. Locke (Oceana I),

(D.D.C. 2011).

Plaintiffs deem that requirement unsatisfied because the Service “does not monitor or track

dead discards against a predetermined annual mortality limit.” ECF No. 38 at 19. Instead, as

Plaintiffs explain, the Service monitors the amount of fish that recreational anglers bring to shore

because the quotas are expressed in those terms—based on an estimated corresponding number of

dead discards.

at 19–20. That method, they contend, is too inaccurate to ensure accountability.

ECF No. 20 at 41–42.

Plaintiffs rely on language from Oceana I, but that case does not support their position.

The court in Oceana I required an accountability measure to be “mandatory and sufficiently spe-

cific.”

. But the quota system at issue there—unlike that in this case—

“require[d] in-season bycatch reports that measure discards in near real time.”

. That

was so because the sector allocations under that management plan explicitly used discards as an

independent input. See Amendment 16 to the Northeast Multispecies Fishery Management Plan,

, 18276 (Apr. 9, 2010). Thus, much of the discussion in Oceana I concerned

31 the level of bycatch monitoring by “at-sea observers,” who assessed the accuracy of anglers’ by-

catch reports. See 831 F. Supp. 2d at 110–13. In that context, the word “mandatory” meant that

the Service had to “dispatch” observers at a certain rate.

. It has no comparable meaning

here, where the estimation of total fishing mortality occurs in the aggregate, not by requiring com-

pliance by individual anglers. Moreover, that case recognized that the level of specificity needed

depends on “the task at hand.”

. In the end, the language Plaintiffs pluck from Oceana I

is not the relevant legal standard, and comparison between the cases is inapt because of their very

different management plans.

Given the quota system in place here, the Court sees no reason to find the accountability

measures inadequate. The recreational quota here is expressed in landings, and that number ac-

counts for other sources of fishing mortality via estimations that originate with the latest stock

assessment. AR 17422. The Service monitors recreational landings, and when landings are pro-

jected to reach the quota, the Service must “close the recreational sector for the remainder of the

fishing year.”

(e)(2)(i). So long as the estimates of total fishing mortality are

consistent, that procedure will ensure the recreational sector stays beneath the total fishing mortal-

ity projected in the last stock assessment and incorporated into the recreational catch limit. Lack-

ing any reason to believe those estimates produce inconsistent results, the Court finds the account-

ability measures in place adequate. 10

10 Plaintiffs also assert that the Service violated its own regulations by failing to account for dead discards in setting the overfishing limit—i.e., the amount of catch above which the stock’s maximum sustainable yield diminishes. ECF No. 20 at 39–40; ECF No. 38 at 21. They observe that the regulatory definition of the overfishing limit refers to an “amount of catch,” not merely to landings. See

(e)(2)(i)(D). But the overfishing limit recommended by the scientific council, as the Service has explained, accounts for all sources of mortality, including bycatch, because the stock assessment factors in that mortality. AR 7357, 17422. Because the annual catch limits are based on that overfishing limit, the annual catch limits account for bycatch

32 2. Standardized Bycatch Reporting Methodology

Management plans also must “establish a standardized reporting methodology to assess the

amount and type of bycatch occurring in the fishery, and include conservation and management

measures that, to the extent practicable and in the following priority . . . (A) minimize bycatch;

and (B) minimize the mortality of bycatch which cannot be avoided.”

(a)(11).

Plaintiffs claim that the applicable fishery management plan contains no satisfactory reporting

methodology because “the recreational sector is not required to report bycatch and existing volun-

tary surveys are not reliable.” ECF No. 20 at 42.

That phrasing betrays the fundamental flaw in this claim: It concerns the whole manage-

ment plan, not A53. The amendment “does not purport to establish a bycatch-reporting method-

ology . . . and therefore does nothing to alter the [existing] bycatch-reporting methodology.”

Oceana I,

. In fact, A53 mentions the bycatch reporting methodology only

to explain the results of a recent review of separately enacted methodologies. See AR 17425. As

the Court has already explained, that scant reference did not reopen the issue for consideration. So

this claim “is not properly before the Court.” Oceana I,

.

C. Arbitrary and Capricious Review

Plaintiffs’ next five challenges are not tied to any specific provision of the MSA. Instead,

they point to some of the Service’s “actions, findings, and conclusions” while considering A53 as

in the same fashion. AR 17422. Plaintiffs provide no reason to think that the further step of setting an overfishing limit, after the annual catch limits are set, that more directly accounts for bycatch would help “ensure accountability” with those “annual catch limits.” See

(a)(15). Section 1853(a)(15) requires “only the establishment of [annual catch limits and accountability measures] such that overfishing does not occur.” Oceana II, 24 F. Supp. 3d at 63. The Service’s regulations recognize that this goal can be accomplished “as long as estimates of bycatch and any other fishing mortality not accounted for in the landings are incorporated into the determination of” the annual catch limit.

(f)(3)(i). That approach was used here, and it satisfies Section 1853(a)(15).

33 arbitrary and capricious. ECF No. 38 at 26. The Court’s review of these claims is limited to

ensuring that the Service “examine[d] the relevant data and articulate[d] a satisfactory explanation

for its action including a rational connection between the facts found and the choice made.” State

Farm,

(quotation omitted). In these five instances, it did.

1. Inconsistent Economic Analysis

The first of those challenges concerns the role of economic analysis in A53. The Service

calculated the expected changes in commercial and recreational-charter-ship revenue, benefit to

consumers, and benefit to recreational anglers for each alternative. AR 8071–77. The result was

an estimate of the total change in net economic benefit that each option would cause. AR 8077.

Because of the reduction in total catch, each option was expected to reduce net economic benefit.

See

But alternative 3, which became A53’s allocation, was predicted to reduce economic ben-

efit the least. See

So the Service concluded that it was “expected to result in the greatest

economic net benefits.” AR 17424.

Plaintiffs think that statement conflicts with prior analyses. ECF No. 20 at 46–47. Specif-

ically, they remind the Service that, when considering an earlier amendment, it disclaimed the

possibility of inferring changes to the recreational sector’s net economic benefit based on reallo-

cations. AR 11991. Its rationale was that the open-access-management system for that sector did

not permit the inference that a fish goes to the angler who values it most highly. See

So,

Plaintiffs say, the Service’s estimation of net economic benefit when analyzing A53’s proposed

alternatives was a sudden, unexplained change of position. ECF No. 20 at 47.

The Service denied any inconsistency when it promulgated A53. It said, essentially, that

the statements relate to slightly different analytical questions. AR 17432. The prior statement

described an inability to maximize net economic benefit, it explained, while the A53 analysis relied

on the ability to compare estimated changes in net economic benefit among alternatives.

34 Plaintiffs see no difference in those applications. ECF No. 20 at 47; ECF No. 38 at 27–28.

They point out that the earlier analysis rejected the idea that the Service could assess changes in

net economic benefit to the recreational sector at all. See AR 11991.

To the extent the Service’s explanation conflicts with prior analyses, the inconsistency is

not serious enough to warrant vacatur. Inconsistencies in evaluations of the facts violate the APA

if the agency’s explanation “was so incomplete and conclusory as to fall below the standards of

reasoned decisionmaking” and the agency “relied heavily” on the analysis in question. Gen. Chem.

Corp. v. United States,

, 855 (D.C. Cir. 1987). Neither condition is true here.

The Service explained that it considered a technique of economic analysis useful in one

situation and not another. Even assuming the situations are not meaningfully different, that change

would not present a “logical[ ] inconsisten[cy]” in its explanation, but a poor use of the technical

tools that are “appropriately entrusted to [its] expertise.” Cf. Public Health Rsch. Grp. v. Pizzella,

, 29–30 (D.D.C. 2021). And the Service’s explanation articulates a distinction

between overall maximization of economic benefit and comparison of benefit among alternatives

that can be applied in “a rational and consistent manner.” Gen. Chem. Corp., 817 F.2d at 854.

In any event, the administrative record does not show that the Service “relied heavily” on

its economic analysis. Gen. Chem. Corp., 817 F.2d at 855. The stated purpose of A53 was to

allocate red grouper based on new estimates of the sectors’ participation in the fishery during a

particular period, not to maximize net economic benefit. AR 7981. The only plausible reference

to economic analysis in the stated “need” for A53 is in the expressed desire to achieve the “opti-

mum yield on a continuing basis” Id. But deciding which yield is “optimal” requires many con-

siderations beyond economic analysis. And the stated need also recited the requisites of using the

“best scientific information available”—i.e., FES—of ensuring that catch limits reflect historical

35 fishing behavior, and of achieving consistency between the data used to set catch limits and that

used to monitor catch. Id. The record reflects that the Service selected A53’s allocation mainly

because it wished to keep its policy consistent by basing the allocation on “the same timeframe as

[A30B]” while also using “FES landings” to set the limits—not because of its projected economic

effects. AR 7967.

2. Failure to Account for New Findings

Plaintiffs’ second challenge arises from the Service’s updated understanding, thanks to

FES, of the proportion of caught fish that recreational anglers discard. ECF No. 20 at 47–48. That

new information, they reason, made it arbitrary to base A53’s allocation on relative participation

during the same years used by A30B. Id. After all, in adopting A30B, the Service noted that the

dead-discard rates of the sectors were then estimated to be roughly equal. AR 10483.

The Court disagrees. As the Court has explained more than once, the Service factored its

new assessment of the dead-discard rate into A53 in other ways, such as by adjusting the catch

limits. Its analysis of that information is woven throughout A53. See, e.g., AR 7961, 7969, 7978,

7988, 8016. Its explanation for selecting that period is “satisfactory.” State Farm,

.

3. Acting from an Unlawful Baseline

Plaintiffs’ third challenge attacks the Service’s 2019 emergency allocation. Noting that the

rule then in place did not mention either “CHTS” or “FES” “units,” Plaintiffs complain that the

Service allowed recreational anglers to catch more fish than the rule allowed (at least if measured

in “FES units”). ECF No. 20 at 48–49. They also assert that implementing the emergency rule

“obscured” and “tainted” the analysis supporting A53. ECF No. 20 at 49; ECF No. 38 at 29.

This claim is way off the mark. To the extent Plaintiffs claim the emergency rule or its

enforcement was unlawful, their challenge is untimely because A53’s allocation displaced that of

36 the emergency rule.

(f)(1). And to the extent they claim the Service’s explana-

tion was confusing or inconsistent because of the emergency rule, they are wrong. It has always

been clear that the Service understands a limit of 1.0 million pounds of red grouper in CHTS units

and a limit of 2.1 million pounds of red grouper in FES units to allow the same amount of real-

world catch. AR 7967. That clarifies any purported inconsistency or ambiguity in the explana-

tion—whether or not the Service is right. Accordingly, this claim must fail.

4. Failure to Use the Best-Available Landings Data

Plaintiffs’ fourth challenge takes aim at the data used to estimate landings. They say these

data are “manipulate[d],” that they produce multiple estimates among which the Service “picks

and chooses,” that some data are not public, that the estimation process is “byzantine and incoher-

ent,” and that the estimates conflict with those produced by a Florida survey. ECF No. 20 at 50–51.

For those flaws, they contend, the Service has yet to provide “a satisfactory explanation.” ECF

No. 38 at 30.

The Court declines the invitation to second-guess the agency’s statistical estimation proce-

dures. The Service explained how it estimates landings and provided links to fuller, scholarly

descriptions of its methods. AR 17426. Its decisions about the most appropriate estimation pro-

cedures are due “an extreme degree of deference.” Huls Am. Inc. v. Browner,

(D.C. Cir. 1996). At bottom, Plaintiffs provide no basis within the Court’s competence to over-

come that deference. And they cite no legal authority suggesting the Service has to make the

“confidential” parts of its data public. AR 17426. So this claim fails.

5. Use of a Flawed Calibration Model

Similarly, Plaintiffs’ fifth and final APA challenge targets the model used to convert be-

tween CHTS and FES. They point to the changing attributes of CHTS over time, to the inability

externally to verify the accuracy of any model based on self-reported data, and to the inherent

37 difficulty of using a model “to retrospectively revise landings estimates from decades earlier.”

ECF No. 20 at 51–53. They say this approach was not “credible” and that the Service “has not

provided a rational explanation” to the contrary. Id. at 53.

Again, Plaintiffs have identified nothing to overcome the deference due to the Service’s

judgment on technical matters. A scientific-review panel considered the attributes of the model

that the Plaintiffs identify. See AR 18693–98. It concluded that the model is “a reasonable and

scientifically defensible estimation approach.” AR 18693. The relevant scientific committee then

tested the model when it conducted the 2019 stock assessment. See AR 17423. And both that

committee and the Service have evaluated the model, including the critiques of Plaintiffs and oth-

ers, and concluded that it produces “the best scientific information available.” Id. The Court

cannot disagree. So this claim, too, fails.

D. NEPA

Plaintiffs’ final set of claims arises under NEPA, which requires agencies to “assess the

environmental consequences of ‘major [f]ederal actions’ by following certain procedures during

the decision-making process.” Nevada v. Dep’t of Energy,

(D.C. Cir. 2006) (quot-

ing

(2)(C)). The requirements relevant here compel the Service to prepare an

environmental impact statement that details the expected consequences of a proposed action, to

take a “hard look” at those consequences, and to “consider all reasonable alternatives” that would

advance the agency’s goals. Flaherty v. Raimondo,

, 85–87 (D.D.C. 2021).

Plaintiffs bring two NEPA challenges. First, they assert that the Service did not truly weigh

alternatives when it adopted A53. ECF No. 20 at 54. In other words, they say the agency prede-

termined the outcome.

Second, they argue the range of alternatives the Service putatively

considered was unreasonable.

. Neither challenge has merit.

38 The first challenge requires Plaintiffs to show that the Service “irreversibly and irretrieva-

bly commit[ted] itself to a [covered] plan of action . . . before [it] completed [an] environmental

analysis.” Loper Bright Enters. v. Raimondo,

, 121 (D.D.C. 2021) (emphasis

deleted). As Plaintiffs concede, that is a high bar. ECF No. 20 at 54. They purport to have cleared

it because the Service not only committed to the policy of A53 before performing the required

analysis, but it actually “implemented the very action being evaluated.”

Plaintiffs refer to the

emergency rule that contained the same allocation proportions as A53.

Plaintiffs’ theory doesn’t hold water. The emergency rule did not bind the Service’s hands

on A53’s allocation. When it enacted A53, it was free to select any allocation consistent with the

national standards and other applicable law. Plaintiffs have no authority suggesting that the emer-

gency rule did anything “irreversible and irretrievable.” ECF No. 20 at 54. Indeed, the whole

point of the rule was to create an “interim” system that would “remain in place until the next stock

assessment.” AR 7977.

Plaintiffs’ second NEPA theory fares no better. They complain that the Service did not

consider alternatives in which discards were included in the overfishing limit. Such an approach,

they say, would have furthered the Service’s goals more than the alternatives that were considered.

ECF No. 20 at 55.

In rejecting a prior, similar challenge, the Court has already explained why that is not so.

In short, it would not have furthered the Service’s goals to factor the bycatch consequences of the

allocation into the overfishing limit because the same effect is captured elsewhere. An agency

need not consider an alternative that does not further its goals. See Oceana I, 831 F. Supp. 2d

at 126–27. So, after considering the agency’s stated goals, the Court finds the Service’s range of

alternatives reasonable, and Plaintiffs’ challenge fails.

39 Conclusion

For the above reasons, the Court will deny Plaintiffs’ motion for summary judgment and

grant Defendants’ and Defendant-Intervenors’ cross-motions for summary judgment. A separate

order will issue.

/s/ Timothy J. Kelly TIMOTHY J. KELLY United States District Judge Date: January 6, 2023

40