Natural Resources Defense Council, Inc.; Oceana, Inc., Plaintiffs-appellants, v. National Marine Fisheries Service; Donald Evans, Secretary of Commerce; National Oceanic and Atmospheric Administration, Defendants-appellees, Andwest Coast Seafood Processors Association; Fishermen's Marketing Association, Defendants-intervenors-appellees, 421 F.3d 872 (9th Cir. 2005)

Appellee National Marine Fisheries Service ("the Agency") set 2002 fishing limits for four species of Pacific groundfish that are commonly sold as "red snapper." Appellant Natural Resources Defense Council ("NRDC"), an environmental organization, brought suit in federal district court challenging the four limits as violating the Magnuson-Stevens Fishery Conservation and Management Act ("the Magnuson Act" or "the Act"), 16 U.S.C. §§ 1801 et seq., which directs that the Agency prevent overfishing; the Administrative Procedure Act ("APA"), 5 U.S.C. §§ 501 et seq., which directs agencies to consider relevant factors in setting such limits; and the National Environmental Policy Act ("NEPA"), 42 U.S.C. §§ 4321 et seq., which directs agencies to prepare adequate environmental analyses when undertaking such actions. The district court granted summary judgment to the Agency and intervenor-appellees Fishermen's Marketing Association and West Coast Seafood Processors Association ("Intervenors"). Because we conclude that the 2002 darkblotched rockfish limit was based on an impermissible construction of the Act, we reverse and remand; we affirm the limits as to the other three species.

A. The National Marine Fisheries Service, the Magnuson Act, Section 1854 and the National Standards Guidelines

Congress enacted the Magnuson Act to "conserve and manage the fishery resources found off the coasts of the United States." 16 U.S.C. § 1801(b) (1).¹  The Agency is charged with developing and implementing rebuilding plans for overfished fish species. § 1854.²  In 1996, Congress amended the Act by passing the Sustainable Fisheries Act ("SFA"). Pub. L. No. 104-297, 110 Stat. 3559 (1996). The SFA added new requirements to the Act to accelerate the rebuilding of overfished species.

The Act, as amended by the SFA, contains a provision the proper interpretation of which is the main subject of this appeal. Section 1854 of the Act provides in part that when any species is found to be overfished, the Agency must approve a rebuilding plan that:

(A) specif [ies] a time period for ending overfishing and rebuilding the fishery that shall —

(i) be as short as possible, taking into account the status and biology of any overfished stock of fish, the needs of fishing communities, . . . and the interaction of the overfished stock of fish within the marine ecosystem; and

(ii) not exceed 10 years, except in cases where the biology of the stock of fish, [or] other environmental conditions. . . dictate otherwise.

§ 1854(e) (4).

The Act also sets forth a series of "national standards" with which any rebuilding plans must be "consistent," and provides for the establishment of National Standards Guidelines ("NSGs") that must be "based on the national standards" for use in "assist [ing] in the development of fishery management plans." §§ 1851(a), (b). The Act provides that NSGs "shall not have the force and effect of law." Id.

There is some ambiguity to § 1854(e) (4). Section 1854(e) (4) (i) specifies that the rebuilding time period be as "short as possible," but also directs that the Agency "tak [e] into account the status and biology of [the] . . . overfished stock" and "the needs of fishing communities." Section 1854(e) (4) (ii) in turn plainly mandates that the rebuilding plan be no longer than 10 years, so long as biologically or environmentally possible.³  However, if it is not possible to rebuild within 10 years, the Act is not clear as to the exact limits on the length of the rebuilding period.

Seeking to clarify the proper interpretation of § 1854(e) (4), the Agency in 1997 sought "comment on whether or not it is correct in its interpretation that the duration of rebuilding programs should not be unspecified and, if so, what factors should be considered in determining that duration." See 62 Fed.Reg. 67,610 (Dec. 29, 1997). The Agency propounded two alternate interpretations for public comment: that whenever it would take longer than 10 years to rebuild an overfished species, either (1) all fishing of that species would be banned until the rebuilding was complete or (2) the Agency would set a ceiling on the rebuilding duration that would be reached by adding the shortest possible time to rebuild plus "one mean generation time . . . based on the species' life-history characteristics." Id. at 67,609-10. A "mean generation time" is a scientific term, not mentioned in the Act itself, measuring how long it will take for an average mature fish to be replaced by its offspring. After notice and comment, the Agency adopted the second interpretation in a NSG ("the 1998 NSG"). See 50 C.F.R. § 600.310(e) (4) (ii) (B). The Agency reasoned that:

[f]or stocks that will take more than 10 years to rebuild, the guidelines [adopted] impose an outside limit that is objective, measurable, and linked to the biology of the particular species. . . . The guidelines strike a balance between the Congressional directive to rebuild stocks as quickly as possible, and the desire . . . to minimize adverse economic effects on fishing communities. For stocks that cannot be rebuilt within 10 years, the guideline allows flexibility in setting the rebuilding schedule beyond the no-fishing mortality period, but places a reasonable, species-specific cap on that flexibility by limiting the extension to one mean generation time.

63 Fed.Reg. 24,217 (May 1, 1998).

B. The 2001 and 2002 Limits for Darkblotched Rockfish

The Pacific Coast Groundfish Fishery is one of the fisheries the Agency oversees, covering the bottom-feeding fish species dwelling in the waters off the coasts of California, Oregon and Washington. In 2000, the Agency assessed the status of one species of Pacific groundfish within the fishery — darkblotched rockfish. It found that the species was at 22% of its unfished population level (its predicted level absent any fishing), and therefore concluded that the species was "overfished" within the meaning of the Act. 66 Fed.Reg. 2,347, 2349-50 (Jan. 11, 2001). The Agency further concluded that the species could be rebuilt in 10 years or less, triggering § 1854(e) (4) (ii)'s mandatory requirement that the rebuilding take place within 10 years. The Agency then set a 130 metric ton "fishing harvest level," or quota, i.e., a set limit of darkblotched rockfish that could be fished in 2001.

In 2001, the Agency updated its assessment of darkblotched rockfish and concluded that it had significantly overestimated the health of the species. The Agency now estimated that the species was almost twice as depleted as previously thought — it was at only 12% of its unfished population level. In the Agency's calculations, rebuilding therefore could not be accomplished within 10 years; the minimum period for rebuilding was now 14 years.

This increased rebuilding time meant, by necessity, that the rebuilding plan was no longer limited by § 1854(e) (4) (ii)'s mandatory 10-year cap; instead, the only applicable statutory time limit was § 1854(e) (4) (i)'s command that the rebuilding period be "as short as possible." Further, according to the interpretation of the Act set forth in the 1998 NSG, the revised minimum rebuilding period triggered a new ceiling that was the 14-year period plus "one mean generation time," which in the case of the long-lived darkblotched rockfish was 33 years. The Agency, in short, switched from operating under the statutory constraint of 10 years rebuilding time to a new constraint, dictated by the 1998 NSG, of 47 years. The Agency then set a "target" rebuilding time of 34 years, and in accordance with this target, raised the fishing level harvest for 2002 from the previous year's 130 metric tons to 168 metric tons.⁴ 

NRDC brought suit alleging that the new quota violated the Act, the Administrative Procedure Act and the National Environmental Policy Act. The district court concluded that the quota violated none of these statutes and granted summary judgment for the Agency. Natural Res. Def. Council, Inc. v. Nat'l Marine Fisheries Serv., 280 F. Supp. 2d 1007, 1014-15 (N.D. Cal. 2003).

C. The Agency's 2002 Specifications for Three Other Groundfish Species

The Agency also set 2002 quotas for three other overfished groundfish species — bocaccio, cowcod and canary rockfish — that were identical to the levels set in 2001, despite evidence that fishing of these overfished species in the prior two years had been significantly higher than that allowed by the previous year's quotas. The Agency reasoned that because it did not have newly available data as to the status of these species (owing to its policy of conducting stock assessments every three years), its response to the evidence of overfishing would be to put in place interim measures (such as establishing no-fishing zones in certain areas), and then to set new quotas once the next assessment was completed.

NRDC charged that the Agency's failure to adjust the quotas violated the Act, APA and NEPA. The district court also granted the Agency summary judgment on these claims. 280 F. Supp. 2d at 1017-1018.

We review de novo the district court's grant of summary judgment. Turtle Island Restoration Network v. Nat'l Marine Fisheries Serv., 340 F.3d 969, 973 (9th Cir. 2003). "De novo review of a district court's judgment concerning a decision of an administrative agency means the court views the case from the same position as the district court." Id.

The Administrative Procedure Act dictates that we should "hold unlawful and set aside agency action . . . [that is] arbitrary, capricious, an abuse of discretion, or otherwise not in accordance with law." 5 U.S.C. § 706(2) (A); see also § 1855(f) (1) (B); Midwater Trawlers Co-op v. Dep't of Commerce, 282 F.3d 710, 716 (9th Cir. 2002) (recognizing that the Magnuson Act adopts APA's standard of review). We must also "determine whether the agency articulated a rational connection between the facts found and the choice made." Ariz. Cattle Growers' Ass'n v. United States Fish & Wildlife, 273 F.3d 1229, 1236 (9th Cir. 2001) (citation omitted).

We should not defer to an agency's interpretation of a statute if Congress' intent can be clearly ascertained through analysis of the language, purpose and structure of the statute. Chevron, U.S.A., Inc. v. Natural Res. Def. Council, Inc., 467 U.S. 837, 842-43, 104 S. Ct. 2778, 81 L. Ed. 2d 694 (1984). If, however, Congress' intent is not clear, and if "Congress delegated authority to the agency generally to make rules carrying the force of law, and [ ] the agency interpretation claiming deference was promulgated in the exercise of that authority," United States v. Mead Corp., 533 U.S. 218, 226-27, 121 S. Ct. 2164, 150 L. Ed. 2d 292 (2001), then we must defer to the agency's construction of the statute so long as "the agency's answer is based on a permissible construction of the statute." Chevron, 467 U.S. at 843, 104 S. Ct. 2778. If the Mead requirements for Chevron deference are not met, we review the agency's interpretation under the Skidmore standard, whereby the interpretation is "entitled not to deference but to a lesser `respect' based on the persuasiveness of the agency decision." Wilderness Soc'y v. U.S. Fish & Wildlife Serv., 353 F.3d 1051, 1067(9th Cir. 2003); Skidmore v. Swift & Co., 323 U.S. 134, 65 S. Ct. 161, 89 L. Ed. 124 (1944).

The Agency arrived at its increased 2002 darkblotched rockfish quota by applying its interpretation of § 1854(e) (4) of the Magnuson Act as set forth in the 1998 NSG. NRDC argues that this interpretation of the Act is not entitled to Chevron deference for two separate reasons.

First, NRDC argues that Congress' intent in this section of the Act is clear, thereby precluding the need for any deference to the Agency's interpretation of the statute. We disagree. As we noted above, § 1854(e) (4) (ii) is explicit that if a species can be rebuilt within 10 years, it must be. But § 1854(e) (4) (i), which states that the rebuilding period must be as "short as possible, taking into account the status and biology of any overfished stock of fish [and] the needs of fishing communities," introduces an ambiguity into the calculus. When it is not biologically possible to rebuild within 10 years, may the Agency extend the rebuilding period beyond the shortest possible rebuilding time to account for the needs of fishing communities? It would be possible to resolve the ambiguity by concluding that the Act as a whole makes it clear that the needs of fishing communities are perfectly aligned with the environmental goal of rebuilding fish stocks in as short a time as possible. But if this were the case, the language "the needs of fishing communities" would be redundant (as these needs would be no different than the need to rebuild stocks in as short a time as possible). But see Natural Res. Def. Council, Inc. v. Daley, 209 F.3d 747, 753 (D.C. Cir. 2000) (" [W]e reject the District Court's suggestion that there is a conflict between the [Act's] expressed commitments to conservation and to mitigating adverse economic impacts.").⁵  There is therefore an ambiguity in this part of the statute, requiring interpretation. See Chevron, 467 U.S. at 843, 104 S. Ct. 2778(holding that " [if] Congress has not directly addressed the precise question at issue," it is necessary to move to the second step of the Chevron analysis).

NRDC next argues that because the Act explicitly provides that NSGs do not have the force of law, Chevron deference is not appropriate. See Mead, 533 U.S. at 226-27, 121 S. Ct. 2164(holding Chevron deference to be appropriate only if "Congress delegated authority to the agency generally to make rules carrying the force of law, and [ ] the agency interpretation claiming deference was promulgated in the exercise of that authority"). The Agency responds that although the 1998 NSG does not have the force of law, the 2002 darkblotched rockfish quota itself — which is what is actually being challenged here — is a binding regulation that does have the force of law, requiring Chevron deference, and that to hold otherwise would mean punishing the Agency for taking the additional step of setting out the interpretation in an NSG.

We need not resolve this question here, because even under the Chevron standard of review, the 2002 quota was based on an impermissible construction of the Act. We therefore will assume that Chevron review is appropriate even as to the 1998 NSG's statutory interpretation that was applied to reach the quota, without deciding the issue.

2. The 2002 Quota Is Based on an Impermissible Construction of the Act

Under Chevron, we must determine whether "the agency's [quota] is based on a permissible construction of the statute." Chevron, 467 U.S. at 843, 104 S. Ct. 2778. Chevron review is also described as determining whether the quota reflects "a reasonable interpretation" of the statute. Id. at 844, 104 S. Ct. 2778.

The interpretation of § 1854(e) (4) stated in the 1998 NSG, as applied in the 2002 quota, is not a permissible (or reasonable) construction of the statute; it is directly at odds with the text and purpose of the Act.⁶  Section 1801 of the Act contains its "Findings, purposes and policy." The "Findings" section states that the nation's fishery resources "constitute valuable and renewable natural resources," that many of these species' "survival is threatened" and that others' survival will soon be threatened by "increased fishing pressure,. . . the inadequacy of fishery resource conservation and management practices and controls." §§ 1801(a) (1), (2). The next subsection recognizes "commercial and recreational fishing" as a "major source of employment" that "contributes significantly to the economy of the Nation." § 1801(a) (3). But even here, the Act urges that the economies of many coastal areas "have been badly damaged by the overfishing of fishery resources at an ever-increasing rate over the past decade." Id. The Act goes on to explain that " [i]f placed under sound management before overfishing has caused irreversible effects, the fisheries can be conserved and maintained so as to provide optimum yields on a continuing basis." § 1801(a) (5). These observations lead to the conclusion that " [a] national program for the conservation and management of the fishery resources of the United States is necessary to prevent overfishing, to rebuild overfished stocks, to insure conservation, to facilitate long-term protection of essential fish habitats, and to realize the full potential of the Nation's fishery resources." § 1801(a) (6). The "purposes" section adds that " [i]t is therefore declared to be the purposes of the Congress in this chapter. . . to take immediate action to conserve and manage the fishery resources . . . ." § 1801(b) (1).

The purpose of the Act is clearly to give conservation of fisheries priority over short-term economic interests. See Daley, 209 F.3d at 753(" [U]nder the [Act], the [Agency] must give priority to conservation measures."). The Act sets this priority in part because the longer-term economic interests of fishing communities are aligned with the conservation goals set forth in the Act. Without immediate efforts at rebuilding depleted fisheries, the very long-term survival of those fishing communities is in doubt. See id. This background provides helpful context for interpreting § 1854. However, even if we turn to the plain language of § 1854(e) (4) and, without such context, ask how its two subsections interact, we still must reject the interpretation of the Act contained in the 1998 NSG as it was applied to this species.

Section 1854 contains two significant mandates that constrain the Agency's options in adopting a rebuilding plan for an overfished species. First, the time period must be "as short as possible," although the Agency may take into account the status and biology of the overfished species and the needs of fishing communities. See § 1854(e) (4) (i). Subsection (i)'s commands apply to all rebuilding periods, whatever their length. Second, Congress specified a presumptive cap of 10 years on any rebuilding period, subject to exceptional circumstances beyond the Agency's control — such as an international treaty or, relevant here, "the biology of the stock of fish." See § 1854(e) (4) (ii).

We have noted some ambiguity in subsection (i)'s mandate to rebuild a species in "as short [a time period] as possible" while giving consideration to "the needs of fishing communities." The natural reading of this language, however, is that Congress intended to ensure that overfished species were rebuilt as quickly as possible, but wanted to leave some leeway to avoid disastrous short-term consequences for fishing communities. To use an example relevant here, even if a fishing community is actively seeking not to fish for a certain species, it will inevitably catch some of the overfished species in the process of fishing for other, more plentiful fish — what is known as "bycatch." Because almost no groundfish that are caught as bycatch survive even if they are thrown back into the ocean, an absolute ban on catching any of a species of groundfish could mean a total moratorium on all fishing in the parts of the fishery containing groundfish, with obvious adverse consequences for fishing communities. Section 1854(e) (4) (i), then, allows the Agency to set limited quotas that would account for the short-term needs of fishing communities (for example, to allow for some fishing of plentiful species despite the inevitability of bycatch), even though this would mean that the rebuilding period would take longer than it would under a total fishing ban.⁷ 

Reading subsection (i) in this light, it is apparent that Congress intended subsection (ii) as a limit on the Agency's discretion. The Agency may consider the short-term economic needs of fishing communities in establishing rebuilding periods, but may not use those needs to go beyond the 10-year cap set by subsection (ii). To breach this cap, the Agency may only consider circumstances that "dictate" doing so. One such circumstance, albeit not relevant here, would be an international agreement. Another that is relevant is "the biology of the stock of fish" — that is, when the current number of fish in the fishery and the amount of time required for the species to regenerate make it impossible to rebuild the stock within 10 years, even with a total moratorium on fishing. In such cases, subsection (ii) recognizes that the presumptive 10-year cap cannot apply. That said, it is manifestly unreasonable to conclude, as the Agency apparently has, that Congress intended in such circumstances to relieve the Agency of its continuing obligation to rebuild the species in a time frame that is "as short as possible."

The 2002 quota was not based on a permissible construction of the Act, because the Agency altered dramatically the balance between the needs of a species and of fishing communities with no statutorily grounded justification.⁸  NRDC argues that if the rebuilding period must exceed 10 years, the Act mandates a total moratorium on all fishing — the alternative interpretation of § 1854 that the Agency rejected when it adopted the 1998 NSG. Although NRDC's interpretation of the statute is reasonable, it is not the only reasonable one. It is also reasonable to conclude that the needs of fishing communities may still be taken into account even when the biology of the fish dictates exceeding the 10-year cap — so long as the weight given is proportionate to the weight the Agency might give to such needs in rebuilding periods under 10 years. This interpretation would allow the Agency's rebuilding periods to account for short-term concerns such as bycatch in the same manner whether the rebuilding period exceeds 10 years or not.

The 2002 darkblotched rockfish quota is patently unreasonable, however, and reflects no such measured proportionality. Freed from the 10-year cap because of the biology of the rockfish (its long regeneration time and its dire condition), the Agency simply applied the 1998 NSG's formulaic approach and increased the annual take. In 2001, the Agency set a quota of 130 million tons of darkblotched rockfish because it believed the species had been reduced to only 22% of its unfished population. When its revised estimate revealed that the species was doing much worse, the Agency expanded the fishing of the species from 130 million tons to 168 million tons, a 29% increase. Whatever the outer limits of the range of permissible constructions of the Act, we are certain that what lies beyond them is an interpretation allowing the Agency, upon discovering that a species is in significantly worse shape than previously thought, to increase dramatically the fishing pressure on that species. Increasing the annual take in these circumstances is simply incompatible with making the rebuilding period as short as possible.

We are not prepared to accept NRDC's argument that once the 10-year cap is lifted because the biology of the fish dictates it, the Act in turn dictates that the Agency can no longer consider the short-term economic needs of fishing communities at all. Such an argument, although plausible, does not appear to give due consideration to the continuing operation of subsection (i)'s command to take the needs of fishing communities into account. But neither are we prepared to accept the Agency's interpretation, which would ignore the primary mandate of subsection (i) — that the rebuilding period be "as short as possible." At least as applied here, the Agency's interpretation not only increased the fishing take by almost 30% but extended the maximum rebuilding period from less than 10 years to 47 years. Plainly, the Act does not contemplate that the Agency grant the least protection to the fish species in the worst shape.

The arguments of the Agency and Intervenors regarding potentially dire consequences for fishing communities seem persuasive at all only because they assume that the sole alternative is NRDC's strict moratorium. The district court made this same flawed assumption:

Faced with a choice between an interpretation of the [Act] that requires a moratorium on harvesting of fish species that take more than ten years to regenerate naturally, and an interpretation that permits limited harvesting over the course of a longer rebuilding period, [the Agency] selected . . . the latter interpretation. In light of [the Act's] dual conservationist and commercial objectives, an interpretation that accommodates both objectives, rather than selecting one to the exclusion of the other, is permissible.

280 F. Supp. 2d at 1014. The Agency was "faced with [this] choice" only because it proposed these two extreme interpretations, and no others.⁹ 

Our rejection of the Agency's interpretation is compelled by the language of § 1854, which requires that rebuilding take place in "as short [a time] as possible" and, if biologically possible, in less than 10 years. § 1854(e) (4). That simple command cannot be reconciled with a rebuilding period that is from 20 to 33 years longer than the biologically shortest possible rebuilding period (and that increases the annual take in the meanwhile). We hold that even granting the Agency some leeway in extending rebuilding periods when the 10-year cap is not applicable, the 2002 darkblotched rockfish quota was based on an impermissible construction of the Act.¹⁰ 

B. The 2002 Limits for Three Other Groundfish Species Do Not Violate the APA or the NEPA

NRDC additionally argues that the 2002 levels for three other groundfish species violate the APA and the NEPA. The Administrative Procedure Act requires that courts determine if agency actions are "arbitrary, capricious, an abuse of discretion, or otherwise not in accordance with law." 5 U.S.C. § 706(2) (A). The National Environmental Policy Act requires agencies to consider the environmental consequences of an action before taking it. See Robertson v. Methow Valley Citizens Council, 490 U.S. 332, 349, 109 S. Ct. 1835, 104 L. Ed. 2d 351 (1989). A reviewing court's ultimate NEPA inquiry is whether an agency has taken the required "hard look" at the environmental consequences of its action. See Greenpeace Action v. Franklin, 14 F.3d 1324, 1332 (9th Cir. 1993).

Bocaccio, cowcod and canary rockfish are three additional Pacific groundfish species that have been found to be overfished. When it set 2002 levels for these fish, the Agency simply carried over the levels from the previous year. The Agency concluded that there was no new information on these stocks to warrant changing the quotas. NRDC argues that because the Agency was aware that the actual amount of these fish that had been caught in previous years far exceeded the set quotas, the Agency should have reduced the 2002 quotas to compensate — and that its failure to do so was arbitrary and capricious in violation of the APA, and reflected a failure to take the NEPA's "hard look." The Agency argues that over- or under-harvests for a single year are accounted for through alternate mechanisms. Assessments are conducted only every three years because of budgetary constraints, so quota revisions likewise take place every three years. In the meantime, however, additional "management measures" are undertaken, such as restrictions on fishing in specific areas within the fishery, or on fishing during certain parts of the year when there is a greater chance of bycatch.

The district court concluded that: [The Agency's] decision to maintain harvest limits at their 2001 levels was reasonably connected to — indeed, was dictated by — the agency's policy of resetting harvest limits only after conducting a stock reassessment. In turn, that policy, which is a product of limited resources available to the agency to manage eighty-two different fish species, was neither an abuse of discretion nor contrary to law. . . . [T]he Court [also] finds that the EA's analysis was adequate to permit informed decision-making under the circumstances.

280 F. Supp. 2d at 1017 & n. 4.

We agree. Even if there are other reasonable approaches to dealing with the problem of exceeding quotas, we cannot say that the Agency's actions were "arbitrary, capricious," "contrary to law" or that they did not reflect a sufficiently "hard look." We therefore affirm the district court on these claims.

We reverse the district court's holding that the Agency did not violate the Magnuson Act in setting its 2002 fishing quota for darkblotched rockfish. We remand to the district court for any further proceedings consistent with this opinion. See Ocean Advocates v. United States Army Corps of Engineers, 402 F.3d 846, 871 (9th Cir. 2005) (remanding to district court to consider remedy in first instance).

We affirm the district court's holding as to the 2002 limits for the other three species of groundfish. The parties shall bear their own costs.

AFFIRMED in part, REVERSED in part and REMANDED.