Nat Resrc Def Cncl v. Daley, William M.

United States Court of Appeals FOR THE DISTRICT OF COLUMBIA CIRCUIT Argued March 24, 2000 Decided April 25, 2000 No. 99-5308 Natural Resources Defense Council, Inc., et al., Appellants v. William M. Daley, In his official capacity as Secretary of the United States Department of Commerce, et al., Appellees Pacific Marine Conservation Council and Alaska Marine Conservation Council, Amicus Curiae Appeal from the United States District Court for the District of Columbia (No. 99cv00221) Monica B. Goldberg argued the cause for appellants. With her on the briefs were Stephen E. Roady, Eric A. Bilsky, and Sarah Chasis. Deborah A. Sivas was on the brief for amicus curiae Pacific Marine Conservation Council and Alaska Marine Conserva- tion Council. James Eichner, Attorney, United States Department of Justice, argued the cause for appellees. With him on the brief were Lois J. Schiffer, Assistant Attorney General, and David C. Shilton, Attorney. Before: Edwards, Chief Judge, Henderson, Circuit Judge, and Buckley, Senior Circuit Judge. Opinion for the Court filed by Chief Judge Edwards. Edwards, Chief Judge: Paralichthys dentatus, or summer flounder, a commercially valuable species of flounder, dwell off the Atlantic coast and are harvested primarily between May and October from North Carolina to Maine. The sum- mer flounder fishery is an "overfished" fishery, in the process of recovering from severe depletion prevalent during the late 1980s and early 1990s. The Secretary of Commerce, advised by the National Marine Fisheries Service ("the Service"), the principal appellee in this case, annually sets a fishing quota limiting each year's summer flounder catch, pursuant to the Magnuson-Stevens Fishery Conservation and Management Act ("the Fishery Act"), 16 U.S.C. ss 1801-1883 (1994 & Supp. IV 1998). This case involves appellants' challenge to the Service's quota for the 1999 summer flounder harvest. Before the District Court, appellants alleged that the 1999 quota did not provide sufficient assurance that it would meet the conservation goals of the Fishery Act and attendant regulations. Appellants also claimed that the Service's con- clusion that the quota had no significant environmental im- pact was based on an inadequate environmental assessment, thereby violating the National Environmental Policy Act ("NEPA"). On cross-motions for summary judgment, the District Court granted judgment in favor of appellees. See Natural Resources Defense Council, Inc. v. Daley, 62 F. Supp. 2d 102 (D.D.C. 1999). We reverse the District Court and remand the case to the Service for further proceedings consistent with this opinion. The 1999 quota, when adopted, had a documented 18% likeli- hood of meeting the statute's conservation goals. We hold that, under the Fishery Act, the disputed quota is insufficient to meet Congress' mandate to the Service to prevent overfish- ing and to assure that specific conservation goals are met. We also hold that the Service's proposal to supplement the quota with other purportedly protective measures does not satisfactorily ameliorate the quota's glaring deficiencies. Be- cause of our disposition on these grounds, we have no need to reach appellants' NEPA claims. I. BACKGROUND A. Regulatory Background The Fishery Act was enacted to establish a federal-regional partnership to manage fishery resources. Under the statute, there are eight Regional Fishery Management Councils "to exercise sound judgment in the stewardship of fishery re- sources." 16 U.S.C. ss 1801(b)(5), 1852(a) (Supp. IV 1998). Management Councils propose and monitor fishery manage- ment plans "which will achieve and maintain, on a continuing basis, the optimum yield from each fishery." Id. s 1801(b)(4) (1994). Management Councils submit management plans to the Secretary of Commerce (functionally the Service), who may then adopt them through notice and comment rulemak- ing. See id. s 1854(a) (Supp. IV 1998). An "optimum yield" under the statute is defined as the "maximum sustainable yield from the fishery." Id. s 1802(28)(B) (Supp. IV 1998). If a fishery is "overfished," the management plan must "provide[ ] for rebuilding to a level consistent with" the maximum sustainable yield. Id. s 1802(28)(C). A fishery is "overfished" if the rate of fishing mortality "jeopardizes the capacity of a fishery to produce the maximum sustainable yield on a continuing basis." Id. s 1802(29). The Service defines overfishing and optimum yield accord- ing to the fishing mortality rate ("F"). F represents that part of a fish species' total mortality rate that is attributable to harvesting by humans, whether through capture or dis- card. Fish are "discarded" for many reasons, including, for example, when they are the wrong species, undersized, or not valuable enough. Values for F can range anywhere from 0 to over 2, and only indirectly represent the amount of fish captured by industry. For instance, an F of 1.4 means that about 20% of all summer flounder that are alive at year 1 will be alive at year 2. There is a specific F, termed "Fmax," that is defined as that fishing mortality rate that will maximize the harvest of a single class of fish over its entire life span. Overfishing is fishing in excess of Fmax. See Amendment 7 To The Fishery Management Plan for The Summer Flounder Fishery at 9 (May 1995), reprinted in Joint Appendix ("J.A.") 316. Therefore, the basic goal of a management plan is to achieve Fmax, thereby preventing overfishing and assuring optimum yield. B. The Summer Flounder Fishing Quota From a commercial standpoint, the summer flounder is one of the most important species of flounder in the United States. All parties agree that the summer flounder fishery is "overfished" and has been for some time. The Mid-Atlantic Fishery Management Council ("MAFMC"), covering New York, New Jersey, Delaware, Pennsylvania, Maryland, Virgi- nia, and North Carolina, developed the original summer flounder management plan with the assistance of two other regional Management Councils and the Atlantic States Ma- rine Fisheries Commission ("the Commission"), a consortium of 15 coastal states and the District of Columbia. The Service approved the original management plan in 1988; however, the Service has amended the plan several times. At the time relevant to the instant case, the plan was designed to achieve a fishing mortality rate equal to Fmax by 1998. Pursuant to the management plan, the Service must set a quota each year fixing the total weight of summer flounder that may be harvested by commercial and recreational fish- ers. This quota is referred to as the "total allowable land- ings" for the year, or "TAL." The Service allocates 60% of the TAL to commercial fisheries and 40% of the quota to recreational fisheries, and states receive allocations based upon their share of the summer flounder fishery. States may subdivide their allocated commercial quota between "inciden- tal" and "directed" catch. Directed fisheries intentionally harvest summer flounder. Fishers who catch juvenile flound- er, or who are part of the directed fishery for another species and catch summer flounder unintentionally, have harvested incidental catch. The TAL must meet several requirements. It must be consistent with the 10 national standards of fishery conserva- tion and management set out in the Fishery Act. See 16 U.S.C. s 1851(a)(1)-(10) (1994 & Supp. IV 1998). Most rele- vant to the instant case, the quota must embody conservation measures that "shall prevent overfishing while achieving, on a continuing basis, the optimum yield from each fishery for the United States fishing industry." Id. s 1851(a)(1) (1994). The quota must also be "consistent with" the fishery management plan. See id. s 1854(b)(1). Finally, under the applicable regulations, the Regional Administrator of the Service must annually adopt a final rule "implement[ing] the measures necessary to assure that the applicable specified F will not be exceeded." 50 C.F.R. s 648.100(c) (1999) (emphasis added). The "applicable specified F" is also referred to as the "target F." There is a relatively direct relationship between the TAL and the likelihood of achieving the target F. In general, the higher the TAL, the less likely a plan is to achieve the target F. In other words, the lower the target F, the lower the TAL must be to attain the target F. The basic dispute between the parties concerns whether the 1999 TAL provides a sufficient guarantee that the target F for summer flounder will be achieved. For 1999, the summer flounder fishery management plan mandated a target F equivalent to Fmax, which was 0.24. The Summer Flounder Monitoring Committee, a MAFMC com- mittee, had recommended a TAL of 14.645 million pounds, while MAFMC had recommended a TAL of 20.20 million pounds. The Service rejected MAFMC's recommendation as "unacceptably risk-prone" for several reasons: (1) it had an "unacceptably low probability" of 3% of achieving the target F; (2) it had a 50% probability of achieving an F of 0.36, which was "significantly higher" than the target F; (3) the proposal relied on unpredictable data; and (4) MAFMC had "yet to specify a harvest level that has achieved the annual target F." Fisheries of the Northeastern United States; Summer Flounder, Scup, and Black Sea Bass Fisheries, 63 Fed. Reg. 56,135, 56,136 (1998) (to be codified at 50 C.F.R. pt. 648) (proposed Oct. 21, 1998) ("Proposed TAL"). The Service also rejected the Summer Flounder Monitoring Committee's recommendation of a 14.645 million pound TAL. Although the Committee's recommendation had a 50% chance of achiev- ing the target F, the Service rejected the proposal without any meaningful explanation. On October 21, 1998, the Service proposed a TAL of 18.52 million pounds. See id. All parties agree that, at most, the Service's proposal afforded only an 18% likelihood of achiev- ing the target F. The Service also proposed an incidental catch restriction "to address discards in this fishery that should further reduce the overall mortality." Id. This mea- sure provided that, within the commercial fishery, 32.7% of the allocated quota be committed to incidental catch. In the end, then, the Service proposed a TAL of 7.41 million pounds for recreational harvest, 7.47 million pounds for directed commercial harvesting, and 3.64 million pounds for incidental commercial catch, for a total of 18.52 million pounds. See id. The Service also considered recent changes in minimum mesh size. On this point, the Service noted that, while MAFMC felt that the "recently adopted mesh provision requiring 5.5 inch" mesh throughout the net would "substantially reduce discard and discard mortality," the alleged benefits of mesh had yet to be verified by anyone. Id. Between the time of proposal of the 1999 TAL and its adoption, the Service concluded that it did not have the authority to impose any incidental catch restrictions on the states. Therefore, the Service merely recommended that the states adopt the incidental catch proposal, making the propos- al entirely voluntary. The Commission, the body represent- ing 15 coastal states and the District of Columbia, also declined to command the states to adopt the proposal. Ac- cording to an advisor to the Service's Assistant Administrator for Fisheries, this development "result[ed] in an unknown but probably substantial reduction in the likelihood that [MAFMC's] rebuilding schedule will be achieved," and he therefore recommended that the Service adopt the Summer Flounder Monitoring Committee's recommended 14.645 mil- lion pound TAL. See Memo from Gary Matlock to Rolland Schmitten (Nov. 25, 1998), reprinted in J.A. 208. The Service rejected this recommendation and, on Decem- ber 31, 1998, issued the final TAL, adopting its initial propos- al. The Service acknowledged that the Summer Flounder Monitoring Committee's recommended quota had a 50% chance of achieving the target F, while the Service's TAL had only an 18% chance of achieving the target F. See Fisheries of the Northeastern United States; Summer Flounder, Scup, and Black Sea Bass Fisheries, 63 Fed. Reg. 72,203, 72,203-04 (1998) (codified at 50 C.F.R. pt. 648) ("Final TAL"). The Service also recognized that the incidental catch provisions were entirely voluntary. See id. at 72,204. The Service simply recommended that states adopt the additional inciden- tal catch provisions "[t]o improve the probability of achieving the target [F]." Id. Nowhere did the Service analyze the effect on fishing mortality of shifting from a mandatory to a voluntary incidental catch provision. The Service responded to comments that the TAL did not sufficiently assure achievement of the target F by stating that: (1) the TAL had a higher probability of meeting the target F than MAFMC's 20.2 million pound recommendation; and (2) the incidental catch recommendations "would improve the likelihood that the target fishing mortality rate would be attained." Id. at 72,206. In response to other comments, the Service suggested that the 5.5 inch minimum mesh provision might ameliorate other mortality concerns, but acknowledged that the requirement had not been in effect long enough to determine its efficacy. See id. at 72,208. Appellants filed suit in District Court on January 29, 1999, seeking, inter alia, (1) a declaratory judgment that defen- dants violated the Fishery Act, the Administrative Procedure Act ("APA"), and NEPA, and (2) remand to the agency to impose a new summer flounder TAL. See Complaint at 25- 26, reprinted in J.A. 52-53. The District Court upheld the Service's adoption of the 18.52 million pound TAL, deferring to the agency under Chevron U.S.A. Inc. v. Natural Re- sources Defense Council, Inc., 467 U.S. 837 (1984). The District Court first determined that ss 1851 (a)(1) and (a)(8) in the Fishery Act evinced competing interests between ad- vancing conservation and minimizing adverse economic effects and that Congress offered no insight as to how to balance these concerns. See Natural Resources Defense Council, 62 F. Supp. 2d at 106-07. In addition, the trial court found that the Fishery Act expressed no clear intent as to the particular level of certainty a TAL must guarantee to be consistent with 16 U.S.C. s 1851(a)(1). See id. at 107. Given these perceived ambiguities, the District Court deferred to the Service pursu- ant to Chevron Step Two. This appeal followed. II. ANALYSIS As we recently held in Associated Builders & Contractors, Inc. v. Herman, 166 F.3d 1248 (D.C. Cir. 1999), [i]n a case like the instant one, in which the District Court reviewed an agency action under the [APA], we review the administrative action directly. See Troy Corp. v. Browner, 120 F.3d 277, 281 (D.C. Cir. 1997); Gas Appliance Mfrs. v. Department of Energy, 998 F.2d 1041, 1045 (D.C. Cir. 1993). In other words, we accord no particular deference to the judgment of the District Court. See Gas Appliance Mfrs., 998 F.2d at 1045. Rather, on an independent review of the record, we will uphold [the agency's] decision unless we find it to be "arbitrary, capricious, an abuse of discretion, or other- wise not in accordance with law." 5 U.S.C. s 706(2)(A) (1994). Id. at 1254. As for the Service's disputed interpretations of the Fishery Act, we are guided by the Supreme Court's seminal decision in Chevron U.S.A., Inc., [467 U.S. at 837], [which] governs review of agency interpretation of a statute which the agency administers. Under the first step of Chevron, the reviewing court "must first exhaust the 'traditional tools of statutory construction' to determine whether Congress has spoken to the precise question at issue." Natural Resources Defense Council, Inc. v. Browner, 57 F.3d 1122, 1125 (D.C. Cir. 1995) (quoting Chevron, 467 U.S. at 843 n.9). The traditional tools include examina- tion of the statute's text, legislative history, and struc- ture, see Southern California Edison Co. v. FERC, 116 F.3d 507, 515 (D.C. Cir. 1997); as well as its purpose, see First Nat'l Bank & Trust v. National Credit Union, 90 F.3d 525, 529-30 (D.C. Cir. 1996). This inquiry using the traditional tools of construction may be characterized as a search for the plain meaning of the statute. If this search yields a clear result, then Congress has expressed its intention as to the question, and deference is not appropriate. See Hammontree v. NLRB, 894 F.2d 438, 441 (D.C. Cir. 1990). If, however, "the statute is silent or ambiguous with respect to the specific issue," Chev- ron, 467 U.S. at 843, Congress has not spoken clearly, and a permissible agency interpretation of the statute merits judicial deference. Id. Bell Atlantic Tele. Cos. v. FCC, 131 F.3d 1044, 1047 (D.C. Cir. 1997). Although agencies are entitled to deferential review under Chevron Step Two, our judicial function is neither rote nor meaningless: [W]e will defer to [an agency's] interpretation[ ] if [it is] reasonable and consistent with the statutory purpose and legislative history. See Troy Corp. v. Browner, 120 F.3d 277, 285 (D.C. Cir. 1997) (noting that an agency's inter- pretation must be "reasonable and consistent with the statutory purpose"); City of Cleveland v. U.S. Nuclear Regulatory Comm'n, 68 F.3d 1361, 1367 (D.C. Cir. 1995) (providing that an agency's interpretation must be "rea- sonable and consistent with the statutory scheme and legislative history"). However, a court will not uphold [an agency's] interpretation "that diverges from any real- istic meaning of the statute." Massachusetts v. Depart- ment of Transp., 93 F.3d 890, 893 (D.C. Cir. 1996). GTE Serv. Corp. v. FCC, 205 F.3d 416, 421 (D.C. Cir. 2000). This case presents a situation in which the Service's quota for the 1999 summer flounder harvest so completely diverges from any realistic meaning of the Fishery Act that it cannot survive scrutiny under Chevron Step Two. As an initial matter, we reject the District Court's sugges- tion that there is a conflict between the Fishery Act's ex- pressed commitments to conservation and to mitigating ad- verse economic impacts. Compare 16 U.S.C. s 1851(a)(1) (directing agency to "prevent overfishing" and ensure "the optimum yield from each fishery"); with id. s 1851(a)(8) (directing agency to "minimize adverse economic impacts" on fishing communities). The Government concedes, and we agree, that, under the Fishery Act, the Service must give priority to conservation measures. It is only when two different plans achieve similar conservation measures that the Service takes into consideration adverse economic conse- quences. This is confirmed both by the statute's plain lan- guage and the regulations issued pursuant to the statute. See id. s 1851(a)(8) (requiring fishery management plans, "consistent with the conservation requirements of this chap- ter," to take into account the effect of management plans on fishing communities) (emphasis added); 50 C.F.R. s 600.345(b)(1) (1999) ("[W]here two alternatives achieve sim- ilar conservation goals, the alternative that ... minimizes the adverse impacts on [fishing] communities would be the pre- ferred alternative.") (emphasis added). The real issue in this case is whether the 1999 TAL satisfied the conservation goals of the Fishery Act, the man- agement plan, and the Service's regulations. In considering this question, it is important to recall that the Service oper- ates under constraints from three different sources. First, the statute requires the Service to act both to "prevent overfishing" and to attain "optimum yield." 16 U.S.C. s 1851(a)(1). Overfishing is commonly understood as fishing that results in an F in excess of Fmax. Since Fmax for 1999 was equivalent to 0.24, this constraint required the Service to issue regulations to prevent F from exceeding 0.24. Second, any quota must be "consistent with" the fishery management plan adopted by the Service. See id. s 1854(b)(1). In this case the fishery management plan called for an F of 0.24. Therefore, the quota had be to "consistent with" achieving that F. Third, the Service is required to adopt a quota "necessary to assure that the applicable specified F will not be exceeded." 50 C.F.R. s 648.100(c). The "applicable speci- fied F" for 1999 was Fmax, or 0.24. All of these constraints, then, collapse into an inquiry as to whether the Service's quota was "consistent with" and at the level "necessary to assure" the achievement of an F of 0.24, and whether it reasonably could be expected to "prevent" an F greater than 0.24. In other words, the question is whether the quota, as approved, sufficiently ensured that it would achieve an F of 0.24. Appellants argue that the quota violates applicable standards under both Chevron Step One and Chevron Step Two. Because we find appellants' Chevron Step Two arguments convincing, we have no need to reach their alternative argument that the Service violated NEPA by relying on an inadequate environmental assessment in pro- mulgating the final rule. Appellants' Chevron Step One "plain meaning" argument is virtually indistinguishable from their Chevron Step Two rea- sonableness argument. Appellants acknowledge that the statutory terms "assure," "prevent," and "consistent with" do not mandate a precise quota figure. However, appellants contend that a TAL with only an 18% likelihood of achieving the target F is so inherently unreasonable that it defies the plain meaning of the statute. This is an appealing argument on the facts of this case, because, as we explain below, the Service's action is largely incomprehensible when one consid- ers the principal purposes of the Fishery Act. Nonetheless, we still view this case as governed by Chevron Step Two. The statute does not prescribe a precise quota figure, so there is no plain meaning on this point. Rather, we must look to see whether the agency's disputed action reflects a reasonable and permissible construction of the statute. In light of what the statute does require, short of a specific quota figure, it is clear here that the Service's position fails the test of Chevron Step Two. The 1999 quota is unreasonable, plain and simple. Govern- ment counsel conceded at oral argument that, to meet its statutory and regulatory mandate, the Service must have a "fairly high level of confidence that the quota it recommends will not result in an F greater than [the target F]." Fisher- men's Dock Coop., Inc. v. Brown, 75 F.3d 164, 169-70 (4th Cir. 1996). We agree. We also hold that, at the very least, this means that "to assure" the achievement of the target F, to "prevent overfishing," and to "be consistent with" the fishery management plan, the TAL must have had at least a 50% chance of attaining an F of 0.24. This is not a surprising result, because in related contexts, the Service has articulated precisely this standard. See National Marine Fisheries Ser- vice, Final Fishery Management Plan for Atlantic Tunas, Swordfish and Sharks, Vol. I, at 288, reprinted in J.A. 382 (April 1999) (concluding that the Service should choose man- agement measures that have "at least a 50-percent confi- dence in target reference points," and when choosing between two alternatives with a greater than 50% probability, should choose the higher "unless there are strong reasons to do otherwise"). The disputed 1999 TAL had at most an 18% likelihood of achieving the target F. Viewed differently, it had at least an 82% chance of resulting in an F greater than the target F. Only in Superman Comics' Bizarro world, where reality is turned upside down, could the Service reasonably conclude that a measure that is at least four times as likely to fail as to succeed offers a "fairly high level of confidence." Rather than argue that the quota alone provided enough assurance, the Service contends instead that two additional measures were adopted to increase the likelihood of achieving the target F. These measures were: (1) the provision relat- ing to minimum mesh size; and (2) the recommendation that states voluntarily allocate a certain portion of the directed commercial fishery toward incidental catch. There is nothing in this record, however, to indicate that the proposals on mesh size and voluntary state action would improve the level of confidence so as to assure a reasonable likelihood of achieving the target F. The Service's reliance on its provision regarding minimum mesh size for fishing nets is rather perplexing. We do not question the Service's rational conclusion that it is important to reduce the number of undersized flounder being captured, given recent observations, in a species with a potential 20 year life span, that very few adult fish survive past three years of age. See Projection for 1998 Summer Flounder Quota at 2 (Aug. 1998), reprinted in J.A. 93. At the time the 1999 TAL was proposed, however, the Service acknowledged that the mesh size provision's "benefits have not yet been analyzed." Proposed TAL, 63 Fed. Reg. at 56,136. In fact, the Service apparently placed little stock in MAFMC's predic- tion that the minimum mesh size of 5.5 inches would reduce the number of undersized fish caught. See id. In the final rule, the Service stated only that the minimum mesh size provision was "intended to address" discarding due to under- sized catch; the Service acknowledged, however, that the mesh provision had "not been in operation long enough to determine if an adjustment to the mesh size is warranted." Final TAL, 63 Fed. Reg. at 72,208. In short, there are no meaningful data (or even well-founded predictions) to support the assertion that a larger mesh size would reduce the number of undersized fish caught. And the Service conduct- ed no analysis whatsoever to determine the likely effect of this measure on the probability of meeting the target F. There is certainly nothing in the record to indicate that the larger mesh size would make it likely that the TAL had at least a 50% chance of achieving the target F. The Service's second recommendation, that states set aside a certain percentage of the commercial fishing quota for incidental catch instead of directed commercial catch, also fails to ameliorate the deficient 18% figure. First, in conclud- ing that the TAL had an 18% likelihood of achieving the target F, the Service assumed that at least 10% of the commercial fishing quota would be allocated to incidental catch. When defending its proposal to allocate 32.7% of the commercial quota to incidental catch against a comment that instead suggested a 10% figure, the Service observed that "[a] 10-percent incidental catch allocation in combination with the 18.52-million [pound] ... TAL would result in a less than 18- percent probability of achieving the target F." Id. at 72,211 (emphasis added). Therefore, at least some of the incidental catch proposal's assumed positive effects were already ac- counted for in the 18% starting probability. The agency's "double-counting" here indicates that the Service overstated the positive effects that might come from the incidental catch recommendation. The second, and more serious, flaw in the Service's reliance on its incidental catch proposal is that the proposal is merely a recommendation to the states, not a mandatory require- ment. The Service initially assumed that the incidental catch proposal would be mandatory. When it was revised from a mandatory to voluntary proposal, however, the Service never assessed the impact of the change. Indeed, the record is conspicuously silent on this point, almost as if the change never occurred. At oral argument before this court, counsel for the Government asserted that the Service could reason- ably conclude that the states would comply with the recom- mendation on incidental catch. But counsel conceded that there is absolutely no demonstrated history in the relations between the federal and state agencies to support such an assumption, and there are no present assurances from the states that they will comply with the Service's recommenda- tion. Indeed, there is evidence in the record to suggest resistance from some states to the Service's incidental catch proposal. See id. at 72,207, 72,209-10 (reflecting comments from Connecticut, Massachusetts, New Jersey, New York, North Carolina, and Virginia stating, in sum, that the inciden- tal catch allocation was too high, unenforceable, and beyond the Service's power). We are left only with the Service's unsupported conclusion that the incidental catch provision "increases the probability of meeting the target F." Approv- al of the Final Rule to Implement the 1999 Specifications for the Summer Flounder, Scup, and Black Sea Bass Fisheries- Decision Memorandum at 3 (Dec. 14, 1998), reprinted in J.A. 277. This is manifestly insufficient. As we noted at the outset of this opinion, the Service's quota for the 1999 summer flounder harvest so completely "diverges from any realistic meaning" of the Fishery Act that it cannot survive scrutiny under Chevron Step Two. See GTE Serv. Corp., 205 F.3d at 421. The Service resists this result by suggesting that we owe deference to the agency's "scientific" judgments. See Br. for Appellees at 33. While this may be so, we do not hear cases merely to rubber stamp agency actions. To play that role would be "tantamount to abdicating the judiciary's responsibility under the Administra- tive Procedure Act." A.L. Pharma, Inc. v. Shalala, 62 F.3d 1484, 1491 (D.C. Cir. 1995). The Service cannot rely on "reminders that its scientific determinations are entitled to deference" in the absence of reasoned analysis "to 'cogently explain' " why its additional recommended measures satisfied the Fishery Act's requirements. Id. at 1492 (quoting Motor Vehicle Mfrs. Ass'n, Inc. v. State Farm Mut. Auto. Ins. Co., 463 U.S. 29, 48 (1983)). Indeed, we can divine no scientific judgment upon which the Service concluded that its measures would satisfy its statutory mandate. Here, the adopted quota guaranteed only an 18% probabili- ty of achieving the principal conservation goal of the summer flounder fishery management plan. The Service offered nei- ther analysis nor data to support its claim that the two additional measures aside from the quota would increase that assurance beyond the at-least-50% likelihood required by statute and regulation. III. CONCLUSION For the reasons articulated herein, we reverse the District Court's judgment and remand the case to the Service for further proceedings consistent with this opinion.