North Carolina Fisheries Association, Inc. et al v. Gutierrez

MEMORANDUM

D.D.C.September 17, 2007

UNITED STATES DISTRICT COURT FOR THE DISTRICT OF COLUMBIA NORTH CAROLINA FISHERIES ASS’N, INC., et al., Plaintiffs, v. THE HONORABLE CARLOS GUTIERREZ, in his official capacity as the Secretary of Commerce, Defendant. ) ) ) ) ) ) ) ) ) ) ) ) ) ) ) Civil Action No. 06-cv-01815-JDB DEFENDANT’S REMEDY PROPOSAL As directed by the Court, Defendant Carlos Gutierrez, in his official capacity as the Secretary of Commerce (hereinafter referred to as the National Marine Fisheries Service (“NMFS”)), submits the following proposal regarding the appropriate remedy in the above- captioned case. This proposal is supported by the Declaration of Dr. Roy E. Crabtree, Southeast Regional Administrator for NMFS. As the Court requested, the parties conferred during the 30 days following the Court’s ruling, culminating in a face-to-face meeting on September 10, 2007 in Washington, D.C. at the offices of undersigned counsel. The parties were unable to reach agreement on the appropriate remedy in this case and accordingly NMFS hereby submits its proposal on remedy and respectfully requests that the Court enter the attached proposed order. Case 1:06-cv-01815-JDB Document 35 Filed 09/17/2007 Page 1 of 18 1 I. INTRODUCTION In its Memorandum Opinion entered on August 17, 2007, the Court held that Amendment 13C to the South Atlantic Snapper-Grouper Fishery Management Plan (“Amendment 13C”) complied with National Standards 2, 4, and 8 of the Magnuson-Stevens Act (“Magnuson Act”) as well as the Regulatory Flexibility Act (“RFA”), and therefore granted summary judgment to NMFS on those challenges. However, the Court found that Amendment 13C was “unlawful in so far as it imposes restrictions to end the overfishing of snowy grouper and black sea bass, but does not include a plan to rebuild those species.” Memorandum Opinion at 59, Dckt. No. 34 (hereinafter “Mem. Op.”). Quite simply, the appropriate remedy for the lack of rebuilding plans is for the Court to order that rebuilding plans be implemented by a date certain. As explained more fully herein, and in the accompanying Declaration of Dr. Roy E. Crabtree, NMFS’ Southeast Regional Administrator, NMFS has developed a realistic and expeditious schedule for doing just that. Under NMFS’ proposed remedy, NMFS will commit to working with the South Atlantic Fishery Management Council (“the Council”) to ensure that rebuilding plans for snowy grouper and black sea bass are implemented in six months’ time, by March 31, 2008. This remedy is “true to the important conservation goals of the MSA . . . responsive to the realities of the administrative process, and . . . affords plaintiffs meaningful relief.” Mem. Op. at 65. II. ARGUMENT A. The Appropriate Remedy In this Case is for the Court to Order NMFS to Act on Rebuilding Plans by a Date Certain. As explained in NMFS’ opening brief, NMFS and the South Atlantic Council opted to implement Amendment 13C without including rebuilding plans for snowy grouper and black sea bass to ensure that the Council’s consideration of those measures did not delay action to end Case 1:06-cv-01815-JDB Document 35 Filed 09/17/2007 Page 2 of 18 2 overfishing. See NMFS’ Mem. at 12. This, it was believed, was consistent with Congress’ primary goal in passing the Magnuson Act, which is to “address overfishing and mandate the sustainable conservation of threatened fish stocks.” Legacy Fishing Co. v. Gutierrez, No. 06-0835 (JR), 2007 WL 861143, *2 (D.D.C. March 20, 2007). Despite the fact that the Council continued to work on rebuilding plans for snowy grouper and black sea bass after taking final action on Amendment 13C, Plaintiffs filed the instant lawsuit challenging Amendment 13C on multiple fronts, including the argument Amendment 13C was invalid because it did not include both measures to end overfishing and to rebuild the affected stocks. See Mem. Op. at 50. Although the Court found that Amendment 13C complied with the Magnuson Act’s National Standards 2, 4, and 8 and the RFA, the Court held that “where a species is found to be both experiencing overfishing and already overfished, the MSA requires that any plan amendment designed to halt the overfishing also include a plan to rebuild the overfished stocks.” See id. at 28 (holding that Plaintiffs’ National Standard Two claim did not “hold[] water”), 41 (holding that Plaintiff’s National Standard Four claim “fails”), 47 (holding that “the Secretary acted in conformity with National Standard 8"), 49 (holding that “Defendant is therefore entitled to summary judgment” on Plaintiff’s RFA claim), 51. The Court therefore ruled that Amendment 13C is “unlawful in so far as it imposes restrictions to end the overfishing of snowy grouper and black sea bass, but does not include a plan to rebuild those species.” Mem. Op. at 51, 59. The appropriate remedy in this case, therefore, is to implement a plan to rebuild the snowy grouper and black sea bass stocks. Neither the Magnuson Act nor NMFS’ National Case 1:06-cv-01815-JDB Document 35 Filed 09/17/2007 Page 3 of 18 1/ The Guidelines for National Standard 1 state that, if a stock or stock complex is overfished, remedial action is required “to rebuild the stock or stock complex to the MSY [maximum sustainable yield] level within an appropriate time frame.” 50 C.F.R. § 600.310(e)(3)(ii). 2/ The Magnuson Act defines “optimum yield” as the amount of fish that will “provide the greatest overall benefit to the Nation, particularly with respect to food production and recreational opportunities, and taking into account the protection of marine ecosystems.” 16 U.S.C. § 1802(28); 50 C.F.R. § 600.10. For overfished stocks, optimum yield “provides for rebuilding to a level consistent with producing the maximum sustainable yield in such fishery.” Id. 3/ Maximum Sustainable Yield is “the largest long-term average catch or yield that can be taken from a stock or stock complex under prevailing ecological and environmental conditions.” 50 C.F.R. § 600.310(c)(1)(i). 3 Standard Guidelines1/ (“Guidelines”) define what constitutes a plan to “rebuild the affected stocks of fish.” Therefore, in developing rebuilding plans, NMFS looks to its “Technical Guidance On the Use of Precautionary Approaches to Implementing National Standard 1 of the Magnuson-Stevens Fishery Conservation and Management Act” (“Technical Guidance”). AR 676, available at www.nmfs.noaa.gov/sfa/NSGtkgd.pdf (copy attached hereto); Crabtree Decl. ¶ 4. The Technical Guidance was developed in 1998 a team of NMFS scientists to provide, in pertinent part, guidance on the use of a “precautionary approach” to specify optimum yield2/ (“OY”) in a manner that is consistent with NMFS’ Guidelines. See, e.g. 50 C.F.R. § 600.310(f)(5). In particular, the Guidelines for National Standard One state that, when a stock’s biomass (i.e., weight) falls below the minimum stock size threshold, remedial action is required “to rebuild the stock or stock complex to the MSY3/ [maximum sustainable yield] level within an appropriate time frame.” Id. at § 600.310(e)(3)(ii). Accordingly, the Technical Guidance defines a “rebuilding plan” as “a strategy of selecting fishing mortality rates or equivalent catches that are expected to increase the stock size to the MSY level within a specified period of time.” Tech. Guidance at 36. Rebuilding plans Case 1:06-cv-01815-JDB Document 35 Filed 09/17/2007 Page 4 of 18 4 typically include “four key elements”: “(1) An estimate of the average spawning biomass “(BMSY”); (2) a rebuilding time period; (3) a rebuilding MSY trajectory; and (4) a transition from rebuilding to more optimal management.” Tech. Guidance at 4. In plain terms, this means that a rebuilding plan specifies the biomass of fish in a population that will constitute a rebuilt population (i.e., the biomass at MSY), the time period for achieving a rebuilt population, the allowable catch that can be permitted until the population is rebuilt, and the allowable catch that will produce the optimum yield once the population has been rebuilt. Crabtree Decl. ¶¶ 5-8. In this case, Amendment 13C contains reductions in allowable catch that are sufficient to end overfishing and begin rebuilding the snowy grouper and black sea bass stocks. Id. ¶ 9. However, Amendment 13C did not contain rebuilding plans for snowy grouper and black sea bass per the Technical Guidance, such as an updated estimate of the biomass at MSY, the rebuilding time period, the rebuilding trajectory (or strategy), or the value of optimum yield that can be used for transition to more optimal management once the rebuilding time period has ended and the stock has reached the biomass at MSY. Id. ¶ 9. Implementing rebuilding plans – and thus bringing Amendment 13C into compliance with the Magnuson Act – requires that the Fishery Management Plan for the Snapper-Grouper Fishery of the South Atlantic Region be amended. Id. ¶ 10. As NMFS explained in its opening brief, amendments to fishery management plans (“FMPs”) are prepared by the regional fishery management councils – in this case the South Atlantic Council – and submitted to NMFS, the Secretary’s designee, for approval. NMFS’ Mem. at 2. The Council currently is considering the necessary rebuilding plan components for snowy grouper and black sea bass in the draft of Amendment 15 to the Snapper-Grouper FMP. Crabtree Decl. ¶ 10. Specifically, Amendment 15 includes Management Reference Points, Case 1:06-cv-01815-JDB Document 35 Filed 09/17/2007 Page 5 of 18 5 Rebuilding Schedule Alternatives, and Rebuilding Strategy Alternatives for snowy grouper (2.1.1.1, 2.1.1.2. and 2.1.1.3, respectively) and black sea bass (2.1.5.1, 2.1.5.2 and 2.1.5.3, respectively). See Public Hearing Draft, Snapper Grouper Amendment 15, available at http://www.safmc.net/Portals/6/Meetings/Council/BriefingBook/September%202007/SGComm/ SGAmend15091007.pdf. The current preferred rebuilding schedule alternatives utilize the maximum time period allowed under the Magnuson Act to rebuild the stocks – 34 years for snowy grouper and 10 years for black sea bass. Crabtree Decl. ¶ 19. The Council will make this draft available to the public at the September 2007 Council meeting, which commences on September 17, 2007. Id. ¶ 13. In addition to alternatives for rebuilding plans for snowy grouper and black sea bass, the draft of Amendment 15 includes a laundry list of additional measures addressing other species and issues unrelated to snowy grouper or black sea bass rebuilding plans. Id. ¶ 11. The Court has held, however, that “the Secretary’s statutory duty to include a rebuilding plan in the plan amendment applies only to species that have been designated overfished: here, snowy grouper and black sea bass.” Mem. Op. at 60. The Court also has expressed a desire that the “remedy [for] the absence of a rebuilding plan” be implemented in a “timely” fashion. Mem. Op. at 64-65. NMFS believes that the only reasonable means available to ensure compliance with the Court’s Opinion and timely remedy the absence of rebuilding plans is to place the rebuilding plans for snowy grouper and black sea bass into a stand-alone amendment. NMFS, however, is not empowered under the Magnuson Act to compel the Council to take any particular course of action or adhere to any particular schedule. Therefore, NMFS will commit to make a motion at the September 2007 Council Meeting to remove the snowy grouper and black sea bass management reference point alternatives, rebuilding schedule alternatives, and Case 1:06-cv-01815-JDB Document 35 Filed 09/17/2007 Page 6 of 18 6 rebuilding strategy alternatives from Amendment 15 and place them into a stand-alone amendment to address rebuilding, titled Amendment 15A to the Snapper Grouper FMP. Crabtree Decl. ¶ 13; see Public Hearing Draft, Snapper Grouper Amendment 15A, available at http://www.safmc.net/Portals/6/Meetings/Council/BriefingBook/September%202007/SGComm/ SGAmend15A091007.pdf. Doing so will allow the Council to approve Amendment 15A for public hearings at its September Meeting, and then take final action to submit Amendment 15A to the Secretary of Commerce at the next Council meeting, beginning on December 3, 2007. Id. ¶¶ 15-17. NMFS also will propose to the Council at its September meeting that the remaining measures and alternatives in Amendment 15 become Amendment 15B to the Snapper Grouper FMP and proceed toward completion. Id. ¶ 13. NMFS anticipates that the Council will approve NMFS’ motion to expedite action on rebuilding plans and will adhere to the following schedule: • September 2007 Council Meeting (9/17-9/21): Public hearing draft of rebuilding plans made available as Amendment 15A; • October 12, 2007: A Notice of Availability of a Draft Supplemental Environmental Impact Statement (“DSEIS”) regarding Amendment 15A sent to the Environmental Protection Agency (“EPA”) pursuant to the National Environmental Policy Act (“NEPA”), 42 U.S.C. § 4321, et seq., to be published in the Federal Register on or before October 19, 2007; • December 2007 Council Meeting (12/3-12/7): Council takes final action on Amendment 15A and submits it for Secretarial review; and • March 31, 2008: NMFS implements Amendment 15A, if approved. See Crabtree Decl. ¶¶ 15-17. The schedule outlined above expedites action on rebuilding plans and thus brings Amendment 13C into compliance with the Magnuson Act in the shortest possible length of time – some six months – given several time constraints of the administrative process. Accord Mem. Case 1:06-cv-01815-JDB Document 35 Filed 09/17/2007 Page 7 of 18 7 Op. at 65. For instance, Council business is conducted at public meetings that are held only four times per year – in March, June, September, and December. Crabtree Decl. ¶ 13. It is imperative that the Council and NMFS utilize an open and transparent rulemaking process that provides the maximum opportunity for public participation, as amendments to the South Atlantic Snapper-Grouper Fishery Management Plan affect not only the Plaintiffs in this case, but fishermen and fishing communities throughout North Carolina, South Carolina, Georgia, and Florida as well. 16 U.S.C. § 1852(a)(1)(C). Any remedy in this case needs to consider the impacts upon the entirety of the regulated community and allow everyone an opportunity to participate. In addition, there are several statutorily-mandated public comment periods that the Council must comply with when implementing a FMP amendment. NEPA likely will require that a DSEIS regarding Amendment 15A be made available for a 45-day public comment period. 23 C.F.R. § 771.123(i). According to the schedule proposed above, the mandatory 45 day public comment period would end on December 3, 2007, the first day of the Council's December Meeting, allowing the Council to consider public comments on the DSEIS and on the Public Hearing Draft of Amendment 15A before the Council takes final action at its December Meeting. Crabtree Decl. ¶ 15. In addition, the Magnuson Act requires that, once NMFS receives the final Amendment 15A from the Council, NMFS must publish a notice of availability in the Federal Register and allow a 60-day period for public review and comment. 16 U.S.C. § 1854(a)(1)(B). Because the reductions in harvest contained Amendment 13C are sufficient to end overfishing and begin rebuilding of snowy grouper and black sea bass, it is likely that no regulatory text will be necessary to implement Amendment 15A, thus obviating the need for a proposed and final rule. Crabtree Decl. ¶ 17. Following the 60 day public comment period, the Secretary would Case 1:06-cv-01815-JDB Document 35 Filed 09/17/2007 Page 8 of 18 8 have 30 days to approve, disapprove, or partially approve Amendment 15A. 16 U.S.C. § 1854(a)(1). Assuming Secretarial approval of Amendment 15A, the schedule proposed above would allow NMFS to remedy the absence of rebuilding plans no later than March 31, 2008. Crabtree Decl. ¶ 17. Adherence to the schedule above would give Plaintiffs meaningful relief because, as explained above, the preferred rebuilding schedule alternatives allow the maximum number of fish to be caught during the rebuilding period that the Magnuson Act allows. Id. ¶ 19. Furthermore, the rebuilding strategies reduce harvest gradually, ending overfishing during 2008. Such a scenario minimizes, to the extent practicable, the short-term economic harm possible to Plaintiffs. Id. Because the rebuilding strategies specify the maximum allowable harvest that will rebuild the stock in the longest possible rebuilding schedule, the amount of snowy grouper and black sea bass allowed to be harvested is expected to increase after 2008. Id. The Court can feel confident that the Council and NMFS will adhere to the schedule outlined above for implementing Amendment 15A. As Dr. Crabtree’s Declaration explains, the Council’s work on rebuilding plans for snowy grouper and black sea bass is nearly complete. Crabtree Decl. ¶ 10. As such, there is no reason to expect that the Council will not be able to adhere to the schedule outlined above. Should the Court so desire, NMFS is willing to submit periodic status reports to the Court at each pertinent point in the proposed schedule to inform the Court of the Council’s progress in achieving implementation of rebuilding plans. Moreover, NMFS would be subject to a Court-ordered deadline to act on rebuilding plans by a date certain. In the unlikely event that the Council is unable to meet the deadlines set forth in the proposed schedule, NMFS is prepared to immediately develop Amendment 15A into a Secretarial Amendment to the Snapper Grouper FMP under authority of § 304(c) of the Magnuson-Stevens Case 1:06-cv-01815-JDB Document 35 Filed 09/17/2007 Page 9 of 18 9 Act, 16 U.S.C. § 1854(c). Crabtree Decl. ¶ 18. Because such a Secretarial amendment would still be subject to NEPA’s 45 day comment period and the Magnuson Act’s 60 day comment period, 16 U.S.C. § 1854(c)(4)(B), the timeframe for such action would be virtually the same as that proposed above. Thus, under either scenario, NMFS is willing to commit to implementing rebuilding plans by March 31, 2008. To summarize, the Court found Amendment 13C was unlawful insofar as it lacks a plan to rebuild the snowy grouper and black sea bass stocks. NMFS has proposed a schedule for taking action on rebuilding plans in as little time as is possible under the time constraints of the administrative process. NMFS therefore respectfully requests that the Court enter the proposed order submitted herewith, which requires rebuilding plans for snowy grouper and black sea bass to be implemented by March 31, 2008. B. The Appropriate Remedy In this Case Does Not Include Extraneous Management Measures Unrelated to Rebuilding. As explained above, the Court held that Amendment 13C was “unlawful in so far as it imposes restrictions to end the overfishing of snowy grouper and black sea bass, but does not include a plan to rebuild those species.” Mem. Op. at 59. Thus, the appropriate remedy for this violation is to implement rebuilding plans. It is evident from Plaintiffs’ briefs filed in this case that Plaintiffs are laboring under a fundamental misconception of what a rebuilding plan is. For instance, Plaintiffs have asserted that they believe a rebuilding plan would include measures such as a ban on the sale of recreationally caught fish (Pls’ Mem. at 33) and regional allocations (Pls’ Mem. at 22). Although Plaintiffs clearly would like the Council to consider these issues, it would not be appropriate for the Court to include these measures as part of the remedy in this case. As Case 1:06-cv-01815-JDB Document 35 Filed 09/17/2007 Page 10 of 18 10 explained above, a rebuilding plan is “a strategy of selecting fishing mortality rates or equivalent catches that are expected to increase the stock size to the MSY level within a specified period of time.” Tech. Guidance at 36. Banning the sale of recreationally caught fish or implementing regional allocations would have no effect on fishing mortality rates and would do nothing to increase stock sizes. Rather, these are management measures that allocate the allowable catch among various participants in the fishery. Plaintiffs advocate these management measures because they believe the measures would benefit them economically, not because they have anything to do with rebuilding the overfished stocks. The Court already has considered Plaintiffs’ arguments that the Council and NMFS failed to mitigate the economic impacts of Amendment 13C by deferring action on these very management measures: [P]laintiffs point first to two alternatives that they say the Secretary dismissed “out of hand.” The first (once again) is a measure to ensure that snowy grouper caught and sold by recreational fishermen not be counted against the hard quota applicable to commercial fishing. Pls.’ Mem. at 37; Pls.’ Reply at 24. Second, plaintiffs take issue with the Council’s brief consideration of a proposal – advanced by Dr. Daniel, among others -- to allocate fishing privileges based on historical catch numbers per region. Pls.’ Mem. at 39-40; Pls.’ Reply at 25-26. Mem. Op. at 43. The Court rejected Plaintiffs’ arguments, holding that deferring action on those management measures was consistent with National Standards 4 and 8. See id. at 43-44 (noting that “[b]arring sales by recreational fishermen may have been a sound idea, but . . . [t]he Secretary's failure to address this one problem, in short, does not amount to arbitrary and capricious conduct in violation of National Standard 8" and that with respect to regional allocations, “[f]ar from being ‘absurd and unsupported,’ Pls.’ Reply at 26, the Council’s reading of National Standard 4 was a rational one that survives review under the deferential Case 1:06-cv-01815-JDB Document 35 Filed 09/17/2007 Page 11 of 18 11 arbitrary-and-capricious standard”). Thus, Plaintiffs cannot contend that the National Standards mandate that the management measures they advocate be included in Amendment 15A. In any event, as explained above, the preferred rebuilding schedule alternatives minimize the economic impacts on the regulated community to the greatest extent allowable under the Magnuson Act by utilizing the maximum time period allowed under the statute to rebuild the stocks. Crabtree Decl. ¶ 19. Plaintiffs’ summary judgment briefs identify no authority to support their contention that management measures such as recreational sale and/or regional allocations must be included in a rebuilding plan. In stark contrast to Plaintiffs’ unsupported arguments, NMFS has developed a permissible definition of “rebuilding plan” in its Technical Guidance, which has been in place for nearly 10 years. NMFS’ “expert interpretation” of the Magnuson Act is entitled to deference. As the D.C. Circuit recently explained, under Chevron U.S.A. v. Natural Resources Defense Council, 467 U.S. 837, 842-43 (1984), “[i]f Congress has spoken to the question at issue, then that is the end of the matter . . . [b]ut if Congress has left a gap in the statute or the text of the statute is ambiguous, then the court must determine if the agency's interpretation is permissible, and if so, the court must defer to it.” Citizens Exposing Truth About Casinos v. Kempthorne, 492 F.3d 460, 465 (D.C. Cir. 2007) (citing Chevron, 467 U.S. at 842-843). In this case, Congress has commanded NMFS and the Council to implement a plan amendment “to end overfishing in the fishery and to rebuild the affected stocks of fish,” 16 U.S.C. § 1854(e)(3)(A), however Congress has not defined what constitutes a rebuilding plan, leaving that task to NMFS as the expert agency. Creating rebuilding plans for overfished stocks is highly technical and requires substantial expertise in fisheries biology and management. NMFS’ definition of “rebuilding Case 1:06-cv-01815-JDB Document 35 Filed 09/17/2007 Page 12 of 18 12 plan” as contained in its Technical Guidance is a permissible construction of the statutory language. This is particularly evident when “the words of [the Magnuson Act are] read in their context and with a view to their place in the overall statutory scheme.” Nat’l Ass’n of Home Builders v. Defenders of Wildlife, 127 S.Ct. 2518, 2534 (2007) (citing FDA v. Brown & Williamson Tobacco Corp., 529 U.S. 120, 132-133 (2000). With respect to overfished stocks, Congress has established a statutory scheme to ensure that overfishing is ended and that the stocks are rebuilt in a time that is “as short as possible.” 16 U.S.C. §§ 1854(e)(4). Rebuilding overfished stocks requires selecting rebuilding targets, timeframes, and fishing mortality rates that will reach those targets. NMFS’ Technical Guidance establishes these necessary criteria for rebuilding. Indeed, to this point in the litigation Plaintiffs have not contended that either NMFS’ definition of a rebuilding plan or the draft rebuilding plan alternatives contained in Amendment 15 are insufficient to rebuild the overfished stocks. Plaintiffs’ argument that rebuilding plans must include management measures that do not rebuild the overfished stocks is not a permissible construction of the statute and also is inconsistent with the Court’s Memorandum Opinion. In any event, the Council is poised to act on the management issue of recreational sale independently of this litigation, in what will become Amendment 15B (item 2.1.7). Crabtree Decl. ¶¶ 11, 21. The Council’s current preferred alternative is to prohibit the sale or purchase of recreationally caught fish, which Plaintiffs have asserted will provide them with economic relief. The soon-to-be Amendment 15B also contains additional measures of interest to Plaintiffs, including allocation alternatives and accountability measures for both the recreational and commercial fisheries. Id. ¶ 21. Amendment 15B does not currently include regional allocation alternatives for snowy grouper, however NMFS will propose to the Council at its September 2007 Meeting that management alternatives to address this issue be included as well. Id. Case 1:06-cv-01815-JDB Document 35 Filed 09/17/2007 Page 13 of 18 13 Depending on the alternatives the Council chooses to include in Amendment 15B and the timing of the mandatory comment periods, the Council could take final action to submit Amendment 15B to the Secretary of Commerce at either its December 2007 Meeting or in 2008 Meetings. Id. ¶ 22. Thus, it is neither appropriate nor necessary to include these extraneous management measures as part of the rebuilding plans for snowy grouper and black sea bass. To summarize, the Court has ruled that Amendment 13C unlawfully lacked rebuilding plans for snowy grouper and black sea bass. Simply stated, the appropriate remedy for the lack of rebuilding plans is for the Court to order that rebuilding plans be implemented by a date certain. NMFS has offered the most expeditious schedule possible for implementing rebuilding plans given the constraints of the administrative process. Plaintiffs are simply mistaken to the extent they believe rebuilding plans must include management measures unrelated to rebuilding overfished stocks. Plaintiffs may complain that implementing rebuilding plans will not provide them with as much economic relief as would the management measures they advocate, however the fact of the matter is that Plaintiffs filed suit because Amendment 13C lacked rebuilding plans. NMFS is now proposing to provide these very plans by March 2008. The Court already has considered Plaintiffs’ claims that Amendment 13C failed to comply with National Standards 4 and 8 and found them to be without merit. Moreover, the Council is poised to act on the issue of recreational sale independently of this litigation. Therefore, it would not be appropriate for the Court to include management measures such as recreational sale or regional allocations as part of the remedy in this case. C. The Court Should Leave Amendment 13C in Place in the Interim. As NMFS has explained above, Plaintiffs sued for the lack of rebuilding plans, and NMFS has proposed a remedy in which rebuilding plans are implemented as expeditiously as Case 1:06-cv-01815-JDB Document 35 Filed 09/17/2007 Page 14 of 18 14 possible, in some six months’ time. Once rebuilding plans are implemented, Amendment 13C will be in full compliance with the Magnuson Act. The Court should reject any proposal by Plaintiffs to roll back Amendment 13C’s overfishing regulations to the first year quota levels until rebuilding plans are implemented. See Pls’ Opp. at 29-30. As the Court noted in its Memorandum Opinion, the D.C. Circuit “disfavor[s] invalidating a regulation in whole or part where doing so would cause unwarranted disruption to the regulated industry or the parties.” Mem. Op. at 63. Indeed, in this case “[v]acating or otherwise enjoining the operation of the affected regulations would be highly disruptive, difficult to administer, and would be contrary to the MSA’s goal of expeditiously ending overfishing.” Id. at 65. As NMFS explained in its briefs, such a remedy would be irrational in that it places in further peril the already-depleted resource for the Council’s delay in implementing plans to rebuild it. See NMFS’ Mem. at 28-29. Furthermore, allowing overfishing of overfished stocks to continue would defeat congressional intent, which is to “address overfishing and mandate the sustainable conservation of threatened fish stocks.” Legacy Fishing Co., 2007 WL 861143 at *2. Rolling back the overfishing regulations would be an illogical remedy for the lack of rebuilding plans, as it would do nothing to hasten their implementation; NMFS has already proposed the most expeditious schedule possible under the applicable statutory notice and comment requirements. NMFS’ proposed remedy should obviate the need for the Court to order additional relief to “hold the agency’s feet to the fire.” Mem. Op. at 64. Indeed, rolling back the overfishing regulations would have the opposite effect. Crabtree Decl. ¶ 20. As such, Plaintiffs’ requested interim relief is directly antithetical to the Court’s desire that the “remedy [for] the absence of a rebuilding plan” be implemented in a “timely” Case 1:06-cv-01815-JDB Document 35 Filed 09/17/2007 Page 15 of 18 15 fashion. Mem. Op. at 64-65. In addition, as NMFS has noted, rolling back Amendment 13C’s overfishing regulations would result in substantially greater reductions in the allowable catch of snowy grouper than is called for under Amendment 13C’s third year step-down, thereby exacerbating the long-term economic impacts. See 71 Fed. Reg. 55096, 55101 (Sept. 21, 2006), AR 5619 (“continued overfishing would ultimately threaten the long-term viability of these fisheries, resulting in increased levels of business failure and adverse community change”). To summarize, rolling back the overfishing regulations would do a great disservice to the resource and defeat the primary purpose of the Magnuson Act, which is to end overfishing and rebuild overfished stocks. As such, the Court should not order such interim relief as part of the remedy in this case. Plaintiffs’ proposed interim relief suggests that Plaintiffs’ true concern is not the absence of rebuilding plans, but avoiding the measures necessary to end overfishing. The Court already has rejected Plaintiffs’ challenges to Amendment 13C’s overfishing regulations. Plaintiffs should not be allowed to use rebuilding plans as a pretext for reviving their challenges to the overfishing regulations. The appropriate remedy for lack of rebuilding plans is to implement rebuilding plans. As NMFS has explained above, the schedule proposed for implementing rebuilding plans is realistic and one that the Council and NMFS can meet. III. CONCLUSION For the reasons set forth above, NMFS respectfully requests that the Court enter the proposed order submitted herewith. Case 1:06-cv-01815-JDB Document 35 Filed 09/17/2007 Page 16 of 18 16 Respectfully submitted this 17th of September, 2007. RONALD J. TENPAS Acting Assistant Attorney General JEAN E. WILLIAMS Chief, Wildlife & Marine Resources Section s/ Robert P. Williams ROBERT P. WILLIAMS, Trial Attorney U.S. Department of Justice Wildlife & Marine Resources Section Benjamin Franklin Station, P.O. Box 7369 Washington, DC 20044-7369 (202) 305-0210 (ph) (202) 305-0275 (fx) robert.p.williams@usdoj.gov Attorneys for Federal Defendant Of Counsel: Monica A. Smit-Brunello Department of Commerce, NOAA Office of General Counsel Southeast Regional Office 263 13th Avenue South, Suite #177 St. Petersburg, Florida 33701 (727) 824-5361 (727) 824-5376 (fax) Case 1:06-cv-01815-JDB Document 35 Filed 09/17/2007 Page 17 of 18 CERTIFICATE OF SERVICE I hereby certify that on September 17, 2007, the foregoing will be electronically filed with the Clerk of Court using the CM/ECF system, which will generate automatic service of such filing upon all parties registered to receive such notice, including the following: David E. Frulla Email: dfrulla@kelleydrye.com Shaun M. Gehan Kelley Drye & Warren LLP 3050 K Street, NW – Suite 400 Washington, DC 20007 Counsel for Plaintiffs J. Allen Jernigan Email: ajern@ncdoj.gov North Carolina Department of Justice Attorney General’s Office P.O. Box 629 Raleigh, NC 27602 Counsel for Amicus Curiae s/ Robert P. Williams ROBERT P. WILLIAMS, Trial Attorney Case 1:06-cv-01815-JDB Document 35 Filed 09/17/2007 Page 18 of 18 IN THE UNITED STATES DISTRICT COURT FOR THE DISTRICT OF COLUMBIA NORTH CAROLINA FISHERIES ASS'N, ) INC., et al., ) ) ) ) v. THE HONORABLE CARLOS GUTIERREZ, in his official capacity as the Secretary of Commerce, ) Defendant. 1 Plaintiffs, Civil Action No. 06-cv-0 1 8 1 5 JDB DECLARATION OF ROY E. CRABTREE I, ROY E. CRABTREE, declare as follows: 1. I am the Southeast Regional Administrator of the National Marine Fisheries Service (NMFS), St. Petersburg, Florida. In this capacity, I am responsible for the development of policy and the implementation of science and management programs for the living marine resources of the southeastern United States. I represent the Secretary of Commerce on the South Atlantic Fishery Management Council, the Gulf of Mexico Fishery Management Council, the Caribbean Fishery Management Council, and in other regional activities. I supervise the NMFS personnel in the Southeast Region who are charged with the implementation of fishery management plans. 2. I submit this declaration to provide information regarding an appropriate remedy in Case 1:06-cv-01815-JDB Document 35-2 Filed 09/17/2007 Page 1 of 8 this case in light of the Court's August 17,2007, Order. As explained more fully below, NMFS proposes a schedule for implementing suitable rebuilding plans for snowy grouper and black sea bass as quickly as possible in compliance with the Magnuson-Stevens Fishery Conservation and Management Act ("Magnuson Act") and all other applicable law. Rebuilding Plans 3. The Magnuson Act requires rebuilding plans for fisheries that are overfished. National Standard 1 of the Magnuson Act requires that "[c]onservation and management measures shall prevent overfishing while achieving, on a continuing basis, the optimum yield fiom each fishery for the United States fishing industry." 16 U.S.C. 9 185 1 (a)(l). 4. NMFS has developed a Technical Guidance on the Use of Precautionary Approaches to Implementing National Standard 1 of the Magnuson Act to provide guidance to fisheries managers tasked with developing rebuilding plans. The Technical Guidance states: "[A] rebuilding plan is a strategy of selecting fishing mortality rates or equivalent catches that are expected to increase the stock size to the MSY level within a specified period of time." The Technical Guidance further explains that the components of rebuilding plans "typically include: (a) an estimate of Bmsy [the biomass at maximum sustainable yield (MSY)], (b) a rebuilding period, (c) a rebuilding trajectory, and (d) a transition fiom rebuilding to more 'optimal' management. Administrative Record p. 676, Technical Guidance p. 36. 5. The biomass at MSY is the total stock size of a fish population that can produce MSY. Thus, including an estimate of the biomass at MSY in a rebuilding plan provides the rebuilding target for the overfished stock. 6. The rebuilding period is the period of time within which the stock will increase to the Case 1:06-cv-01815-JDB Document 35-2 Filed 09/17/2007 Page 2 of 8 rebuilding target. If an overfished population can increase and reach the rebuilding target within 10 years, then the maximum rebuilding time period is 10 years. If the population cannot rebuild within 10 years, then the maximum rebuilding time period is 10 years plus 1 generation time (defined by the reproductive capabilities of the stock). 7. The rebuilding trajectory, also known as a rebuilding strategy, specifies the total allowable catch that can be taken from the population during the rebuilding time period. The rebuilding strategy starts the year the rebuilding schedule begins and ends when population reaches the rebuilding target. 8. The transition fiom rebuilding to more optimal management means that once the population is rebuilt, management can shift fiom a recovery mode to a strategy that promotes the optimum yield. Amendment 1 3 C 9. Amendment 13C to the Fishery Management Plan for the Snapper-Grouper Fishery of the South Atlantic region ("Amendment 13C") includes the reductions in harvest for recreational and commercial fisheries that are suficient to end overfishing and begin to rebuild both snowy grouper and black sea bass. Amendment 13C phased-in reductions to the commercial quota over a three year period, with the first year reductions in commercial quota having begun in 2006 and ?he third year reductions beginning in 2008. However, Amendment 13C did not contain rebuilding plans for snowy grouper and black sea bass, as explained in NMFS' Technical Guidance. NMFS' Plan for Compliance With the Court's Opinion 10. The South Atlantic Fishery Management Council (the "Council"), with assistance Case 1:06-cv-01815-JDB Document 35-2 Filed 09/17/2007 Page 3 of 8 from NMFS, has developed Draft Amendment 15 to the Fishery Management Plan for the Snapper Grouper Fishery of the South Atlantic Region ("Snapper Grouper FMP"). Amendment 15 contains rebuilding plans for snowy grouper and black sea bass, as set forth in the NMFS Technical Guidance discussed above. Specifically, Amendment 15 includes alternatives for specifying MSY and the optimum yield (management reference points), rebuilding schedules and rebuilding strategies. The Council's work on rebuilding plans for snowy grouper and black sea bass is nearly complete. 1 1. In addition to rebuilding plans, Amendment 15 also contains numerous other unrelated measures, such as: snowy grouper allocations, golden tilefish management reference points, changes to the golden tilefish fishing year and trip limits, establishing a deepwater snapper grouper unit along with management measures to reduce bycatch, red porgy management reference point alternatives, red porgy rebuilding strategies, red porgy allocation, black sea bass pot limits, commercial harvest overages for the overfished fisheries, recreational harvest ovefages for the overfished fisheries, a prohibition on the sale of recreationally harvested fish, monitoring and assessing bycatch, minimizing the incidental take impact on sea turtles and smalltooth sawfish, permit renewal time periods, and the transferability of permits. 12. NMFS is aware that, in its Memorandum Opinion, the Court found that Amendment 13C did not comply with the Magnuson Act insofar as it lacked rebuilding plans for snowy grouper and black sea bass, and that the Court expressed a desire to have rebuilding plans implemented in a "timely" fashion. NMFS is also aware that the Court found that the Secretary's statutory duty to implement rebuilding plans applies only to the overfished species of snowy grouper and black sea bass. NMFS believes that the only reasonable means available to ensure Case 1:06-cv-01815-JDB Document 35-2 Filed 09/17/2007 Page 4 of 8 compliance with the Court's Opinion and provide a timely remedy to the absence of rebuilding plans is to place the snowy grouper and black sea bass alternatives for rebuilding schedules, rebuilding strategies, and management reference points into a stand-alone amendment. 13. The South Atlantic Council meets to conduct business four times per year - in March, June, September, and December. The Council's September 2007 Meeting will commence on September 17,2007. At that meeting, NMFS will make a motion to remove the snowy grouper and black sea bass management reference points alternatives, rebuilding schedule alternatives, and rebuilding strategy alternatives fkom Amendment 15 and place them in a stand-alone amendment to address rebuilding, titled Amendment 15A to the Snapper Grouper FMP. This will allow the Council to approve Amendment 15A for public hearings at its September Meeting, and then take final action to submit Amendment 15A to the Secretary of Commerce at the next Council meeting, which begins on December 3,2007. NMFS will propose to the Council that the remaining measures and alternatives in Amendment 15 become Amendment 15B to the Snapper Grouper FMP and proceed toward completion. 14. Since Amendment 15A will be a stand-alone rebuilding amendment, NMFS will discuss with the Council at its September Meeting whether or not it would be appropriate to include rebuilding strategies and management reference points for red porgy -- currently in Amendment 15 -- into Amendment 15A as well. 15. Because the actions in Amendment 15A may be expected to have a significant impact on the human environment, NMFS will prepare a Draft Supplemental Environmental Impact Statement (DSEIS) pursuant to the National Environmental Policy Act (NEPA). NMFS anticipates that the Notice of Availability of the DSEIS will be filed with the Environmental Case 1:06-cv-01815-JDB Document 35-2 Filed 09/17/2007 Page 5 of 8 Protection Agency ("EPA") on October 12,2007, and published in the Federal Register on October 19,2007. NEPA's statutorily-mandated 45-day public comment period would end on December 3,2007, the first day of the Council's December 2007 Meeting. Thus, NMFS anticipates that the Council will have considered public comments on the DSEIS and on the Public Hearing Draft of Amendment 15A prior to the December 2007 Meeting, allowing the Council to take final action Amendment 15A at that time. 16. Once the Council takes final action in December 2007 and submits Amendment 15A fo fhe Secretary for approval, NMFS will &mediately commence its review of Amendment 15A and publish a notice in the Federal Register to announce that Amendment 15A is available for a 60-day public comment period, as per section 304(a)(l) of the Magnuson-Stevens Act. 17. Since Amendment 13C has reductions in harvest that are sufficient to end overfishing and begin rebuilding of snowy grouper and black sea bass, and Amendment 15A currently contains management reference points, rebuilding schedules and rebuilding strategies and no management measures, it is likely that no regulatory text would be necessary to implement the Council's preferred alternatives in Amendment 15A, thus obviating the need for a proposed and final rule. Under the Magnuson Act, following the 60-day public comment period, the Secretary would have 30 days to approve, disapprove or partially approve Amendment 15A, as per 304(a)(3) of the Magnuson Act. Assuming Secretarial approval of Amendment 15A, NMFS would be able to remedy the absence of rebuilding plans by March 3 1,2008. These rebuilding plans will guide future management actions as the stocks recover. 18. NMFS has every belief that the Council will adhere to the schedule described above and take final action to submit Amendment 15A to the Secretary in December 2007. However, if Case 1:06-cv-01815-JDB Document 35-2 Filed 09/17/2007 Page 6 of 8 the Council does not.take final action in December, NMFS is prepared to immediately develop Amendment 15A into a Secretarial Amendment to the Snapper Grouper FMP that contains suitable rebuilding plans and management reference points for snowy grouper and black sea bass, under authority of section 304(c) of the Magnuson Act. 19. Amendment 15A will remedy the lack of rebuilding plans, while seeking to limit adverse impacts on the fishing community because the Council's current preferred rebuilding schedule alternatives allow for the longest period of time within the legal constraints of the Magnuson Act to rebuild the population (34 years for snowy grouper and 10 years for black sea bass). Such scenarios provide the least amount of short term economic impact possible to the fishery participants under the Magnuson Act. After 2008, the Council's preferred rebuilding alternatives allow an increase in the harvest of snowy grouper and black sea bass as those stocks rebuild. 20. The timeline discussed above for remedying the absence of the rebuilding plans for snowy grouper and black sea bass is dependent upon the overfishing regulations contained in Amendment 13C remaining in effect. Vacating Amendment 13C or rolling back the quotas to the Amendment 13C's first year levels, as has been suggested by Plaintiffs in this case, would significantly delay the implementation of rebuilding plans. Specifically, rolling back the quotas will not end overfishing during 2008 and thus would require that the rebuilding plans (both the rebuilding schedule and strategies) be recalculated for both snowy grouper and black sea bass. That, in turn, would require additional time to revise the alternatives and analyses and then bring them back before the Council for consideration. Additionally, rolling back the quotas to first year levels would likely require the addition of management measures to Amendment 15A to Case 1:06-cv-01815-JDB Document 35-2 Filed 09/17/2007 Page 7 of 8 further reduce the allowable harvest of snowy grouper and black sea bass to end overfishing and continue to rebuild the stocks, as required by the Magnuson Act. Therefore, NMFS believes that any remedy should not change the existing measures in Amendment 13C. 21. In addition to rebuilding plans, what will become Amendment 15B contains management measures of interest to the Plaintiffs in this litigation, including a preferred alternative to prohibit the sale of recreationally harvested snapper grouper species, allocation alternatives, and accountability measures for both the recreational and commercial fisheries. At the September 2007 Meeting, NMFS will propose additional management measures of interest to Plaintiffs, such as regional quota alternatives for snowy grouper, an item of interest to the Plaintiffs that to date has not been analyzed by the Council. 22. Currently, the Council expects to approve the measures that would be in Amendment 15B for public hearing at its September 2007 Meeting. NMFS expects that Amendment 15B will require a DSEIS to comply with NEPA. Depending on the alternatives the Council chooses to include in Amendment 15B, the need for additional scoping under NEPA, and the public comment periods required by applicable law, the Council could take final action to submit Amendment 15B to the Secretary of Commerce at either its December meeting, or in 2008. I declare under penalty of perjury that the foregoing is true and correct. - I Regional Administrator, Southeast Region, NMFS Case 1:06-cv-01815-JDB Document 35-2 Filed 09/17/2007 Page 8 of 8 Technical Guidance On the Use of Precautionary Approaches to Implementing National Standard 1 of the Magnuson-Stevens Fishery Conservation and Management Act Prepared for the National Marine Fisheries Service by V. R. Restrepo (Convener), G. G. Thompson, P. M. Mace, W. L. Gabriel, L. L. Low, A. D. MacCall, R. D. Methot, J. E. Powers, B. L. Taylor, P. R. Wade, and J. F. Witzig. NOAA Technical Memorandum NMFS-F/SPO-## July 17, 1998 U.S. Department of Commerce William M. Daley, Secretary National Oceanic and Atmospheric Administration D. James Baker, Under Secretary for Oceans and Atmosphere National Marine Fisheries Service Rolland E. Schmitten, Assistant Administrator for Fisheries Case 1:06-cv-01815-JDB Document 35-3 Filed 09/17/2007 Page 1 of 56 iv CONTENTS PREFACE 1 EXECUTIVE SUMMARY 2 1. INTRODUCTION 5 1.1 The MSFCMA and the National Standard Guidelines, 5 1.1.1 The MSY Control Rule and Status Determination Criteria, 5 1.1.2 The Precautionary Approach in Specifying Management Targets, 7 1.2 The Precautionary Approach in Fisheries Management, 8 1.2.1 Evolution: International Agreements, 8 1.2.2 The Overall Scope of the Precautionary Approach, 9 1.3 Control Rules and Reference Points in the Context of the PA, 12 2. LIMIT CONTROL RULES AND STATUS DETERMINATION CRITERIA 15 2.1 General Approach, 15 2.1.1 Control Rules, 15 2.1.2 MSY Control Rules and the SDC, 17 2.1.3 Choosing an MSY Control Rule, 19 2.1.4 Recommended Default MSY Control Rule, 19 2.1.5 The Role of Selectivity, 21 2.2 Situations Requiring the Use of Proxies, 21 2.2.1 Data-Moderate Situations, 22 2.2.2 Data-Poor Situations, 25 2.3 Multispecies Considerations in Implementing MSY Control Rules, 26 3. TARGET CONTROL RULES 28 3.1 A Decision-Theoretic Approach, 29 3.2 A General Simulation Framework, 31 3.3 Recommended Default Target, 34 3.3.1 Data-Moderate and Data-Poor Situations, 36 3.4 Rebuilding from Overfished Status, 36 3.5 Special Considerations, 39 3.5.1 Mixed Stock Complexes, 39 3.5.2 Environmental Fluctuations, 39 3.5.3 Stock Definition Issues, 40 3.5.4 Special Life Histories, 41 3.5.5 Data Issues, 41 3.5.6 New Fisheries, 43 3.5.7 Other Precautionary Tactics, 43 CONCLUDING REMARKS 44 ACKNOWLEDGMENTS 44 REFERENCES 45 APPENDIX A. Equilibrium Implications of Fishing at 75% F 49MSY APPENDIX B. Glossary 51 Case 1:06-cv-01815-JDB Document 35-3 Filed 09/17/2007 Page 2 of 56 1 PREFACE The Magnuson-Stevens Fishery Conservation and Management Act (MSFCMA) contains a set of ten National Standards for fishery conservation and management. National Standard 1 states, "Conservation and management measures shall prevent overfishing while achieving, on a continuing basis, the optimum yield from each fishery for the United States fishing industry." The MSFCMA requires the Secretary of Commerce to "establish advisory guidelines (which shall not have the force and effect of law), based on the national standards, to assist in the development of fishery management plans." These advisory guidelines, known as the National Standard Guidelines (NSGs), were first published in the Federal Register as a proposed rule on August 4, 1997, and revised in the final rule published on May 1, 1998. Section 600.310 of the guidelines contains the text pertaining to National Standard 1. Because the NSGs were written for a non-technical audience, they do not provide detailed guidance for the stock assessment scientists who will ultimately be requested to develop many of the conservation and management measures called for, particularly in the Section relating to National Standard 1, and particularly in light of the widely perceived need to adopt a precautionary approach to the management of marine fisheries. The main purpose of this paper is therefore to provide technical guidance on the use of precautionary approaches to implementing National Standard 1 of the MSFCMA in accordance with the NSGs. This paper was prepared by a team of scientists from the National Marine Fisheries Service (NMFS) with experience in conducting stock assessments, providing scientific advice for fishery management, and developing precautionary approaches to managing fisheries. The technical guidance provided below is partly the product of their combined expertise. In addition, this guidance also reflects the work and group discussions of over 80 scientists who participated in the Fifth NMFS National Stock Assessment Workshop (February 24-26, 1998, Key Largo, Florida), which focused on the theme “Providing Scientific Advice to Implement the Precautionary Approach under the MSFCMA.” Proceedings from that workshop will be published in a complementary NOAA Technical Memorandum. This technical guidance is provided essentially for those aspects of scientific fishery management advice that have biological underpinnings, such as the response of fish populations to exploitation. The drafting team recognizes that there are many other important aspects to managing fisheries, such as socioeconomic factors, which are key to defining optimum yield, and which Fishery Management Councils must consider. Unfortunately, no formal operational protocol is routinely used to incorporate socioeconomic benchmarks into management advice. As such, the implementation of the MSFCMA would benefit greatly from complementary guidelines that address non- biological aspects of fisheries management in a quantitative framework. Case 1:06-cv-01815-JDB Document 35-3 Filed 09/17/2007 Page 3 of 56 2 EXECUTIVE SUMMARY The 1998 Guidelines for National Standard 1 (Optimum Yield) of the Magnuson- Stevens Fishery Conservation and Management Act, 50 CFR Part 600, state: “In general, Councils should adopt a precautionary approach to specification of OY.” Because of the technical nature of the task, NMFS convened a panel of scientists to provide technical guidance on specifying OY that is consistent with the Guidelines (NSGs). The technical guidance is contained in this document. The precautionary approach implements conservation measures even in the absence of scientific certainty that fish stocks are being overexploited. In a fisheries context, the precautionary approach is receiving considerable attention throughout the world primarily because the collapse of many fishery resources is perceived to be due to the inability to implement timely conservation measures without scientific proof of overfishing. Thus, the precautionary approach is essentially a reversal of the “burden of proof”. The precautionary approach in fisheries is multi-faceted and broad in scope. The discussions in this document are not so broad in scope, and are limited to providing guidance to managers and scientists for specifying OY and for developing reference points to guide management decisions. A common element in the application of the precautionary approach to fisheries management worldwide is the definition of “limits” intended to safeguard the long-term productivity of a stock. Several international agreements and documents that deal with the precautionary approach identify maximum sustainable yield (MSY) levels as a minimum standard for defining management limits. The Magnuson-Stevens Act encompasses this concept in that it constrains OY to be no greater than MSY. The NSGs identify two limits for fishery management (referred to as “thresholds”) that are necessary to maintain a stock within safe levels, capable of producing MSY: A maximum fishing mortality threshold (MFMT) and a minimum stock size threshold (MSST). The MFMT and MSST are intended for use as benchmarks to decide if a stock or stock complex is being overfished or is in an overfished state. In the NSGs, these two limits are intrinsically linked through an “MSY Control Rule” that specifies how fishing mortality or catches could vary as a function of stock biomass in order to achieve yields close to MSY. If the maximum fishing mortality limit is reduced as biomass decreases, then the minimum stock size limit decreases (although the MSST cannot become lower than ½ of the equilibrium biomass under a constant-fishing mortality MSY control rule). Thus, the shape of the MSY control rule is an important consideration for developing status determination criteria for overfishing. A default MSY control rule is recommended in Section 2 of this document. Noting that Councils have considerable flexibility in defining the shape of the MSY control rule for each stock under their jurisdiction, and that different control rule shapes pertain to different management objectives, the recommended default could be used in the absence of more specific analyses. The default makes use of estimates of the constant fishing mortality rate resulting in MSY, F , and of the corresponding average spawningMSY Case 1:06-cv-01815-JDB Document 35-3 Filed 09/17/2007 Page 4 of 56 3 biomass, B . The limit F, MFMT, is set equal to F at higher stock sizes; if the stockMSY MSY decreases much below B , the limit F is reduced proportionately (the reduction starts atMSY a fraction of B related to the level of natural mortality). It is anticipated that estimatesMSY of F and B will be either unavailable or unreliable for many stocks. For this reason,MSY MSY Section 2 also presents a discussion of useful proxies. Another common element in the application of the precautionary approach to fisheries management worldwide is the specification of “targets” that are safely below limits. Setting OY at its limit (MSY in the Magnuson-Stevens Act) would not normally be precautionary because there could be a high probability of exceeding the limit year after year. Under the precautionary approach, the target should be set below the limit aking uncertainty and other management objectives into consideration. Development of control rules requires communication between fisheries managers, scientists, industry and the public. If performance criteria for target control rules can be defined, then a range of alternative control rules can be developed and evaluated in terms of precautionary behavior and other desirable economic or operational characteristics for management, once precautionary constraints have been met. Control rules are pre-agreed plans for making management decisions based on stock size. The pre-agreed nature of the measures ensures that management actions are implemented without delay, and it is possible to respond rapidly to changing conditions. As with MSY control rules, Councils have considerable flexibility in defining targets. Section 3 presents a recommended default target control rule that could be used in the absence of more specific analyses. The default sets the target fishing mortality rate 25% below the default limit proposed in Section 2. The 25% reduction constitutes a safety margin that may not perform well for all stocks in terms of preventing overfishing. The performance of the default target can only be evaluated on a case-by-case basis and will depend on (a) the accuracy and precision of stock size, B and F estimates, (b)MSY MSY natural variability in population dynamics, and (c) errors in the implementation of management regulations. Age-structured deterministic models suggest that, for a large combination of life history parameters, the recommended default can result in high stock sizes (around 130% of B ) at the expense of relatively small foregone yields (achievingMSY around 95% of MSY). It is recognized that no single policy can fully address all of the considerations to be encountered in the wide variety of fisheries subject to the Magnuson- Stevens Act. Nevertheless, the default target will be useful in a variety of situations and should at least serve to encourage development of more suitable policies for individual fisheries. The default target control rule may not be applicable for many stocks that are already below the MSST (i.e., that are already overfished). In such cases, the NSGs require that special plans be implemented to rebuild the stocks up to the B l vel within aMSY time period that is related to the stock’s productivity. This document does not propose a default rebuilding plan, because the time to rebuilding may depend on each stock’s current level of depletion. Instead, the document presents the four key elements that should be considered in rebuilding plans: An estimate of B , a rebuilding time period, a rebuildingMSY trajectory, and a transition from rebuilding to more optimal management. The default target control rule may be adapted into a rebuilding plan for each overfished stock, for Case 1:06-cv-01815-JDB Document 35-3 Filed 09/17/2007 Page 5 of 56 4 example, by allowing only a very low fishing mortality when the stock is below the MSST in order to rebuild the stock within the rebuilding time period. This document also discusses a number of special considerations, such as changes in the selectivity of fishing gear, mixed-stock situations, changes in productivity due to the environment, and the appropriateness of various proxies for MSY-related parameters. One consideration of particular importance relates to setting limits and targets for data- poor stocks, i.e., those having very limited information. While the document provides defaults for these cases as well, it is imperative to improve the ability to make informed decisions through enhanced data collection and analyses. Specification of MSY control rules, status determination criteria, and precautionary target control rules is a challenging exercise. Key to this process is communication between managers, scientists, users and the public. In the face of conflicting objectives (avoiding overfishing while achieving high long-term yields), it is essential to understand the tradeoffs associated with alternative control rules and the importance of the weights assigned to the different objectives or performance criteria. Simulation frameworks can facilitate the ncessary interaction. In addition, simulation tools should be used to examine the performance of management systems as a whole, including data collection, assessments, control rules, and implementation of management tactics. Case 1:06-cv-01815-JDB Document 35-3 Filed 09/17/2007 Page 6 of 56 MSY and other terms that appear throughout this document are defined in the Glossary (Appendix B).1 5 1. INTRODUCTION 1.1 The MSFCMA and the National Standard Guidelines 1.1.1 The MSY Control Rule and Status Determination Criteria1 A brief recap of key points from §600.310 of the NSGs will help to focus the task at hand. In discussing the concept of maximum sustainable yield (MSY), the NSGs include the following definitions in paragraph (c)(1): "MSY is the largest long-term average catch or yield that can be taken from a stock or stock complex under prevailing ecological and environmental conditions." "MSY control rule means a harvest strategy which, if implemented, would be expected to result in a long-term average catch approximating MSY." "MSY stock size means the long-term average size of the stock or stock complex, measured in terms of spawning biomass or other appropriate units, that would be achieved under an MSY control rule in which the fishing mortality rate is constant." Paragraph (c)(2) expands upon the meaning and importance of the MSY control rule, providing considerable flexibility in the process: "Because MSY is a theoretical concept, its estimation in practice is conditional on the choice of an MSY control rule. In choosing an MSY control rule, Councils should be guided by the characteristics of the fishery, the FMP's objectives, and the best scientific information available. The simplest MSY control rule is to remove a constant catch in each year that the estimated stock size exceeds an appropriate lower bound, where this catch is chosen so as to maximize the resulting long-term average yield. Other examples include the following: Remove a constant fraction of the biomass in each year, where this fraction is chosen so as to maximize the resulting long-term average yield; allow a constant level of escapement in each year, where this level is chosen so as to maximize the resulting long-term average yield; vary the fishing mortality rate as a continuous function of stock size, where the parameters of this function are constant and chosen so as to maximize the resulting long-term average yield. In any MSY control rule, a given stock size is associated with a given level of fishing mortality and a given level of potential harvest, where the long-term average of these potential harvests provides an estimate of MSY." Although the MSFCMA mandates use of MSY, paragraph (c)(3) of the NSGs allows for cases in which MSY cannot be estimated directly: "When data are insufficient to estimate MSY directly, Councils should adopt other measures of productive capacity that can serve as reasonable proxies for MSY, to the extent possible. Examples include various reference points defined in terms of Case 1:06-cv-01815-JDB Document 35-3 Filed 09/17/2007 Page 7 of 56 6 relative spawning per recruit. For instance, the fishing mortality rate that reduces the long-term average level of spawning per recruit to 30-40 percent of the long- term average that would be expected in the absence of fishing may be a reasonable proxy for the MSY fishing mortality rate. The long-term average stock size obtained by fishing year after year at this rate under average recruitment may be a reasonable proxy for the MSY stock size, and the long-term average catch so obtained may be a reasonable proxy for MSY. The natural mortality rate may also be a reasonable proxy for the MSY fishing mortality rate. If a reliable estimate of pristine stock size (i.e., the long-term average stock size that would be expected in the absence of fishing) is available, a stock size approximately 40 percent of this value may be a reasonable proxy for the MSY stock size, and the product of this stock size and the natural mortality rate may be a reasonable proxy for MSY." In discussing the concept of overfishing, the NSGs use the MSY control rule to define a pair of "status determination criteria" (SDC) in paragraph (d)(2): "Each FMP must specify, to the extent possible, objective and measurable status determination criteria for each stock or stock complex covered by that FMP and provide an analysis of how the status determination criteria were chosen and how they relate to reproductive potential. Status determination criteria must be expressed in a way that enables the Council and the Secretary to monitor the stock or stock complex and determine annually whether overfishing is occurring and whether the stock or stock complex is overfished. In all cases, status determination criteria must specify both of the following: "(i) A maximum fishing mortality threshold or reasonable proxy thereof. The fishing mortality threshold may be expressed either as a single number or as a function of spawning biomass or other measure of productive capacity. The fishing mortality threshold must not exceed the fishing mortality rate or level associated with the relevant MSY control rule. Exceeding the fishing mortality threshold for a period of 1 year or more constitutes overfishing. "(ii) A minimum stock size threshold or reasonable proxy thereof. The stock size threshold should be expressed in terms of spawning biomass or other measure of productive capacity. To the extent possible, the stock size threshold should equal whichever of the following is greater: One-half the MSY stock size, or the minimum stock size at which rebuilding to the MSY level would be expected to occur within 10 years if the stock or stock complex were exploited at the maximum fishing mortality threshold specified under paragraph (d)(2)(i) of this section. Should the actual size of the stock or stock complex in a given year fall below this threshold, the stock or stock complex is considered overfished." Case 1:06-cv-01815-JDB Document 35-3 Filed 09/17/2007 Page 8 of 56 7 Section 2 of this document focuses on technical guidance for establishing MSY and limit control rules and the associated minimum stock size and maximum fishing mortality thresholds. 1.1.2 The Precautionary Approach in Specifying Management Targets The MSFCMA does not use the term "precautionary approach" per se. However, in discussing the concept of optimum yield (OY), the NSGs call for the use of a precautionary approach in paragraph (f)(5): "In general, Councils should adopt a precautionary approach to specification of OY. A precautionary approach is characterized by three features: "(i) Target reference points, such as OY, should be set safely below limit reference points, such as the catch level associated with the fishing mortality rate or level defined by the status determination criteria. Because it is a target reference point, OY does not constitute an absolute ceiling, but rather a desired result. An FMP must contain conservation and management measures to achieve OY, and provisions for information collection that are designed to determine the degree to which OY is achieved on a continuing basis--that is, to result in a long-term average catch equal to the long-term average OY, while meeting the status determination criteria. These measures should allow for practical and effective implementation and enforcement of the management regime, so that the harvest is allowed to reach OY, but not to exceed OY by a substantial amount. The Secretary has an obligation to implement and enforce the FMP so that OY is achieved. If management measures prove unenforceable--or too restrictive, or not rigorous enough to realize OY--they should be modified; an alternative is to reexamine the adequacy of the OY specification. Exceeding OY does not necessarily constitute overfishing. However, even if no overfishing resulted from exceeding OY, continual harvest at a level above OY would violate national standard 1, because OY was not achieved on a continuing basis. "(ii) A stock or stock complex that is below the size that would produce MSY should be harvested at a lower rate or level of fishing mortality than if the stock or stock complex were above the size that would produce MSY. "(iii) Criteria used to set target catch levels should be explicitly risk averse, so that greater uncertainty regarding the status or productive capacity of a stock or stock complex corresponds to greater caution in setting target catch levels. Part of the OY may be held as a reserve to allow for factors such as uncertainties in estimates of stock size and DAH. If an OY reserve is established, an adequate mechanism should be included in the FMP to permit timely release of the reserve to domestic or foreign fishermen, if necessary." Section 3 of this document focuses on technical guidance for specifying Case 1:06-cv-01815-JDB Document 35-3 Filed 09/17/2007 Page 9 of 56 8 precautionary targets that would be consistent with the NSGs. The subsection below provides more comprehensive information on the precautionary approach as it has been and is being considered in different fisheries fora, and discusses elements of the approach that are not identified in the National Standard 1 Guidelines. 1.2 The Precautionary Approach in Fisheries Management 1.2.1 Evolution: International Agreements The United Nations Convention on the Law of the Sea (1982) provided several mechanisms to promote responsible management of marine fisheries; however, it was not until the 1990s that work began on developing a precautionary approach to fisheries management. In 1991, the Committee on Fisheries (COFI) of the Food and Agriculture Organization (FAO) requested FAO to develop an International Code of Conduct for Fisheries. Subsequently, FAO and the government of Mexico sponsored an International Conference on Responsible Fishing, held in Cancun in May 1992. Resolutions formulated in Cancun were presented at the United Nations Conference on Environment and Development (UNCED) in Rio de Janeiro in June 1992. The Rio meeting highlighted the importance of the precautionary approach in the Rio Declaration and Agenda 21. For example, Principle 15 of the Rio Declaration states that “in order to protect the environment, the precautionary approach shall be widely applied by States according to their capabilities. Where there are threats of serious or irreversible damage, lack of full scientific certainty shall not be used as a reason for postponing cost-effective measures to prevent environmental degradation.” Several binding and non-binding agreements embodying the precautionary approach were developed and concluded over the period 1991-1996. The most comprehensive of these is the FAO Code of Conduct for Responsible Fisheries, concluded in late 1995 (FAO 1995a). The Code of Conduct addresses six key themes: Fisheries management, fishing operations, aquaculture development, integration of fisheries into coastal area management, post-harvest practices and trade, and fisheries research. In total, there are 19 general principles and 210 standards in the Code. While a precautionary approach is integral to all themes, it is applied particularly to fisheries management, as detailed in Article 7.5. Paragraph 7.5.1 includes a statement to the effect that: “States should apply the precautionary approach widely to conservation, management, and exploitation of living aquatic resources in order to protect them and preserve the aquatic environment.” The same paragraph also emphasizes that the absence of adequate scientific information is not a reason for postponing or failing to take conservation and management measures. The remaining paragraphs include similar provisions to those in Article 6 of the UN Straddling Stocks Agreement (see below); for example, determination of stock- specific target and limit reference points (Caddy and Mahon 1995), the need to take action if they are exceeded, and the need to take account of uncertainties and impacts on non- target and associated or dependent species. In addition, guidelines are given for adopting a cautious approach in the case of new or exploratory fisheries, and for implementing Case 1:06-cv-01815-JDB Document 35-3 Filed 09/17/2007 Page 10 of 56 9 emergency management measures when resources are seriously threatened due to environmental factors or fishing activity. The Code of Conduct is a voluntary, non-binding agreement. However, it contains sections that are similar to those in two binding agreements: The Agreement to Promote Compliance with International Conservation and Management Measures by Fishing Vessels on the High Seas (the Compliance Agreement), and the Agreement for the Implementation of the Provisions of the United Nations Convention on the Law of the Sea of 10 December 1982 Relating to the Conservation and Management of Straddling Fish Stocks and Highly Migratory Fish Stocks (the Straddling Stocks Agreement; UN 1995). The Compliance Agreement was adopted at the FAO Conference at the 27th session in November 1993. The agreement specifies the obligations of Parties whose fishing vessels fish on the high seas, including the obligation to ensure that such vessels do not undermine international fishery conservation and management measures. The Compliance Agreement is considered to be an integral part of the Code of Conduct. The United States implemented the Compliance Agreement through the High Seas Fishing Vessel Compliance Act of 1995. The Straddling Stocks Agreement was negotiated over a similar period to the Code of Conduct and the content and wording on many issues, including those related to the precautionary approach and General Principles, is similar to that in the Code of Conduct. Although the Straddling Stocks Agreement is strictly applicable to straddling fish stocks and highly migratory fish stocks, much of it is also relevant to fishing within national exclusive economic zones. Annex II of the Straddling Stocks Agreement (UN 1995) provides guidelines for the application of precautionary reference points. Paragraph 2 states, “Two types of precautionary reference points should be used: conservation, or limit, reference points and management, or target, reference points.” Paragraph 5 stipulates, “Fishery management strategies shall ensure that the risk of exceeding limit reference points is very low,” and imposes the further constraint that target reference points should not be exceeded on average. Paragraph 7 states that “The fishing mortality rate which generates maximum sustainable yield should be regarded as a minimum standard for limit reference points.” This combination of requirements implies that fishing mortality should always be well below the level associated with maximum sustainable yield (F ).MSY More detailed treatments of the historical development of the precautionary approach are contained in ICES (1997a), Serchuk et. al. (1997), Thompson and Mace (1997), and Mace and Gabriel (in prep.). 1.2.2 The Overall Scope of the Precautionary Approach According to the Code of Conduct (FAO 1995a), precaution is required in development planning, management, research, technology development and transfer, legal and institutional frameworks, fish capture and processing, fisheries enhancement, and aquaculture. Thus the precautionary approach is multi-faceted and broad in scope. The 1995 FAO Technical Guidelines on the Precautionary Approach (FAO 1995b) Case 1:06-cv-01815-JDB Document 35-3 Filed 09/17/2007 Page 11 of 56 10 groups guidelines on the precautionary approach into three primary subject areas of relevance to capture fisheries: Fisheries management, fisheries research, and fisheries technology. The next three subsections summarize the main issues covered under each area and, while they do not include every aspect of the guidelines, they highlight the large number and diversity of issues involved. Fisheries Management The precautionary approach to fisheries management requires: prudent foresight; taking into account unknown uncertainty by being more conservative; establishment of legal or social frameworks for all fisheries, including rules to control access, data reporting requirements, and management planning processes; implementation of interim measures that safeguard resources until management plans are finalized; avoidance of undesirable or unacceptable outcomes such as overexploitation of resources, overdevelopment of harvesting capacity, loss of biodiversity, major physical disturbances of sensitive biotopes, and social or economic dislocations; explicit specification of management objectives including operational targets and constraints; prospective evaluation; and sound procedures for implementation, monitoring and enforcement. Fisheries Research Research needed to implement precautionary management should strive to: provide data and analyses of relevance to fisheries management; emphasize the roles that fisheries scientists and others must play in helping managers develop objectives; provide scientific evaluation of consequences of management actions; develop operational targets, constraints and criteria that are both scientifically usable and managerially relevant; incorporate both biological and socio-economic elements; ensure that data are accurate and complete; monitor fisheries; conduct research on which management processes and decision structures work best; incorporate uncertainty into assessments and management; address reversibility and irreversibility in ecosystems; formulate implementation guidelines; be multi-disciplinary in nature, including social, economic, and environmental sciences, and addressing management institutions and decision-making processes; and investigate environmentally-friendly fishing gears. Case 1:06-cv-01815-JDB Document 35-3 Filed 09/17/2007 Page 12 of 56 11 Fisheries Technology A precautionary approach to fisheries technology would: not use technology to cause capacity to increase further in already overcapitalized fisheries; use technology to improve sustainability, prevent damage to the environment, improve economic and social benefits, and improve safety; evaluate the effects of new technologies and gears; educate fishers and consumers towards responsible practices; consider impacts on non-target species and ecosystems; evaluate fishing gears with respect to selectivity by size and species, survival of escapees, ghost fishing, effects on habitat, contamination, pollution, generation of debris, safety and occupational hazards, user conflicts, employment, monitoring and enforcement costs, techno-economic factors (infrastructure and service requirements, product quality), and legal factors (existing legislation, international agreements, civil liberties); consider proper procedures for introducing new technology or changes to existing technology; promote research to encourage improvement of existing technologies and to encourage development of appropriate new technologies, and; encourage research into responsible fisheries technology. From these three lists, it is obvious that biological reference points and control rules are but one part in the overall framework of the precautionary approach. Although in some respects they can be considered a primary focus of any precautionary management strategy, they need to be put in proper perspective. Other needs may be just as important; for example, development of access control systems to ensure that fishing capacity is commensurate with resource productivity, evaluation of alternative management systems and institutions, improvements in the quality and reliability of data, improved monitoring and enforcement, design of "environmentally-friendly" fishing gear, and education of fishers and consumers. Regarding research in support of management decisions, it is important that decisions made in stock assessments regarding model choice, estimation techniques and selection of parameters be transparent. Care should be taken when using the term “precautionary” in relation to the science underpinning advice to managers. The scientists’ primary role is to provide scientifically-based options that managers can use to achieve management goals. It is perfectly reasonable for managers to select a "precautionary" management target (e.g., F = lower 80% CI of the probability distribution for F ) based on advice from scientists that this choice will achieve the managementMSY objectives, but it is not reasonable for scientists to add non-transparent conservatism or precaution into the estimation process (e.g., by claiming that the lower 80% CI of the distribution of F is the best estimate of F ).MSY MSY Case 1:06-cv-01815-JDB Document 35-3 Filed 09/17/2007 Page 13 of 56 12 1.3 Control Rules and Reference Points in the Context of the Precautionary Approach According to the Code of Conduct for Responsible Fisheries (FAO 1995a), “States and subregional or regional fisheries management organizations and arrangements should, on the basis of the best scientific evidence available, inter alia, determine: “stock specific target reference points, and, at the same time, the action to be taken if they are exceeded; and “stock-specific limit reference points, and, at the same time, the action to be taken if they are exceeded; when a limit reference point is approached, measures should be taken to ensure that it will not be exceeded.” Thus, two critical components of precautionary management are the specification of limit and target reference points, and pre-agreed management measures to be implemented as a function of stock conditions relative to those reference points. The pre- agreed nature of the measures ensures that management actions are implemented without delay, and it is possible to respond rapidly to changing conditions. Otherwise, management actions could be dependent on the achievement of consensus while stock conditions continue to deteriorate. The MSFCMA makes it clear that effective management actions must be implemented promptly. Limit reference points are intended to constrain harvests so that the stock remains within safe biological limits, and is capable of producing maximum sustainable yield. Management should proceed so that the risk of exceeding the limit reference points is very low. The minimum standard for limit reference points should be the fishing mortality rate that generates MSY, according to Annex II of the Straddling Stocks Agreement. This is consistent with the revised MSFCMA, which states that the terms “overfishing” and “overfished” mean a rate or level of fishing mortality that jeopardizes the stocks’s capacity to produce MSY. Thus, the MSFCMA definition of overfishing and the Annex II standards for precautionary limit reference points both imply that F should be an upperMSY bound on fishing mortality, although the MSFCMA does not define F as an undesirableMSY outcome to be avoided. [NOTE: Nomenclature within the National Standard Guidelines differs somewhat from that in various FAO documents. Limit reference points in the FAO text correspond to threshold levels in the National Standard Guidelines and in some literature, such as the review of overfishing definitions by Rosenberg et. al. (1994). In the FAO text and much of the international literature, the word threshold is used in the context of establishing “buffers”, to trigger action before limit reference points are reached. Such buffers are not equivalent to the thresholds defined in the NSGs, but are analogous to the “interim thresholds” referred to in the preamble to the final rule issuing the NSGs. This document uses the word limit in the same sense as the FAO text. However, in order to maintain consistency with the language of the NSGs, “threshold” is used when referring specifically to the limit reference points that define the act overfishing and an overfished state in the NSGs --the Maximum Fishing Mortality Threshold, MFMT, and the Minimum Stock Size Threshold, MSST--] Case 1:06-cv-01815-JDB Document 35-3 Filed 09/17/2007 Page 14 of 56 13 Target reference points are intended to achieve management objectives, and represent desirable outcomes to be attained. Target reference points should not be exceeded more than 50% of the time, nor on average. A target biomass level for stocks that require rebuilding could be the biomass that would produce MSY. The FAO guidelines on the precautionary approach (FAO 1995b) indicate that the constraints of limit reference points have precedence over targets, and target reference points may require adjustment so that the probability of violating the constraints while meeting the target would be small. The idea that limits have precedence over targets is consistent with the revised MSFCMA, in which OY corresponds to a target level, but is constrained to be less than or equal to MSY. A control rule describes a variable over which management has some direct control as a function of some other variable(s) related to the status of the stock. In many discussions of the topic, a control rule describes a reference fishing mortality rate as a function of stock size, and such is the main focus of Sections 2 and 3 of this paper. In general, however, control rules do not have to be cast in terms of fishing mortality rates or biomass levels. Simply put, a control rule seeks to identify measures of “good” and “bad” stock condition (by comparing perceived stock status with biological reference points), as well as the actions that will make the stock condition change from “bad” to “good.” There are two types of precautionary elements that can be considered in implementing a control rule for management targets: The reference points to be used, and the type of management reaction to be implemented. The degree of precaution achieved in implementing such a control rule is determined by a combination of the probability of going from a “good” stock condition to a “bad” one (overfishing), and the action to be taken when the stock is overfished. Naturally, the current stock condition affects the probability of overfishing, and hence the degree of precaution. Development of control rules requires interaction between fisheries managers and scientists. In addition, public participation is important because the public and fishing industry are more inclined to support management measures on which they have been consulted and which they understand clearly (FAO 1995b). If managers can define acceptable performance criteria for target control rules, then a range of alternative control rules can be developed and evaluated in terms of precautionary behavior and other desirable economic or operational characteristics for management, once precautionary constraints have been met (this approach is explained in Section 3.2). For example, performance criteria could be formulated as the application of a target control rule with “probability of less than X% of reducing the resource below Y% of K within a period of Z years” (Butterworth and Bergh 1993). The effects of other criteria, e.g., “no more than W% change in catch from year to year” could also be evaluated once precautionary constraints were met. An alternative to maximizing performance, constrained by the degree of precaution defined by managers, is to define performance itself in terms of precaution (i.e., the approach in Section 3.1) so that precaution is built directly into optimizing the management objective. With either approach, it is clear that the nature of tradeoffs between the various performance criteria of interest requires substantial interaction between managers and scientists, and open consultation with the public. Target control rules will vary depending on the quality and quantity of available Case 1:06-cv-01815-JDB Document 35-3 Filed 09/17/2007 Page 15 of 56 14 data, as well. Thus, it is unreasonable to expect that target control rules will be perfectly uniform over all stocks. Specification of objectives and performance criteria will enable the development of control rules that will have more acceptable operational implications and still meet precautionary criteria. Rebuilding plans are special forms of target control rules, to be implemented when stocks have fallen below limit biomass levels. Rebuilding plans should include quantifiable milestones to measure progress toward recovery during the plan’s implementation. The precautionary approach counsels that rebuilding action be undertaken imm diately, rather than deferred to the end of the proposed rebuilding period. Case 1:06-cv-01815-JDB Document 35-3 Filed 09/17/2007 Page 16 of 56 F(B) a bln(B) , F (B) a bmin(0,Bc) , F (B) ac max(B,c) bmin(0 ,Bc) , 15 2. LIMIT CONTROL RULES AND STATUS DETERMINATION CRITERIA This section provides technical guidance for specifying what the National Standard Guidelines refer to as “MSY control rules” (Section 1.1.1), which are used to set the criteria for determining whether a stock is being overfished or the stock is in an overfished state. Also included are recommended defaults for cases lacking detailed analyses, and guidance on the use of proxies. In presenting these defaults, our intention is not to inhibit the use of other control rules, but rather to suggest a useful starting point or a “fall-back” position. 2.1 General Approach 2.1.1 Control Rules A control rule describes a variable over which management has some direct control as a function of some other variable(s) related to the status of the stock. That is, the control rule represents a pre-agreed plan for adjusting management actions depending on the condition of the stock. In broad terms, the management actions may be designed as strategies to achieve (a) a fixed exploitation rate (to harvest a constant fraction of the stock each year), (b) constant escapement (e.g., to maintain a constant spawning stock size), or (c) constant catch. However, control rules do not have to adhere strictly to any of these three strategies, and managers may prefer control rules that achieve different results depending on the condition of the stock. In many discussions of the topic, a control rule describes a reference fishing mortality rate F as a function of stock size B, although it is also possible to use catch as the dependent variable. In fact, either option can be expressed in terms of the other, and it is useful to present both. Figure 1 illustrates three possible functional forms for target control rules in terms of both fishing mortality and catch: The two-parameter "logarithmic" form the three-parameter "linear-linear" form and the three-parameter "linear-hyperbolic" form where a, b and c are parameters that determine the magnitude of F depending on the value of B. Case 1:06-cv-01815-JDB Document 35-3 Filed 09/17/2007 Page 17 of 56 0 0.1 0.2 0.3 0.4 0 5 10 15 20 25 30 Stock Size Logarithmic 0 1 2 3 4 5 6 0 5 10 15 20 25 30 Stock Size Logarithmic 0 0.1 0.2 0.3 0.4 0 5 10 15 20 25 30 Stock Size Linear-Linear 0 1 2 3 4 5 6 0 5 10 15 20 25 30 Stock Size Linear-Linear 0 0.1 0.2 0.3 0.4 0 5 10 15 20 25 30 Stock Size Linear-Hyperbolic 0 1 2 3 4 5 6 0 5 10 15 20 25 30 Stock Size Linear-Hyperbolic 16 Figure 1. Some families of control rules. Each panel shows a family of control rules conforming to a particular functional form and passing through a common (arbitrary) point. The logarithmic form forces the fishing mortality rate to vary smoothly with stock size. The linear-linear form forces the fishing mortality rate to be constant when the stock exceeds a specified size. The linear-hyperbolic form forces the catch to be constant when the stock exceeds a specified size (for the special case where catch is computed as the product of stock size and the fishing mortality rate). Figure 1 shows six examples for each form of control rule, where the six examples of the linear-linear form (middle panels of Figure 1) are indistinguishable from one another at values of B>c, as are the six examples of the linear-hyperbolic form (lower panels of Figure 1). The control rules shown in Figure 1 are only a subset of the many shapes possible that could be specified. For instance, an asymptotic (mono-molecular) equation would be Case 1:06-cv-01815-JDB Document 35-3 Filed 09/17/2007 Page 18 of 56 17 an alternative to the smooth logarithmic control rule in which F would be capped at high levels of biomass. 2.1.2 MSY Control Rules and the Status Determination Criteria A special case of control rule is the MSY control rule. Referring to control rules of the type described above and illustrated in the left half of Figure 1, NMFS' guidelines for National Standard 1 state that such an MSY control rule gives "...fishing mortality rate as a continuous function of stock size, where the parameters of this function are constant and chosen so as to maximize the resulting long-term average yield." For example, any of the control rules listed above could be transformed into an MSY control rule by fixing the value of one or perhaps two of the control parameters (say, b in the case of the logarithmic control rule or b and c in the case of the linear-linear or linear-hyperbolic control rules) independently and setting the remaining control parameter (say, a) at the value that maximizes long-term average yield, conditional on the value of the independent control parameter(s) (see Section 3.1). For example, in either the logarithmic or linear-linear forms, setting b=0 gives a control rule in which the fishing mortality rate is equal to the constant (i.e., a control rule in which fishing mortality is independent of stock size). Setting a at the value that maximizes long-term average yield for this special case results in a very simple form of MSY control rule. However, substituting the same value of a into a control rule where b>0 would generally not result in an MSY control rule, because the yield-maximizing value of one control parameter will typically be dependent on the value of the other(s) (Thompson in prep.). Under the guidelines for National Standard 1, the MSY control rule serves two important purposes: (1) It constitutes the maximum fishing mortality threshold (MFMT), above which overfishing is considered to be occurring; and (2) it determines the minimum stock size threshold (MSST), below which the stock is considered overfished. Thus, the MSY control rule is key to defining limit reference points. The role of the MSY control rule in determining the MSST can be seen in the following definition: “To the extent possible, the stock size threshold should equal whichever of the following is greater: One-half the MSY stock size, or the minimum stock size at which rebuilding to the MSY level would be expected to occur within 10 years if the stock or stock complex were exploited at the maximum fishing mortality threshold ...” For example, all of the logarithmic control rules shown in the upper-left panel of Figure 1 happen to constitute MSY control rules under a particular model (Thompson in prep.). These control rules are reproduced in Figure 2 together with a set of vertical dotted lines, each of which indicates the minimum stock size at which rebuilding to the MSY level would be expected to occur within 10 years if the stock were consistently exploited according to the corresponding MSY control rule. The vertical dotted line labeled "A" corresponds to the control rule labeled "A," the vertical dotted line labeled "B" corresponds to the control rule labeled "B," and so forth. The more the control rule departs from the horizontal (control rule "F"), the lower the stock can fall and still be Case 1:06-cv-01815-JDB Document 35-3 Filed 09/17/2007 Page 19 of 56 0 0.05 0.1 0.15 0.2 0.25 0.3 0 5 10 15 20 25 30 Stock Size A B C D E F A B C D E F 18 expected to recover within 10 years. This result conforms with intuition, because curves with greater departure from the horizontal exert less fishing pressure at low stock sizes, thus increasing the rate of rebuilding at those stock sizes. Figure 2. Example MSY control rules (solid curves) and associated stock sizes at which rebuilding would be expected within 10 years (dotted lines). The curve labeled "A" is associated with the line labeled "A," etc. The dependence of the MSST on the MSY control rule is also illustrated in Figure 3 for a linear-linear type of control rule. Here, the MSY control rule sets MFMT constant for biomass levels above B and decreases it linearly with biomass below B . The solidMSY MSY lines labeled a, b and c represent three such MSY control rules and the dashed lines indicate the corresponding MSST levels (shown in relative units), i.e., the values of biomass at which rebuilding to B would take 10 years when fishing at the MFMT (inMSY reality, the actual position of these levels will vary with the life-history characteristics of the species in question). The ascending parts of these example control rules can be interpreted as built-in plans for rebuilding from the MSST to B — for a fixedMSY rebuilding time period (e.g., 10 years), the stronger reductions in limit fishing mortality at low biomass allow for rebuilding from lower biomass limits. Case 1:06-cv-01815-JDB Document 35-3 Filed 09/17/2007 Page 20 of 56 0.00 0.25 0.50 0.75 1.00 0.00 0.25 0.50 0.75 1.00 1.25 1.50 B / BMSY R el at iv e F a b c 19 Figure 3 Hypothetical example illustrating the relationship between Minimum Stock Size Threshold (intersection of the dashed lines with the X-axis) and a linear-linear MSY control rule (solid lines, which define the Maximum Fishing Mortality Threshold). Each of the three rules labeled a, b and c, is scaled relative to its own maximum. 2.1.3 Choosing an MSY Control Rule One factor that might go into choosing an MSY control rule is the resulting location of the MSST. For example, if a Council wished to minimize the range of stock sizes within which special rebuilding plans would be required, it would probably opt for an MSY control rule that afforded a good deal of "built-in" rebuilding, that is, an MSY control rule in which fishing mortality was greatly decreased at low stock sizes. Of course, in no case could the MSST fall below one-half of the MSY level. Another factor that might go into choosing an MSY control rule is the tradeoff between magnitude of yield and constancy of yield. In general, a horizontal MSY control rule (e.g., control rule "F" in Figure 2) would be expected to result in a lower long-term average yield but a less variable yield than an MSY control rule in which fishing mortality was strongly related to stock size (e.g., control rule "A" Figure 2). Councils have considerable flexibility in choosing how to weight their preferences for these and other performance criteria. NMFS' guidelines for National Standard 1 give the following advice: "In choosing an MSY control rule, Councils should be guided by the characteristics of the fishery, the FMP's objectives, and the best scientific information available." 2.1.4 Recommended Default MSY Control Rule As implied above, specifying an MSY control rule is a flexible process that should involve a great deal of communication between scientists and managers so that the tradeoffs between the relevant performance criteria are understood. Due to the demands Case 1:06-cv-01815-JDB Document 35-3 Filed 09/17/2007 Page 21 of 56 F(B) FMSYB c BMSY for all B c BMSY F(B) FMSY for all B c BMSY, 0 0.25 0.5 0.75 1 0 0.25 0.5 0.75 1 1.25 1.5 B / BMSY F / F M S Y M 20 imposed by the timetable of required FMP amendments or other factors, it is desirable to propose a limit control rule that can be used as a default for defining SDC in the absence of more detailed analyses. We recommend a default MSY control rule of the form (see Figure 4): where c=max(1-M, 1/2), F is the fishing mortality rate that maximizes long-term yieldMSY under a constant-F policy, and B is the equilibrium biomass expected when fishingMSY constantly at F . Setting c=max(1-M, 1/2), where M is the natural mortality rate of theMSY exploited age classes, seems reasonable insofar as one would expect a stock fished at FMSY to fluctuate around B on a scale related to M (small fluctuations for low M and largeMSY fluctuations for high M). Figure 4. Recommended default MSY control rule. Note that a control rule of this shape, and parameterized as suggested, may not exactly achieve the maximum long-term yield. The reason for this is that, in an MSY control rule of this form, F(B) would be somewhat larger than F in the flat part of theMSY function (the degree of departure from F is likely to be small in many cases, but isMSY unknown a priori in the absence of detailed analyses). Nevertheless, F(B) can be used to define an approximate MFMT. As noted in Section 2.1.2, the MSST is determined in part by the MSY control rule and is constrained to be greater than ½B . However, for a given MSY control rule,MSY the precise location of the MSST with respect to B may depend on the dynamics of theMSY Case 1:06-cv-01815-JDB Document 35-3 Filed 09/17/2007 Page 22 of 56 21 particular stock. Estimating the location of the MSST with respect to the MSY stock size can be fairly difficult in some situations and may require the use of simulation tools. If needed, we recommend that the point cB in the default MSY control rule be used as aMSY default proxy for the MSST. 2.1.5 The Role of Selectivity A fact often overlooked is that the enumeration of MSY depends on partial recruitment patterns. In theory, assuming no variability in life-history parameters, there could be a "global" MSY that can be achieved by totally avoiding fishing until each cohort reaches the age (size) at which losses due to natural mortality exceed contributions from growth and reproduction, and then harvesting all fish of that age (size) instantaneously. However, such knife-edge selection and deterministic life-history parameters are unrealistic, such that the “global” MSY referred to by the NSGs should be treated as a purely theoretical concept. Calculations of MSY are generally based on the current partial recruitment pattern exhibited by the fishery. "Partial recruitment" patterns reflect both the relative availability of fish of different ages or sizes (i.e., their distribution in time and space relative to that of the fishery) and of the relative selectivity of fish of different ages or sizes exhibited by the mix of gears used in the fishery. For any particular partial recruitment pattern, there is a unique estimate of MSY (all other things being constant). What this means is that estimates of MSY will change if management actions or environmental factors alter the partial recruitment of the fishery in any way. Management actions that can affect MSY include reallocation of quotas between sectors, increases or decreases in size limits, gear modifications and seasonal changes in the fishery. Environmental factors that can alter MSY include those that influence growth rates and other life history characteristics, and those that influence fish movements and distribution, and therefore availability. Estimates of MSY can vary over a large range due to these factors. It is often possible to substantially increase sustainable yields by changing the selectivity pattern to improve yield per recruit. Similarly, potential sustainable yield is dissipated when the fishery is managed in such a way that yield per recruit is reduced, even though management may still be based on “MSY.” Clearly, the magnitude of MSY is an important management issue, as is the exploitation pattern, since it affects the magnitude of MSY. Indeed, these are important issues in developing rebuilding plans for overfished stocks. However, initial specification of control rules should be based upon existing partial recruitment patterns, i.e., the existing mix of gears, allocation decisions and management regulations. If the partial recruitment pattern used for defining the MFMT is substantially different from that in the fishery, then the Councils and the Secretary will be unable to monitor and evaluate the condition of the stock relative to the definition of overfishing. 2.2 Situations Requiring the Use of Proxies As noted in Section 1.1, the MSFCMA allows for the use of proxies in situations where there is insufficient knowledge to implement approaches such as that in Section 2.1. In general, proxies will be needed when MSY-related parameters cannot be estimated from available data, or when their estimated values are deemed to be unreliable for various Case 1:06-cv-01815-JDB Document 35-3 Filed 09/17/2007 Page 23 of 56 22 reasons (e.g., extremely low precision, insufficient contrast in the data, or inadequate models). This documents refers to “data-moderate” and “data-poor” situations as those that require the use of proxies. There are no standards for measuring the level of data richness for a stock. This document offers the following guidance to categorize stocks (note that cases involving a stock complex are likely to be of mixed data richness): Data-rich cases: Reliable estimates of MSY-related quantities and current stock size are available. Control rules typically involve parameters such as F , B ,MSY MSY etc. Stock assessments may be sophisticated, and provide a reasonably complete accounting of uncertainty. Data-moderate cases: Reliable estimates of MSY-related quantities are either unavailable or of limited use due to peculiar life history, poor data contrast, or high recruitment variability, but reliable estimates of current stock size and all critical life history (e.g., growth) and fishery (e.g., selectivity) parameters are available. Control rules typically involve parameters such as F , B , etc., or other35% 35% proxies for MSY-related benchmarks. Stock assessments may range from simple to sophisticated and uncertainty can be reasonably characterized and quantified. (It should be noted that there may be cases when proxies would be useful in “data- rich” situations, i.e., when the proxies are believed to be more robust or reliable than the estimates of MSY parameters. Thus, the term “data-moderate” might be better interpreted as meaning “information-moderate”). Data-poor cases: Reliable estimates of MSY-related quantities are unavailable, as are reliable estimates of either current stock size or certain critical life history or fishery parameters. Control rules typically involve parameters such as M, hi torical average catch, etc. Stock assessments are minimal, and measurements of uncertainty may be qualitative rather than quantitative. The list of proxies presented in the following sections is not all-inclusive and scientists are encouraged to develop and examine alternatives. 2.2.1 Data-Moderate Situations The most widely used biological reference points are those derived from age- structured stock-recruitment models or surplus production models (MSY, F , f ), yieldMSY MSY per recruit analysis (F and F ), spawning per recruit analysis (various percentages of0.1 max maximum SPR and associated fishing mortality rates such as F , F , F , and F ),20% 30% 35% 40% and stock-recruitment relationships (slope at the origin, or the spawning biomass below which recruitment markedly drops) (Caddy and Mahon 1995). In general, reference points from YPR and SPR analyses are the simplest to calculate because they require fewer inputs (stock recruitment data in particular). For this reason, YPR and SPR reference points are often used as proxies for other reference points that do require stock and recruitment data. Proxies for FMSY Case 1:06-cv-01815-JDB Document 35-3 Filed 09/17/2007 Page 24 of 56 23 F was one of the earliest measures used as a proxy for F . However, it wasmax MSY often believed to be an overestimate of F , because it does not account for the fact thatMSY recruitment must decline at some point for low spawning stock sizes, and because F ismax unreasonably large (or even infinite) for some sets of growth and mortality parameters. Computer models have also demonstrated that F typically overestimates F if amax MSY Beverton-Holt (1957) stock-recruitment relationship applies, although F can sometimesMSY exceed F with a Ricker (1958) curve. F (Gulland and Boerema 1973) was developedmax 0.1 as an alternative to F which could result in nearly the same yield per recruit but withmax lower levels of exploitation. Today, F is commonly interpreted as a conservative or0.1 cautious proxy for F , although this is not always the case (Mace 1994; Mace andMSY Sissenwine 1993). Another class of reference points that has gained prominence are those based on F . In particular, values in the range F to F have frequently been used to%SPR 20% 30% characterize recruitment overfishing thresholds (Rosenberg et. al.1994), while values in the range F to F have been used as proxies for F . These uses are supported by30% 40% MSY Goodyear (1993); by Mace and Sissenwine (1993), who advocated F as a recruitment20% overfishing threshold for well-known stocks with at least average resilience and F as a30% recruitment overfishing threshold for less well-known stocks or those believed to have low resilience; and by Clark (1991; 1993), who advocated F as a robust estimator of F35% MSY applicable over a wide range of life histories, or F if there is strong serial correlation in40% recruitment. Note, however, that much of the work on F has presupposed a moderate%SPR amount of resilience to fishing pressure. Moderate resilience may not be a viable assumption for long-lived species and those with low reproductive output. For example, recent analyses of west coast rockfish (Sebastes pp.) stocks are showing the high SPR levels in the range of 50% to 60% are needed to sustain these fisheries (A. MacCall, personal communication). Similar high SPR levels may be nec ssary to protect many species of sharks and other species that have low productivity. F (Sissenwine and Shepherd 1987) may be a useful proxy for different biologicalmed reference points, depending on the level of exploitation of the stock from which the stock- recruitment data were estimated. If the stock has been maintained near B , then FMSY med may be considered a reasonable proxy for F .MSY Proxies for BMSY The equilibrium biomasses corresponding to the above-mentioned fishing mortality reference points can be used as proxies for B . In addition, B has been approximatedMSY MSY by various percentages of the unfished biomass, B , usually in the range 30-60% B (higher0 0 percentages being used for less resilient species, and lower percentages for more resilient species). Referring (in the preamble) to estimates based on two shapes of production models, the NSGs recommend 0.4B as a reasonable proxy for B . However, this value0 MSY may be too low for species with low fecundity such as many species of sharks. B can also be approximated by the mean recruitment (R ) multiplied by eitherMSY mean (a) the level of spawning per recruit at F — namely SPR(F ), or some proxy thereof;MSY MSY or (b) 30-60% SPR (the percentage being determined by the stock’s resilience toF=0 fishing). The danger with using the first approach to develop an MSY control rule of the Case 1:06-cv-01815-JDB Document 35-3 Filed 09/17/2007 Page 25 of 56 24 type in Section 2.1.4 is that, if F is overestimated, then SPR(F ) and B will both beMSY MSY MSY underestimated. Thus, the MFMT could be too high and the MSST too low. If catch and CPUE data are available, production models may provide useful proxies, such as CPUE , which can be used as a relative index of B (in addition, theMSY MSY nominal effort (e.g., in boat-months) corresponding to F can be used as a relative indexMSY of F ).MSY Proxies for B0 Where B is unknown, it can be approximated by the product of average0 recruitment and SPR (Myers et al. 1994). However, this approximation may beF=0 unrealistic because it assumes that there have been no density-dependent changes in growth, survival, or age at maturity during the “fishing down” period. Proxies for MSY The equilibrium yield corresponding to the above-mentioned F and/or B reference points can be used as a proxy for MSY. Inadequate proxies for F and BMSY MSY The literature offers a number of estimators of, or approximations to, the “ultimate” limit reference point at which a stock is likely to collapse (variously called F , F , F (Mace 1994), F (ICES 1997a)). In terms of fishing mortality, theseextinction ext crash estimators include F (if calculated from data collected during a period when the stockmed was overexploited), F (the fishing mortality corresponding to the 90th percentile ofhigh survival ratios), F , and F (the fishing mortality corresponding to the lowest observed20% loss spawning stock — Cook in press). In terms of biomass, these estimators include some definitions of MBAL (the minimum biologically acceptable level of spawning biomass; Serchuk and Grainger 1992), B (the spawning biomass corresponding to 50% of the50%R maximum recruitment in a stock recruitment relationship; Mace 1994; Myers et al. 1994), B (the biomass corresponding to the intersection of the 90th percentile of90%R,90%R/S observed recruitment and the 90th percentile of survival; Serebryakov 1991; Shepherd 1991), and B (the biomass corresponding to the lowest observed spawning stock; ICESloss 1997a). In the absence of a reasonable basis for it, the use of these estimators as proxies for F or B should be avoided because they are likely to be poor approximations. MSY MSY Recommended data-moderate defaults The recommended data-moderate default MSY control rule is that of Section 2.1.4, using proxies for F and B as described below.MSY MSY It is recommended that fishing mortality rates in the range F to F be used as30% 60% general default proxies for F , when the latter cannot be reliably estimated. In theMSY absence of data and analyses that can be used to justify alternative approaches, it is recommended that F be used for stocks believed to have relatively high resilience, F30% 40% for stocks believed to have low to moderate resilience, and F for stocks with "average"35% resilience (Mace and Sissenwine 1993). For stocks with very low productivity (such as rockfish and most elasmobranchs), fishing mortality rates in the range F to F are50% 60% recommended as proxies for F . Less-preferred alternatives (in order of preference) areMSY Case 1:06-cv-01815-JDB Document 35-3 Filed 09/17/2007 Page 26 of 56 25 to use F , M, F , or F (however, if F is calculated from data collected when the0.1 max med med stock was fluctuating around B , then it would be a good proxy for F ). TheMSY MSY equilibrium or average biomass levels corresponding to these fishing mortality rates should then be used as proxies for B , in the same order of preference. The default limit controlMSY rule would then be defined with fishing mortality set to this default level when biomass exceeds (1-M)*B or ½ B , whichever is greater, and would decline linearly to zero forMSY MSY biomass levels below this level (see Figure 4). The recommended default MSST corresponds to ½ B (the absolute lowest limit triggering the need for a rebuilding plan)MSY for species with M 0.5; but occurs at a larger biomass for species with smaller M. 2.2.2 Data-Poor Situations If there are insufficient or inadequate data to conduct YPR and SPR analyses, or if estimates of F and B cannot be obtained for comparison with YPR and SPR reference points, there are few options for defining meaningful targets and limits. Priority should be given to bringing the knowledge base at least up to “data-moderate” standards. Proxies for FMSY The natural mortality rate M has often been considered to be a conservative estimate of F ; however, it is becoming more and more frequently advocated as a targetMSY or limit for fisheries with a modest amount of information. In fact, in several fisheries, F=0.8*M and F=0.75*M have been suggested as default limi s for data-poor cases (Thompson 1993, NMFS 1996). Proxies for BMSY The equilibrium biomass corresponding to F=M or F=0.8*M can be used as a proxy for B . However, in most data-poor situations, it will not be possible to calculateMSY this quantity. Proxies for B0 Some function of CPUE might conceivably be used as a relative index of initial biomass. If information (perhaps anecdotal) exists on resource conditions prior to or shortly after the onset of fishing, some inferences of initial biomass (B ) may be possible. 0 Because the geographic area occupied by a stock may contract with declines in abundance, the contrast between present and early geographic distributions of the resource may be used to obtain a rough approximation of pre-fishery abundance. Early sport fishing records may provide useful information on resource conditions prior to intense exploitation (MacCall 1996). Estimates of early CPUE may relate to B , but care must be0 taken to correct for the general tendency for CPUE to underestimate declines in resource abundance. For example, this may require geographic stratification, correction for temporal changes in fleet composition (e.g., loss of less efficient vessels as catch rate declines) and a variety of behavioral and biological interactions (see Section 3.5.5). Nonequilibrium production modeling (Hilborn and Walters 1992; Prager 1994) also may provide an inference of initial CPUE for the fishery. Proxies for MSY If there is no reliable information available to estimate fishing mortality or biomass Case 1:06-cv-01815-JDB Document 35-3 Filed 09/17/2007 Page 27 of 56 26 reference points, it may be reasonable to use the historical average catch as a proxy for MSY, taking care to select a period when there is no evidence that abundance was declining. Recommended data-poor defaults In data-poor cases it is recommended that the default limit control rule be implemented by multiplying the average catch from a time period when there is no quantitative or qualitative evidence of declining abundance (“Recent Catch”) by a factor depending on a qualitative estimate of relative stock size: Above B : Limit catch = 1.00*(Recent catch).MSY Above MSST but below B : Limit catch = 0.67*(Recent catch).MSY Below MSST (i.e., overfished): Limit catch = 0.33*(Recent catch). The multipliers 1.0, 0.67 and 0.33 were derived by dividing the default precautionary target multipliers in Section 3.3.1 by 0.75, in order to maintain the 0.75 ratio recommended as the default distance between the limit and target reference points for stocks above (1-M)*B . Since it probably will not be possible to determine stockMSY status relative to B analytically, an approach based on "informed judgement" (e.g., aMSY Delphi approach) may be necessary. 2.3 Multispecies Considerations in Implementing MSY Control Rules Under the National Standard Guidelines, MSY is to be specified for each stock in a mixed-stock fishery, and if this is not possible, then “MSY may be specified on the basis of one or more species as an indicator for the mixed stock as a whole or for the fishery as a whole.” Because productivity (growth, recruitment and mortality) of each species in a stock complex is likely to be different, there will be no single value of F that applies toMSY all species within the assemblage. Likewise, catchability (vulnerability) of each co-occurring species by the gear is likely to be different. Thus, fishing rates for co-occurring species are not going to be reduced by equal amounts if effort within the fishery is reduced. Consequently, it will be difficult if not impossible to obtain F andMSY B for several species simultaneously. Depending on which stock (or stocks) within theMSY mixed-stock complex serve as indicators for the complex as a whole, remaining stocks within the complex may be variously over- or under-exploited with respect to their individual MSY levels. If the indicator stock is more productive than other species within the mixed-stock complex, some stocks within the complex may not be able to withstand the same level of fishing effort associated with the MSY control rule for the indicator species, and a precautionary approach becomes warranted in the face of uncertainty about productivity of non-indicator stocks (Section 3.5.1). Those stocks may be potentially at risk for protection under the Endangered Species Act (ESA) if the fishery continues to overfish those stocks, while maintaining productive indicator stocks at MSY levels. The National Standard Guidelines allow exceptions to the requirement to prevent overfishing in the case of a mixed-stock complex. If one species in the complex is harvested at OY, overfishing of other components in the complex may occur if (1) long-term net benefits to the Nation will be obtained an (2) similar long-term net benefits Case 1:06-cv-01815-JDB Document 35-3 Filed 09/17/2007 Page 28 of 56 27 cannot be obtained by modification of fleet behavior or gear characteristics or other operational characteristics to prevent overfishing a d (3) the resulting fishing mortality rate will not cause any stock or ecologically significant unit to require protection under the ESA. Case 1:06-cv-01815-JDB Document 35-3 Filed 09/17/2007 Page 29 of 56 28 3. TARGET CONTROL RULES NMFS' guidelines for National Standard 1 state, "Target reference points, such as OY, should be set safely below limit reference points, such as the catch level associated with the fishing mortality rate or level defined by the status determination criteria." They also state, "...target harvest levels may be prescribed on the basis of an OY control rule similar to the MSY control rule ... but designed to achieve OY on average, rather than MSY. The annual harvest level obtained under an OY control rule must always be less than or equal to the harvest level that would be obtained under the MSY control rule." The words “safely below” in the first quotation have a clear precautionary connotation as elaborated in the National Standard 1 text cited in Section 1.1.2. This section provides technical guidance for developing target control rules. As noted in the Preface, this technical guidance for defining management targets does not incorporate socioeconomic considerations other than aversion to the risk of overfishing. In terms of accounting for uncertainty, two main approaches have been proposed for establishing a target control rule. Both employ probabilistic treatments of uncertainty, but differ in how probability is used. The first approach can be viewed as "decision- theoretic" because it uses the principles of decision theory to establish a target, given a specified level of relative risk aversion. The greater the level of relative risk aversion, the more conservative the precautionary target control rule will be. For example, if a substantial over-estimate of allowable harvest is perceived to be much more undesirable than an under-estimate of equal magnitude, the implied level of relative risk aversion is higher, and the resulting target fishing mortality will be lower, than if the two mis- estimates were perceived to be equally undesirable. In this approach, risk is defined as "expected loss" and is viewed as an objective function to be minimized. A risk-averse target control rule established under a decision-theoretic approach will also necessarily imply some probability of exceeding the limit, but this probability will generally vary on a case-by-case basis, even under a fixed level of relative risk aversion. The second approach can be considered as "frequentist" because it uses the frequency of violating the limit to establish a target, given a specified time frame and a critical frequency level. The lower the critical frequency level, the more conservative the target control rule will be. For example, if it is unacceptable to have more than a 5% chance of violating the limit at any time within a 20-year period, the resulting target control rule will be more conservative than if it were acceptable to have a 10% chance of violating the limit within the same time period. In this approach, risk is defined as "frequency of violation" and is viewed as a constraint to be satisfied. A target control rule established under a frequentist approach will also necessarily imply some level of relative risk aversion, but this level will generally vary on a case-by-case basis, even under a fixed critical frequency level. Case 1:06-cv-01815-JDB Document 35-3 Filed 09/17/2007 Page 30 of 56 Pr(B1 B(t) B2) B2 B1 gB(B ;B0; t) dB . gB(B ;B0;t) gB(B) F(B) c dln(B) Y(Bc,d) gB(Bc,d; ;B0; t) gB(Bc,d;) 29 In Section 3.1 below, an example of a precautionary target control rule developed under the decision-theoretic approach is given. In Section 3.2, a general simulation framework, applicable to both the decision-theoretic and frequentist approaches, is presented. 3.1 A Decision-Theoretic Approach The distinction between limit and target control rules can be thought of as a distinction between levels of relative risk aversion, and development of both limit and target control rules considered as an optimization problem in a decision-theoretic context. For example, a limit control rule might be defined by the optimum derived under a risk- neutral attitude, while a target control rule might be defined by the optimum derived under a risk-averse attitude. A simple and intuitive way to characterize this difference is in terms of stationary (i.e., long-term) yield: A risk-neutral solution maximizes the expectation of stationary yield (MESY) while a risk-averse solution maximizes the expectation of log stationary yield (MELSY; Thompson 1992 and 1996). When computing these expectations, uncertainty in parameter values should be considered along with uncertainty due to recruitment variability and other natural processes. In the absence of fishing, stock size B at time t can theoretically range anywhere from zero to infinity, with some stock sizes being more probable than others. Stock size can be modeled as a probability density function (pdf) with parameter vector and an initial condition B (in this section, B is not used to denote pristine stock size, but rather0 0 the stock size at the start of a population projection). Thus, given an initial condition B=B , the probability that stock size falls between B and B at time t may be written in0 1 2 terms of the "transition distribution" as follows: As t approaches infinity, g describes the "stationary distribution" of stock size,B which can be written as . Next, consider a function which uses a parameter vector to map stock size B into a fishing mortality rate F. Such a function constitutes a control rule. A simple but useful control rule may be specified by two parameters, c and d (for example, the logarithmic form ). For any control rule, yield Y will be a function of stock size conditional on the parameters of the control rule. The stationary distribution of stock size will also be conditional on the same control rule parameters. In the case of the two- parameter control rule, yield can be written as , the transition distribution of stock size as , and the stationary distribution of stock size as . Risk-neutral Optimization A risk-neutral approach can be useful in defining a limit control rule. A risk- neutral solution maximizes the expectation of stationary yield (MESY) for one of the Case 1:06-cv-01815-JDB Document 35-3 Filed 09/17/2007 Page 31 of 56 dB dt aB 1 ln B b (c d ln(B)) B, 30 parameters of the control rule (for example c), conditional on the other parameters (for example d) being fixed, while simultaneously accounting for parameter uncertainty. The solution can be denoted by c (d), meaning the optimum value of parameter c of theMESY control rule that maximizes long-term yield conditional on parameter d. Mathematically, the solution is found by maximizing the marginal arithmetic mean long-term yield, A (c,d)Y with respect to c. This is achieved by differentiating the marginal arithmetic mean yield with respect to c, setting the resulting expression equal to zero, and solving with respect to c. The arithmetic mean yield can generally be computed by projecting the population over a long time horizon. Analytical expressions for arithmetic mean yield can also be obtained for some simple models; in many cases, the solution for c (d) will need to beMESY found numerically. Risk-averse Optimization A risk-averse approach can be useful for defining a target control rule. A risk- averse solution maximizes the expectation of log stationary yield (MELSY) for one of the parameters of the control rule conditional on the other parameters being fixed, while also accounting for parameter uncertainty. Continuing with the example of optimizing c in a two-parameter control rule, the solution can be denoted by c (d), and is found by maximizing the marginal geometricMELSY mean yield, G (c,d) with respect to c. As with A (c,d), the geometric mean yield can beY Y computed by means of simulation, or, in some simple cases, analytically. An Example Thompson (in prep.) provides a detailed example of using the decision-theoretic approach to define limit and target control rules based on maximizing the expected stationary yield or expected log stationary yield. In the deterministic case of that example, the population dynamics of the stock are regulated by a Gompertz-Fox model. The control rule is the two-parameter logarithmic form, giving the expression for change in population size as where a is a growth rate and b is a scale parameter. By recasting the model as a stochastic differential equation that incorporates natural variability, analytical expressions can be derived for the risk-neutral and risk-averse solutions presented above (note, however, that the decision-theoretic approach is not limited to cases where an analytical solution is available, as the same approach can be followed using simulation tools such as those of Section 3.2). Figure 5 presents examples of limit and target control rules developed with the decision-theoretic approach for two levels of parameter uncertainty. The control rules shown in Figure 5 have the desirable precautionary property that the buffer between the limit and the target fishing mortality increases as the level of uncertainty surrounding parameter estimates increases. Case 1:06-cv-01815-JDB Document 35-3 Filed 09/17/2007 Page 32 of 56 0.0 0.5 1.0 1.5 0.0 0.5 1.0 1.5 2.0 2.5 3.0 Relative stock size R el at iv e fis hi ng m or ta lit y 0.0 0.5 1.0 1.5 0.0 0.5 1.0 1.5 2.0 2.5 3.0 Relative stock size R el at iv e fis hi ng m or ta lit y Low uncertainty High uncertainty 31 Fi gure 5. Example limit (dashed lines) and target (solid lines) control rules in a particular model derived with a decision-theoretic approach. The size of the buffer between the limit and target control rules is dictated by the amount of parameter uncertainty (compare upper and lower panels). 3.2 A General Simulation Framework A fishery management strategy is the combination of data collection, stock assessment, control rules, and technical measures for implementing the harvest controls. Considerable work has been undertaken to develop simulation methods to evaluate the performance of management strategies (e.g.,de la Mare 1986; see Kirkwood and Smith 1996), with much attention often given to the way the various components of a strategy may interact with each other over time. For example, in a recent review of stock assessment methods, the National Research Council stated that “Both harvesting strategies and decision rules for regulatory actions have to be evaluated simultaneously to determine their combined ability to sustain stocks” (NRC 1998). Case 1:06-cv-01815-JDB Document 35-3 Filed 09/17/2007 Page 33 of 56 Underlying Fishery System Performance Statistics error to data Observed Data assessment Perceived Fishery System control rule Perceived Performance Statistics Fishery Tactics 32 Figure 6. Schematic representation of a simulation framework for evaluating management strategies. Modified, with permission, from Section 4 of ICES (1997b). The conceptual framework depicted in Figure 6 (taken from ICES 1997b), illustrates a flexible simulation approach for evaluating management strategies. The general technique is to simulate a “true” underlying fishery system of known characteristics, including natural variability. Monte Carlo simulation is used to sample observations with measurement error from the underlying system, and the sample observations are then used in a stock assessment. This allows repeated realizations of the “perceived” system, which may or may not differ substantially from the “true” system (depending partly on the degree of similarity between the true population dynamics and those assumed in the assessment procedure). Using a pre-specified target control rule (e.g., to set the Total Allowable Catch equal to the catch obtained by harvesting the perceived population at the F rate), a regulatory strategy can then be translated intoMSY specific fishery tactics (e.g., catch allocations for different fishing sectors). These tactics in turn affect the real underlying system in the next iteration, and so on. A key step in the evaluation process is to identify the performance criteria that will be examined (see also Section 1.3). In the case of rebuilding an overfished stock, an important performance criterion might be the probability that BB after X years (e.g.,MSY 10 years) of implementing a target control rule (a similar approach was used in the guidelines for estimating “potential biological removals” [PBR] for the implementation of the 1996 amendments to the Marine Mammal Protection Act; Wade and Angliss 1997). In most applications, multiple criteria will probably need to be examined, such as the probability that the stock remains above MSST, the average annual yield, and the interannual variability in yield. Inclusion of multiple criteria is particularly useful when there are conflicting goals, such as preventing the stock from falling below B while atMSY the same time achieving yields as close to MSY as possible. Figure 7 depicts an example from ICES (1997b), in which simulation starts with a stock at an equilibrium biomass equal to ½B , the limit F is set to F , and the precautionary target F is set below FMSY MSY MSY Case 1:06-cv-01815-JDB Document 35-3 Filed 09/17/2007 Page 34 of 56 0 25 50 75 100 0 20 40 60 80 100 Percent reduction in limit F P ro b( re co ve ry ) or Y ie ld yield % recovery 33 by a given percentage. The figure illustrates the tradeoffs between increasing the chances of rebuilding in a 10-year period and sacrificing average yield. Figure 7 Tradeoffs between two conflicting performance criteria: Rebuilding an overfished stock and maximizing average yield during a 10-year period. Hypothetical example taken from ICES (1997b), data set 7, with limit F = F .MSY Simulation results such as those depicted in Figure 7 can be used to infer the degree of precaution required to achieve a desired outcome. In the example above, if at least a 50% probability of rebuilding to B was desired, then the rebuilding target FMSY should be set at about ½F . Thus, the simulation approach can help determine how farMSY apart (or how “safely below”) targets have to be from limits to achieve management goals. In general, simulations should be conducted on a case-by-case basis to account for: - Growth, reproductive and recruitment dynamics of the stock, including variability (process error); - Initial conditions, including age-structure; - Selectivity of the fishing gear(s); - Types of observations sampled (e.g., age-structure data) and their variability; - Stock assessment method used; - Estimation of biological reference points (e.g., limit F) and their uncertainty; and - Potential biases in the implementation of regulations determined by the control rule. The simulation approach can also be used to evaluate the benefits to management from reduced uncertainty (Powers and Restrepo 1993). Figure 8 shows that the probability-of-rebuilding curve (from the previous example) is shifted upwards when there is increased in precision regarding current stock status and F .MSY Case 1:06-cv-01815-JDB Document 35-3 Filed 09/17/2007 Page 35 of 56 0 25 50 75 100 0 20 40 60 80 100 Percent reduction in limit F P ro b( re co ve ry ) increased precision reduced precision F(B) 0.75 FMSYB c BMSY for all B c BMSY F(B) 0.75FMSY for all B c BMSY, 34 Figure 8 The effect of increased precision on the rebuilding example of Figure 7. 3.3 Recommended Default Target Ideally, target control rules should be developed using approaches such as those in Sections 3.1 or 3.2. In setting a precautionary target control rule by means of the “frequentist” approach (Sections 3 and 3.2), we recommend that the probability of exceeding the MFMT be not greater than 20%-30%, and certainly smaller than 50%. Absent such analyses or a risk-averse solution as explained in Section 3.1, the following default target control rule is recommended. The recommended target control rule (Figure 9) sets the target fishing mortality rate 25 percent below the limit fishing mortality (MFMT) recommended in Section 2.1.4. In equation form, the recommended default target is: where c, F and B are as defined in Section 2.1.4.MSY MSY The default provides a safety margin (or buffer) to ensure that the realized F does not exceed MFMT. The default target control rule also facilitates rebuilding of stocks by reducing F proportionately at stock sizes below (1-M)B . In some cases, however, theMSY rebuilding rate from the default target will be insufficient to rebuild an overfished stock to B within the time period allowed by the NSGs (depending on the life historyMSY characteristics of the stock and the level of depletion). In such cases, stronger conservation measures will be required, as explained in Section 3.4. Case 1:06-cv-01815-JDB Document 35-3 Filed 09/17/2007 Page 36 of 56 0 0.25 0.5 0.75 1 0 0.25 0.5 0.75 1 1.25 1.5 B / BMSY F / F M S Y M 35 Figure 9. Recommended target (solid line) and limit (dashed line) control rules. The target may only be applicable for biomass levels at or above the minimum stock size threshold because it may not allow for sufficient rebuilding for stocks that are already overfished. Such stocks may require a more conservative target control rule for rebuilding (Section 3.4). The equilibrium consequences of fishing at the default 75% F were evaluatedMSY using the deterministic model of Mace (1994) (see Appendix A). The results of this exercise indicate that fishing at 75% F would result in equilibrium yields of 94% MSYMSY or higher, and equilibrium biomass levels between 125% and 131% B -- a relativelyMSY small sacrifice in yield for a relatively large gain in biomass (Table A1). Although it is likely that results would diverge for more complex models (e.g., those in which the ages of maturity and recruitment differed substantially, or those incorporating stochasticity), the calculations indicate that relatively small sacrifices in yields will result in relatively much larger gains in stock biomass. Increased biomass should in turn result in a number of benefits to the fishery, including increased CPUE, decreased costs of fishing, and decreased risk to the stock. Relative to fishing at F , fishing at 75% F will reduce theMSY MSY probability that a stock will decline to ½ B .MSY The deterministic simulation results presented in Appendix A should not be taken as being strictly applicable to every situation. Variability in the population dynamics parameters of a stock will affect the performance of fishing at 75% F . As well, theMSY evaluation only pertains to cases where F can be reliably estimated. As such, theMSY performance of the default target will depend on the robustness with which F can beMSY estimated or approximated. Simulation tools such as those discussed in Section 3.2 could be used to investigate these issues. It is recognized that no single policy can fully address all of the considerations to be encountered in the wide variety of fisheries subject to the MSFCMA. To the extent that this default target control rule may be inappropriate, it should at least serve to encourage development of more suitable policies for individual fisheries. Case 1:06-cv-01815-JDB Document 35-3 Filed 09/17/2007 Page 37 of 56 The MSFCMA requires that the rebuilding time period be as short as possible and not to exceed 10 years2 with a few exceptions, including cases where the biology of the stock or other environmental conditions dictate otherwise. 36 3.3.1 Data-Moderate and Data-Poor Situations In data-moderate cases, the default target control rule may require the use of appropriate proxies for reference points such as those presented in Section 2.2. In data-poor cases, the default policy may be interpreted qualitatively as follows: Above B Target catch = 0.75*(Recent catch).MSY Above MSST but below B Target catch = 0.50*(Recent catch).MSY Below MSST (i.e., overfished) Target catch = 0.25*(Recent catch). Determination of the status of biomass relative to B preferably involvesMSY quantitative analysis, but in data-poor cases, applicable analytic methods may not be particularly sophisticated and include a variety of stock assessment methods developed in the 1960s and 1970s (e.g., Gulland 1983). In cases of severe data limit ions, qualitative approaches may be necessary, including expert opinion and consensus-building methods (see also Section 2.2.2). 3.4. Rebuilding from Overfished Status The National Standard 1 guidelines indicate that once biomass falls below the minimum stock size threshold (MSST), then remedial action is required “to rebuild the stock or stock complex to the MSY level within an appropriate time frame.” Th refore, recommendations are presented here for determining the adequacy and efficacy of rebuilding plans. A rebuilding plan is a strategy of selecting fishing mortality rates or equivalent catches that are expected to increase the stock size to the MSY level within a specified period of time. Components for a rebuilding plan typically include: (a) an estimate of B , (b) a rebuilding period, (c) a rebuilding trajectory, and (d) a transition fromMSY rebuilding to more “optimal” management (Powers 1996). Specifying a control rule in terms of fishing mortality rate and biomass incorporates these components. Species life history characteristics will affect rebuilding plans in several ways. Some stocks may possess low productivity and will be incapable of recovering within 10 years , even in the absence of fishing mortality. Alternatively, a stock may be highly2 productive, in which case a rebuilding plan of 10 years will not be precautionary, i.e. the stock has the capability of reaching B well before 10 years.MSY Often productivity is correlated with the mean generation time of a stock (defined below), which is why the final rule issuing the NSGs link the maximum rebuilding time period to generation time when rebuilding cannot be achieved in 10 years. The minimum possible rebuilding period is constrained by a stock’s status relative to B and itsMSY biological productivity. Linking the rebuilding period with generation time is important because it highlights the time span in the future during which recruitment will begin to depend primarily upon fish that have yet to be born, as opposed to spawners that already Case 1:06-cv-01815-JDB Document 35-3 Filed 09/17/2007 Page 38 of 56 G A a1 aEaNa A a1 EaNa , Na N1exp( a1 j1 Mj) , 37 exist. Rebuilding rates will also be affected by the partial recruitment pattern. Generally, greater rebuilding rates are possible by reducing mortality rates on juveniles than by equal mortality rate reductions on adult fish. However, this depends upon the relative growth and natural mortality between the age groups. For all overfished resources, the overarching principle is that initial actions must provide a very high probability of preventing further stock declines and have a high probability of immediate improvement. Delaying action is not precautionary. Generation time Although the NSGs do not provide a definition of generation time, various definitions exist in the scientific literature (Caswell 1989). In the context of stock rebuilding time horizons, the definition of generation time used could refer to an unfished state. We recommend that the default definition of generation time, G, be (Goodyear 1995): where a denotes age, A is the oldest age expected in a pristine (unfished) condition, E isa the mean fecundity at age of females, and N is the average number of females per recruita alive at age a in the absence of fishing, i.e., where M is the natural mortality rate. These expressions should be computed on an equilibrium per-recruit basis, i.e., setting N = 1. When fecundity data are not available, G1 can be computed by replacing E with an age-specific vector of maturity ratios times bodya weight (as commonly used to compute spawning biomass). The rebuilding plan In the absence of data and analyses that can be used to justify alternative approaches, we recommend that a default rebuilding plan for stocks below the MSST be based upon the precautionary target control rule of Section 3.3 with the following extensions: 1) The maximum rebuilding period, T , should be 10 years, unless T (themax min expected time to rebuilding under zero fishing mortality) is greater than 10 years, when T should be equal to T plus one mean generation time.max min Case 1:06-cv-01815-JDB Document 35-3 Filed 09/17/2007 Page 39 of 56 0 0.25 0.5 0.75 1 0 0.25 0.5 0.75 1 1.25 1.5 B / BMSY F / F M S Y a b c 38 2) The target rebuilding time period, T , should be as short as possible andtarget lower than T (although it could be adjusted up to T under themax max circumstances described in §600.310(e)(4) of the NSGs). We suggest that T not exceed the midpoint between T and T ; and,target min max 3) If the stock is well below the MSST (e.g., B ½MSST), it may be necessary to set the fishing mortality rate as close to zero as possible (i.e., to that associated with unavoidable levels of bycatch) for a number of years. Figure 10 illustrates what a rebuilding plan might look like for a severely- overfished stock. In region a, the rebuilding plan’s F is set to zero. In region b, between ½MSST and B , the rebuilding F is set to 75% of the target F in the control rule ofMSY Section 3.3. In region c, the stock is rebuilt and the F is set again to the target of Section 3.3. Whether or not a zero F in region a and a 75% reduction in region b satisfy the requirement for rebuilding within the target time period largely depends on the initial level of stock depletion and the stock’s productivity. Figure 10. Example of a rebuilding plan (solid line) for a severely-overfished stock. The dotted and dashed lines represent the recommended default limit and target control rules of Sections 2.1.4 and 3.3, respectively. The regions a, b and c represent three phases in the rebuilding plan: part a is designed to initiate rebuilding with high probability; part b is designed to accelerate rebuilding compared to the rate of rebuilding that is built into the target control rule of Section 3.3; part c represents a transition to more “optimal” management. The role of uncertainty Accounting for uncertainty in stock dynamics, current stock status and recruitment variability is important in developing rebuilding plans (Rosenberg and Restrepo 1994). As such, we suggest that the rebuilding plan should be designed to possess a 50% — or higher — chance of achieving B within T years, and a 90% — or higher — chanceMSY target of achieving B within T years.MSY max The intent of the MSFCMA is that overfished stocks be rebuilt quickly. For this Case 1:06-cv-01815-JDB Document 35-3 Filed 09/17/2007 Page 40 of 56 39 reason, stock rebuilding should be monitored closely so that adjustments can be made when rebuilding milestones are not being met for whatever reason. For example, if target rebuilding Fs are exceeded due to quota over-runs, subsequent target Fs should typically be adjusted downwards to put the stock back on the rebuilding time table. The magnitude and variability of future recruitment will affect the realized rebuilding trajectory. In cases when one or more very large year classes appear, it may be tempting to utilize them to increase short-term yield at the expense of slower stock rebuilding, hoping that subsequent year classes will be of similar — or at least average — magnitude. Such action would not be precautionary. Furthermore, the resulting change in fishing mortality would depart from the pre-agreed nature of the rebuilding control rule and therefore be inconsistent with the rebuilding plan. 3.5 Special Considerations 3.5.1 Mixed-Stock Complexes The National Standard Guidelines provide for specification of a fishery-wide OY for a mixed-stock fishery, where management measures for separate target harvest levels for individual stocks may be specified, but are not required. Although the guidelines recommend that the sum of individual target levels be less than the fishery-wide OY, if individual OY levels are not specified, the entire OY could be removed from one or a few unproductive stock components and overfishing of these components would occur. Clearly, a precautionary approach should be used to minimize the risk of removing the least productive components in the mixed-stock fishery. Biological reference points (or proxies) and precautionary target control rules for each stock in a mixed-stock complex should be developed whenever possible, even though information may be limited. At a minimum, fishing mortality should not exce d the limit (MFMT) for any individual stock in a mixed-stock complex, except as provided under the very stringent criteria specified in §600.310(d)(6) of the NSGs. The relevant target control rule should be implemented, regardless of the level of information from which the rule was developed. This should lessen the possibility of reducing less-productive stocks to levels at which they would require protection under the ESA, especially if relatively little were known about those stocks. 3.5.2 Environmental Fluctuations Fish stocks undergo natural fluctuations in abundance. These fluctuations are principally due to year-to-year changes in recruitment which are often environmentally induced. Environmental influences can be inter-decadal in nature, with a low level of predictability. Harvest policies should prepare for these natural swings in abundance, which may be greater than half to double the target level of abundance. It is convenient to classify the impacts of recruitment variability (independent of stock size) on implementation of target control rules into one of three types: A. Short-term (year-to-year) fluctuations in recruitment are frequently difficult to measure until the fish have been in the population for several years. This causes uncertainty in the estimation of current stock abundance, thus introducing some random Case 1:06-cv-01815-JDB Document 35-3 Filed 09/17/2007 Page 41 of 56 40 error in the implementation of the control rule. B. Medium-term (3-10 year; Francis and Hare 1994, Jacobson and MacCall 1995) fluctuations in recruitment can impact rebuilding time frames. While the expected time to rebuilding may be calculated to be, say, less than 10 years, the actual time to rebuilding will be shorter or longer depending on the actual sequence of recruitments over the 10- year period. When recruitment is highly variable, the actual time to rebuilding will usually also be highly variable. This is one of the reasons why it is important to account for future recruitment uncertainty in developing rebuilding plans. C. Longer-term (decadal) climate conditions appear to impact recruitment dynamics (Alheit and Hagen 1997, MacCall 1996), producing prolonged periods with above-average (or below-average) recruitment. In an evolutionary sense, fish stocks have adapted to this pattern, and harvest policies should attempt to preserve this adaptation. It may be therefore necessary to design control rules that conserve spawning stock abundance during prolonged periods of poor recruitment to preserve a stock’s capability to produce higher recruitment when environmental conditions improve. In some cases, environmental effects may be directly integrated into the stock assessment and the control rule. However, one should be cautious in interpreting a long run of good or poor recruitments as indicative of an environmentally-driven change in stock productivity. In particular, for a period of declining abundance, the “burden of proof” should initially rest on demonstrating that the environment (as opposed to fishing) caused the decline, and that, therefore, the target control rule should be modified. However, if productivity has in fact declined, more conservative limit and target reference points will be needed . 3.5.3 Stock Definition Issues A “stock” or “stock complex” is a management unit in the sense of the Magnuson-Stevens Act's first definition of the term “fishery”: “One or more stocks of fish that can be treated as a unit for purposes of conservation and management and that are identified on the basis of geographic, scientific, technical, recreational, or economic characteristics.” Defining a "stock" on a scientific basis is a very difficult task. Many types of information are used to identify stocks: Distribution and movements, population trends, morphological differences, genetic differences, contaminants and natural isotope loads, parasite differences, and oceanographic habitat differences. Evidence of morphological or genetic differences in animals from different geographic regions normally indicates that the populations are reproductively isolated. Separate management is usually appropriate when such differences are found. Failure to detect differences experimentally, however, does not mean the opposite. Dispersal rates, though sufficiently high to homogenize morphological or genetic differences detectable experimentally between putative populations, may still be insufficient to deliver enough recruits from an unexploited population (source) to an adjacent exploited population (sink) to prevent local extinctions leading to contraction or fragmentation of range. When the distribution of fishing effort corresponds spatially with the density of the target species, management errors caused by improper stock definition are likely to be Case 1:06-cv-01815-JDB Document 35-3 Filed 09/17/2007 Page 42 of 56 41 small. However, for multispecies fisheries and particularly for by-caught species, fishing effort may be concentrated in only a portion of a species' range. The risk of local depletion leading to range contraction or fragmentation is particularly high for long-lived species with high site fidelity. Careful consideration needs to be given to how stocks are defined scientifically. In the absence of adequate information on stock structure, a species' range within an ocean should be divided into stocks that represent useful management units. Examples of such management units include distinct oceanographic regions, semi-isolated habitat areas, and areas of higher density of the species that are separated by relatively lower density areas. 3.5.4 Special Life Histories Delayed maturity, where fish become vulnerable to fishing before they are reproductively mature, can pose a risk of recruitment overfishing. Proxy policies such as F and F=M may be too high in such cases. SPR-based policies such as F ccount for0.1 35% impacts on spawning potential and tend to provide more precaution in this respect (Clark 1991; Goodyear 1993). Protandric hermaphrodites may be considered as cases of late sexual maturity, and an SPR approach based on female maturity schedules should be adequate. Species with life stages or behaviors that are highly vulnerable to fishing merit precautionary management. Groupers may be protogynous hermaphrodites, and form very large and predictable spawning aggregations that render them highly vulnerable to fishing, risking both depletion and disturbed population structure due to targeting on large males (Bannerot et. al. 1987). Precaution might require severe reductions in fishing pressure, and perhaps a ban on fishing during these vulnerable time periods. No-fishing areas (a.k.a. Marine Protected Areas) could also be appropriate for these species. Fishes with low frequency variability in recruitment or with rare large recruitments may also require a precautionary reduction in fishing. Clark (1993) showed that an F40% SPR-based fishing rate is preferable to his generally recommended F policy if there is35% high serial correlation in annual recruitment. Management of rarely-recruiting species should adopt a very high SPR so that sufficient biomass survives the intervals between major recruitment events. Similarly, certain taxa (e.g., elasmobranchs) that are highly vulnerable to fishing due to their low productivity should be managed to ensure very high SPR. 3.5.5 Data Issues The precautionary approach dictates that greater caution be used in the face of greater uncertainty. Thus, improved knowledge of stock dynamics and of the effects of fishing should result in higher benefits to the Nation through higher yields and lower risks of stock depletion (the relative benefits and costs of enhanced research can be evaluated with the methods presented in Sections 3.1 and 3.2). As noted by FAO (1995b, section 4.2), a precautionary approach “requires explicit specification of the information needed to achieve the management objectives, taking account of the management structure, as well as of the processes required to Case 1:06-cv-01815-JDB Document 35-3 Filed 09/17/2007 Page 43 of 56 42 ensure that these needs are met.” Data should be collected to improve data quality from a lower tier to a higher tier level of data richness. Logbooks from commercial fishing operations may be useful, whereby daily fishing logs would record target catch and bycatch amount, by species, by fishing statistical area, by gear type, and by units of fishing effort. Any self-report information, such as that contained in logbooks, should be verifiable. Improved data collection systems should also be implemented for recreational fisheries. Scientific observer coverage should also be encouraged, whenever feasible, for independent scientific sampling of commercial and recreational catches. Scientific (fishery-independent) surveys should also be conducted to estimate the distribution, relative or absolute abundance, age/length frequency, and other relevant biological characteristics of the stocks to improve data quality to a higher data quality tier. An important aspect of fishery-independent monitoring is that it can form the basis for addressing issues and questions that are not necessarily of immediate concern but may become important in the future. Another important data issue is that of the appropriateness of certain types of data for use in assessment models. Although catch per unit of effort (CPUE) has a long history of use as a fishery-based index of abundance, it also has often proved insensitive to changes in true abundance, particularly when not properly standardized, and its uncritical use has contributed to the collapse of major world fisheries, including the northern cod (Hutchings 1996). Walters and Ludwig (1994) go so far as to say “We flatly recommend that catch/effort data never be used as a direct abundance index (assumed proportional to stock size).” Given the dangers of unvalidated CPUE, the precautionary approach would call for the burden of proof to be placed on demonstrating that CPUE is linearly related to abundance. Patterns such as that shown in Hutchings (1996) and other studies suggest that CPUE often varies approximately in proportion to the square root of abundance. Thus, in cases where a nonlinear relationship between catchability and stock biomass is suspected, it may be necessary to transform CPUE (e.g., by squaring it) before using it as an index of abundance (MacCall in prep.). In addition, standardization of CPUE series may fail to account for increases in fishing power due to the unavailability of appropriate data on gear/vessel configuration and fishing tactics for use in the analyses. In such cases, it is risky to assume that catchability remains constant over time and it may be necessary to adjust CPUE (e.g., by assuming a 3%-5% increase in fishing power per year) before using it as an index of abundance. Such adjustments to CPUE data, while difficult to justify in the absence of direct evidence, may be necessary to reduce the chances of overly- optimistic perceptions of stock status. These risks should be clearly communicated to managers and the public so that they understand that the CPUE adjustments may be necessary in order to avoid serious biases in the assessment. Of course, the preferred remedial action to take is to develop accurate fishery-independent indices of stock abundance. 3.5.6 New Fisheries New fisheries should be viewed as data-poor cases. Initially, fishing should be largely exploratory in nature, and aimed at gathering sufficient information to bring the level of information content up to at least data-moderate standards. New fisheries present Case 1:06-cv-01815-JDB Document 35-3 Filed 09/17/2007 Page 44 of 56 43 opportunities to estimate life history parameters such as natural mortality, which should be considered when planning for data collection. It is precautionary to develop new fisheries gradually from an unexploited state to a fully-exploited state over a period of more than one generation time in order to obtain information from intermediate stock sizes that may be vital to determining B . FAO (1995b, section 3.5) contains other recommendationsMSY for a precautionary approach to managing new fisheries. 3.5.7 Other Precautionary Tactics A number of fishery management tools (or tactics) possess precautionary properties and may be useful mechanisms to ensure that limit reference points are not exceeded. For example, allowing fish to spawn at least once before becoming vulnerable to the fishing gear adds a measure of protection against biased estimates of stock status (Myers and Mertz 1998). Marine Protected Areas (MPAs), wherein all fishing is prohibited, are an extension of area closures, and include precautionary properties (Bohnsack 1996). MPAs may allow a segment of the resource to preserve its unexploited life history, age structure, ecological relationships, etc., in the presence of exploitation. MPAs have limited benefit for highly mobile resources such as pelagic fishes. Somewhat analogous to an MPA is a “biomass reserve”, where a fixed amount of the resource is set aside before applying a target management measure such as F . This alternative approach may reduce the need for35% precise specification of SPR in F policies, offsets imprecision in stock assessments,%SPR and may be especially useful in managing rarely recruiting species that are easily subject to depletion. Other tactics that may have precautionary properties include: (a) Use of "clean" gear types to minimize impacts of fisheries on the stocks, (b) restrictions on the physical characteristics of gear (such as mesh size, hook size, and other physical characteristics) to minimize impacts of fisheries on the stocks and damage to the habitat, (c) modifying fishing characteristics to minimize impacts of fisheries on the stocks and damage to the habitat, and (d) modifying fishing seasons to achieve conservation goals. Adoption of any of the above or similar conservative tactics into an FMP does not guarantee that the NSGs’ recommendations for achieving National Standard 1 will be satisfied. Nevertheless, it is important to consider these as management options that possess desirable conservation properties. Case 1:06-cv-01815-JDB Document 35-3 Filed 09/17/2007 Page 45 of 56 44 CONCLUDING REMARKS Specification of status determination criteria and target control rules is a challenging exercise. Key to this process is communication among managers, scientists, industry and the public. In the face of conflicting objectives, it is essential to understand the tradeoffs associated with alternative control rules and the importance of the weights assigned to the different objectives or performance criteria. Simulation frameworks of the type highlighted in Section 3.2 can facilitate these interactions. Simulation tools should also be used to examine the performance of management systems as a whole, including data collection, assessments, control rules, and implementation of management tactics. ACKNOWLEDGMENTS The authors are grateful to M. Fogarty, P. Goodyear, L. Kell, F. Serchuk and K. Stokes for their detailed reviews of earlier drafts. Case 1:06-cv-01815-JDB Document 35-3 Filed 09/17/2007 Page 46 of 56 45 REFERENCES Alheit, J. and E. Hagen. 1997. Long-term climate forcing of European herring and sardine populations. Fish. Oceanogr. 6:130-139. Bannerot, S., W. W. Fox, Jr., and J. E. Powers. 1987. Reproductive strategies and the management of snappers and groupers in the Gulf of Mexico and Caribbean. pp. 561-603 In: J. J. Polovina and S. Ralston [eds.] Tropical snappers and groupers: Biology and fisheries management. Westview Press, 659p. Beverton, R. J. H., and S. J. Holt. 1957. On the dynamics of exploited fish populations. Fish. Invest. Ser. 2, vol. 19. U.K. Ministry of Agriculture, Food and Fisheries, London. Bohnsack, J. A. 1996. Maintenance and recovery of reef fishery productivity. Pp. 283-313 In N. V. C. Polunin and C. M. Roberts [eds.] Reef Fisheries. Chapman and Hall, London. Butterworth, D. S . and M. O. Bergh. 1993. The development of a management procedure for the South African anchovy resource. p. 83-99 in S. J. Smith, J. J. Hunt and D. Rivard [ed.]. Risk evaluation and biological reference points for fisheries management. Can. Spec. Publ. Fish. Aquat. Sci. 120. Caddy, J. F., and R. Mahon. 1995. Reference points for fishery management. FAO Fish. Tech. Pap. 347. 82 p. Caswell, H. 1989. Matrix population models. Sinauer Associates, Sunderland, MA. 328 p. Clark, W. G. 1991. Groundfish exploitation rates based on life history parameters. Can. J. Fish. Aquat. Sci. 48:734-750. Clark, W. G. 1993. The effect of recruitment variabil ty on the choice of a target level of spawning biomass per recruit. Pp. 233-246 In G. Kruse, D. M. Eggers, R. J. Marasco, C. Pautzke and T. Quinn II [eds.] Proceedings of the International Symposium on Management Strategies for Exploited Fish Populations. Alaska Sea Grant College Program, P. O. Box 755040, Fairbanks AK. Cook, R. M. in press. A sustainability criterion for the exploitation of North Sea cod. ICES J. Mar. Sci. (1998). de la Mare, W.K. 1986. On the management of exploited whale populations. Ph.D. Thesis, University of York. Food and Agriculture Organization (FAO). 1995a. Code of conduct for responsible fisheries. FAO Mimeo. Rome, FAO. 14 p. Food and Agriculture Organization (FAO). 1995b. Precautionary approach to fisheries. Part I: Guidelines on the precautionary approach to capture fisheries and species introductions. Elaborated by the Technical Consultation on the Precautionary Approach to Capture Fisheries (Including Species Introductions). Lysekil, Sweden, 6-13 June 1995. FAO Tech. Pap. 350, Part 1. 52 p. Francis, R. C. And S. R. Hare. 1994. Decadal-scale regime shifts in the large marine ecosystems of the North-east Pacific: a case for historical science. Fish. Oceanogr. 3:279-291. Goodyear, C.P. 1993. Spawning stock biomass per recruit in fisheries management: foundation and current use. P. 67-81 in S.J. Smith, J.J. Hunt, and D. Rivard [eds.] Risk evaluation and biological reference points for fisheries management. Can. Spec. Publ. Fish. Aquat. Sci. 120. Goodyear, C.P. 1995. Red snapper in U.S. waters of the Gulf of Mexico. NMFS/SEFSC Contribution MIA-95/96-05. Case 1:06-cv-01815-JDB Document 35-3 Filed 09/17/2007 Page 47 of 56 46 Gulland, J. A. 1983. Fish stock assessment: A manual of basic methods. John Wiley and Sons, New York. 223p. Gulland, J.A. and L.K. Borema. 1973. Scientific advice on catch levels. Fish. Bull. 71:325-335. Hilborn, R., and C. J. Walters. 1992. Quantitative fisheries stock assessment: choice, dynamics and uncertainty. Chapman and Hall, N.Y. 570 p. Hutchings, J. A. 1996. Spatial and temporal variation in the density of northern cod and a review of hypotheses for the stock’s collapse. Can. J. Fish. Aquat. Sci. 53:943-962. International Council for the Exploration of the Sea (ICES) 1997a. Report of the study group on the precautionary approach to fisheries management. ICES CM 1997/Assess:7. 37pp. International Council for the Exploration of the Sea (ICES). 1997b. Report of the Comprehensive Fishery Evaluation Working Group. ICES CM 1997/Assess:15. 140 pp. Jacobson, L. D. And A. D. MacCall. 1995. Stock-recruitment models for Pacific sardine (Sardinops sagax). Can. J. Fish. Aquat. Sci. 52:566-577. Kirkwood, G.P., and A.D.M. Smith. 1996. Assessing the precautionary nature of fishery management strategies. FAO Fish. Tech. Pap. 350(2): 141-158. MacCall, A.D. 1996. Patterns of low-frequency variabil ty in fish populations of the California Current. CalCOFI Reps. 37:100-110. MacCall, A.D. In prep. Use of decision tables to develop a precautionary approach to problems in behavior, life history and recruitment variability. Proceedings of the 5th National Stock Assessment Workshop, V. Restrepo [ed.]. NOAA Tech. Memo. NMFS-F/SPO-##. NMFS, Office of Science and Technology, 1315 East-West Highway, Silver Spring, MD 20910. Mace, P.M. 1994. Relationships between common biological reference points used as thresholds and targets of fisheries management strategies. Can. J. Fish. Aquat. Sci. 51:110-122. Mace, P.M. and W. Gabriel. In prep. Evolution, scope and current applications of the precautionary approach in fisheries. Proceedings of the 5th National Stock Assessment Workshop, V. Restrepo [ed.]. NOAA Tech. Memo. NMFS-F/SPO-##. NMFS, Office of Science and Technology, 1315 East-West Highway, Silver Spring, MD 20910. Mace, P.M. and M.P. Sissenwine. 1989. Biological reference points for New Zealand fisheries assessments. New Zealand Fish. Assess. Res. Doc. 89/11. 11pp. Mace, P.M. and M.P. Sissenwine. 1993. How much spawning per recruit is enough? In S.J. Smith, J.J. Hunt and D. Rivard [eds.] Risk evaluation and biological reference points for fisheries management. Can. Spec. Publ. Fish. and Aquat. Sci. 120:101-118. Myers, R.A., and G. Mertz. 1998. The limits of exploitation: A precautionary approach. Ecol. Appl. 8: S165-S169. Myers, R.A., A.A. Rosenberg, P.M. Mace, N. Barrowman, and V.R. Restrepo. 1994. In search of thresholds for recruitment overfishing. ICES J. Mar. Sci. 51:191-205. National Marine Fisheries Service (NMFS). 1996. Environmental Assessment/Regulatory Impact Review for Amendment 44 to the Fishery Management Plan for the Groundfish Fishery of the Bering Sea and Aleutian Islands Area and Amendment 44 to the Fishery Management Plan for the Groundfish Fishery of the Gulf of Alaska to Redefine Acceptable Biological Catch and Overfishing, Appendix B. AKFSC, NMFS, 7600 Sand Point Way NE., Seattle, WA 98115-0070. National Research Council (NRC). 1998. Improving Fish Stock Assessments. National Academy Case 1:06-cv-01815-JDB Document 35-3 Filed 09/17/2007 Page 48 of 56 47 Press, Washington D.C. 177 pp. Powers, J.E. 1996. Benchmark requirements for recovering fish stocks. N. Amer. J. Fish. Manage. 16:495-504. Powers, J.E., and V.R. Restrepo. 1993. Evaluation of stock assessment research for Gulf of Mexico king mackerel: Benefits and costs to management. N. Amer. J. Fish. Manage. 13: 15-26. Prager, M. H. 1994. A suite of extensions to a nonequilibrium surplus-production model. Fish. Bull. 92: 374-389. Ricker, W. E. 1958. Handbook of computations for biological statistics of fish populations. Fish. Res. Bd. Canada, Bull. No. 119. Rosenberg, A., P. Mace, G. Thompson, G. Darcy, W. Clark, J. Collie, W. Gabriel, A. MacCall, R. Methot, J. Powers, V. Restrepo, T. Wainwright, L. Botsford, J. Hoenig, and K. Stokes. 1994. Scientific review of definitions of overfishing in U.S. Fishery Management Plans. NOAA Tech. Memo. NMFS-F/SPO-17. 205 pp. Rosenberg, A.A., and V.R. Restrepo. 1994. Uncertainty and risk evaluation in stock assessment advice for U.S. marine fisheries. Can. J. Fish. Aquat. Sci. 51: 2715-2720. Serchuk, F.M. and R.J.R. Grainger. 1992. Development of the basis and form of ICES fisheries management advice: historic background (1976-1990) and the new form of ACFM advice. ICES C.M. 1992/Assess:20. 13pp. Serchuk, F., D. Rivard, J. Casey and R. Mayo. 1997. Report of the ad hoc working group of the NAFO Scientific Council on the precautionary approach. NAFO SCS Document 97/12. Serebryakov, V.P. 1991. Predicting year-class strength under uncertainties related to survival in the early life history of some North Atlantic commercial fish. NAFO Sci. Counc. Stud. 16:49-56. Shepherd, J.G. 1991. Report of special session. NAFO Sci. Counc. Stud. 16:7-12. Sissenwine, M.P., and J.G. Shepherd. 1987. An alternative perspective on recruitment overfishing and biological reference points. Can. J. Fish. Aquat. Sci. 44: 913-918. Thompson, G. G. 1992. A Bayesian approach to management advice when stock-recruitment parameters are uncertain. Fish. Bull. 90:561-573. Thompson, G. G. 1993. A proposal for a threshold stock size and maximum fishing mortality rate. In S. J. Smith, J. J. Hunt, and D. Rivard [eds.] Risk evaluation and biological reference points for fisheries management, p. 303-320. Can. Spec. Publ. Fish. Aquat. Sci. 120. Thompson, G. G. 1996. Risk-averse optimal harvesting in a biomass dynamic model. Environmental Assessment/Regulatory Impact Review for Amendment 44 to the Fishery Management Plan for the Groundfish Fishery of the Bering Sea and Aleutian Islands Area and Amendment 44 to the Fishery Management Plan for the Groundfish Fishery of the Gulf of Alaska to Redefine Acceptable Biological Catch and Overfishing, Appendix B. AKFSC, NMFS, 7600 Sand Point Way NE., Seattle, WA 98115-0070. Thompson, G. G. In prep. Optimizing Harvest Control Rules in the Presence of Natural Variability and Parameter Uncertainty. Proceedings of the 5th National Stock Assessment Workshop, V. Restrepo [ed.]. NOAA Tech. Memo. NMFS-F/SPO-##. NMFS, Office of Science and Technology, 1315 East-West Highway, Silver Spring, MD 20910. National Marine Fisheries Service, Office of Science and Technology, 1315 East-West Highway, Case 1:06-cv-01815-JDB Document 35-3 Filed 09/17/2007 Page 49 of 56 48 Silver Spring, MD 20910. Thompson, G. G., and P. M. Mace. 1997. The evolution of precautionary approaches to fisheries management, with focus on the United States. NAFO SCR Doc. 97/26. United Nations (UN). 1995. Agreement for the implementation of the provisions of the United Nations convention on the law of the sea of 10 December 1982 relating to the conservation and management of straddling fish stocks and highly migratory fish stocks. United Nations Conference on Straddling Fish Stocks and Highly Migratory Fish Stocks. Sixth session. New York, 24 July - 4 August 1995. A/CONF. 164/37 8 September 1995. Wade, P. R., and R. P. Angliss. 1997. Guidelines for assessing marine mammal stocks: report of the GAMMS Workshop April 3-5, 1996, Seattle, Washington. NOAA Tech. Memo. NMFS-OPR-12, 93 p. Walters, C. And D. Ludwig. 1994. Calculation of Bayes posterior probability distributions for key population parameters. Can. J. Fish. Aquat. Sci. 51:713-722. Case 1:06-cv-01815-JDB Document 35-3 Filed 09/17/2007 Page 50 of 56 49 APPENDIX A Equilibrium Implications of Fishing at 75% FMSY The simple, deterministic model described in Mace (1994) was used to evaluate the consequences of fishing at the default target of 75% F . Since the calculations wereMSY deterministic and the equilibrium biomass associated with a fishing mortality rate below F will always exceed B , it was not necessary to take explicit account of the behaviorMSY MSY of the default target at biomass levels below B . This model is age-structured withMSY natural mortality constant over all ages, knife-edge recruitment and maturity, growth rates represented by a von Bertalanffy growth function, and recruitment represented by either a Beverton-Holt relationship or a Ricker relationship. The procedures used to run the model were the same as those described in Mace (1994), except that the outputs of primary interest were the equilibrium yield at 75% F (abbreviated Y75), the equilibriumMSY biomass at 75% F (B75), the ratio Y75/MSY, and the ratio B75/B . Since theMSY MSY biomass is calculated as the average level present during the course of the fishing year, the ratio B75/B is equivalent to 1.333*(Y75/MSY). These calculations were performed forMSY all combinations of natural mortality (M) = 0.1, 0.2, and 0.3; Brody growth coefficient in von-Bertalanffy equation (K) = 0.1, 0.2, and 0.3; age of recruitment (t ) equal to age ofr maturity (t ), both knife-edged at ages 3, 5, 7, and 9 years; and extinction parameter () =m 0.05, 0.10, 0.15, 0.20, 0.25, 0.30, 0.35, 0.40, 0.45, 0.50 (where 100* represents the level of %SPR corresponding to the slope at the origin of a stock-recruitment relationship) with a Beverton-Holt stock-recruitment relationship for which maximum (asymptotic) recruitment was fixed at 10 recruits for all parameter combinations. Additional runs8 combining M and/or K = 0.4 with the other parameter values were also conducted. Even though some of these parameter combinations resulted in rather unlikely sets of life history characteristics, the ratios calculated were remarkably consistent across parameter combinations: Y75/MSY ranged between 0.949 and 0.983 and B75/BMSY ranged between 1.265 and 1.311. Selected results for these and other variables are shown in Table A1. Similar calculations were conducted for a Ricker stock-recruitment function with maximum recruitment fixed at 10 . Parameter values and combinations were the same as8 those used with the Beverton Holt stock-recruitment function, except that only one age of recruitment was used (t = 5). For this formulation, Y75/MSY ranged between 0.940 andr 0.963, and B75/B ranged between 1.253 and 1.284 (Table A1).MSY Case 1:06-cv-01815-JDB Document 35-3 Filed 09/17/2007 Page 51 of 56 50 Table A1. Equilibrium yield and biomass levels corresponding to F and 0.75 FMSY MSY (selected results from 600 parameter and model combinations). SRR: stock-recruitment relationship (B-H = Beverton-Holt, R = Ricker). 0.75* Y75/ B75/ SRR M K tτ r FMSY FMSY MSY BMSY Y75 MSY BMSY B-H 0.1 0.1 0.05 5 0.091 0.068 12096 133565 11770 0.973 1.298 B-H 0.1 0.1 0.20 5 0.051 0.038 7223 141068 6941 0.961 1.281 B-H 0.1 0.1 0.50 5 0.022 0.016 2279 105381 2175 0.955 1.273 B-H 0.1 0.2 0.05 5 0.147 0.110 30719 209012 30007 0.977 1.302 B-H 0.1 0.2 0.20 5 0.074 0.056 17594 237692 16946 0.963 1.284 B-H 0.1 0.3 0.05 5 0.200 0.150 45966 229351 45008 0.979 1.306 B-H 0.1 0.3 0.20 5 0.091 0.068 25388 278511 24494 0.965 1.286 B-H 0.2 0.1 0.05 5 0.189 0.141 7042 37333 6873 0.976 1.301 B-H 0.2 0.1 0.20 5 0.099 0.075 4120 41422 3964 0.962 1.283 B-H 0.2 0.2 0.05 9 0.501 0.375 45113 90125 44315 0.982 1.310 B-H 0.2 0.2 0.05 5 0.300 0.225 23231 77558 22744 0.979 1.306 B-H 0.2 0.2 0.05 3 0.194 0.145 13215 68123 12873 0.974 1.299 B-H 0.2 0.2 0.20 9 0.195 0.146 23811 122170 23012 0.967 1.289 B-H 0.2 0.2 0.20 5 0.141 0.106 13090 92667 12619 0.964 1.285 B-H 0.2 0.2 0.20 3 0.107 0.080 7831 73125 7529 0.961 1.282 B-H 0.2 0.2 0.50 9 0.069 0.052 6897 99668 6568 0.952 1.270 B-H 0.2 0.2 0.50 5 0.055 0.041 3961 72352 3764 0.950 1.267 B-H 0.2 0.2 0.50 3 0.045 0.034 2456 54969 2331 0.949 1.266 B-H 0.2 0.3 0.05 5 0.405 0.304 39200 96819 38446 0.981 1.308 B-H 0.2 0.3 0.20 5 0.175 0.131 21411 122555 20667 0.965 1.287 B-H 0.3 0.1 0.05 5 0.329 0.246 5447 16579 5331 0.979 1.305 B-H 0.3 0.1 0.20 5 0.159 0.119 3105 19555 2992 0.964 1.285 B-H 0.3 0.2 0.05 5 0.499 0.374 20371 40864 19984 0.981 1.308 B-H 0.3 0.2 0.20 5 0.217 0.163 11226 51639 10833 0.965 1.287 B-H 0.3 0.3 0.05 9 0.926 0.695 61113 65962 60059 0.983 1.310 B-H 0.3 0.3 0.05 5 0.651 0.489 36410 55889 35756 0.982 1.309 B-H 0.3 0.3 0.05 3 0.395 0.297 19438 49150 19011 0.978 1.304 B-H 0.3 0.3 0.20 9 0.337 0.253 31391 93032 30363 0.967 1.290 B-H 0.3 0.3 0.20 5 0.264 0.198 19555 73941 18888 0.966 1.288 B-H 0.3 0.3 0.20 3 0.195 0.146 11114 57070 10707 0.963 1.285 B-H 0.3 0.3 0.50 9 0.115 0.087 8917 77240 8492 0.952 1.270 B-H 0.3 0.3 0.50 5 0.096 0.072 5738 59609 5458 0.951 1.268 B-H 0.3 0.3 0.50 3 0.077 0.058 3399 44086 3228 0.950 1.267 R 0.2 0.2 0.05 5 0.669 0.502 30262 45243 29096 0.962 1.282 R 0.2 0.2 0.20 5 0.190 0.142 23630 124380 22459 0.950 1.267 R 0.2 0.2 0.50 5 0.061 0.045 9037 149062 8522 0.943 1.257 R 0.3 0.3 0.05 5 1.458 1.094 50728 34784 48840 0.963 1.284 R 0.3 0.3 0.20 5 0.358 0.268 35826 100105 34121 0.952 1.270 R 0.3 0.3 0.50 5 0.107 0.080 13120 122951 12385 0.944 1.259 Case 1:06-cv-01815-JDB Document 35-3 Filed 09/17/2007 Page 52 of 56 51 APPENDIX B Glossary Availability . Refers to the distribution of fish of different ages or sizes relative to that of the fishery. B. Biomass, measured in terms of spawning capacity (in weight) or other appropriate units of production. B . Virgin stock biomass, i.e. the long-term average biomass value expected in the absence of0 fishing mortality. In Section 3.1, B is used as the biomass at the start of a population0 projection. B . Long-term average biomass that would be achieved if fishing at a constant fishing mortalityMSY rate equal to F .MSY BRP (Biological Reference Point). Benchmarks against which the abundance of the stock or the fishing mortality rate can be measured, in order to determine its status. BRPs can be categorized as limits or targets, depending on their intended use (see also Reference Points). There are also socio-economic reference points, but those are not treated in any detail in this document. Catchability . Proportion of the stock removed by one unit of effective fishing effort (typically age-specific due to differences in selectivity and availability by age). Control Rule. Describes a plan for pre-agreed management actions as a function of variables related to the status of the stock. For example, a control rule can specify how F or yield should vary with biomass. In the NSGs, the “MSY control rule” is used to determine the limit fishing mortality, MFMT. Control rules are also known as “decision rules” or “harvest control laws” in some of the scientific literature. CPUE (Catch per Unit of Effort). Measures the relative success of fishing operations, but is also sometimes used a proxy for relative abundance based on the assumption that CPUE is linearly related to stock size. The use of CPUE that has not been properly standardized for temporal-spatial changes in catchability is highly undesirable. DAH (Domestic Annual Harvest). ESA (Endangered Species Act). F. Instantaneous fishing mortality rate. Measures the effective fishing intensity for a given partial recruitment pattern. F . Fishing mortality at which the slope of equilibrium yield per recruit (YPR) is reduced to 10%0.1 of the slope when F=0. F . Fishing mortality rate corresponding to an equilibrium SPR equal to the inverse of the 90high th percentile observed survival ratio. F . Fishing mortality rate corresponding to an equilibrium SPR equal to the inverse of the 10low th percentile observed survival ratio. F . Fishing mortality at which the slope of equilibrium yield per recruit (YPR) is zero (may bemax undefined in some cases where the YPR-F curve is asymptotic). F . Fishing mortality rate corresponding to an equilibrium SPR equal to the inverse of themed median observed survival ratio. f . Effective fishing effort corresponding to F .MSY MSY Case 1:06-cv-01815-JDB Document 35-3 Filed 09/17/2007 Page 53 of 56 52 F . Fishing mortality rate which, if applied constantly, would result in MSY.MSY F (also F , F ). Fishing mortality rate corresponding to an equilibrium SPR equal to the extinction crash inverse of the survival ratio at the origin of the stock-recruitment relationship. A stock fished at or above this level for a prolonged period of time is expected to collapse. F . Fishing mortality rate that results in x% equilibrium spawning potential ratio.x% FMP (Fishery Management Plan). A plan containing conservation and management measures for fishery resources, and other provisions required by the MSFCMA, developed by the Fishery Management Councils or the Secretary of Commerce. Generation Time. In the context of the NSGs, generation time is a measure of the time required for a female to produce a reproductively-active female offspring for use in setting maximum allowable rebuilding time periods. Several estimators of generation time are available in the literature, and one is presented in Section 3.4. Limit Reference Points. Benchmarks used to indicate when harvests should be constrained substantially so that the stock remains within safe biological limits. The probability of exceeding limits should be low. In much of the NSGs, limits are referred to as thresholds. In much of the international literature (e.g., FAO documents), “thresholds” are used as buffer points that signal when a limit is being approached. M. Instantaneous natural mortality rate. MESY (Maximum expected stationary yield). Maximum statistical expectation of long-term yield, considering uncertainties in parameter values and natural (process) variability. MELSY (Maximum expected log stationary yield). Maximum statistical expectation of the logarithm of long-term yield, considering uncertainties in parameter values and natural (process) variability. MFMT (Maximum Fishing Mortality Threshold). SDC for determining if overfishing is occurring. It will usually be equivalent to the F corresponding to the MSY Control Rule. MSFCMA (Magnuson-Stevens Fishery Conservation and Management Act). U.S. Public Law 94-265, as amended through October 11, 1996. Available as NOAA Technical Memorandum NMFS-F/SPO-23, 1996. MSST (Minimum Stock Size Threshold). The greater of (a) ½B , or (b) the minimum stock sizeMSY at which rebuilding to B will occur within 10 years of fishing at the MFMT. MSSTMSY should be measured in terms of spawning biomass or other appropriate measures of productive capacity. MSY (Maximum Sustainable Yield). Largest long-term average yield (catch) that can be taken from a stock (or stock complex) under prevailing ecological and environmental conditions. Any estimate of MSY depends on the population dynamics of the stock, the characteristics of the fisheries (e.g. gear selectivity), and the control rule used. In much of the traditional fisheries literature, MSY is estimated with a control rule in which F is independent of stock size. In the language of the NSGs, estimates of MSY will change depending on the shape of the control rule, but B and F pertain only to a constant-F control rule.MSY MSY NSGs (National Standard Guidelines). Advisory guidelines developed by NMFS, based on the National Standards of the MSFCMA, intended to assist in the development of FMPs. Case 1:06-cv-01815-JDB Document 35-3 Filed 09/17/2007 Page 54 of 56 Copies of the NSGs and other relevant documents that have appeared in the Federal Register can be3 obtained in the Web at http://www.nmfs.gov/sfa. 53 Published in the Federal Register first as proposed rule on August 4, 1997, and then3 revised as final rule on May 1, 1998. Overfished. According to the NSGs, an overfished stock or stock complex is one “whose size is sufficiently small that a change in management practices is required in order to achieve an appropriate level and rate of rebuilding.” A stock or stock complex is considered overfished when its size falls below the MSST. A rebuilding plan is required for stocks that are overfished. Overfishing. According to the NSGs, “overfishing occurs whenever a stock or stock complex is subjected to a rate or level of fishing mortality that jeopardizes the capacity of a stock or stock complex to produce MSY on a continuing basis.” Overfishing is occurring if the MFMT is exceeded for 1 year or more. OY (Optimum Yield). The amount of fish that will provide the greatest overall benefit to the Nation, particularly with respect to food production and recreational opportunities and taking into account the protection of marine ecosystems. MSY constitutes a “ceiling” for OY. OY may be lower than MSY, depending on relevant economic, social, or ecological factors. In the case of an overfished fishery, OY should provide for rebuilding to B .MSY Partial Recruitment. Patterns of relative vulnerability of fish of different sizes or ages due to the combined effects of selectivity and availability. Rebuilding Plan. A plan that must be designed to recover stocks to the B level within 10 yearsMSY when they are overfished (i.e. when B < MSST). Normally, the 10 years would refer to an expected time to rebuilding in a probabilistic sense. Recent Catch. In the context of this document, this term should be interpreted as the average catch during a time period (e.g., 5 years) for which there is evidence of stable abundance. As this type of information is unlikely to be available in many data-poor cases, scientists could carefully consider defining Recent Catch as the median catch during the last 5, 10 or 15 years. Reference Points. Values of parameters (e.g. B , F , F ) that are useful benchmarks forMSY MSY 0.1 guiding management decisions. Biological reference points are typically limits that should not be exceeded with significant probability (e.g. MSST) or targets for management (e.g. OY). Risk. The probability of an event times the cost associated with the event (loss function). Sometimes “risk” is simply used to denote the probability of an undesirable result (e.g. the risk of biomass falling below MSST). SDC (Status Determination Criteria). Objective and measurable criteria used to determine if a stock is being overfished or is in an overfished state according to NSGs. Selectivity. Measures the relative vulnerability of different age (size) classes to the fishing gears(s). SPR (1). Spawning output Per Recruit: Amount of per-capita spawning biomass (or other appropriate measure of reproductive output) obtained at a given value of F, conditional on values of partial recruitment, growth, maturity (and/or fecundity) and natural mortality. (2). Spawning Potential Ratio: The expected lifetime spawning output per recruit relative to the spawning output that would be realized in the absence of fishing, often expressed as Case 1:06-cv-01815-JDB Document 35-3 Filed 09/17/2007 Page 55 of 56 54 a percentage. Throughout this document, references to the second definition are associated with a percentage (%) sign. Survival Ratios. Ratios of recruits to spawners (or spawning biomass) in a stock-recruitment analysis. Target Reference Points. Benchmarks used to guide management objectives for achieving a desirable outcome (e.g. OY). Target reference points should not be exceeded on average. Uncertainty. Uncertainty results from a lack of perfect knowledge of many factors that affect stock assessments, estimation of reference points, and management. Rosenberg and Restrepo (1994) identify 5 types: Measurement error (in observed quantities), process error (or natural population variability), model error (mis-specification of assumed values or model structure), estimation error (in population parameters or reference points, due to any of the preceding types of errors), and implementation error (or the inability to achieve targets exactly for whatever reason). YPR (Yield per Recruit). Amount of per-capita yield obtained at a given value of F, conditional on values of partial recruitment, growth and natural mortality. Case 1:06-cv-01815-JDB Document 35-3 Filed 09/17/2007 Page 56 of 56 UNITED STATES DISTRICT COURT FOR THE DISTRICT OF COLUMBIA NORTH CAROLINA FISHERIES ASS’N, INC., et al., Plaintiffs, v. THE HONORABLE CARLOS GUTIERREZ, in his official capacity as the Secretary of Commerce, Defendant. ) ) ) ) ) ) ) ) ) ) ) ) ) ) ) Civil Action No. 06-cv-01815-JDB [PROPOSED] ORDER Upon consideration of [19] plaintiffs’ motion for summary judgment, [21] defendant’s cross-motion for summary judgment, [27] the brief filed by the State of North Carolina as amicus curiae, the arguments presented by the parties at the motions hearing held on June 14, 2007, [31, 32] the parties’ supplemental briefs, the parties’ remedy proposals, and the administrative record, and for the reasons stated in the Memorandum Opinion issued on August 17, 2007, it is this ____ day of September, 2007, hereby ORDERED that defendant shall implement rebuilding plans for snowy grouper and black sea bass by March 31, 2008. Date: ____________________ _______________________ Honorable John D. Bates United States District Judge Case 1:06-cv-01815-JDB Document 35-4 Filed 09/17/2007 Page 1 of 1