International Pacifi c Halibut Commission Eighty-fi fth Annual Meeting

Schedule of Sessions ...... 2

Public Session Agenda ...... 3

The Pacifi c halibut fi shery, 2008 ...... 5 Heather L. Gilroy

Analysis of PIT tag recoveries through 2008 ...... 21 Raymond A. Webster

Removal fi shing to estimate catch probability: preliminary data analysis ...... 27 Raymond A. Webster

Exploring effects of fi shing and migration on the distribution of Pacifi c halibut ...... 37 Juan L. Valero and Steven R. Hare

IPHC Biomass Apportionment Workshop summary and responses to signifi cant questions arising at the Workshop ...... 51 Steven R. Hare, Ray A. Webster, Juan L. Valero, and Bruce M. Leaman

Assessment of the Pacifi c halibut stock at the end of 2008 ...... 83 Steven R. Hare and William G. Clark

IPHC Staff regulatory proposals: 2009 ...... 137 Bruce M. Leaman and Heather L. Gilroy

Appendix I. IPHC Research program: Review of 2008 projects and preliminary proposals for 2009 ...... 143 IPHC Staff

Appendix II. Tide predictions ...... 171

1 2009 IPHC ANNUAL MEETING HANDOUT International Pacifi c Halibut Commission Eighty-fi fth Annual Meeting Vancouver Marriott Pinnacle Downtown Hotel Vancouver, B.C. January 12-16, 2009

Schedule of Sessions

Monday - January 12 Room Location p.m. 5:00 - 8:30 United States Delegation Meeting Pinnacle III 5:00 - 7:30 Canadian Delegation Meeting Pinnacle I & II

Tuesday - January 13 a.m. 8:00 - 9:00 IPHC Executive Session I Shaughnessy Salon 9:00 - 1:00 Public Session I Pinnacle I & II p.m. 2:30 - 5:00 IPHC Administrative Session I Shaughnessy Salon 2:30 - 5:00 Conference Board (CB) Session I Pinnacle II & III 2:30 - 5:00 Processor Advisory Group (PAG) I Point Grey 7 :00 - 9:30 IPHC Reception Pinnacle Foyer & I (Note: Reception open to all IPHC meeting participants) Wednesday - January 14 a.m. 8:30 - 5:00 IPHC Administrative Session II Shaughnessy Salon 8:30 - 5:00 Conference Board Session II Pinnacle II & III 8:30 - 5:00 Processor Advisory Group II Point Grey p.m. 5:30 - 7:30 HANA Reception Renaissance Harbourside Ballroom (Note: Reception open to commercial industry and agency staff only) Thursday - January 15 a.m. 8:30 - 9:30 IPHC, CB, and PAG Joint Session Pinnacle II & III 9:30 - 12:00 IPHC Administrative Session III Shaughnessy Salon p.m. 1:30 - 5:00 IPHC Administrative Session IV Shaughnessy Salon

Friday - January 16 a.m. 7:30 - 8:30 IPHC Executive Session II Pinnacle I 9:00 - 11:00 IPHC Meeting (public welcome) Pinnacle I

• IPHC offi ce is located in Hollyburn • Breakout room available in Kitsilano (contact IPHC staff for availability)

2 2009 IPHC ANNUAL MEETING HANDOUT International Pacifi c Halibut Commission Eighty-fi fth Annual Meeting Vancouver Marriott Pinnacle Downtown Hotel Vancouver, B.C. January 12-16, 2009

Public Session Agenda January 13, 2009 Pinnacle I & II

9:00 a.m. Opening of Meeting Chairperson and Vice-chairperson opening remarks Introductions

9:15 Director’s Remarks

Staff Presentations The fi shery PIT tagging and removal experiment Impact of migration and local removals on exploitable biomass Questions and issues arising from Apportionment Workshop Stock assessment and harvest policy Staff regulatory proposals: 2009

10:15 Coffee

10:30 Staff presentations continued

12:00 Questions and discussion

1:00 p.m. Announcements and Adjournment

7:00-9:30 IPHC Reception (no host) – Pinnacle Foyer & I

3 2009 IPHC ANNUAL MEETING HANDOUT Notes

4 2009 IPHC ANNUAL MEETING HANDOUT The Pacifi c halibut fi shery, 2008

Heather L. Gilroy

The removals of halibut off the Pacifi c coast totaled 84 million pounds in 2008. This is lower than in recent yeas as the catch has exceeded 90 million pounds annually since 1997. The removals include commercial catch, sport catch, bycatch mortality, personal use, wastage, and International Pacifi c Halibut Commission (IPHC or Commission) research catch (Table 1). In addition to data compiled by the IPHC, other sources of halibut data include federal and state agencies. All 2008 data are net weight and considered preliminary at this time. When the Commission was established in 1923 to manage and conserve the halibut resource, the commercial fi shery, which dates back to the late 1800s, was the primary source of removals. Currently, the commercial catch continues to represent the largest removal. Through the years, estimates of other removals have been added progressively: bycatch mortality in the 1960s, sport catch in the late 1970s, wastage in the 1980s, and personal use (subsistence) in the 1990s. All legal and sublegal-sized halibut removals are estimated and tracked as part of the total removals and are accounted for in the IPHC’s harvest policy and stock assessment. The legal-size limit for the commercial halibut fi shery is 32 inches or greater. Accurate reporting of removals is essential to stock assessment and for determining the recommended allowable catch. Removals are estimated most accurately using a scale weight, such as with the commercial fi shery, or a scientifi cally-based estimation procedure. Over the years, the Commission has worked with different agencies to improve the estimation procedures in attempting to accurately account for all removals. Personal use removals include the Washington State treaty tribal ceremonial and subsistence fi shery, the British Columbia Food, Social, and Ceremonial (FSC) fi shery, and the Alaska subsistence fi shery. The FSC fi shery catch had been estimated at 300,000 pounds since 1993 but in 2008 a new estimate of 405,000 pounds was provided by the Canadian Department of Fisheries and Oceans (DFO). The new FSC catch estimate is a sum of all allocations within the FSC communal licenses including recent treaty settlements (Gary Logan, DFO, pers. comm.). The Alaska subsistence has been just over 1 million pounds since the implementation of the program in 2003. Wastage accounts for the loss of halibut from the commercial fi shery. It includes legal-sized halibut killed by lost and abandoned longline gear and a proportion of the sublegal-sized halibut that are discarded due to regulation and die. In 2008, the removals for commercial catch, sport catch, bycatch mortality, personal use, wastage, and IPHC research were 57.9, 10.8, 10.6, 1.5, 2.5, and 0.7 million pounds respectively. This report reviews commercial and sport fi sheries, bycatch mortality estimates, and progress on the allocation programs by each country.

Allocation programs and totals The authority for allocation among the user groups in each country is the responsibility of each national government. Currently, both the United States and Canadian governments have allocation plans within a single regulatory area or within smaller local areas. For Area 2B, DFO has had an allocation framework for the commercial and recreational sectors since 2005, where the recreational sector is allocated a 12 percent “ceiling” of a combined commercial/recreational

5 2009 IPHC ANNUAL MEETING HANDOUT catch limit. When managed to the allocation ceilings, both sectors’ catch will fl uctuate with stock abundance. For 2008, the Commission adopted a catch limit for Area 2B of 9.0 million pounds for the combined recreational and commercial fi sheries. An additional 19,000 pounds was added for the projected commercial fi shery wastage, resulting in a total catch limit of 9.019 million pounds. DFO then allocated to the commercial fl eet 88 percent of the total catch limit and reduced it by 19,000 pounds to account for the wastage, which resulted in the commercial allocation of 7,917,720 pounds. In 2007, the underage/overage program resulted in a 337,674 pound surplus roll-over to the 2008 catch limit, for an adjusted 2008 commercial fi shery catch limit of 8,255,394 pounds. The remaining 1.082 million pounds of the combined catch limit was allocated to the recreational sector. The 2008 combined commercial and sport catch was 8,000 pounds under the combined adjusted catch limit (Table 2). For managing the sport charter fi sheries in Areas 2C and 3A, the North Pacifi c Fishery Management Council (NPFMC) adopted a Guideline Harvest Level (GHL) program, which went into effect in September, 2003. The GHL program included a provision that the GHL declines by specifi ed increments if halibut abundance declines, but the GHL will not increase above the original level. In 2008, the Area 2C GHL of 931,000 pounds was exceeded by over 100 percent and the Area 3A sport charter fi shery harvest was very close to the GHL (Table 3). The NPFMC has recently adopted a Catch Sharing Plan (CSP) to allocate between the commercial and sport charter sectors in Areas 2C and 3A. The CSP has specifi c allocations which vary depending on the magnitude of a combined commercial/sport charter catch limit adopted by the Commission. It is unlikely that any plan will be in place before 2010. The one area where comprehensive user group allocation occurs is in Area 2A (Washington, Oregon, and California). The Commission determines the total allowable catch for all user groups, and the Pacifi c Fishery Management Council (PFMC) allocates the harvest among user groups according to a CSP. The Commission annually approves the CSP, which determines the catch limits for the different fi sheries. There are three commercial fi sheries (directed, incidental with salmon troll, and incidental with limited-entry sablefi sh longline), a treaty Indian fi shery, and two sport divisions (with nine sub-area sport fi sheries). The 2008 total catch (1.19 million pounds) for commercial, sport, and treaty Indian users was slightly under the area catch limit (Table 4). The Commission adopts biologically-based catch limits for all individual regulatory areas and for Areas 4CDE combined. IPHC considers Area 4CDE to be one biological unit. A CSP developed by the NPFMC specifi es individual catch limits for Areas 4C, 4D, and 4E. This CSP also allows Area 4D Community Development Quota (CDQ) to be harvested in Area 4E, and Area 4C quota shares to be fi shed in Areas 4C or 4D. The total commercial catch of 3.855 million pounds was under the combined Area 4CDE catch limit (3.89 million pounds).

Detailed catch data The commercial fi shery A detailed summary of fi shing seasons, catch limits, and catch by IPHC regulatory area (Fig. 1) is provided in Table 5. The expanded catch limit represents the IPHC catch limit with adjustments from the underage and overage programs from the previous year’s quota share program. The commercial catch occurs in: an open-access fi shery, two incidental catch fi sheries, and a treaty Indian fi shery in Area 2A; the quota share (QS) fi sheries in British Columbia and Alaska; and the Metlakatla fi shery within the Annette Island Reserve in Southeast Alaska. 6 2009 IPHC ANNUAL MEETING HANDOUT Area 2A The IPHC issued licenses to sport charter and commercial vessels in Area 2A. In 2008, a total of 570 Area 2A vessel licenses were issued: 135 licenses for the incidental commercial catch of halibut during the salmon troll fi shery, 296 for the directed commercial fi shery and the incidental halibut during sablefi sh fi shery, and 139 for the sport charter fi shery. Three fewer sport licenses were issued in 2008 (139) than in 2007 (142). There was an increase in number of licenses issued between 2007 and 2008 for the directed commercial/incidental during sablefi sh fi shery (+71) and a decrease for the incidental halibut during the salmon troll season (-160). The change within the commercial fi sheries refl ects the 2008 closure of the salmon troll fi shery south of Cape Falcon in the state of Oregon; the closure prompted salmon troll fi shers to obtain licenses for the directed halibut fi shery as an alternative. In the incidental commercial halibut fi shery conducted during the salmon troll season, the allowable incidental catch ratio was one halibut per two chinook salmon (Oncorhynchus tshawytscha), plus an “extra” halibut per landing. In addition, the total number of incidental halibut per vessel per landing could not exceed 35. The 1:2 ratio of halibut to chinook in 2008 was a drop from the 1:3 from 2000 to 2007. These ratios increased over the years, from the 1:20 ratio seen in the fi rst year of the program (1995). The incidental commercial halibut fi shery during the salmon troll season opened on May 1 and closed on November 15 when the commercial halibut fi shery closed for the year. The halibut catch was 56% (21,022 pounds) under the catch limit. The directed commercial fi shery consisted of four 10-hour fi shing periods with fi shing period limits (Table 6). The fi shing period limits were assigned by vessel class and, for the fi rst two openings, H-class vessels received 9,000 pounds per opening. The limits for the third fi shing period remained high, with H-class vessels receiving 8,000 pounds. The fourth and fi nal fi shing period had a signifi cantly lower catch limit with H-class vessels receiving 1,200 pounds. The total directed commercial catch was 3% (7,300 pounds) over the catch limit, compared to 12% under in 2007. The incidental halibut fi shery during the limited-entry longline sablefi sh season opened May 1 and closed on October 31 with the closure of the sablefi sh season. Incidental landings of halibut in this fi shery were restricted to 100 pounds (dressed weight) of halibut for every 1,000 pounds (dressed weight) of sablefi sh landed, and up to two “extra” halibut in excess of the 100 pounds per 1,000-pound ratio per landing. The catch was 50% (35,039 pounds) under the 70,000 pound catch limit. The decrease in the incidental halibut catch was proportional to the decrease observed in sablefi sh catch during the 2008 season. Since 2005, the Treaty Indian tribes have agreed upon a management plan that includes allocations to tribes or groups of tribes. In the tribal fi shery, 75% of the commercial catch limit was allocated to the separately managed fi shery and was taken between March 8 and June 3. The remaining catch limit (25%) was allocated to the restricted fi shery, subject to daily limits of 500 pounds per vessel. The total tribal commercial catch was 7.5% (29,879 pounds) over the catch limit.

Area 2C Metlakatla fi shery The Metlakatla Indian Community was authorized by the United States government to conduct a commercial halibut fi shery within the Annette Islands Reserve. Eleven 48-hour fi shing periods took place between May 9 and October 5, producing a total catch of 40,855 pounds, which was included in the Area 2C commercial catch. The catch was about 1,000 pounds more than last year’s

7 2009 IPHC ANNUAL MEETING HANDOUT catch of 39,252 pounds. The total catch has varied over time, from a high of 126,000 pounds in 1996 to a low of 12,000 pounds in 1998.

The Quota Share fi sheries The Individual Fishing Quota/Community Development Quota (IFQ/CDQ) halibut and sablefi sh fi sheries in Alaska have been in effect since 1995. The National Marine Fisheries Service (NMFS) Restricted Access Management (RAM) Division allocates halibut QS to recipients by IPHC Regulatory Area. Quota share transfers were permitted with restrictions on the amount of QS a person could hold and the amount that could be fi shed per vessel. As of the end of the 2008 fi shery, RAM reported that 2,911 persons held quota shares, down from the initial 4,830 persons at the start of the program in 1995 (Table 7). In addition, the number of vessels catching halibut was decreased by approximately 36% since the implementation of the QS fi shery in Alaska. The total 2008 catch from the IFQ/CDQ halibut fi shery for the waters off Alaska was 49.5 million pounds, 3% under the catch limit (not adjusted). For Areas 2C and 3A, the commercial QS catch was within 1% of the catch limit; the Area 3B commercial catch was within 2% of the catch limit. For Area 4, Area 4A’s catch was within 5% and Area 4B’s was within 8% of the catch limit. The individual catch limits adopted for Regulatory Areas 4C, 4D, and 4E were determined by the NPFMC catch sharing plan. As mentioned previously, this CSP allowed Area 4D CDQ to be harvested in Area 4E, and Area 4C IFQ and CDQ to be fi shed in Areas 4C or 4D. These two regulations were the reason the catch in Area 4D exceeded the catch limit. The total commercial catch of 3.855 million pounds was under the combined Area 4CDE catch limit (3.89 million pounds). The IPHC adopted a combined sport and commercial catch limit of 9.0 million pounds for Area 2B, to be allocated to commercial and sport groups by DFO. After the adjustments, the total expanded commercial catch limit was 8.255 million pounds. The Area 2B commercial catch was under the catch limit by 6% percent; it was 4% under in 2007, and within 1% prior to 2007. When the initial halibut IVQ program was implemented in 1991, four hundred and thirty- fi ve vessels received IVQs. Each initial IVQ was split into two shares called blocks. Numerous changes have been made since then, including fi rst allowing temporary block transfers (1993), then permanent block and IVQ transfers (1999), and then implementing an integrated groundfi sh fi shery (2006). With the various changes, the number of vessels catching halibut in Area 2B decreased to 195 vessels in 2008 (Table 7). In 2006, DFO implemented a Groundfi sh Integrated Fisheries Management Plan (Plan) to meet conservation needs, including addressing rockfi sh conservation concerns and improving catch monitoring. This Plan was developed with consultation by the groundfi sh industry and other stakeholders through the Commercial Groundfi sh Integrated Advisory Committee (CGIAC). A pilot program was developed by a sub-committee of the CGIAC and implemented in 2006. With the implementation of this three-year pilot program, signifi cant changes were made to the longline groundfi sh fi sheries, including the halibut fi shery. The pilot fi shery included IQs for all hook and line groundfi sh fi sheries, transferability with limits between license holders, 100% at-sea and dockside monitoring, and vessel accountability for all catch, both landed and discarded. A key component of the Plan was the 100% monitoring through logbook recordings, video camera coverage, and dockside coverage. A newly designed logbook, which allowed the recording of all retained and discarded species, was used to compare to the video recordings. This year (2008) was the fi nal year of the three-year pilot program. However, DFO has decided to continue

8 2009 IPHC ANNUAL MEETING HANDOUT with the Plan for one more year, after which a complete review will be performed. IPHC will be reviewing how the Plan has affected the halibut fl eet dynamics and fi shing patterns. Data are not yet available to report on any changes to fi shing patterns.

Landing patterns The 2008 QS fi shery landings were spread over nine months of the year (Table 8). On a month-to-month comparison, August had the highest landed pounds for Alaska landings, which was a change, as for the last seven years May had received the highest poundage. August landings represented 17% of the total catch for Alaska. Contrary to volumes landed in 2007, April of 2008 was the busiest month for poundage delivered in British Columbia. In 2008, fourteen percent of the Area 2B catch was landed in April compared with 21% during the busiest month (March) last year. Homer received over 9.1 million pounds of halibut, or about 18% of the commercial Alaskan catch. Kodiak and Seward received the second and third largest landing volumes, each moving between 11%-17% of the Alaskan commercial catch. In southeast Alaska, Sitka received 2.8 million pounds, Petersburg 2.1 million pounds, and Juneau 1.9 million pounds. As in previous years, not a high percent (2.3%) of the Alaskan QS catch was landed outside of Alaska. Commercial trips from Area 2B were delivered into 15 different ports in 2008 down from 18 ports in 2007. The ports of Prince Rupert/Port Edward, Port Hardy, and Vancouver were the major landing locations, receiving about 90% of the Area 2B commercial catch. Port Hardy and Prince Rupert/Port Edward received about 40% and 43% of the B.C. commercial landings, respectively. The landing of live halibut from Area 2B was legally allowed by DFO and resulted in a total landing weight of 23,226 pounds. Live fi sh landings have ranged from a low of 7,900 pounds in 1998 to a high of 103,000 pounds in 1999.

The sport fi shery harvest Sport harvest estimates are provided to IPHC by state and federal agencies.

Washington, Oregon, and California harvest The Area 2A sport harvest of 457,151 pounds (Blood 2008) was under the allocation by 3% (Table 9). In Washington, the sport fi shery closed prior to September, which was earlier than in 2007, and the harvests in the primary areas were less than the catch limits. Also, the Washington south coast average weight was lower this year. Washington Department of Fish &Wildlife (WDF&W) will change the 2009 season in an effort to more effectively fi sh those areas. The Oregon sport fi shery closed 6,000 pounds, or 3%, under the catch limit. The Oregon spring fi shery went well into July until anglers turned to albacore fi shing, as they have done in the past. In September, the daily bag limit was increased to a two-fi sh limit which attracted more anglers, however the catch was still under the limit.

British Columbia harvest DFO provided the fi nal 2007 sport harvest estimate of 1.556 million pounds and the preliminary 2008 estimates of 1.536 million pounds (Table 10). The revised 2007 catch exceeded the sport fi shery allocation (1.381 million pounds) by 13%, or 175,000 pounds. In 2008, DFO implemented additional accounting methods, including weekly reporting by lodges, to move towards better in-season estimation procedures, which also allowed for earlier

9 2009 IPHC ANNUAL MEETING HANDOUT current year estimates. The 2008 estimate has been provided to IPHC, however the methodology has not been submitted for review. DFO instituted several management restrictions on the sport fi shery in 2008. First, the opening of the sport season was delayed to March 1 instead of the traditional February 1. Additionally, the bag limit between March 1 and May 30 was reduced to one halibut. It reverted back to a two- fi sh daily bag limit on June 1. The sport season closed on October 31, two months earlier than scheduled. Despite these actions the harvest was 0.5 million pounds over the initial sport allocation. DFO used funds available from previous year’s leasing of sport allocation to the commercial sector to purchase commercial quota share (145,000 pounds) to reduce the sport overage to 309,000, or 25%, in 2008. The estimate for Washington anglers fi shing in Canadian waters and landing in Washington State ports in 2008 has been provided by WDF&W as 4,778 halibut. This is a decrease from 9.977 and 13,045 halibut landed in 2007 and 2006, respectively. There were reports that fi shing was slower off Neah Bay and there were area closures that made it more diffi cult to fi sh in the area.

Alaskan harvest Estimates of the sport fi shery harvest off Alaska are supplied to IPHC by the Alaska Department of Fish and Game (ADF&G). The 2008 projected sport harvest for Alaska is 8.773 million pounds and the fi nal 2007 estimate is 9.4 million pounds, the highest Alaska sport catch on record. The projections for 2008 were made using projection models, the Statewide Harvest Survey (SWHS), and average weights by area based on assumptions of linear trends. As noted earlier, there is a GHL for the charter boat fi shery in Areas 2C and 3A. At the end of 2006, the harvest estimates for the 2006 guided sport charter fi shery in Areas 2C and 3A indicate that the GHL was exceeded in both areas. At the 2007 IPHC Annual Meeting, the Commission, with the support of its advisory bodies, passed a regulation for the 2007 fi shing season to reduce the daily bag limit from two fi sh to one for sport guided charter fi shing in Area 2C from June 15-July 31, and for Area 3A from June 15-30. The U.S government (Departments of State and Commerce) did not adopt this regulation, however, and the Secretary of Commerce assured the Commission that NMFS would implement regulations in 2007 to reduce halibut mortality in Area 2C. Subsequently, NMFS restricted the daily bag limit on sport charter vessels to two halibut per day in Area 2C, with the requirement that at least one of the two fi sh be no more than 32 inches in length. This regulation was effective on June 1, 2007 and remained in effect for the remainder of the season. Additionally, ADF&G prohibited retention of halibut by charter skippers and crew in both Areas 2C and 3A. The goal of the NMFS action was to reduce the harvest to the GHL with minimal impact on the sport charter fi shery, and to achieve a reduction in catch similar to what was anticipated by the Commission’s action. The fi nal estimates for the 2007 guided sport chartered fi shery were over the GHL in Areas 2C and 3A by 34% and 10%, respectively. For 2008, the NPFMC adopted a one-fi sh bag limit for the Area 2C and NMFS implemented regulations which went into effect on June 1. A preliminary court injunction was granted while a judge reviewed the case on June 4, 10, and 20. On June 20, a temporary court injunction was granted and the daily bag limit reverted back to two fi sh provided that one of the two fi sh was no longer than 32 inches. The projection for the 2008 charter fi shery harvest in Areas 2C and 3A was 1.914 and 3.603 million pounds, respectively. The charter sector in Area 2C exceeded the GHL of 0.931 million pounds by over 100%, whereas the Area 3A charter fi shery was under its GHL (3.65 million pounds).

10 2009 IPHC ANNUAL MEETING HANDOUT Bycatch mortality Halibut that are caught in other commercial fi sheries are required to be returned to the sea with minimal injury. Some halibut still die and this fraction is known as bycatch mortality. The discard mortality rate (DMR) is the rate used to determine what fraction of the bycatch dies. The DMR is typically based on the condition of the halibut at release, and the condition is determined by at-sea observers when present. IPHC receives bycatch estimates, size, and release condition data from the observer programs which operate off Canada and the U.S. Not all fi sheries are observed, therefore bycatch rates and DMRs from similar fi sheries are used to calculate bycatch mortality in unobserved fi sheries. In 2008, bycatch mortality was estimated at 10.7 million pounds (Fig. 2), a 12% decrease from 2007, and the lowest since 1986 (Williams 2009). Historically, bycatch mortality had increased to 20 millions pounds in 1992 with the growth and expansion of the Alaska groundfi sh fi sheries. In 2008, the majority of the decrease is due to lower bycatch in the Alaska groundfi sh fi sheries, with an additional decrease in Area 2B trawl fi sheries. In Area 2A, the bycatch estimate is supplied to IPHC by NMFS. The 2007 estimate is 0.298 million pounds, and was rolled over for 2008, as the 2008 estimate is not yet available. The bycatch decreased from 2006 to 2007 by 47%, however the DMRs were generally higher due to a new methodology which uses condition data collected by observers. The trawl effort decreased in depths less than 150 fathoms and in the north coast where the fi shery was closed. This decreased effort accounted for the reduced bycatch mortality which is lower than it has been in the last ten years. Area 2B bycatch mortality of 0.13 million pounds for 2008 is well below the 10-year average of 0.24 million pounds, since the implementation of the Individual Bycatch Quota program. DFO attributes the decrease to lower trawl effort in the summer months and increased off-bottom fi shing for other species. The Area 3 bycatch mortality was 4.3 million pounds, slightly below the 10-year average. Area 4 bycatch was 5.6 million pounds, signifi cantly lower than the last 10-year average (7.1 million pounds). The reduction is due to the new fi shery cooperatives approved by the NPFMC which allowed fl exibility with a slower pace of fi shing. Counter to the trawl reduction there was an increase in bycatch mortality in the hook-and-line cod fi shery. All federally-managed groundfi sh fi sheries conducted off Alaska operate under halibut bycatch mortality limits. The limits are divided by gear type (trawl, longline, pot) and time period. NMFS closes directed fi shing and/or areas when the limits are reached.

References Blood, C. L. 2009. 2008 Sport fi shery. Int. Pac. Halibut Comm. Report of Assessment and Research Activities 2008: 49-58.

Gilory, H. L and Hare, S. R. 2009. Wastage of halibut in the commercial halibut fi shery. Int. Pac. Halibut Comm. Report of Assessment and Reseach Activities 2008: 59-62.

Williams, G.H. 2009. Incidental catch and mortality of Pacifi c halibut, 1962-2008. Int. Pac. Halibut Comm. Report of Assessemnt and Research Activities 2008: 299-312.

Williams, G. H. 2009. The personal use harvest of Pacifi c halibut in 2008. Int. Pac. Halibut Comm. Report of Assessment and Research Activities 2008: 63-66. 11 2009 IPHC ANNUAL MEETING HANDOUT Table 1. The preliminary estimates of 2008 removals of Pacifi c halibut by regulatory area (thousands of pounds, net weight). Area 2A 2B 2C 3A 3B 4 Total Commercial 700 7,721 6,148 24,021 10,762 8,542 57,894 Sport 457 1,536 3,083 5,629 18 43 10,766 Bycatch Mortality: Legal-sized fi sh 1411 67 216 1,058 485 2,259 4,226 Sublegal-sized fi sh 1571 64 128 1,906 853 3,329 6,437 Personal Use2 303 405 525 372 48 1094 1,489 Wastage: Legal-sized fi sh <1 23 12 63 4 38 140 Sublegal-sized fi sh 15 262 212 924 681 243 2,337 IPHC Research 7 73 61 355 131 105 732 Total 1,507 10,151 10,385 34,328 12,982 14,668 84,021 1 Area 2A bycatch is the 2007 estimate as the 2008 estimate will not be available until 2009. 2 Includes 2007 Alaskan subsistence harvest estimates. 3 Treaty Indian ceremonial and subsistence fi sh authorized in the 2008 catch sharing plan. 4 Includes 19,000 pounds of sublegal halibut retained in the 2007 Area 4DE Community Development Quota.

Table 2. The Area 2B Pacifi c halibut catch limits allocated by the Canadian Department of Fisheries and Oceans and the catch estimates (thousand of pounds, net weight), 2007 - 2008. Fishery 2008 Allocation 2008 Catch1 2007 Allocation 2007 Catch Commercial fi shery 7,918 7,721 10,089 9,694 Sport fi shery 1,0822 1,5362 1,381 1,556 Total allocation/ catch 9,000 9,257 11,470 11,250 IPHC research catch 73 78 Previous year carryover3 338 -37 Total 9,338 9,330 11,433 11,328 1 Preliminary 2 Quota shares transfer of 145,000 pounds from the commercial fi shery to the sport fi shery as of December 17. 3 Adjustment for carryover/overage amount from commercial fi shery

12 2009 IPHC ANNUAL MEETING HANDOUT Table 3. The Areas 2C and 3A sport charter halibut harvest and Guideline Harvest Level (thousands of pounds, net weight), 2006 - 2008. Area 2C Area 2C Area 3A Area 3A Year GHL Catch GHL Catch 2006 1,432 1,804 3,650 3,664 2007 1,432 1,918 3,650 4,002 2008 931 1,914 3,650 3,603

Table 4. The Area 2A 2008 catch limits allocated by the Pacifi c Fishery Management Council catch sharing plan and preliminary catch estimates (pounds, net weight). Area and Fishery Catch Limit Catch Non-treaty directed commercial 213,674 220,590 Non-treaty incidental commercial with salmon troll fi shery 37,707 16,685 Non-treaty incidental commercial with sablefi sh fi shery 70,000 34,961

Treaty Indian commercial 397,000 426,879 Treaty Indian ceremonial and subsistence 30,000 30,000

Sport - North of Columbia River 220,238 216,297 Sport - South of Columbia River 251,381 240,853

Total 1,220,000 1,186,265 IPHC research catch 7,000 Total 1,220,000 1,193,265

13 2009 IPHC ANNUAL MEETING HANDOUT Table 5. Commercial fi shing periods, number of fi shing days, catch limit, commercial, research and total catch (thousands of pounds, net weight) by regulatory area for the 2008 Pacifi c halibut commercial fi shery (preliminary, landings as of November 26, 2008). Catch No. of Commercial Research Total Area 2A Fishing Period Limit Days Catch Catch Catch treaty Indian March 8 – June 3 88 326.0 Restricted: Mar 17 – Apr 15 30 101.0 total 397.0 427.0 0 427.0 Commercial Incidental in May 1 – Nov 15 37.7 199 16.7 0 16.7 Salmon fi shery

Incidental in May 1- Oct 31 70.0 184 35.0 0 35.0 Sablefi sh fi shery Directed June 111 213.7 10-hours 68.0 7 June 251 “ 73.0 July 91 “ 57.0 July 231 “ 23.0 total 221.0 228.0 2A Total 718.4 699.7 7 706.7 Catch Adjusted Commercial Research Total Area Fishing Period Limit Catch Limit2 Catch Catch Catch 2B 3/08 – 11/15 7,918.0 8,255 7,721.03 73 7,794.0 2C 3/08 – 11/15 6,210.0 6,432 6,148.04 61 6,209.0 3A 3/08 – 11/15 24,220.0 24,415 24,021.0 355 24,376.0 3B 3/08 – 11/15 10,900.0 10,995 10,762.0 131 10,893.0 4A 3/08 – 11/15 3,100.0 3,170 2,962.0 49 3,011.0 4B 3/08 – 11/15 1,860.0 1,886 1,725.0 40 1,765.0 4C 3/08 – 11/15 1,769.0 1,825 723.05 0 723.0 4D 3/08 – 11/15 1769.0 1,809 2,544.05, 6 16 2,560.0 4E 3/08 – 11/15 352.0 352 588.06 0 588.0 Alaska Total 50,180.0 50,880 49,473.0 652 50,125.0 Grand Total 58,816.4 59,139 57,893.7 732 58,625.7 1 Fishing period limits by vessel class. 2 Includes adjustments from the underage and overage programs 3 Includes the pounds that were landed by Native communal commercial licenses (FL licenses). 4 Includes pounds taken by Metlakatla Indians during additional fi shing within reservation waters. 5 Area 4C IFQ and CDQ can be fi shing in Area 4D by NMFS and IPHC regulations 6 Area 4D CDQ can be fi shed and 4E by NMFS and IPHC regulations

14 2009 IPHC ANNUAL MEETING HANDOUT Table 6. The fi shing period limits (net weight) by vessel class used in the 2008 directed commercial halibut fi shery in Area 2A. Vessel Class Fishing Periods (pounds) Letter Feet June 11 June 25 July 9 July 23 A 0-25 755 755 670 200 B 26-30 945 945 840 200 C 31-35 1,510 1,510 1,345 200 D 36-40 4,165 4,165 3,705 560 E 42-45 4,480 4,480 3,985 600 F 46-50 5,365 5,365 4,770 715 G 51-55 5,985 5,985 5,320 800 H 56+ 9,000 9,000 8,000 1,200

Table 7. The number of vessels catching halibut since the implementation of the Area 2B IVQ fi shery and number of vessels and permit holders catching halibut since the implementation of the Alaska IFQ. Number of vessels Number of vessels Number of quota share Year fi shing in Area 2B fi shing in Alaska holders in Alaska 1991 439 no IFQ fi shery no IFQ fi shery 1992 433 no IFQ fi shery no IFQ fi shery 1993 355 no IFQ fi shery no IFQ fi shery 1994 318 no IFQ fi shery no IFQ fi shery 1995 295 2,206 4,830 1996 278 2,130 NA 1997 284 2,163 NA 1998 287 1,802 NA 1999 268 1,847 NA 2000 238 1,841 3,541 2001 239 1,728 3,507 2002 216 1,643 3,500 2003 225 1,592 3,435 2004 219 1,513 3,315 2005 222 1,495 3,239 2006 214 1,476 3,210 2007 210 1,499 3,076 2008 195 1,401 2,907

15 2009 IPHC ANNUAL MEETING HANDOUT Table 8. The total pounds (thousands, net weight, preliminary) of 2008 commercial qutoa share landings of Pacifi c halibut by regulatory area and month. Regulatory Area March April May June July Aug. Sept. Oct. Nov. Total 2B1 1,077 1,098 995 556 895 993 1,048 601 458 7,721 2C2 937 736 962 946 586 827 774 228 152 6,148 3A 2,633 3,483 3,978 3,118 2,411 3,029 2,750 1,996 623 24,021 3B 117 488 1,569 2,161 2,038 2,061 990 849 489 10,762 4A 2203 367 643 789 482 4614 2,962 4B 633 153 307 394 415 208 1854 1,725 4CDE 673 955 1,298 868 61 3,855 Alaska Total 3,687 4,770 6,882 7,572 7,027 8,419 6,072 3,780 1,264 49,473 Total 4,764 5,868 7,877 8,128 7,922 9,412 7,120 4,381 1,722 57,194 1 Based on landing ratios from DFO website. 2 Area 2C included Metlakatla Fishery landings 3 Weight combined with previous month for confi dentiality purposes. 4 Weight combined with following month for confi dentiality purposes.

Table 9. The Area 2A 2008 halibut sport catch limits and catch estimates (in pounds, net weight) by subarea. Subarea Catch limit Harvest estimate Over/under WA Inside Waters 59,354 59,3541 0 WA North Coast 109,991 106,852 -3,139 WA South Coast 44,700 40,398 -4,302 Columbia River 18,762 17,899 -863 OR Cent. Coast (spring, May-June) 159,5772 119,656 -39,921 OR Cent. Coast (summer, Aug-Sept) 53,192 93,618 +40,426 OR Coast (<40 fathoms) 18,502 11,833 -6,669 OR/CA (south of Humbug Mt.) 7,541 7,5411 0 Total 471,619 457,151 -14,468 1 An estimate is not yet available, so the catch limit is assumed to have been taken. 2 Oregon’s Central Coast spring all-depth fi shery underage of 39,921 pounds was transferred to the summer all-depth fi shery, increasing that quota to 93,113 pounds.

16 2009 IPHC ANNUAL MEETING HANDOUT Table 10. Halibut sport catch (millions of pounds, net weight) by regulatory area, 1995-2008. Year Area 2A Area 2B Area 2C Area 3A Area 3B Area 4 Alaska Total 1995 0.236 0.887 1.759 4.511 0.022 0.055 6.347 7.470 1996 0.229 0.887 2.129 4.740 0.021 0.077 6.967 8.084 1997 0.355 0.887 2.172 5.514 0.028 0.069 7.783 9.025 1998 0.383 0.887 2.501 4.702 0.017 0.096 7.316 8.585 1999 0.338 0.859 1.843 4.228 0.017 0.094 6.182 7.379 2000 0.344 1.021 2.258 5.305 0.015 0.073 7.651 9.017 2001 0.446 1.015 1.925 4.675 0.016 0.029 6.645 8.106 2002 0.399 1.260 2.090 4.202 0.013 0.048 6.353 8.011 2003 0.404 1.218 2.258 5.427 0.009 0.031 7.725 9.348 2004 0.487 1.613 2.937 5.606 0.007 0.053 8.603 10.703 2005 0.484 1.841 2.798 5.672 0.014 0.050 8.534 10.860 2006 0.516 1.773 2.526 5.337 0.014 0.046 7.923 10.212 2007 0.504 1.556 3.049 6.283 0.025 0.044 9.401 11.461 20081 0.457 1.536 3.083 5.629 0.018 0.043 8.773 10.766 1 Only Area 2A and 2B are current; all other areas are projected harvests.

Table 11. Halibut sport catch (millions of pounds, net weight) for charter and private anglers, and the Guideline Harvest Level (GHL) for Areas 2C and 3A, 2003 -2008. Area 2C Area 3A Year GHL Charter Private Total1 GHL Charter Private Total1 2003 1.432 1.412 0.846 2.258 3.650 3.382 2.046 5.427 2004 1.432 1.750 1.187 2.937 3.650 3.668 1.937 5.606 2005 1.432 1.952 0.845 2.798 3.650 3.689 1.984 5.672 2006 1.432 1.804 0.722 2.526 3.650 3.664 1.674 5.338 2007 1.432 1.918 1.131 3.049 3.650 4.002 2.281 6.283 2008 0.932 1.914 1.169 3.083 3.650 3.603 2.026 5.629 1 Discrepancies in totals due to rounding

17 2009 IPHC ANNUAL MEETING HANDOUT Table 12. Estimates (thousands of pounds, net weight) of bycatch mortality of halibut by year and area, 1999 - 2008. Area 1999 2000 2001 2002 2003 2004 2005 2006 2007 20081 Area 2A 987 822 837 553 503 302 459 387 298 298 Area 2B 193 230 177 244 244 251 346 294 319 131 Area 2C 358 395 341 340 341 362 340 341 342 344 Area 3A 2,885 2,892 3,009 2,194 3,180 3,671 3,220 2,975 2,843 2,964 Area 3B 1,737 1,510 1,675 1,924 1,734 1,274 1,126 1,400 1,115 1,338 Area 4 7,684 7,441 7,120 7,273 6,822 6,735 7,692 7,491 7,262 5,588 TOTAL 13,844 13,290 13,159 12,528 12,824 12,595 13,183 12,888 12,179 10,663 1 Preliminary

18 2009 IPHC ANNUAL MEETING HANDOUT 170°E 180° 170°W 160°W 150°W 140°W 130°W 120°W 65°N 65°N Russia Alaska

60°N Bering Sea 60°N 4E 4D 4C 3A Kodiak Is. Closed B r 2C it 55°N 3B is 55°N 4A h 4B C Queen Charlotte Is. o Aleutian Is. lu 4A m b Gulf of Alaska 2B i 50°N a 50°N 4B Vancouver Is.

2A 45°N 45°N

170°E 180° 170°W 160°W 150°W 140°W 130°W 120°W

Figure 1. IPHC regulatory areas for the 2008 fi shery.

25 Area 4 Area 3 20 Area 2

15

10

Bycatch Mortality (Mlbs).... Mortality Bycatch 5

0 1962 1966 1970 1974 1978 1982 1986 1990 1994 1998 2002 2006 Year

Figure 2. Bycatch mortality of Pacifi c halibut by IPHC regulatory area, 1962-2008.

19 2009 IPHC ANNUAL MEETING HANDOUT 20 2009 IPHC ANNUAL MEETING HANDOUT Analysis of PIT tag recoveries through 2008

Raymond A. Webster

Abstract Patterns of tag-recoveries in 2008 were similar to 2007, with continued low recoveries in western areas and in Area 2B. A model was fi tted this year that allowed for different migration rates for small and large fi sh, and this provided a better fi t than one which assumed migration rates are the same for all fi sh. Patterns of recoveries and estimates of migration rates continue to show strong eastward movement of fi sh from Areas 4A and 3B, but less clear movement patterns from other areas.

Introduction In 2003 the International Pacifi c Halibut Commission (IPHC) staff marked with PIT tags and released all fi sh caught on three skates of gear at all setline survey stations coastwide, totaling almost 44,000 fi sh (Kaimmer and Geernaert 2004). The release was repeated in 2004 in Areas 2B and 3A, totaling another 23,000 fi sh (Williams et al. 2005). In each year from 2003 to 2008, samplers in the ports scanned a substantial part of the landings to recover tags (Forsberg 2009). The primary purpose of this large project is to estimate the harvest rate of fully selected halibut by the commercial fi shery, but the tag-recovery data also permit estimates of length- specifi c selectivity schedules, rates of migration among regulatory areas, and, in principle, the rate of natural mortality. This report updates the analyses of the PIT tag recovery data to include recoveries in 2008. Some alternative models were considered this year, including a model that allowed migration to depend on fi sh length. Some results presented here differ from those shown in the 2008 Report of Assessment and Research Activities (RARA) (Webster 2009) due to minor coding corrections.

Data analysis PIT tag recoveries for 2008 are discussed in detail in Forsberg (2009). In total, 422 tags were recovered, with 261 from the 2003 releases and 161 from the 2004 releases (Table 1). Table 2 shows recoveries of tagged fi sh from the 2003 and 2004 releases in 2008 for comparison with the corresponding tables from previous years (Clark 2006, Webster 2008). The overall pattern of recoveries is similar to 2007. In particular, very few Area 4A tags are now recovered in Area 4A, largely because the releases have now migrated eastward. Total Area 2B recoveries were again low compared to years prior to 2007, and are now comparable to recoveries of Area 4D tags, an area with similar scanning rates (Forsberg, 2009) but far fewer releases. The data are analysed by fi tting tag-recovery models. Natural mortality is fi xed at a rate of 0.15, tag-loss is taken to be 3% per year, and based on previous tag seeding work, we assume 97% of scanned tags are detected. Other sources of fi shing mortality – sport catch, wastage, personal use, and bycatch – are accounted for as components of total fi shing mortality in the models. We fi tted a sequence of plausible models to the recovery data from Areas 2B-4A. Data from other areas were excluded because of their sparse nature. The two best fi tting models we

21 2009 IPHC ANNUAL MEETING HANDOUT considered both included two migration matrices. In one model, also fi tted last year (Webster 2008), migration differed between the year following tag-release and subsequent years. The other model had different migration rates for small fi sh (< 90cm) and large fi sh (> 90cm). The estimates we present are for the model with different migration rates for fi sh less than and over 90 cm in length (Fig. 1). The migration estimates for Area 4A in particular differ greatly between the size groups, with much greater migration from Area 4A to Area 3A estimated for smaller fi sh, although the estimates for Area 4A fi sh under 90 cm are very imprecise due to sparse data. The estimate of the migration rate from Area 3B to Area 3A is also higher for smaller fi sh, but estimates for other areas are similar across the two groups. Although we estimate that a smaller proportion of fi sh larger than 90 cm migrate each year, estimated rates of up to 10% per area confi rm that migration is ongoing for larger fi sh. This was also apparent in the raw data, with rates of tag recoveries out of the area of release showing movement of tagged fi sh across all release length classes (Table 3).

Net migration The tag-recovery modelling leads to estimates of annual rates of emigration from each regulatory area. We can estimate net annual migration by applying the migration estimates from the model to estimates of the number of legal-sized fi sh. Using population estimates from the 2008 stock assessment and migration rates estimated under a model with a single migration matrix (i.e., migration not a function of length or time), we obtain the net migration rates shown in Table 4. We estimate strong net eastward migration from Area 4A , at a rate of 15% per year, while Area 2B receives the greatest percentage of migrants relative to its population, with an estimate of around 7% annual inward migration. Other net migration estimates are smaller, with values refl ecting a combination of emigration and immigration. These estimated net migration rates apply to legal-sized fi sh only. The analysis described above showed that migration of smaller fi sh is estimated to be greater than larger fi sh, and we also expect higher rates of migration for fi sh too small to have been captured and tagged as part of this study. This, combined with fi sh becoming part of the legal-sized population due to growth, means regulatory areas with net outward migration of legal-sized fi sh can still maintain stable population levels.

Discussion Over the course of this study, recoveries of tagged fi sh out of release area have shown that migration is an ongoing process not restricted to smaller, younger fi sh. This is supported by the results of our tag-recovery modelling, which show continued migration of fi sh greater than 90 cm in length at estimated rates of up to 10% per year for each regulatory area. Rates of migration are estimated to be even greater for smaller fi sh in western areas. The recovery data are too sparse for most regulatory areas to permit a more detailed look at the relationship between halibut size and migration probability, but the raw data imply this relationship varies greatly among areas. As in 2007, the best fi tting model of those we considered was one which showed little or no movement of fi sh out of areas in the eastern Gulf of Alaska after the fi rst year following releases of the tags. The 2008 data continue to show support for the possibility that while migration is an important and ongoing process eastward from Areas 4A and 3B, there may be little or no annual emigration out of areas to the east (see Webster 2008 for details). Note that values shown in Figure

22 2009 IPHC ANNUAL MEETING HANDOUT 1 are essentially averages over the course of the study, and do not account for possible differences over time in migration rates. Recovery data are too sparse to fi t a model which includes both size and year effects on migration rates. Clark (2006) also estimated rates of net migration, but used model predictions of the 2004 tag distribution as measures of relative population sizes. This approach leads to a much higher estimate of net migration into Area 2B, largely due to Area 2B releases (and hence 2004 predicted numbers) accounting for a smaller proportion of the coastwide total than corresponding estimates of population size from the coastwide stock assessment. Presently, estimates of net migration are calculated in terms of percentages of numbers of fi sh in the population. In future work, we will also devise estimates of migration rates in terms of the amount of biomass moving among regulatory areas in a year.

Reference Clark, W. G. 2006. Analysis of PIT tag recoveries through 2005. Int. Pac. Halibut Comm. Report of Assessment and Research Activities 2005: 123-134.

Hare, S. R and Clark, W. G. 2009 Assessment of the Pacifi c halibut stock at the end of 2008. Int. Pac. Halibut Comm. Report of Assessment and Research Activities 2008: 137-202.

Forsberg, J. E. 2009. Portside and survey vessel sampling for recovered PIT tags in Pacifi c halibut. Int. Pac. Halibut Comm. Report of Assessment and Research Activities 2008: 403-430.

Kaimmer, S. M., and Geernaert, T. O. 2004. 2003 PIT tagging: tagging equipment and protocol, and tag shedding studies. Int. Pac. Halibut Comm. Report of Assessment and Research Activities 2003: 351-360.

Webster, R. A. 2008. Analysis of PIT tag recoveries through 2007. Int. Pac. Halibut Comm. Report of Assessment and Research Activities 2007: 259-273.

Williams, G. H., Geernaert, T. O., and Chen, D. 2005. PIT tagging in 2004. Int. Pac. Halibut Comm. Report of Assessment and Research Activities 2004: 345-350.

23 2009 IPHC ANNUAL MEETING HANDOUT Table 1. Summary of PIT tag recoveries by release and recovery year. Year of Year of recovery release 2003 2004 2005 2006 2007 2008 Total 2003 86 383 463 429 293 261 1,915 2004 0 45 297 251 193 161 947 Total 86 428 760 680 486 422 2,862

Table 2a. Recoveries in 2008 of fi sh released coastwide in 2003. Area of Releases Recoveries by area in 2008 release in 2003 4D 4C 4B 4A 3B 3A 2C 2B 2A Total 4D 9791100200000 13 4B 1,128000002100 3 4A 3,4520003213221 23 3B 15,28200004341611 92 3A 16,4130000475120 82 2C 3,6750000012610 28 2B 2,7170000001140 15 2A 299000000032 5 Total 43,945 11 0 0 5 49 132 37 23 4 261

Table 2b. Recoveries in 2008 of fi sh released in Areas 2B and 3A in 2004. Area of Releases Recoveries by area in 2008 release in 2003 4D 4C 4B 4A 3B 3A 2C 2B 2A Total 3A 20,34100008136300147 2B 3,0850000010121 14 Total 23,426000081373121161

24 2009 IPHC ANNUAL MEETING HANDOUT Table 3. Proportion of PIT tag recoveries made outside of the release area, by release area and length at release. Release Area Length at release 4D 4A 3B 3A 2C 2B Coastwide <81 cm 0.00* 0.79 0.36 0.19 0.28 0.02 0.27 81-100 cm 0.06 0.69 0.37 0.10 0.11 0.07 0.17 101-120 cm 0.23 0.67 0.39 0.08 0.11 0.10 0.16 >120 cm 0.25 0.25* 0.47 0.07 0.12 0.10 0.11 * Fewer than 10 recoveries

Table 4. Estimates of net annual migration rates (%) for legal-sized halibut. Area Emigration1 Immigration2 Net migration 4A 18.4 3.2 -15.2 3B 11.5 8.6 -2.8 3A 5.8 8.6 2.8 2C 10.2 5.6 -4.6 2B 1.7 8.3 6.6 1Estimates of mean annual emigration rates from tag-recovery modelling for 2004-08. 2Based on legal-sized population numbers for January 1, 2008 estimated using the coastwide stock assessment.

25 2009 IPHC ANNUAL MEETING HANDOUT 4A, length<90cm 4A, length>=90cm

90.1 (3)

49.3 (15.7) 37.3 (12.8)

7.9 (6.2) 1.1 (1.5) 4.4 (2.5) 3.1 (2.1) 2.9 (1.3) 1.7 (0.6) 2.2 (1.2) migration rate (%) migration rate (%) 04080 04080

4A 3B 3A 2C 2B 4A 3B 3A 2C 2B 3B, length<90cm 3B, length>=90cm

85 (2.5) 91 (1.2)

13.4 (2.4) 8.2 (1.1) 0.2 (0.1) 0.9 (0.3) 0.4 (0.2) 0 (0) 0.4 (0.1) 0.4 (0.3) migration rate (%) migration rate (%) 04080 04080

4A 3B 3A 2C 2B 4A 3B 3A 2C 2B 3A, length<90cm 3A, length>=90cm

95.4 (0.9) 93.2 (1.2)

0 (0.1) 4.1 (0.9) 0.2 (0.1) 0.3 (0.1) 0 (0) 5.9 (1.2) 0.4 (0.1) 0.5 (0.1) migration rate (%) migration rate (%) 04080 04080

4A 3B 3A 2C 2B 4A 3B 3A 2C 2B 2C, length<90cm 2C, length>=90cm

88.9 (2.3) 90.8 (2.8)

8 (2) 0 (0) 0 (0) 3.1 (1.6) 0 (0) 0 (0) 2.8 (1.2) 6.4 (2.6) migration rate (%) migration rate (%) 04080 04080

4A 3B 3A 2C 2B 4A 3B 3A 2C 2B 2B, length<90cm 2B, length>=90cm 99.4 (0.6) 98.1 (0.6)

0.3 (0.2) 0 (0) 0 (0.3) 0.3 (0.5) 0 (0) 0 (0) 0.5 (0.4) 1.4 (0.5) migration rate (%) migration rate (%) 04080 04080

4A 3B 3A 2C 2B 4A 3B 3A 2C 2B

Figure 1. Estimated annual rates of migration from each regulatory area (rows) for fi sh with length under 90 cm, and for those with length greater than or equal to 90 cm.

26 2009 IPHC ANNUAL MEETING HANDOUT Removal fi shing to estimate catch probability: preliminary data analysis

Raymond A. Webster

Abstract This project examined whether or not removal sampling was a useful technique for estimating catch probability and modeling the factors affecting catchability of Pacifi c halibut. Sampling was done at 20 setline survey stations in the eastern part of Area 3A, and at each station a sequence of fi ve sets was made on consecutive days. The expectation was that catch would decrease over the fi ve days of fi shing, and the rate of decrease would provide information on catch probability. We anticipated some movement into and out of the catchable population during the fi ve days, and developed models that allowed for some degree of local migration. Little, if any, decline in catch was observed, and the results of preliminary modeling imply that there was such large movement into the catchable population following removals that useful estimates of catch probability are impossible to obtain. Our conclusion is that removal sampling is not a promising approach for modeling the catchability of Pacifi c halibut.

Introduction Catch per unit effort (CPUE) computed from the IPHC setline survey data is used as the basis of an index of population size or biomass in each regulatory area. However, CPUE is a function not only of the number of halibut, but also of their probability of capture. The capture probability may vary according to many factors, such as environmental covariates (depth, temperature), individual covariates (sex, maturity, size, prior injuries), and design variables (location of set, time of day or year, length of soak). Failure to account for potential variation in capture probability means CPUE can be a misleading index of abundance. Removal sampling of closed populations is a long established technique for estimating population size while accounting for catch probability. When applied to fi sh sampling, the same population is repeatedly sampled over multiple occasions in quick succession. For halibut on the IPHC setline survey, this would require several consecutive sets to be made at a survey station. The basic idea is that the catch at a station will decline on each successive set as more of the local population is removed, and modeling the rate of decline will allow us to estimate the number of fi sh that were present prior to the fi rst set. Standard removal models are equivalent to the behavioural response model of closed population capture-recapture modeling (Pollock 1990), but with the practical advantage that animals do not need to be tagged and recaptured on multiple occasions. Depending on local abundance, catch probability, and the number of consecutive sets, information at each individual station may be poor. However, when data from multiple stations are combined using hierarchical modeling, it is possible to obtain good estimates of relative abundance, capture probability, and their relationships with measured covariates for the whole regulatory area (Forsyth et al. 2005, Webster et al. 2008). Prior to the study, concern was raised about movement into and out of the catchable population (local migration) during removal sampling, violating the standard model assumption of population

27 2009 IPHC ANNUAL MEETING HANDOUT closure. To accommodate the possibility of some movement, we modifi ed the usual migration models to allow for a fi xed daily rate of migration. Simulation studies showed that it was possible to estimate both the rate of migration and catch probability with little bias provided that migration was low to moderate and catch probability was not too low. We considered migration rates of r=0.5 or less, where r=1 means the migrants replace removed fi sh on average each day, and r=0 means there is no migration. Only when an animal’s daily catch probability, p, was low (0.25 in our simulation study) did bias become a problem and, even then, estimators were great improvements over standard removal estimators. In this study, we trialed removal sampling as a technique for directly estimating catch probability at a sample of setline survey stations in one regulatory area in 2008. For the selected stations, a sequence of multiple consecutive sets was made. The fi rst set was the standard survey set. Subsequent sets used the same methods and collected the same data on halibut and station conditions as standard survey sets, with the exception that otoliths were only collected during the fi rst day’s set for the setline survey. To maximize the number of stations we could include in the study, stations were selected in clusters of four (Figs. 1 and 2), allowing four stations to be fi shed consecutively within a single day. Five clusters were selected for the study, three in the Yakutat setline survey region, and two in the Prince William Sound survey region. The clusters were selected by generating all possible designs with the stated number of clusters of four stations, discarding any that were considered operationally impractical, and selecting a design at random from the remaining designs. During the course of the data collection, one cluster was replaced due to the presence of a commercial fi shing vessel on one of the stations of the original cluster (Fig. 1). On days 1-4, all halibut offal and obviously dead or mortally wounded bycatch were retained and not dumped near the survey stations. This was so that offal and dumped bycatch did not attract fi sh from outside the area around the removal fi shing locations. Following the fi rst day (the survey set), only legal-sized fi sh were sampled, while sublegals were measured and released. Soak times and station order were maintained within standard survey limits. While there were predetermined limits for disruptions (e.g., poor weather) no such disruptions occurred during the removal sampling.

Halibut catch Table 1 and Figure 3 summarize catch of legal-sized halibut by cluster and fi shing day, while Table 2 gives mean length of legal-sized fi sh. Overall, catch was greatest on the second day, and declined on average thereafter, with lowest catch on day 5. Changes in catch over the fi ve days varied among the fi ve removal clusters: cluster 1 had no obvious trend in catch; cluster 2 had highest catch during the fi rst two days; clusters 3 and 4 both caught more legal halibut on the second and third days; and cluster 5 had highest catch on days 4 and 5. There was no clear trend in mean fi sh length over the fi ve days, with average length over all clusters remaining stable. Catch of sublegal fi sh was highly variable across days and among clusters. Like legal halibut, the peak in catch was on the second day, largely due to high catches on cluster 3 stations. However, catch of sublegals increased in the last two days on clusters 4 and 5, leading to total catches for those days that were little different from day 2. We should note that because sublegals were released, these data may include some animals previously captured, making interpretation of any patterns more diffi cult.

28 2009 IPHC ANNUAL MEETING HANDOUT Bycatch and returned bait Increases in competition by other species for bait can lead to decreases in catch probability for halibut as less bait becomes available for the target species. High levels of bycatch can therefore depress catches of halibut, while low amounts of returned bait can mean that fewer halibut had the opportunity to be caught. Table 4 shows the average percentage of hooks with bycatch for each cluster, by fi shing day, based on counts from the fi rst 20 hooks of each skate. On all clusters, bycatch peaked on the fourth day, with an overall mean of 29%, and was lowest on average on the fi nal day, at 25%. As might be expected, the reverse was true for returned baits, with fewest on the fourth day (8% on average) and the most on the fi nal day (13% on average).

Preliminary modeling The hope was that we could use removal sampling to give us estimates of catch probability, and model its relationship with variables such as depth, weather conditions, tidal fl ux, and halibut size and sex. Successful modeling depended on observing a declining catch and on the rate of migration not being too high. As the rate of migration approaches 1, it becomes harder to distinguish high catchability and low local abundance from low catchabilty and high local abundance. The daily catches of legal-sized halibut (Table 1) showed little average decline over the fi ve days, with some clusters showing no decline at all. As a fi rst attempt at estimating catch probability, we fi tted a simple removal with migration model to the total counts from Table 1. The migration rate for this model was estimated as rˆ = 0.96 (95% CI: 0.71-1.00), almost at the daily replacement rate of 1, and estimates of daily migrants ranged from 351 between the second and third daily sets and 318 after the fourth set. With such large daily movement of animals into the catchable population, useful estimation of catch probability is almost impossible, and the estimate of catch probability was pˆ = 0.53, with a very wide 95% CI of 0.11-0.99. We have also attempted to fi t more complex models that include relationships between catch probability and individual station covariates using Bayesian hierarchical modelling. The Markov chain Monte Carlo algorithms we used to fi t such models do not converge, and estimation of model parameters has been unsuccessful to date. This is not very surprising given that for many if not most stations, there was no apparent decline in legal catch over the fi ve fi shing days, and for some stations there was even an increase in catch. Modelling of these data is ongoing, but given the observed data, we are not optimistic that we will obtain useful results.

Discussion Because of apparently very high rates of movement of fi sh into the catchable population around the survey station, removal sampling does not look like a promising method for estimation of catch probability of halibut or for exploration of the relationships of catchability with station or individual halibut level covariates. We anticipated some movement towards the survey lines between fi shing days, either because of the attractive effect of baited lines, the replacement of depleted habitat with new fi sh, or the movement of groups of fi sh passing through the area around a station. However, rates of movement of the order of that estimated in our preliminary model fi tting makes useful inference on catchability almost impossible. The models we have tried to fi t to date assume catch probability of an individual fi sh is constant with time. There are a number of reasons why this might not be true. One suggestion is

29 2009 IPHC ANNUAL MEETING HANDOUT that the size distribution of catchable fi sh could change over the fi ve days, possibly due to larger, more catchable fi sh being caught fi rst. If this were the case, it might be expected that catches would be more likely to decline than maintain the same level on average, even in the presence of high migration. In any case, mean lengths remained very stable during the fi ve fi shing days, and changing fi sh size does not seem to be a factor. Changes in bycatch can also affect catchability: as competition with halibut for available baits increases, an individual halibut’s chance of being caught decreases. Thus stable halibut catches could be explained by declining bycatch over the fi ve days (and similarly by increased bait return) rather than a migration rate close to 1. However, except for a drop on the fi nal day in all fi ve clusters, bycatch either remained stable or increased over the removal period. Another suggestion was that changes in local abundance could be due to tidal fl uctuations, specifi cally, that abundance increases during building tides and decreases during declining tides. We have yet to look closely at tidal data during the removal fi shing period, but information from staff on the vessel conducting the study was that fi shing was carried out over both building and declining tides.

References Forsyth, D. M., Link, W. A., Webster, R., Nugent, G. and Warburton, B. 2005. Nonlinearity and seasonal bias in an index of brushtail possum abundance. Journal of Wildlife Management, 69: 976-981.

Pollock, K. H., Nichols, J. D., Brownie, C. and Hines, J. E. 1990. Statistical Inference for Capture- Recapture Experiments. Wildlife Monographs, 107.

Webster, R. A., Pollock, K. H., Ghosh, S. K. and Hankin, D. G. 2008. Bayesian spatial modeling of data from unit-count surveys of fi sh in streams. Transactions of the American Fisheries Society, 137: 438-453.

30 2009 IPHC ANNUAL MEETING HANDOUT Table 1. Legal-sized catch by cluster, sex, and removal fi shing day. Totals for each cluster include a small number of fi sh of unknown sex. Cluster Day 1 Day 2 Day 3 Day 4 Day 5 Female 54 50 54 54 55 1 Male 45725 Total 58 55 61 56 60 Female 53 52 24 33 35 2 Male 54342 Total 59 59 28 43 37 Female 128 160 147 139 112 3 Male 11 16 17 14 13 Total 142 178 165 153 126 Female 18 32 24 22 18 4 Male 11 9 7 13 17 Total 29 42 31 35 35 Female 28 37 48 50 33 5 Male 5 8 7 10 10 Total 33 45 55 60 43 Total 321 379 340 347 301

Table 2. Average length (cm) of legal-sized catch by cluster, sex, and removal fi shing day. Means for each cluster include a small number of fi sh of unknown sex. Cluster Day 1 Day 2 Day 3 Day 4 Day 5 Female 104 106 105 100 108 1 Male 90 101 85 90 92 All 103 106 104 96 106 Female 104 107 107 101 103 2 Male 88 94 87 92 89 All 103 105 105 99 100 Female 105 100 106 105 102 3 Male 89 87 91 85 93 All 103 99 105 100 101 Female 103 103 106 91 106 4 Male 84 87 92 88 92 All 102 99 104 90 103 Female 110 103 108 93 97 5 Male 94 86 87 88 90 All 108 102 105 90 95 All 104 103 103 102 102

31 2009 IPHC ANNUAL MEETING HANDOUT Table 3. Catch of sublegal halibut by cluster and fi shing day. Cluster Day 1 Day 2 Day 3 Day 4 Day 5 1 3522334022 2 5150234229 3 127 180 144 106 104 4 33 53 54 105 129 5 1419172135 All 260 324 271 314 319

Table 4. Percentage of hooks catching other species (bycatch) by cluster and fi shing day. Data come from the fi rst 20 hooks of each skate of each set. Cluster Day 1 Day 2 Day 3 Day 4 Day 5 1 1821243423 2 3638374135 3 1816161812 4 2624232726 5 2324252721 All 24 24 25 29 23

Table 5. Percentage of returned hooks with baits by cluster and fi shing day. Data come from the fi rst 20 hooks of each skate of each set. Cluster Day 1 Day 2 Day 3 Day 4 Day 5 1 3831251528 2 78336 3 1414191318 4 85422 5 135413 All 12 12 11 8 13

32 2009 IPHC ANNUAL MEETING HANDOUT 145°W 144°W 143°W 142°W 141°W 140°W

4101 4100 60°N (! (! 60°N

40994098 4097 4096 4095 4094 4093 4092 (! " " S S (! (! (!

4091 40904089 4088 4087 4086 4085 40844083 4082 4081 4080 4079 (! (! " " S S (! (! *# *# (! (! (!

4078 4077 4076 4075 4074 4073 4072 *# *# (! (! (! (! (!

4071 4070 4069 4068 4067 4066 4065 4064 (! (! (! (! (! (! (! (!

4063 4062 4061 4060 4059 4058 4057 (! (! (! (! (! (! (!

4056 4055 4054 4053 59°N (! (! (! (! 59°N

4052 4051 Legend (! (!

*# Cluster 1 (! Cluster 2 S Cluster 3 not fished " Cluster 3 fished

145°W 144°W 143°W 142°W 141°W 140°W

Figure 1. Survey stations in the Yakutat region selected for the removal sampling pilot study.

33 2009 IPHC ANNUAL MEETING HANDOUT 148°W 147°W 146°W 145°W

61°N 61°N

4146 4145 (! (!

4144 4143 (! (!

4141 4140 4139 4138 4142 (! (! (! (! (!

4137 (!

4136 4135 4134 4133 4132 (! (! (! (! (!

4131 4130 4129 4128 4127 4126 4125 60°N ^ ^ (! (! (! (! (! 60°N

4124 4123 4122 4121 4120 4119 ^ ^ (! (! (! (!

4118 4117 4116 4115 4114 4113 4112 (! (! (! (! (! (! (!

4111 4110 4109 4108 4107 4106 (! (! % % (! (!

Legend 4105 4104 4103 4102 (! % % (! % Cluster 4 ^ Cluster 5

148°W 147°W 146°W 145°W Figure 2. Survey stations in the Prince William Sound region selected for the removal sampling pilot study.

34 2009 IPHC ANNUAL MEETING HANDOUT ●

● ●

●

●

● total ● cluster 1 ● cluster 2 cluster 3 cluster 4 cluster 5 sized catch (count) − legal

● ●●● ● ● ● ● ● ● 0 100 200 300 400

12345

day

Figure 3. Catch of legal-sized halibut by cluster and fi shing day.

35 2009 IPHC ANNUAL MEETING HANDOUT 36 2009 IPHC ANNUAL MEETING HANDOUT Exploring effects of fi shing and migration on the distribution of Pacifi c halibut

Juan L. Valero and Steven R. Hare

Abstract Evidence of continuing migration beyond the age of recruitment to the fi shery has recently led to the adoption of a coastwide method of assessment for Pacifi c halibut. The coastwide assessment currently in use has resulted in an estimated coastwide harvest rate near the target. However, area- specifi c harvest rates are estimated to have been more than twice the target in eastern areas and less than the target in western areas. In order to illustrate the effects of migration and fi shing on the distribution and population structure of Pacifi c halibut we developed a simulation model with a user-friendly graphical user interface (GUI). The GUI allows the user to specify different migration patterns and fi shing levels, run scenarios, and visualize results of the simulations. The underlying model of the GUI is an age-structured migratory model. Scenarios using migratory and fi shing patterns close to those observed for Pacifi c halibut suggest that the current spatial distribution of halibut differs from that expected under no fi shing conditions or under a spatially uniform harvest rate, with higher levels of depletion in eastern areas relative to western areas. Preliminary results are consistent with recent and historical fi shing patterns, recent coastwide assessment estimates of realized harvest rates, and historical estimates of halibut distribution. Further work is needed to evaluate the effects of different levels of uncertainty and spatial variability in both population and fi shery dynamics on the management of Pacifi c halibut.

Introduction The Pacifi c halibut (Hippoglossus stenolepis) is widely distributed in the north Pacifi c from Hokaido, Japan to northern California, U.S. The International Pacifi c Halibut Commission (IPHC) studies and manages halibut from the Bering Sea to northern California (Fig. 1). Halibut abundance changes along its geographic range, with the current center of abundance located around Kodiak Island (IPHC Regulatory Area 3A, Fig. 1) in the Gulf of Alaska (Clark and Hare 2006). There are also seasonal changes in halibut distribution resulting from spawning migrations. During summer, halibut are distributed on the continental shelf but during the winter mature halibut migrate to spawning grounds located in deeper waters. Recent archival tagging has identifi ed winter spawning migrations as long as 1200 km, as well as some degree of site fi delity to summer areas after the winter spawning migration (Loher 2008). After spawning, halibut eggs and larvae are carried by prevailing currents north and westward towards the western Gulf of Alaska and the Bering Sea. Juvenile halibut undertake an eastward-southward migration that counters the drift of eggs and larvae (Skud 1977). Until recently, it was assumed that this migration was completed by about age six or seven when halibut begin to become vulnerable to the fi shery. However, recent passive integrated transponder (PIT) tagging data have provided evidence of continuing ontogenetic halibut migration beyond age eight (Webster and Clark 2007). The counterclockwise (northward-westward) drift of eggs and larvae and the clockwise (eastward-southward) ontogenetic migration of juvenile and adult halibut is a type of “compensatory

37 2009 IPHC ANNUAL MEETING HANDOUT emigration” (Skud 1977). This process is expected to have evolved along with other biological traits of Pacifi c halibut on an evolutionary time scale, and has strong implications for population abundance and distribution. Another process affecting the abundance and distribution of Pacifi c halibut is fi shing. A number of Pacifi c coast Indian tribes fi shed halibut long before the arrival of Europeans to the Pacifi c northwest, but we expect that its impact on halibut dynamics was limited compared with the subsequent development of the commercial fi shery. The Pacifi c halibut commercial fi shery is considered to have started in 1888 near Cape Flattery off the northwest coast of Washington State (current IPHC Regulatory Area 2A, Fig. 1). The fi shery expanded rapidly, both in spatial extent and into deeper waters as fi shing areas became depleted and fi shermen expanded toward new areas. Understanding the effect of the fi shery on halibut population structure, abundance, and distribution was identifi ed as a crucial topic for management early on in the history of halibut research (Thompson 1916). From a conservation perspective, it was recognized early that the quantitative distribution of the species must be considered since all possible sources of eggs and young are important, whether at the limit of the species range or at the center (Thompson and Van Cleve 1936). The IPHC harvest policy uses the same target harvest rate for all areas (except areas of special concern) with the goals of altering as little as possible the relative distribution of halibut along its geographic range, and to have halibut encounter the same exploitation rate wherever they might be fi shed. For many years, this harvest policy was implemented using closed- area assessments under the assumption of no net migration of legal-sized fi sh between regulatory areas. In 2006, the IPHC staff and an external scientifi c peer review panel recognized the biases of the closed-area (CA) approach in light of the evidence of continuing migration of legal-size halibut, and moved to a coastwide (CW) assessment approach (Clark and Hare 2007). The CW assessment estimated recent coastwide realized harvest rates near the target harvest rate. However, realized harvest rates on an area basis are estimated to have been more than twice the target in eastern areas, and less than the target in western areas. The combination of complex migration throughout the halibut life history and varying temporal and spatial fi shing effects set the canvas for several key questions:

Q1: Is the current relative distribution of halibut abundance similar to that expected under no fi shing conditions? Q2: Are the unbalanced harvest rates (higher in the east than in the west) estimated by the coastwide model consistent with the dynamics and current population structure of the stock? Q3: Does applying the same fi shing mortality rate in all areas achieve the goal of not changing the relative distribution of abundance when taking into account migration? Q4: What is the effect of fi shing on “upstream” areas relative to effects of local fi shing in “downstream” areas?

Answering these questions based solely on observed data or experimental results is logistically prohibitive or impossible, even in a historically documented and data-rich fi shery such as the Pacifi c halibut. In order to answer these questions we need more than observed quantities. Another approach is to use all the available information on halibut population and fi shery dynamics to build a model with characteristics as close to reality as possible. We can then use this model to evaluate the effects of its different components and processes (such as migration, fi shing) on observable quantities (such as abundance, age structure, etc.) under different scenarios. In order to illustrate the effects of migration and fi shing on the distribution and population structure of Pacifi c halibut,

38 2009 IPHC ANNUAL MEETING HANDOUT we developed a simulation model with a user-friendly GUI (Graphical User Interface) widget. The widget allows the user to specify different migration patterns and fi shing levels, run scenarios, and visualize the results of the simulations. This report summarizes the structure of the fi rst release version of the widget as well as some preliminary results. A more detailed description of the widget, underlying model and results, as well as installation instructions can be found in Valero and Hare (2009). The fi rst version of the widget has many simplifying assumptions and basic alternatives for relevant processes as described in following sections. A second version of the widget is under development at the time of writing of this report.

Model description The widget consists of a GUI that allows the user to specify scenarios with different levels of migration and fi shing mortality rates (both as coastwide or area specifi c rates) using graphical elements such as buttons, tabs, and sliders (Fig. 2). The computations associated with the specifi ed scenario are made by an underlying simulation model and the results are presented graphically on the widget. The widget’s underlying model is an age-structured, multi-area model with migration. Sex-specifi c growth, maturity at age, and selectivity (survey and commercial) at size were assumed to be the same among areas. The model includes six areas corresponding to IPHC Regulatory Areas 4A, 3B, 3A, 2C, 2B, and 2A (Fig. 1). These areas were selected based on data availability, but the model can be expanded to include the remaining regulatory areas. The model includes ages eight (recruits) to fi fty, recruitment is assumed to be time-invariant and given by the average number of 8-year old halibut as estimated by CA stock assessments from 1996 to 2005. Migration is modeled as an annual movement of halibut between adjacent areas in an eastward direction only. Migration is assumed to occur at the beginning of the year, before any mortality occurs. Analysis of traditional (Quinn et al. 1985) and PIT tag (Webster and Clark 2007) recoveries suggest that the fraction of fi sh migrating is a function of fi sh size/age, with smaller/younger fi sh more likely to migrate than larger/older fi sh. Therefore, we specify greater migration rates for younger halibut. Selectivity curves at length are transformed to selectivity curves at age. Model scenarios run for 100 years, reaching equilibrium within 20 years in most cases. Results for scenarios run are illustrated in the widget using several tabs. Figures 2 to 7 illustrate results for a scenario set with migration and fi shing mortality rates at values compatible with migration studies and recent CW assessments. This is only one of many potential scenarios a user can specify, the goal is not to provide the best depiction of reality (that is the goal of assessment models) but instead to explore the effect of different processes (migration, fi shing) on the dynamics of the stock through simulation modeling. A map is displayed with circles proportional to the fraction of coastwide distribution of spawning biomass (Fig. 2). The “SBio dist.” tab (Fig. 2) illustrates the distribution of spawning biomass projected under the scenario run as well as the percentage change from the distribution expected under the same migration scenario but without fi shing. The “SBio depl.” tab (Fig. 3) illustrates area-specifi c (columns), coastwide (black horizontal line) spawning biomass depletion and spawning biomass limit (20% of expected spawning biomass under no fi shing, dashed horizontal line). The “Catch” tab (Fig. 4) illustrates the projected distribution of exploitable biomass and catch. Tabs “Ages Surv.” (Fig. 5) and “Ages Catch” (Fig. 6) illustrate the percentage of halibut older than 20 years in the IPHC survey and commercial catch as projected under the scenario run and those observed in 2007. The “HR” tab (Fig. 7) illustrates harvest rates

39 2009 IPHC ANNUAL MEETING HANDOUT corresponding to the fi shing mortalities specifi ed in the scenario run and recent harvest rates estimated by coastwide assessments.

Preliminary widget runs In order to illustrate the use of the widget, this section summarizes preliminary results obtained by running different scenarios relevant to the four questions asked in the Introduction. For a more detailed description of scenarios and results, see Valero and Hare (2009). It is important to note that the widget scenarios are based on simplifi ed migration and fi shing patterns, rates used to set up the scenarios should not be interpreted as fi nal values but rather values consistent with results of ongoing research. A similar caveat applies to widget results: the goal is to capture patterns in the dynamics/structure of the stocks and to explore the sensitivity of results to varying assumptions, rather than to obtain a defi nitive picture of a particular area.

Q1: Is the current relative distribution of halibut abundance similar to that expected under no fi shing conditions?

Widget scenarios with no fi shing and migration patterns consistent with results of tag- recovery analyses suggest higher than currently observed relative halibut abundance in the eastern part of the stock, particularly in Area 2B. Two differences regarding the percentage of older fi sh emerge between the widget projections and those observed. First, the widget projects much larger percentages of older fi sh in general than those observed, which is expected since the projection is in the absence of fi shing whereas the observed percentages refl ect fi shing effects. In addition, a larger percentage of older halibut is projected for the eastern part of the stock given the continued eastward migration of older fi sh. However, the observed percentage of older fi sh has the opposite pattern, with older fi sh in the western part of the stock. This point leads us to the next question:

Q2: Are the unbalanced harvest rates (higher in the east than in the west) estimated by the coastwide model consistent with the dynamics and current population structure of the stock?

The scenario run previously resulted in differences between the widget projections and observed quantities of interest such as relative distribution of abundance and percentage of older fi sh between the eastern and western parts of the stock. Since the previous scenario did not include fi shing, some of those differences are expected. To explore the potential effect of fi shing on discrepancies between projections and observed quantities we can set up another scenario with fi shing mortality set at F = 0.23 (corresponding to a harvest rate close to 0.20, the current target harvest rate for the Pacifi c halibut fi shery). Under this scenario, the projected percentage of older fi sh in the western part of the stock is smaller than observed and the opposite happens in the eastern part of the stock. This is expected since the eastern part of the stock has had a longer history of fi shing than the western part of the stock. Additional scenarios with unbalanced fi shing mortality rates, higher on the east than on the west, result in better concordance between widget projections and observed proportions of older fi sh (Figs. 5, 6). Harvest rates doubling (Areas 2C, 2A) or tripling (Area 2B) the target harvest rate are required to obtain projections with the low percentage of older halibut currently observed in eastern areas. Harvest rates set at (Area 3A) or below (Areas 4A, 3B) the

40 2009 IPHC ANNUAL MEETING HANDOUT target harvest rate result in high agreement between projected and observed percentages of older fi sh. In summary, preliminary explorations suggest that unbalanced harvest rates between the west and the east of the same magnitude as estimated by the coastwide assessment can explain the marked differences in age structure between the different parts of the stock. Projections under this level of unbalanced harvest rates result in marked departures of the spawning biomass distribution relative to unfi shed conditions and in highly heterogeneous depletion levels.

Q3: Does applying the same fi shing mortality rate in all areas achieve the goal of not changing the relative distribution of abundance when taking into account migration?

Projections using the same fi shing mortality rate for all areas result in only small departures from the relative distribution of spawning biomass expected under no fi shing. This pattern is similar under different migration patterns within the range of moderate migration rates of halibut and the target harvest rate levels used historically in its fi shery.

Q4: What is the effect of fi shing on “upstream” areas relative to effects of local fi shing in “downstream” areas?

Preliminary explorations suggest that varying levels of fi shing mortality in the western part of the stock does not alter the dynamics and age structure of eastern areas. Decreasing exploitation rates to zero (no fi shing) in western Areas 4A and 3B has no noticeable effect in the percentage of older fi sh in eastern areas if their own exploitation rates stay at the high historical levels. If anything, the relative spawning biomass contribution of eastern areas decreases given that unexploited western areas would increase their spawning biomass contribution.

Discussion and on-going work Preliminary explorations of the simulation model presented here suggest that the current distribution of halibut abundance differs from that expected under no fi shing conditions. Simulations under the assumptions described in this work project higher abundances in the eastern part of the stock (particularly in Area 2B) in the absence of fi shing. This is consistent with early estimates of relative distribution of halibut abundance during the development of the fi shery (Thompson and Van Cleve 1936). Lower abundances and smaller percentages of older fi sh in the eastern areas are consistent with higher historical exploitation rates estimated by the coastwide assessment approach. A common fi shing mortality rate for all areas results in only marginal departures from the relative distribution of unfi shed spawning biomass under different migration patterns within the range of moderate migration rates of halibut and the target harvest rate levels used historically in its fi shery. Varying exploitation rates in the western areas have little effect on the dynamics of eastern areas, if exploitation rates in the east remain as high as estimated by the coastwide assessment. The widget described here has many simplifying assumptions and basic options for relevant processes such as same size at age relationship for all areas, migration only between adjacent areas, time invariant recruitment and fi shing mortality rates. A second version of the widget is under development at the time of writing of this report. The new version includes additional options such as fully-specifi ed migration matrices among areas, different relationships between

41 2009 IPHC ANNUAL MEETING HANDOUT migration and age, temporal trends in exploitation rates and recruitment, different recruitment spatial distributions, and area-specifi c size at age relationships.

Glossary Exploitable Biomass: Biomass of halibut selected by the fi shery.

GUI: Graphical User Interface, interface that allows people to interact with computers by manipulating graphical elements such as icons, buttons, etc.

Juvenile: Halibut after larval development, ages 0 to seven.

Ontogenetic migration: Migration that occurs over the lifetime of an organism. In the case of Pacifi c halibut it is the ongoing migration resulting in the annual relocation of fi sh from one area to another, predominantly in an eastward-southward (clockwise) direction.

PIT: Passive Integrated Transponder tag.

PAT: Pop-up Archival Transmitting tag.

Recruit: Halibut at age eight.

Spawning biomass: Biomass of mature female halibut.

Spawning biomass depletion: percentage of current spawning biomass relative to that expected without fi shing.

Spawning migration: Seasonal (summer-winter) migration of mature halibut to and from spawning areas.

Widget: element of a graphical user interface that displays information changeable by the user.

References Clark, W. C. and Hare, S. R. 2006. Assessment and management of Pacifi c halibut: data, methods and policy. Int. Pac. Halibut Comm. Sci. Rep. 83.

Clark, W. C. and Hare, S. R. 2007. Motivation and plan for a coastwide assessment. Int. Pac. Halibut Comm. Report of Assessement and Research Activities 2006:83-96.

Loher, T. 2008. Homing and summer feeding site fi delity of Pacifi c halibut (Hippoglossus stenolepis) in the Gulf of Alaska, established using satellite-transmiting archival tags. Fish. Res. 92: 63-89.

42 2009 IPHC ANNUAL MEETING HANDOUT Quinn, T. J., Deriso, R. B. and Hoag, S. H. 1985 Methods of population assessment of Pacifi c halibut. Int. Pac. Halibut Comm. Sci. Rep. 72.

Skud, B. E. 1977. Regulations of the Pacifi c halibut fi shery, 1924-1976. Int. Pac. Halibut Comm. Tech. Rep. No. 15.

Thompson, W. F. 1916. Statistics of the halibut fi shery in the Pacifi c: Their bearing on the biology of the species and the condition of the banks. British Columbia Fisheries Department, 1915: 65-126.

Thompson, W. F. and Van Cleve, R. 1936. Life history of the Pacifi c halibut - Distribution and early life history. Int. Fisheries Comm. Rep. 9.

Valero, J. L. and Hare, S. R. 2009. Exploring effects of fi shing and migration on the distribution of Pacifi c halibut. Int. Pac. Halibut Comm. Report of Assessment and Research Activities 2008: 265-298.

Webster, R. A. and Clark W. G. 2007. Analysis of PIT tag recoveries through 2006. Int. Pac. Halibut Commission. Report on Assessment and Research Activities 2006: 129-138.

43 2009 IPHC ANNUAL MEETING HANDOUT Figure 1. IPHC Regulatory Areas. Areas included in the “Migration and Fishing Widget” are circled.

44 2009 IPHC ANNUAL MEETING HANDOUT Figure 2. Widget screenshot. The slider bars are user controls specifying migration and fi shing mortality rates. The map displays the equilibrium distribution of spawning biomass. Results of the scenario run are illustrated under different tabs, in this case showing the distribution of spawning biomass at equilibrium and percentage change from the corresponding projection with no fi shing.

45 2009 IPHC ANNUAL MEETING HANDOUT Figure 3. Widget screenshot showing area-specifi c (columns), coastwide (black horizontal line) spawning biomass depletion and spawning biomass limit (20% of expected spawning biomass under no fi shing, dashed horizontal line).

46 2009 IPHC ANNUAL MEETING HANDOUT Figure 4. Widget screenshot showing projected distribution of exploitable biomass and catch.

47 2009 IPHC ANNUAL MEETING HANDOUT Figure 5. Widget screenshot showing the area-specifi c percentage of halibut older than 20 years as projected under the scenario run and the percentages observed in the 2007 IPHC survey.

48 2009 IPHC ANNUAL MEETING HANDOUT Figure 6. Widget screenshot showing the area-specifi c percentage of halibut older than 20 years as projected under the scenario run and the percentages observed in the 2007 commercial catch.

49 2009 IPHC ANNUAL MEETING HANDOUT Figure 7. Widget screenshot area specifi c harvest rates corresponding to the fi shing mortality pattern specifi ed in the scenario run and recent area-specifi c harvest rates estimated by recent coastwide assessments.

50 2009 IPHC ANNUAL MEETING HANDOUT IPHC Biomass Apportionment Workshop summary and responses to signifi cant questions arising at the Workshop

Steven R. Hare, Ray A. Webster, Juan L. Valero, and Bruce M. Leaman

At the International Pacifi c Halibut Commission (IPHC) 2008 Annual Meeting, the Commission endorsed the staff’s recommended coastwide assessment approach to estimating exploitable biomass, which was a departure from the closed area assessments used previously. The staff also employed IPHC survey data and bottom area to apportion this coastwide biomass estimate into biomass estimates for individual Regulatory Areas and subsequently calculate recommended catch limits. The Commission chose to employ the survey apportionment for adopting 2008 catch limits for most Regulatory Areas but directed the staff to convene a workshop with harvesters and agency personnel to update understanding of fi sh movements based on tagging studies, further examine the survey-based apportionment, as well as examine alternate methods of biomass apportionment. The workshop was held on September 4, 2008 in Bellevue, Washington, chaired by an independent scientist, and attended by approximately 110 people, including IPHC Commissioners, agency staff, processors, as well as commercial and recreational harvesters. Background material was made available via the Commission’s website prior to the meeting, at:

http://www.iphc.washington.edu/halcom/meetings/workshop2008/baw2008.htm

The goals of the workshop were to: • Explain the basis for current assessment framework and survey apportionment method • Explore merits and impacts of alternate apportionment schema • Identify improvements to current apportionment approach

Staff presentations included a review of PIT tagging results and implications, review of assessment approach and harvest policy, consideration of the necessary characteristics for candidate apportionment methods, explanation and evaluation of alternate apportionment methods, and presentation of a simulation tool for examining impacts of assumptions about harvest policy and fi sh movement. All workshop presentations and a summary of the workshop are available on the Commission’s website. http://www.iphc.washington.edu/halcom/meetings/workshop2008/summary/apporwrkshp3.pdf

In addition, the workshop resulted in a number of signifi cant comments and questions, for which the IPHC staff has compiled detailed responses. Those responses follow and are also presented on the IPHC website, and in the 2008 Report of Assessment and Research Activities.

51 2009 IPHC ANNUAL MEETING HANDOUT 1. If the exploitation rates in Area 2, particularly Area 2B, have really been as high as 40- 50% then how can we still have stock left to fi sh on? Exploitation rate is removals divided by exploitable biomass. Exploitable biomass is a fraction (generally around 20-25%) of total resident biomass. It is quite incorrect to think that 40-50% of the halibut in Area 2 are being captured every year. This is particularly true of the younger aged fi sh (ages 6-11) as selectivity is quite low at those ages and quite a small fraction of resident fi sh ages 6-11 are taken annually. The catches in Area 2 have a much higher proportion of younger fi sh than the catches in Area 3 and 4. Fish of this age are growing relatively rapidly and their weight gain is the main addition annually to the surplus production and exploitable biomass in Area 2. Areas 3 and 4, which have a broader age spectrum, have a larger component of the exploitable biomass derived from older fi sh. The exploitable biomass in Area 2 is thus of a younger average age than in other areas and the high exploitation rate maintains the younger age distribution we see in the survey and commercial age compositions. Figures 1 and 2 illustrate the relative levels of estimated exploitable and unexploitable biomass and numbers, and actual removals in Area 2B, both for the past fi ve years as well as the detailed age and sex composition for 2007. All estimates are derived using relative survey catch rates among areas and the area/sex/age proportions are applied to the coastwide estimate of total numbers. Summing across just age groups 8 and older, exploitable biomass has ranged between 20 and 30% of total biomass during the past fi ve years. For comparison purposes, similar graphs for Area 3A are provided in Figures 1 and 3. Finally, while fi shing is still occurring in Area 2, the catch rates have declined precipitously. The survey CPUE in Area 2B declined by 59% between 1997 and 2007; Area 2C survey CPUE declined by 66%. Put another way, the catch rates in Area 2 are just 30-40% of what they were a decade ago. The combination of sharply declining survey catch rates, and young age composition of the stock is not indicative of sustainable fi shing and generally supports our estimates of too high harvest rates in Area 2.

2. The applicability of the coastwide survey data for biomass apportionment depends on catchability being equal, or nearly so, among all regulatory areas. Is this actually true? How can it be verifi ed? The assumption of equal catchability across regulatory areas is critical to use of the survey in apportioning biomass. Equal catchability implies that survey catch rates (“CPUE”) will be similar in all areas given the same density of fi sh. Equivalently, it also implies that areas of lower density will have proportionally lower catch rates. As far as we have been able to statistically detect, this assumption appears to be reasonable. At the very least, it is the default assumption and we would need to demonstrate an actual and consistent difference in catchability among areas to weight certain areas differently. The best means of verifying consistent catchability among different areas is probably to conduct a large scale trawl/setline survey experiment using paired stations. Figure 4 provides a new method of examining the catchability assumption. To prepare the fi gure, survey catch rates, in number of fi sh per survey unit of effort, were computed for males and female halibut ages 6 to 25. These catch rates were then weighted by bottom area to establish relative proportions by regulatory area for all ages and both sexes. This method uses no assessment model, it simply looks at survey derived relative catch rates. The catch rates are divided by selectivity at age, which differs by region according to relative size at age. There are several items of note in Figure 4. First, it can be seen that Area 2 (shaded blue in the fi gures) has

52 2009 IPHC ANNUAL MEETING HANDOUT around 20% of the youngest fi sh (ages 6-8). However, the share of older fi sh in Area 2 steadily declines with age (particularly among males) while the relative shares in Area 3 steadily increase. What these trends mean is that either: catchability differs by age among regulatory areas (i.e., that the relative share of older fi sh is similar among areas but older fi sh have a lower catchability in Area 2) or the catchability is similar among ages and there are fewer older fi sh in Area 2 due to increased mortality (presumably from fi shing). While differential catchability on the whole is possible among regulatory areas, it is more diffi cult to postulate a viable difference based solely on age across areas. We also note that the sablefi sh biomass apportionment is done in precisely the same manner as IPHC staff are proposing for halibut (with the exception that commercial CPUE is also part of the allocation calculation) – and there is no differential weighting by area for differences in catchability. In personal discussions with the sablefi sh assessment team, no reasons to suspect area differences in catchability were suggested.

3. NMFS survey data for Areas 3B and 3A were used to validate the assessment but not those from NMFS surveys in Areas 2C or 2A, nor those from DFO surveys in Area 2B. Why not? A comparison of NMFS swept area estimates and IPHC abundance estimates for Area 2C is illustrated in Figure 5. In general, the IPHC estimates are higher, by a factor of about two, than the NMFS estimates. There are two points at very high abundance where the NMFS estimates are higher and these are the 2005 and 2007 estimates for 60-69 cm halibut. Given the unsuitability of much of Area 2C bottom area for trawling, it is highly likely that the NMFS trawl survey does not survey all halibut habitat. This would result in an underestimate of halibut numbers in the NMFS trawl survey. We will inquire as to the availability of trawl swept area estimates of halibut abundance in 2A and 2B and make the same comparison, if possible at the Annual Meeting

4. The fi shery in various regulatory areas catches different proportions of the available catch limit prior to the median date of the IPHC survey, yet the IPHC does not account for these differences. What is the basis for not requiring some correction to the IPHC survey CPUE index based on those removals? If the fi shery has an effect on the biomass, why doesn’t it have an effect on the IPHC survey? Does the assessment of this impact incorporate all removals prior to the median date (both directed and incidental)? Should we scale survey CPUE by the proportion of the annual removals taken before the survey? In most regulatory areas, the commercial catch has little or no effect on commercial CPUE within a fi shing season (Webster and Clark, 2008, Figs. 3-7). The obvious exception is Area 2B, where commercial CPUE shows an initial decline each year with increasing cumulative commercial catch. A smaller decline is also apparent in some years in Area 2C. To apply a correction to survey CPUE based on proportion of catch taken prior to each year’s survey, you would need information on the relationship between survey CPUE and this proportion in each regulatory area. The proportions themselves should not be used as correction factors, as this would require making unsupportable assumptions about the strength of the relationship between survey CPUE and proportion of catch taken prior to median survey date. Data since 1998 show no apparent relationship between survey CPUE and proportion of commercial catch in any regulatory area (Webster and Clark, 2008, Fig. 1). However, in most areas, survey CPUE is highly variable from year to year, and the proportions are in quite a narrow range, making it very diffi cult to determine if a relationship exists, and impossible to devise a sensible correction factor. There

53 2009 IPHC ANNUAL MEETING HANDOUT is a greater range of proportions for some statistical areas within each regulatory area, and so we looked at whether, for any regulatory area, there is a clear pattern in the relationships between statistical area survey CPUE and proportion of commercial catch taken prior to the median survey date for a statistical area. It is diffi cult to discern consistent patterns in any regulatory area, with each containing some statistical areas which show signs of a negative relationship and those that show an apparent positive relationship of survey CPUE with proportion commercial catch taken before median survey date (Fig. 6). Other removals, such as sport, bycatch and personal use cannot be included in these analyses because we do not have data on these removals by date. In summary, there are no clear relationships between survey CPUE and proportion of commercial catch taken prior to the median survey date. The high variability in survey CPUE and often narrow range of the proportions mean we cannot conclude that there is no effect of catch taken prior to median survey date on survey CPUE, but if there is an effect, we cannot quantify it with suffi cient precision to use it as basis for calculating correction factors for survey CPUE.

Webster, R. A., and Clark, W. G. 2008. Questions about fi shery-survey interactions. Int. Pac. Halibut Comm. Report of Assessment and Research Activities 2007: 229-243.

5. Some believe the commercial fi shing in the vicinity of the survey stations affects CPUE. IPHC says they see no evidence of it, but is the method of determining an effect sensitive enough to detect an effect? Can you run some tests to assess the sensitivity of your tests and what effect would removals have? One of the ways we looked at the effect of commercial catch on survey CPUE was to plot station CPUE against commercial catch taken within a certain number of days prior to the set and within a specifi ed distance. If commercial catch close in time and space to the survey set had a negative effect on station CPUE, then we would expect to see decreasing CPUE with increasing catch. To see how sensitive such plots are to a relationship between station CPUE and commercial catch, we simulated data that mimics observed data (in terms of mean and variance) assuming relationships of varying strengths. A detailed summary for Area 2C is available on request; results for other areas were almost identical. Our conclusion is that if a relationship is not strong enough to be detected visually in such plots, the effect on survey CPUE will be extremely small (a reduction of 2% or less). The reason for this is that relatively few survey stations have any commercial catch taken nearby (we looked at catches taken within 5 days prior to and 5 nmi of the set), and most of the remainder have relatively little nearby catch. This means that survey CPUE is very robust even under simulated strong negative effects of commercial catch.

6. We use the estimates of F from PIT tagging in the tag recovery model but not in the assessment; why not? Tag recoveries were very low in almost all regulatory areas, and with natural mortality fi xed at 0.15 in all areas, this leads to estimates of F (commercial fi shing mortality) from tag-recovery modelling that are lower, sometimes much lower, than we think is credible. Although we still do not fully understand why, it is very likely the tag-recovery estimates of F are inaccurate, and for this reason they should not be used in the stock assessment modelling.

54 2009 IPHC ANNUAL MEETING HANDOUT The reverse question has also been posed: if we don’t believe the F values from the PIT tag modelling, can we run the models with the Fs from the stock assessment, and how will that affect migration estimates? To examine this question we fi xed Fs so that their relative magnitude among areas refl ects the relative magnitude of the stock assessment values. The most striking result is that estimates of migration from Area 4A are much higher under this model. However, the model fi ts poorly, as forcing Area 4A to have higher relative F leads to fi tted tag-recoveries that are much higher than those observed for this area. We do not think there is much merit to fi tting models that constrain the Fs in such a way. Estimated migration rates are effectively annual out-of-release- area recovery rates corrected for differences in tag-recovery probabilities among areas. Whatever the cause of the low tag-recovery rates, they are still what we observe, and it is appropriate when estimating migration to fi t models that approximate well the observed tag-recovery probabilities. The question is not if these observed recoveries are correct (they are what they are), but rather if they’re not caused by low Fs as the model says, what is causing them?

7. The survey apportionment is unfair and we don’t support it. The survey apportionment method is the most objective, consistent and standardized method available to us to partition the biomass. There are legitimate questions about potential regional differences in catchability but nothing that we’ve found to date has caused us to question it as the most scientifi cally defensible method of accurately apportioning coastwide biomass. It is also important to understand that the apportionment process addresses biological stock management goals, not social equity. Economic impacts are clearly a consequence of the assessment and application of a harvest policy.

8. Releases of PIT tags in Area 2A coincided with changes in the regulated distribution of the commercial fi sheries, with major halibut grounds closed due to bycatch concerns for other species, especially rockfi shes. This should affect the recoveries of tags. The Area 2A non-trawl RCAs cover an extensive portion of the halibut habitat and include the majority of the IPHC setline survey stations, at which tags were released in 2003. The closures, however, apply only until the end of August. This means that all tagged fi sh still have the potential to be captured by the commercial fl eet for at least part of the fi shing season. If overall harvest is reduced because of the closures, then that would lead to fewer recoveries of tagged fi sh, and corresponding decreases in estimates of rates of commercial fi shing mortality (if we had suffi cient data to attempt to estimate these). In general, changes in fi shing mortality are something the tagging study is designed to measure, although with few releases in Area 2A, this would be diffi cult even if recovery rates were comparable to northern areas.

9. Movement patterns from PIT recoveries refl ect ‘artifi cial’ regulatory area boundaries. What do the data indicate if they are partitioned into minor statistical areas? The question here seems to be whether “migration” is simply a back and forth movement of fi sh across the boundaries between regulatory areas. The raw recovery data show that Area 4A and 3B releases are recovered in statistical areas throughout areas to the east, with no apparent concentration of recoveries on the boundaries of these two areas with their neighbours. The only cases where there appears to be more recoveries near the boundary are 3A releases recovered in 3B, and 2C releases recovered in 2B.

55 2009 IPHC ANNUAL MEETING HANDOUT Further, if migration was simply due to boundary movements between neighbouring areas, we would expect net migration to be close to zero. The tag-recovery models produce direct estimates of emigration only: to estimate immigration rates, and hence net migration, we need to know the relative population sizes of each regulatory area. Assuming that tag-release numbers in 2003 are proportional to population size, we estimate large annual rates of net migration into 2B (+18%) and out of 4A (-13%), with net migration rates for other areas ranging from -5% to +4%.

10. Will additional PIT tagging resolve uncertainty around migration estimates? Additional PIT tagging will not be considered until we understand why recovery rates of tags have been so low in most regulatory areas. However, if there are more data, you will get more precise estimates, and for the purpose of estimating migration rates, future studies should have a greater proportion of tags released in areas such as Area 4 and Area 2A. The primary purpose of the 2003/2004 experiment was to estimate exploitation rates, hence population size, rather than migration rates, so the experimental design was optimized for that purpose.

11. How do recovery data from surveys compare with those from commercial fi sheries? There are relatively few survey recoveries (81 in 2006 and 52 in 2007), but we can see both similarities and differences (Tables 1-4). As with commercial recoveries, in both years combined, there have been very few recoveries in Area 4 (3 in total, and none in Area 4B). Unlike commercial recoveries, few recoveries have been made outside of the area of release – just three, or 2% of the total, compared with around 16% to date for commercial recoveries. This is likely to be an artifact of revisiting release sites on the survey. Fish that return to a survey site are more likely to be recovered than those that have moved somewhere else, which is more likely to be a location between stations than another station.

12. Should IPHC apply differential mortality estimates to tagged fi sh based on depth of capture during tagging? It has also been suggested that migration rates could also be made a function of depth strata on release. These are things we can explore in this year’s tag-recovery modeling.

13. Do estimates of migration rates from the tag-recapture model include tagging mortality? Tag-induced mortality was assumed to be “negligible” (i.e., fi xed at zero) in the model, based on the results of holding tank studies. A working value of 2.5% has been suggested, and adding this into the model has the same effect as increasing natural mortality or tag loss rate. That is, provided it’s the same for all areas, estimates of commercial fi shing mortality will be affected, but migration estimates will not be signifi cantly affected. Migration estimates depend on the relative amounts of mortality in each area, something barely affected by changing a parameter that is the same for each area. To check this, we fi tted a model with a 2.5% annual tag-induced mortality rate, and as expected, while the Fs changed, changes in migration rate estimates were miniscule.

14. The high bycatch in the Bering Sea is affecting recoveries of PIT tags. Most of the bycatch in the Bering Sea occurs on the fl ats and is comprised mainly of small fi sh, fi sh smaller than those generally caught and tagged on our setline survey (e.g., 80% of the halibut captured in the 2007 NMFS trawl survey of the Bering Sea were < 55 cm). Further, with

56 2009 IPHC ANNUAL MEETING HANDOUT the exception of Area 4D, almost all PIT tagging was done “downstream’ of where the majority of trawl bycatch is taken. These two factors make it unlikely that many PIT tags were taken in the bycatch. Also, legal-sized bycatch is accounted for in the tag-recovery modelling as part of non- commercial removals. For Bering Sea bycatch to be a cause of low tag-recoveries in the commercial catch in Area 4 (rather than reduced commercial catch overall), tagged fi sh would need to have been preferentially selected by bycatch over untagged fi sh, something that is hard to imagine occurring.

15. Is the decline in recruitment in Area 2 caused by the increase in harvest of halibut in Areas 4 and 3? How can that be demonstrated? Actually there is no indication of a decline in recruitment (as numbers of 8 year old halibut) in Area 2 for the last 10 years. Figure 7 shows the numbers of 8 year old halibut based on survey catch rates and bottom area for each regulatory area (top) and aggregated areas (bottom) for years 1998 to 2008. Estimates are corrected for differences in selectivity due to regional differences in size at age. The most recent recruitment estimates (2005 to 2008) include fi sh that have not recruited fully to the survey and therefore their estimates are less precise than those for earlier years. Even excluding those most recent recruitment estimates, the overall trend for all regulatory areas is of an increase or at least a stable recruitment level since 1998 (Fig. 7).

16. The ‘arithmetic’ slide showing effects of undetected migrations was not easily understood. This slide will be redesigned for all future presentations. The main point of the slide was that emigration had the same effect on estimating population size as an increase in natural mortality while immigration has the same effect as a decrease in natural mortality. It is well known, and easily demonstrated, that the natural mortality rate has the effect of scaling a population number upwards or downwards depending on its magnitude. A lower rate than that assumed infl ates numbers (as occurs in Area 2) while a higher rate than assumed decreases estimates (as in Area 4).

17. Does a mismatch of the timing of migration and the timing of the survey create biases? The survey occurs at midyear, when abundance is approximately at its average value for the year. By basing apportionment on a midyear apportionment, an equal rate of instantaneous fi shing mortality among regulatory areas is achieved. The mathematical details supporting a mid-year based apportionment are available in a paper titled “Effect of migration on achievement of proportional harvest under a system of survey apportionment of total catch” by William G. Clark and can found on pages 217-219 of the 2007 IPHC Report of Assessment and Research Activities.

18. Don’t ratchet up harvest in ‘source’ areas for migration – is cutting off migration to ‘destination’ areas. In order to answer this question we need to understand the combined impacts of fi shing and migration on the dynamics of halibut across IPHC regulatory areas. There is increasing evidence indicating that western areas have historically been exploited at lower harvest rates than areas in the eastern part of the stock. Average harvest rates estimated from the coast-wide assessment from 2001 to 2007 range from 0.15 (4A) to 0.17 (3B) in the west

57 2009 IPHC ANNUAL MEETING HANDOUT part of the stock and from 0.31 (2A) to 0.54 (2B) in the eastern part of the stock. The effect that different harvest rates in western areas may have on eastern areas (given the general pattern of eastern migration) can be illustrated with one of the widgets presented at the workshop. The widget simulates halibut population dynamics for areas 4A to 2A under different migration and fi shing rates. For more details on the widget and underlying model please refer to the widget tutorial and presentation by Valero during the Apportionment Workshop. Figures 8 to 12 show run results of 5 relevant scenarios to answer this question. All 5 scenarios assume a migration pattern similar to the estimated from PIT analysis. The “unfi shed conditions” scenario is shown in Figure 8. Simulated spawning biomass distribution is shown in the top panel, expected and observed (for the 2007 IPHC Survey) percentage of halibut older than 20yr in the survey are shown in the bottom left panel, and realized harvest rate estimated from the coastwide assessment is shown in the bottom right panel. Figure 9 represents the expected results under uniform harvest rates (0.2) across regulatory areas. Under this scenario the spawning biomass distribution is barely altered (top panel), with changes of less than 3% with respect to the unfi shed scenario. The percentage of simulated halibut older than 20yr decreases in all areas (as expected under exploitation) but the overall distribution pattern is similar to unfi shed conditions with older halibut in the western part of the stock (bottom left panel). However, the observed distribution of halibut older than 20yr (older halibut in the west than in the east) is dramatically different than that simulated under a balanced harvest rate (older halibut in the west); suggesting this scenario does not adequately capture observed conditions. Scenarios with unbalanced exploitation rates that are similar to those estimated by the coastwide model (Fig. 10, bottom right panel) capture the differential distribution of older halibut (Fig. 10, bottom left panel) and indicate a decrease in spawning biomass contribution for eastern areas (Fig. 10, top panel). Increasing harvest rates up to the target rate (0.2) in the western areas decreases the percentage of older fi sh in the western areas but have no noticeable effect either on the percentage of older fi sh (Fig. 11, bottom left panel) or the low contribution of spawning biomass in eastern areas (Fig. 11, top panel) if their exploitation rate stay at the levels estimated by the coastwide assessment (Fig. 11, bottom right panel). Conversely, decreasing exploitation rates to 0 (no fi shing) in western areas has no noticeable effect in the percentage of older fi sh in eastern areas if they if their exploitation rates stay at the high historical levels (Fig. 12, bottom panels). If anything, the relative spawning biomass contribution of eastern areas decreases given that unexploited western areas would increase their spawning biomass contribution. In summary, the widget simulations suggest that varying exploitation rates in the western areas have little effect on the dynamics of eastern areas if exploitation rates in the east remain as high as estimated by the coastwide assessment. Uniform (balanced) exploitation rates across the distribution of the stock do not alter the relative distribution of halibut biomass under migration patterns and exploitation rates on the order of what is expected for the halibut stock and its fi shery.

19. Show that survey data refl ect local abundance. A survey, by design, refl ects local abundance of fi sh multiplied by their selectivity (which is determined by the length of the fi sh). It uses the same standardized gear and bait at each station and presents the only consistent measure of abundance across areas that is available. If local abundance is mainly comprised of smaller fi sh, i.e., smaller than the legal size limit, the survey will

58 2009 IPHC ANNUAL MEETING HANDOUT produce a low estimate of exploitable biomass. Harvests are set on the basis of exploitable fi sh, not small fi sh, so the survey is the appropriate apportionment method.

20. There is a mismatch between survey CPUE and commercial CPUE among areas, i.e., they don’t track the same and they differ in ratio – is that evidence of catchability differences and why should the ratio be different? The commercial fl eet and the survey fi sh in very different ways and one would not expect the indices to have the same catch rate nor necessarily identical trends. The survey fi shes the same locations with the same amount of gear every year and measures how the grand average within each regulatory area changes over time. The commercial fl eet may or may not fi sh the same grounds but will constantly move to fi nd the highest concentration of fi sh. Thus the commercial index is likely to show less decrease when overall biomass declines, a situation termed “hyperstability” in the fi sheries literature. There is no means of identifying catchability differences among areas from the relative survey and commercial catch rates. What the difference in magnitude between the two may illustrate is the extent to which commercial fi shing is tightly clustered in some regulatory areas.

21. How does the IPHC account for terrain differences and gear differences among areas? The IPHC does not account for terrain differences. The design of the survey factors all forms of habitat into its method of calculating the CPUE index. Commercial CPUE is computed from fi xed hook gear in all areas, though in Areas 2A and 2B snap hook gear is also included. Catch rates in the commercial fl eet are adjusted for hook spacing.

22. The IPHC did some historical experimental comparisons of catchability differences between Area 3 and Area 2. Those studies indicated catchability differences, which the IPHC accepted; why are differences among areas now being rejected? The IPHC used a catchability correction between Areas 2B and 2C/3A/3B during the mid 1980s to mid 1990s in fi tting assessment models. The correction ranged from 1.0 (no correction) to 1.5 for Area 2B and 0.75 for Areas 2C/3A/3B. This practice was discontinued as more elaborate models were developed that allowed for freely varying catchability coeffi cients. The two studies that report on the trawl/setline catchability experiments are both replete with discussion of the very large variances (in part due to small sample sizes) in estimation of regional catchability differences. The experiments were restricted to trawlable grounds, only fi sh in the size range 82-99 cm were used (larger fi sh outswam the trawls), and there was no means of insuring that trawl catchability between the different areas was constant. For all these reasons, the conclusion that different catchabilities existed was always couched in general and highly-qualifi ed terms. In addition, the experiments were conducted in the mid 1980s, a time when dogfi sh were very prevalent in Area 2B and the experimental data refl ect such. Nowadays, the catch rate of dogfi sh is actually higher in 3A than in 2B. The presence of dogfi sh was generally cited as the cause of the catchability difference. As the difference in density of dogfi sh has since disappeared, it would make no sense to apply very uncertain results from experiments conducted 20 years ago to the situation today. That said, the time is right to revisit the question of whether a regional difference in catchability can be quantifi ed by an updated trawl/setline study.

59 2009 IPHC ANNUAL MEETING HANDOUT 23. The biggest fl aw in survey apportionment is that it doesn’t account for migration. This question was addressed in the answer to Question 17.

24. Use a different survey pattern in Area 2A because the grid pattern doesn’t match bottom area distribution. An analysis of station bottom depth and actual bottom depth distribution initially showed that Area 2A had a few too many shallow stations and too few deep stations. However, reanalysis of the data (summing station depths into correctly sized bins) eliminates most of the difference in the two distributions and does not indicate that a correction to the survey CPUE should be made.

25. Show that your tests of potential effects of commercial fi shing on survey data are sensitive enough to detect real differences and show the impact of those differences. This is addressed in the answer to Question 5.

26. Were the data previously presented just catch or did they include all removals? All data presented as catches were total removals with the exception of one slide that illustrated historical commercial catches from 1895-present. The apportionment shares method that referenced historical catches in fact used total removals from each area to calculate the share.

27. Let’s use a 15-yr average of commercial catch distribution by area as the basis for apportionment. Will that address stock management needs? Figure 13 shows total CEY (Constant Exploitation Yield), Catch Limit Recommendations and Realized Harvest Rates resulting from the suggested apportionment based on a 15 year average of commercial catch distribution among Halibut regulatory areas. Figure 14 summarizes the effect of alternative apportionment methods including the 15 year average on Realized Harvest Rates. The use of any form of catch history has the dual characteristics of both refl ecting whatever annual catch limits were at the time and incorporating no feedback for changes in the stock. In the case of simply refl ecting catch limits, those data will have built into them any shortcomings of the stock assessments at the time. This is particularly important when thinking of catch data as an apportionment tool for a coastwide biomass estimate because the catches in question were assigned on the basis of the closed-area assessments, which we now know to have been in error because of continuing movement of fi sh among areas. It was the conclusion of the 2007 Assessment Workshop and the external peer review that we cannot use a closed-area assessment for the stock, which was accepted by the Commission. Therefore, these catch limits cannot be expected to refl ect true stock biomass distribution. The lack of feedback in using a fi xed average catch distribution is signifi cant because such an apportionment will never change, in spite of any changes in the distribution of exploitable biomass. That those changes have, and will, occur is amply demonstrated by both the catch history by area and the area-specifi c biological characteristics of the stock, which are not dependent on any estimation or model-based process. Any improvements to individual areas resulting from decreases in current exploitation rates will never be refl ected in fi xed catch-averaging.

28. Use a blend of survey and commercial data in the apportionment index. This point was also raised during the 2007 IPHC Stock Assessment Workshop and it was addressed in its Final Report, page 17, Question 22:

60 2009 IPHC ANNUAL MEETING HANDOUT http://www.iphc.washington.edu/halcom/meetings/workshop2007/SAW07Reportfi nal.pdf

In most fi sheries the use of CPUE (survey or commercial) as an index for abundance has traditionally assumed a proportional relationship to abundance. However, that relationship is not always proportional and CPUE can decline faster than abundance (hyper-depletion) or slower than abundance (hyper-stability). Commercial CPUE indices are likely to be hyper-stable, that is: they tend to remain at stable values even when abundance drops. Some of the reasons behind commercial CPUE hyper-stability are the ability of fi shermen to obtain relatively high CPUE even when abundance is declining by distributing their effort to productive areas, and post-fi shing redistribution of fi sh resulting in concentration of fi sh in areas preferred by fi shers. Commercial CPUE is also likely to be biased due to changes in gear, technology and fi shers’ behavior. Some of the major world fi shery collapses (from forage fi sh to marine mammals) have been attributed to hyper-stability of commercial CPUE and its use as an index of abundance. In contrast to commercial CPUE, survey CPUE is a better index of abundance by being highly standardized and consistent both in spatial coverage and time of the year, gear and overall sampling methodology. Mixing commercial and survey CPUE will incorporate the potential biases of commercial CPUE into the resultant apportionment index. In theory, this could be addressed by giving different weights to each CPUE index. However, it would likely be diffi cult to fi nd a consistent and objective way to assign weights between CPUE indices.

29. Can the IPHC demonstrate the benefi ts to coastwide and regulatory area management? We take this question to mean one of two things: 1) the benefi ts of coastwide vs. closed- area assessments, because under either assessment scenario the Commission continues to enact regulatory area management, i.e., catch limits are set by individual regulatory areas; or 2) should we use the same exploitation rate in every regulatory area? The fi rst question of coastwide vs. closed-area assessment was the topic of the 2007 workshop and our external peer review, reported here: http://www.iphc.washington.edu/halcom/meetings/workshop2007/wrkshp2007.htm

As a result of staff reports and the external peer review, the Commission accepted the need for a coastwide assessment framework at the 2008 Annual Meeting. Accordingly, and given the inability to assess exploitable biomass accurately using closed-area assessments, the staff therefore uses this coastwide assessment approach. The errors resulting from incorrectly using the closed- area approach were detailed at the 2007 workshop. The second question is a question of basic biology. The two scenarios are: using a common exploitation rate, or using individual area-specifi c exploitation rates. We have shown in previous analyses that the Bering Sea Regulatory Areas constitute an area of lower productivity and are using a target harvest rate of 0.15, with all Areas from 4A eastwards using a target harvest rate of 0.2. It is fair to question whether we need to use a constant exploitation rate within each of these two production areas. It is clear from all previous tagging studies that there is substantial movement of fi sh for annual reproductive activities, with a general movement toward the central Gulf of Alaska. The relative contribution of the individual regulatory areas to the coastwide spawning biomass is far from equivalent and refl ects the spawning biomass in each area. What is of great concern in a biological sense is that the composition of the spawning biomass in each area differs not just in the magnitude of that spawning biomass, but also in its composition (Fig. 15). As a

61 2009 IPHC ANNUAL MEETING HANDOUT result of very high recent exploitation rates in Area 2, the spawning biomass is composed almost exclusively of fi sh < 20 yr old, and there is a large difference in reproductive capacity between the spawning biomasses in Areas 2 and 3. Over the past several years, removals from Area 2 have been around 32% of the coastwide removals, but Area 2 contributes just 20% of the female spawning biomass. This truncated age composition in Area 2 means that the reproductive value (the per unit spawning contribution over the fi sh’s lifetime in the exploitable stock) of recruiting fi sh in this Area is substantially below that of recruiting fi sh in Area 3 (or Area 4). Why is this disparity important? Halibut have evolved in response to a specifi c set of environmental challenges that has resulted in selection for its mode of lifetime reproductive contribution. This means that the long-term distribution of halibut biomass is dependent upon, among other things, a common contribution in terms of the reproductive value of fi sh from all areas. Persistence of the halibut stocks in individual areas is therefore a function of their reproductive contribution. The high exploitation rates in Area 2 relative to other areas have substantially altered the reproductive contribution from Area 2, both in terms of historical contribution and also that which would occur in the absence of exploitation. To ignore this change, is to ignore the necessity for the reproductive characteristics that have evolved within the halibut population. We do not know with certainty whether the reproductive contribution from a given regulatory area has a direct bearing on the reproductive success in recruitment to that area. However, basic biological principles argue that we should maintain the relative contributions to the reproductive output of the stock. If we do not do so, we are undertaking a substantial gamble that these evolved characteristics of the halibut population have no meaning in its persistence. The Precautionary Approach to fi sheries management requires that we respect these biological principles in our management strategy and the Commission’s harvest policy enacts that approach.

30. Migration widget uses constant recruitment pattern; simulate something more realistic. The Migration-Fishing widget presented at the Apportionment Workshop assumes that recruitment does not change during the simulated model run (100 years). Area specifi c recruitment was based on the area specifi c average recruitment estimates for the 1995-2005 period derived from closed-area assessments. This was only a fi rst approximation and we are currently working with more realistic recruitment scenarios that will be presented at upcoming meetings.

31. Can the IPHC show the potential mid-term impact of estimates of the projected strong incoming year-classes? Average recruitment for the 1988 to 1997 year classes is estimated to be around 12.5 million eight year olds. Estimates of recruitment numbers for the 1999 and 2000 year classes are around 31.5 million eight year old halibut. In Figure 16, the expected contribution of these different levels of recruitment is illustrated. From an average recruitment of 12.5 million halibut, average annual yield at a harvest rate of 0.20 would be approximately 5.4 million pounds over 10 years with catch peaking when the fi sh are 13 years of age (5 years from present). From an average recruitment of 31.5 million halibut, average annual yield over a ten year horizon (at a harvest rate of 0.20) would be 13.5 million pounds, again peaking at age 13

32. Hook competition not equal across areas but the IPHC is ignoring it. This is equivalent to a differential catchability argument among regulatory areas. We did make an analysis of hook competition (see pages 211-215 in the 2007 RARA) and found little

62 2009 IPHC ANNUAL MEETING HANDOUT difference in hook competition – in terms of baits still remaining – for Areas 2B through 4A. There is some evidence of increased hook competition in 2A and decreased hook competition in Areas 4B and 4CDE. Adjusting CPUE on the basis of this study remains an option but the differences are relatively minor.

33. The PIT tag data show almost no movement between Area 3A and 2C – that can’t be true. It isn’t true. We presented estimates of annual emigration rates from the tag-recovery modelling that showed a rate of emigration from Area 3A to Area 2C that, to the nearest 1%, was zero. That fi gure does not tell the full story. The estimated emigration rate from Area 3A to 2C is actually 0.4% (SD=0.1%). While this seems very small, Area 3A has a much larger population than Area 2C and the number of fi sh moving into 2C from 3A is a much larger proportion of the Area 2C population size than it is of 3A population. The tag-recovery model estimated an annual migration rate of 3% from 2C into 3A. Assuming that tag-release numbers are proportional to population size (see Question 9 above), we get an estimate of annual immigration into 2C from 3A that is around 2%, not large, but not zero either. The model estimate of annual migration from 3A to 2B is also 0.4%, implying that 3A makes a similar contribution to migration into 2B.

34. The ages on the slide of average survey age and %> 20 y old don’t match data on p.517 of 2007 RARA. Why not? The slight mismatch between the slides in the presentation and the data on p. 517 of the RARA is mainly due to the display of combined sex age data in the presentation while the RARA listed data segregated by sex. A very small source of difference comes from an updating of ages following quality control of ages that is done after initially data are collected for the RARA.

35. Is there information on survey catchability from survey recoveries of PIT tags? In the PIT tag study, all halibut caught on the fi rst three skates of the 2003 setline survey were tagged and released, resulting in tagging rates that were proportional to survey catch for each area. If catchability is the same in each regulatory area, this design should also lead to releases that are proportional to area abundance. If the equal catchability assumption holds, then we can expect survey tag recoveries per scanned fi sh to be approximately equal among areas. A lower rate of recoveries would be evidence of lower survey catchability in the release year, while a higher rate would imply higher catchability. Taken at face value, the combined 2006-07 survey recovery rates imply higher survey catchability in Area 2B, and lower catchability in Areas 4A and 3A (Table 5). However, the rates are not precisely estimated, and we cannot rule out other factors affecting these values, for example, greater site fi delity of tagged fi sh in some areas than others. Nevertheless, if the concern is that survey CPUE apportionment is unfair on Area 2 because catchability is lower in that area, these data provide no evidence to support that concern.

63 2009 IPHC ANNUAL MEETING HANDOUT Table 1. Survey recoveries in 2006 of PIT tags released in 2003, by release and recovery area. Release Area Recovery Area (2006) (2003) 4D 4A 3B 3A 2C 2B Total 4D 1 1 4A 0 3B 16 2 18 3A 12 12 2C 6 6 2B 6 6 Total 1 0 16 14 6 6 43

Table 2. Survey recoveries in 2006 of PIT tags released in 2004, by release and recovery area. Release Area Recovery Area (2006) (2004) 4D 4A 3B 3A 2C 2B Total 3A 34 34 2B 4 4 Total 0 0 0 34 0 4 38

Table 3. Survey recoveries in 2007 of PIT tags released in 2003, by release and recovery area. Release Area Recovery Area (2007) (2003) 4D 4A 3B 3A 2C 2B Total 4D 0 4A 2 1 3 3B 11 11 3A 7 7 2C 1 1 2B 5 5 Total 0 2 11 8 1 5 27

Table 4. Survey recoveries in 2007 of PIT tags released in 2004, by release and recovery area. Release Area Recovery Area (2007) (2004) 4D 4A 3B 3A 2C 2B Total 3A 24 24 2B 1 1 Total 0 0 0 24 0 1 25

64 2009 IPHC ANNUAL MEETING HANDOUT Table 5. Setline survey recovery rates of 2003 releases per 10,000 fi sh scanned calculated from combined 2006 and 2007 scanning data Legal halibut Recoveries Catch scanned Tag per 10,000 area (10,000s) recoveries scanned (SD) 4D 1,561 1 6 (6) 4A 4,404 2 5 (3) 3B 25,107 27 11 (2) 3A 42,818 22 5 (1) 2C 7,314 7 10 (4) 2B 4,627 11 24 (7) Coastwide 90,974 70 8 (1)

65 2009 IPHC ANNUAL MEETING HANDOUT NOTE: The following fi gures make extensive use of color, which is not conveyed well in black and white. Full-color fi gures can be viewed here: http://www.iphc.washington.edu/halcom/meetings/workshop2008/summary/qncomm.pdf

2B 3A

14 40 UnexploitableNos. UnexploitableNos. 12 ExploitableNos. 35 ExploitableNos. TotalRemovals(Nos.)

8+) TotalRemovals(Nos.)  8+) 10  30 (age  (age  25 8 20 halibut  halibut 6  of  of  15 4 10 Millions 2 Millions 5

0 0 2004 2005 2006 2007 2008 2004 2005 2006 2007 2008

160 500 UnexploitableBiomass. UnexploitableBiomass. 450 140 ExploitableBiomass ExploitableBiomass

8+) 400 8+)

 TotalRemovals(weight) 120  TotalRemovals(weight) 350 (age (age  100  300 80 250 pounds pounds  

of 200 of  60  150 40 100 Millions 20 Millions 50 0 0 2004 2005 2006 2007 2008 2004 2005 2006 2007 2008

Figure 1. Estimates of halibut numbers (top plots) and weight (bottom plots) in Area 2B (on the left) and 3A (on the right) for 2004-2008. The plots show estimated numbers and weight of exploitable halibut, unexploitable halibut and actual removals for 2007. The biomass only counts fi sh aged 8 and older thus the unexploitable biomass and numbers are understated (there is almost no commercial catch of fi sh younger than 8).

66 2009 IPHC ANNUAL MEETING HANDOUT 2BFemalesintheseaandcatch 2BMalesintheseaandcatch 0.5 bynumber 0.5 bynumber

UnexploitableNos. 0.4 UnexploitableNos. 0.4 ExploitableNos. ExploitableNos.

halibut 0.3 halibut

0.3   2007CatchNos. of

of 2007CatchNos.   0.2 0.2 Millions Millions 0.1 0.1

0.0 0.0 10 12 14 16 18 20 22 24 10 12 14 16 18 20 22 24 Age Age 2BFemalesintheseaandcatch 2BMalesintheseaandcatch 7.0 byweight 5.0 byweight 4.5 6.0 UnexploitableBiomass UnexploitableBiomass 4.0 5.0 3.5 ExploitableBiomass. ExploitableBiomass

pounds 3.0 pounds   4.0 2007Catchlbs.

2007Catchlbs. of of

 2.5  3.0 2.0 2.0 1.5 Millions

Millions 1.0 1.0 0.5 0.0 0.0 10 12 14 16 18 20 22 24 10 12 14 16 18 20 22 24 Age Age

Figure 2. Breakdown of estimated exploitable and unexploitable numbers and biomass and actual removals of halibut by age and sex in Area 2B in 2007. Numbers are shown only for ages 10+ as inclusion of ages 8 and 9 would make values for older ages almost imperceptible. For comparison, the estimated number of age 8 females is approximately 2.7 million, compared to an estimated total of 0.63 million 10 year olds.

67 2009 IPHC ANNUAL MEETING HANDOUT 3AFemalesintheseaandcatch 3AMalesintheseaandcatch 2.0 bynumber 2.0 bynumber

UnexploitableNos. 1.5 UnexploitableNos. 1.5 ExploitableNos. ExploitableNos. halibut halibut   2007CatchNos. of

of 2007CatchNos.  1.0  1.0

0.5 0.5 Millions Millions

0.0 0.0 10 12 14 16 18 20 22 24 10 12 14 16 18 20 22 24 Age Age 3AFemalesintheseaandcatch 3AMalesintheseaandcatch 35.0 byweight 20.0 byweight

30.0 UnexploitableBiomass UnexploitableBiomass 15.0 25.0 ExploitableBiomass. ExploitableBiomass pounds pounds   20.0 2007Catchlbs.

2007Catchlbs. of of

 10.0  15.0

10.0 5.0 Millions Millions 5.0 0.0 0.0 10 12 14 16 18 20 22 24 10 12 14 16 18 20 22 24 Age Age

Figure 3. Breakdown of estimated exploitable and unexploitable numbers and biomass and actual removals of halibut by age and sex in Area 3A in 2007. Numbers are shown only for ages 10+ as inclusion of ages 8 and 9 would make values for older ages almost imperceptible. For comparison, the estimated number of age 8 females is approximately 4.3 million, compared to an estimated total of 1.51 million 10 year olds.

68 2009 IPHC ANNUAL MEETING HANDOUT 2008Females 4EBS 100% 90% 4D 80% 4B 70% 4A 60% 50% 3B 40% 3A 30% 20% 2C 10% 2B 0% 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 2A

2008Males 4EBS 100% 90% 4D 80% 4B 70% 4A 60% 50% 3B 40% 3A 30% 20% 2C 10% 2B 0% 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 2A

Figure 4. Estimated relative distribution of halibut ages 6 to 25 (which is a plus group) among IPHC Regulatory Areas. Estimates are derived from area weighted catch rates, corrected for differences in selectivity due to regional differences in size at age, on the IPHC setline survey and are not derived from a model.

69 2009 IPHC ANNUAL MEETING HANDOUT 0e+00 2e+06 4e+06 6e+06 8e+06 0e+00 1e+06 2e+06 3e+06 4e+06 5e+06 NMFS trawlNMFS survey swept area estimates - Millions of halibut Stock assessment estimates - Millions of halibut

Figure 5. Comparison of population estimates between the IPHC stock assessment and NMFS trawl survey swept area estimates for regulatory area 2C. Each dot represents one year’s estimate of fi sh in a 10 cm group (60, 70, 80, 90, 100, 110, 120+). A 1:1 line is illustrated for reference.

70 2009 IPHC ANNUAL MEETING HANDOUT o d ls. Area 2C Area 4A Proportion of catch prior to median survey date survey median to prior catch of Proportion date survey median to prior catch of Proportion

0.0 0.2 0.4 0.6 0.8 1.0 0.0 0.2 0.4 0.6 0.8 1.0

0 0 0 0 500 400 300 200 100 0 1500 1000 500 0

Setline Stat Area CPUE (lbs/skate) CPUE Area Stat Setline (lbs/skate) CPUE Area Stat Setline Area 2B Area 3B Proportion of catch prior to median survey date Proportion of catch prior to median survey date

0.0 0.2 0.4 0.6 0.8 1.0 0.0 0.2 0.4 0.6 0.8 1.0

010102020300 250 200 150 100 50 600 500 400 300 200 100

Setline Stat Area CPUE (lbs/skate) CPUE Area Stat Setline (lbs/skate) CPUE Area Stat Setline Area 2A Area 3A Proportion of catch prior to median survey date Proportion of catch prior to median survey date

0.0 0.2 0.4 0.6 0.8 1.0 0.0 0.2 0.4 0.6 0.8 1.0

010150 100 50 0 500 400 300 200 100

Setline Stat Area CPUE (lbs/skate) CPUE Area Stat Setline (lbs/skate) CPUE Area Stat Setline Figure 6. Relationship between survey CPUE at a statistical area and proportion of commercial catch taken in that area prior t prior catch taken in that area of commercial and proportion 6. Relationship between survey CPUE at a statistical area Figure estimate The lines are (1998-07). in one year one statistical area the CPUE for its median survey date. Each point represents distinguished by colours and symbo are statistical areas different Data for each statistical area. lines for regression linear

71 2009 IPHC ANNUAL MEETING HANDOUT 14

12

4EBS 10 4D 4B 8 4A 3B 6 3A 2C 2B 4 2A Age 8 halibut (Millions) halibut 8 Age 2

0 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 Year

20

18

16

14

12 4 10 3 2 8

6

4 Age 8 halibut (Millions) halibut 8 Age

2

0 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 Year

Figure 7. Halibut recruitment as numbers of 8 year olds for each regulatory area (top) and aggregated area (bottom) for years 1998 to 2008. Recruitment was calculated from survey catch rates and bottom area, corrected for differences in selectivity due to regional differences in size at age.

72 2009 IPHC ANNUAL MEETING HANDOUT Figure 8. Widget screenshot for Scenario 1, “Unfi shed conditions”.

73 2009 IPHC ANNUAL MEETING HANDOUT Figure 9. Widget screenshot for Scenario 2, “Proportional harvest rate”.

74 2009 IPHC ANNUAL MEETING HANDOUT Figure 10. Widget screenshot for Scenario 3, “Historical realized harvest rates”.

75 2009 IPHC ANNUAL MEETING HANDOUT Figure 11. Widget screenshot for Scenario 4, “Target harvest rate in Western areas”.

76 2009 IPHC ANNUAL MEETING HANDOUT Figure 12. Widget screenshot for Scenario 5, “No fi shing in Western areas”.

77 2009 IPHC ANNUAL MEETING HANDOUT Figure 13. Widget screenshot illustrating Total CEY (Constant Exploitation Yield), Catch Limit Recommendation and Realized Harvest Rate resulting from an apportionment based on a 15 year average of commercial catch distribution among halibut regulatory areas.

78 2009 IPHC ANNUAL MEETING HANDOUT Figure 14. Widget screenshot summarizing the effect of alternative apportionment methods on Realized Harvest Rates among halibut regulatory areas.

79 2009 IPHC ANNUAL MEETING HANDOUT 2008SpawningBiomasscomposition

1% 2% 3% 6% 9% 2A 2B 5% 10% 2C 3A 3B 22% 4A 4B 4D 42% 4EBS

Averageageofspawners,byarea 14

13

12

11

10

9 2A 2B 2C 3A 3B 4A 4B 4D 4EBS

Figure 15. Characteristics of halibut female spawning biomass. Top plots shows estimated distribution of spawning biomass by regulatory area, as estimated from survey catches of female halibut. Bottom plots shows average age of spawning females by regulatory area. The thick line in the bottom plot is the coastwide average age.

80 2009 IPHC ANNUAL MEETING HANDOUT Numbers

35 30 Avg. R 25 High R 20 15 10 No. Halibut (M) Halibut No. 5 0 8 9 10 11 12 13 14 15 16 17

Exploitable Biomass

120

100

80

60

40

20

Contribution to EBio (M lbs) 0 8 9 10 11 12 13 14 15 16 17

Weight in Catch

25.0

20.0

15.0

10.0

5.0

Contribution to yield (M lbs) (M yield to Contribution 0.0 8 9 10 11 12 13 14 15 16 17 Age

Figure 16. Illustration of the relative contributions to exploitable biomass and yield for two different levels of recruitment. The top panel shows the decline in numbers at age. The middle panel shows expected annual contribution to exploitable biomass at age and the bottom panel shows expected contribution to catch (yield). Age is synonymous with year (e.g., 8 or 2008) and both trajectories assume an annual harvest rate of 0.20.

81 2009 IPHC ANNUAL MEETING HANDOUT 82 2009 IPHC ANNUAL MEETING HANDOUT Assessment of the Pacifi c halibut stock at the end of 2008

Steven R. Hare and William G. Clark

Abstract As has been the case since 2006, the IPHC stock assessment model was fi t to a coastwide dataset to estimate total exploitable biomass. The coastwide exploitable biomass was then apportioned among regulatory areas in accordance with survey estimates of relative abundance, corrected for regional hook competition. Coastwide exploitable biomass in 2009 is estimated to be 325 million pounds, down from the 361 million pounds estimated last year. Virtually all of the decrease is due to lower survey and commercial catch rates of legal-sized halibut. Projections based on the currently estimated age compositions suggest that the exploitable and female spawning biomasses will increase over the next several years as a sequence of strong year classes recruit to the legal- sized component of the population.

Introduction Each year the International Pacifi c Halibut Commission (IPHC) staff assesses the abundance and potential yield of Pacifi c halibut using all available data from the commercial fi shery and scientifi c surveys (Appendix A). A biological target level for total removals from each regulatory area is calculated by applying a fi xed harvest rate to the estimate of exploitable biomass in that area. This target level is called the “constant exploitation yield” or CEY for that area in the coming year. The corresponding target level for catches in directed fi sheries subject to allocation is called the fi shery CEY. It comprises the commercial setline catch in all areas plus the sport catch in Area 2B, and the sport plus ceremonial and subsistence catches in Area 2A. It is calculated by subtracting from the total CEY an estimate of all unallocated removals - bycatch of legal-sized fi sh, wastage of legal-sized fi sh in the halibut fi shery, fi sh taken for personal use, and sport catch except in Areas 2A and 2B. Staff recommendations for catch limits in each area are based on the estimates of fi shery CEY but may be higher or lower depending on a number of statistical, biological, and policy considerations. Similarly, the Commission’s fi nal quota decisions are based on the staff’s recommendations but may be higher or lower. For many years the staff assessed the stock in each regulatory area by fi tting a model to the data from that area (Clark and Hare 2008). This procedure relied on the assumption that the stock of fi sh of catchable size in each area was closed, meaning that net migration was negligible. A growing body of evidence from both the assessments (Clark and Hare 2007) and the ongoing mark- recapture experiment (Webster and Clark 2007, Webster 2008, Webster 2009a) is consistent with a continuing eastward net migration of catchable fi sh from the western Gulf of Alaska (Areas 3B and 4) to the eastern side of the stock (Area 2). The effect of this migration on the closed-area stock assessments was to produce underestimates of abundance in the western areas and overestimates in the eastern areas. To some extent this has almost certainly been the case for some time, meaning that exploitation rates were well above the target level in Area 2 and a disproportionate share of the catches have been taken from there. In order to obtain an unbiased estimate of the coastwide stock beginning with the 2006 assessment, the staff built a coastwide data set and fi tted the model to it. Exploitable biomass in

83 2009 IPHC ANNUAL MEETING HANDOUT each regulatory area was estimated by partitioning, or apportioning, the total in proportion to an estimate of stock distribution derived from the setline survey catch rates (CPUE). Specifi cally, an index of abundance in each area was calculated by multiplying survey CPUE (running 3-year average) by total bottom area between 0 and 300 fm (Hare 2008). The logic of this index is that survey CPUE can be regarded as an index of density, so multiplying it by bottom area gives a quantity proportional to total abundance. This year an adjustment was applied to the index for each area, derived on the basis of hook competition. The estimated proportion in each area is then the adjusted index value for that area divided by the sum of the adjusted index values.

Observations from the survey and commercial fi shery The IPHC collects data from a variety of sources to characterize the status and population trends in all regulatory areas, and assist in fi tting a population assessment model. Some of the more important datasets are summarized herein. Total removals from the halibut populations come from seven categories: commercial catch (IPHC survey catch is included in this category), sport catch, legal-sized bycatch (from a variety of fi sheries targeting species other than halibut), personal use, legal-sized wastage from the commercial fi shery, sublegal-sized bycatch from non-target fi sheries, and sublegal-sized wastage from the commercial fi shery. Detailed descriptions of each category are contained in the Fishery Removals section of the annual Report of Assessment and Research Activities. The 2008 regulatory area total removals are illustrated in Figure 1, coastwide total removals from 1935 to 2008 are illustrated in Figure 2 (and listed in Appendix Table A1). Commercial catch is separately listed in Appendix Table A2. On a coastwide basis, total removals are at their lowest level since 1996. The pattern of changes between 1996 removals and 2008 removals has been quite different among regulatory areas, however. The current Standardized Stock Assessment (SSA) survey has been conducted since 1996 in almost all areas and in all years. The exceptions are the Eastern Bering Sea shelf which was surveyed only in 2006; Area 2A which was not surveyed in 1996, 1998, and 2000, the Area 4D edge which was not surveyed in 1996, 1998 and 1999, and Area 4A and 4B which were not surveyed in 1996. Stations are placed on a 10-nautical mile grid between depths of 20 and 275 fathoms, resulting in a total of approximately 1280 stations. A key indicator of stock status in each regulatory area is the weight of legal-sized (> 32 inch) halibut caught per standardized skate, termed the survey CPUE (Fig. 3 and Appendix Table A3). Survey CPUE has declined by over 50% on a coastwide basis over the past 10 years. While the rate of decline has differed among areas, there has been a substantial decrease in CPUE in all areas, indicative of a consistent coastwide decline in exploitable biomass. The survey catch of halibut is sampled to obtain biological information about the stock including sex and age distribution. (The regulatory area distributions are included in the Area Summaries section at the end of this document). The age structure of the population is of considerable interest for a variety of reasons. These distributions indicate the relative abundance of fi sh available to the fi shery, relative contributions to the female spawning biomass, etc. In 2008, there is a general tendency for an older age structure in the western areas, relative to the eastern areas. In particular, the lack of fi sh older than 20 years is noted for Area 2. Area 3B presents a somewhat anomalous age distribution in that it more closely resembles Area 2 than Area 3A or Area 4 distributions. The reasons for this are presently unclear although the estimated rate of fi shing mortality is not

84 2009 IPHC ANNUAL MEETING HANDOUT excessive and there appears to be substantial recruitment into this area. The staff will be conducting an extensive investigation of this area in the 2009 assessment. Sex and age-specifi c catch rates are also computed; these are discussed and plotted in the section on Assessment model fi t. The second major component of the annual IPHC data collection is sampling the commercial catch. From commercial fi shing logs, commercial CPUE is computed for each regulatory area (Fig. 4 and Appendix Table A4). As with the survey CPUE, there has been a consistent coastwide decline in commercial CPUE though not quite as pronounced. This is not unexpected however, as commercial fi shers tend to move their effort to maintain their catch rate, whereas the survey maintains the same fi shing locations every year. Approximately 1500 otoliths are collected and aged from each regulatory area (smaller samples in Areas 2A and 4B). Because commercially caught halibut are gutted at sea, the sex of halibut is unknown when sampled at the port of landing. As with the survey age samples, the fi sh in Area 2 are, on average, several years younger than fi sh caught in Areas 3 and 4. Here, as well, Area 3B is anomalous in that the average age of fi sh is closer to the Area 2 average. Every year, the IPHC places a sampler aboard the National Marine Fisheries Service (NMFS) Eastern Bering Sea (EBS) groundfi sh/crab trawl survey. The sampler collects biological data on the halibut catches, taking lengths of almost all halibut caught and selecting a subsample for aging. Due to the high cost, and very low catch rate, of setline surveying halibut in the EBS, the IPHC does not conduct the SSA grid survey in that region. While the IPHC survey does operate along the Area 4D edge, that region is not indicative of densities and trends across the broad shelf. For the purposes of apportionment, it is vital that a measure of density for the EBS shelf be derived each year, and the NMFS groundfi sh trawl survey is leveraged to allow just such an estimate. The NMFS survey generates swept area estimates of abundance for the entire shelf (Fig. 5). In 2006, the IPHC added 100 extra stations to the SSA grid survey and placed these across the shelf to obtain an estimate of shelf-wide density. In that year, mean density was estimated to be 18.1 pounds per standardized survey skate. That estimate of density is tied to the NMFS trawl survey to provide the annually varying estimate of density. We feel this method is valid for the following reason. From the NMFS trawl survey we actually obtain swept area estimates of abundance at length. We then apply the stock assessment estimated survey selectivity at length schedule to the full catch to provide an index of survey catch rate, comparable to the SSA survey fi shing gear. Figure 6 illustrates how the length frequency distribution resulting from this treatment of trawl survey data compares to the actual length frequencies collected in the 2006 IPHC special EBS setline survey. In this manner, we are able to obtain, for a fraction of the cost it would take to survey the EBS with a setline survey, a highly reliable index of halibut abundance across the EBS fl ats. As can be noted from the time series, the EBS is also showing a strong decline in halibut abundance over the past decade, with an estimated decline of more than 50%. Part of the coastwide decline in exploitable biomass can be attributed to a decline in size at age. For a given number of halibut in the population, a smaller size at age results in a smaller cumulative biomass. Figure 7a shows how the average weights of halibut in survey and commercial catches have changed over the past 12 years. Average weight has declined by 25% in the survey catches and 33% in the commercial catches. While the decline could be due to a decline in average age of the fi sh in the catches (since younger fi sh are smaller), Figure 7b shows this has not been the case as average age in both the survey and commercial catch has actually increased by several years. Trends, by regulatory area, in average age and average weight are illustrated in Figure 8.

85 2009 IPHC ANNUAL MEETING HANDOUT Description of the assessment model For the fi rst time in ten years, a new lead analyst (author SRH) has taken over the assessment (from author WGC, who retired in 2008 and had been the lead analyst). In addition, since last year’s acceptance of a coastwide stock assessment model, much of the focus of the staff and the industry is now on how the coastwide estimate of exploitable biomass is apportioned among regulatory areas. For both these reasons, the assessment model for 2008 is identical to that used for the 2007 assessment. This model has been essentially unchanged since 2003. It has been thoroughly described in an IPHC Scientifi c Report (Clark and Hare 2006) and was subjected to an external peer review by two scientists from the Center for Independent Experts. In the interest of brevity, little discussion is presented here of the model itself. Interested readers are referred to Clark and Hare (2006, 2007b, 2008) for full details. Until 2006, estimates of halibut abundance were made using closed area models for all areas except Areas 2A and 4CDE. Area 2A leveraged the Area 2B assessment and relative survey CPUE, while Area 4CDE relied upon the NMFS EBS trawl estimates of swept area abundance. The closed area models are no longer considered reliable but for the sake of comparison they are still fi tted to data and provide abundance estimates. The EBio estimates produced by the closed area fi ts are contained in the summary tables listed in the section on coastwide abundance apportionment.

Alternative model fi ts As was done in 2007, four versions of the basic assessment model were fi tted. The main difference for three of the models concerned how survey catchability (which is referred to as “q” below) was parameterized. The fourth variant excluded commercial CPUE from the model fi t and is considered to be similar to many of the NMFS groundfi sh assessment models. The models are summarized as such: (i) Survey q constant: catchability is a single fi xed (though estimated) value in all years. (ii) Survey q drift: survey catchability estimated for each year, but with a penalty of 0.05 on log differences. This is similar to the treatment of commercial catchability. (iii) Survey q trendless drift: same as Survey q drift, but with the additional requirement that a regression of estimated survey catchability on year have zero slope. This means that survey catchability was allowed to vary but not to show any trend over time. This was last year’s production model. (iv) No commercial CPUE: Commercial CPUE is not included in the likelihood. Table 1 shows features of the candidate model fi ts and some others. The best fi t, indicated by a delta AIC score of zero is the survey q drift model. Nearly as good a fi t is provided by last year’s production model, survey q trendless drift. The two other model fi ts are signifi cantly worse. The exploitable biomass estimate produced by all four models covers a very narrow range. As in 2007, the survey q trendless drift model is selected as the production model and the coastwide exploitable biomass estimate of 325 million pounds forms the basis for apportionment among regulatory areas.

Effect of the 2008 data on abundance estimates Coastwide survey CPUE declined by 9% and commercial CPUE declined by 8% from 2007 to 2008 (Figs. 3 and 4; Appendix A tables A3 and A4). As a result, the 2008 coastwide model fi t is

86 2009 IPHC ANNUAL MEETING HANDOUT revised downwards, by about 20%, from the estimate of abundance at the beginning of 2008 made in the 2007 assessment (Table 2). At the same time the 2008 fi t shows an increase in abundance, of about 12%, between the beginning of 2008 and the beginning of 2009. The net result is an estimated decline of 10% between the 2008 beginning of year exploitable biomass and the 2009 beginning of year exploitable biomass.

Evaluation of the assessment Quality of fi ts The model predicts survey CPUE at sex/age and commercial catch at age very well (not shown). There is no apparent pattern to the residuals from the fi ts, although the model initially underestimates slightly the early strength of the 1987 year class. The model is successfully predicting the increasing number of fi sh aged 25 and older, particularly males, which are appearing in both the survey and commercial catches. The very low growth rate for male halibut means that many are not recruiting to the fi shery until they are older than 25. This “plus” group is poised to increase even more in the new few years as the remains of the very large 1987 and 1988 year classes reach 25 years of age. The series of total survey and commercial CPUE are also predicted closely (Fig. 9, middle panels).

Estimates of recruitment, exploitable biomass, and spawning biomass Exploitable biomass (EBio) at the beginning of 2009 is estimated to be 325 million pounds and female spawning biomass (SBio) is estimated to be 315 million pounds. EBio is down by about 10% from the beginning of year 2008, while SBio is a bit over 3% higher than the 2008 beginning of year value estimated in the 2007 assessment. EBio and SBio are both estimated to have declined continuously between 1998 and 2007. EBio continued to decline in 2008, the model estimates that both are now on the increase, with SBio bottoming out in 2007 and EBio bottoming out in 2008. However, the 2007 assessment estimated that the low point for both was reached in 2007 and 2008 was the beginning of the turn around. This point is discussed more fully in the “Retrospective performance” section. Recruitment (measured as age-eight fi sh in the year of assessment) has varied between 8 and 40 million halibut since the 1988 year class, with a mean of 17.4 million. The 1989 to 1993 year classes, presently 15 to 19 years old and the main target of the commercial fi shery for the past several years, are all estimated to have been well below average. The sharply declining biomass over the past decade has resulted from these small year classes replacing earlier year classes that were much larger, especially the 1987 and 1988 year classes. A hopeful sign, and the explanation for the projected increase in 2009 biomasses, is the estimation that the 1998, 1999 and 2000 year classes all appear well above average. The extent to which these year classes will contribute to EBio over the next few years depends on the growth rate which, as has been frequently noted, continues to decline. Figure 9 (top panels) illustrates estimated recruitment and biomass trends since 1996.

Estimates of uncertainty There are a number of ways of estimating the uncertainty associated with a given model fi t and biomass estimate. They are all unsatisfactory in that they are conditioned on the correctness of the model, and in fact it is the choice of one model rather than another that is the major source of uncertainty in assessments. This is well illustrated by the difference in area-specifi c biomass

87 2009 IPHC ANNUAL MEETING HANDOUT estimates between the coastwide and closed-area fi ts of the IPHC model. One standard method of illustrating uncertainty around an estimate, for a given model, is the likelihood profi le. The bottom panels in Figure 9 show the likelihood profi le for both the exploitable biomass as well as the female spawning biomass. The 95% confi dence interval (C.I.) for EBio is 286 to 368 million pounds, while the 95% C.I. for the female spawning biomass is 274 to 359 million pounds. Confi dence intervals for the recruitment estimates were also computed and are plotted with the recruitment estimates (Fig. 9, top panel).

Retrospective performance Each year’s model fi t estimates the abundance and other parameters for all years in the data series. One hopes that the present assessment will closely match the biomass trajectory estimated by the previous year’s assessment. To the extent that it does not, the assessment is said to have poor retrospective performance. Our assessment has not tracked very well for the last few years. Each year the assessment has revised downward the previous year’s biomass estimates (Fig. 10a), meaning that biomass was overestimated then and may be overestimated now if the cause of the retrospective problem lies somewhere within the model. There is some precedent for that; the assessment models in use in the mid 1990s and the early 2000s showed strong retrospective patterns that turned out to be the result of misspecifi ed selectivity (age- rather than length-based). There is also the possibility that the retrospective pattern is caused in some way by the external estimation of the sex composition of the commercial catch, or by the internal prediction of surface age compositions prior to 2002 through the application of an age misclassifi cation matrix (Clark and Hare 2006). Problems of this sort with the assessment machinery would manifest themselves as systematic revisions of the estimated relative strength of the year-classes present in the stock. That was true of the retrospective patterns caused by the misspecifi cation of selectivity in the past: incoming year-classes would at fi rst be estimated as weak because catch rates were low, but the real reason was low selectivity rather than low abundance. When they were later caught in large numbers, the estimates of relative year-class strength increased. The retrospective estimates of year class strength as plotted in Figure 10b. There is some evidence of a systematic revision of estimates of year class strength as the 1994 through 1998 year class have all trended downward for the last fi ve assessments. The pattern does not hold for the 1999 year class strength estimates. We also note that the retrospective pattern has changed this year compared to the past several years. The 2008 EBio trajectory essentially overlays the 2007 EBio trajectory, with the exception of the 2007 estimate which again showed a decline. Also, the span of the revised estimates has narrowed. The difference between the 2005 EBio, as estimated using data up to 2004, and the 2008 assessment estimate of the 2005 EBio differ by just 15%, which is generally within the error range of a good stock assessment. Causes of retrospective behavior are notoriously diffi cult to diagnose. In the case of halibut it appears to result from lower CPUE rates than expected, given the estimated mortality rate. This could be due, for example, to a trend in natural (or undocumented fi shing) mortality, or a trend in catchability. The catchability explanation is unlikely, however, given that a model which permits catchability to show a trend produces assessment estimates that differ little from models with tightly constrained catchability. To summarize, there is ongoing retrospective behavior in the halibut assessment. The magnitude of the behavior is relatively small and the trend of successively lowering all earlier EBio estimates essentially ended this year. We do not feel the retrospective

88 2009 IPHC ANNUAL MEETING HANDOUT behavior weakens the assessment in any way, and analyses of the recognized patterns will be ongoing.

Harvest policy, status relative to reference points and biomass projections The IPHC has developed, refi ned and utilized a constant harvest rate policy since the 1980’s. The policy was fully described in Clark and Hare (2006) and further modifi ed as described in Hare and Clark (2008). Stated succinctly, the policy is to harvest 20% of the coastwide exploitable biomass when the spawning biomass is estimated to be above 30% of the unfi shed level. The harvest rate is linearly decreased towards a rate of zero as the spawning biomass approaches 20% of the unfi shed level. This combination of harvest rate and precautionary levels of biomass protection have, in simulation studies, provided a large fraction of maximum available yield while minimizing risk to the spawning biomass. Since the early 2000s, and in common with many fi sheries management agencies, the harvest policy has incorporated a measure designed to avoid rapid increases or decreases in catch limits, which can arise from a variety of factors including true changes in stock level as well as perceived changes resulting from changes in the assessment model. The adjustment, termed “Slow Up Fast Down” (SUFD) results in a target harvest rate of 20% but a realized rate usually a bit different. The SUFD approach is somewhat different from other agencies in that it is asymmetric around the target value, i.e., the catch limit responds more strongly to estimated decreases in biomass than to estimated increases. This occurs for two reasons: fi rst, the assessment generally has a better information base for estimating decreasing biomass compared with increasing biomass; and second, such an asymmetric policy follows the Precautionary Approach.

The unfi shed female spawning biomass (Bunfi shed) is computed by multiplying spawning biomass per recruit (SBR, from an unproductive regime) and average coastwide age-six recruitment (excluding the four most recent years). The SBR value, computed from Area 2B/2C/3A size at age data from the 1960s and 1970s is 118.5 lbs per age-six recruit. Average coastwide recruitment

for the 1990-1999 year classes (computed at age-six) is 23.3 million. This gives a Bunfi shed of 878 million pounds, a B20 of 176 million, a B30 of 263 million pounds, and the 2009 female spawning biomass value of 315 million pounds establishes Bcurrent as 35% of Bunfi shed (Fig. 11), down from the

2008 beginning of year estimate of Bcurrent of 40%. The revised trajectory of SBio suggests that the

female spawning biomass has been very close to the B30 level, the point at which the harvest rate would start being curtailed. On an annually estimated basis, however, the stock has not been that low; it is only retrospectively that we estimate the spawning biomass to have gotten so close to the reference point threshold. One problem with this method of establishing reference points is that the threshold and limit are dynamic, changing each year as the estimate of average recruitment changes. In this year’s calculation the very strong 1999 year class was included among the year

classes used to compute average recruitment, hence Bunfi shed increased from the 2008 estimate of

748 million pounds to this year’s estimate of 878 million pounds. The corresponding B20 and B30 values also increased, thus even though SBio is estimated to have increased between 2008 and

2009, the Bcurrent value declined. This situation will exacerbate next year if the 2000 year class, which presently appears to be almost as large as the 1999 year class enters the calculation. This seems paradoxical that an increasing SBio appears to be dropping closer to the reference point threshold. One solution to this paradox is to use a fi xed set of year classes to estimate average

89 2009 IPHC ANNUAL MEETING HANDOUT recruitment, in the same way that SBR is computed from a set of size at age estimates. Staff will explore modifi cations to the determination of reference points in the next year. In addition to monitoring the status of the female spawning biomass relative to reference points, success at achieving the target harvest rate is also documented (Fig. 12). The target harvest rate over the past decade for halibut has generally been 0.20. Exceptions include a briefl y increased rate to 0.225 and 0.25 between 2004 and 2006, and a lower rate of 0.15 in Areas 4B and 4CDE. On a coastwide basis, however, recent realized harvest rates have hovered around 0.25. A sizable portion of this above target harvest rate comes from the retrospective revision of exploitable biomass estimates. Thus, while the intended target rate has been around 0.20, with catch limits based on such a rate, a retrospective revision of exploitable biomass, when combined with unchanged estimates of total removals generates the higher estimated harvest rates. A smaller portion of the above target results from the SUFD adjustment which prevents catch limits dropping fully to the target level indicated by contemporary estimates of exploitable biomass. Estimates of realized harvest rate among individual regulatory areas require use of an apportionment method. The staff favors use of bottom area-weighted survey CPUE adjusted for hook competition (discussed below). Using this apportionment method, regulatory area realized harvest rates are illustrated in Figure 13. Realized harvest rates are estimated to be at, or above target in Area 4 (where target harvest rate is 0.15), at target in Area 3, and substantially above target in Area 2. The annual stock assessment produces an estimate of the total number of male and female halibut, ages 6 and older, in the ocean (Fig. 14). With this set of numbers and assuming that life history parameters, such as size at age and maturity at age, remain close to what they are today, we can make biomass and yield projections for several years into the future. Because the age range of halibut in the catch is generally in the 10-20 year old range, estimates of recruitment – which are often imprecise – do not much infl uence the projections. The time series of abundance shown in Figure 14 illustrate the strength of the celebrated 1987, and to a lesser extent 1988, year classes. As was true last year, the current assessment suggests that two extremely large year classes – 1999 and 2000 – are poised to enter the exploitable biomass over the next few years. Presently, both year classes look to be larger – in terms of numbers – than the 1987 and 1988 year classes. However, it is important to note that size at age is much smaller now than it was 20 years ago. This has two important ramifi cations – fi rst it means that the 1999 and 2000 year classes are only just beginning to reach the exploitable size range and, therefore, their true numbers in the population are still quite uncertain. Secondly, it also means that for a given number of halibut, their collective biomass will be lower. Currently, a large fraction of males never reach the minimum size limit and thus never enter the exploitable biomass. It remains to be seen just how these year classes will develop. If we assume that size at age remains at the values seen this year, then the projections for both the exploitable biomass and spawning biomass are very optimistic (Fig. 15) and indicate that the declines we have seen over the past decade are on the verge of reversing. It important to note that total removals should still remain at around 20% of the exploitable biomass and not be kept high in anticipation of future increases. As happened in the mid 1990s, when the biomass rises, higher catch limits will follow.

Apportioning the coastwide biomass among regulatory areas The staff believes that survey CPUE-based apportionment is the most objective and consistent method of estimating the biomass distribution among areas and therefore the best distribution

90 2009 IPHC ANNUAL MEETING HANDOUT of total CEY, if the aim is proportional harvest. The validity of the survey CPUE apportioning requires that survey catchability – the relationship between density and CPUE – be roughly equal among areas. In 2007, several checks for area differences in catchability were made (Clark 2008a, Clark 2008b, Clark 2008c, Webster 2009b) but little compelling evidence suggesting signifi cant differences was found. The exception was in Area 2A where a preliminary analysis suggested that uneven station distribution, in relation to bottom depth, resulted in a 40% lower catchability. The other factor that indicated potential area differences concerned hook competition and whether areas had different catchabilities as a result of fewer baited hooks being available to halibut. Both of those factors have been reconsidered for this year.

Station depth distribution The IPHC survey stations are set on a 10-nmi grid between the depths of 20 and 275 fathoms. Ideally, such an arrangement should lead to stations having the same physical and oceanic characteristics as the entire bottom area within each regulatory area. As CPUE is affected by a myriad of factors that vary with depth, a simple mean CPUE computed from all stations should be the same as one computed from a depth weighted CPUE. Figure 16 illustrates how closely survey station depths relate to the cumulative bottom depth distribution. With the exception of Area 4B where survey stations are disproportionally deep, station depth distribution closely matches bottom depth distribution. Minor differences are also noted in Area 2C, which has a slight surplus of deep stations and Area 4A which has a slight surplus of shallow stations. Survey stations were stratifi ed by depth interval and mean CPUE values were computed for each interval. These depth- stratifi ed CPUEs were weighted by the amount of bottom area to compute a depth stratifi ed mean CPUE (Fig. 17). In computing the stratifi ed means, it was necessary to fi nd depth ranges such that adequate numbers of stations contributed to the mean calculation, otherwise a biased computation could occur from undue infl uence of a small number of stations. In fact, this is what occurred in Area 2A when depth stratifi ed means were computed. This year, the depth intervals were chosen such that 10 stations were included in each depth stratum. The resultant depth stratifi ed means are very close to the simple survey means. The largest difference is in Area 4B but the difference is not statistically meaningful. Thus, for 2008, no depth correction is made to the survey CPUE.

Hook competition Catchability of halibut is affected by the presence of other bait takers, a process known as hook competition. If the average number of baits available to halibut varies substantially among regions, this would be reason to adjust survey CPUE. An analytical method for determining the level of hook competition and a correction factor for such competition was presented in 2007 (Clark 2008a). The following section is reprinted from Clark (2008a):

Mathematically the process of baits being removed from a longline by different species is the same as the process of fi sh being removed from a population by different fi sheries and natural predators. We can represent each kind of bait taker as removing a certain proportion

of the baits per unit time, so that the number of baits Bi taken by a given species i during a soak time T is given by the familiar catch equation:

BFBii=⋅⋅−0 ()1exp() −⋅ ZTZ

where Fi is the instantaneous rate of bait removal by species i, B0 is the initial

91 2009 IPHC ANNUAL MEETING HANDOUT number of baited hooks, and ZF= ∑ j is the sum of the instantaneous rates applied by all bait takers. j The instantaneous rate of bait removal by halibut can be taken to be proportional to the local density of halibut, and depending on size and gear selectivity some proportion of halibut that take a bait will also be hooked and caught, so the catch per skate of halibut

Ch will be proportional to the density of halibut Dh multiplied by the last term in the bait removal equation:

CkBkFBhhh=⋅ =⋅ ⋅00 ⋅−()1exp() − ZZkDB =′ ⋅ h ⋅ ⋅−() 1exp() − ZZ

where k and k′ are constants of proportionality. In this equation, ()1exp−−()Z is the fraction of baits removed by all takers during the active period, and ()1exp−−()ZZ is the average number of baits remaining over the course of the active period as a proportion of the initial number. If this term is the same in all areas, then survey CPUE is a consistent index of density across areas. Otherwise survey CPUE does not index density consistently across areas. Equivalently, if the fraction of baits taken is the same in all areas, then survey CPUE is a consistent index of density. It is interesting to note that the effect of hook competition on the comparability of survey CPUE is wholly determined by the total bait removal rate Z. The species composition of the bait takers makes no difference. If 80% of the baits are taken in both Area X and Area Y (meaning that Z is the same), and the catch in Area X is all halibut and the catch in Area Y is half halibut and half dogfi sh, the survey CPUEs of halibut in the two areas will accurately refl ect the relative densities of halibut.

Figure 18 shows hook occupancy rates for years 2006-2008. The catch rate (hook occupancy) varies widely for different species among the areas. The important rate however is the number of baits remaining. It is this amount, and assuming an instantaneous rate of removal, that determines average number of baits available to halibut. Areas where the number of baits remaining is higher than the Coastwide total have higher catchability while areas with fewer baits remaining have lower catchability. A hook competition correction factor is computed by dividing the coastwide value of average baits (1-exp(-Z))/Z by the area-specifi c value of average baits. Thus lower catchability will result in a correction factor greater than 1 (survey CPUE is increased) while higher catchability has the opposite result. Figure 19 shows the range of hook correction factors by area from 1996 to 2008. Areas 2A, 4B, and 4D are signifi cantly different than 1.0 while the other regions range slightly above and below 1.0. For this year, staff recommends adopting the hook correction factor as a means of adjusting survey CPUE within each regulatory area. A running three-year mean is used so that trends in competition can be tracked. The correction factors used for weighting survey CPUE in 2008 are listed in Table 3.

92 2009 IPHC ANNUAL MEETING HANDOUT Methods of apportioning biomass Last year, staff recommended apportioning the coastwide biomass using area weighted survey CPUE. This year, staff recommends the same method though with a hook competition correction factor applied. The staff examined several candidate methods, including those brought forward in various meetings, as well as via email, for apportioning the biomass and determining Total and Fishery CEY using these alternative methods. The full complement of apportionment methods for which staff compiled CEY estimates are as follows: 1. Survey CPUE x Bottom Area. This method uses a three-year average of survey CPUE multiplied by bottom area to develop an index of relative abundance. Each area’s portion of the coastwide biomass is its index divided by the coastwide sum of the indices. 2. Survey CPUE x Bottom Area, hook competition correction applied. Same as above but regulatory area survey CPUE average is multiplied by the hook competition correction factor listed in Table 3. 3. 2008 Closed-Area Assessment proportions applied to Coastwide Total EBio. The relative area abundances as computed in the closed-area assessment are applied to the coastwide estimate of Exploitable Biomass. Relative abundance estimate for Area 2A leverages Area 2B using relative survey CPUE while Area 4CDE biomass is computed using NMFS swept area estimates of abundance. 4. 2008 Closed-Area Assessment proportions applied to Closed Area Total EBio. This is the method used up until 2006 and is the only method that doesn’t use the Coastwide Total of EBio. 5. Relative Proportion of age-eight halibut as estimated in the closed area assessments. The logic is that this represents numbers of fi sh that would have eventually ended up in each area even though they may have been elsewhere at age-eight. 6. Share of Total Removals (3-year average). This method averages removals by area for the past three years and each regulatory area’s biomass is average removals divided by coastwide average removals. 7. Share of Total Removals (10-year average). Same as above but using a 10-year average. 8. Share of Total Removals (15-year average). Same as above but using a 15-year average. 9. Share of Bottom Area. Bottom area is computed for each regulatory area (0-300 fathoms) and biomass is apportioned according to each area’s share of bottom area. This method excludes the EBS outside of Area 4C. 10. Commercial CPUE x Bottom Area. Same as method 1, but using commercial CPUE instead of survey CPUE.

Area-apportioned biomass, total and fi shery constant exploitation yields With the 10 different methods of apportioning biomass, 10 sets of area-apportioned exploitable biomass, total and fi shery CEY can be computed. All of the methods utilize the same table of Other Removals – deducted from Total CEY to obtain Fishery CEY. The Other Removals are listed in Table 4. The staff recommended method of apportioning biomass, Method 2 – survey CPUE,

93 2009 IPHC ANNUAL MEETING HANDOUT adjusted for hook competition and area-weighted leads to the area-specifi c Exploitable Biomass, Total and Fishery CEY values listed in Table 5. For comparison purposes, the corresponding 2007 estimates are shown in Table 6. There are two differences between 2007 and 2008 – no hook competition correction was used in 2007, though a depth correction was applied to Area 2A and which has now been removed. Also, the recommended target harvest rate for Area 4A has been lowered from 0.20 to 0.15. The reasons for this recommendation are discussed in the Area Summary for 4A. The area shares for each of the 10 apportionment methods are listed in Table 7. The EBio totals for each area are listed in Table 8, Total CEYs are listed in Table 9, and Fishery CEYs are listed in Table 10. The target harvest rates used to compute Fishery CEYs are 0.20 for Areas 2 and 3 and 0.15 for all of Area 4. Within the tables, apportionment method No. 4, which solely relies upon the closed area assessments, has a different EBio (and 28 million pounds higher) total that the other 9 methods.

Area summaries The coastwide assessment indicates that the exploitable biomass of halibut has declined approximately 50% over the past decade. This declining trend is seen almost all of the area-specifi c survey and commercial CPUE indices. But the breadth and reasons behind the declines vary by area. The following is a region by region discussion of the trends and grouping of diagnostic plots to assess the past and present removals, stock trends, and prospects for each area. For each of the areas, six plots are illustrated. These include the following: 1. Total removals – illustrated by category (commercial catch, sport, etc.) 2. Sublegal bycatch – An estimate of lost commercial yield due to sublegal bycatch is also given. Note that the lost yield from bycatch in any given year is an estimate of future lost yield summed across several years. Methodology for estimating sublegal bycatch, lost production and computing surplus production are in the process of being documented (Hare, in prep.). 3. Surplus production. Stated simply, surplus production is the amount of total catch that, when taken exactly, keeps the exploitable biomass at the same level from one year to the next. If the biomass increases, then total catch (termed “removals”) was less than surplus production. If the biomass declines, then removals were greater than surplus production. Long term declines in biomass result from removals exceeding surplus production; stock building results from taking less than surplus production. 4. CPUE and effort – Long term trends in commercial fi shing effort and CPUE. 5. Abundance indices – these include survey CPUE, Coastwide assessment with survey partitioning and closed area assessments. 6. 2008 age structure obtained from survey samples.

Taken in total, these indicators convey a comprehensive picture for each area and serve as a helpful reference when discussing each regulatory area.

Area 2 Area 2A, 2B and 2C indices are illustrated in Figures 20, 21 and 22, respectively. Between 1997 and 2006, total removals were stable in all three areas, averaging 1.6 million pounds in Area

94 2009 IPHC ANNUAL MEETING HANDOUT 2A, 13.5 million pounds in Area 2B and 12.4 million pounds in Area 2C. Removals declined sharply in 2007 and 2008, in response to the revised view of relative halibut abundance in Area 2. Sublegal bycatch, and subsequent lost yield to the sport and commercial fi sheries, is estimated to be rather low, though legal-sized bycatch in Area 2A still represents a sizable portion of total removals. Surplus production estimates suggest that removals have exceeded surplus production in Area 2 for most of the past decade. Commercial effort has steadily increased in Area 2A for almost a decade but was relatively level in Areas 2B and 2C, and in fact declined over the past two years. Indices of abundance all suggest a steady decline in biomass in all three areas, though the Area 2B survey setline CPUE increased nearly 50% in 2008. All three areas saw declines of more than 50% in survey CPUE between 1996 and 2007, and declines continued for 2A and 2C. As is the case with the coastwide estimate of abundance, a small increase in EBio is projected for the beginning of 2009. The age structure of fi sh caught in Area 2 is noticeably younger than in Areas 3 and 4. Mean age is around 11 years of age, with little difference between males and females. All the indices are consistent with a picture of a steadily declining exploitable biomass in Area 2. The reasons for the decline are likely twofold. The fi rst is the passing through of the two very large year classes of 1987 and 1988. Every assessment over the past decade has shown that those two year classes were very strong in comparison to the surrounding year classes. Now that those two year classes are 20 years old, their contribution to the exploitable biomass and catches has sharply declined and the drop in biomass is to be expected as they are replaced by year classes of lesser magnitude. Removals have been generally larger than surplus production and this prevents rebuilding of regional stocks. Our present view of Area 2 is that harvest rates have been much higher than the target rate of 0.20 over the past decade and are not sustainable, particularly with the passage of the 1987 and 1988 year classes. There are signs that two or three large year classes are set to enter the exploitable biomass, however, the exploitable biomass will not increase as long as harvest rates remain high. Finally, Area 2 presently accounts for 28% of total removals coastwide but contributes just 17% to the female spawning biomass, a byproduct of the young age of halibut in this Area.

Area 3 Area 3A and 3B indices are illustrated in Figures 23 and 24 respectively. While these two areas occupy the central area of distribution of the halibut stock at present, they have substantially different exploitation histories over the past 10-20 years. Area 3A removals, both the total as well as the individual components (commercial, sport, bycatch) have been very stable over the past 10 years. Commercial effort has also seen relatively little variation. The CPUE indices show a slow decline with a drop of 20% in the commercial and 33% in the survey between 1998 and 2008. Removals have been very close to estimated surplus production when averaged over the past seven years, although there has been large annual variation in the proportion of the surplus production removed. The coastwide assessment estimates a decline of 16% in the EBio over the past 10 years. Area 3B saw a large increase in removals beginning in 1996 which peaked in 2002 and has dropped sharply since. Commercial fi shing effort more than tripled in the seven years after 1996 and then declined modestly over the past four years before increasing again in 2008. We estimate that removals greatly exceeded surplus production between 1998 and at least 2006. Commercial and survey CPUE both dropped by a bit more than 50% between 1998 and 2008. The coastwide assessment suggests biomass dropped by 55% between 1998 and 2008. Area 3A has a much broader spectrum of ages in the population than is seen in Area 2. Average age for females

95 2009 IPHC ANNUAL MEETING HANDOUT in survey catches is 13 and for males is 16 years of age. Area 3B, however, is more similar to Area 2 in age distribution than to Area 3A. Area 3A has the appearance of being the most stable of the IPHC regulatory areas. The area has been fully exploited for many decades and there is a wealth of data detailing its population dynamics. The area also sits at the current center of halibut distribution and it appears that emigration is roughly equal to immigration resulting in an effectively closed population. Like Area 2, Area 3A benefi ted from the very large year classes of 1987 and 1988 and the slow decline in exploitable biomass is the result of those year classes dying off. The biomass remains in a healthy state and should continue to support removals of the magnitude seen over the past 2-3 decades. The situation in Area 3B is different. Area 3B was relatively lightly fi shed until the mid 1990s. With the introduction of a regular survey, quotas were incrementally increased from 4 million pounds to a high of 17 million pounds. Predictably, catch rates declined steadily. Our view of Area 3B is that the area had an accumulated “surplus” biomass that could be (and was) taken but the level of catches was not sustainable. The area has now been fi shed down and the average annual yield will be somewhere in between the low levels of the mid 1990s and the high levels of 5-6 years ago. As the area is also centrally located, we apply the dynamics of Areas 2 and 3A and believe that a constant harvest rate of 0.20 is appropriate for the region. The coastwide assessment suggests that harvests have been in the 0.15 to 0.20 range over the past six years.

Area 4 Area 4A, 4B and 4CDE indices are illustrated in Figures 25, 26 and 27, respectively. These areas have roughly similar commercial exploitation histories over the past decade and show similar trends. In these areas, commercial catches increased from around 1.5 million pounds to around 4-5 million pounds between 1996 and 2001. Catches have since declined, most strongly in Areas 4B and 4CDE where a lower target harvest rate of 0.15 was applied the past few years. Commercial effort mirrored the rise in removals from 1996-2001, however the drop in effort was not nearly as sharp as the drop in catches, and the drop in commercial CPUE is evident in the time series. Survey CPUE in Area 4A has declined around 70% over the past decade while Area 4B is down 50% over the same time period; the decline in Area 4D survey CPUE is around 40% (there is no survey index for 4C or 4E). The coastwide assessment indicates an exploitable biomass decline of 61% for Area 4A, 68% for Area 4B, and 43% for Area 4CDE. The situation in Area 4 is somewhat like Area 3B only more exaggerated. Area 4 was very lightly exploited up until the mid 1990s. With the onset of surveys, quotas were quickly increased and the accumulated surplus biomass quickly removed. Catches of 4-5 million pounds in each area are clearly not sustainable, as was stated by the IPHC staff when higher catch limits were recommended. In Area 4B, where catch limits were dropped most strongly, there is evidence of a reversal in the strong biomass decline. Over the past three years, the CPUE indices have actually increased slightly and the two assessments estimate a level time trend in exploitable biomass. The target harvest rate was reduced to 0.15 in Area 4CDE in 2004 and in Area 4B in 2005. While Area 4CDE still shows continuing signs of decline, the situation in Area 4B is much more promising. The Area 4B survey CPUE increased for the fourth consecutive year and total removals now appear to be less than surplus production. This year, staff is recommending lowering the target harvest rate for Area 4A to 0.15, in line with the rest of Area 4. Sublegal bycatch remains very large relative to removals and lost annual yield to the commercial fi shery is on the order of 1.5 million pounds. Additionally, Area 4A is a

96 2009 IPHC ANNUAL MEETING HANDOUT net exporter of fi sh, likely receiving little immigration from the rest of Area 4 while emigration has been seen to be quite large (Webster 2009). Yield per recruit calculations for Area 4A, based on estimated average recruitment suggest sustainable yield is no greater than 3 million pounds; an F40 harvest policy for Area 4A gives a recommended harvest rate of 0.15. All of these factors together suggest that removals continue to be too high in Area 4A and a lower target harvest rate is required. The hope is that Area 4A will respond as Area 4B has and the stock will curtail its steep decline and begin to increase, perhaps with assistance from the anticipated large 1999 and 2000 year classes and removals will then increase commensurately.

Acknowledgements We wish to acknowledge the many samplers, age readers, data entry personnel and other IPHC staff who are responsible for collecting and quality control checking the data upon which the halibut assessment depends so strongly. A great deal of effort is expended on both on the setline survey as well as in the port sampling programs and the assessment staff appreciates the time constraints involved in having the data available days after the fi shery ends, in time for the annual stock assessment.

References Clark, W. G. 2008a. Effect of station depth distribution on survey CPUE. Int. Pac. Halibut Comm. Report of Research and Assessment Activities 2007: 205-210

Clark, W. G. 2008b. Effect of hook competition on survey CPUE. Int. Pac. Halibut Comm. Report of Research and Assessment Activities 2007: 211-215.

Clark, W. G. 2008c. Comparison of setline and trawl survey catch rates in different areas. Int. Pac. Halibut Comm. Report of Research and Assessment Activities 2007: 221-228.

Clark, W. G., and Hare, S. R. 2006. Assessment and management of Pacifi c halibut: data, methods, and policy. Int. Pac. Halibut Comm. Sci. Rep. 83.

Clark, W. G., and Hare, S. R. 2007. Motivation and plan for a coastwide stock assessment. Int. Pac. Halibut Comm. Report of Assessment and Research Activities 2006: 83-96.

Clark, W. G., and Hare, S. R. 2008. Assessment of the Pacifi c halibut stock at the end of 2007. Int. Pac. Halibut Comm. Report of Research and Assessment Activities 2007: 177-203.

Hare, S. R. In prep. Estimates of halibut surplus production and lost production.

Hare, S. R. 2008. New estimates of bottom area in IPHC regulatory areas. Int. Pac. Halibut Comm. Report of Assessmnent and Research Activities 2007: 297-307.

Hare, S. R. and Clark, W. G. 2007 IPHC harvest policy analysis: past, present, and future considerations. Int. Pac. Halibut Comm. Report of Assessmnent and Research Activities 2007: 275-295.

97 2009 IPHC ANNUAL MEETING HANDOUT Webster, R. 2008. Analysis of PIT tag recoveries through 2007. Int. Pac. Halibut Comm. Report of Assessment and Research Activities 2007: 259-274.

Webster, R. 2009a. Analysis of PIT tag recoveries through 2008. Int. Pac. Halibut Comm. Report of Assessment and Research Activities 2008:213-220.

Webster, R. 2009b. Further examination of fi sheries-survey interactions.. Int. Pac. Halibut Comm. Report of Assessment and Research Activities 2008: 203-212.

Webster, R., and Clark, W. G. 2007. Analysis of PIT tag recoveries through 2006. Int. Pac. Halibut Comm. Report of Assessment and Research Activities 2006:129-138.

98 2009 IPHC ANNUAL MEETING HANDOUT Table 1. Alternative coastwide model fi ts. The AIC value is relative units compared to the model with the lowest AIC scare. Number of Exploitable Model parameters Deviance Δ AIC biomass Survey q constant 153 534 + 82 320 Survey q drift 164 512 0 322 Survey q trendless drift 164 513 + 3 325 No fi t to commercial CPUE 153 530 + 65 316

Table 2. Effect of the 2008 data on closed-area and coastwide abundance estimates. 2008 ebio 2008 ebio 2008 ebio 2009 ebio 2007 assessment 2007 assessment 2008 assessment 2008 assessment Area Data as of 11/07 Data as of 11/08 Data as of 11/08 Data as of 11/08 Coastwide assessment: 361 360 290 325

Table 3. Hook correction factors applied to survey CPUE in partitioning coastwide biomass among regulatory areas. The factors represent 2006-2008 hook occupancy data. 2A 2B 2C 3A 3B 4A 4B 4CDE 1.112 1.009 1.050 1.048 1.087 1.024 0.845 0.732

99 2009 IPHC ANNUAL MEETING HANDOUT gures for Areas 2C and 3A are actual catches not GHL levels as in Table 5. Table levels as in actual catches not GHL are 2C and 3A Areas for gures fi 0.4570.1410.030 1.5200.001 0.0670.629 0.405 3.083 0.023 0.2160.142 2.015 0.525 5.629 0.012 1.058 0.495 3.836 0.372 0.0180.015 0.063 0.485 3.836 7.122 0.048 0.043 0.262 0.004 0.496 7.122 0.555 0.0150.157 0.000 0.212 0.012 0.211 0.555 0.566 0.002 0.064 0.924 0.012 0.000 0.566 0.225 0.128 1.552 10.750 0.681 0.092 0.225 4.226 1.905 0.014 1.489 0.133 1.658 0.141 0.852 16.606 1.658 0.019 14.599 0.814 0.176 0.091 2.337 2.337 6.432 Area 2AArea 2B Area 2C Area 3A Area 3B Area 4A Area 4B Area 4CDE Area Total Sport catch Legal-sized bycatch Personal use Legal-sized wastage Total excl.sport catch Total and 2B 2A Areas in Sublegal discard mortality information; (shown for not taken off total CEY) Sublegal bycatch mortality information; (shown for Not taken off total CEY) Table 4. Other removals in detail. Sport catch removals 4. Other Table

100 2009 IPHC ANNUAL MEETING HANDOUT nd 2B sport catch is included nd 2B sport catch is included 0.53 0.75 0.33 2.01 13.68 3 6.71 3 t with survey apportionment of the total biomass estimate among regulatory areas, and adjusted by t with survey apportionment of the total biomass estimate among regulatory areas. “Area fi fi 4.7 25.6 32.5 144.8 74.0 21.3 20.2 37.9 361 3.2 27.0 27.9 140.0 68.8 18.5 15.4 24.2 325 0.290.65 0.471.22 4.65 2.59 9.00 3.92 6.21 22.25 24.22 14.27 10.90 3.51 3.10 2.71 1.86 3.68 55.64 3.89 60.40 ts. 0.1420.500 0.495 4.919 2.710 2.864 7.169 20.839 0.555 13.202 0.566 2.204 0.225 2.085 1.658 1.966 13.520 48.579 fi Area 2AArea 2B Area 2C Area 3A Area 3B Area 4A Area 4B Area 4CDE Area Total Area 2AArea 2B Area 2C Area 3A Area 3B Area 4A Area 4B Area 4CDE Area Total 1 1

2 2 2, 3 2 shery CEY CEY shery shery CEY CEY shery exploitable biomass fi fi exploitable biomass

2008 2008 2009 2008 catch limit shery CEY rather than in other removals. shery CEY rather than in other removals. shery CEY 2009 fi fi

Proportion of total harvest rateTarget CEYTotal Other removals 0.010 0.20 0.083 0.20 0.086 0.642 0.431 0.20 5.414 0.212 5.574 0.20 0.057 28.008 0.20 13.757 Proportion of total 0.047 harvest rateTarget 0.15 CEYTotal 2.770 Other removals 0.074 2.310 0.15 0.013 1.000 0.20 0.071 3.624 0.15 62.099 0.20 0.090 <0.20 0.94 0.401 0.20 5.12 0.205 0.20 6.50 0.059 0.20 0.056 28.96 0.20 14.80 0.105 0.15 4.26 1.000 3.03 0.15 <0.20 5.69 69.30

Coastwide assessment Coastwide assessment

“Coastwide assessment” refers to the coastwide model a Areas 2A “Other removals” comprise legal-sized wastage, bycatch, personal use, and in most areas sport catch. In Area 3A. Area 2C and 3.650 M lbs. in of 0.788 M lbs. in Assumes GHL “Coastwide assessment” refers to the coastwide model a Areas 2A “Other removals” comprise legal-sized wastage, bycatch, personal use, and in most areas sport catch. In in in estimated hook competition correction factors. assessments” are the closed-area model Table 5. Estimates of 2009 exploitable biomass and CEY from the 2008 assessment from 5. Estimates of 2009 exploitable biomass and CEY Table Notes: 1 2 3 the 2007 assessment (2007 RARA, p. 185). from 6. Estimates of 2008 exploitable biomass and CEY Table Notes: 1 2

101 2009 IPHC ANNUAL MEETING HANDOUT 0.009 0.0830.010 0.082 0.0830.012 0.415 0.086 0.1140.012 0.196 0.431 0.123 0.114 0.056 0.212 0.4490.013 0.123 0.057 0.057 0.107 0.118 0.4490.018 0.047 0.027 0.106 0.107 0.102 0.1490.018 0.063 0.483 1.000 0.027 0.074 0.151 0.1510.018 0.122 1.000 0.063 0.104 0.408 0.141 0.1570.066 0.049 1.000 0.135 0.357 0.148 0.1590.028 0.024 0.104 0.047 0.169 0.371 0.082 0.123 1.000 0.023 0.056 0.148 0.256 0.085 0.055 1.000 0.039 0.054 0.154 0.401 0.070 0.038 0.094 0.162 1.000 0.069 0.078 0.082 1.000 0.067 0.069 1.000 0.111 1.000 0.080 1.000 RuleSurvey apportionment (CPUE x bottom area) 2A Area 2B Area 2C Area 3A Area 3B Area 4A Area 4B Area 4CDE Area Total Survey apportionment (with hook correction) 2009 exploitable biomass from assessments 2008 closed-area assessment sum) (CW 2009 exploitable biomass from assessments 2008 closed-area assessment sum) (CA from Historical recruitment assessments 2007 closed-area (1987-1996) of total removals Share avg.) (3 year of total removals Share avg.) (10 year of total removals Share avg.) (15 year of bottom 0-300 fm Share (excl. EBS shelf outside 4C) apportionment Commercial (CPUE x bottom area) Table 7. Shares of total CEY by area according to various apportionment methods. by area of total CEY 7. Shares Table

102 2009 IPHC ANNUAL MEETING HANDOUT 2.911 27.071 26.778 134.822 63.845 18.193 18.389 32.992 325.000 3.208 27.0723.975 27.868 36.966 140.041 40.053 68.7844.316 145.858 18.465 40.138 34.928 15.401 43.490 8.8514.140 158.374 20.468 24.162 38.502 37.926 325.000 34.5875.946 9.610 33.901 156.928 325.000 48.361 22.2255.933 39.662 48.916 49.0925.694 15.767 132.471 36.810 45.755 50.941 353.000 43.951 7.657 116.099 48.002 15.123 55.0058.967 120.731 27.758 7.498 18.260 40.109 48.082 325.000 12.575 17.744 17.429 22.732 130.376 12.495 325.000 22.281 52.771 325.000 26.778 21.626 325.000 22.409 25.845 325.000 21.446 51.820 26.676 83.174 50.064 30.492 25.196 36.133 325.000 RuleSurvey apportionment (CPUE x bottom area) Survey apportionment (with hook correction) 2A Area 2B Area 2C Area 3A Area 3B Area 4A Area 4B Area 4CDE Area Total 2009 exploitable biomass 2009 exploitable biomass from assessments 2008 closed-area assessment sum) (CW 2009 exploitable biomass from assessments 2008 closed-area assessment sum) (CA from Historical recruitment assessments 2007 closed-area (1987-1996) of total removals Share avg.) (3 year of total removals Share avg.) (10 year of total removals Share avg.) (15 year of bottom 0-300 fm Share (excl. EBS shelf outside 4C) apportionment Commercial (CPUE x bottom area) Table 8. Exploitable biomass by area according to various apportionment methods. 8. Exploitable biomass by area Table

103 2009 IPHC ANNUAL MEETING HANDOUT 0.580.64 5.410.79 5.41 5.36 7.39 5.57 26.960.86 8.01 28.01 12.77 8.030.83 29.17 13.76 2.73 8.70 7.701.18 6.99 2.77 2.76 31.67 6.921.18 9.60 1.33 2.31 7.59 31.391.14 4.95 9.80 9.81 3.07 1.444.29 10.18 61.52 7.93 3.62 9.16 26.541.79 10.36 62.10 3.33 9.61 2.36 5.09 23.23 8.76 8.02 5.34 24.15 61.84 10.99 1.15 2.27 5.52 16.63 3.55 9.61 2.74 67.15 1.14 10.01 26.08 4.16 2.61 1.89 62.44 10.55 4.57 1.88 3.43 4.02 3.78 3.35 62.72 62.34 3.25 3.36 5.42 62.42 60.41 3.88 61.25 RuleSurvey apportionment (CPUE x bottom area) 2A Area 2B Area 2C Area 3A Area 3B Area 4A Area 4B Area 4CDE Area Total Survey apportionment (with hook correction) 2009 exploitable biomass from assessments 2008 closed-area assessment sum) (CW 2009 exploitable biomass from assessments 2008 closed-area assessment sum) (CA from Historical recruitment assessments 2007 closed-area (1987-1996) of total removals Share avg.) (3 year of total removals Share avg.) (10 year of total removals Share avg.) (15 year of bottom 0-300 fm Share (excl. EBS shelf outside 4C) apportionment Commercial (CPUE x bottom area) Table 9. Total CEY by area according to various apportionment methods. by area CEY Total 9. Table

104 2009 IPHC ANNUAL MEETING HANDOUT 0.4400.500 4.9190.653 4.919 2.646 6.898 2.864 19.7950.721 12.214 5.301 20.839 7.533 2.163 13.202 22.0030.686 5.988 2.204 2.533 6.431 7.205 24.506 2.085 3.2911.047 0.762 4.207 7.030 1.9661.045 9.177 2.845 24.217 48.001 0.8760.997 9.323 7.073 3.427 7.377 48.579 3.1094.147 9.693 6.441 19.325 1.799 3.864 8.235 48.319 1.651 9.869 6.890 16.051 0.924 10.446 1.703 7.527 2.625 16.977 2.506 53.626 2.173 9.061 0.900 0.000 9.466 1.661 2.048 1.752 48.921 18.906 9.458 1.684 9.999 1.649 4.008 49.212 3.451 1.586 3.554 48.824 3.136 3.762 48.902 2.219 46.889 46.890 RuleSurvey apportionment (CPUE x bottom area) Survey apportionment (with hook correction) 2009 exploitable biomass from assessments 2008 closed-area assessment sum) (CW 2009 exploitable biomass from assessments 2008 closed-area 2A Area assessment sum) (CA 2B Area from Historical recruitment 2C Area assessments 2007 closed-area 3A Area 3B Area (1987-1996) 4A Area of total removals Share 4B Area avg.) (3 year 4CDE Area of total removals Share Total avg.) (10 year of total removals Share avg.) (15 year of bottom 0-300 fm Share (excl. EBS shelf outside 4C) apportionment Commercial (CPUE x bottom area) Table 10. Fishery CEY by area according to various apportionment methods. by area 10. Fishery CEY Table

105 2009 IPHC ANNUAL MEETING HANDOUT Wastage (<81 cm) Bycatch (<81 cm) 30 Wastage (81 cm+) Personal use Bycatch (81 cm+) 25 Sport Commercial 20

15 Removals (M lbs) 10

5

0 4CDE 4B 4A 3B 3A 2C 2B 2A

Figure 1. Total removals by type and regulatory area for 2008.

106 2009 IPHC ANNUAL MEETING HANDOUT Wastage (<81 cm) Bycatch (<81 cm) Wastage (81 cm+ ) Personal use Bycatch (81 cm+ ) Sport Commercial Removals (M lbs) Removals 0 20406080100 1935 1945 1955 1965 1975 1985 1995 2005

Figure 2. Total removals coastwide for the period 1935-2008.

107 2009 IPHC ANNUAL MEETING HANDOUT CPUE (lbs/skate) 2A 2B 2C 0 1020304050 0 50 100 150 0 100 200 300 400

1996 2000 2004 2008 1996 2000 2004 2008 1996 2000 2004 2008

CPUE (lbs/skate) 3A 3B 4A 0 100 200 300 300 0100 0 50 150 250

1996 2000 2004 2008 1996 2000 2004 2008 1996 2000 2004 2008

CPUE (lbs/skate) 4B 4D Total 0 50 150 250 0 50 150 250 050100150

1996 2000 2004 2008 1996 2000 2004 2008 1996 2000 2004 2008

Figure 3. Survey CPUE (weight of legal-sized halibut per standardized skate of gear) by regulatory area. The dots indicate the area-wide average; the vertical bars represent +/ 2 standard errors of the mean. The gray line is a smoother to illustrate trend; it is not an assessment model fi t to the CPUE data. The total is computed by area-weighting the individual area CPUE time series.

108 2009 IPHC ANNUAL MEETING HANDOUT 2A 2B 2C

CPUE (lbs/skate) OAIVQ JC JC JC OAIFQ 0 100 200 300 400 050100200 0 100 200 300 400

1975 1985 1995 2005 1975 1985 1995 2005 1975 1985 1995 2005

3A 3B 4A CPUE (lbs/skate) JC OAIFQ JC OAIFQ JC OAIFQ 1030500 300 0100 0 200 400 600 0 200 400 600

1975 1985 1995 2005 1975 1985 1995 2005 1975 1985 1995 2005

4B 4D Total CPUE (lbs/skate) JC OAIFQ JC OAIFQ JC OAIFQ 0 100 200 300 0 400 800 1200 0100200300400

1975 1985 1995 2005 1975 1985 1995 2005 1975 1985 1995 2005

Figure 4. Commercial CPUE by regulatory area. The dots indicate the area-wide average; the vertical bars represent +/ 2 standard errors of the mean. The gray line is a smoother to illustrate trend; it is not an assessment model fi t to the CPUE data. The total is computed by area-weighting the individual area CPUE time series. The dashed vertical lines indicate transitions between J and C hook, between open access (OA) and Individual Vessel Quotas in Area 2B, and between open access and Individual Fishing Quotas in Areas 2C, 3, and 4.

109 2009 IPHC ANNUAL MEETING HANDOUT 300 Total Biomass

Survey EBio 250 Scaled IPHC CPUE

200 30 150 25 18.1 20

Million lbs Million 100 12.8 15 8.8 10 50 5 0 0 IPHC Survey CPUE (net lbs) 1985 1990 1995 2000 2005

Figure 5. Swept-area estimates of halibut abundance from the NMFS EBS trawl survey. The red dots and error bars represent mean and 95% confi dence interval for the total abundance; the blue diamonds are error bars represent mean and 95% confi dence interval for abundance with survey selectivity applied to the total biomass (termed survey EBio). The inverted purple triangles represent the estimated density of legal-sized halibut (per standardized skate of gear) across the shelf; this index is scaled to the survey EBio trend (see text for full details).

110 2009 IPHC ANNUAL MEETING HANDOUT NMFS trawl survey (2005í07) proportion in length class length in proportion 0.00 0.10 0.20 0 20 40 60 80 100 120 140 160 180 10 cm length class (lower bound)

NMFS trawl survey (2005í07): IPHC survey selectivity proportion in length class 0.00 0.10 0.20 0 20 40 60 80 100 120 140 160 180 10 cm length class (lower bound)

IPHC shelf survey (2006) proportion length in class 0.00 0.10 0.20 0 20 40 60 80 100 120 140 160 180 10 cm length class (lower bound)

Figure 6. Comparison of NMFS trawl survey and IPHC length frequency compositions. The top panel shows the length frequency composition for all halibut caught by the NMFS trawl gear for years 2005-7. the middle panel shows the frequency distribution of lengths after the IPHC setline selectivity curve is applied to raw counts. The bottom panel illustrates the length composition of halibut in the 2006 IPHC shelf survey.

111 2009 IPHC ANNUAL MEETING HANDOUT a) Average weight trend

35 1996 av. wt. = 35.7 2008 av. wt. = 23.6 30

25

20

15 Average wt. in catch 1997 av. wt. = 21.1 Commercial Survey 10 2008 av. wt. = 15.3 1995 2000 2005 2010 b) Average age trend

14

13

12

11

Average age in catch Average Commercial Survey 10 1995 2000 2005 2010

Figure 7. Average weight (panel a) and average weight (panel b) trends for the coastwide halibut stock for 1996 to 2008.

112 2009 IPHC ANNUAL MEETING HANDOUT 18 2A 4A 18 2A 4A 2B 4B 2B 4B 2C 4D 2C 4D 3A Total 3A Total 3A 3B 3B 16 16 4D 4A 4B 4B 4D Total 3A 14 14 3B

Total 2C Average age Average Average age Average 3B 2B 12 4A 12 2A 2C 2A 2B 10 10 1998 2000 2002 2004 2006 2008 1998 2000 2002 2004 2006 2008

40 2A 4A 40 2A 4A 2B 4B 2B 4B 2C 4D 2C 4D 35 3A Total 35 3A Total 3B 3B 4B 30 30 2C

4A 25 25 4D Total 3A 4B 3B2B 20 20 2A

4D Average wt. (net lbs) Average wt.Average (net lbs) 2C 3A Total 15 15 2A2B 4A 3B 10 10 1998 2000 2002 2004 2006 2008 1998 2000 2002 2004 2006 2008

Figure 8. Trends in average age (top panels) and average weight (bottom panels) in survey catches (left panels) and commercial catches (right panels).

113 2009 IPHC ANNUAL MEETING HANDOUT Age 8 Recruitment Biomass trends 1400 40 ) 1200 ) 1000 30 800 TBio8+ EBi o 20 600 SBi o 400

10 Biomass (M lb Age 8 recruits 8 Age (M 200 0 0 1988 1992 1996 2000 1995 2000 2005 2010

CPUE in No. of halibut CPUE in Wt. of halibut

max cpue = 15. 2 max cpue = 425 15 2008 cpue = 10. 7 400 2008 cpue = 248

10 300 200 max cpue = 166 2008 cpue = 72 5 max cpue = 9. 6 2008 cpue = 6. 9 100 Commercial Commercial CPUE in legal weight legal in CPUE CPUE in total number total in CPUE 0 Survey 0 Survey 1995 2000 2005 2010 1995 2000 2005 2010

2009 EBio likelihood profile 2009 SBio likelihood profile 08 08 í í

5% C.I. 95% C.I. 5% C.I. 95% C.I. 08 2.0e 08 2.0e í í

324.6 315.1 Rel. Probability Rel. Probability 0.0e+00 1.0e 260 300 340 380 0.0e+00250 1.0e 300 350 Exploitable biomass (M lbs) Female spawning biomass (M lbs)

Figure 9. Features of the 2008 halibut coastwide assessment.

114 2009 IPHC ANNUAL MEETING HANDOUT a) Retrospective pattern in EBio

600 500 2004 M lbs 2005 í 400 20062007 300 2008 200

Biomass 100 0 1996 2000 2004 2008 Year

b) Retrospective pattern in age 8 recruitment

40 2004 2005 30 2006 2007 2008 20

10 M Age 8 recruits M 0 1988 1991 1994 1997 2000 Year class

Figure 10. Retrospective behavior of 2008 halibut assessment model. The top panel illustrates the effect on estimates of EBio by sequentially removing years of data. The bottom panel illustrates the efect on estimation of age eight recruitment. Note that the most recent year class (2001) is only estimated in the 2008 assessment, the 2000 year class in the 2007 and 2008 assessments, and so on.

115 2009 IPHC ANNUAL MEETING HANDOUT 1000 Bunfished

800

600

400 presently at B35 B30

Female SBio (M lbs) 200 B20

0 1995 2000 2005 2010

Figure 11. Status of female spawning biomass. See text for details.

116 2009 IPHC ANNUAL MEETING HANDOUT 2.0 OK, but Lower HR 1.5 Bad

20062008 2004

1.0 2002

2000 1998

1996 0.5 HR relative to target Bad Good

0.0 0.0 0.5 1.0 1.5 2.0 2.5 3.0 SBio relative to B20

Figure 12. Trend and status of halibut management relative to reference points. Horizontal

axis indicates female spawning biomass (SBio) relative to B20 (value of 1.0) and B30 (value of 1.5). Vertical axis illustrates realized harvest rate relative to a target harvest rate of 0.20 (value of 1.0) and the previous target harvest rate of 0.25 (value of 1.25).

117 2009 IPHC ANNUAL MEETING HANDOUT 1996 2003 0.5 1997 2004 1998 2005 1999 2006 0.4 2000 2007 2001 2008 2002 0.3

Harvest rate 0.2

0.1

0.0 4CDE4B 4A 3B 3A 2C 2B 2A CW

Figure 13. Summary of realized harvest rates from the coastwide assessment, using survey CPUE weighted by hook competition to partition biomass among areas. The dashed horizontal lines indicates target harvest rate by regulatory area.

118 2009 IPHC ANNUAL MEETING HANDOUT Figure 14. Coastwide population estimates of halibut numbers at age. Several large year classes are highlighted.

119 2009 IPHC ANNUAL MEETING HANDOUT 600 Exploitable biomass 500

M lbs 400 í

300 Female spawning biomass 200 Biomass

100

0 1995 2000 2005 2010 2015

Figure 15. Projected exploitable and spawning biomasses for the coastwide population of halibut.

120 2009 IPHC ANNUAL MEETING HANDOUT 1.0 1.0 1.0 2A 2B 2C 0.8 0.8 0.8

0.6 0.6 0.6

0.4 0.4 0.4

0.2 0.2 0.2

Cumulative proportion 0.0 Cumulative proportion 0.0 Cumulative proportion 0.0 50 100 150 200 250 50 100 150 200 250 50 100 150 200 250 Depth (fm) Depth (fm) Depth (fm)

1.0 1.0 1.0 3A 3B 4A 0.8 0.8 0.8

0.6 0.6 0.6

0.4 0.4 0.4

0.2 0.2 0.2

Cumulative proportion 0.0 Cumulative proportion 0.0 Cumulative proportion 0.0 50 100 150 200 250 50 100 150 200 250 50 100 150 200 250 Depth (fm) Depth (fm) Depth (fm)

1.0 1.0 4B 4D Depth distributions by area 0.8 0.8

0.6 0.6 Bottom depth distribution

0.4 0.4 Station depth distribution

0.2 0.2 20 fathom depth interval

Cumulative proportion 0.0 Cumulative proportion 0.0 50 100 150 200 250 50 100 150 200 250 Depth (fm) Depth (fm)

Figure 16. Cumulative distribution of bottom depth and survey station depth by regulatory area.

121 2009 IPHC ANNUAL MEETING HANDOUT 70 500 60 2A 2B 2C 150 400 50 40 300 100 30 200 20 50 100 CPUE (lb/skate) 10 Ratio = 1.01 +/í 0.31 Ratio = 0.94 +/í 0.12 Ratio = 0.96 +/í 0.05 0 0 0 1996 2000 2004 2008 1996 2000 2004 2008 1996 2000 2004 2008

3A 3B 4A 300 400 300

300 200 200 200 100 100 100 CPUE (lb/skate) Ratio = 1.02 +/í 0.04 Ratio = 1.01 +/í 0.05 Ratio = 0.95 +/í 0.09 0 0 0 1996 2000 2004 2008 1996 2000 2004 2008 1996 2000 2004 2008

350 4B 4D 300 300 250 Mean survey CPUE 200 200 150 Unstratified 100 100 Stratified CPUE (lb/skate) 50 Ratio = 0.92 +/í 0.15 Ratio = 1.16 +/í 0.66 0 0 1996 2000 2004 2008 1996 2000 2004 2008

Figure 17. Survey CPUE plotted as simple mean (unstratifi ed, gray dots) and depth-stratifi ed CPUE (yellow line). The errors bars are +/- two standard errors of the mean for the un- stratfi ed mean

122 2009 IPHC ANNUAL MEETING HANDOUT Bait 2A 2B 2C Halibut Dogfish Sablefish Pacific cod Other catch Skin/empty

01020 50 01020 50 01020 50

Bait 3A 3B 4A Halibut Dogfish Sablefish Pacific cod Other catch Skin/empty

01020 50 01020 50 01020 50

Bait 4B 4CDE Coastwide Halibut Dogfish Sablefish Pacific cod Other catch Skin/empty

01020 50 01020 50 01020 50 Percentage of hooks

Figure 18. Hook occupancy by regulatory area, 2006-2008 data combined.

123 2009 IPHC ANNUAL MEETING HANDOUT 2.0

1.5

1.0

CPUE correction factor 0.5

0.0 2A 2B 2C 3A 3B 4A 4B 4D All

Figure 19. Boxplot of hook competition correction factors for the period 1996-2008. Correction factors were computed for each year of survey data for a maximum of 13 values for any regulatory area.

124 2009 IPHC ANNUAL MEETING HANDOUT Female Male Total Removals ASP F avg. = 11.3 T avg. = 11.3 M avg. = 11.4 o Age Age structure 10 15 20 25 1996 2000 2004 2008

1.5 1.0 0.5 0.0

0.20 0.15 0.10 0.05 0.00

Removals or ASP (M lbs) (M ASP or Removals

Proportion EBio (M lbs) (M EBio 8 6 4 2 0 Survey CPUE CW assessment Sublegal bycatch Surplus production Lost commercial yield bycatch Sublegal Abundance indices 1985 1990 1995 2000 2005 2010 1996 2000 2004 2008

0

50 40 30 20 10

0.25 0.20 0.15 0.10 0.05 0.00 (lbs/skate) CPUE

M (lbs) M Effort (’000 Skates) (’000 Effort 10 8 6 4 2 0 Area 2A summary ) ) Comm. CPUE Comm. Effort Comm. Removals CPUE & effort Wastage (<81 cm) (<81 Wastage cm) (<81 Bycatch cm+ (81 Wastage Personal use Personal cm+ (81 Bycatch Sport Commercial 1985 1990 1995 2000 2005 2010 1974 1980 1986 1992 1998 2004 0

500 400 300 200 100

2.0 1.5 1.0 0.5 0.0 (lbs/skate) CPUE Removals (M lbs) (M Removals Figure 20. Summary of removals, production, effort, abundance indices and age structure for Area 2A. Area for effort, abundance indices and age structure production, 20. Summary of removals, Figure

125 2009 IPHC ANNUAL MEETING HANDOUT Female Male Total Removals ASP F avg. = 10.6 T avg. = 10.8 M avg. = 11.1 o Age Age structure 10 15 20 25 1996 2000 2004 2008 5 0

15 10

0.20 0.15 0.10 0.05 0.00

Removals or ASP (M lbs) (M ASP or Removals

Proportion EBio (M lbs) (M EBio 60 50 40 30 20 10 0 Survey CPUE Survey assessment CW CA assessment Sublegal bycatch Surplus production Lost commercial yield commercial Lost bycatch Sublegal Abundance indices 1985 1990 1995 2000 2005 2010 1996 2000 2004 2008 0

50

150 100

0.20 0.15 0.10 0.05 0.00 (lbs/skate) CPUE

M (lbs) M Effort (’000 Skates) (’000 Effort 120 100 80 60 40 20 0 Area 2B summary Comm. CPUE Comm. Effort Removals CPUE & effort Wastage (<81 cm) (<81 Wastage cm) (<81 Bycatch cm+) (81 Wastage use Personal (81 cm+) Bycatch Sport Commercial 1985 1990 1995 2000 2005 2010 1974 1980 1986 1992 1998 2004 0

50

5 0

300 250 200 150 100 15 10 (lbs/skate) CPUE Removals (M lbs) (M Removals Figure 21. Summary of removals, production, effort, abundance indices and age structure for Area 2B. Area for effort, abundance indices and age structure production, 21. Summary of removals, Figure

126 2009 IPHC ANNUAL MEETING HANDOUT Female Male Total Removals ASP F avg. = 11.4 T avg. = 11.4 M avg. = 11.5 o Age Age structure 10 15 20 25 1996 2000 2004 2008 5 0

15 10

0.20 0.15 0.10 0.05 0.00

Removals or ASP (M lbs) (M ASP or Removals

Proportion EBio (M lbs) (M EBio 60 40 20 0 Survey CPUE CW assessment CA assessment Sublegal bycatch Surplus production Lost commercial yield Lost commercial bycatch Sublegal Abundance indices 1985 1990 1995 2000 2005 2010 1996 2000 2004 2008 0

500 400 300 200 100 0.25 0.20 0.15 0.10 0.05 0.00 (lbs/skate) CPUE

M (lbs) M Effort (’000 Skates) (’000 Effort 70 60 50 40 30 20 10 0 Area 2C summary Comm. CPUE Comm. Effort Comm. Removals CPUE & effort Wastage (<81 cm) cm) Bycatch (<81 cm+) (81 Wastage use Personal Bycatch (81 cm+) Sport Commercial 1985 1990 1995 2000 2005 2010 1974 1980 1986 1992 1998 2004 0 5 0

15 10 400 300 200 100 CPUE (lbs/skate) CPUE Removals (M lbs) (M Removals Figure 22. Summary of removals, production, effort, abundance indices and age structure for Area 2C. Area for effort, abundance indices and age structure production, 22. Summary of removals, Figure

127 2009 IPHC ANNUAL MEETING HANDOUT Female Male Total Removals ASP F = 13.0 avg. T avg. = 14.2 T avg. M avg. =M avg. 16.0 o Age Age structure 10 15 20 25 1996 2000 2004 2008 0

40 30 20 10

0.10 0.08 0.06 0.04 0.02 0.00

Removals or ASP (M lbs) (M ASP or Removals Proportion EBio (M lbs) (M EBio 300 250 200 150 100 50 0 Survey CPUE Survey assessment CW assessment CA Sublegal bycatch Surplus production Lostyield commercial bycatch Sublegal Abundance indices 1985 1990 1995 2000 2005 2010 1996 2000 2004 2008 0

2.5 2.0 1.5 1.0 0.5 0.0

400 300 200 100 M (lbs) M CPUE (lbs/skate) CPUE Effort (’000 Skates) (’000 Effort 100 80 60 40 20 0 Area 3A summary ) ) Comm. CPUE Effort Comm. Removals CPUE & effort Wastage (<81 cm) (<81 Wastage cm) Bycatch (<81 cm+ (81 Wastage Personal use Personal Bycatch (81 cm+ Sport Commercial 1985 1990 1995 2000 2005 2010 1974 1980 1986 1992 1998 2004 0 0

50 40 30 20 10

600 500 400 300 200 100 CPUE (lbs/skate) CPUE Removals (M lbs) (M Removals Figure 23. Summary of removals, production, effort, abundance indices and age structure for Area 3A. Area for effort, abundance indices and age structure production, 23. Summary of removals, Figure

128 2009 IPHC ANNUAL MEETING HANDOUT Female Male Total Removals ASP F avg. = 11.1 T avg. = 12.7 M avg. = 14.4 o Age Age structure 10 15 20 25 1996 2000 2004 2008 5 0

15 10

0.15 0.10 0.05 0.00

Removals or ASP (M lbs) (M ASP or Removals

Proportion EBio (M lbs) (M EBio 150 100 50 0 Survey CPUE Survey assessment CW CA assessment Sublegal bycatch Surplus production Lost commercial yield Lost commercial bycatch Sublegal Abundance indices 1985 1990 1995 2000 2005 2010 1996 2000 2004 2008 0

1.5 1.0 0.5 0.0

500 400 300 200 100 CPUE (lbs/skate) CPUE M (lbs) M Effort (’000 Skates) (’000 Effort 50 40 30 20 10 0 Area 3B summary Comm. CPUE Comm. Effort Removals CPUE & effort Wastage (<81 cm) (<81 Wastage cm) (<81 Bycatch cm+) (81 Wastage use Personal (81 Bycatch cm+) Sport Commercial 1985 1990 1995 2000 2005 2010 1974 1980 1986 1992 1998 2004 0

5 0

800 600 400 200 20 15 10 CPUE (lbs/skate) CPUE Removals (M lbs) (M Removals Figure 24. Summary of removals, production, effort, abundance indices and age structure for Area 3B. Area for effort, abundance indices and age structure production, 24. Summary of removals, Figure

129 2009 IPHC ANNUAL MEETING HANDOUT Female Male Total Removals ASP F avg. = 10.7 T avg. = 12.2 T avg. = M avg. = 13.5 o Age Age structure 10 15 20 25 1996 2000 2004 2008 6 4 2 0 2

í

0.15 0.10 0.05 0.00

Removals or ASP (M lbs) (M ASP or Removals

Proportion EBio (M lbs) (M EBio 50 40 30 20 10 0 Survey CPUE CW assessment CA assessment Sublegal bycatch Surplus production Lost commercial yield Lost commercial bycatch Sublegal Abundance indices 1985 1990 1995 2000 2005 2010 1996 2000 2004 2008 0

2.5 2.0 1.5 1.0 0.5 0.0

500 400 300 200 100 CPUE (lbs/skate) CPUE M (lbs) M Effort (’000 Skates) (’000 Effort 15 10 5 0 Area 4A summary Comm. CPUE Comm. Comm. Effort Removals CPUE & effort Wastage (<81 cm) (<81 Wastage Bycatch cm) (<81 cm+) (81 Wastage use Personal Bycatch (81 cm+) Sport Commercial 1985 1990 1995 2000 2005 2010 1974 1980 1986 1992 1998 2004 0

8 6 4 2 0

800 600 400 200 CPUE (lbs/skate) CPUE Removals (M lbs) (M Removals Figure 25. Summary of removals, production, effort, abundance indices and age structure for Area 4A Area for effort, abundance indices and age structure production, 25. Summary of removals, Figure

130 2009 IPHC ANNUAL MEETING HANDOUT Female Male Total Removals ASP F avg. = 12.6 F avg. = T avg. = 14.6 M avg. = 15.8 o Age Age structure 10 15 20 25 1996 2000 2004 2008 4 2 0 2 4

í í

0.10 0.08 0.06 0.04 0.02 0.00

Removals or ASP (M lbs) (M ASP or Removals

Proportion EBio (M lbs) (M EBio 50 40 30 20 10 0 Survey CPUE Survey assessment CW CA assessment Sublegal bycatch Surplus production Abundance indices Lost commercial yield bycatch Sublegal 1985 1990 1995 2000 2005 2010 1996 2000 2004 2008 0

50

0.3 0.2 0.1 0.0

300 250 200 150 100 CPUE (lbs/skate) CPUE M (lbs) M Effort (’000 Skates) (’000 Effort Area 4B summary 25 20 15 10 5 0 ) ) Comm. CPUE Comm. Effort Removals CPUE & effort Sport Commercial Wastage (<81 cm) Wastage (<81 cm) Bycatch (<81 Wastage cm+ (81 use Personal Bycatch (81 cm+ 1985 1990 1995 2000 2005 2010 1974 1980 1986 1992 1998 2004 0

5 4 3 2 1 0

500 400 300 200 100 CPUE (lbs/skate) CPUE Removals (M lbs) (M Removals Figure 26. Summary of removals, production, effort, abundance indices and age structure for Area 4B. Area for effort, abundance indices and age structure production, 26. Summary of removals, Figure

131 2009 IPHC ANNUAL MEETING HANDOUT Female Male Total Removals ASP F avg. = 13.4 T avg. = 14.5 M avg. = 16.1 o Age Age structure 10 15 20 25 1996 2000 2004 2008 5 0

10

0.15 0.10 0.05 0.00

Removals or ASP (M lbs) (M ASP or Removals

Proportion EBio (M lbs) (M EBio 80 60 40 20 0 CW assessment CW 4D Survey trawl EBS Sublegal bycatch Surplus production Abundance indices Lost commercial yield Sublegal bycatch 1985 1990 1995 2000 2005 2010 1996 2000 2004 2008 0

4 3 2 1 0

800 600 400 200 M (lbs) M CPUE (lbs/skate) CPUE Effort (’000 Skates) (’000 Effort 15 10 5 0 Area 4CDE summary 4D C.CPUE 4C C.CPUE 4D Effort 4C Effort Removals CPUE & effort Wastage (<81 cm) (<81 Wastage cm) (<81 Bycatch Wastage (81 cm+) use Personal cm+) (81 Bycatch Sport Commercial 1985 1990 1995 2000 2005 2010 1974 1980 1986 1992 1998 2004 0 8 6 4 2 0

800 600 400 200 CPUE (lbs/skate) CPUE Removals (M lbs) (M Removals Figure 27. Summary of removals, production, effort, abundance indices and age structure for Area 4CDE. Area for effort, abundance indices and age structure production, 27. Summary of removals, Figure

132 2009 IPHC ANNUAL MEETING HANDOUT Appendix A. Selected fi shery and survey data summaries.

Table A1. Total removals (million pounds, net weight). Removals include commercial catch, IPHC survey catches, sport catch, personal use catch, legal-size bycatch and legal-sized wastage. Removals do not include sublegal-sized bycatch or sublegal-sized wastage. 2A 2B 2C 3A 3B 4 4A 4B 4CDE Total 1974 0.77 5.52 5.97 12.67 4.49 2.60 32.02 1975 0.71 8.03 6.69 13.21 4.22 1.73 34.59 1976 0.49 8.22 6.03 13.78 4.67 1.90 35.10 1977 0.48 6.16 3.67 12.20 4.73 3.20 30.44 1978 0.36 5.17 4.62 13.02 2.63 4.75 30.54 1979 0.32 5.56 5.34 16.19 1.08 4.82 33.31 1980 0.29 6.17 3.99 17.39 1.15 6.42 35.41 1981 0.47 6.20 4.73 18.97 1.55 1.55 0.94 3.08 37.49 1982 0.51 5.87 4.19 17.44 6.48 1.89 0.38 2.12 38.88 1983 0.58 5.78 7.13 17.16 8.96 3.10 1.66 2.14 46.52 1984 0.80 9.63 6.70 22.30 7.61 1.61 1.40 2.46 52.51 1985 0.94 11.30 10.31 23.78 11.43 2.32 1.57 2.75 64.40 1986 1.17 12.17 11.98 37.24 9.42 4.21 0.61 3.61 80.41 1987 1.29 13.48 12.04 36.47 8.50 4.50 1.90 3.39 81.55 1988 0.99 13.93 12.85 44.75 7.25 2.78 2.03 3.24 87.82 1989 1.06 11.52 11.49 40.00 8.47 1.54 2.97 2.49 79.54 1990 0.79 10.10 11.98 36.23 10.13 3.28 1.73 3.98 78.21 1991 0.78 8.83 11.95 32.42 13.46 3.44 1.81 4.46 77.15 1992 0.98 9.09 12.68 34.46 9.98 3.63 3.02 3.53 77.35 1993 1.05 12.00 13.74 30.59 8.46 2.92 2.48 3.14 74.38 1994 0.84 11.18 13.11 32.86 4.83 3.20 2.59 3.37 71.98 1995 0.93 11.55 9.80 24.51 4.02 2.86 1.84 3.56 59.06 1996 1.01 10.93 11.28 26.11 4.70 2.49 2.56 4.50 63.58 1997 1.26 13.75 12.37 31.86 9.92 3.94 3.55 5.49 82.14 1998 1.69 14.53 13.15 32.12 12.00 4.82 3.25 5.48 87.03 1999 1.57 14.01 12.45 31.02 14.69 5.57 3.94 6.56 89.80 2000 1.49 12.29 11.17 26.00 16.15 6.23 5.30 6.30 84.91 2001 1.79 11.80 10.72 27.97 17.04 5.82 4.88 6.87 86.89 2002 1.65 13.82 11.05 28.62 18.10 5.86 4.29 6.26 89.65 2003 1.61 13.48 11.47 29.70 17.80 5.61 4.10 5.46 89.22 2004 1.71 14.25 14.03 32.77 15.91 4.17 3.03 4.90 90.77 2005 1.51 14.70 14.20 33.61 13.62 3.96 2.26 5.78 89.64 2006 1.56 14.27 13.85 32.59 11.37 4.07 1.83 5.46 84.99 2007 1.51 11.81 12.35 34.20 9.79 3.56 1.75 5.85 80.82 2008 1.34 9.81 10.10 31.51 11.45 3.59 1.99 5.54 75.33

133 2009 IPHC ANNUAL MEETING HANDOUT Table A2. Commercial catch (million pounds, net weight). Figures include IPHC research catches. Sport catch in Areas 2A and 2B is not included in this table. 2A 2B 2C 3A 3B 4 4A 4B 4C 4D 4E Total 1974 0.52 4.62 5.60 8.19 1.67 0.71 ------21.31 1975 0.46 7.13 6.24 10.60 2.56 0.63 ------27.62 1976 0.24 7.28 5.53 11.04 2.73 0.72 ------27.54 1977 0.21 5.43 3.19 8.64 3.19 1.22 ------21.88 1978 0.10 4.61 4.32 10.30 1.32 1.35 ------22.00 1979 0.05 4.86 4.53 11.34 0.39 1.37 ------22.54 1980 0.02 5.65 3.24 11.97 0.28 0.71 ------21.87 1981 0.20 5.66 4.01 14.23 0.45 --- 0.49 0.39 0.30 0.01 0.00 25.74 1982 0.21 5.54 3.50 13.52 4.80 --- 1.17 0.01 0.24 0.00 0.01 29.01 1983 0.26 5.44 6.38 14.14 7.75 --- 2.50 1.34 0.42 0.15 0.01 38.39 1984 0.43 9.05 5.87 19.77 6.69 --- 1.05 1.10 0.58 0.39 0.04 44.97 1985 0.49 10.39 9.21 20.84 10.89 --- 1.72 1.24 0.62 0.67 0.04 56.10 1986 0.58 11.22 10.61 32.80 8.82 --- 3.38 0.26 0.69 1.22 0.04 69.63 1987 0.59 12.25 10.68 31.31 7.76 --- 3.69 1.50 0.88 0.70 0.11 69.47 1988 0.49 12.86 11.36 37.86 7.08 --- 1.93 1.59 0.71 0.45 0.01 74.34 1989 0.47 10.43 9.53 33.74 7.84 --- 1.02 2.65 0.57 0.67 0.01 66.95 1990 0.32 8.57 9.73 28.85 8.69 --- 2.50 1.33 0.53 1.00 0.06 61.60 1991 0.36 7.19 8.69 22.93 11.93 --- 2.26 1.51 0.68 1.44 0.10 57.08 1992 0.44 7.63 9.82 26.78 8.62 --- 2.70 2.32 0.79 0.73 0.07 59.89 1993 0.50 10.63 11.29 22.74 7.86 --- 2.56 1.96 0.83 0.84 0.06 59.27 1994 0.37 9.91 10.38 24.84 3.86 --- 1.80 2.02 0.72 0.71 0.12 54.73 1995 0.30 9.62 7.77 18.34 3.12 --- 1.62 1.68 0.67 0.64 0.13 43.88 1996 0.30 9.54 8.87 19.69 3.66 --- 1.70 2.07 0.68 0.71 0.12 47.34 1997 0.41 12.42 9.92 24.63 9.07 --- 2.91 3.32 1.12 1.15 0.25 65.20 1998 0.46 13.17 10.20 25.70 11.16 --- 3.42 2.90 1.26 1.31 0.19 69.76 1999 0.45 12.70 10.14 25.32 13.84 --- 4.37 3.57 1.76 1.89 0.26 74.31 2000 0.48 10.81 8.44 19.27 15.41 --- 5.16 4.69 1.74 1.93 0.35 68.29 2001 0.68 10.29 8.40 21.54 16.34 --- 5.01 4.47 1.65 1.84 0.48 70.70 2002 0.85 12.07 8.60 23.13 17.31 --- 5.09 4.08 1.21 1.75 0.56 74.66 2003 0.82 11.79 8.41 22.75 17.23 --- 5.02 3.86 0.89 1.96 0.42 73.19 2004 0.88 12.16 10.23 25.17 15.46 --- 3.56 2.72 0.95 1.66 0.31 73.11 2005 0.80 12.33 10.63 26.03 13.17 3.40 1.98 0.53 2.58 0.37 71.82 2006 0.83 12.01 10.49 25.71 10.79 --- 3.33 1.59 0.49 2.37 0.37 67.98 2007 0.79 9.77 8.47 26.49 9.25 --- 2.83 1.42 0.55 2.72 0.58 62.87 2008 0.71 7.79 6.21 24.38 10.89 --- 3.01 1.77 0.72 2.56 0.59 58.63

134 2009 IPHC ANNUAL MEETING HANDOUT Table A3. IPHC setline survey CPUE of legal sized fi sh in weight (net pounds per skate). Figures refer to entire areas. For cases where only part of an area was fi shed (e.g., northern 2B, western 3A), the CPUE shown is an adjusted value. No hook corrections are applied; J-hook values are raw J-hook catch rates. Area 4EBS is the eastern Bering Sea shelf, fi rst surveyed in 2006. For other years, the 4EBS CPUE is a constructed value based on the NMFS trawl survey and the single 2006 setline data point. 2A 2B 2C 3A 3B 4A 4B 4C 4D 4EBS Total J-hook surveys: 1974 ------1975 ------1976 ------1977 --- 13 --- 58 ------1978 --- 18 --- 27 ------1979 --- NA --- 41 ------1980 --- 25 --- 76 ------1981 --- 16 --- 131 ------1982 --- 21 114 130 ------1983 --- 18 142 119 ------1984 --- 25 --- 176 ------C-hook surveys: 1984 --- 57 260 361 ------7 --- 1985 --- 42 260 378 ------8 --- 1986 --- 38 283 305 ------9 --- 1987 --- NA ------10 --- 1988 --- NA ------20 --- 1989 --- NA ------13 --- 1990 --- NA ------14 --- 1991 --- NA ------12 --- 1992 --- NA ------11 --- 1993 --- 93 --- 261 ------22 --- 1994 --- NA --- 254 ------17 --- 1995 29 148 --- 300 ------20 --- 1996 --- 156 306 317 352 ------25 --- 1997 35 139 411 331 414 245 282 71 111 23 166 1998 --- 82 232 281 435 299 216 ------30 157 1999 37 88 204 241 438 290 203 ------27 147 2000 --- 93 233 272 373 276 216 --- 213 20 142 2001 41 102 237 256 357 199 171 --- 197 21 133 2002 33 92 261 299 297 168 119 --- 263 13 128 2003 22 73 223 229 262 154 104 --- 195 18 108 2004 27 86 173 270 236 137 73 --- 132 18 106 2005 28 72 171 276 211 107 86 --- 69 17 99 2006 16 59 144 232 181 84 95 --- 63 18 86 2007 19 57 140 212 191 66 87 --- 57 13 79 2008 18 88 108 189 126 83 103 --- 68 9 72

135 2009 IPHC ANNUAL MEETING HANDOUT Table A4. Commercial CPUE (net pounds per skate). Values before 1984 are raw J-hook catch rates, with no hook correction. 1983 is excluded because it consists of a mixture of J- and C- hook data. No value is shown for area/years after 1980 with fewer than 500 skates of reported catch/effort data. Total column recomputed in 2007 with new bottom area numbers. 2A 2B 2C 3A 3B 4A 4B 4C 4D 4E Total J-hook CPUE: 1974 59 64 57 65 57 ------1975 59 68 53 66 68 ------1976 33 53 42 60 65 ------1977 83 61 45 61 73 ------1978 39 63 56 78 53 ------1979 50 48 80 86 37 ------1980 37 65 79 118 113 ------1981 33 67 145 142 160 158 99 110 ------1982 22 68 167 170 217 103 --- 91 ------1983 ------C-hook CPUE: 1984 63 148 314 524 475 366 161 --- 197 --- 357 1985 62 147 370 537 602 333 234 --- 330 --- 400 1986 60 120 302 522 515 265 --- 427 239 --- 356 1987 57 131 260 504 476 341 220 384 ------349 1988 134 137 281 503 655 453 224 --- 201 --- 392 1989 124 134 258 455 590 409 268 331 384 --- 376 1990 168 175 269 353 484 434 209 288 381 --- 334 1991 158 148 233 319 466 471 329 223 398 --- 328 1992 115 171 230 397 440 372 278 249 412 --- 336 1993 147 208 256 393 514 463 218 257 851 --- 392 1994 93 215 207 353 377 463 198 167 480 --- 326 1995 116 219 234 416 476 349 189 --- 475 --- 351 1996 159 226 238 473 556 515 269 ------413 1997 226 241 246 458 562 483 275 335 671 --- 419 1998 194 232 236 451 611 525 287 287 627 --- 425 1999 --- 213 199 437 538 500 310 270 535 --- 394 2000 263 229 186 443 577 547 318 223 556 --- 412 2001 169 226 196 469 431 474 270 203 511 --- 379 2002 181 222 244 507 399 402 245 148 503 --- 378 2003 184 231 233 487 364 355 196 105 389 --- 349 2004 145 212 240 485 328 315 202 120 444 --- 340 2005 155 197 203 446 293 301 238 91 379 --- 314 2006 147 202 170 403 292 241 218 72 280 --- 284 2007 94 197 160 398 257 206 230 65 237 --- 269 2008 69 174 163 359 232 205 211 88 251 --- 248

136 2009 IPHC ANNUAL MEETING HANDOUT IPHC Staff regulatory proposals: 2009

Bruce M. Leaman and Heather L. Gilroy

In making catch limit recommendations for 2009, the staff has considered the results of the analytic assessment, changes in the commercial and survey indices used to monitor the stock, estimated recruitment of incoming year classes, and a harvest policy that refl ects coastwide policy goals. The staff also drew on the outcome of both the September 2008 Biomass Apportionment Workshop and recent regional meetings with industry. Detailed results of these additional investigations are reported in the 2008 Report of Assessment and Research Activities. Ongoing tag returns from the coastwide PIT tagging program continue to demonstrate that regulatory areas cannot be treated as closed management units, as had previously been the case. Changes in the stock biomass as indicated by our analytic assessment as well as changes in relative abundance indices from our surveys and the commercial fi shery were also infl uential in our recommendations for 2009. With the exceptions of Areas 2C, 4A, 4C, and 4D, commercial catch per unit effort (CPUE) in 2008 decreased from 2007 values; declines were generally around -10%. The 2008 IPHC assessment survey CPUE values increased in Areas 2B, 4A, 4B, and 4D but decreased in all other areas, with particularly sharp declines in Areas 2C and 3B. The analysis of optimum harvest rates for the coastwide assessment conducted in 2006 resulted in a target harvest rate of 20% of coastwide exploitable biomass. The staff examined multiple alternatives, including industry suggestions, for apportioning the estimated coastwide exploitable biomass among regulatory areas and concluded that the use of the IPHC assessment survey data offered the most standardized and consistent data with which to achieve this partitioning. However, the staff also recognized some regional differences in hook competition with other species and applied an adjustment to accommodate that feature. Accordingly, the distribution of biomass, as determined by the three-year average CPUE of legal-sized fi sh obtained on the stock assessment survey adjusted for hook competition, was used to partition the coastwide exploitable biomass estimate into regulatory area biomass totals. Following reanalysis of the depth distribution of survey data compared with bottom depth distribution the staff also removed an adjustment that was applied in Area 2A for the 2008 apportionment. While the 20% harvest rate is appropriate for the majority of the stock, a harvest rate of 15% is indicated by the analysis of productivity for Areas 4B and 4CDE conducted in 2005, and a similar analysis for Area 4A conducted in 2008. Therefore, staff recommended catch limits for Area 4 use a 15% harvest rate. Fishery statistics and biological characteristics of halibut in Area 3B are also of some concern to staff and a detailed analysis of this area will be conducted in 2009.

Catch limit recommendations The staff recommendations totaling 54.08 million pounds for 2009 are presented in Table 1. The Area 2A recommendation includes all removals (commercial, treaty Tribes, and sport) allocated by the Pacifi c Fishery Management Council's (PFMC's) Catch Sharing Plan. Area 4CDE is treated as a single regulatory unit by the Commission, although the North Pacifi c Fishery Management Council’s (NPFMC's) Catch Sharing Plan allocates the Commission catch limit into limits for the individual regulatory areas. The Area 2B catch limit recommendation includes totals for the

137 2009 IPHC ANNUAL MEETING HANDOUT commercial and sport fi sheries. The Canadian Department of Fisheries and Oceans will allocate the adopted catch limit between the sport and commercial fi sheries. For Areas 2C and 3A the catch limit recommendation includes the use of the NPFMC and National Marine Fisheries Service (NMFS) authorized Guideline Harvest Levels (GHL) for the halibut recreational charter fi sheries of 0.788 Mlb and 3.650 Mlb, respectively, as the projected removals by that sector for 2009. Staff notes that the GHL of 0.788 Mlb for Area 2C is lowered from the 2008 GHL by NMFS regulatory requirements, based on the lower CEY for Area 2C in 2009. The catch limit recommendations are made with the assumption that both Canada and the U.S. will manage to their domestic targets for sport fi sh catch and the national parties are cautioned that any departure from these assumed levels of removal by the recreational sector will compromise achievement of Commission harvest targets for 2009. The staff notes with great concern that actions by the two national parties in 2008 were insuffi cient to achieve the sector removal targets established by the management agencies of the parties. In Area 2C, successful legal challenges to the NMFS regulations designed to achieve the NPFMC-approved GHL for the charter halibut fi shery resulted in a charter harvest more than double the target level of 0.931 Mlb. In Area 2B, regulations to restrict removals by the recreational fi shery were insuffi cient to meet the reductions required by a reduced total catch limit and only an underharvest of commercial IVQs allowed the total removals to remain within the total catch limit. Both national parties are urged to implement regulations and monitoring that will achieve the adopted Commission catch limit regulations. The use of a coastwide assessment and apportionment of coastwide biomass based on survey estimates of distribution creates some substantial changes in Total Constant Exploitation Yield (Total CEY) and recommended catch limits among areas, compared to previous assessments. Lower recommended catch limits are identifi ed for Areas 2, 3A, 4A, and 4CDE while Areas 3B and 4B have somewhat higher recommended catch limits. These differences are associated with the different distribution of biomass associated with survey apportionment of a coastwide total biomass, compared with the previous biomass distribution estimated from closed-area assessments, application of target harvest rates, as well as CPUE changes in both the survey and the commercial fi shery. As noted in the 2007 stock assessment, the distribution of biomass based on survey estimates is more consistent with other estimates of biomass distribution that are independent of the stock assessment. The staff continues to recommend a slow rate of increase in catch limits when estimated CEY is increasing and a more rapid reduction of catch limits when CEY is decreasing (a Slow Up – Fast Down policy). For Areas 2, 3A, 4A, and 4CDE the staff recommends catch limits that are lower by one-half of the difference between 2008 catch limits and the estimated fi shery CEYs for 2009. For Areas 3B, and 4B, the staff recommends an increase over the 2008 catch limit equivalent to one-third of the difference between the 2008 catch limit and the estimated 2009 fi shery CEY. The staff recognizes that adoption of the coastwide assessment and survey apportionment results in a signifi cant shift in the estimated distribution of exploitable biomass. This analysis concludes that exploitation rates on the eastern portion of the stock have been too high in the past decade, resulting in lower biomass in Area 2 than would be realized if harvest rates had been near the target level. In the longer term, a lowered harvest rate will permit rebuilding of the exploitable biomass in Area 2 and an increase in available yield. The pace of that rebuilding will be affected by the strength of year classes recruiting to the fi shery over the next several years and staff is particularly concerned with the need to realize the advantage from the 1999-2000 year classes that can be achieved through reduced harvest rates in Area 2.

138 2009 IPHC ANNUAL MEETING HANDOUT Fishing periods As in the past years, the staff recommends March 15 to November 15 opening and closing dates for the quota share fi shing season. This recommendation is a compromise between minimizing interceptions of migrating fi sh and providing opportunity for market presence of wild halibut. The Area 2A fi shery should also occur within this period. For the Area 2A directed commercial fi shery, the staff is surveying the directed commercial license holders to determine their preferred starting date from the following: May 13, May 27, June 10, or June 24. The staff will likely support the directed commercial harvester’s choice for the fi rst directed commercial opening date and as in past years, will recommend a series of 10- hour periods, with fi shing period limits to ensure that the catch limit is not exceeded. The size of the fi shing period limits will be determined when more information on fl eet participation is available.

Catch sharing plans: Area 2A and 4CDE The Commission does not make allocative decisions within regulatory areas or among different user groups. However, for Areas 2A and 4CDE the staff recommends that the Commission endorse the catch sharing plans developed by the PFMC and NPFMC for these areas, respectively.

Proposed changes to the IPHC regulations Fishing in multiple IPHC Regulatory Areas (4A, 4B, 4C, or 4D) In 2008, regulations were approved by the Commission that allowed fi shing in multiple regulatory areas (4A, 4B, 4C, or 4D), provided a certifi ed observer was on board; or a Vessel Monitoring System was on board and the vessel did not possess at any time on board more halibut than the IFQ allowed for the area currently fi shed. In both cases, the halibut need to be identifi able by regulatory area. The IPHC regulation as written in the U.S. Federal Register requires the IFQ limit for the area being fi shed be met whether fi shing occurred in any other area fi rst, when an observer is on board, which was not the intent. We recommend changing the text to ensure the intent of the regulation is correct.

72-hour restriction in Area 2A IPHC staff and Area 2A enforcement offi cers recommend changing the Area 2A regulation that restricts setline gear use during the 72-hour period immediately before the opening. The change would add a restriction that vessels and skippers fi shing before the 72 hours immediately prior to the opening would have to offl oad their catch or submit to a hold inspection by an authorized offi cer before taking part in the halibut opening. Without this change, enforcement cannot tell if fi sh onboard was caught before or within the 72-hour period. In addition, this recommendation changes the regulation from a restriction for setline vessels to all vessels fi shing in the directed commercial fi shery.

Alaska sport fi shing regulations The 2008 Alaska sport regulations stated that no person shall possess on board a fi shing vessel, including charter vessels and pleasure craft used for fi shing, halibut that has been fi lleted, mutilated, or otherwise disfi gured in any manner except that each halibut may be cut into no more than 2 ventral, 2 dorsal pieces, and 2 cheeks, with skin on. Problems arose in 2008 that

139 2009 IPHC ANNUAL MEETING HANDOUT made the daily and possession limits unenforceable. Based on consultations by the IPHC staff with Alaska Department of Fish and Game, NMFS, and NOAA Enforcement staff, the IPHC staff is recommending changing the regulation to reference ‘vessel’ instead of ‘fi shing vessel’. Additionally, the IPHC staff is recommending adding an exemption so that halibut in excess of the possession limit may be possessed on the vessel for the transportation of halibut, when the vessel is not carrying sport fi shing gear as defi ned in the regulations.

References Hare, S.R. and Clark, W.G. 2009. Assessment of the Pacifi c halibut stock at the end of 2008. Int. Pac. Halibut Comm. Report of Assessment and Research Activities 2008: 137-202.

140 2009 IPHC ANNUAL MEETING HANDOUT Table 1. Staff recommended catch limits for 2009 by IPHC regulatory area (million lbs, net weight). Recommendations based on coastwide assessment of exploitable biomass with survey apportionment to regulatory area exploitable biomass. Removal data are preliminary. Also shown are the 2008 catch limits for comparison. 2008 2008 2009 Reg Exploitable Harvest Total Other Catch Fishery 2009 Catch Limit Area biomass rate CEY Removals Limit CEY Recommendation 2A 3.21 20.0% 0.64 0.14 1.22 0.50 0.86 1,4 2B 27.07 20.0% 5.41 0.50 9.00 4.92 6.96 2,4 2C 27.87 20.0% 5.57 2.71 6.21 2.86 4.54 4 3A 140.04 20.0% 28.01 7.17 24.22 20.84 22.53 4 3B 68.78 20.0% 13.76 0.56 10.90 13.20 11.67 3 4A 18.47 15.0% 2.77 0.57 3.10 2.20 2.65 4 4B 15.40 15.0% 2.31 0.23 1.86 2.09 1.94 3 4CDE 24.16 15.0% 3.62 1.66 3.89 1.97 2.93 4 Total 325.00 19.1% 62.10 13.52 60.40 48.58 54.08 Note: Exploitable biomass is coastwide assessment, survey partitioning; amended 2C sport 1 Catch limits and Fishery CEY for 2A includes commercial, sport, treaty subsistence catches 2 Catch limits and Fishery CEY for 2B includes commercial and sport catch. 3 Calculated as 2008 catch limit plus 1/3 of the difference between 2009 Fishery CEY and 2008 Catch Limit. 4 Calculated as 2008 Catch Limit minus ½ of the difference between 2009 Fishery CEY and 2008 Catch Limit. Assumes GHL of 0.788 Mlb in Area 2C, 3.65 Mlb in Area 3A under Other Removals; reduced GHL in Area 2C, based on NMFS-regulated GHL reduction associated with reduced Total CEY. Other removals for 2C and 3A are adding projected unguided harvest to the applicable GHL.

141 2009 IPHC ANNUAL MEETING HANDOUT 1 6 -- 7 Annual Meeting 66,990 59,240 5 CATCH 1,309 1,369 1,311 1,186 3,329 3,278 2,786 2,962 1,9233,459 1,542 3,199 1,369 3,830 1,725 3,855 14,089 13,723 11,250 9,257 10,489 10,397 8,346 6,148 25,228 25,238 26,133 24,021 12,874 10,565 9,047 10,762 2004 2005 2006 20072004 2008 2005 2006 2007 2008 1,395 3,473 2,683 2,889 10,114 13,700 24,717 15,180 1 1 9,000 2004 2005 2006 2007 2008 SCSC S CS C S C

4 2008 ESTIMATED CEYESTIMATED RECOMMENDATIONS STAFF c halibut by IPHC regulatory area (in thousands of pounds, net weight), 2004-2008. area c halibut by IPHC regulatory fi shery CEY, staff recommended catch limits, catch limits, catch, percent of catch limits taken, survey and catch limits, catch, percent staff recommended shery CEY, fi 2004 2005 2006 2007 2008 1,890 1,170 1,490 6601,480 650 1,330 1,480 1,380 1,330 1,340 1,380 1,220 1,020 1,000 94 98 99 98 97 27 145 28 155 16 147 19 94 18 69 15,780 12,700 13,200 6,220 4,65013,800 13,800 13,250 13,250 13,220 13,220 11,470 9,720 8,060 75,505 73,819 69,860 65,170 60,400 73,910 72,700 69,311 64,072 59,916 2004 2005 2006 2007 PERCENT OF CATCH LIMIT TAKEN LIMIT OF CATCH PERCENT (C) CPUE AND COMMERCIAL (S) SURVEY AVERAGE shery CEY and staff recommendations from coastwide stock assessment with survey partitioning to area as presented in the 2009 and staff shery CEY fi 2 2 2 3 3 2B2C3A3B4A 994B 96 106 99 97 96 104 100 99 95 98 98 98 100 97 97 85 98 100 98 103 92 86 98 99 96 99 212 173 270 95 99 96 240 485 72 236 142 93 197 171 276 328 315 203 73 446 211 59 111 202 293 144 232 301 202 170 86 403 181 57 238 88 140 212 292 241 197 191 95 160 398 66 218 88 108 257 189 206 87 174 163 126 359 230 83 232 103 205 211 2C3A3B4A 17,0304B 29,980 15,600 11,800 26,300 3,470 10,700 10,330 2,810 24,940 3,400 8,570 4,980 1,700 27,630 3,250 16,770 3,920 1,070 22,250 14,270 5,230 11,310 25,600 2,560 15,600 3,510 10,930 25,470 2,700 13,150 10,630 3,470 25,200 2,810 10,860 26,010 7,810 3,440 2,260 12,830 24,220 3,350 6,210 10,900 1,670 3,980 1,970 3,100 1,860 2C 10,500 10,930 10,630 8,510 6,210 3A 25,060 25,470 25,200 26,200 24,220 3B 15,600 13,150 10,860 9,220 10,900 4A 3,470 3,440 3,350 2,890 3,100 4B 2,810 2,260 1,670 1,440 1,860 2B 2B 2A 2A 2A Area 2004 2005 2006 2007 2008 Area Area Total 89,880 72,170 65,950 67,900 55,630 76,910 73,820 69,860 Total 4CDE 76 87 90 93 99 ------4CDE 3,390 4,400 3,110 3,850 3,680 3,390 3,990 3,550 3,650 3,890 4CDE 3,785 3,989 3,550 4,100 3,890 Regulatory LIMITS CATCH Regulatory Regulatory Total 97 98 99 98 99 Catch limits do not include additional pounds from underage/overage programs Estimated includes commercial catch, sport and treaty ceremonial subsistence catch Area 2A Area 2B includes commercial and sport catch Catch does not include IPHC research catch 2008 data are preliminary CPUE based on small sample; refer to Hare and Clark (2009) Area 2A lbs/skate: Handout. Table 2. Estimated Table 1 2 3 4 5 6 7 commercial CPUE of Paci commercial

142 2009 IPHC ANNUAL MEETING HANDOUT Appendix I. IPHC Research program: Review of 2008 projects and preliminary proposals for 2009

IPHC Staff

Introduction This document reviews research conducted by the IPHC staff in the past year and proposed for the upcoming year. The report is divided into two sections, with the fi rst section reviewing the status of research projects conducted in 2008. The second section presents the preliminary staff research proposals for 2009. Information is provided on when each project was initiated, the anticipated completion date, the annual cost, a description of the costs, and the purpose of the project. This report does not include ongoing staff tasks such as data collection and processing that are necessary for the management of the fi shery. Research projects are organized into three funding categories that refl ect availability and source of research funds. Limited research requiring direct fi nancial support from the Commission is possible under the basic $3.64 million (as of 2008) government appropriations, although a number of programs can be conducted using only the staff resources that are supported by the appropriations. The three funding categories are:

1) Funded Research: Necessary research projects of high priority that can only be conducted with revenues generated by survey fi shing in 2008, and/or carry-over from 2007; 2) Contracts and Grants: Agreements with other parties to conduct specifi c research. In this case, contracts and grants are shown for projects where the IPHC staff is the principle investigator; and 3) Research conducted without direct funding: Necessary research projects of high priority that can be conducted within the IPHC budget.

Nearly all of the research done by the staff is directed toward one of three continuing objectives of the Commission:

i) Improving the annual stock assessment and quota recommendations; ii) Developing information on current management issues; and iii) Adding to knowledge of the biology and life history of halibut.

In each of these areas our routine work program applies the best information and methods available, and our research program aims to improve the information and methods by answering the most important outstanding questions.

143 2009 IPHC ANNUAL MEETING HANDOUT SECTION I: REVIEW OF RESEARCH CONDUCTED IN 2008 Research conducted by the IPHC staff during 2008 continued in three basic areas: life history, fi sh movements, and general biology. Most of the projects were conducted as part of the normal staff duties, with no additional funding required outside of staff salaries. Funding for projects outside of staff salaries came from supplemental funding, and these projects are outlined below. Detailed reports can be found in the current Report of Assessment and Research Activities (RARA).

Overview The PIT tag scan sampling program (#413) continued in 2008, with samplers in eight Alaskan and four B.C. ports. Additionally, IPHC received state and tribal assistance in scanning in nine ports in Washington and Oregon. Through 2 November, almost 23 million pounds (40% of total landings) have been scanned. The number of tags recovered in 2008 totals 255 from the 2003 primary experiment and 159 from the 2004 experiment. Once again, very few tags were recovered by the fi shery in Area 4. Achieving desired scanning goals for Area 4B trips was a challenge in 2008, as many vessels shifted their landings to unstaffed ports. Overall, the number of recoveries continued to decline from the previous year, which is typical as natural mortality affects the remaining fi sh at large. 2009 is the last year of scheduled scanning for tags released in this experiment. Effort on the Genetics project (#621) continued in 2008 with additional sample analysis under the supervision of Dr. Lorenz Hauser at the University of Washington’s Marine Molecular Biology Laboratory. The project was initiated in 2002 to investigate the genetic population structure in the northeast Pacifi c using non-selected nuclear microsatellite markers, and in 2004 the study was expanded to spawning groups from British Columbia, the central Gulf of Alaska, and southeast Bering Sea. Although initial analysis suggested population differentiation, interpretation of results was complicated by very low FST values and the fact that genetic studies conducted without temporal replicates are in danger of detecting false positives. In February 2007, winter charters were conducted to resample the locations visited in 2004, and analysis of those samples continued through December, 2007. Increasing the number of microsatellites analyzed and fi lling gaps in the data set resulted in little change from prior work, although increasing sample sizes did result in increased levels of signifi cance. At the end of 2007 we deemed the microsatellite work to be largely complete, with the conclusion that these markers may simply not be very powerful for detecting population structure in a species, such as halibut, with large amounts of larval mixing. At the same time, we lost our project technician. However, in May of 2008 we were fortunate to fi nd a full- time post-doctoral researcher (Dr. Heather Galindo, formerly at Stanford University) to dedicate herself to the program. During the relatively short period of her employment, she has managed to make considerable progress investigating some other genetic markers that may be more powerful than microsatellites. She has screened 25 “expressed sequence tags”, found in regions of DNA that are responsible for coding proteins and that may therefore be driven by evolution in different directions in different ocean basins. She has also sequenced four mitochondrial DNA (mtDNA) regions, which are maternally-inherited and have proven useful in other species for investigating sex-biased migration and demographics in relation to climate change. This work is expected to continue through this fall and winter. A PAT-tagging project (#622.12) was initiated in 2008 to investigate why so few PIT tags were recovered from Area 4 and investigate the possibility that eastward migration is higher south

144 2009 IPHC ANNUAL MEETING HANDOUT of Unimak Pass than north of it. Of particular importance is the fact that PAT tags do not need to be physically recaptured in order to generate accurate endpoint locations, thereby eliminating spatial recovery biases arising from regional differences in reporting, or tag detection. In addition, archived depth data broadcast along with the endpoint locations can be used to determine the timing and duration of seasonal migratory phases between shallow and deep-water habitats, and fi ne-scale depth data downloaded from physically-recovered tags can be used to defi ne periods of presumed active spawning (that is, egg release). These data can be useful for assessing match- mismatch between commercial fi shery season opening and closing dates and the actual timing of seasonal migration and spawning periods, on a regional basis. The latter represents a geographic extension of PAT tag studies conducted in Areas 2 & 3 and recently expanded via external archival tagging. During the 2008 setline survey, 115 halibut were tagged throughout the Bering Sea and Aleutian Islands with tags programmed to report location exactly 365 days after tagging. At the beginning of October, one tag had been recovered by the targeted fi shery and another ten had released prematurely, leaving a total of 104 tags still in the water. The pilot study (Project 650.12) on use of archival tags on halibut, which began in 2006, continued in 2008 with the release of 162 archival tags in Area 2B during August-September. The tags utilized an external mount which allows for easier detection and recovery than internally- implanted tags. Veterinary ultrasound was used to determine gender of fi sh prior to tagging and proved to be a rapid (approximately 6 seconds per fi sh) and highly accurate method (0% error in a 45-fi sh test) of noninvasive sex-determination. An equal number of males and females, ranging from 71-151 cm fork length, were tagged in three regions with temperature-depth logging archival tags: 42 off northwestern Vancouver Island, 60 off the southern Queen Charlotte Islands, and 60 off the northwestern Queen Charlottes. Four fi sh (all female; 74-108 cm FL) were internally- implanted with temperature-depth-light logging tags that were left over at the end of the 2007 holding experiment. By the end of September, a single tag had been recovered in northern BC (near the Whaleback) by a commercial trawl vessel. Conducted in tandem with the release of archival tags was a gear experiment examining relative catch rates on swivel and non-swivel gear. This experiment (#655.11) sought to determine whether the presence of swivels on setline gear has an effect on the catch of halibut or bycatch species. With a new charter vessel, the F/V Van Isle, we completed 36 gear sets, with each set comprised of 4 pairs of swivel and non-swivel skates. Fishing was conducted in Area 2B, during August and September 2008. The non-swivel gear was the standard gear used on the assessment surveys. The swivel gear used 24-inch gangions made of 400-lb test monofi lament with swivels attached to the hook. These in turn were tied to the mainline with a short nylon becket. Both gears used 16/0 hooks. Preliminary analysis suggests that there is no catch difference in terms of weight of legal-sized halibut caught, but there may be a difference in the catch rate of sublegals. We are still conducting the analysis of catches of rockfi sh and other bycatch species. Removal sampling can be a useful technique for directly estimating catch probability, so a pilot fi eld experiment (#656.11) was conducted in 2008 to examine the effectiveness for halibut. Sampling was done at 20 setline survey stations in the eastern part of Area 3A. At each station, a sequence of fi ve sets was made on consecutive days following the general setline survey protocol, with the fi rst set being the standard survey set. The expectation was that catch would decrease over the fi ve days of fi shing, and the rate of decrease would provide information on catch probability. While standard removal models assume a closed population during the sample period, we anticipated some movement into and out of the catchable population during the fi ve

145 2009 IPHC ANNUAL MEETING HANDOUT days, and developed models that allowed for some degree of local migration. A formal analysis is in progress, but the raw data show that, on average, catch did not decrease over the fi ve days as had been expected. Possible explanations are that the local populations were so large that each set removed only a very small fraction of the population, or that local migration was such that the removed fi sh were replaced with new migrants on a daily basis. Whatever the cause, with no decline in catch, distinguishing catchability from other factors affecting catch size becomes extremely diffi cult. Although fi nal conclusions will await the results of a formal analysis, the raw data imply that removal sampling is not a promising approach for estimating catchability of Pacifi c halibut. A study to verify, by direct observation, the halibut hooking success curve for halibut on setlines took place in 2008 (#652.12) in Area 3A as a follow-up to work conducted in 2007. The goal in this latest experiment was to derive a similar curve for the smaller 14/0 hooks. We completed 65 deployments with the 14/0 hooks, and completed another 40 with the 16/0 hooks. This represents over 64 hours of viewing hooks on bottom. The 2008 data are currently being reviewed and edited, and results from the combined studies will be published next year. Combined with observations on mouth dimensions, the comparison of the two hook sizes should give us valuable insights to the physical parameters which can be used to describe the halibut selectivity curve. The ability to use electronic (video) monitoring in the halibut fi shery off Alaska was the focus of Project 654.11-84. The long term focus of this research is to improve the understanding of the ecosystem impacts of halibut fi shing through improved monitoring of longline fi shery bycatch and to provide data on mortality of bycatch species for input to stock assessments. In this study, we compared and evaluated the effectiveness of electronic monitoring (EM) and the current North Pacifi c Groundfi sh Observer Program (NPGOP) monitoring methods to operate effectively in a commercial longline (hook-and-line) setting in Alaskan waters. This was a cooperative study with the commercial fi shing industry, IPHC, NMFS, and PSMFC, and relied on our ability to sample on various vessel confi gurations. The project was not as successful in getting as many vessels to participate as expected. Only four vessels, three in the Area 3A and 1 in Area 4, took an observer and accompanying video system. Over 230 sets were monitored among the four vessels. Archipelago Marine Research (Victoria BC) is currently analyzing the video data for the subsequent comparison with the observer sampling. We have gained agreement from the NPRB to extend the fi eld activities into 2009, so that additional vessels deployments can be accomplished, which should result in a more complete data set for analysis of applicability of the technology to different vessel size classes. Analytical results should be available in early 2009. IPHC has been involved in several cooperative studies which utilize the summer assessment survey and the catch of non-halibut species. Since 2002, the IPHC has worked cooperatively with both the Washington Department of Fish and Wildlife (WDFW) and Oregon Department of Fish and Wildlife (ODFW) to collect rockfi sh (Sebastes spp.) bycatch data. All rockfi sh caught on operations in 2A are retained and marked externally with a Floy T-bar anchor tag and the tag number is recorded with the set (and recently the skate number) information. All marked fi sh are retained so state biologists can collect additional data shore-side. Marketable fi sh are sold. IPHC then provides each agency with the effort information collected as part of the normal survey data collection. In 2008, IPHC worked with WDFW and ODFW to fi sh supplemental stations designed to further enhance the understanding of rockfi sh status in these areas. In each state, the stations’ locations and design were specifi ed by the state agency involved. Eighteen stations were fi shed off

146 2009 IPHC ANNUAL MEETING HANDOUT Washington (a continuation of similar studies from the previous 2 years) and twenty stations off Oregon. Three skates of gear were fi shed at each station as a precautionary approach due to the exploratory nature of these stations and concerns about overfi shing yelloweye rockfi sh (Sebastes ruberrimus). Activities at each station were identical to those on standard IPHC stations except that halibut were only sampled for length and prior hooking injury and then released alive; rockfi sh were handled as described above. A summary of this project will be submitted to each state by the end of November 2008. Depending on those results, there may be similar cooperative work conducted in future years. In 2008, IPHC worked with the Department of Fisheries and Oceans Canada (DFO) to provide a third biologist on IPHC survey vessels to collect hook by hook occupancy information for all species, and otoliths, maturities, and lengths for rockfi sh except thornyheads. This is the sixth year of this cooperative program and continued collaboration is anticipated. In 2008, IPHC worked cooperatively with Alaska Department of Fish and Game (ADFG) to provide a third biologist on IPHC survey vessels in the Fairweather, Sitka, Ommaney and Ketchikan charter regions to collect hook by hook occupancy information for all species, and otoliths, sex, and lengths for yelloweye rockfi sh. This project built upon cooperative work started with ADFG in 2007 and future collaboration is anticipated. IPHC collected length frequency data on incidentally caught Pacifi c cod (Gadus macrocephalus) in the 4A Edge and 4D Edge charter regions. This project was initiated at the request of NMFS- AFSC Pacifi c cod assessment team and part of a developing effort to collect bycatch information on Pacifi c cod in the western regions of our survey, where it makes up the largest component of our survey bycatch. Finally, IPHC hired an intern (Project 618) in 2008. The intern program generally includes various pre-determined offi ce tasks as well as being assigned a research project which the intern designs and executes. A fi nal report and presentation are given at the conclusion of the employment term and results are included in that year’s RARA. In 2008, we went a different direction and hired a fi lm student from the University of Montana to make a port sampling training video. The student constructed a story board, traveled to several ports for footage, conducted interviews, and then put it all together. The project was a success. For 2009, we plan to go back to the traditional science student format, but may seek out specialties in the future if a particular project requires expertise not available on the staff.

2008 Grants and contract research NMFS Auke Bay Lab (ABL) has had a sablefi sh data collection program for several years and IPHC has assisted NMFS with the program. In 2003/2004, the program was reviewed and modifi ed to meet the IPHC confi dentiality policy and to encompass all vessels rather than just vessels greater than 60 feet. Under a Statement of Work (SOW), NMFS contracts IPHC to collect and review information on sablefi sh catches (Project 628.00) during the IPHC port sampler’s logbook interview. Sablefi sh data are entered by IPHC staff, edited, and an electronic summary provided to the ABL scientists. Vessels are assigned a unique code in the summarized data to preserve confi dentiality. The SOW was renewed for 2009. IPHC also received several grants in 2008. NMFS provided a grant for the incremental increase in port sampling costs due to the IFQ program (Project 300.00-81). IPHC also received a grant from NPRB to partially cover its costs associated with the study examining the use of electronic

147 2009 IPHC ANNUAL MEETING HANDOUT monitoring (video) of the halibut fi shery off Alaska (#654.11-84). Additional payments from the NPRB grant are expected as the project continues. IPHC also received funds from APICDA and CBSFA to offset a potion of the cost of the large release of PAT tags in Area 4 in 2008 (#622.12).

148 2009 IPHC ANNUAL MEETING HANDOUT Budget Summary for 2008 Projects

Project FY08 FY08 No. Project Title Budget Expenses Field Experiments 413.00 PIT tagging study: Fifth year of tag recovery and scanning 414,235 410,890 416.00 PIT tagging study: Double tag expt. 400 311 419.00 Eastern Bering Sea assessment survey 01 01 604.00 NMFS trawl survey: At-sea data collection 21,414 24,483 Field Experiments Total $ 436,049 $ 435,684 Other Research 610.11 Water column profi ler (annual survey) 2,550 3,526 610.12 Water column profi ler (Oregon grid) 2,050 736 610.13 Water Column profi ler (coastwide) 376,142 0 618.00 Undergraduate internships 17,109 13,479 620.00 Otolith elemental fi ngerprinting 0 498 621.00 Genetic population structure – lab work by UW 101,130 55,412 622.11 PAT Tags: Summer 2006 releases (Areas 4A and 4D) 17,815 2,259 622.12 PAT Tags: Summer 2008 releases (Area 4) 369,370 425,023 630.00 Sleeper shark genetics studies 6,400 5,118 636.00 Analysis of gonad staging on IPHC setline surveys 24,500 0 642.00 Assessment of mercury and contaminants in Pacifi c halibut 300 423 646.11 PAT tags – summer 2006 releases (Area 2A) 0 500 646.12 PAT tags – summer 2006 releases (Area 2B) 0 2,251 650.11 Archival tags – holding studies 4,424 2,599 650.12 Archival tags – pilot releases (Area 2B) 194,775 185,772 651.00 Effect of magnets and rare metals on catch rates of Spiny dogfi sh 0 5,815 652.12 Hooking success 49,856 63,732 653.00 Species identifi cation of amphipods frequenting Pacifi c halibut 1,000 0 654.11 Alaska fi shery electronic monitoring – pilot study 94,300 79,849 655.11 Swivel gear comparison 118,439 83,514 656.11 Removal fi shing – pilot study 94,702 193,264 Other Research Total $ 1,474,862 $ 1,123,409 GRAND TOTAL $ 1,910,911 $ 1,559,093 Note: Values shown do not include any revenues generated from the sale of fi sh or other cost offsets. 1 Eastern Bering Sea Survey approved in concept by Commissioners, contingent on expenses being similar to staff projections (~$350 k) and availability of vessels. The Commission received insuffi cient bids to complete survey.

149 2009 IPHC ANNUAL MEETING HANDOUT Other 2008 Research – Contracts and Grants Granting agency shown in parentheses Project FY08 Project Title No. Income 300.00-81 AK port sampling grant (NMFS) 225,980 420.00 DFO Rockfi sh data collection (DFO) 14,968 622.12 PAT Tags: Summer 2008 releases (Area 4) (CBSFA, APICDA) 105,000 628.00 AK catcher vessel logbook and sablefi sh data collection (NMFS) 35,900 654.11-84 Alaska fi shery electronic monitoring – pilot study (NPRB) 11,863 GRAND TOTAL $ 393,711

150 2009 IPHC ANNUAL MEETING HANDOUT 2008 Research Publications IPHC staff noted in Bold type.

Kaimmer, S.M. and Stoner, A.W. 2008. Field investigation of rare-earth metal as a deterrent to spiny dogfi sh in the Pacifi c halibut fi shery. Fish. Res. 94(1):43-47. Loher, T. 2008. Homing and summer feeding site fi delity of Pacifi c halibut (Hippoglossus stenolepis) in the Gulf of Alaska, established using satellite-transmitting archival tags. Fish. Res. 92:63-69. Loher, T. and Seitz, A.C. 2008. Characterization of active spawning season and depth for eastern Pacifi c halibut (Hippoglossus stenolepis), and evidence of probable skipped spawning. J. Northw. Atl. Fish. Sci. 41:23-36. McElderry, H.I., Reidy, R.D. and Pahti, D.F. 2008. A pilot study to evaluate the use of electronic monitoring on a Bering Sea groundfi sh factory trawler. Int. Pac. Halibut Comm. Tech. Rep. 51:29 p. Moukhametov, I.N., Orlov, A.M., and Leaman, B.M. 2008. Diet of Pacifi c halibut (Hippoglossus stenolepis) in the northwestern Pacifi c Ocean. Int. Pac. Halibut Comm. Tech. Rep. 52:24 p. Seitz, A.C., Loher, T., and Nielsen, J.L. 2008. Seasonal movements and environmental conditions experienced by Pacifi c halibut along the Aleutian Islands, examined by pop-up satellite tags. Int. Pac. Halibut Comm. Sci. Rep. 85, 24p. Stoner, A.W., and Kaimmer, S.M. 2008. Reducing elasmobranch bycatch: Laboratory investigation of rare earth metal and magnetic deterrents with spiny dogfi sh and Pacifi c halibut, Fish. Res. 92(2-3):162-168. Valero J.L., Lee, B., Armstrong, D., Orensanz, L., Parma, A., Hilborn, R., Sizemore, B., and Palzer, T. 2008. Population dynamics and historic trends of geoduck clams under episodic low dissolved oxygen conditions in Hood Canal. J. Shellfi sh Res. 2:462-463. Valero, J.L. and Lasta, M.L. 2008. Estimating survival of discarded scallops in the Patagonian scallop fi shery: Comment on “Survival of Patagonian scallop (Zygochlamys patagonica, King and Broderip, 1832) after the size selection process on commercial fi shing vessels”, by Bremec et al. 2004. Fish. Res. 90: 313–315. Webster, R.A., Pollock, K.H., and Simons, T.R. 2008. Bayesian spatial modeling of data from bird surveys. Journal of Agricultural, Biological and Environmental Statistics 13, 121--139. Webster, R.A., Pollock, K.H., Ghosh, S.K. and Hankin, D.G. 2008. Bayesian spatial modeling of data from unit-count surveys of fi sh in streams. Trans. Amer. Fish. Soc. 137:438-453. Yoshizaki, J., Pollock, K.H., Brownie, C., and Webster, R.A. (in press), Modeling misidentifi cation errors in capture-recapture studies using photographic tags of evolving marks. Ecology.

151 2009 IPHC ANNUAL MEETING HANDOUT SECTION II: RESEARCH PROPOSED FOR 2009

Projects to be carried out in 2009 consist of a continuation of several projects currently underway. Selected continuing projects include:

1. PIT tag scan sampling program (Project 413.00) - PIT tag recovery efforts will continue in 2009 with the scan sampling program. Scanning will also continue on the assessment survey vessels. Additionally, the staff is looking at adding a sampler to Sand Point for several months in mid-summer to enhance recoveries of 3B tags. No other changes are planned for port coverage or duration of sampling. Planning for this activity is based on a March 1 – November 15 season. At this time, 2009 is expected to be the fi nal year of scanning.

2. Water column profi lers (Project 610.11 and 610.12) – The fi rst profi ler was deployed on an IPHC survey vessel in 2003, and a second started in 2007. The profi lers measure temperature, salinity, dissolved oxygen, pH, and fl orescence and will be deployed at each station during the summer assessment survey.

3. Genetics (Project 621.00) - The study of the population genetic structure will continue in 2009 with the sample testing and analysis supervised by Dr. Lorenz Hauser (UW Marine Molecular Biology Laboratory). The FY2009 budget will allow for continued sample analysis by a technician in Dr. Hauser’s lab. We also are broadening the geographic scope by seeking samples from Atlantic halibut and also samples which have been offered by TINRO scientists in Vladivostok. Additionally, several publications covering the results to date will be authored.

4. Bycatch characterization in the Pacifi c halibut fi shery off Alaska: A fi eld test of electronic monitoring technology (Project 654.11-84) – This study, which began in 2008, is focused on examining the use of electronic monitoring technology in commercial fi shing conditions, where a broad range of environmental and physical factors affect the vessel operations. The amount of coverage achieved in 2008 was disappointingly low, and we will continue with additional effort in early 2009. Partial funding for this project has already been secured through the North Pacifi c Research Board, and is being carried out in cooperation with the NMFS Groundfi sh Observer Program, the Pacifi c States Marine Fisheries Commission (PSMFC), and the Fishing Vessel Owner’s Association of Seattle.

Staff will also continue with other long-standing projects in 2009. These include the collaborative work on contaminants with ADEC (#642.00), placement of IPHC staff on the NMFS summer trawl surveys (#604.00), and the undergraduate internship program (#618.00). Projects for estimating bycatch on the setline surveys in Area 2A with WDFW and ODFW, Area 2B with DFO, and in Area with ADF&G will continue in 2009. Additionally, IPHC will collect Pacifi c cod information for NMFS in predetermined areas within the Bering Sea. In addition, projects conducted under contract to other agencies or through research grants will be continued in 2009. IPHC port sampling activities in Alaska will continue being augmented by a grant from NMFS (Project 300.00-81), and IPHC port samplers in Alaska will collect sablefi sh

152 2009 IPHC ANNUAL MEETING HANDOUT logbook data for the NMFS Auke Bay lab (Project 628.00). NPRB has provided a grant to offset some of the costs associated with the electronic monitoring pilot study (Project 654.11-84).

Six new funded projects are proposed for 2009:

1. Coastwide use of water column profi lers – IPHC was awarded a NOAA grant to acquire a suffi cient number of water column profi lers such that they can be deployed on every IPHC survey vessel, and we will begin this coastwide collection of oceanographic data in 2009. The profi lers will measure temperature, salinity, dissolved oxygen, pH, and fl orescence and will be deployed at each station during the summer assessment survey, accompanying the existing profi lers owned by IPHC. The grant provides funding through FY2010 and covers the costs for acquiring and maintaining the devices as well as editing and publishing the data. The total value of the grant is $537,035 through September 2011. This project replaces the deferred FY2008 610.13 project with the award of the grant.

2. Bering Sea age validation study utilizing 14C radiocarbon - Radiocarbon, or 14C bomb carbon, has been used successfully in the past on several fi sh species as a validation of absolute age assignment. This project would be a collaborative study between IPHC and the NMFS Alaska Fishery Science center as a follow-up to the 2003 aging study conducted on Gulf of Alaska halibut otoliths. Costs to complete the study are based on otolith preparation and accelerated mass spectroscopy (AMS) time, with expenses shared by both agencies.

3. Holding tank experiments examining mounting protocols for external archival tags – This study will investigate alternate mounting protocols for the externally-mounted archival tags. The staff is anticipating utilizing facilities and staff at the Oregon Coast Aquarium, as was done in 2006. The results would support the anticipated future use of this type of technology.

4. Pacifi c Ocean Shelf Tracking (POST) study collaboration – IPHC staff have jointly proposed funding with POST, ADF&G, and NMFS to collaborate on the set-up of a line of remote receivers placed on the shelf in southern Area 2C. The receivers would track movements of halibut and other species carrying transponding tags. The 2009 work would include a pilot project to test the technical capability of the tag-receiver confi guration for deepwater species such as halibut and sablefi sh.

153 2009 IPHC ANNUAL MEETING HANDOUT 2009 Proposed Projects - Budget Summary

Project FY09 Project Title No. Budget Field Experiments 413.00 PIT tagging study: Year 6 of tag recovery and scanning 447,433 416.00 PIT tagging study: Double tag expt. 400 419.00 Eastern Bering Sea assessment survey 466,067 604.00 NMFS trawl survey: At-sea data collection 43,413 Field Experiments Total $ 957,313 Other Research – Continuing 610.11 Water column profi ler (General survey) 1,500 610.12 Water column profi ler (Oregon) 3,700 618.00 Undergraduate internships 9,375 621.00 Genetic population structure – lab work by UW 137,770 622.12 PAT tagging: Summer 2008 releases (Area 4) 29,835 636.00 Histology: Analysis of gonad staging 24,500 642.00 Assessment of mercury and contaminants in Pacifi c halibut 1,000 650.12 Archival tags: Pilot studies (2008 releases in Area 2B) 18,000 653.00 Species identifi cation of amphipods frequenting Pacifi c halibut 1,000 654.11-84 Alaska fi shery electronic monitoring: Pilot study 14,000 Subtotal $ 240,680 Other Research – New 1 Water column profi ler (Coastwide) 395,301 2 Bering Sea age validation study utilizing 14C radiocarbon 14,900 3 Archival tags: Mounting protocols 37,283 4 POST study: IPHC participation 40,000 Subtotal $ 487,484 Other Research Total $ 728,164 GRAND TOTAL $ 1,685,447 Note: Values shown do not include any revenues generated from the sale of fi sh or other cost offsets.

154 2009 IPHC ANNUAL MEETING HANDOUT 2009 Proposed Projects - Budget Summary (cont’d)

Other 2009 Research – Contracts and Grants Granting agency shown in parentheses Project FY09 Project Title No. Income 300.00-81 AK port sampling grant (NMFS) $ 200,757 610.13-81 Water column profi ler grant (NMFS) 395,301 628.00 AK catcher vessel logbook and sablefi sh data collection (NMFS) 40,000 654.11-84 Alaska fi shery electronic monitoring – pilot study (NPRB) 32,638 GRAND TOTAL $ 668,696

155 2009 IPHC ANNUAL MEETING HANDOUT Continuing Research

Project 413.00: PIT tagging study: Year 6 of tag recovery and scanning Cost: $ 447,433 Start Date: 2003 Anticipated ending: 2009 Personnel: J. Forsberg, C. Blood, G. Williams, S. Hare, A. Ranta, scan samplers

Scanning for PIT tags will continue in 2009. IPHC will hire samplers for Alaskan ports, while contracting with AMR for the Canadian ports and continuing to seek state and tribal assistance in Area 2A. Sampler duties include scanning commercial deliveries for PIT tags, and conducting regular tests of detection and piece (fi sh) counts to measure accuracy of sample data. Project costs are expected to be about the same as last year. Modest increases are expected in salaries (cola), our contract with AMR for sampling in Area 2B, and the possible addition of scanning in Sand Point.

Project 419.00: Eastern Bering Sea assessment survey Cost: $ 466,067 Start Date: 2006 Anticipated ending: continuing Personnel: C. Dykstra, T. Geernaert, E. Soderlund, sea samplers

IPHC proposes to conduct a standardized grid survey of the eastern Bering Sea to better characterize the biology, relative abundance, and range of halibut in that area. The survey also serves to calibrate results from the long-term NMFS trawl survey, to provide an index of halibut abundance in the area. The area, which was fi rst surveyed in 2006, stretches from the shelf edge eastward to inner Bristol Bay and from the Alaskan Peninsula northward to 60 nmi north of St. Matthew Island (55° 20’ N to 61° 30’ N and from 159° 33’ W to 177° 23’ W). To survey such a vast area effi ciently, the systematic station layout has been altered to enable completion of these charter regions in a time frame comparable to that of the standard survey. Additionally, in the interests of effi ciency, trip length restrictions as well as requirements for fi sh retention have been altered. For the majority of the area covered, paired stations, 10 nmi apart, are placed at 60 nmi by 60 nmi intervals. One station of each pair corresponds to a trawl station on the annual National Marine Fisheries Service (NMFS) Eastern Bering Sea trawl survey, thereby enabling comparison between the two surveys. An additional 29 stations were placed on the standard 10 nmi by 10 nmi confi guration of our traditional surveys around St. Matthew (nine stations), St. Paul (10 stations), and St. George (10 stations) Islands. These island stations were incorporated into the annual surveys in 2007 and 2008. The entire area was split into three regions for bidding purposes, each containing stations from both design layouts. Bids for this survey were requested in 2008 but suffi cient bids were not received.

156 2009 IPHC ANNUAL MEETING HANDOUT Project 604.00: NMFS trawl survey: At-sea data collection Cost: $ 43,413 Start Date: 1996 Anticipated ending: Continuing Personnel: L. Sadorus, A. Ranta, S. Hare

A series of NMFS trawl survey data on halibut, parallel to our setline data, is extremely valuable to IPHC as a second fi shery-independent data source for stock assessment. Trawl data are particularly useful because they include large numbers of juveniles (ages 3-7 yr) that do not appear in large numbers in the setline survey. Otoliths have been collected on the NMFS surveys since 1996 and provide relevant age information. These data are incorporated into a copy of the NMFS haul data, expanded to estimates of relative abundance and age/size composition by IPHC area (NMFS calculates estimates by INPFC area), and stored in a database at IPHC. Project cost is comprised of personnel and travel. In 2009, NMFS will be conducting both the Bering Sea and Gulf of Alaska shelf surveys and the IPHC plans to have one biologist aboard each survey.

Project 610.11: Water column profi ler project (General survey) Cost: $ 1,500 Project 610.12: Water column profi ler project (Oregon) Cost: $ 3,700 Start date: 2000, 2007 (respectively) Anticipated ending: Continuing Personnel: L. Sadorus, S. Hare, P. Stabeno (NMFS PMEL)

The IPHC maintains one of the most extensive sampling platforms in the north Pacifi c. This platform provides enormous potential for collection of valuable oceanographic data. In particular, understanding the dynamics of the structure of the mixed layer depth – a major GLOBEC goal - requires in situ vertical profi ling. Since 2001, IPHC has successfully deployed a SeaBird SBE- 19 water column profi ler during the annual stock assessment survey (#610.11). A second profi ler was added to the program in 2007 (#610.12). Annual costs are directed towards maintenance and calibration of the profi lers.

Project 618.00: Undergraduate Internship Cost: $ 9,375 (One intern) Start Date: 2002 Anticipated duration: Continuing Personnel: L. Sadorus, T. Loher, other staff support as needed

One undergraduate will be selected through the intern/co-op programs at regional universities and colleges to do a combination of offi ce and at-sea work based out of the Commission offi ces during the summer months. The program includes various pre-determined offi ce tasks as well as being assigned a research project then designing and executing said project. A fi nal report and presentation are given at the conclusion of the employment term.

157 2009 IPHC ANNUAL MEETING HANDOUT Project 621.00: Genetic population structure of Pacifi c halibut assessed via nuclear microsatellite diversity – lab work by UW Cost: $ 137,770 Start: 2002 Anticipated Ending: Continuing Personnel: T. Loher, L. Hauser (UW-MMBL), other staff as needed

The eastern north Pacifi c halibut resource is presently managed under the assumption that a single fully mixed population exists from California through the eastern Bering Sea. This belief rests largely upon studies that indicate that drift of larvae to the northwest is balanced by migration of juveniles and adults to the southeast, over broad geographic expanses. In 2002, a project was initiated to investigate genetic population structure in the northeast Pacifi c using non-selected nuclear microsatellite markers, and in 2004 the study was expanded to spawning groups from British Columbia, the central Gulf of Alaska, and southeast Bering Sea. Although initial analysis suggested population differentiation, interpretation of results was complicated by very low FST values and the fact that genetic studies conducted without temporal replicates are in danger of detecting false positives. In February of 2007, winter charters were conducted to resample the locations visited in 2004, and analysis of those samples continued through December, 2007. Increasing the number of microsatellites analyzed and fi lling gaps in the data set resulted in little change from prior work, although increasing sample sizes did result in increased levels of signifi cance. At the end of 2007 we deemed the microsatellite work to be largely complete, with the conclusion that these markers may simply not be very powerful for detecting population structure in a species, such as halibut, with large amounts of larval mixing. At the same time, we lost our project technician. However, in May of 2008 we were fortunate to fi nd a full-time post-doctoral researcher (Dr. Heather Galindo, formerly at Stanford University) to dedicate herself to the program. During the relatively short period of her employment, she has managed to make considerable progress investigating some other genetic markers that may be more powerful than microsatellites. She has screened 25 “expressed sequence tags”, found in regions of DNA that are responsible for coding proteins and that may therefore be driven by evolution in different directions in different ocean basins. She has also sequenced four mitochondrial DNA (mtDNA) regions, which are maternally-inherited and have proven useful in other species for investigating sex-biased migration and demographics in relation to climate change. This work is expected to continue through this fall and winter.

Project 622.12: PAT tagging: summer 2008 releases (Area 4) Cost: $ 29,835 Start: 2008 Anticipated Ending: 2010 Personnel: T. Loher, A. Seitz (UAF), sea samplers

A PAT-tagging project was initiated in 2008 to investigate why so few PIT tags were recovered from Area 4 and investigate the possibility that eastward migration is higher south of Unimak Pass than north of it. During the 2008 setline survey, 115 halibut were tagged throughout the Bering Sea and Aleutian Islands with tags programmed to report location exactly 365 days after tagging. At the beginning of October, one tag had been recovered by the targeted fi shery and another ten

158 2009 IPHC ANNUAL MEETING HANDOUT had released prematurely, leaving a total of 104 tags still in the water. Project costs in 2009 are for anticipated satellite transmission time when the tags pop to the surface 365 days following release.

Project 636.00: Histology: Analysis of gonad staging Cost: $ 24,500 Start: 2004 Anticipated Ending: Continuing Personnel: T. Geernaert, C. Dykstra, other staff as needed

The IPHC Stock Assessment surveys assess maturity of halibut based on visual criteria established in the early 1990’s and modifi ed in 1995. These survey data combined with the age data are important components in the stock assessment model. Four maturity stages are presently assigned to female halibut; immature (F1), maturing (F2), spawning (F3) and resting (F4). Once a female halibut has spawned, the gonad transitions to a resting phase, back to maturing, and then to spawning again. Our criteria for classifi cation also assume that the immature (F1) stage is only seen with immature fi sh but we are seeing anomalies during the survey that could question this assumption. Mature females are seen as small as legal size (82 cm) but, area-wide, there have been several large 100+ cm females whose gonadal characteristics classify them as immature (never spawned). The SSA survey data also suggest that fi sh in the southern latitudes (Area 2B) mature earlier and possibly spawn earlier that fi sh in the northern latitudes (Area 3A and west). The timing and duration of these events are not clearly understood. We would like to re-evaluate our classifi cation criteria and examine the stages and gonadal tissue development more closely. In 2003 preliminary histological work on the female gonads was initiated. We developed a sampling design and collection protocols for the 2004 surveys. In 2004, during winter and summer surveys, female gonads from three different regions, in each stage of development, were collected. Three different histological subsamples have been prepared and we are presently standardizing the sample sites on the gonad for the fi nal slide preparation. We have collected nearly 240 gonad pairs and will be analyzing multiple sites from each sample.

Project 642.00: Assessment of mercury and contaminants in Pacifi c halibut Cost: $ 1,000 Start Date: 2002 Anticipated ending: Continuing Personnel: C. Dykstra, Alaska Department of Environmental Conservation (ADEC)

For the last few years, health offi cials and media have raised the profi le of pollutant contamination in fi sh (methyl mercury, PCB’s, pesticides). Since 2002, the IPHC has been working collaboratively with the Alaska Department of Environmental Conservation (ADEC) to collect halibut tissue samples to be analyzed for heavy metal and organic pollutant loading. The principal results from the 2002 collection led the Alaska Division of Public Health in 2003 to conclude that the concentrations of heavy metals in Alaskan Pacifi c halibut are not a public health concern. In 2004 the fi rst results regarding organic pollutants (PCB’s, pesticides) were released demonstrating

159 2009 IPHC ANNUAL MEETING HANDOUT that halibut had the lowest concentrations of the fi ve species (including salmon and sablefi sh) examined. The Alaska Division of Public Health updated their advice on fi sh consumption in 2007 with some restrictions on the number of meals of halibut for women of child bearing age and young children. Since 2002 the IPHC has submitted 981 samples for testing by ADEC. The mean level of total mercury for these samples has been 0.340 ppm (for comparison the FDA limit of concern for methyl mercury is 1.000 ppm, the EPA and the CFIA level of concern is 0.500 ppm) ranging from non-detectable to 1.947 ppm. The IPHC and ADEC are continuing to qualify the data with physical parameters (age, size, and weight) and additional analyses will be done on the samples. ADEC and EPA planned on going ahead with this study regardless of IPHC input. Our involvement in the project has allowed us to provide input on study design, sampling protocols in the fi eld, etc., which will make the resultant information much more robust. Sampling continued in 2008 with a targeted collection of 60 samples (15 fi sh between 10- 20 lbs, 15 fi sh between 20-40 lbs, 15 fi sh between 40-100 lbs, and 15 fi sh greater than 100 lbs.) from each of three regions (Sitka, Prince William Sound, and upper Cook Inlet) during the setline survey. ADEC has expressed interest in further assessments of contaminant occurrence in halibut in 2009.

Project 650.12: Archival tagging: Pilot studies (Area 2B releases) Cost: $ 18,000 Start Date: 2006 Anticipated ending: Continuing Personnel: T. Loher

This study is investigating migratory behavior and environmental conditions experienced by two components of stock: small adult (primarily male) and late pre-recruit halibut, as well as larger adults including reproductive females. The work is a complement to PAT (satellite) tagging studies and seeks to expand our knowledge to components of the population that have not been studied with PAT tags due to apparent size constraints (i.e., males and pre-recruits) and to obtain multi-year data for larger fi sh. The objectives for each stock component are slightly different, but do not require separate studies. Externally attached, rather than surgically implanted, archival tags are being used. The tags were applied to all females above 90 cm and all fi sh above 100 cm during August-September, 2008 in Area 2B. Project costs in 2009 are for the anticipated recoveries. Premium rewards are being offered to encourage recoveries.

Project 653.00: Species identifi cation of amphipods frequenting Pacifi c halibut Cost: $ 1,000 Start Date: 2006 Anticipated ending: Continuing Personnel: B. Leaman, E. Soderlund

The project intends to document the occurrence and virulence of attacks by predatory amphipods on halibut caught on IPHC surveys and, by inference, the commercial fi shery. The

160 2009 IPHC ANNUAL MEETING HANDOUT commercial industry suffers annual losses of product due to amphipod predation and must adjust its fi shing locations and practices in response to predation. Harvester discussions indicate that predation sites are both known and ephemeral, and the virulence may vary interannually at a given site. The specifi c identity of the amphipods has not been established and it is probable that more than one species is involved. Harvesters are interested in both documentation of predation areas for avoidance, as well as gaining an understanding of the dynamics of the species at given sites, i.e., are there cycles of abundance that respond to other factors. Data were collected on all stations during the 2004, 2005, and 2006 stock assessment surveys as part of standard protocol, recording incidence of sand fl ea predation, and the extent and virulence of the predation. The 2006 was the last year of data collection for this stage of the project. The 2007 summer intern performed initial analysis of interannual occurrence and virulence. Additional work will be directed at correlated variables.

Project 654.11-84: Alaska fi shery electronic monitoring – pilot study Cost: $ 14,000 in 2009 ($94,300 for total project) Revenue: $43,220 (grant from NPRB) Start Date: 2008 Anticipated ending: 2009 Personnel: G. Williams, B. Leaman, B. Karp and J. Cahalan (NMFS Observer Program)

Bycatch rates of other species in the Pacifi c halibut fi shery are not well estimated, so electronic monitoring has been suggested as one option to collect such information. The majority of vessels operating in this fi shery are not required to have observer monitoring due to their size. The long term focus of this research is to improve the understanding of the ecosystem impacts of halibut fi shing through improved monitoring of longline fi shery bycatch and to provide data on mortality of bycatch species for input to stock assessments. In this study, we will be comparing and evaluating the effectiveness of electronic monitoring (EM) and the current North Pacifi c Groundfi sh Observer Program (NPGOP) monitoring methods to operate effectively in a commercial longline (hook- and-line) setting in Alaskan waters. This was a cooperative study with the commercial fi shing industry and relied on our ability to sample on various vessel confi gurations. The project was not as successful in 2008 in getting many vessels to participate as expected. Only four vessels, three in the Area 3A and 1 in Area 4, took an observer and accompanying video system. Over 230 sets were monitored among the four vessels. Archipelago Marine Research (Victoria BC) is currently analyzing the video data for the subsequent comparison with the observer sampling. Additional vessel monitoring has been approved for spring 2009, which should provide a more complete data set for analysis during 2009.

161 2009 IPHC ANNUAL MEETING HANDOUT Proposed New Research

1. Coastwide deployment of water column profi lers Cost: $ 395,301 Revenue: $395,301 (NOAA Grant) Start date: 2009 Anticipated ending: Continuing, although the grant ends in Sept., 2011 Personnel: L. Sadorus, S. Hare, P. Stabeno (NMFS PMEL)

In 2008, IPHC was awarded a grant from NOAA for the purchase of water column profi lers for each survey vessel and therefore coastwide collection of oceanographic data. The profi lers will measure temperature, salinity, dissolved oxygen, pH, and fl orescence at each station. These data will provide an annual snapshot of near shore oceanic conditions as well as valuable observational data for modeling and biological studies on recruitment and growth variability. IPHC staff will be working with the Pacifi c Marine Environmental Lab (PMEL) on data processing and storage. The grant provides funding through FY2011 and covers the costs for acquiring and maintaining the devices as well as editing and publishing the data. The total value of the grant is $537,035 through September 2011. This project replaces the deferred FY2008 610.13 project with the award of the grant.

2. Bering Sea age validation study utilizing 14C radiocarbon Cost: $ 14,900 Start Date: 2009 Anticipated ending: 2010 Personnel: S. Wischniowski, T. Loher, NMFS personnel

Radiocarbon, or 14C bomb carbon, has been used successfully in the past on several fi sh species as a validation of absolute age assignment. This project would be a collaborative study between IPHC and the NMFS Alaska Fishery Science center as a follow-up to the 2003 aging study conducted on Gulf of Alaska halibut otoliths. Costs to complete the study are based on otolith preparation and accelerated mass spectroscopy (AMS) time, with expenses shared by both agencies.

3. Archival tags: Holding tank experiments examining mounting protocols Cost: $ 37,283 Start Date: 2009 Anticipated ending: 2010 Personnel: T. Loher

This study will investigate alternate mounting protocols for the externally-mounted archival tags. The 2008 releases in Area 2B were our fi rst experience with using an external mount, and that process suggested some revisions and improvements could be possible which would reduce any

162 2009 IPHC ANNUAL MEETING HANDOUT effect the tags may have on the fi sh’s behavior. Additional improvements to tag design may also be helpful in creating a different mounting device. A total of 30 halibut ranging from 75-90 cm FL will be captured via hook-and-line and transported live to an appropriate holding facility. The IPHC staff is anticipating utilizing facilities and staff at the Oregon Coast Aquarium, as was done in 2006. Following tagging, fi sh will be reared for 12-18 months, treated regularly for parasites, examined regularly to assess healing and/or relative infection rates among mounting types, and behavior monitored. At the end of the holding period, fi sh will be measured to assess relative growth among treatment groups, and tags will be removed to examine the effects of the tag mounts on the tissue and musculature at the attachment site, or internal interactions in the case of an internal-external-streamer modifi cation. The results would support the anticipated use of this type of technology in subsequent years.

4. POST Study: IPHC participation Cost: $ 40,000 Start Date: 2009 Anticipated ending: Continuing Personnel: B. Leaman, other staff as needed, ADF&G, NMFS

The Pacifi c Ocean Shelf Tracking (POST) Project, part of the Census of Marine Life and fl agship for the international Ocean Tracking Network, provides a tool for scientists to track the movement of marine animals along the West Coast of North America. POST uses acoustic transmitters implanted in various species and a series of receivers running in lines across the continental shelf to provide movement data. A recent call for proposals by the North Pacifi c Research Board for an Integrated Ecosystem Research Program (IERP) creates a funding opportunity for a collaborative effort by IPHC with NMFS and ADF&G to become participants in the Pacifi c Ocean Shelf Tracking (POST) initiative. The POST program seeks agencies to purchase the receivers to equip a sensor track, or lines, with matching funds available. For IPHC, the idea is to conduct a proof-of-concept study, to see if the receivers can indeed pick up a demersal fi sh like a halibut. ADF&G and NMFS are interested in the applications to sablefi sh. The agency group agreed the benefi ts were suffi cient to move forward with a range test - as a fi rst step - focused primarily in the area around Cape Ommaney. Funding would be used for purchase of acoustic tags and/or deployment of receivers. If this concept proves sound, a joint proposal by the three agencies to the NPRB and POST would occur for 2010.

163 2009 IPHC ANNUAL MEETING HANDOUT Other 2009 Research – Contracts and Grants

Project 300.00-81: Alaska port sampling Cost: Staff salaries Revenue: $ 200,757 Granting agency: NMFS Start Date: 2002 Anticipated ending: Continuing Personnel: H. Gilroy, M. Larsen, L. Hutton

The commercial fi shery port sampling program hires samplers to collect otoliths, halibut lengths, fi shing logbook information and landed weight data. The U.S. program includes staffi ng eight Alaskan ports and Bellingham, Washington. The samplers act as the liaison between the fi shing industry and the Commission staff in Seattle. The Commission is responsible for the overall assessment and management of the halibut fi shery and the data collected are necessary for stock assessment. The U.S. Government adopted the Individual Fishing Quota (IFQ) allocation program in 1995. This grant provides funds to the IPHC for the incremental cost to the Commission sampling program due to the IFQ program. The grant is generated from the NMFS IFQ Fee Collection Program.

Project 610.13-81: Coastwide deployment of water column profi lers Cost: $ 395,301 Revenue: $395,301 (NOAA Grant) Start date: 2009 Anticipated ending: Continuing, although the grant ends in Sept., 2010 Personnel: L. Sadorus, S. Hare, P. Stabeno (NMFS PMEL)

In 2008, IPHC was awarded a grant from NOAA for the purchase of water column profi lers for each survey vessel and therefore coastwide collection of oceanographic data. The profi lers will measure temperature, salinity, dissolved oxygen, pH, and fl orescence at each station. These data will provide an annual snapshot of near shore oceanic conditions as well as valuable observational data for modeling and biological studies on recruitment and growth variability. IPHC staff will be working with the Pacifi c Marine Environmental Lab (PMEL) on data processing and storage. The grant provides funding through FY2011 and covers the costs for acquiring and maintaining the devices as well as editing and publishing the data. The total value of the grant is $537,035 through September 2011. This project replaces the deferred FY2008 610.13 project with the award of the grant.

164 2009 IPHC ANNUAL MEETING HANDOUT Project 628.00: Alaska catcher vessel logbook and sablefi sh data collection Cost: Staff salaries Revenue: $ 40,000 Granting agency: NMFS Start Date: 1999 Anticipated ending: Continuing Personnel: L. Hutton, H. Gilroy, A. Taheri, port samplers

IPHC and NMFS Auke Bay Lab (ABL) have a sablefi sh data collection program. The program was reviewed and modifi ed in 2003/2004 to meet the IPHC confi dentiality policy and to encompass all vessels rather than just vessels greater than 60 feet. Under a Statement of Work, NMFS contracted IPHC staff to interview the IFQ fi shers to review and collect the sablefi sh information in addition to the halibut information. Logbook data are entered by IPHC staff, matched with landings records, and provided electronically with a summary to the ABL scientists. In the summarized data, the vessels are assigned a unique code to preserve confi dentiality.

Project 654.11-84: Alaska fi shery electronic monitoring – pilot study Cost: $ 14,000 in 2009 ($94,300 for total project) Revenue: $ 32,638 in 2009 (Grant from NPRB for $43,220) Start Date: 2008 Anticipated ending: 2009 Personnel: G. Williams, B. Leaman, B. Karp and J. Cahalan (NMFS Observer Program)

This study compares and evaluates the effectiveness of electronic monitoring (EM) and the current North Pacifi c Groundfi sh Observer Program (NPGOP) monitoring methods to operate effectively in a commercial longline (hook-and-line) setting in Alaskan waters. This was a cooperative study with the commercial fi shing industry. IPHC is partnering with the NMFS North Pacifi c Groundfi sh Observer Program and Pacifi c States Marine Fish Commission in this study.

Research Conducted Without Direct Funding

1. The 2008 stock assessment Cost: Staff salaries Personnel: S. Hare, B. Clark

The annual stock assessment process comprises a large amount of work including preparation of IPHC data, estimation of bycatch by length in other fi sheries, model development and validation, model fi tting, examination of residuals, comparison of alternative model specifi cations, sensitivity tests, evaluation of harvest strategy, incidental analyses, and reporting.

165 2009 IPHC ANNUAL MEETING HANDOUT 2. Development of IPHC harvest policy Cost: Staff salaries Personnel: S. Hare, J. Valero

Staff quota recommendations are calculated by applying a judiciously chosen harvest rate to an estimate of present exploitable biomass. The constant harvest rate policy was developed on the basis of its performance over a long time horizon and with the explicit goal of avoiding reaching the minimum stock sizes seen in the 1930s and 1970s. In 2003 the staff proposed a conditional constant catch policy under which total removals would be capped at a chosen ceiling level at high biomass levels, while a constant harvest rate policy would continue to be employed at low and intermediate stock levels. The Commission did not adopt the proposed policy at the 2004 annual meeting, and a staff/industry workshop on harvest policy alternatives in September 2004 showed general satisfaction with the present policy, as mediated in practice by the judgment exercised by the Director in developing staff recommendations and by the Commission in fi nally setting catch limits. In 2004, an explicit lower limit on spawning biomass and a threshold below which the harvest rate will be reduced—were added to the constant harvest rate policy. As a result the target constant harvest rate for the core IPHC areas (2B, 2C, and 3A) was lowered to 22.5% from 25%. The staff will continue to evaluate the constant harvest rate policy. In particular, we will recalculate the optimum harvest rate itself in light of our present understanding of stock dynamics and new information on commercial length-specifi c selectivity coming from the PIT tag experiment.

3. Development of a robust management procedure Cost: Staff salaries Personnel: S. Hare, B. Leaman, J. Valero

Staff catch limit recommendations are derived from the annual stock assessment by applying a constant harvest rate to the estimates of exploitable biomass, in the belief that the assessment model is correctly specifi ed and the estimates are accurate. In fact there are a number of structural uncertainties about the model, and the assessment itself has become highly complicated, so it is vulnerable to any small error in data compilation or programming. The aim of this project is to develop a procedure for deriving catch limit recommendations that would achieve the desired harvest policy, potentially relying on much simpler calculations and at the same time be effective across a range of uncertainties about stock, fi shery and management behavior. Such procedures have been developed for other fi sheries and it is appropriate to investigate their application to halibut management.

4. Estimation of halibut abundance from mark-recapture data Cost: Staff salaries Personnel: R. Webster, B. Leaman, S. Hare

The IPHC has conducted many tagging programs since the 1920s. IPHC has also conducted at least fi ve reviews of these programs, again with differing objectives. However, many of these reviews did not account for the issues of non-reporting or differential reporting of tags by areas,

166 2009 IPHC ANNUAL MEETING HANDOUT fi shing effort effects on recovery probabilities, the relationship of initial tag releases and the density of fi sh in given areas, and the effect of seasonal migratory patterns on the analysis of recoveries were not always considered. A changed paradigm for the area-specifi c impacts of juvenile bycatch, questions concerning the effects of changing seasonal distribution of fi shing effort, potential halibut distribution changes with climatic shifts, and the utility of juvenile surveys in specifi c areas have all prompted concerns about halibut movements. The staff marked the catch of three skates at each survey station coastwide in 2003 and in Areas 2B and 3A in 2004. Preliminary analysis of the 2004 recoveries showed good agreement with the stock assessment in Areas 2B and 2C, but farther west the mark-recapture estimates were much higher than the assessment estimates. Recoveries in 2005-2007 followed a similar pattern. The 2008 recoveries will be added to the analysis this year.

5. Whales sightings by the commercial fi shery off Alaska while hauling gear Personnel: L. Hutton, port samplers Cost: Staff Salaries Start: 2007 End Date: Under review

Beginning in 2007, IPHC U.S. commercial fi shery logbooks were modifi ed to facilitate the collection of whale sightings while hauling gear. This is a pilot project to determine if sighting information can be collected and if accurate information can be obtained. This information could help to set a baseline for sightings. A review of other programs will be conducted to see where information is collected, available, and stored. Additionally whale sighting data collected during the grid survey can be reviewed along with known depredation activity on the grid survey sets. Data will be reviewed and results provided in the 2009 RARA.

6. Seabird occurrence project Cost: Staff salaries Start Date: 2002 Anticipated ending: Continuing Personnel: T. Geernaert, Washington State Sea Grant

During the stock assessment surveys, sea samplers count the number of seabirds in the vicinity of the vessels following gear retrieval. Sampling after the haul addresses the question of where and when certain seabird species occur. These data have been used to identify appropriate seabird deterrent requirements in certain geographic locations. Data have also been collected, using the same protocol, on the NMFS and ADF&G sablefi sh surveys. IPHC has developed a database to store seabird occurrence data and the collection project is ongoing.

167 2009 IPHC ANNUAL MEETING HANDOUT 7. Seabird data repository (Project 643.00) Cost: Staff salaries Start Date: 2005 Anticipated ending: Continuing Personnel: T. Geernaert

This project encompasses the storage by IPHC of various types of seabird data collected on agency surveys, including the seabird occurrence project conducted on the assessment surveys. Although IPHC has been collecting these data on its assessment survey since 2002, other agencies are just beginning the same procedure. IPHC’s head-start with these types of data led other programs to seek the effi ciencies provided by IPHC in data management for optimal use by analysts and managers. A grant from Washington Sea Grant funded this activity in previous years.

8. Estimates of bycatch on the assessment surveys Cost: Staff Salaries Revenue: To be determined Start Date: 2003 Anticipated ending: Continuing Personnel: C. Dykstra, Survey Team, DFO, ADF&G, WDFW and ODFW personnel

Area 2A Since 2002, the IPHC has worked cooperatively with both the Washington Department of Fish and Wildlife (WDFW) and Oregon Department of Fish and Wildlife (ODFW) to collect rockfi sh (Sebastes spp.) bycatch data. All rockfi sh caught on operations in 2A are retained and marked externally with a Floy T-bar anchor tag and the tag number is recorded with the set (and recently the skate number) information. All marked fi sh are retained so state biologists can collect additional data shore-side. Marketable fi sh are sold. IPHC then provides each agency with the effort information collected as part of the normal survey data collection. In 2008, IPHC worked with WDFW and ODFW to fi sh supplemental stations designed to further enhance the understanding of rockfi sh status in these areas. In each state, the stations’ locations and design were specifi ed by the state agency involved. Eighteen stations were fi shed off Washington (a continuation of similar studies from the previous 2 years) and twenty stations off Oregon. Three skates of gear were fi shed at each station as a precautionary approach due to the exploratory nature of these stations and concerns about overfi shing yelloweye rockfi sh (Sebastes ruberrimus). Activities at each station were identical to those on standard IPHC stations except that halibut were only sampled for length and prior hooking injury and then released alive; rockfi sh were handled as described above. A summary of this project will be submitted to each state by the end of November 2008. Depending on those results, there may be similar cooperative work conducted in future years.

Area 2B In 2008, IPHC worked with the Department of Fisheries and Oceans Canada (DFO) to provide a third biologist on IPHC survey vessels to collect hook by hook occupancy information for all species, and otoliths, maturities, and lengths for rockfi sh except thornyheads. This is the sixth year of this cooperative program and continued collaboration is anticipated.

168 2009 IPHC ANNUAL MEETING HANDOUT Area 2C and eastern 3A In 2008, IPHC worked cooperatively with Alaska Department of Fish and Game (ADFG) to provide a third biologist on IPHC survey vessels in the Fairweather, Sitka, Ommaney and Ketchikan charter regions to collect hook by hook occupancy information for all species, and otoliths, sex, and lengths for yelloweye rockfi sh. This project built upon cooperative work started with ADFG in 2007 and future collaboration is anticipated.

Area 4 IPHC collected length frequency data on incidentally caught Pacifi c cod (Gadus macrocephalus) in the 4A Edge and 4D Edge charter regions. This project was initiated at the request of NMFS- AFSC Pacifi c cod assessment team and part of a developing effort to collect bycatch information on Pacifi c cod in the western regions of our survey, where it makes up the largest component of our survey bycatch.

9. Electronic reporting project for commercial landings in Alaska Cost: $30,000 (covered under Catch Statistics budget: A30-7131-30) Start Date: 2002 Anticipated ending: Continuing Personnel: H. Gilroy, L. Hutton, T. Kong, A. Tesfatsion, H. Tran

IPHC, ADF&G, and NMFS staffs have continued to refi ne the web-based Interagency Electronic Reporting System (IERS). For halibut, the system reduces duplicative reporting resulting from the current requirements of completing ADF&G fi sh tickets, NMFS/RAM quota share reports, and has been operational since May 2006. The application (eLandings) records data elements required by regulations, prints fi sh tickets, and connects with the NMFS quota share database. The appropriate data from IERS is being sent to the agencies for their internal databases. Industry personnel and agency staff have provided feedback on the operation and the application is continuously being modifi ed including incorporation additional fi sheries and tender landings. Agency staffs have been to yearly trainings or workshops on the program. In 2009, the focus will be on continued training with the processors. Costs represent system maintenance costs, software purchase and development, steering committee and travel costs.

10. Electronic logbooks Cost: $50,000 (Covered under Catch Statistics budget: A10-7131-30) Start: 2009 (postponed from 2008) End Date: Pilot project Personnel: H. Gilroy, L. Hutton, K. MacTavish

IPHC and NMFS/AKR are collaborating to determine the feasibility of an electronic logbook and establish the specifi cations needed for contractors. The Commission will also be reviewing other programs to decide if another geographic location (Area 2B or 2A) is an appropriate place to start an electronic logbook program.

169 2009 IPHC ANNUAL MEETING HANDOUT Research Topics for 2010

1. Design of experiment to examine coastwide catchability of longline survey gear Personnel: B. Leaman, S. Hare, R. Webster, J. Valero

The issue of common catchability for IPHC setline survey gear across all areas of the coast is a key component of the use of survey data for biomass apportionment. While we have no indication that catchability is highly biased in any area, we acknowledge that the existing data comparisons with which catchability has been assessed are highly variable. This high variance contributes to questions about the validity of the assumed commonality of catchability across the coast. Historical evaluations of catchability have employed trawl-setline comparison fi shing. However, those experiments produced the high variance noted, plus the fact that trawls cannot be used in all areas where longline fi shing occurs and that trawls have their own separate gear selectivity by halibut size. Thus, the trawl-setline comparisons are both spatially and selectivity compromised. It therefore is highly unlikely that any experiment using a trawl-setline comparison will produce results that will be either convincing or spatially inclusive. The staff believes that a different approach to estimating setline catchability is required. Its fundamental feature must include a method to obtain an unbiased estimate of halibut density, wherever halibut are fi shed, so that fi shing using IPHC survey gear can then be evaluated coastwide. At present, we do not know what that method might be. It could include, among other approaches, remote sensing, short-term tag-recapture fi shing, acoustic tagging, or other technology- based methods. We believe that such an experiment demands careful planning and may include pilot studies of candidate methods. In any event, the results of such a coastwide experiment are important to the Commission’s management process and it should be only be conducted after adequate planning, and using methods with a higher probability of producing accurate results than those methods presently available.

170 2009 IPHC ANNUAL MEETING HANDOUT Appendix II. Tide Predictions

A. Tillamook Bay, OR – May and June, 2009 B. Kodiak, AK – February 15-March 21 (Source: http://tbone.biol.sc.edu/tide/) A. Tillamook Bay, Bay City May 2009

High High High Low Low Date Day Time/Height Time/Height Time/Height Time/Height Time/Height May 1 Fri 00:57 / 2.48 ft 06:15 / 6.28 ft 13:56 / -0.47 ft 20:29 / 5.53 ft May 2 Sat 02:23 / 2.26 ft 07:34 / 5.78 ft 14:58 / -0.11 ft 21:24 / 5.86 ft

May 3 Sun 03:45 / 1.78 ft 08:58 / 5.45 ft 15:57 / 0.24 ft 22:11 / 6.25 ft May 4 Mon 04:54 / 1.14 ft 10:17 / 5.31 ft 16:50 / 0.59 ft 22:54 / 6.62 ft May 5 Tue 05:52 / 0.48 ft 11:27 / 5.31 ft 17:39 / 0.94 ft 23:32 / 6.92 ft May 6 Wed 06:41 / -0.11 ft 12:27 / 5.36 ft 18:23 / 1.29 ft May 7 Thu 00:07 / 7.11 ft 07:25 / -0.57 ft 13:21 / 5.42 ft 19:05 / 1.60 ft May 8 Fri 00:41 / 7.18 ft 08:05 / -0.86 ft 14:10 / 5.44 ft 19:45 / 1.88 ft May 9 Sat 01:15 / 7.15 ft 08:43 / -0.99 ft 14:55 / 5.42 ft 20:23 / 2.12 ft

May 10 Sun 01:48 / 7.02 ft 09:22 / -0.98 ft 15:39 / 5.34 ft 21:01 / 2.32 ft May 11 Mon 02:22 / 6.83 ft 10:00 / -0.86 ft 16:22 / 5.22 ft 21:39 / 2.48 ft May 12 Tue 02:58 / 6.58 ft 10:40 / -0.66 ft 17:07 / 5.08 ft 22:20 / 2.61 ft May 13 Wed 03:36 / 6.29 ft 11:21 / -0.41 ft 17:55 / 4.96 ft 23:06 / 2.71 ft May 14 Thu 04:18 / 5.94 ft 12:06 / -0.14 ft 18:45 / 4.92 ft May 15 Fri 00:02 / 2.76 ft 05:06 / 5.56 ft 12:52 / 0.14 ft 19:34 / 4.97 ft May 16 Sat 01:09 / 2.70 ft 06:03 / 5.15 ft 13:40 / 0.42 ft 20:21 / 5.15 ft

May 17 Sun 02:24 / 2.47 ft 07:13 / 4.78 ft 14:29 / 0.69 ft 21:01 / 5.42 ft May 18 Mon 03:34 / 2.04 ft 08:31 / 4.55 ft 15:18 / 0.95 ft 21:38 / 5.77 ft May 19 Tue 04:32 / 1.45 ft 09:48 / 4.52 ft 16:05 / 1.21 ft 22:12 / 6.18 ft May 20 Wed 05:22 / 0.76 ft 10:57 / 4.66 ft 16:51 / 1.46 ft 22:47 / 6.62 ft May 21 Thu 06:08 / 0.04 ft 11:58 / 4.90 ft 17:37 / 1.68 ft 23:23 / 7.06 ft May 22 Fri 06:52 / -0.65 ft 12:54 / 5.17 ft 18:23 / 1.87 ft May 23 Sat 00:02 / 7.46 ft 07:36 / -1.23 ft 13:46 / 5.41 ft 19:09 / 2.02 ft

May 24 Sun 00:43 / 7.77 ft 08:21 / -1.66 ft 14:37 / 5.59 ft 19:57 / 2.13 ft May 25 Mon 01:28 / 7.96 ft 09:08 / -1.90 ft 15:27 / 5.69 ft 20:46 / 2.20 ft May 26 Tue 02:15 / 7.96 ft 09:56 / -1.92 ft 16:18 / 5.74 ft 21:39 / 2.24 ft May 27 Wed 03:05 / 7.76 ft 10:46 / -1.75 ft 17:10 / 5.78 ft 22:38 / 2.25 ft May 28 Thu 04:00 / 7.33 ft 11:37 / -1.40 ft 18:03 / 5.86 ft 23:44 / 2.19 ft May 29 Fri 05:00 / 6.71 ft 12:30 / -0.92 ft 18:56 / 6.02 ft May 30 Sat 00:57 / 2.02 ft 06:07 / 6.00 ft 13:23 / -0.36 ft 19:48 / 6.23 ft

May 31 Sun 02:16 / 1.68 ft 07:24 / 5.32 ft 14:17 / 0.24 ft 20:38 / 6.49 ft

171 2009 IPHC ANNUAL MEETING HANDOUT Tillamook Bay, Bay City June 2009

High High High Low Low Date Day Time/Height Time/Height Time/Height Time/Height Time/Height Jun 1 Mon 03:32 / 1.18 ft 08:48 / 4.84 ft 15:11 / 0.83 ft 21:25 / 6.73 ft Jun 2 Tue 04:40 / 0.60 ft 10:12 / 4.64 ft 16:05 / 1.36 ft 22:09 / 6.92 ft Jun 3 Wed 05:37 / 0.05 ft 11:27 / 4.68 ft 16:58 / 1.79 ft 22:51 / 7.04 ft Jun 4 Thu 06:27 / -0.39 ft 12:31 / 4.83 ft 17:49 / 2.12 ft 23:30 / 7.09 ft Jun 5 Fri 07:11 / -0.71 ft 13:25 / 5.01 ft 18:36 / 2.35 ft Jun 6 Sat 00:09 / 7.08 ft 07:51 / -0.89 ft 14:11 / 5.15 ft 19:20 / 2.49 ft

Jun 7 Sun 00:46 / 7.02 ft 08:29 / -0.96 ft 14:52 / 5.24 ft 20:02 / 2.57 ft Jun 8 Mon 01:23 / 6.94 ft 09:06 / -0.95 ft 15:31 / 5.28 ft 20:41 / 2.61 ft Jun 9 Tue 01:59 / 6.82 ft 09:42 / -0.86 ft 16:09 / 5.29 ft 21:21 / 2.63 ft Jun 10 Wed 02:37 / 6.65 ft 10:18 / -0.71 ft 16:46 / 5.30 ft 22:03 / 2.63 ft Jun 11 Thu 03:15 / 6.41 ft 10:54 / -0.51 ft 17:23 / 5.33 ft 22:48 / 2.62 ft Jun 12 Fri 03:55 / 6.08 ft 11:30 / -0.25 ft 18:00 / 5.40 ft 23:40 / 2.56 ft Jun 13 Sat 04:40 / 5.66 ft 12:07 / 0.07 ft 18:38 / 5.52 ft

Jun 14 Sun 00:38 / 2.41 ft 05:31 / 5.18 ft 12:44 / 0.45 ft 19:14 / 5.71 ft Jun 15 Mon 01:42 / 2.13 ft 06:34 / 4.70 ft 13:24 / 0.86 ft 19:51 / 5.96 ft Jun 16 Tue 02:48 / 1.70 ft 07:50 / 4.33 ft 14:08 / 1.29 ft 20:30 / 6.27 ft Jun 17 Wed 03:51 / 1.11 ft 09:15 / 4.19 ft 14:57 / 1.70 ft 21:10 / 6.63 ft Jun 18 Thu 04:48 / 0.44 ft 10:36 / 4.32 ft 15:52 / 2.04 ft 21:54 / 7.03 ft Jun 19 Fri 05:40 / -0.25 ft 11:46 / 4.62 ft 16:50 / 2.29 ft 22:41 / 7.44 ft Jun 20 Sat 06:30 / -0.90 ft 12:46 / 4.99 ft 17:48 / 2.41 ft 23:30 / 7.82 ft

Jun 21 Sun 07:19 / -1.43 ft 13:38 / 5.35 ft 18:44 / 2.41 ft Jun 22 Mon 00:20 / 8.13 ft 08:07 / -1.80 ft 14:27 / 5.65 ft 19:40 / 2.33 ft Jun 23 Tue 01:12 / 8.29 ft 08:54 / -1.98 ft 15:13 / 5.91 ft 20:35 / 2.20 ft Jun 24 Wed 02:04 / 8.24 ft 09:41 / -1.93 ft 15:59 / 6.14 ft 21:32 / 2.04 ft Jun 25 Thu 02:58 / 7.95 ft 10:28 / -1.66 ft 16:44 / 6.35 ft 22:31 / 1.87 ft Jun 26 Fri 03:53 / 7.40 ft 11:14 / -1.19 ft 17:30 / 6.54 ft 23:35 / 1.69 ft Jun 27 Sat 04:52 / 6.65 ft 12:00 / -0.57 ft 18:16 / 6.72 ft

Jun 28 Sun 00:43 / 1.46 ft 05:57 / 5.82 ft 12:46 / 0.14 ft 19:03 / 6.86 ft Jun 29 Mon 01:54 / 1.17 ft 07:11 / 5.05 ft 13:34 / 0.87 ft 19:50 / 6.94 ft Jun 30 Tue 03:07 / 0.81 ft 08:36 / 4.54 ft 14:25 / 1.55 ft 20:38 / 6.97 ft

172 2009 IPHC ANNUAL MEETING HANDOUT B. Kodiak, Port of Kodiak 15 February 2009 - 21 February 2009

Sun Sat Kodiak, Port of Kodiak 22 February 2009 - 28 February 2009

Sun Sat

173 2009 IPHC ANNUAL MEETING HANDOUT Kodiak, Port of Kodiak 1 March 2009 - 7 March 2009

Sun Sat

Kodiak, Port of Kodiak 8 March 2009 - 14 March 2009

Sun Sat

174 2009 IPHC ANNUAL MEETING HANDOUT Kodiak, Port of Kodiak 15 March 2009 - 21 March 2009

Sun Sat

175 2009 IPHC ANNUAL MEETING HANDOUT 176 2009 IPHC ANNUAL MEETING HANDOUT