REFM Report OND '99

Status of Stocks and Multispecies Assessment Program

The Status of Stocks and Multispecies Assessment Program (SSMA)Program completed stock assessments for major groundfish stocks in the Bering Sea, Aleutian Islands, and Gulf of Alaska. Stock assessments utilized information from a variety of sources including survey biomass estimates provided by the RACE Division, fishery catch estimates from the Alaska Regional Office, fishery-dependent biological samples from the REFM North Pacific Groundfish Observer Program, and fishery and survey age composition data from the REFM Age and Growth Program. Final SAFE (stock assessment and fishery evaluation) reports were prepared for the North Pacific Fishery Management Council (NPFMC) plan team meetings. The assessments resulted in recommended levels of Acceptable Biological Catch (ABC) which were presented by the plan team chairpersons to the NPFMC. Final ABC levels were set by the Council following the recommendations of its advisory committees.

Scientists from the SSMA Program participated in preparing many of the assessment chapters for the Gulf of Alaska and Bering Sea/Aleutian Islands regions. Scientists from the Auke Bay Laboratory (ABL), the RACE Division, and the Alaska Department of Fish and Game (ADF&G)also authored or contributed to the development of the SAFE chapters.

In 1999, 14 stocks or stock complexes in the Bering Sea/Aleutian Islands region were assessed: walleye pollock, Pacific cod, yellowfin sole, Greenland turbot, rock sole, flathead sole, arrowtooth flounder, other flatfish (includes 16 species of flatfish), eastern Bering Sea (EBS) Pacific ocean perch, Aleutian Islands Pacific ocean perch, other rockfish, sablefish, Atka mackerel, and squid and other species. In the Gulf of Alaska, 11 stocks or stock complexes were assessed: walleye pollock, Pacific cod, arrowtooth flounder, other flatfish, pelagic shelf rockfish, slope rockfish, demersal shelf rockfish, sablefish, thornyheads, Atka mackerel, and other species. Major findings for selected assessments are summarized below, grouped by management area and species. Refer to Auke Bay’s section in this issue for information on the assessments of Alaskan sablefish and Gulf of Alaska slope rockfish and pelagic shelf rockfish.

The values presented below reflect information found in the SAFE documents developed by the NPFMC teams for groundfish fisheries in the Gulf of Alaska and Bering Sea/Aleutian Islands. It should be noted that the plan team’s ABC and TAC recommendations are reviewed by the NPFMC advisory bodies and the Council sets the final TAC. Thus, the values presented below may differ from the final Council recommendations. For information on final groundfish specifications, please consult the NPFMC’s web page at www.fakr.noaa.gov/npfmc.

Bering Sea Walleye Pollock
The 1999 NMFS bottom trawl and echo-integration trawl (EIT)survey biomass estimates and bottom trawl population numbers at age were included in the assessment. The bottom-trawl survey biomass estimate for 1999 was 3.57 million metric tons (t), up 61% from the 1998 estimate of 2.21 million t. An EIT survey was conducted over the EBS shelf in 1999. The biomass estimate from the 1999 EIT survey was 3.29 million t, up 27% from the 1997 estimate (the last year that an EIT survey was conducted in this region). As expected, the most abundant age group in this survey was the 1996 year class. While both survey methods showed promising increases in abundance, the water temperatures during the survey period (summer 1999) were the coldest experienced in all the years that NMFS has conducted summer surveys of this region. These temperature anomalies may have affected the availability of pollock to the survey gear and, consequently, may affect the biomass estimates.

The NMFS observer samples of pollock age and size composition were evaluated for the 1998 fishery, and these data were included in the analysis. The estimates of weight-at-age from the fishery were also revised for 1998. The total catch estimate was updated for 1998; and for 1999, the catch was assumed to be equal to the 1999 TAC (992,000 t).

No substantive changes to the assessment model were made compared to the 1998 SAFE chapter. Two minor modifications were made to the assessment model configuration. The annual estimates of bottom trawl survey variances were used. Previously the average variance over all years had been used. Control over the time period used to fit the stock-recruitment parameters was added to allow the user to fit the current climate condition (since 1977). Previously stock recruitment parameters were fitted using the entire time series.

ABC harvest alternatives based on F40% and the harmonic mean value of F_msy were 1.102 million t and 1.197 million t, respectively, based on the reference model. The corresponding year 2000 overfishing level (OFL) alternatives for the reference model were 1.471 million t and 1.678 million t under F35% and Fmsy, respectively. Issues extraneous to these harvest levels include current uncertainty about changes in harvest rates on the EBS stock outside of the U.S. EEZ (Exclusive Economic Zone), particularly for prerecruit age groups, continuing decline in Steller sea lion populations in adjacent areas, and recent observations that the biological and physical environment is changing. Some of the uncertainty in current and future environmental conditions was factored into the assessment and appropriate precautionary measures were implemented.

Aleutian Islands Walleye Pollock
Harvest levels for Aleutian Islands walleye pollock were computed in the same way as last year. The recommended ABC and OFL for Aleutian Islands pollock were 23,760 t and 31,680 t, respectively. For the Bogoslof region, three methods for harvest level determinations were presented. These methods provided estimates of ABC that ranged from 14,200 t (the old method) to 110,000 t based on an age-structured model.

Bering Sea-Aleutian Islands Pacific Cod
Relative to the November edition of last year’s Bering Sea-Aleutian Islands (BSAI) SAFE report, the following substantive changes to the input data were made in the Pacific cod stock assessment. Size composition data from the 1998 and January-August 1999 commercial fisheries were incorporated into the model. Size composition data from the 1999 EBS bottom trawl survey were incorporated. The biomass estimate from the 1999 EBS bottom trawl survey was incorporated (the 1999 estimate of 583,259 t was up about 9% from the 1998 estimate). Weight-at-length data from recent EBS bottom trawl surveys were incorporated.

No changes in the assessment model were made in 1999. Results of assessment modeling showed that the estimated 2000 spawning biomass for the BSAI stock was 357,000 t, up about 8% from last year’s estimate for 1999 and up about 21% from last year’s FABC projection for 2000. The estimated 2000 total age 3+ biomass for the BSAI stock was 1.3 million t, up about 7% from last year’s estimate for 1999 and up about 21% from last year’s FABC projection for 2000. The recommended 2000 ABC for the BSAI stock was 193,000 t, up about 9% from last year’s recommendation for 1999 and up about 18% from last year’s FABC projection for 2000. As in previous years, this recommended ABC was based on a harvest strategy that formally addresses some of the uncertainty surrounding parameters of M (natural mortality) and Q (catchability coefficient). The strategy relies on a Bayesian meta-analysis. The recommended ABC for the EBS portion of the BSAI was 165,000 t. The estimated 2000 OFL for the BSAI stock was 240,000 t, down about 9% from last year’s estimate for 1999 and up about 7% from last year’s FOFL projection for 2000.

Bering Sea Rock Sole
Evidence of density-dependent effects on growth of rock sole was presented in the 1999 assessment. In conjunction with the large and steady increase in the rock sole stock size since the early 1980s, it was found that there was also a corresponding decrease in size-at-age for both sexes. Weight-at-age decreased as the population increased and expanded westward toward the shelf edge. Updated values of weight-at-age were used in this assessment to model the population dynamics of the rock sole population and were compared to results obtained from the constant growth model used in past assessments.

The projected age 2+ biomass for 2000 was 2,073,600 t. The projected female spawning biomass for 2000 was 675,500 t. The recommended 2000 ABC was 229,500 t based on an F40% (0.154) harvest level. This ABC was 25% lower than the 1999 ABC recommendation but was consistent with 1998 projected trends in biomass. The 2000 overfishing level was 272,000 t based on an F35% (0.186) harvest level.

Bering Sea Yellowfin Sole
The trawl survey biomass estimate for 1999 was more than 1.0 million t below the biomass estimate in 1998. This was believed to be the result of a change in fish distribution and behavior in response to an unusually cold year in the Bering Sea, rather than a large decrease in population abundance. The 1999 ABC for yellowfin sole was 212,000 t, and only 59,000 t were caught.

The projected age 2+ biomass for 2000 was 2,815,600 t. The projected female spawning biomass for 2000 was 789,300 t. The recommended 2000 ABC was 190,600 t based on an F40% (0.11) harvest level. This estimate was 10% below the 1999 ABC recommendation and was consistent with 1998 projected trends in biomass. The 2000 overfishing level was 226,000 t based on an F35% (0.13) harvest level.

Eastern Bering Sea and Aleutian Islands Greenland Turbot
The following changes were made since last year’s assessment: new summary estimates of retained and discarded Greenland turbot by different target fisheries, an update to the estimated catch levels by gear type in recent years, and new length frequency and biomass data from the 1999 NMFS EBS shelf survey.

Conditions do not appear to have changed substantively over the past several years. For example, the abundance of Greenland turbot from the EBS shelf-trawl survey has found only spotty quantities with very few small fish that were common in the late 1970s and early 1980s. The majority of the catch has shifted to longline gear in recent years. The assessment model analysis was similar to last year’s but with a slightly higher estimated overall abundance. We attribute this to a slightly improved fit to the longline survey data trend. The target stock size (B40%, female spawning biomass) was estimated at about 81,200 t while the projected year 2000 spawning biomass was about 150,800 t. The adjusted yield projection from F40% computations was estimated at 34,700 t for 2000 (233,000 t total age 1+ biomass). Given the continued downward abundance trend and no sign of recruitment to the EBS shelf, extra caution was warranted. Therefore, we recommended that the ABC be set to 25% of the maximum FABC value giving 9,300 t. This low level was recommended until further information on the source of Greenland turbot production is found. Namely, whether or not recruitment to the adult slope population is still occurring even though the bottom trawl estimates of small Greenland turbot on the shelf has continued to decline since the early 1980s. Survey plans for the year 2000 are to include the slope region of the EBS. This should help assess the prime area where Greenland turbot are found. As additional survey information becomes available and signs of recruitment (perhaps from areas other than the shelf) are apparent, then we believe that the full ABC or increases in harvest may be appropriate for this species.

Bering Sea Arrowtooth Flounder
The 1999 EBS bottom trawl biomass estimate for arrowtooth flounder was 243,800 t, a 29% decrease from the 1998 trawl survey biomass estimate of 344,900 t.

In the 1998 assessment model, proportions of the length composition from each size bin were adjusted to reflect the preponderance of females found in survey catches. Unfortunately, the synthesis model does not allow for the assumption of unequal sex composition and gave very low male selectivity values, which resulted in much higher biomass values. This was considered an undesirable result, and this exercise was not repeated for the 1999 assessment. Thus, the 1998 projected values for 2000 were not comparable to this year’s assessment results.

The projected age 1+ total biomass for 2000 was 784,630 t. The projected female spawning biomass for 2000 was 496,000 t. The recommended 2000 ABC for BSAI arrowtooth flounder was 130,500 t based on an F0.40 (0.22) harvest level. The 2000 OFL was 160,200 t based on a F0.35 (0.27) harvest level.

Bering Sea Flathead Sole
The 1999 EBS bottom trawl biomass estimate for flathead sole was 394,806 t, a 43% decrease from the 1998 biomass estimate of 692,237 t.

No changes were made to the assessment model used to assess flathead sole. The recommended ABC for 2000 was based upon the F40% fishing mortality rate. The year 2000 projections for flathead sole increased relative to the 1998 SAFE because the actual fishing mortality in 1999 was less that the F40% rate.

The recommended ABC for flathead sole was 73,537 t, a 13% increase from last year’s projection of year 2000 catch. The projected female spawning biomass for 2000 was 261,342 t, an 11% increase from the year 2000 projection made last year.

Bering Sea Other Flatfish

The following table contains the most recent bottom trawl survey biomass estimate in metric tons (t) for Alaska plaice and miscellaneous flatfish, plus the miscellaneous species of the other flatfish chapter.

The assessment methodology for the other flatfish SAFE chapter was identical to last year’s with the exception of the Alaska plaice assessment (within the other flatfish chapter), in which the software program AD Model Builder was used.

The table below contains the year 2000 projections of ABC and spawning biomass from the 1998 and 1999 stock assessments for BSAI Alaska plaice. The yield projections were based upon the F40% fishing mortality rate. The year 2000 projections for Alaska plaice increased relative to the 1998 SAFE because the actual fishing mortality in 1999 was less that the F40% rate.

Bering Sea/Aleutian Islands Pacific Ocean Perch and Other Red Rockfish
The methods used to assess Pacific ocean perch (POP) were identical to last year. The following table contains the most recent bottom trawl survey biomass estimate in metric tons (t) for the Pacific ocean perch and a comparison to previous survey estimates.

Current biomass estimates for other red rockfish (northern rockfish, rougheye rockfish, shortraker rockfish, and sharpchin rockfish) were obtained by averaging the estimated biomass in the recent U.S. domestic survey (ranging from 1988 to 1997). In contrast with previous years, earlier surveys were not considered because of concerns with consistency with recent surveys. This resulted in removing an unusually high biomass estimate for the EBS northern rockfish, and thus lowered the estimated biomass.

The year 2000 ABCs for AI Pacific ocean perch and EBS Pacific ocean perch are shown below.

The recommended ABCs for other red rockfish were obtained from multiplying 0.75 times the product of the natural mortality rate and the average biomass from the U.S. domestic trawl surveys. A summary of the recommended 2000 ABC with the recommended 1999 ABC is shown below.

Bering Sea/Aleutian Islands Other Rockfish
The methods used to assess other rockfish were identical to last year. ABC was obtained by multiplying 0.75 times the product of the natural mortality rate and the average biomass from the U.S. domestic trawl surveys. Because the most recent survey was conducted in 1997, the estimated mean biomass and recommended ABC were identical to those in the 1998 SAFE. The estimated biomass of other rockfish in the EBS and AI was 7,029 t and 13,041 t, respectively. The recommended ABC for the EBS and AI was 369 t and 685 t, respectively.

Bering Sea/AleutianIslands Atka Mackerel
Relative to the November 1998 SAFE report, the following substantive changes have been made in the current draft of the Atka mackerel chapter. The 1999 TAC was used as a proxy for the 1999 catch. The most recent survey biomass estimate (1997) was down about 50% relative to the 1994 survey estimate.

No substantive changes were made in the assessment model. A stochastic projection model was used to project 2000 biomass and yield, and to evaluate alternative harvest policies for stocks managed under Tiers 1, 2, or 3 of Amendment 56.

The mean recruitment from the stochastic projections was 622 million recruits, which gives an estimated B40% level of 154,300 t. The projected age 3+ biomass at the beginning of 2000 was 565,150 t. The projected female spawning biomass for 2000 was 162,500 t; representing a 0.9% decrease relative to last year’s 2000 estimate. The projected 2000 yields at F40% and FABC were 102,700 and 78,500 t, respectively. The 2000 recommended ABC was 78,500 t corresponding to F = 0.26, that represents a 7% increase relative to the 1999 ABC and an 11% increase relative to last year’s projected 2000 yield. These increases in the face of a declining stock trend, were due to a change in harvest strategy and the use of stochastic rather than constant recruitment in the projections. The projected 2000 OFL at F35% was 119,300 t.

Bering Sea/Aleutian Islands Squid and Other Species
The “other species” management group has been established to account for species which are currently of slight economic value and upon which there is little, if any, directed fishing. However, these species could have economic value in the future, and many are important components of the ecosystem as prey for commercial fish species, marine mammals and birds. Squid was considered separately from the “other species” management group, which includes sculpins, skates, sharks, and octopus. Smelts were removed from the “other species” group and moved to the forage fish group beginning in 1999 as a result of fishery management plan (FMP) amendments 36 and 39 to the Bering Sea and Aleutian Islands and Gulf of Alaska groundfish FMPs.

The recommended ABC for squid in the year 2000 was calculated as 0.75 times the average catch from 1978-95, or 1,970 t; this was unchanged from 1999. The recommended OFL for squid in the year 2000 was calculated as the average catch from 1978 to 1995, or 2,624 t; this was also unchanged from 1999.

The recommended ABC for the other species complex in the year 2000 was calculated as the average catch from 1977 to 1998, or 26,800 t. This was unchanged from 1999. The recommended OFL for the other species complex in the year 2000 was 122,100 t, which was calculated by multiplying an estimate of the natural mortality rate (M=0.2) by a biomass estimate for the complex (average of the last three EBS bottom trawl surveys plus the most recent Aleutian Islands bottom trawl survey = 610,400 t). The OFL and biomass estimate for other species for the year 2000 were 5% lower than those calculated for 1999.

Gulf of Alaska Walleye Pollock
There were several changes made in the input data to the stock assessment model used this year. Biomass estimates and length composition from the 1999 triennial bottom trawl survey were included. The 1999 triennial survey estimate of pollock biomass west of long. 140EW was 606,295 t, a 9% decline from 1996. Several factors complicate interpretation of the 1999 biomass estimate. First, bottom temperatures in 1999 were generally cooler than during other triennial surveys. The vulnerability of pollock to the bottom trawl survey may be temperature-related, either due to changes in vertical distribution or due to inshore-offshore changes in spatial distribution. Second, estimates of total biomass and biomass distribution between statistical areas were strongly affected by a few large tows. The coefficient of variation of the gulf-wide biomass estimate was 0.38, more than double any previous triennial survey. The largest concentration of pollock biomass was estimated to be in the Shumigan area (72%), followed by the Kodiak (18%), Chirikof (9%) and West Yakutat INPFC areas (1%). Furthermore, additional age data were used in the 1999 assessment. Pollock age composition from the 1997 and 1998 Shelikof Strait EIT surveys were included. Age composition and catch from the 1998 fishery were included. Biomass and length composition data from the ADF&G coastal bottom trawl survey for 1989-98 were evaluated for consistency with other information, and were included in the base-run model.

Gulf of Alaska pollock were assessed using an age-structured population model developed using ADModel Builder (a C++ software language extension and automatic differentiation library).
The stock assessment was extended eastward to 140EW longitude to coincide with the area open for trawling in the Gulf of Alaska (Statistical Areas 610-640). The annual catch and triennial bottom trawl biomass estimates were revised to correspond to this area. This change addresses the NPFMCs Scientific and Statistical Committee’s comments about the discrepancy between assessment and management areas in the eastern Gulf of Alaska.

A separate ABC for pollock in the no-trawl zone in southeast Alaska (East Yakutat and Southeastern INPFC areas) was obtained under Tier 5 of NPFMC harvest policy.

To address Plan Team and SSC concerns about unsurveyed pollock in Prince William Sound (PWS), the biomass from the triennial survey and the ADF&G survey in PWS in 1999 were added to obtain a total biomass estimate. NMFS vessels surveyed the northern gulf within 2 weeks of the ADF&G survey in PWS. The adjustment factor for the 1999 surveys (PWS + Triennial)/Triennial was applied to earlier triennial surveys and increased the biomass time series by 1.05%.

There were several changes in assessment results this year. Spawning biomass in 2000 was projected to be 214,900 t, which was 35% of the unfished biomass of 619,000 t, thereby placing Gulf pollock in sub-tier “b” of Tier 3 NPFMC harvest guidelines. Age 3+ biomass for the region west of 140EW in 2000 was projected to be 588,000 t for the base-run model and 651,000 t for a model without the ADF&G survey data. These estimates were similar to biomass projections made last year, which ranged from 581,500 t to 674,500 t. The recommended 2000 ABC for the GOA stock was 111,306 t, representing an increase of 28% from the projected 2000 ABC in the 1998 assessment for the central and western GOA (80,139 t). This projection was based on an adjusted F40% rate. The higher projected 2000 ABC was caused by several factors: a larger assessment area including PWS and all of West Yakutat; increased abundance of the 1995 year class based on 1998 fishery age composition data (492 million as compared to 181 million in the 1998 assessment); and the use of average fishery selectivity for 1992-99 both to calculate SPR rates and to project harvests. Last year’s assessment used the 1985-98 average selectivity to calculate SPR rates, and 1994-98 average selectivity to project yields and the correction of an error in the calculation of SPR rates. The estimated 2000 OFL for the GOA stock was 130,758 t. For pollock in Southeast Alaska (East Yakutat and Southeastern INPFC areas), the ABC recommendation was 6,460 t., an increase of 2% for a comparable area in the 1998 assessment). Various approaches to apportioning the ABC using triennial survey biomass were evaluated. A regional allocation based on a four survey average was recommended (41.0% Shumigan, 24.4% Chirikof, 32.1% Kodiak, 2.5% West Yakutat INPFC areas). Since the assessment now explicitly includes the pollock biomass in Prince William Sound, the harvest guideline for PWS pollock should be subtracted from the total ABC prior to regional allocation.

Gulf of Alaska Pacific Cod
Relative to the November edition of last year’s GOA SAFE report, the following substantive changes have been made in the Pacific cod stock assessment.

Several new sources of data were used in the 1999 GOA cod assessment. Size composition data from the 1998 and January-August 1999 commercial fisheries were incorporated into the model. Size composition data from the 1999 GOA bottom trawl survey were incorporated. The biomass estimate from the 1999 GOA bottom trawl survey was incorporated (the 1999 estimate of 305,823 t was down about 43% from the 1996 estimate). Weight-at-length data from recent GOA bottom trawl surveys were incorporated.

There were no changes in the assessment model this year. The estimated 2000 spawning biomass for the GOA stock was 111,000 t, down about 15% from last year’s estimate for 1999 and down about 3% from last year’s FABC projection for 2000. The estimated 2000 total age 3+ biomass for the GOA stock was 567,000 t, down about 13% from last year’s estimate for 1999 and up about 7% from last year’s FABC projection for 2000. The recommended 2000 ABC for the GOA stock was 76,400 t, down about 16% from last year’s recommendation for 1999 and down about 10% from last year’s FABC projection for 2000. The estimated 2000 OFL for the GOA stock was 102,000 t, down about 24% from last year’s estimate for 1999 and down about 11% from last year’s FOFL projection for 2000.

Gulf of Alaska Arrowtooth Flounder
Catch for 1998 was updated, and an age-based model run using different natural mortality values for males and females with added survey age composition data from 1984, 1987, and 1990. The 1999 NMFS survey biomass estimate and length composition were added to the model. The natural mortality for males was set higher than for females to obtain a sex ratio of about 70% female in the population. Length composition data were fit using a fixed length-age transition matrix. The 1998 assessment used the length-based synthesis model, estimating growth parameters in the model. In the 1998 assessment the natural mortality estimates and the selectivities by length were the same for males and females.

Survey biomass estimates from halibut trawl surveys in the 1960s, groundfish trawl surveys in the 1970s and NMFS triennial trawl surveys from 1984 to 1999 were used. Selectivities were estimated by a smooth function that was constrained to be monotonically increasing with age for the fishery and survey. The estimated biomass from the model increased from about 206,600 t in 1961 to 1.69 million t in 1994, then declined to about 1.592 million t in 1999. In the 1998 assessment the survey selectivities by age for males was lower than females resulting in a higher population biomass than in the current assessment (about 2,126,000 t in 1999). In the current model the survey selectivities by age for males and females are more similar, resulting in population biomass estimates that are closer to the survey biomass estimates.

The 2000 yield using F40% was 145,361 t. OFL using F35% was 173,915 t. The 1999 ABC using F40% was 217,106 t.

Gulf Of Alaska Flatfish
Catch has been updated through 9 October 1999. ABCs were estimated using the 1999 survey biomass estimates for all species except Greenland turbot and deep-sea sole where the mean catch from 1978 to 1995 was used.

Catches for species in the deep-water or shallow-water groups were estimated from 1978 to 1998 by multiplying the group catch estimate by the estimate of the fraction of each species in the catch based on observer data.

The 1999 triennial trawl survey biomass was used as current biomass for calculation of ABC. Rock sole ABC was estimated using F40% calculated using the Bering Sea rock sole maturity schedule. Greenland turbot and deep-sea sole ABC and OFL were calculated using average catch. ABCs for other flatfish except rock sole were estimated using F = 0.75 M. The ABC for Dover sole was estimated using the 1999 survey biomass which should be more applicable than the 1996 survey, since it covered the depth range of Dover sole (to 1,000 m). The 1990 to 1996 surveys included depths to 500 m only, which did not survey the population depth range of Dover sole.

Gulf of Alaska Atka Mackerel

There were very few changes made to the current year’s Atka mackerel assessment. Catch data were updated and research catches were included. There were no changes in the assessment methodology and no changes in assessment results.

In this assessment several issues were highlighted. There was no reliable estimate of current biomass from the Gulf of Alaska bottom trawl survey. Using Tier 6 criteria, OFL was set equal to the average catch from 1978 to 1995, which equals 6,200 t. It was recommended that the ABC for Atka mackerel in the GOA be 600 t, enough to satisfy only the anticipated bycatch needs of other trawl fisheries, principally those for Pacific cod, rockfish, and pollock. This recommendation was based on the following reasons: a) there was no reliable estimate of current biomass, b) Leslie estimates of local population sizes suggest that abundance has declined significantly in localized areas from 1992-94, and c) the species has exhibited vulnerability to fishing pressure in the past.

Gulf of Alaska Thornyheads
This year we updated the model introduced in 1997 with available recent data, including 1998 harvest levels by gear, and biomass estimates from the 1999 NMFS triennial trawl survey. Alternate models examined assumptions regarding recruitment variability and survey catchability. Results from this year’s analyses were similar to last year’s, although harvest levels have increased slightly for next year under the F40% fishing mortality.

The following table summarizes the ABC recommendations and status of spawning biomass level for the past few years relative to the current assessment.

Gulf of Alaska Other Species
The first assessment of Gulf of Alaska other species was introduced in 1999. This assessment considered fishing impacts on sharks, skates, sculpins, smelts, octopi, and squids in 1990-98 (smelts were removed in 1999). In the past, TAC for this category has been set at 5% of the sum of all GOA target species TACs. The proposed assessment model was a simple state-space model incorporating both observation error and process error in biomass and exploitation rate estimates. Input data included catch estimates by species group from 1990-98, and 1984-99 GOA triennial trawl survey biomass estimates for each species group. Assessment scientists also examined catch and biomass estimates for grenadiers, which are not included in the other species category.

Catches of fish and invertebrates of the other species category were very small compared to those of target species in the Gulf of Alaska. It appears unlikely that the observed 1990-98 bycatch of other species has had a negative effect on biomass at the species group level, according to the available trawl survey data. Grenadiers, while not included in the other species category, may be the dominant fish in deeper habitats and were caught in sufficient numbers to warrant additional attention, especially because they may be very long- lived species. Data limitations were severe for all species groups in this category, and further investigation is necessary to ensure that all components of the other species complex are not adversely affected by groundfish fisheries. Furthermore, if target fisheries develop for any component of the other species group, effective management will be extremely difficult with the current limited information.

Although changing the procedure for establishing the TAC of other species requires an amendment to the GOA FMP, we propose separate ABC and OFL levels for each species group within other species to ensure that less productive groups are not over harvested. These individual ABCs sum to slightly less than the recent aggregate TACs in the range of 14,000 t, but observed catches in each of the categories have never exceeded these proposed ABCs in the domestic fishery, with the exception of octopus catches in 1992 and 1997. We believe that cephalopod biomass was substantially underestimated by the bottom trawl survey, resulting in overly conservative estimates of ABC and OFL for these species groups; but we have no other data on which to base recommendations.

Understanding other species population dynamics is fundamental to describing ecosystem structure and function in the Gulf of Alaska, because each group in other species plays an important ecological role. The species groups in this category occupy all marine habitats from pelagic to benthic, near shore to open ocean, and shallow to slope waters. Regardless of management decisions regarding TAC and the future structure for other species, it is essential that we continue to improve species identification, survey sampling, and biological data collection for the species in this category if we hope to ensure their continued conservation.

By Anne Hollowed.

RESOURCE ECOLOGY AND ECOSYSTEM MODELING PROGRAM

Stomachs collected totaled 1,010 for the West Coast. Laboratory analysis was performed on 1,423 groundfish stomachs from the eastern Bering Sea, 522 from the Aleutian Islands region, and 160 from Gulf of Alaska. Nine observers returned with stomach samples (602 walleye pollock) collected from the eastern Bering Sea.

Diets of Pacific and Giant Grenadier

In a collaborative investigation with Jeff Drazen (Scripps Institution of Oceanography) and Jerry Hoff (RACE Division), the feeding habits of Pacific grenadier (Coryphaenoides acrolepis) and giant grenadier (Albatrossia pectoralis) from the continental slope off Washington, Oregon, and California were analyzed. A total of 497 Pacific grenadier and 617 giant grenadier were used in this analysis. We found an ontogenetic shift in the diet of Pacific grenadier from benthic epifauna and infauna (amphipods, cumaceans, and polychaetes) to nekton and scavenged material (Gnathophausia spp., squid and fish). We found a decrease in squid and an increase in midwater fishes and scavenged material in the diet with increasing size of the giant grenadier examined. These shifts were attributed primarily to increasing mouth size and locomotory ability. Significant differences were found between the diets of Pacific and giant grenadier where they co-occur, indicating some degree of niche separation between the two species.

By Troy Buckley.

Winter Feeding Habits of Pacific Whiting (Merluccius productus)

In January 1997, 226 stomach samples from Pacific whiting were collected during a spawning survey off southern California. Tows were made in midwater at depths of 165 to 400 m over bottom depths of 750 to 1800 m. The gonads of Pacific hake were categorized by their stage of development based on gross characteristics of the whole organ. In this study, 80% of the stomachs were empty and 20% contained prey items. In stomachs that contained food, fishes dominated the weight composition (88%) and numerical composition (52%) of the diet and had a high frequency of occurrence (60%). All of the fish were myctophids (lanternfishes) except for one incidence of cannibalism. Myctophids have been found in other studies, but not as the dominant prey. The high percentage of empty stomachs was found in all reproductive stages including immature Pacific hake. We believe our results reflect a lack of other prey in the study area due to seasonal, geographical, and interannual influences.

By Geana Tyler.

PICES Science: 1999

The PICES (North Pacific Science Organization) Eighth Annual Meeting was held 8-17 October 1999 in Vladivostok, Russia. One hundred and fifty-one oral presentations and 99 posters were given at the meeting. More than 300 scientists attended the event.

The year 1999 was important for PICES because of several ‘firsts’ in the area of international collaborative field and laboratory work by the PICES scientific community. In 1999 a practical workshop was held in Vancouver Harbor by scientists of the Marine Environmental Quality Committee’s Working Group 8 on Practical Assessment Methodology. This was the first time scientists from all PICES-member countries participated in a mutual field and laboratory effort. The workshop set the stage for future collaborative work in the study of marine environmental quality. Similarly, the Physical Oceanography and Climate Committee’s Working Group 13 on Carbon Dioxide in the North Pacific also held a multinational technical workshop in 1999. Their intercalibration exercise for laboratory measurements of carbon dioxide will ensure the high quality of North Pacific carbon dioxide measurements in the future, allow multinational synthesis of carbon dioxide studies, and lead to improved understanding of carbon cycle processes. Finally, the PICES-GLOBEC CCCC Program successfully obtained funding from the North Pacific Marine Research Program to perform a 2-year study to initiate continuous plankton recorder (CPR) monitoring in the north. The next challenge will be to find a way to maintain this monitoring as a long-term PICES effort.

Many PICES scientific efforts were completed in 1999 through publication in either the PICES Scientific Report Series, in special journal volumes, or as books. The proceedings of the 1997 Science Board Symposium on “Ecosystem dynamics in the eastern and western gyres of the subarctic Pacific” was published in volume 43(2-4) of Progress in Oceanography. The final endpoint of the Science Board’s Working Group 5 on the Bering Sea came with the publication of the book “Dynamics of the Bering Sea” by University of Alaska Sea Grant. Three volumes of the PICES Scientific Report series were produced in 1999: Volume 10 has the proceedings of the 1998 Science Board Symposium on El Niño 1997-98 events; Volume 11 has the proceedings of the 1998 PICES GLOBEC CCCC Program MODEL, REX, and MONITOR workshops; and Volume 12 has the proceedings of the second PICES workshop on the Okhotsk Sea and adjacent areas.

Several working groups completed their work in 1999 and are preparing to publish results in 2000. The Physical Oceanography and Climate Committee’s Working Group 10 on Circulation and Ventilation in the Japan/East Sea will place their report on the PICES web site as a revisable, living document. The Fishery Science Committee’s Working Group 12 on Crabs and Shrimp and Biological Oceanography Committee’s Working Group 11 on Consumption of Marine Resources by Marine Mammals and Seabirds will also be working toward publication of their final results in the year 2000.

The PICES-GLOBEC CCCC Program continues its work on integrating and stimulating national GLOBEC research efforts in the North Pacific. The Regional Experiments (REX) Task Team is presently focusing on comparative work on herring in the North Pacific. They completed an interesting workshop on “Herring and Euphausiids” and are planning a workshop for 2000 on herring population trends and trophodynamics. The Basin Scale Studies (BASS) Task Team has an Iron Fertilization Panel that is planning international field experiments in the subarctic North Pacific to understand the role of iron in influencing production. The MODEL Task Team is undertaking two workshops in 2000, one to build lower trophic level models in several areas of the North Pacific and the second one to link these models to upper trophic level models. The MONITOR Task Team completed a successful workshop on the Global Ocean Observing System (GOOS) and to examine the future role of PICES in this growing international program. As an outcome of the workshop, PICES will be developing an action plan that will outline how PICES will be taking an active and leading role in the implementation of GOOS at a North Pacific level.

New collaborations and working groups will begin in the year 2000. An advisory panel on CPR surveys in the North Pacific was formed, which will advise the CCCC-MONITOR Task Team on the design of its CPR experiments and work. Marine mammal and bird experts now have their own Advisory Panel under the Biological Oceanography Committee (BIO). They will be providing scientific advice to BIO and CCCC, and providing leadership to marine mammal and bird researchers in the North Pacific in ecosystem research. The Marine Environmental Quality Committee has just formed a Working Group to examine the ecology of harmful algal blooms in the North Pacific, and the Fisheries Science Committee has formed a Working Group to consider the implications of climate change to fisheries management.

PICES IX will be held 20-28 October 2000 in Hakodate, Japan.

By Pat Livingston.

Total production figures were 27,010 with 8,450 test ages and 206 examined and determined to be unageable.

At the request of the Stock Assessment Program, the Observer Program has changed the sampling of fisheries to small random samples collected by all at-sea observers. This will affect the Age and Growth Program in two ways. First, during the second half of 1999 and into 2000 extra handling of otoliths requiring re-labeling and re-storing of vials will be required. Also, the number of species and numbers requested for ageing may change as a result of the random sampling method.

Joint research between REFM and NMML using radionuclides to age gray whales and bowhead whales was presented via a poster at the 13th Biennial Conference on the Biology of Marine Mammals in Wailea, Hawaii, held 28 November - 3 December 1999. The title of the poster was “The Potential use of Pb-210/Ra-226 Disequilibria for Age Determination of Mysticete Whales,” By Craig Kastelle, Kim Shelden, and Daniel Kimura (.pdf, 1.63 MB).

By Dan Kimura.

U.S. GROUNDFISH OBSERVER PROGRAM

During the fourth quarter of 1999, 127 observers were trained, briefed, and equipped for deployment to fishing and processing vessels and shore-side plants in the Gulf of Alaska, Bering Sea, and Aleutian Islands region. They sampled aboard 219 fishing and processing vessels and at 14 shore-side processing plants. These observers were trained or briefed in various locations. The Observer Program in Seattle trained 17 first-time observers and briefed another 18 observers with prior experience. The University of Alaska Anchorage (UAA) Observer Training Center briefed 30 observers and trained another 34. At the Observer Program’s field offices in Dutch Harbor and Kodiak, 4 observers were briefed and 24 were excused from briefing because they had just completed a cruise successfully and were returning immediately to the field. The fourth quarter 1999 observer workforce comprised 40% new observers and 60% experienced observers.

The Observer Program conducted a total of 212 debriefings during the fourth quarter of 1999. Nine debriefings were held in Kodiak, 13 in Dutch Harbor, 25 in Anchorage, and 165 in Seattle.

Highlights of 1999: CDQ and AFA Fisheries

Implementation of the expanded Community Development Quota (CDQ) program and of provisions of the recently enacted American Fisheries Act (AFA) continued during 1999. The CDQ program was developed for the purpose of allocating fishery resources to eligible western Alaska communities to provide the means for starting or supporting commercial fishery activities that would result in ongoing, regionally-based, commercial fishery, or related businesses. The CDQ program was initiated in 1992 with pollock and expanded to include fixed-gear halibut and sablefish in 1995. In 1998, it was further expanded to include multiple species of groundfish and crab (MSCDQ). In 1999, NMFS was responsible for monitoring the groundfish (including pollock and sablefish) and halibut CDQs, and the state of Alaska was responsible for monitoring the crab CDQs.

The AFA, enacted by Congress in late 1998, made changes to the pollock fishery in the Bering Sea and Aleutian Islands region. These changes reallocated fish among industry segments, provided for the formation of fishing cooperatives, and increased observer coverage levels on some components of the fleet. The offshore component of the fleet organized a fishing cooperative in 1999 and has received increased, mandatory observer coverage. More recently, the Observer Program has been involved in implementing aspects of the AFA related to shore-side pollock. The shore-side component is more complex than offshore and may involve NMFS regulatory actions and a changing role for the observer.

The MSCDQ and AFA catch accounting for offshore processors is based entirely on data collected by observers and, unlike the open access fisheries, where observer data is used to manage a fleet-wide quota, industry participants in the MSCDQ and AFA fisheries require individual accounting of fish harvested in each haul or set. This change in expectations placed on observers, their data, and the Observer Program in general has required much Observer Program staff effort in developing special selection criteria and training requirements for observers, developing new sampling strategies and regulations to enhance the observer’s working environment, and changes to the data collection and data management software systems.

Observer Program Review

An extensive, independent review of the Observer Program is being carried out by Marine Resources Assessment Group (MRAG) Americas, Inc. MRAG is an independent consulting firm which provides professional advice and services for the management of marine fisheries throughout the world. The purpose of the review is to provide recommendations for changes in Program operations and organization which might improve the Program’s ability to meet its mission and goals. Their final report is due later in 2000.

By Bob Maier.

SOCIOECONOMIC ASSESSMENTS PROGRAM

During the past quarter, Center economists have been involved principally in activities in support of the Pacific and North Pacific Fishery Management Councils and other cooperative efforts within the agency and beyond. Five research activities are reported on below.

Cost, Earnings, and Employment Survey

Final approval for the cost, earnings, and employment survey of the Alaska BSAI pollock industry was secured from the Office of Management and Budget (see “Economics and Management of the Alaskan Groundfish Fishery: a New Data Collection Project,” AFSC Quarterly Report, April-May-June 1999.) Before questionnaires are sent out, a letter will be sent to all potential respondents indicating that they will receive a questionnaire in approximately one week. Two weeks after the questionnaires are mailed, a follow-up letter will be sent to nonrespondents. A second follow-up letter will be sent two week later to individuals who have still not responded. There will also be follow-up phone calls or e-mails in some cases. Initial letters will be sent out 4 January and questionnaires on 10 January.

Spatial Modeling of Groundfish Fisheries

The Socioeconomic Assessment Program assisted with the bioeconomic model of the New England Groundfish Fisheries. A price model for groundfish developed at the Northeast Fisheries Science Center was integrated into the simulations to allow fish prices to vary with supply. Simulations were then rerun. The major conclusion from this work was that the integration of downward sloping demand curves had negligible impacts on the evaluation of the benefits derived from alternative area closures. While area closure tend to reduce output and increase price in the short-run, they have the opposite effect in the long-run, and the price effects largely cancel out when considering the present value of future revenues derived from the fishery with and without area closures. The primary conclusions from the research remain unchanged: 1) by redirecting fishing effort to less productive fishing areas, area closure can reduce effective fishing effort substantially; 2) area closures appear to offer a means of increasing revenues only when effort levels are excessive, but greater gains can usually be made by reducing nominal fishing effort or through direct controls on total catch levels; 3) area closures can provide habitat protection and allow for increases in equilibrium spawning stock biomass at a relatively low cost in forgone revenues even when effort is already close to MSY levels; 4) the location and shape of area closures will have important effects not only on how they affect overall harvest and revenue streams, but which user groups and individuals benefit. These effects are often not intuitively obvious, since they often result indirectly from redistribution of fishing effort.

Common Property Institutions

The Socioeconomic Assessment Program in collaboration with the NMFS Alaska Regional Office is initiating a study of common property institutions that have been developed by managers and by user groups in the BSAI groundfish fisheries to describe the process by which the groundfish resources off Alaska have been gradually transformed from essentially open-access to more manageable common property resources. In particular, the study focuses on the de facto partitioning of the larger groundfish fishery into a number of smaller fisheries as delineated by vessel size, gear, mode of operation, etc. and discusses how this may have facilitated the development by user groups of cooperative solutions to problems such as rent dissipation and excessive bycatch. Selected common property institutions that have been created in these fisheries are described in terms of the problems they were designed to solve, their success or lack of success in doing so, and the factors that influenced both emergence and success of these institutions. The institutions examined in the study include the North Pacific Fishery Management Council, the Community Development Quota system, the offshore Bering Sea pollock cooperative, and the cooperative bycatch control system known as Sea-State used by the non-pollock factory trawler fleet.

Fishery Participation Rate Modeling

Members of the Socioeconomic Assessment Program conducted econometric analysis for and coauthored the working paper “Modeling the Effect of Fishery Attributes on Participation Rates: The Kenai Peninsula Marine Sport Fishery.” The paper was prepared for use by NPFMC staff for use in an analysis of potential regulations affecting the Alaskan recreational halibut fishery.

Knowledge about factors that influence recreational anglers’ participation decisions is important to fishery managers. Changes in fishery regulations, environmental quality, and resource abundance, as well as trip cost, can affect the expected net benefit associated with a fishing trip, and therefore participation decisions. Models of participation decisions are consequently directly linked to natural resource damage assessment, the measurement of recreational benefits, and economic impact analysis. This study shows how hybrid stated preference questions can be used to estimate the affects of fishery attributes on participation rates. The stated preference method is a natural choice for such circumstances because anglers’ participation decisions will likely depend on multiple trip attributes. This approach allows for the simulation of a wide variety of alternative policy scenarios, many of which would not be possible using data from observed fishing activity. The non-linear model specification allows for substitution and complementary effects across attributes and the possibility of non-linear marginal utility.

Excess Processing Capacity

Program members also assisted in estimating the amount of excess capacity that exists in the in-shore pollock processing sector. The estimates were then used in an analysis of the American Fisheries Act (AFA) conducted by economists at the University of Washington.

Part of the analysis preformed at the Center was based on how AFA has affected the rate of pollock production in the at-sea sector between the years 1998 and 1999. This was an effort to determine whether and to what degree AFA co-ops have led to a reduction in the processing rate. The analysis was based on 17 separate catcher/processors vessels that processed Alaska pollock in both 1998 and 1999 (the year before and after the AFA took affect). Based on this analysis, the 1999 harvest and production rate was between 65 and 75 % of the 1998 rate. These results were then used to formulate a similar type of prediction for the in-shore sector.

It was estimated that the BSAI in-shore pollock processors would use less than 50 percent of their processing capacity in 2000. Therefore, they could increase their production by more than 100 % without increasing their capacity.

By Joe Terry.

Resource Ecology & Fisheries Management

Resource Ecology &
Fisheries Management