RACE Report OND '99

1999 Gulf of Alaska Biennial Groundfish Assessment Survey

The first in the new series of biennial bottom trawl surveys of Gulf of Alaska (GOA) continental shelf and slope groundfish resources was conducted by the Resource Assessment and Conservation Engineering (RACE) Division from 10 May through 28 July 28 1999. Prior to the 1999 survey, groundfish resources in the GOA had been surveyed by the RACE Division on a triennial schedule with surveys conducted in 1984, 1987, 1990, 1993 and 1996. The triennial surveys covered the continental shelf of the GOA in each of the five previous surveys but only included the continental slope in 1984 and 1987. Beginning in 1999, the biennial bottom trawl survey effort in Alaska will alternate between surveys of the GOA shelf and slope one year followed by surveys of the Aleutian Islands shelf and Bering Sea continental slope the next year. Bering Sea shelf groundfish and crab resources will continue to be surveyed on an annual basis. This report summarizes the general sampling operations and preliminary results of the 1999 GOA shelf and slope bottom trawl survey.

Vessels and Gear
Survey sampling was conducted aboard the chartered commercial trawlers Vesteraalen, Dominator, and Morning Star. All three charter vessels are house-forward trawlers with stern ramps, multiple net storage reels (mounted forward of the working deck and/or aft over the stern ramp), telescoping deck cranes, propeller nozzles, and paired, controlled-tension hydraulic trawl winches with 1,280 m to 2,190 m of 2.54 cm diameter steel cable. The Vesteraalen and Dominator are both 38 m in overall length (LOA) and powered by single, 1,700 and 2,000 continuous horsepower (HP) main engines, respectively. The Morning Star is 45 m LOA and propelled by a 1,700 HP main engine.

The 75-day survey period from 10 May to 23 July was divided into four legs of 18-19 days each. The Vesteraalen charter period was extended an additional 5 days from 24 July to 28 July to complete stations in the eastern GOA. Sampling operations began near the Islands of Four Mountains (long.170°W) and extended eastward throughout the GOA on the continental shelf and upper continental slope to Dixon Entrance (long.132°30’W). Sampling occurred at pre-selected stations, or nearby alternate stations, in depths ranging from 16 m to 946 m (Figure 1).

The primary purpose of the biennial groundfish surveys is to build a standardized time series of data designed to assess, describe, and monitor the distribution, abundance, and biological condition of various GOA groundfish stocks. Specific objectives of the 1999 survey were to:

Standard RACE Division Poly-Nor’eastern high opening bottom trawls, rigged with roller gear, were used exclusively for sampling the selected survey stations. Experimental sampling was conducted with reinforced Poly-Nor’eastern nets equipped with tire gear at sites previously found to be too rugged for the standard roller gear trawl, however these data were not used to estimate relative abundance and biomass. Steel V-doors with dimensions of 1.83 x 2.74 m, and weighing approximately 800 kg each were used to open the net.

The GOA biennial survey area is dominated by the continental shelf (depths to 200 m) which is crossed by numerous gullies or troughs, some as deep as 300 m. The continental shelf represents approximately 71% of the 319,884 km2 survey area with associated gullies representing an additional 17%. The width of the shelf area varies from approximately 18.5 km in the Unalaska Island- Islands of Four Mountains region to 185 km off the Kenai Peninsula (Figure 1). Irregular bathymetric features along with an extensive and complex shoreline region, provide a rich diversity of habitat types for many juvenile and adult groundfish such as walleye pollock (Theragra chalcogramma), Pacific cod (Gadus macrocephalus), juvenile sablefish (Anoplopoma fimbria), a wide variety of flatfish species including Pacific halibut (Hippoglossus stenolepis), rockfish species such as Pacific ocean perch (Sebastes alutus), and many invertebrates including several commercial crab species, scallops, and pandalid shrimp.

The continental slope (excluding major gullies) between the depths of 200 m and 1,000 m represents approximately 12% of the survey area. Often steep and rugged, the slope region provides habitat for species such as rougheye and shortraker rockfishes (Sebastes aleutianus and S. borealis), adult sablefish, shortspine thornyheads (Sebastolobus alascanus), Dover sole (Microstomus pacificus) and grenadiers (Macrouridae).

Survey Design and Methods
As in past years, a stratified random survey design was employed in the 1999 biennial survey. The GOA survey region was divided into 59 strata based on bathymetry, major geographic features, and International North Pacific Fisheries Commission (INPFC) statistical areas. Strata range from near-shore areas adjacent to coastlines on the continental shelf to 1,000 m depths on the upper continental slope.

A modified Neyman optimum allocation strategy based on data from the 1990, 1993, and 1996 triennial surveys was used to allocate effort between strata. Neyman optimum allocation calculations were made for each of the principal groundfish species in each survey year based on that year’s survey data and the estimated time to perform a tow in a given stratum as the cost variable (deeper tows take longer to execute, therefore cost more). A mean of the resulting proportions was then calculated, resulting in an estimate of optimal allocation for each of the principal groundfish species. A weighted mean of these values was then calculated using each species’ mean biomass as the weighting variable. Within each major depth interval (1-100 m, 101-200 m, etc.), the number of stations was then summed resulting in an optimal allocation between depth intervals. This number of tows was then reallocated between strata, proportional to the area of each stratum so that the sampling density was constant within each depth interval throughout the survey area.

The target duration for on-bottom time for a standard trawl haul was 15 minutes. Trawling time on bottom was estimated during the tow using real-time net configuration data (wingspread and headrope height) acoustically transmitted to the vessel. GPS data were collected every 2 seconds throughout the tow. Temperature and depth were recorded every 6 seconds by a bathythermograph attached to the trawl headrope. A bottom contact device, based on a tilt sensor attached to the fishing line to detect contact with the bottom, collected data every 6 seconds. The final tow durations, start and end times, and geographical positions were estimated from all information collected from each tow.

Catches were sorted to species, weighed, and enumerated according to standard AFSC and RACE Division protocol. Extensive size composition data were collected with barcode-based recording devices and downloaded to computer database files after each tow. A variety of biological data including age structures (mostly otoliths), lengths, and weights of individual specimens were collected and entered in the computer database.

Results
Sampling proceeded from west to east. In the Shumagin and Southeastern INPFC areas, an equal number of stations from each stratum were randomly assigned to each of the three vessels after accounting for tows deeper than 700 m being sampled exclusively by the Morning Star and stations between 500 m and 700 m being sampled equally by the Morning Star and Dominator. Stations in the Chirikof, Kodiak, and Yakutat INPFC areas were assigned to minimize the running time between stations for each vessel. Some pre-assigned stations were not sampled due to unsuitable bottom conditions. In cases where trawlable bottom could not be found at a given station, a preselected alternate location was sampled. Of the 832 attempted standard survey tows, 764 tows were successfully completed ranging in depth from 16 m to 946 m.

Sea surface temperatures and bathythermograph recordings were collected at nearly every trawl site. The average sea surface temperature for the entire survey period was 8.0^oC with temperatures increasing from 4.1^oC in May, to 7.1^oC in June and 11.5^oC in July. The mean bottom temperature for the survey period was 5.1^oC with temperatures increasing from an average of 3.6^oC in May to 5.0^oC in June and 5.8^oC in July. Bottom temperatures decreased with increasing depth from 5.2^oC at depths shallower than 100 m to 3.2^oC between 900 and 1,000 m of bottom depth.

Arrowtooth flounder (Atheresthes stomias), Pacific ocean perch, Pacific halibut, walleye pollock, giant grenadiers (Albatrossia pectoralis) and Pacific cod, were the dominant species in trawl catches representing nearly 70% of the 338,932 kg total survey catch. Distribution and relative abundance highlights for these major species include:

Throughout the survey, biological data were collected from a wide variety of species with length measurements being the most common. Nearly 219,000 fish representing 74 species were measured for length including approximately 56,000 arrowtooth flounder, 24,000 pollock, 18,000 shortspine thornyheads, 16,000 flathead sole, 13,000 Pacific ocean perch and 12,000 rex sole. Approximately 11,000 otoliths were collected from 22 species along with 12,000 length-weight observations representing 39 species. In addition, over 2,400 stomach samples were collected from four major predator species: arrowtooth flounder, Pacific cod, Pacific halibut, and walleye pollock.

By Eric Brown.

Trawl Impact Studies in Eastern Bering Sea

The AFSC’s trawl impact project examines possible adverse effects of bottom trawls on soft-bottom benthos in the eastern Bering Sea. Earlier studies revealed chronic effects on community diversity and on individual macrofauna populations. However, interpretation of these findings and effective use for management purposes requires some understanding of the underlying processes. To address this need, a new multiyear study in the Crab and Halibut Protection Zone 1 (also known as management area 512) is being planned. Scheduled to begin in summer 2000, it will investigate acute effects and recovery from a single repetitive trawling event. Detailed physical information and historical trawl effort data have been assembled to identify suitable experimental sites. Epifauna and infauna data collected in 1996 and 1997 are being analyzed to identify appropriate sample sizes for the research trawl (epifauna) and benthic grab (infauna) sampling efforts. Sidescan sonar, acoustic seabed classification and subsampling of benthic grabs will be used to characterize physical and chemical effects (in collaboration with scientists at the University of Alaska Fairbanks).

The before-after/control-impact (BACI) experimental design involves repeated sampling of specific sites to compare biotic and abiotic conditions before and after trawling. This requires accurate real-time positioning of sampling gear and the commercial trawl used to impact the experimental corridor. In May 1998, three ultra-short baseline (USBL) systems were tested in a fixed short baseline (SBL) tracking array maintained in Puget Sound by the U.S. Navy Naval Undersea Warfare Center Division Keyport. Using a chartered Bering Sea trawler operating under representative study conditions, this project demonstrated the feasibility of real-time trawl positioning. Comparison with Navy SBL fixes provided estimates of USBL positioning error for two systems, the Nautronix ATS II (3.7 m) and the ORE Trackpoint II Plus (5.9 m). When all other sources of error (e.g., errors due to GPS, the gyro and sound velocity estimates) are considered, along-track errors of 8.4 m and 10.6 m and cross-track errors of 11.6 m and 13.8 m, respectively, can be expected for a trawl fishing in 60-65 m of water. An over-the-side hydrophone pole suitable for chartered fishing vessels was also developed and tested. Complete details are available in “Report and Analysis of Test Data from 1998 Trawl Position Gear Trials in the Puget Sound Dabob Bay, Washington” incorporating technical input from all contractors and is available from Robert McConnaughey.

By Robert McConnaughey.

Influence of Bottom Temperature on Distribution and Abundance Estimates from EBS Crab/Groundfish Bottom Trawl Survey

Figure 2. Average bottom temperature among stations sampled in the standard area of the AFSC eastern Bering Sea crab/groundfish survey. High density stations, those on the border of the 20 x 20 nautical mile grid, were excluded. Error bars indicate 95% confidence intervals.

Bottom temperatures during the 1999 eastern Bering Sea crab/groundfish bottom trawl survey were the coldest among these surveys to date (Figure 2 above). Spring-summer surveys in the eastern Bering Sea have been conducted by the AFSC since 1975 and on an annual basis since 1979. Although the survey in 1999 began approximately 2-weeks earlier than previous years, the time difference did not account for the extreme difference in bottom temperatures.

Figure 3. Survey biomass estimates of major flatfish species
in the eastern Bering Sea from 1982 to 1999.

Distributions and biomass estimates for many species examined during the 1999 survey indicate a substantial deviation from previous years. Decreases in biomass from 1998 estimates occurred for major flatfish species such as yellowfin sole (44% decrease), rock sole (22%), flathead sole (44%), and arrowtooth flounder (29%) (Figure 3 above). The biomass estimate for yellowfin sole was the lowest recorded since the survey began using the standard 83-112 eastern bottom trawl in 1982. Alaska plaice was the only major flatfish species in which the estimated biomass did not decrease from 1998 to 1999. The decreased estimates of biomass among flatfish species may in part be due to decreased catchability owing to these species behavior to trawl gear in colder temperatures. If flatfishes are less active in colder than normal waters, the likelihood for escapement under the footrope may increase and a herding effect by the trawl warps may decrease. In addition, for yellowfin sole, their distribution in 1999 appeared closer to shore than in previous years; catch per unit effort values in 1999 were higher among survey stations closest to shore compared with previous years. Because yellowfin sole migrate during spring to nearshore spawning waters (< 30 m), some of which is unavailable to the survey, a change in their distribution may account for the decrease in estimates of yellowfin sole biomass in 1999.

Biomass estimates for major roundfish species such as walleye pollock and Pacific cod increased from 1998 to 1999. Walleye pollock estimates increased from 2.21 to 3.57 million t, and estimates for Pacific cod increased from 558,419 to 583,259 t. While an increase in biomass estimates were somewhat expected, their distributions during 1999 differed from previous years. In 1999, walleye pollock were more concentrated to the northwest area of the survey and appeared to avoid the north central portion where a cold pool of less than 0EC water was well developed. Walleye pollock also had more westerly distributions during other cold years (i.e., 1992, 1994, 1995) compared with warmer years. Similarly, Pacific cod was more concentrated to the west in 1999, with relatively low concentrations in the inner part of the survey (< 50 m bottom depth) compared with previous years.

By Dan Nichol.

West Coast Upper Continental Slope Groundfish Resources

The RACE West Coast Groundfish Team completed a bottom trawl survey of the groundfish resources of the West Coast upper continental slope (WCUCS) between the U.S.-Canada border and Point Arguello (lat. 34°50’N) in waters 183-1,280 m deep from 14 October to 19 November aboard the NOAA ship Miller Freeman. This was the eleventh survey in an ongoing series to monitor long-term trends in the distribution and abundance of WCUCS groundfish populations. Previous groundfish assessment surveys of the West Coast upper continental slope were conducted in 1984, from 1988 to 1993, and from 1995 to 1997. Two hundred seven (207) tows were attempted during the 1999 survey. Out of 208 possible stations, 199 were sampled successfully. Remaining attempted tows were unsuccessful due to poor or uncertain gear performance. Nine stations were abandoned without any attempt because extensive acoustic surveys of the sites revealed they were too rough or steep for successful completion of a standardized tow. Resource information from this survey series is used to assess and manage several WCUCS species including sablefish (Anoplopoma fimbria), shortspine thornyhead (Sebastolobus alascanus), longspine thornyhead (S. altivelis), and Dover sole (Microstomus pacificus).

By Bob Lauth.

Resource Assessment & Conservation Engineering

Resource Assessment &
Conservation Engineering