Link to NMFS Homepage Link to NOAA Homepage Keyboard Navigation Alaska Fisheries Science Center Home banner



2003 Gulf of Alaska Biennial Groundfish Assessment Survey

The third in the series of biennial bottom trawl surveys of Gulf of Alaska (GOA) groundfish resources was conducted from 20 May through 9 August 2003. Prior to establishing a biennial schedule in 1999, the RACE Division had surveyed groundfish resources in the GOA triennially in 1984, 1987, 1990, 1993, and 1996. The GOA triennial surveys covered the continental shelf (out to 500 m depth) but only included portions of the continental slope in 1984 (to 825 m) and 1987 (to 750 m). The biennial surveys since 1999 were designed to cover the continental shelf and slope between the Islands of Four Mountains (long. 170 °W) and Dixon Entrance (U.S.-Canada border in Southeast Alaska) out to the 1,000-m depth contour. While the 1999 survey succeeded in sampling the entire area, the 2001 survey area was reduced due to the Division’s increased survey responsibility in other areas under limited funding. The 2001 survey area did not include the area east of long. 147°W, nor did it extend deeper than 500 m. This year the survey covered the entire geographic extent, but the outermost depth stratum (700-1,000 m) was omitted because the vessels were unable to fish that deep.

Sampling was conducted aboard three chartered commercial trawlers, Sea Storm, Gladiator, and Northwest Explorer. The 75-day survey period was divided into four legs of 18-19 days each. Sampling began near the Islands of Four Mountains and progressed eastward on the continental shelf and slope to the U.S.-Canada border in Southeast Alaska. Stations were allocated among 54 depth and geographic strata and were preselected randomly from a grid of potential sites overlaying the survey area. If rugged bottom or heavy commercial fishing prevented sampling a station, a nearby alternate station was selected. Of the 880 attempted standard survey tows, 809 were successfully completed, ranging in depth from 13 m to 667 m.

The primary focus 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 2003 survey were to:

  1. Define the distribution and relative abundance of the principal groundfish and invertebrate species inhabiting the continental shelf and slope of the GOA
  2. Collect data to define various biological characteristics of major groundfish species, such as age, sex, size, growth rates, length-weight relationships, and feeding habits
  3. Collect integrated fishing performance, net configuration, and position data for all trawl hauls with which to derive precise effort estimates
  4. Collect environmental data such as surface-to-bottom water column temperatures
  5. Collect biological specimens and data requested by scientists from the AFSC or other cooperating research groups.

Results from the survey are preliminary and will be finalized following further examination of the effort data for each tow.

When looking at changes since the last survey in 2001, we can only compare abundance estimates in the central and western subareas (west of long. 147 °W) shallower than 500 m. In that area, the most abundant species in 2003 were, in order, arrowtooth flounder, Pacific halibut, walleye pollock, Pacific ocean perch, Pacific cod, and flathead sole. Since 2001, the estimated abundances of all of these species except Pacific ocean perch increased: arrowtooth flounder by 87% to 2,540,000 metric tons (t), halibut by 50% to 518,000 t, pollock by 84% to 387,000 t, cod by 10% to 283,000 t, flathead sole by 56% to 239,000 t. The abundance estimate of Pacific ocean perch declined by 47% to 356,000 t.

We can compare abundance estimates from the entire survey area between 2003 and 1999 except for species that occur commonly in the deepest stratum (700-1,000 m), such as grenadier. Over the entire GOA survey area, arrowtooth flounder was by far the most abundant species with a total biomass estimate of more than 2.8 million t, a 124% increase over the 1999 estimate. Nearly 78% of its biomass was from the central GOA survey subarea. The second most abundant species was Pacific halibut with a biomass estimate of 634,000 t, an 8% increase since the 1999 survey, with 66% of its biomass coming from the central subarea. Pacific ocean perch ranked third in abundance at 457,000 t, down 37% from 1999 with 62% of its biomass occurring in the central GOA. Walleye pollock ranked fourth in abundance at 425,000 t, a 33% decrease since 1999, with 50% and 41% of its biomass coming from the western and central GOA, respectively.

By Mark Wilkins


picture of survey catch aboard the Arcturus
One of the larger survey catches (primarily walleye pollock) aboard the Arcturus during the 2003 eastern Bering Sea groundfish survey. Photo by Erika Acuna.

2003 Eastern Bering Sea Shelf Crab-Groundfish Bottom Trawl Survey

The Bering Sea Resource Assessment group completed their annual Bering Sea shelf crab-groundfish bottom trawl survey from 2 June to 22 July 2003. This year’s survey was performed aboard the chartered fishing vessels Arcturus and Aldebaran. This marks the 11th consecutive survey using these vessels.

In general, temperatures were warmer than the long-term mean beginning in 1982. Mean bottom temperature was 3.81° C, compared to the long-term value of 2.44° C. Surface values were also higher at a mean of 7.79° C and a long-term mean of 6.60° C.

A total of 399 stations were sampled including 18 special stations from inner Bristol Bay to continue our evaluation of yellowfin sole spawning populations during the sampling period.

Abundance estimates for the major species showed some unexpected changes from 2002 (Table 1 below). Walleye pollock abundance was significantly higher in 2003; increasing to 8.5 million t. Pacific cod decreased slightly from 616,900 t in 2002 to 605,700 t this year.

Table 1.  Bering Sea resource assessment bottom trawl survey biomass estimates.
Species 2000 2001 2002 2003

Pacific cod
Yellowfin sole
Rock sole
Flathead sole
Alaska plaice
Arrowtooth flounder
Greenland turbot


Among the flatfish species only flathead sole decreased slightly in abundance this year at 529,200 t, down from 574,900 t in 2002. Yellowfin sole, rock sole, Alaska plaice, arrowtooth flounder, and Greenland turbot all showed increases in abundance with the most significant increase being in arrowtooth flounder at 553,900 t this year, up from 355,100 t in 2002.

It should be noted that these abundance estimates are the bottom trawl survey biomass estimates and NOT the status of stocks final model estimates.

After the standard survey was completed on 22 July, two important experiments were performed during 23 July - 2 August 2003. The first was a 9-day special project conducted in the southeast Bering Sea which examined trawl footrope selectivity for skates by utilizing an auxiliary net beneath the survey trawl to account for skates escaping under the footrope. The results of this study will be available next year and will estimate capture probability as a function of skate size.

An exploratory study was conducted in late July 2003 aboard the Aldebaran in the Pribilof Canyon area focusing on northern rockfish Sebastes polyspinis. The goal of the 4-day study was to evaluate the feasibility of sampling patchily distributed rockfish with hydroacoustic and bottom trawl methods in which the two methods would be combined to increase the precision of the biomass estimates.

By Erika Acuna.

Bottom Trawl Survey Protocol Experiments

In order to better understand the factors affecting the performance of our standard survey trawls, the Groundfish Assessment Program conducted three trawl performance experiments in September aboard the chartered fishing vessel Vesteraalen, using the standard RACE Bering Sea survey 83-112 bottom trawl. The first experiment investigated the effect of varying trawl wire offset on the trawl bridle and footrope contact with the seafloor. The second experiment studied the effect of three different trawling modes (locked winches, tension-controlled autotrawl, and symmetry-controlled autotrawl) on both trawl footrope and bridle contact with the bottom. The third experiment examined the effect of vessel heave, pitch, and roll on the bridle and footrope contact with the seafloor. Analyses of the data are currently in progress.


quarterly July-Sept 2003 sidebar Contents


Auke Bay Lab

National Marine
Mammal Lab

RACE Division

REFM Division

Quarterly Index

Quarterly Home