Alaska Fishery Research Bulletin Issues, Vol.4 No. 1 - Summer 1997

Diurnal Variation in the Catch of Salmon in Drift Gillnets in Lynn Canal, Alaska

Randolph P. Ericksen and Robert P. Marshall - Vol. 4(1):1-11. 1997.

Operators of 4 commercial drift gillnet vessels kept logbooks of their 1991 fishing effort and catch of sockeye Oncorhynchus nerka, chum O. keta, coho O. kisutch, and chinook O. tshawytscha salmon in Lynn Canal, Alaska. Mann-Whitney tests were used to determine if catch per unit effort (CPUE) for each species varied significantly by ambient light phase (day vs night). Two definitions of day and night were employed: one relative to sunrise and sunset and another relative to civil twilight. All data were temporally stratified into 3 fishing periods of 5 or 6 weeks each. Logbook fishers caught 7,840 sockeye, 6,330 chum, 3,579 coho, and 187 chinook salmon in 1,060 sets between June 16 and October 2. Chinook salmon were only caught incidental to other species, and 82% were classified as small (£8 lb) fish. The chinook harvest rate was significantly higher at night than at day, except during the final (fall) strata when very few chinook salmon were caught. During the peak of directed fisheries for sockeye salmon (mid July through August), CPUE for sockeye salmon was highest during the day. During September, CPUE for chum salmon was also higher during the day. Results provide a solid indication that considerable savings in chinook salmon could result from night closures in drift gillnet fisheries of inside marine waters.

Top of Page

Full Article (PDF 94 kB)

Mark-Recapture Abundance Estimate of Fall-Run Chum Salmon in the Upper Tanana River, Alaska, 1995

Thomas A. Cappiello and Jeffrey F. Bromaghin - Vol. 4(1):12-35. 1997.

The primary objective of this study was to determine the feasibility of a mark-recapture program to estimate inseason and postseason abundance of fall-run chum salmon Oncorhynchus keta in the upper Tanana River. We used 2 fish wheels, each on opposite riverbanks, to capture fish for tagging and 2 additional fish wheels on opposite riverbanks to recover tagged fish approximately 76 km upstream. All chum salmon caught during a daily 6-h schedule were marked with spaghetti tags. From 16 August through 30 September 1995, a total of 3,993 fall-run chum salmon were released with orange tags from the right bank; 181 were released with yellow tags from the left bank. From 18 August to 1 October, the right-bank recovery wheel caught 6,773 chum salmon, of which 103 were recaptures (94 orange; 9 yellow). During the same period, the left-bank recovery wheel caught 3,902 chum salmon, of which 63 were recaptures (55 orange; 8 yellow). Catches from both recovery wheels were pooled because tagged fish were not bank-oriented and there was no statistically significant difference in the marked proportions between each wheel. However, information from yellow-tagged fish was not adequate for estimating abundance. Bailey's closed population model produced a total estimate of 268,173 (SE = 21,597) fall-run chum salmon that passed the tagging site after 16 August. No significant sources of bias from assumption violations were detected. The mean migration rate between tagging and recovery sites was 26 km/d. We concluded that a mark-recapture program using fish wheels for fish capture appears feasible but should continue in a developmental stage to allow further evaluation of its utility under a variety of circumstances. Tagging fish from only the right-bank tagging wheel and tagging more fish, while using 2 wheels for tag recovery, may be necessary to improve precision of the abundance estimate.

Top of Page

Full Article (PDF 468 kB)

Time Series Outlier Analysis: Evidence for Management and Environmental Influences on Sockeye Salmon Catches in Alaska and Northern British Columbia

Edward V. Farley, Jr., and James M. Murphy - Vol. 4(1):36-53. 1997.

Autoregressive, moving average models were fit and outliers were identified for commercial catches of 9 major sockeye salmon Oncorhynchus nerka stocks in Alaska and northern British Columbia. Distinct patterns in the sample autocorrelation and partial autocorrelation functions indicated stock-specific dynamics. Three types of outliers were considered: level-shift, temporary-change, and additive outliers. Most additive outliers were unexplainable and may represent multiplicative survival at several different life history stages. Additive outliers that could be explained resulted from changes in fishing effort. Temporary-change outliers commonly reflected cold winter temperatures in western Alaska during the early 1970s. Four of nine river systems in our analysis had level-shift outliers, and only one of these had a positive shift in the late 1970s. The level-shift outliers, which indicate a long-term shift in catch levels, appeared to be the result of changes in escapement policy rather than an abrupt change in the production dynamics of the North Pacific.

Top of Page

Full Article (PDF 182 kB)

Timing of Occurrence and Length Frequency Profiles of Walleye Pollock Larvae in Resurrection Bay, Alaska, Spring 1991

Linda A. Clayton, A. J. Paul, and Judy M. Paul - Vol. 4(1):54-61. 1994.

In 1991 the major cohorts of walleye pollock Theragra chalcogramma larvae in Resurrection Bay, Alaska, initiated feeding when the copepod nauplii population was still at low winter density levels. First-feeding walleye pollock larvae were most abundant between 16 April and 14 May; thereafter, there were very few recently hatched larvae in the plankton. Copepod nauplii did not begin their annual increase in abundance until mid June. Throughout the spring and summer few walleye pollock larger than the first-feeding stage were observed, suggesting that low prey abundance was limiting larval growth. In 1991 the small late-hatching cohorts had the best feeding conditions in the bay.

Top of Page

Full Article (PDF 154 kB)

Application of a Catch-Survey Analysis to Blue King Crab Stocks Near Pribilof and St. Matthew Islands

J. Zheng, M. C. Murphy, and G. H. Kruse - Vol. 4(1):62-74. 1996.

A catch-survey analysis was conducted on 2 blue king crab Paralithodes platypus stocks near Pribilof and St. Matthew Islands of the eastern Bering Sea. Total annual catch and observed relative survey abundance of legal-sized crabs were classified into recruits and postrecruits by shell age and size and analyzed with a maximum likelihood approach. This analysis separates measurement errors from the true changes in population abundance and converts the relative abundance to an absolute value. The analysis provides smoothed annual estimates of legal blue king crab abundance for both stocks. In both areas, legal male abundances appear to be rebuilding slowly from depressed stock levels of the mid 1980s. Based on a reasonable instantaneous natural mortality range of 0.16 to 0.35 for both crab stocks, (1) postrecruit catchability is about 1.98-2.07 for the Pribilof Islands stock and about 1.20-1.38 for the St. Matthew Island stock, and (2) the ratios of recruit-to-postrecruit catchabilities for both stocks are <1.05. Because of confounded effects among catchability, recruit-to-postrecruit catchability ratio, and natural mortality, additional information is needed to separately estimate 2 of these 3 parameters.

Top of Page

Full Article (PDF 295 kB)

Are Prince William Sound Salmon Hatcheries a Fool's Bargain

William W. Smoker and Tim J. Linley - Vol. 4(1):75-78. 1997. No Abstract.

Top of Page

Full Article (PDF 53 kB)

Call For Data: Bering Sea Ecosystem Metadatabase

Bernard A. Megrey - Vol. 4(1):79-80. 1997. No Abstract.

Top of Page

Full Article (PDF 15 kB)