Alaska Fishery Research Bulletin Issues, Vol.9 No. 1 - Summer 2002

Preseason Forecasts of Pink Salmon Harvests in Southeast Alaska Using Bayesian Model Averaging

Milo D. Adkison - Vol. 9(1):1-8. 2002.

Competing models for generating a forecast of salmon returns may give differing predictions yet have similar credibility. Within a single model, a wide range of alternative parameter values may also have similar credibility. Bayesian methods allow us to incorporate a wide range of models, weighting their forecasts by the posterior probability of each model and parameter combination. I illustrate this procedure by forecasting pink salmon harvests for the year 2002 for northern and southern Southeast Alaska.

Top of Page

Full Article (PDF 402 kB)

A Statistical Approach to Estimating Fish Passage Using a Form of Echo Integration

Pham X. Quang and Harold J. Geiger - Vol. 9(1):9-15. 2002.

Sonar echo integration is widely used to assess fish density when individual targets cannot be tracked or counted. We propose an echo-integration estimator of fish density, which is proportional to the ratio of the echo integral to an estimate of the average, squared echo-voltage amplitude. Our argument is based on the fact, pointed out by Ehrenberg, that amplitudes of returning echoes constitute an inhomogeneous filtered Poisson process. Our estimator, unlike the classical echo integrator, is not based on the thin-shell approximation. It uses thresholded echoes and accommodates background noises, and thus could be more appropriate in riverine environments.

Top of Page

Full Article (PDF 79 kB)

A Review of the Net Selectivity Problem and a Model for Apportioning Species Based on Size-Selective Sampling

Pham X. Quang and Harold J. Geiger - Vol. 9(1):16-26. 2002.

The theory of inference of a size distribution for fish captured with size-biased fishing gears is called the “net selectivity problem.” This family of problems arises because of questions about the underlying distribution of the sampled fish, questions about the nature of the size bias, or perhaps questions about the mixtures of species of different sizes. We discuss these problems in the context of size-biased gillnets, although generalization to other kinds of fishing gear is straightforward. We estimate the length distribution of the captured fish, the size distribution of the fish population, and the parameters of the selectivity curves parametrically, using maximum likelihood. To avoid overparametrization we make strong assumptions about the relationship among the selectivity curves, so that fishing power cannot be estimated in this setup. This assumption is called “geometric similarity.” The normal-normal and gamma-gamma combination models are considered, so called because of the shape of the fish-length distribution and the shape of the selectivity curves. The method is demonstrated with a worked simulation example and the American plaice Hippoglossoides platessoides data of Holst and Moth-Poulsen.

Top of Page

Full Article (PDF 386 kB)

German Participation in Alaska Sport Fisheries in 1998

Mark Herrmann, Laura M. Milner, Kelly L. Giraud, Michelle Skogen Baker, and Rod F. Hiser - Vol. 9(1):27-43. 2002.

This study is the first characterization of German sport fishing anglers in Alaska. A survey was taken of 601 residents of Germany who had purchased a 1998 Alaska fishing license. Of the 601 individuals surveyed, 363 responded, and after eliminating the undeliverable surveys, a response rate of 61.7% was achieved. The German anglers who participated in the Alaska sport fishery are predominantly male, employed, married, and have children. The average age is approximately 45 years old. Fully one-quarter of the surveyed anglers came from the German state of Bavaria. More than half of the German anglers fishing Alaska have visited Alaska multiple times. Reasons for return trips include the Alaska wilderness, abundance of salmon and halibut, beautiful countryside, friendly people, remoteness, solitude, nature, and scenery. Salmon were the most popular targeted species by German anglers, with sockeye Oncorhynchus nerka, coho O. kisutch, and chinook salmon O. tshawytscha leading the way. The results of this study show a high level of satisfaction among German anglers who fish in Alaska. The level of satisfaction was modeled and the parameters were estimated using an ordered probit model. The results of the German angler satisfaction estimation indicate that trip satisfaction was likely to be increased for those fishermen who exceeded their expectations in regard to the number and size of fish caught, were dedicated to fly-fishing, who exhibited pre-trip preferences towards a favorable fishing environment, and who encountered less crowding conditions than expected.

Top of Page

Full Article (PDF 143 kB)

Estimated Mortality of Snow Crabs Chionoecetes opilio Discarded During the Bering Sea Fishery in 1998

Jonathan J. Warrenchuk and Thomas C. Shirley - Vol. 9(1):44-52. 2002.

The mortality of discarded snow crabs Chionoecetes opilio was estimated for the 1998 fishery for snow crabs in the Bering Sea. Estimates of discard mortality were calculated from deadloss of retained crabs, a windchill model of mortality was developed from the results of a laboratory study, and a model that predicts mortality from temperature and windspeed was developed. The 1998 season was used because it had the most complete set of data. No relationship existed between the deadloss reported in the catch delivered to processors and the windchill conditions when the crabs were caught. Mortality of discarded snow crabs was estimated at 3.9% by the windchill model and 19.6% by the temperature and windspeed model.

Top of Page

Full Article (PDF 172 kB)

An Overview of Salmon Bycatch in Alaska Groundfish Fisheries

David Witherell, David Ackley, and Cathy Coon - Vol. 9(1):53-64. 2002.

Chinook salmon Oncorhynchus tshawytscha and chum salmon O. keta are caught incidentally in Alaska groundfish fisheries, primarily in the walleye pollock Theragra chalcogramma trawl fishery. From 1990-2001, an average of 37,819 chinook salmon and 69,332 other salmon species (> 95% are chum salmon) were incidentally caught annually in the Bering Sea and Aleutian Islands groundfish trawl fisheries, and 20,799 chinook salmon and 20,496 other salmon were incidentally caught annually in the Gulf of Alaska trawl fisheries. In 1999 and 2000, chinook salmon bycatch was reduced in the Bering Sea, but increased in the Gulf of Alaska. Chum salmon bycatch has remained relatively stable in recent years. Bycatch is primarily juvenile salmon that are one or two years away from returning to the river of origin as adults. The origin of salmon taken as bycatch includes rivers in western Alaska, Southcentral and Southeast Alaska, Asia, British Columbia, and Washington. Analysis indicates that an incidental catch of 30,000 chinook salmon in Bering Sea and Aleutian Islands groundfish trawl fisheries equates to about 14,581 adult chinook salmon from western Alaska. Similarly, a bycatch of 60,000 chum salmon in Bering Sea and Aleutian Islands groundfish trawl fisheries equates to about 13,120 adult chum salmon from western Alaska. We estimated that, on average, salmon bycatch in Bering Sea and Aleutian Islands groundfish trawl fisheries reduced the western Alaska chum salmon run by less than 0.2%, and reduced the western Alaska chinook salmon run by less than 2.7%. Impacts of salmon bycatch from the Gulf of Alaska groundfish trawl fisheries cannot be estimated at this time.

Top of Page

Full Article (PDF 362 kB)

New Records of the Slender Codling Halargyreus johnsonii Günther, 1862 from the Eastern Bering Sea, Alaska

Gerald R. Hoff - Vol. 9(1):65-67. 2002.

Two specimens of the slender codling Halargyreus johnsonii Günther, 1862 were collected from the eastern Bering Sea, the most northerly records from the eastern North Pacific. The two immature specimens were collected in June of 2000 by bottom trawl during the Alaska Fisheries Science Center’s groundfish survey of the eastern Bering Sea upper continental slope. These two individuals appear similar to all previously reported specimens from the Pacific and extend the range for the species to Alaska. Also reported herein are 5 previously unreported records of H. Johnsonii collected just south of the Gulf of Alaska in the eastern North Pacific.

Top of Page

Full Article (PDF 94 kB)