Abstract.- Northern shrimp Pan- 

 dalus borealis carapace length-fre- 

 quency data collected from shrimp 

 surveys and commercial catches be- 

 tween 1972 and 1986 were studied 

 to estimate age, growth, and mortal- 

 ity in Pavlof Bay, Alaska. Dominant 

 size modes representing 1971 and 

 1975 year-classes were separated 

 from carapace length-frequency dis- 

 tributions by the use of a maximum- 

 likelihood separation technique. The 

 growth rate for the 1975 year-class 

 was significantly greater than that 

 of the 1971 year-class. Because the 

 abundance of the 1975 year-class was 

 lower than that of the 1971 year- 

 class, an inverse relationship of 

 growth to year-class strength was 

 suggested. Age-at-sex transition ap- 

 pears to be related to year-class 

 abundance or overall density as 1975 

 year-class shrimp completed transi- 

 tion at a younger age than that of 

 1971. Increases in natural mortality 

 occurred after the 1971 year-class 

 completed transformation to the fe- 

 male sex at age 6.4, and may indicate 

 a combination of spawning stress 

 and senescence or increasing abun- 

 dance of the predator, Pacific cod 

 Gadus macrocephalus. Dynamic pool 

 yield-per-recruit analysis indicated 

 maximum biomass of year-classes 

 was achieved at or prior to full re- 

 cruitment to the survey and commer- 

 cial fishing gear. Because mortality 

 rates are high, yield-per-recruit is op- 

 timized by harvesting more of the 

 smaller male shrimp. 



Age, Growth, and Mortality 

 of the Northern Shrimp 

 Pandalus borealis Kroyer 

 in Pavlof Bay, Alaska 



Paul J. Anderson 



Kodiak Laboratory. Alaska Fisheries Science Center 

 National Marine Fisheries Service, NOAA 

 P.O. Box 1638, Kodiak, Alaska 99615-1638 



Manuscript accepted 5 July 1991. 

 Fishery Bulletin, U.S. 89:541-553 (1991). 



The Alaska Peninsula region of the 

 western Gulf of Alaska was the site 

 of one of the world's major shrimp 

 fisheries in the 1970s (Anderson and 

 Gaffney 1977). A significant decline 

 in shrimp abundance starting in 1978 

 lead to the closure of most areas to 

 fishing in 1979, and to date the fish- 

 ery has yet to reopen. The cause of 

 this decline and the failure of stocks 

 to recover to fishable levels are ob- 

 scured by a lack of information on 

 the population dynamics of pandalid 

 shrimps in the western Gulf of Alas- 

 ka. In an effort to fill this informa- 

 tion gap, I have estimated the age, 

 growth, and mortality of two year- 

 classes from a stock of northern 

 shrimp Pandalus borealis Kroyer, 

 the major commercial species in the 

 region. 



The Pavlof Bay (Fig. 1) stock of P. 

 borealis was chosen for this study be- 

 cause it has supported a major fish- 

 ery and an extensive database was 

 available. The Bay was a major pro- 

 ducer accounting for over 13,000 

 metric tons (t) of pandalid shrimp 

 landings in 1977 alone. Pandalus 

 borealis made up 11-97% of land- 

 ings, with variable quantities of P. 

 goniurus, P. hypsinotus, and Panda- 

 lopsis dispar making up the remain- 

 der. Pavlof Bay shrimp have been 

 surveyed annually by the National 

 Marine Fisheries Service (NMFS) 

 since 1972. Data collected from these 

 surveys include weight, number, and 

 carapace length-frequencies by sex 

 for P. borealis. The Alaska Depart- 



ment of Fish and Game (ADF&G) has 

 collected species and size-composition 

 data from commercial landings since 

 the beginning of the fishery in 1968.* 



Pandalus borealis in Pavlof Bay 

 are managed as a discrete stock by 

 the ADF&G. Analysis of data from 

 commercial catches and shrimp sur- 

 veys conducted by NMFS and ADF&G 

 suggest that migration between ad- 

 jacent bays along the Alaska Penin- 

 sula is minimal (Jackson et al. 1983). 

 Vertical migrations in concert with 

 ocean currents may act as dispersal 

 mechanisms for pandalids (Barr 

 1970, Pearcy 1970, Gotshall 1972). In 

 Pavlof Bay, effects of ocean currents 

 probably are minimized by barrier 

 islands and shallow entrances that 

 favor confinement and limit immigra- 

 tion (Fig. 1). 



Since there are no known anatom- 

 ical structures for ageing P. borealis, 

 researchers have used length-fre- 

 quency analysis to estimate age and 

 growth. For example, Rasmussen 

 (1953) used dominant year-class 

 modes to interpret growth and sex 

 transformation of P. borealis in Nor- 

 wegian stocks, and Skuladottir (1981) 

 used the positive deviations from 

 long-term length-frequency distribu- 

 tions to identify average lengths-at- 

 age for P. borealis from Icelandic 

 fjords. 



* Commercial landings, species, and size com- 

 position data available from Alaska Depart- 

 ment of Fish and Game, 211 Mission Road, 

 Kodiak, AK 99615. 



541 



