Anderson: Age, growth, and mortality of Pandalus borealis Kroyer 



55! 



absence of harvest, would be achieved at a relatively 

 early age (Fig. 7) and small size. The maxima of bio- 

 mass vectors were attained between age 2.4 (14.3 mm 

 CL) and 3.4 (16.5mm CL) for the 1971 year-class, and 

 at age 3.4 (18.2mm CL) for the 1975 year-class (Fig. 

 7). A management strategy based on maximizing the 

 yield-per-recruit for the 1971 year-class would have 

 resulted in the harvest of mostly male shrimp prior to 

 their transformation to females. A similar finding was 

 also reached by Abramson and Tomlinson (1972) in 

 yield studies of Pandalus jordani. Harvesting more 

 young shrimp may also lead to a possible lower eco- 

 nomic return due to market resistance to small meats. 

 Optimum management of Pandalid shrimp fisheries 

 probably involves a trade-off between size-related 

 economic return and larger yields from the harvest of 

 young shrimp. 



Conclusions 



1 The bathymetric features of the Pavlof Bay region 

 make it an ideal area for studying populations of 

 P. borealis. The confining sills probably allow little im- 

 migration or emigration of shrimp. Thus dominant size 

 modes, representing year-classes, could be followed 

 through time in both the research survey and commer- 

 cial data sets. It is necessary to rely on following 

 dominant size modes to gain insight into population 

 parameters owing to the problem of overlap with less 

 dominant year-classes in the size-modal structure. 



2 The Pavlof Bay population of P. borealis showed 

 significant differences in year-class strength, with the 

 1971 year-class predominating during the study. Domi- 

 nant year-classes are first detected when they are 

 young (1.4 years old). Biotic or abiotic factors that 

 control year-class strength have their greatest effect 

 during the larval and juvenile stages. 



3 There is evidence that growth may be inversely 

 related to year-class strength. Since 1979, 1.4 year-old 

 shrimp have averaged 13-15 mm CL rather than the 

 10-11 mm CL observed prior to 1979. 



4 Age at sex transition was highly variable for the 

 two dominant year-classes studied. The 1971 year-class 

 showed transition over three years, while the 1975 

 year-class completed transformation in one year. 



5 Total mortality rates of Pavlof Bay P. borealis are 

 some of the highest reported for the species (Frechette 

 and Parsons 1983, Teigsmark 1983, Hopkins and 

 Nilssen 1990) and are mostly a reflection of the high 

 natural mortality rates. High natural mortality rates 



for the 1971 year-class are attributed to intense preda- 

 tion by Pacific cod (Albers and Anderson 1985). After 

 the cod population subsided within the Bay, natural 

 mortality, along with total mortality, declined. 



6 Yield could have been maximized by harvesting 

 more young (male) shrimp, since mortality rates were 

 high for the year-classes studied. 



Worldwide, other P. borealis fisheries have ex- 

 perienced similar cycles of high and low abundance 

 (Balsiger 1981) that characterized the rise and fall of 

 western Gulf of Alaska fisheries in the 1970s. The 

 decline of shrimp in Pavlof Bay was probably inevi- 

 table, regardless of the presence of a fishery, since 

 many adjacent areas lightly or seldom fished ex- 

 perienced similar population declines (Anderson and 

 Gaffney 1977). These declines may be directly attrib- 

 utable to the demise of a single or series of strong year- 

 classes. Little is known about the parameters control- 

 ling shrimp natality and its relationship to subsequent 

 recruitment. This problem deserves study, along with 

 the effects of predation (Albers and Anderson 1985) 

 and ecological and environmental parameters (Nunes 

 1984). Continuing shrimp research surveys in Pavlof 

 Bay may lead to understanding the dynamics and 

 perhaps the mechanisms driving the cycle of low and 

 high shrimp abundance in the western Gulf of Alaska. 



Acknowledgments 



Thanks to P. Holmes and J. Nichols from the Alaska 

 Department of Fish and Game for providing commer- 

 cial catch sampling data. D. Somerton, R.S. Otto, 

 D. Kimura, and an anonymous reviewer provided many 

 helpful suggestions in the analysis of shrimp length- 

 frequency data and critically reviewed the manuscript. 

 Thanks also to A. Lindsay from NORFISH Center for 

 Quantitative Science at the University of Washington 

 for his assistance in computer programming. Finally, 

 I owe a deep debt of thanks to the Captains as well as 

 the crew and scientists aboard the R/Vs Oregon, Chap- 

 man, John N. Cobb, and Alaska for their efforts in data 

 collection. 



Citations 



Abramson, N.J. 



1971 Computer programs for fish stock assessment. FAO 

 Fish. Tech. Pap. 101, Rome, 140 p. 



Abramson, N.J., and P.K. Tomlinson 



1972 An application of yield models to a California ocean 

 shrimp population. Fish. Bull, U.S. 70:1021-1041. 



