550 



Fishery Bulletin 89(4). 199! 



Figure 6 



Spawner/recruit curve fit on female Pandalus 

 borealis caught per kilometer as index of spawn- 

 ing biomass, and 1.4 year-olds (lagged one year) 

 as recruit index, captured in Pavlof Bay shrimp 

 surveys 1972-86. Points labeled A and B show 

 effect of strong 1.4-age male year-classes of 1973 

 and 1975, respectively. 



transformation of shrimp has been 

 suggested in this study, the rela- 

 tionship between stock and recruit- 

 ment appears to elude suitable 

 definition. Using the average catch 

 of females per tow as an index of 

 spawning population and the catch 

 rate of 1.4 year-old males as an in- 

 dex of recruits (lagged one year), a 

 Ricker (1975) stock-to-recruit rela- 

 tionship was constructed (Fig. 6). 

 While other combinations representing spawning and 

 recruit biomass were tried, this combination gave the 

 best fit to the recruitment function. Possible reasons 

 for this combination rendering the best fit are that 

 females are almost fully vulnerable to trawl capture 

 and remain near the bottom during the fall survey 

 period. On the other hand, while the 1.4 year-old males 

 are not fully vulnerable to trawl capture, they do seem 

 to represent a good relative index of incoming year- 

 class strength. When dominant year-classes could be 

 identified and followed through time, they were first 

 identified as being relatively strong as 1.4 year-olds 

 (Fig. 2). Since the 1.4 year-old mode is usually well 

 separated from the remaining frequency distribution, 

 weak as well as strong incoming year-classes could be 

 quantified with the mode-separation technique. 



Extreme fluctuations in the abundance of young 

 year-classes make it difficult to devise management 

 strategies. In some years, a relatively small spawning 

 stock may give rise to a large year-class. The fit to the 

 stock recruitment curve (Fig. 6) shows that the major- 

 ity of the data points mostly conform to the fitted 

 relationship. Outlying points labeled as A and B (A, 

 1973 year-class males (age 1.4) vs. brood females of 

 1972; and B, 1975 year-class males (age 1.4) vs. brood 

 females of 1974) illustrate how relatively dominant 1.4 

 year-old male year-classes can push data points well 

 outside the bounds of the curve (Fig. 6). The relative 

 abundance of 1.4 year-old shrimp is probably a func- 

 tion of the variable survival of larvae and juveniles in 

 response to environmental conditions rather than 

 spawning biomass. The inadequacy of the fitted stock 

 to recruit function to describe these important con- 

 tributions to stock biomass means that other methods 











Age 



Figure 7 



Biomass vector (in the absence of fishing) calculated using the 

 technique of Paulik and Bayliff (1967) for the 1971 and 1975 

 year-classes. 



for defining management are needed. Therefore, 

 strategies based on yield-per-recruit of dominant year- 

 classes were examined. 



Incorporating the growth and mortality estimates 

 from this study into a Ricker yield model (Paulik and 

 Bayliff 1967) indicated that maximum biomass, in the 



