Kimura et al.: Tag-recapture data analysis of Pacific cod growth 



275 



x 

 i— 

 o 



>- 



o 



z 



UJ 



3 



o 



10 30 50 70 90 

 LENGTH AT RELEASE (CM) 



o 



UJ 



en 

 u 



j_ 



o 



rr 



O 



1 2 4 6 8 12 16 



TIME TO RECAPTURE (YR) 



Figure 3 



Length-frequency at time of release (top) and 

 relationship between time at liberty and growth 

 (bottom) for female sablefish in Gulf of Alaska. 



Sablefish growth differs between regions, 

 so we analyzed tag-recapture data from the 

 Gulf of Alaska and the west coast of the 

 United States separately (Fig. 1). Sablefish 

 growth differs greatly by sex (Sasaki 1985), 

 so it was necessary to estimate growth pa- 

 rameters for each sex. 



Because the type of fishing gear used for 

 tagging and recapture might affect the ob- 

 served growth increments, we include tables 

 of tagging and recapture gears for the us- 

 able data for all actual datasets (Table 2). 

 Fish in the Gulf are primarily caught using 

 longline gear, while fish off the west coast 

 are primarily caught by trawl and pot gears. 



For Gulf of Alaska sablefish, 133,558 fish 

 were released, with 7946 recoveries between 

 1972 and 1990. Of these recoveries, 1340 



had usable growth-increment information (Figs. 2,3). Longline 

 gear dominated both tag and recovery gears (Table 2). 



Estimated von Bertalanffy parameters for sablefish in the Gulf 

 of Alaska based on tag-recapture data can be compared with von 

 Bertalanffy parameters estimated directly from length-at-age data 

 (Table 3) from specimens collected on longline surveys in 1987 

 and 1989, and aged during 1990 and 1991 (Fig. 4). For this dataset 

 the unweighted Fabens' estimates, the James' estimates, and es- 

 timates based on direct ages gave very similar parameter esti- 

 mates. However, the weighted Fabens' estimates gave estimates 

 quite different from the others. 



