600 



Fishery Bulletin 89(4), 1991 



(0.1987/year for a reporting 

 rate of 1.0) is less than half the 

 estimate obtained when the 

 decreasing shedding-rate 

 model was used (0.4038/year 

 for a reporting rate of 1.0). 



SE model 



The SE model was fit to the 

 tag-return data from experi- 

 ments 2 and 3 (analysis A) and 

 experiments 2 and 4 (analysis 

 B). These data, along with the 

 catch data used to parameter- 

 ize Fs, are shown in Tables 4 

 and 5, respectively. For anal- 

 ysis A, returns from the NSW 

 fishery (fishery 1) were de- 

 fined by recapture positions 

 east of 145°E. SA fishery 

 (fishery 2) returns were de- 

 fined by recapture positions 

 west of 145°E. No returns 

 from these experiments were 

 recorded in WA (defined for 

 these purposes as west of 

 125°E) as commercial tuna 

 fishing in WA did not begin 

 until 1969. For analysis B, 

 there is a slight complicating 

 factor in that southern bluefin 

 were tagged in WA before 

 substantial commercial fishing 

 began in 1969; the majority of 

 the few returns recorded in 

 WA from experiment 4 were 

 from the tagging vessel. How- 

 ever, because the tagged fish 

 were released in the WA area, 

 fishery 2 for analysis B is de- 

 fined as the SA and WA areas 

 combined. For both analyses, 

 fishery 3 is the Japanese long- 

 line fishery with no geograph- 

 ical restrictions. 



As mentioned earlier, the 

 use of catch data to param- 

 eterize F involves an assump- 

 tion that the population is in 

 equilibrium for the duration of 

 the tag-recovery period. This 

 is not an unreasonable assump- 

 tion for the juvenile popula- 

 tion available to the surface 



