Hampton Natural mortality and movement rates of Thunnus maccoyu 



601 



fisheries as long as recruitment was reasonably con- 

 stant during this period. The stability of the surface 

 catches during the 1960s and early 1970s and cohort 

 analysis (Hampton 1989) would suggest that this was 

 the case. Although the adult population declined during 

 several periods since exploitation began, Japanese 

 catch was fairly constant during these tagging experi- 

 ments, and parental biomass also appeared to be rela- 

 tively stable. Therefore, the use of catch data to param- 

 eterize F should not cause major difficulties in this case. 

 In these experiments, tagged southern bluefin were 

 released over a period of years; therefore, returns 

 within a specific time-at-liberty category cannot be 

 related to a catch in any one year. In these cases, the 

 catch (or effort) data used are normally averaged over 

 time, assuming a constant F within each fishery (e.g., 

 Kleiber et al. 1987). This was the approach taken for 

 the Japanese fishery, where catch was in fact quite con- 

 stant over most of the return period (mid-1960s to early 

 1980s). For the Australian surface fisheries, an 8-year 

 moving average (equivalent to the total release period 

 for experiments 2, 3, and 4) was used so that the 

 gradual decline of the NSW fishery and gradual in- 



crease in the SA and WA fisheries (Hampton 1989) 

 could be represented. 



Analysis A The first fit to the data allowed M to vary 

 among the three fisheries. The resulting estimates of 

 the three natural mortality parameters, three catch- 

 ability parameters and four movement parameters and 

 their standard errors, for reporting rates between 1.0 

 and 0.5, are presented in Table 6. There is no a priori 

 reason why reporting rate should be the same for the 

 three fisheries. In the absence of any information on 

 the actual reporting rates, they have been assumed 

 here, for simplicity, to be the same for the three 

 fisheries. 



For a reporting rate of 1.0, the estimates of M for 

 the three fisheries are 0.71/year, 1.5/year, and 0.20/ 

 year, respectively. M 2 and M 3 decrease slightly as 

 reporting rate decreases; however, Mj decreases 

 substantially with decreasing reporting rate. This is 

 because the tag recapture rate in NSW is very high 

 even if a reporting rate of 1.0 is assumed; therefore, 

 even relatively small reductions in reporting rate from 

 1.0 require substantial compensatory changes in Mj . 



