Hampton: Natural mortality and movement rates of Thunnus maccoyii 



609 



noting that the high recovery rate of tags released 

 in NSW would suggest a high reporting rate in that 

 fishery at least. 



The few data available suggest that mortality asso- 

 ciated with tagging southern bluefin tuna is slight. 

 Animals have been reported to survive in good condi- 

 tion for 80 hours in a live bait tank after having been 

 tagged (Robins 1963). Furthermore, Hynd and Lucas 

 (1974) considered that the behavior of the fish imme- 

 diately following tagging was consistent with slight or 

 no tagging mortality. Personal observations in the field 

 also support this hypothesis. If tagging mortality oc- 

 curred and was immediate or nearly so, its effect on 

 the parameter estimates would be slight and identical 

 to that of non-reporting. If there was significant, con- 

 tinuous tag-induced mortality (which is unlikely), M 

 would tend to be overestimated by both methods used 

 in this paper. 



The assumption of constant M has not been tested 

 in this paper, but the possibility that a model with age- 

 dependent M might provide a better fit to the tagging 

 data should not be ruled out. Generalization of the SE 

 model in this regard would be possible and may yield 

 useful results, although similar estimation problems to 

 those of the unconstrained-M fits might occur. 



The final assumption concerns equal vulnerability of 

 the tagged and untagged populations and is possibly 

 the most difficult with which to comply. Strict compli- 

 ance would require that either the fishing effort is dis- 

 tributed randomly with respect to both the tagged and 

 untagged populations, or that the tagged population 

 is distributed randomly with respect to the untagged 

 population. The first possibility is seldom seen in prac- 

 tice because fishermen tend to direct their effort in 

 areas of high fish concentration. If these areas happen 

 to coincide with the areas of tag release (as is often the 

 case), the number of recaptures during the first period 

 after release may be larger than expected. This would 

 result in overestimates of M and other components of 

 tag attrition using the SE method. Similarly, the sec- 

 ond possibility is not usually feasible, as it would re- 

 quire tagged fish to be released over a wide area within 

 a short space of time. However, for highly mobile tunas 

 such as southern bluefin, mixing is likely to be rapid, 

 thus lessening the problem to a large extent. In the case 

 of the NSW fishery, many tagged fish were released 

 prior to the fishing season, which would tend to en- 

 hance mixing. Division of the Australian fishery into 

 two components also helps compliance with this as- 

 sumption. From this point of view, even further strati- 

 fication would have been desirable; however, this would 

 greatly complicate the algebra and probably result in 

 statistical problems with parameter estimation. 



The fishing mortality inflicted on tagged southern 

 bluefin by the Japanese fishery would appear to be 



slight (<0.06/year for assumed reporting rates >0.5) 

 compared with values of 0.15-0.40/year obtained from 

 cohort analysis (Hampton 1989). This discrepancy could 

 result because southern bluefin migrating through 

 Australian coastal waters are subsequently somewhat 

 less available to the longline fishery than the popula- 

 tion in general. Such differential availability to surface 

 and longline fisheries has been noted for other tuna 

 stocks, e.g., populations of yellowfin tuna in the Pacific 

 (Lenarz and Zweifel 1979, Suzuki 1988) and Atlantic 

 Oceans (Fonteneau 1986, Suzuki 1988). While not 

 necessarily affecting the estimates of M, caution should 

 be exercised in the interpretation of estimates of fish- 

 ing mortality derived from these tagging experiments. 

 The results of estimations using the SE method were 

 consistent in suggesting a value of M, in the Japanese 

 fishery at least, of approximately 0.2/year. There was 

 little information in the data regarding M in the 

 Australian surface fisheries; however, a value similar 

 to the above is not an unreasonable, if tentative, con- 

 clusion. Analysis of more recent tagging experiments 

 may, in due course, help to resolve this question. 



Acknowledgments 



This work formed part of a Ph.D. thesis undertaken 

 at the University of New South Wales, Kensington, 

 Australia. The work benefited through numerous dis- 

 cussions with Drs. Geoffrey P. Kirkwood and William 

 S. Hearn of the CSIRO Division of Fisheries, Hobart, 

 Australia. Two anonymous referees reviewed the 

 manuscript and provided valuable suggestions. 



Citations 



Baranov, T.I. 



1918 On the question of the biological basis of fisheries. 

 Nauchn. Issled. Ikhtiol. Inst. Izv. 1:81-128 [in Russ.]. 

 Bard, Y. 



1974 Nonlinear parameter estimation. Academic Press, NY, 

 341 p. 

 Beverton, R.J H., and S.J. Holt 



1957 On the dynamics of exploited fish populations. Fish. In- 

 vest. Ser. II. Mar. Fish. G. B. Minist. Agric. Fish. Food 19. 

 533 p. 

 Caton, A.E., K. McLoughlin. and M.J. Williams 



1990 Southern bluefin tuna: Scientific background to the 

 debate. Department of Primary Industries and Energy, 

 Bureau of Rural Resources Bull. 3, Canberra, Australia, 

 41 p. 

 Courant, R. 



1937 Differential and integral calculus, 2d ed., vol. 1. Blackie 

 and Son Ltd., London, 616 p. 

 Cox, D.R., and D.V. Hinkley 



1974 Theoretical statistics. Chapman and Hall, London, 511 p. 



