SMITH: SIZE CHANGES OF THREE DOLPHIN POPULATIONS 



lower pregnancy rate that probably occurs in 

 older animals. New methods are being developed 

 for age determination, and an effort is being 

 made to apply these methods to age the samples 

 of dolphins. With accurate data on age of ani- 

 mals, a more detailed examination of sampling 

 biases will be undertaken. 



Natural Mortality Rates 



No direct estimates of natural mortality rates 

 exist for the eastern Pacific dolphin populations, 

 as might be obtained from tagging data or from a 

 sampled age structure. Ohsumi (1979) presented 

 a statistical relationship between natural mor- 

 tality rate and body length for cetaceans, from 

 which can be derived an annual, natural mortal- 

 ity rate of around 0.14 for the eastern Pacific 

 dolphin populations. However, this estimate is 

 obtained by extrapolating the relationship out- 

 side the range of his data, and consequently is 

 unreliable. 



Another method of estimating natural mortal- 

 ity rate is from information on gross reproduc- 

 tive rate for a population in equilibrium with its 

 environment, assuming natural mortality does 

 not change with population size. This approach 

 was used in the 1976 workshop (footnote 2). An 

 estimate of gross reproductive rate of 0.09 (Ka- 

 suya et al. 1974) for a population off Japan, 

 thought to be lightly exploited, was used as the 

 natural mortality rate estimate for the eastern 

 tropical Pacific populations. It now appears that 

 the population off Japan had, in fact, been ex- 

 ploited to a greater degree than was thought, and 

 that there is segregation of prepubertal dolphins 

 into separate schools (footnote 3, p. 41). The as- 

 sumption, consequently, of a natural mortality 

 rate of 0.09 is probably not valid. 



In the 1979 workshop (footnote 3), estimates of 

 the gross reproductive rate of lightly exploited 

 Southern Hemisphere populations of spotted and 

 spinner dolphins in the eastern tropical Pacific 

 were used as estimates of natural mortality 

 rates. These rates were 0.098 and 0.067 for spot- 

 ted and spinner dolphins, respectively. 



Net Rates 



Net recruitment rates for the offshore spotted, 

 eastern spinner, and whitebelly spinner dolphin 

 populations can be estimated as the differences 

 between the gross reproductive rate estimates, 

 listed in Table 5, and the corresponding natural 



mortality rate estimates given above. Using 

 method I estimates of pregnancy rates, one ob- 

 tains estimated net reproductive rates of 0.021, 

 0.008, and 0.029 for these three populations, re- 

 spectively. Using method II estimates of preg- 

 nancy rates, one obtains estimates of 0.002, 0.032, 

 and 0.021, respectively. These highly variable 

 estimates are unsatisfactory, because they are 

 based on data with known sample biases, and 

 they differ among populations in unexpected 

 ways. In particular, it is not expected that the net 

 reproductive rate of the whitebelly spinner dol- 

 phin, which has been relatively less exploited, 

 should be higher than that of the more heavily 

 exploited eastern spinner dolphin popula- 

 tion. 



Due to these uncertainties, specific point esti- 

 mates were not obtained by the 1979 workshop 

 participants. Rather, a range of values from 0.0 

 to 0.04 were considered equally likely, given the 

 available information. The lower value of 0.0 was 

 selected by the 1979 workshop to reflect uncer- 

 tainties about unexpected changes in some repro- 

 ductive rates, and the small magnitude of the 

 estimates of net reproductive rates. This range 

 compares with the estimates from the 1976 work- 

 shop of 0.02-0.06, with a midpoint estimate of 

 0.04. Although higher rates of increase of ceta- 

 cean populations have been reported, contrary to 

 the conclusions in the 1979 workshop report, 

 there are no reliable estimates of rates of increase 

 for dolphin populations which can be used with 

 confidence. Pending better information, the 

 range of estimates considered in the two work- 

 shops will be used here, recognizing that higher 

 rates may be possible. 



Rate Dependent on Population Size 



The evidence on which to base an estimate of 

 the value of Z in Equation (3) for dolphin popula- 

 tions is limited. Fowler (1981) argued that for 

 large, long-lived mammals, Z is greater than 

 unity. He based this conclusion on a review of 

 empirical data, primarily from terrestrial popu- 

 lations, and on an analysis of the demographic 

 constraints which come with long life and ex- 

 tended parental care. McCullough (footnote 3, p. 

 8) gave preliminary estimates of maximum net 

 productivity level (MNPL), and hence Z, for four 

 large terrestrial mammal populations. His esti- 

 mates agree with Fowler's conclusions that Z is 

 greater than unity, and that later reproducing 

 animals would have higher values of Z. 



