Archer et al.: Estimates of the incidental kill of Stene/la attenuata attenuate/ calves in the tuna purse-seine fishery 



237 



Fraction of females lactating, 

 by color (1973-78, 1981): Flac 



Number of lactating 



females killed: N h „ 



Tally of females, 

 by color: nj ' 



Tally of dolphins killed, by color and 

 sex, from set log ( 1973-90. 1996-2000) 



Tally of dolphins, 

 by color: n\ 



Fraction of dolphins in age class, 

 by color (1973-78, 1981): F at 



Number of dolphins 

 killed, by age: N a 



Probability of suckling, 

 by age (1989-91): P(milk) a 



Calf deficit: D 



Number of suckling 

 calves killed: N ca y 



Figure 1 



Diagram of the analytical method used to estimate the spotted dolphin (Stenella attenuata attenuata) 

 calf deficit in each set as described in the text. Boxes identify original data that were bootstrapped to 

 produce confidence intervals. Values in parentheses are years for which data were available. 



In a subset of the sets that we examined, every indi- 

 vidual killed had been examined and biological samples 

 had been collected from it; therefore, we knew the actual 

 number of lactating females killed. There were 1108 of 

 these "100% sampled" sets on the northeastern stock, and 

 697 on the western-southern stock from 1973 to 1990. We 

 evaluated the accuracy of our frequency-based method 

 by conducting a paired /-test between our estimate of the 

 number of lactating females and the number observed in 

 each of these sets. 



Stomach-content data were not available for every 

 animal in these 100% -sampled sets; therefore, we did not 

 know the actual number of suckling calves. However, we 

 also used paired /-tests to compare our estimate of the 

 number of suckling calves in each set with the number of 

 animals smaller than 122 cm, which was the estimated 

 length at which the probability of milk in the stomach 

 was 0.5, given the weaning model of Archer and Robertson 

 (in press). Likewise, our estimate of the calf deficit was 

 compared with the deficit as estimated by using a cutoff 

 length of 122 cm. These tests were done to determine if the 

 method in the present study would produce significantly 

 different results from the method used in the previous 

 study Paired /-tests were conducted for each year sepa- 

 rately, as well as for all years combined. A power analysis 

 was also performed for these paired /-tests to determine 

 the minimum detectable difference at which we could re- 

 ject the null hypothesis of no difference between methods 

 given observed sample sizes and variability. 



Results 



The calf deficit as a fraction of the number of dolphins 

 killed (D k ) increased slightly during the mid-1970s but 

 remained relatively constant throughout the rest of the 

 time series at approximately 0.14 missing calves per dol- 

 phin killed for both stocks (Fig. 2). The total calf deficit 

 (D total ) as estimated from the annual kill decreased from 

 highs of approximately 5000 in the mid-1970s down to 

 2000-3000 by the early 1980s (Fig. 3). In the late 1980s, 

 this value increased to approximately 5000 in northeast- 

 ern spotted dolphins (Table 3A) and approximately 2800 

 in the western-southern stock (Table 3B), reflecting an 

 increase in the reported kills. In the last five years of the 

 time series (1996-2000), the estimated total deficit was 

 approximately 60 missing calves. 



The mean deficit per set (Z),) for northeastern spotters 

 over all years was 1.03 missing calves per set, and the me- 

 dian was 0.30 (Fig. 4). For western-southern spotted dol- 

 phins, the mean was 1.28 missing calves per set, and the 

 median was 0.33. The estimated mean deficit per set was 

 approximately 1.5 in the mid-1970s and decreased over 

 time to 0.01-0.02 at the end of the time series (Fig. 4). For 

 both stocks, 75- 95% of lactating females killed were not 

 killed with their calf (Fig. 5). 



In the sets that were 100% -sampled, for all years com- 

 bined, there was no significant difference between the 

 observed and the estimated number of lactating females 

 killed in either stock (Table 4). The results of paired /-tests 



