Perkins and Edwards: Capture rate as a function of school size for Stenella attenuata 



543 



<D 

 T3 



in 



o 



C\J 



lO 



in 



Middle 



■120 



-115 



-110 



-105 



Longitude 



Figure 1 



Geographic stock boundaries for the northeastern offshore stock of the pantropical spotted dolphin 

 iStenella attenuatai. The stock is defined by the region in the ETP north and east of 5'N and 120-W, 

 bounded at 28'N. The two strata pictured are based on those defined by Holt et al. (1987). The inshore 

 and middle strata have total areas of 4,544,000 km- and 2,019,000 km-, respectively. Points represent 

 on-effort sightings from the research vessels, 1986-90 and 1992-93. 



individual dolphin because animals may associate 

 with schools of different sizes at different times. We 

 consider implications for individual dolphins in the 

 "Discussion" section. 



Data and methods 



To quantify the tendency for purse-seiners to set on 

 large schools, we compared the relative frequency 

 with which different sizes of schools are selected by 

 fishermen for encirclement with the relative fre- 

 quency with which schools of various sizes occur 

 naturally. First, we estimated the probability distri- 

 bution of sizes for spotted dolphin schools within the 

 geographic boundaries of the northeast (NE) offshore 

 stock" (Fig. 1 ), using observations from research ves- 

 sels. This distribution models the relative number 

 of schools of each size in the study area. Then we 

 fitted a smooth probability distribution to dolphin 

 school sizes from tuna vessel sets. This distribution 

 models the relative number of times schools of each 



^ Coastal and NE Offshore stocks of the pantropical spotted dol- 

 phin exist sympatrically: however, schools are for the most part 

 distinguishable. 



size were set on. Finally, we used the ratio of the two 

 estimated density ftmctions, suitably scaled, to estimate 

 the average number of times per year a dolphin school 

 of a given size was set on. This estimated effect includes 

 not only the tendency of fishermen to preferentially 

 set on larger schools, but also any tendency to search 

 in areas where large schools may be more prevalent. 



Research vessel sighting data 



In 1986, the U.S. National Marine Fisheries Service 

 (NMFS) initiated a multiyear research program to 

 monitor trends in the abundance of dolphin popula- 

 tions in the ETP. The program used research vessels 

 to conduct line transect surveys in the ETP and 

 record sightings of all cetaceans encountered (Wade 

 and Gerrodette, 1993). The surveys were designed 

 to provide a spatially unbiased sample of the survey 

 area and to be as similar as possible across years. 

 Typically, three independent observers estimated 

 school size and species composition for each sight- 

 ing. For detailed descriptions of the survey design, 

 materials, and methods, see Holt et al. (1987), Hill 

 et al. (1991), and Mangels and Gerrodette (1994). 



The research vessel data used in this study con- 

 sisted of spotted dolphin school sightings from the 



