Wetherbee: Assemblage of deep-sea sharks on Chatham Rise, New Zealand 



191 



Consideration of sharks as an assemblage, which 

 is separate from the rest of Chatham Rise commu- 

 nity, is an artificial division. However, because the 

 primary interest of this study was to describe the 

 abundance and distribution of the sharks on Cha- 

 tham Rise, several ecological indices were employed 

 to compare different locations, depths, and species 

 of shark. Abundance was expressed as density (kg 

 shark/km- ) and was calculated for each species within 

 each trawl based on the total weight of sharks caught, 

 net width, towing speed, and trawl duration. Three 

 features of distribution were examined for sharks: 

 diversity, similarity, and randomness. Diversity was 

 expressed as the number of species of sharks per 

 trawl (Stephens et al. 1984; Garcia et al. 1998). The 

 Bray-Curtis similarity index was used for compari- 

 sons among the ten areas on Chatham Rise, depth 

 intervals, and between the two surveys: 



S=l 



IK->^u|/X<i^.,+^* 



where Yj^ = score for /'^ species in the j^^ sample; 

 and 

 y,^ = score for the /* species in the k^^ sample 

 (Field et al. 1982; Sedberry and Van 

 Dolah 1984). 



This index ranges from (no species in common) to 

 1 (identical species in each sample). Morisitas index 

 of dispersion (/^) was calculated for each species 

 of shark as an indicator of the randomness of their 

 distributions: 



Ia=n{^X--N/N(N-l)\, 



801 



E 

 Z 



1990(281) 

 1993(109) 



Deplh (m) 



m 



o 

 o 

 o 



Time 



Figure 2 



Number of trawls in a orange roughy survey (1990) and 

 in a smooth oreo survey (1993) on Chatham Rise, New 

 Zealand, at various intervals of (A) depth and (B) time 

 of day. Numbers in parentheses are total trawls for each 

 survey. 



where n = number of plots; 



A'^ = total number of individuals counted in 

 all n plots; and 

 IX^ = squares of the number of individuals 

 per plot summed over all plots. 



If the dispersion is random, then I^ = 1.0; if per- 

 fectly uniform, /^ = 0; and if maximally aggregated 

 (all individuals in the same trawl), 7^ - the number 

 of trawls (Brower and Zar, 1984). To minimize the 

 dominant effect of anomalous catches, each value 

 of shark density (kg/km-) was converted to ln(x +1) 

 prior to calculation of these ecological indices (Field 

 et al., 1982; Bianchi, 1991 ) and comparisons of means 

 were made by using either a two-tailed ^test or one- 

 way ANOVA. 



Results 



Seven species of shark belonging to the family Squal- 

 idae and five undescribed species of catsharks in the 

 genus Apristurus (family Scyliorhinidae) were regu- 

 larly captured in deep-water trawls (Table 1). For 

 the purpose of this study, all sharks of the genus 

 Apristurus were treated as a single group. Several 

 other species of shark were captured iChlamydose- 

 lachus anguineus, Centroscymnus coelolepis, Scymn- 

 odon squamulosus, Etmopterus lucifer), but only one 

 or two individuals of each species were captured 

 and these are not discussed further. Although the 

 composition of the catch varied throughout the day, 

 there was no apparent correlation between density 



