Ralston and Tagami: Exploitable biomass of Heterocarpus laevigatas in Hawaiian Is , Part 



499 



Depth (m) 



400 



500 



600 



700 



BOO 



goo 



Kauai 

 KSl Niihau 

 ^M Average 



200 250 300 350 400 450 500 

 Depth Interval (fathom) 



Figure 5 



Distribution of habitat by depth at Kauai and Niihau. The com- 

 posite average distribution is based on poohng habitat at these 

 two sites. 



400 



Depth (m) 

 500 BOO 700 



eoo 



900 



Niihau 



- ■ — ■ Kauai 

 Hawaii 



MLKM 



Oahu 



200 250 300 350 400 450 

 Depth Interval (fathom) 



500 



Figure 6 



Average CPUE oi Heterocarpiis laengatus relative to depth 

 at each site sampled. 



trap-night of fishing. Also presented is the ordinary 

 least-squares regression equation relating these 

 variables. The equation of the line is 



CPUEj = 22.84 - 0.007988[K,], 



with standard errors of the slope and intercept equal 

 to 0.001997 and 1.9368, respectively. The regression 

 is highly significant (Fi,sy= 16.00, P 0.0001). The re- 

 siduals show no obvious departure from linearity, an 

 indication of constant catchability. 



Under the Leslie model, the exploitable biomass 

 at the start of the experiment is defined by the x- 

 intercept, i.e., 2859kg. Because the study site covered 

 1187 ha, this amounts to an initial density of 2.4087 

 kg/ha, which produced an initial catch rate (CPUEo) 

 equal to 22.84 kg/trap-night (i.e., the y-intercept). Then, 

 q expressed on a hectare basis, rather than defined in 

 terms of the study site, is estimated to be 9.4798ha/ 

 trap-night. In real terms, one overnight soak of a 

 single, large pyramidal shrimp trap is estimated to have 

 captured ~0.8% of the shrimp in the entire study area, 

 which is equivalent to all the shrimp in ~9.5ha, were 

 they randomly dispersed. The standard error of q/ha 

 is the product of A (1187 ha) and the standard error 

 of q measured over the study site (0.001997) (Seber 

 1982). 



It is instructive to note that an initial density of 

 2.4087 kg/ha is equivalent to an average of 73 shrimp/ 

 ha (see statistics for mean shrimp weight above). Based 

 on this density, the average utilization of habitat by 

 each shrimp was 137 m^, a remarkable figure given 



the relatively high catch rate encountered at the begin- 

 ning of the depletion experiment (22.83 kg/trap-night). 



Depth-stratified sampling 



Habitat areas by 92m (50fm) depth intervals were 

 estimated for the Oahu, Hawaii, and MLKM sites by 

 assuming that the proportionate distribution of habitat 

 between 366 and 915 m (200-500 fm) at these sites was 

 the same as the composite distribution obtained at 

 Kauai and Niihau. While there were some differences 

 in the distribution of habitat with depth between Kauai 

 and Niihau (Fig. 5, Table 2), they were relatively minor. 

 At both sites the amount of habitat in the 458-549 m 

 (250-300 fm) depth interval was slightly less than in the 

 366-458 m (200-250 fm) interval; but with increasing 

 depth below that, the amount of habitat per 92 m (50 fm) 

 depth interval increased steadily. 



There were, however, marked differences in CPUE 

 with depth among the five localities sampled (Fig. 6, 

 Table 2). Catch rates at Niihau were particularly high 

 relative to the other areas, especially at 458-549 m 

 (250-300 fm). Catch rates at Oahu and MLKM were 

 much lower. The modes of the distributions at Kauai 

 and Niihau were shifted to the shallow end of the depth 

 range, whereas at Hawaii it was shifted deeper. 



Results presented in Table 2 provide estimates of the 

 exploitable biomass (B), as well as variance estimates, 

 for the depth intervals sampled at each site. Although 

 catch rates at Niihau are quite high, the reduced 

 amount of habitat at this island (69,530ha) is sufficient 

 to support only a small stock of shrimp (35.7 MT). Oahu, 

 with its much lower catch rates, has a larger exploit- 



