A MULTISPECIES ANALYSIS OF THE 

 COMMERCIAL DEEP-SEA HANDLINE FISHERY IN HAWAII 



Stephen Ralston 1 and Jeffrey J. Polovina 2 



ABSTRACT 



In the Hawaiian Islands 13 species of bottom fish are commonly harvested in the commercial deep- 

 sea handline fishery. These are all high-level carnivores, including snappers, jacks, and a species of 

 grouper, which are sought in water depths ranging from 60 to 350 m. Cluster analyses performed on 

 the Hawaii Division of Fish and Game commercial catch report data suggest the existence of three 

 bottom fish species groups which apparently segregate on the basis of depth distribution. These 

 groups seem to be stable through time and similar among differing geographic localities. 



Two measures of fishing effort, catch-records and fisherman-days, were compared to determine 

 which is more suitable for use in stock-production analyses. Fisherman-days was selected because, 

 among other reasons, it repeatedly demonstrates a stronger negative correlation with catch per unit 

 effort. 



Application of the Schaefer stock-production model to this multispecies fishery on a species-by- 

 species basis provides an inadequate description of productivity. When catch statistics are 

 aggregated according to the three cluster analysis species groups the results are much improved. In 

 this regard consistently significant results and production estimates were obtained from the Maui- 

 Lanai-Kahoolawe-Molokai bank, a region which presently accounts for about half of the total 

 Hawaii catch. No significant interaction among the cluster groups was detected. When all 13 bottom 

 fish species are analyzed together, the results are in agreement with the preceding analysis. 

 Examining the aggregation process suggests that the model based on the intermediate level of 

 aggregation (cluster groups) explains slightly more of the variation in total catch than does the 

 model which treats all 13 species together. 



We estimate the annual maximum sustainable yield of the commercial deep-sea handline fishery 

 around the Maui-Lanai-Kahoolawe-Molokai bank to be 106 metric tons or about 272 kg/nmi of 100- 

 fathom isobath. Because recreational catch is unaccounted for these figures are considered lower 

 bounds for the gross production obtainable from this type of fishery although currently the 

 commercial fishery is operating close to this maximum-sustainable-yield level. 



Effective management programs for tropical 

 fisheries are difficult to achieve (Pauly 1979). 

 Often attempts at managing these fisheries are 

 based on the application of inappropriate models 

 to sparse data. Both deficiencies are due in part 

 to the multiplicity of fish species inhabiting 

 tropical environments. This great diversity (Sale 

 1977; Talbot et al. 1979) makes it difficult to 

 compile adequate data for all species of interest. 

 The Hawaiian Islands, which straddle the 

 Tropic of Cancer, possess a relatively impover- 

 ished tropical ichthyofauna, yet between 600 and 

 700 species are known from this region (Gosline 

 and Brock 1960). Coupled with high diversity, 



'Fisheries Research Institute, College of Fisheries, 

 University of Washington, Seattle, Wash.; present address: 

 Southwest Fisheries Center Honolulu Laboratory, National 

 Marine Fisheries Service, NOAA, P.O. Box 3830, Honolulu, HI 

 96812. 



2 Southwest Fisheries Center Honolulu Laboratory, National 

 Marine Fisheries Service, NOAA, P.O. Box 3830, Honolulu, HI 

 96812. 



Manuscript accepted January 1982. 

 FISHERY BULLETIN: VOL. 80, NO. 3, 1982. 



many tropical countries lack a refined statistical 

 system for the acquisition and storage of 

 fisheries data. In concert these two limitations 

 impose severe restrictions on the quantity and 

 quality of data which are currently available for 

 the analysis and management of tropical 

 fisheries (Pope 1979). Furthermore, classical 

 fisheries models thus far developed have been 

 directed toward the management of temperate 

 and boreal stocks (Food and Agriculture Organ- 

 ization of the United Nations (FAO) 1978). These 

 models usually treat species as independent 

 management units. It has become apparent that 

 such an approach is often inadequate when ex- 

 trapolated to the tropics where community 

 dynamics become increasingly important (Pauly 

 1979). 



The multispecies approach to managing 

 fisheries exploitation in complex ecosystems has 

 only recently acquired a substantive base in the 

 literature. Early work by Larkin (1963, 1966) 



435 



