FISHERY BULLETIN: VOL. 80, NO. 3 



substantially greater, in spite of the fact that the 

 catchability coefficient for Group I is greater 

 than for Group III. This apparent contradiction 

 can be understood when estimates of carrying 

 capacity and instantaneous growth rate are com- 

 puted for the two groups. Ricker (1975) showed 

 that the virgin shock biomass (Boo) is equal to 

 a/q and the intrinsic rate of natural increase 

 (r) is equal to aq/b, where q is the catchability co- 

 efficient and a and b are the intercept and slope, 

 respectively, of the regression of CPUE on effort. 

 Using these equations the estimate of virgin 

 biomass for Group I at the bank is much less than 

 for Group III whereas the intrinsic rate of 

 natural increase for Group I is nearly double that 

 of Group III, hence, the disparity in catchability 

 coefficients. This manner of evaluating the 

 growth dynamics of the fishery implies that if 

 fishing were to stop abruptly, Group I would 

 recover to pristine levels much sooner than 

 either Group II or III. Thus, this analysis would 

 predict that a form of succession would occur 

 around the MLKM bank if fishing were cur- 

 tailed as a new equilibrium point was ap- 

 proached. Although there is little hope of manip- 

 ulating the system to test this particular 

 prediction of the model, this type of heuristic 

 calculation can provide valuable insights con- 

 cerning the consequences of different manage- 

 ment programs. 



Pope (1979) has shown that in a multispecies 

 fishery an increase in the colinearity of effort 

 values among species or groups will result in a 

 more parabolic-shaped yield curve. Conse- 

 quently, he argues that if fishing pressure is 

 exerted in such away that the fishing mortalities 

 of the various species remain in constant ratio to 

 one another, then the use of the TBSM is a real- 

 istic management option. He points out though, 

 that it cannot be concluded that an MSY esti- 

 mated by application of the model to actual data 

 is anywhere near the global maximum of the 

 system. These considerations bear directly on 

 this study because of the high correlations of 

 fishing effort among the three species groups. 

 Even though MSY from the MLKM bank is 

 estimated to be 106 t/yr it is quite possible that a 

 substantially larger yield could be sustained if it 

 were possible to alter the ratios of fishing 

 mortality among the species groups. This pos- 

 sibility is not unrealistic because these groups 

 seem to be for the most part spatially separated. 

 In principle then, appropriate management 

 action could reduce fishing effort on one group 



while simultaneously increasing that on another, 

 but at present it is impossible to speculate about 

 what the global MSY of the MLKM bank might 

 be. 



One of the least realistic aspects of the TBSM is 

 its inability to adequately model trophic 

 dynamics (Pauly 1979). The addition of Lotka- 

 Volterra interaction terms to the model (Pope 

 1979) is a relatively simplistic attempt to deal 

 with this problem. Pauly (1979) argued that the 

 surplus-yield of fish predator-prey systems may 

 be overestimated by the TBSM because of 

 "prudent predation" by top carnivores. This 

 theory (Slobodkin 1961) would propose that fish 

 predators optimally harvest their fish prey, 

 leaving little or no remaining latent productivity 

 of the prey species for man to utilize. These argu- 

 ments must impose group selectionist reasoning 

 and suffer as a result. Nevertheless, the TBSM 

 assumes that total stock size is greatest in a 

 virgin state, a condition which need not be satis- 

 fied if limitation is internally imposed (May etal. 

 1979). 



Fortunately these considerations do not 

 detract from the value of the present analysis. 

 The six dominant species in the fishery 

 (opakapaka, ulua, uku, onaga, hapu'upu'u, and 

 kahala) are all high-level carnivores and occupy 

 a similar trophic position. No predator-prey re- 

 lationship is known to exist between any of the 13 

 species listed in Table 1, although extensive gut 

 content analyses of all life history stages are 

 currently unavailable. Thus, some of the objec- 

 tionable aspects of the TBSM have been 

 minimized by not including species from differ- 

 ent trophic levels within the same analysis. 

 Predator-prey relationships in a fisheries 

 context are poorly understood at present and will 

 probably require a more dynamic construct than 

 the TBSM is capable of offering (May et al. 1979; 

 Pauly 1979). 



SUMMARY 



Examining the HDFG catch report data shows 

 that the commercial deep-sea handline fishery in 

 the Hawaiian Islands is a multispecies fishery 

 composed principally of 13 species of bottom fish, 

 6 of which comprise 86% of total landings. Snap- 

 pers (Lutjanidae), jacks (Carangidae), and a 

 species of grouper (Serranidae) dominate the 

 catch, all of which are high-level carnivores. 



In the main high islands of the Hawaiian 

 Archipelago (see Figure 2) three bottom fish 



446 



