FISHERY BULLETIN: VOL. 85, NO. 4 



i.ar 



U.S. Catch 



1.0- 



ID 



£1 



a 



(0 

 CJ 



Effort«2 : 



. . Baitboat 



Longl ine 



_... U.S. 



Effort/2 : 



. Baitboat 



Longl ine 



. . U.S. 



Nominal : 



105 



115 



125 



Length (cm) 



Figure 6. — Annual albacore U.S. catch-at-size in numbers predicted by the model under nominal and experimental conditions. 



(expressed as percent) persisted through a series 

 of updates of the model (such as changes in config- 

 uration of geographic strata) and inevitable up- 

 dates and corrections of the fishery data base. 



Though the nominal behavior of the model need 

 not conform precisely to the mean behavior of 

 albacore fisheries in the 1970s (however that 

 might be defined), the behavior should, nonethe- 

 less, be a plausible representation of the albacore 

 fisheries in that period. We have seen, for exam- 

 ple, that our conclusion would be suspect if the 

 actual exploitation rate were considerably higher 

 than the 6% value that we assumed (Fig. 7), but 

 such high exploitation levels would be contrary to 

 the tag return results. 



We have only tested the effects of changes in 

 the magnitude of effort, not changes in seasonal 

 and geographic pattern of effort, which might 

 cause the fleets to overlap much more than they 

 do. However, our experimental treatment of dou- 

 bling the effort of a fleet is tantamount to adding 



u 



U 



c 

 o 



■D 



cr 



5 10 15 



Exploitation rate (%) 



Figure 7. — Sensitivity of interaction to overall exploitation 

 rate. The ordinate is the percent reduction from nominal levels 

 in annual albacore longline catch in weight as a result of dou- 

 bling the baitboat effort. 



710 



