SECKEL: SKIPJACK AND ENVIRONMENT 



nitude as the mean velocity of the salmon relative 

 to the currents and cannot be neglected. In fact, 

 it is important to consider Vf rather than (Vw 

 + Vf) when studying the navigational abilities 

 or behavior of fish. Even if Vw is one order of 

 magnitude smaller than Vf, when travel times 

 of 1 or 2 years are involved, Vw may not be ne- 

 glected and the destination will reflect the effect 

 of the current system. 



An example of Vw being negligible in com- 

 parison with Vf is the travel of albacore across 

 major portions of the North Pacific Ocean. The 

 example where Vf is very much smaller than Vw 

 may be the travel of skipjack from the eastern 

 to the central North Pacific Ocean. 



When skipjack reach the vicinity of islands 

 a fixed reference becomes available and the 

 swimming behavior is likely to become different 

 from that in the open ocean. The fish that were 

 tagged by a sonic device near the Hawaiian Is- 

 lands (Yuen, 1970) are an example of such be- 

 havior. The current field is also affected by the 

 proximity of islands. Although the relative mag- 

 nitudes of the current velocities and swimming 

 velocities may diflfer from the open ocean case, 

 both velocities must still be considered. The 

 travel behavior of skipjack near islands is, how- 

 ever, a different problem from that considered 

 in this paper because the time scale is in the 

 order of hours rather than weeks. 



Finally, the relative magnitude of Vf as com- 

 pared with Vw may vary throughout the travel 

 history of a particular species of fish. This var- 

 iation was documented by Royce et al. (1968) for 

 the case of Pacific salmon and may also apply 

 to albacore. Williams (1972) tends to favor 

 "active" migration of skipjack into the eastern 

 North Pacific fishery. Therefore, Vf may not 

 be small when compared with Vw throughout the 

 travel history of skipjack. 



CONCLUSION 



A plea is made in this paper by way of pro- 

 posing a model, much as was done by Rothschild 

 (1965), to progress from the exploratory phase 

 of skipjack distribution studies to the experi- 

 mental phase. Results from exploration (as- 



semblage of data collected without experimental 

 design) were used to demonstrate empirical as- 

 sociations between the availability of skipjack 

 to the Hawaiian fishery and environmental in- 

 dices. Important to an understanding of the 

 life history is the linkage between environment 

 and the distribution of skipjack that the empiri- 

 cal associations do not provide. 



Insight into the linkage mechanisms is gained 

 if the associations are used as leads to hypotheses 

 or models that can be tested experimentally. 



Modern technology together with the power- 

 ful analytical tools now available make it pos- 

 sible to construct a complete migration-distribu- 

 tion model of skipjack from the eastern and 

 central North Pacific Ocean. In such a model 

 the North Equatorial Current portion would be 

 linked with one of the eastern Pacific models of 

 Williams (1972) and with a larval survival-year 

 class strength model. Swimming velocities of 

 skipjack can be simulated and included in the 

 model. Numerical evaluation of such a model 

 depends upon adequate environmental informa- 

 tion, e.g., large-scale sea-air interaction proces- 

 ses, geostrophic and wind-driven velocities of 

 ocean currents. 



Important elements that were not discussed 

 in this paper must be evaluated. For example, 

 there are the effects of dispersion on the distri- 

 bution due to the random motions of skipjack 

 schools and due to large eddies within the cur- 

 rent system. Currents, either geostrophic or 

 wind driven, are not necessarily constant within 

 the range of vertical movement of skipjack. The 

 current drift must therefore be tuned to the 

 depth range within which skipjack swim. 



The sources for the required environmental 

 information are meteorological observations 

 from ships, that in the future may be supple- 

 mented by buoys. Geostrophic current speeds 

 can be monitored by the use of vertical temper- 

 ature sections obtained from merchant ships reg- 

 ularly traveling specific routes. The dispersive 

 effect of random fish school motions can be de- 

 termined by using a sonic tag to track skipjack 

 as was described by Yuen (1970). The disper- 

 sive effect of eddying currents can be determined 

 from drifting buoys whose positions are mon- 

 itored by satellite. Predicted drift or progress 



785 



