BARKI.EY ET AL: SKIPJACK TUNA HABITAT 



on individual sizes of skipjack tuna composing the 

 catch nor synoptic information on the vertical dis- 

 tribution of temperature and dissolved oxygen in 

 the fishing area. 2) The degree to which catch per 

 unit effort measures fish abundance may vary 

 greatly with gear type, fish size, and environmen- 

 tal conditions. For example, the habitat 

 hypothesis implies that purse seines (which fish 

 the upper 50 m or so of the water column ) should be 

 most effective in those parts of skipjack tuna 

 habitat with a shallow floor, hi fact, the eastern 

 Pacific purse seine fishery operates almost en- 

 tirely in waters with a skipjack tuna habitat-floor 

 at depths of 50 m or less (c.f. our Figure 3 and fig. 1 

 of Matsumoto 1975). Efforts to catch skipjack tuna 

 by purse seining in Hawaiian waters — where the 

 habitat-floor lies at depths near 200 m (Figure 

 3) — have been ineffectual (Murphy and Niska 

 1953 k Green ( 1967) reported strong positive corre- 

 lations between the success of purse seining for 

 eastern Pacific skipjack tuna and yellowfin tuna, 

 Thunnus albacares, and the presence of shallow 

 ( «60 m to top), steep ( >0.55°C cm ' ) thermoclines. 

 3) Commercial fishermen naturally fish only 

 where they expect to find and catch fish; thus, 

 fishing effort tends to be very unevenly distri- 

 buted. 



A partial test of the habitat hypothesis might be 

 achieved through experimental fishing in and 

 near the hatched areas of Figure 6; fishing effort, 

 the sizes of captured skipjack tuna, and vertical 

 distributions of temperature and dissolved oxygen 

 would need to be measured at each fishing loca- 

 tion. But, experimental fishing — even if con- 

 ducted in a thorough and systematic fashion — 

 might not yield a conclusive test of the hypothesis, 

 because there would still be no guarantee that 

 catch per unit effort accurately reflected flsh 

 abundance (reason 2 above). 



We advocate, instead, the application of ul- 

 trasonic telemetry to test the skipjack tuna- 

 habitat hypothesis. Because skipjack tuna tagged 

 with ultrasonic transmitters tend to stay with 

 their school (Yuen 1966) and because skipjack 

 tuna schools tend to be homogeneous with respect 

 to fish size (Brock 1954), the track of a single 

 tagged fish could be taken as representative of a 

 large number of normally behaving, similarly 

 sized fish. Pressure-sensitive ultrasonic transmit- 

 ters ( like that described by Luke et al. 1973 ) would 

 permit continuous monitoring of fish position in 

 all three spatial dimensions through time. 

 Spatial-temporal coordinates offish could then be 



compared with synoptic data on vertical distribu- 

 tions of temperature and dissolved oxygen. A few 

 dozens of such comparisons, for fish of various sizes 

 in waters with diverse vertical distributions of 

 temperature and dissolved oxygen, would consti- 

 tute a valid and sufficient test of the habitat 

 hypothesis. Toward this end, preliminary tele- 

 metry work is now underway at this Laboratory. 



SUMMARY 



Work with captive skipjack tuna at this 

 Laboratory has yielded information on the tem- 

 perature and dissolved oxygen requirements of 

 this species. If these laboratory results apply to 

 skipjack tuna in nature, they provide new insight 

 into the evolution of migration in skipjack tuna 

 populations, make it possible to account for the 

 geographic distribution of skipjack tuna on the 

 basis of environmental conditions, and provide 

 means for predicting their movements in major 

 fisheries such as those of the eastern tropical 

 Pacific. 



In particular, we suggest that only young skip- 

 jack tuna can inhabit tropical surface waters, and 

 that the habitat of adult skipjack tuna in the 

 tropics is the thermocline and not the warmer 

 surface layer, as has generally been thought. 

 Since the thermocline in many areas is too 

 oxygen-poor to support these active fish and the 

 well-oxygenated surface layer is too warm for 

 adult skipjack tuna, only heat-tolerant young 

 skipjack tuna can live in those areas. As they 

 grow, these fish are forced to move into areas 

 where well-oxygenated water of the proper tem- 

 perature is more readily available. 



Up to now, it has not been possible to trace the 

 movements of migrating skipjack tuna largely be- 

 cause they move through areas of many millions of 

 square miles, at unknown depths. Knowledge of 

 their temperature and dissolved oxygen require- 

 ments dramatically reduces the scope of the prob- 

 lem: the fish should be in a well-defined layer of 

 water, of directly and easily measured thickness, 

 whose geographic extent can be sharply defined 

 with either historical or current oceanographic 

 observations. 



ACKNOWLEDGMENTS 



The physiological studies on which this paper is 

 based were supported, in part, by the University of 

 Wisconsin Brittingham Foundation. 



661 



