MURPHY and SHOMURA: PRE-EXPLOITATION ABUNDANCE OF TUNAS 



o 

 o 



X 



o: 



UJ 



a. 



X 



o 

 o 



180' 



170° 160° 



LONGITUDE 



150° 



140° 



Figure 32. — Catch rates of yellowfin tuna by longitude 

 and by time of year (1953). 



and fish may have a firm biological basis in that 

 feeding tunas, by driving small fish and squid 

 to the surface, appear to be an important factor 

 in making food available to birds. 



The observed behavior of working bird flocks 

 substantiates this contention. A bird flock may 

 be first seen when it is feeding over a school, the 

 birds excitedly circling close to the sea surface 

 and frequently plummeting into the sea to cap- 

 ture prey. Subsequently, the school may sound 

 and the birds discontinue feeding, scatter some- 

 what, and rise to a higher elevation as though 

 scouting for the school. As soon as the school 

 appears again at the surface, the birds quickly 

 assemble over it and resume feeding. This re- 

 lationship suggests that the abundance of pe- 

 lagic predators such as tuna may directly affect 

 the abundance of birds by controlling the amount 

 of food available to them. Insofar as this is true, 

 bird abundance should form an index of fish 

 abundance. 



On the other hand, it is difficult to discount 

 completely the possibility of introducing error 

 through the use of birds as an indicator of fish, 

 for the frequency of sightings of both birds and 

 fish is everywhere parallel (Murphy and Ike- 

 hara, 1955), and it seems unlikely a priori that 

 the populations of both need always be parallel. 

 For instance, birds and fish are seen more fre- 

 quently around islands than far at sea. Oceanic 



birds nest on land, thus tending to make islands 

 focal points for their populations; but, is the vi- 

 cinity of land also a focal point for tuna schools? 

 Perhaps so, for nesting birds have an even great- 

 er demand for food than wandering birds, and 

 thus the magnitude of their concentrations 

 around islands must be based on the food supply 

 too and, to the extent that they obtain their food 

 over schooling tunas, on the abundance of schools 

 of surface tunas. 



Aside from the problem of just what surface 

 sightings mean in relation to tuna abundance, 

 the validity of conclusions based on relatively 

 scanty sampling of so vast an area as the cen- 

 tral Pacific (Figure 3) can be questioned. The 

 only available test, a comparison of sightings 

 from the Laboratory's vessels with the seasonal 

 changes in Hawaiian commercial landings (Fig- 

 ure 33), shows that in general a seasonal pre- 

 diction based on our relatively scanty observa- 

 tions would have been borne out by the com- 

 mercial catch. Thus, it appears that rather lim- 

 ited scanning of the sea can yield results com- 

 patible with those from a very large amount of 

 effort (the commercial fishery). 





o 



UJ 



I- 



X 



(/> 



(/) 



-I 



8 



X 



u 



< 

 z 



3 

 H 



< 



2£ 



MAR-MAY JUNE -AUG. SEPT- NOV. DEC- FEB. 



Figure 33. — Comparison between average Hawaiian 

 skipjack tuna landings (1950-52) and surface sightings 

 of skipjack tuna schools (number of schools sighted 

 noted above bar diagrams) from research vessels in the 

 Hawaiian Islands area (data from Murphy and Ikehara, 

 1955). 



903 



