WILLIAMS: MODELS OF MIGRATION OF YOUNG SKIPJACK 



Revillagigedo Islands (as proposed in the active 

 migration model at this time of year). In the 

 period prior to cessation, when the NECC is in- 

 termittent, juvenile fish would increasingly lose 

 their eastward orientation, some being deflected 

 into the weak NEC and with a sufficient north- 

 ward component in the current might passively 

 reach the Revillagigedo Islands, though such con- 

 ditions would be short-lived. 



Should recruitment to the northern fishery not 

 take place at the time of cessation of the NECC, 

 juvenile skipjack at that time being recycled to 

 the central Pacific in the NEC and SEC, then all 

 recruits to the eastern Pacific fisheries would 

 have to be carried to the coast in the NECC in 

 the period May-June to December-January. 

 From long 130°W (eastern limit of spawning 

 area) to 85°W (approximate NECC nearshore 

 terminus) is 2,700 nautical miles and with an 

 average NECC speed of % knot, continuous pas- 

 sive movement over this distance would take 

 about 5 months. This would make the arrival 

 time at long 85 °W October-November to May- 

 June. Obviously transport time could be much 

 longer than this, if, for instance, (i) daylight 

 hours were entirely spent foraging and with 

 near-zero movement eastward, and/or (ii) real 

 transport rates were lower due to the effects of 

 diel vertical migrations. From the NECC ter- 

 minus at about long 85 °W recruitment, even pas- 

 sive, to the initial southern fishery areas would 

 probably take no more than 4-6 weeks, i.e., ar- 

 rival November-January to June-August. But a 

 passive drift from the NECC terminus north- 

 west to the Revillagigedo Islands with surface 

 currents about 1/2 knot would take in the order 

 of 4-5 months, i.e., arrival February-May to Sep- 

 tember-December. However, the recruitment 

 size of the skipjack to both northern and south- 

 ern fisheries appears similar. Several factors 

 might be proposed to account for this fact: (i) 

 the spawning grounds of the fish found in the 

 southern fishery are farther west in the central 

 Pacific than those of the northern fishery, (ii) 

 the early life history stages enter the NECC at 

 different sizes, (iii) behavioral patterns of the 

 juveniles (from the two groups) in the NECC 

 are different, or (iv) the juveniles stay longer 

 in the offshore areas of the southern fishery than 



the northern one. Of these proposals the first 

 seems most likely, and would be understandable 

 if the two fishery groups did represent parts of 

 two spawning groups (northern and southern) 

 as suggested earlier (see page 748) . One feature 

 which could influence the eastern boundary of the 

 groups spawning in the central Pacific, particu- 

 larly the southern one, would be the westward 

 extent of the influence of low near-surface tem- 

 peratures at and south of the equator due to 

 equatorial upwelling and eflfects of Peru Current 

 water. 



A wide degree of geographic overlap of the 

 spawning groups in the central Pacific would 

 alleviate problems of recruitment of young stages 

 of southern spawners into the NECC, as trans- 

 equatorial migration might not then be involved 

 to any extent. In addition, the occurrence of a 

 surface SECC would provide a possible mech- 

 anism by which young stages of southern spawn- 

 ers from south of the equator could be carried 

 passively into the eastern Pacific. However, as 

 previously stated, the existence of a surface 

 SECC has not yet been confirmed in the eastern 

 Pacific. 



Thus, as our knowledge stands at the moment 

 the NECC could be passively carrying juveniles 

 of both spawning groups into the eastern Pacific 

 fisheries but at different sizes. 



GYRAL MIGRATION MODEL (Figure 7) 



The two models described so far are based on 

 the assumption that larval or juvenile fish mi- 

 grate, passively or actively, out of the central 

 Pacific to feed in coastal surface waters of the 

 eastern Pacific and then subsequently return to 

 the central Pacific for first spawning. Between 

 spawnings the adult fish diffuse outwards from 

 the central Pacific to feed but rarely reenter the 

 adolescent feeding grounds. 



Incidental longline catches of skipjack through 

 1967 (Miyake, 1968; Walter M. Matsumoto, per- 

 sonal communication) indicate the widespread 

 distribution of the adults. According to Matsu- 

 moto there is some evidence that areas of high 

 longline catch-per-unit-effort of skipjack in the 

 central-east Pacific show seasonal shifts which 

 tend to coincide with the direction of flow of the 



753 



