The contribution to the surface circulation due to the intermediate 

 flow associated with the salinity minimum is illustrated by the dynamic 

 topography at 250 m. relative to the 500-m. level (fig. 34). It can be 

 seen that the vortices in the vicinity of stations 49 and 51 extend to the 

 depths of the salinity mininnum. The vortex at station 45 is, however, 

 obscured by a northeasterly setting current, the direction of which is 

 reversed south of Oahu and in Kaiwi Channel to continue as the westerly 

 setting current south of Kauai. Similar isopleth patterns in figures 33 

 and 34 again show that the isobaths of the 26. 5 signna-t surface at the 

 salinity minimum give a good indication of flow in this layer. Discrepan- 

 cies can probably be attributed to the fact that the dynamic height at 250 m. 

 relative to the 500-m, level reflects the distribution of mass only in the 

 upper portion of the flow layer associated with the salinity minimuna. 



The salinity distribution on the 24. 5 and the 26. 5 sigma-t surfaces 

 is also indicated by dashed lines in figures 31 and 33 and becomes plaus- 

 ible if studied in conjunction with the dynamic topography at the surface 

 relative to the 500 m. and that at 250 m. relative to the 500-m. level. 

 The usefulness of tracing currents by this method, however, is impaired 

 by the complexity of the system. 



A comparison of the dynamic heights presented in figures 29, 32, 

 and 34 shows that the major features in the two intermediate systems are 

 quite similar and generally have an additive effect on the net surface cir- 

 culatory pattern. The spacing of the contour lines southwest of the island 

 chain is closer in figure 32 than in figure 34, which means that the con- 

 tribution due to the intermediate system associated with the salinity max- 

 innum is of prinnary importance. Northeast of the island chain the contour 

 line spacing is approximately the same in the two figures, indicating that 

 there the two systems contribute equally. 



Differences in the contour line patterns of the two systems (figs. 

 32 and 34) exist north of the islands of Kauai and Oahu and in the vicinity 

 of station 47. The clockwise circulation about the island of Hawaii is 

 also absent in the deeper flow pattern (fig. 34). 



The averaged dynamic height anomalies at 250 m, relative to the 

 500-m. level in each subdivision of the survey area show a net drift 

 toward the island chain from the northeast which is deflected to the 

 northwest. The sanne computations for the surface flow relative to the 

 250-m. level also show a net drift toward the island chain from the north- 

 east, which in this case is deflected toward the southeast. The net motion 

 on the leeward side of the island chain appears negligible in each case. 

 These observations are in agreement with the averaged dynamic heights 



23 



