always in agreement with those obtained from surface dynamic heights. 

 This emphasized the importance of the subsurface distribution of mass. 

 The survey data indicate three major systems within the POFI sampling 

 depths: the system associated with the surface distribution of nnass, 

 that associated with the salinity mciximum, and that associated with the 

 salinity minimum. 



The 24. 5 and 26. 5 sigma-t surfaces were generally associated 

 with the salinity maximum and the salinity minimum, respectively. 

 Their isobaths would give an indication of the direction of flow at these 

 levels. Conventional isentropic analysis, however, such as plotting the 

 variation of a tracer, like salinity, on surfaces of potential density 

 (sigma-t) was not successful in the Hawaiian Islands region. 



The relative importance of the systems associated with the 

 salinity maximum and minimum was further investigated by means of 

 dynamic height anomalies in these layers. The reference levels were 

 the maximum sampling depths (600 m. for cruise 17, 1, 000 m. for 

 cruise 20, and 500 m. for cruise 21) and the approximate depth of the 

 inflection point between the salinity maximum and the salinity minimum 

 on the tennperature-salinity diagram (fig. 32). Thus for cruise 17 the 

 contribution to the net surface flow due to the system associated with 

 the salinity minimum would be represented by the dynamic height anoma- 

 lies at 300 m. relative to the 600-m. level. The contribution to the net 

 surface flow due to the upper layer, associated with the salinity maxinrium 

 and the surface distribution of mass, would be represented by the dynamic 

 height anomalies at the surface relative to 300 m. The latter gives the 

 current directions and magnitudes relative to the currents at 300 m. (The 

 concept of relative motion is familiar to navigators, for because of current 

 nnotion the actual path of a vessel nnay be quite different from either the 

 course steered or the current direction.) This presentation shows the re- 

 lative importance of the distribution of mass associated with two subsurface 

 flow systems, the upper of which would include the effect due to the surface 

 distribution of mass. 



The dynamic topographies of the intermediate water masses 

 associated with the salinity minimum of the three cruises were shown in 

 figures 8, 21, and 34. In each case any net flow through the survey region 

 is obscured, except in figure 8 of cruise 17, where the westerly flow is 

 clearly in evidence. Qualitatively, th& averaged dynamic heights for 

 cruises 20 and 21 show a southwesterly setting drift passing through the 

 survey region. Differences between the averaged dynamic heights in the 

 quadrants of the survey region for each cruise indicate that the strength 

 of currents during the sumnrier months (cruises 17 and 21) is approximately 

 twice that observed during the winter cruise. 



29 



