248 KNAUSS [chap. 10 



As noted previously there are marked seasonal changes in the equatorial 

 currents, and it is of interest to know to what extent these seasonal changes 

 are reflected in the distribution of mass. Using the slope of the thermocline as 

 a criterion for the presence and strength of the Pacific Equatorial Counter- 

 current, Cromwell (1958) and Knauss (1958) have shown that there is qualitative 

 agreement with the seasonal change in the countercurrent and the change in 

 the slope of the thermocline. From their analysis it is not possible to tell 

 whether or not there is any appreciable lag in the response of the mass distribu- 

 tion to the changing currents, or whether the response is complete; that is, 

 whether the countercurrent is always in geostrophic balance. This question is 

 of some interest in view of the recent analysis of Veronis and Stommel (1956) 

 on the response of a baroclinic ocean to a variable wind stress. The Atlantic 

 and Pacific, however, are not the best areas to study this problem. It will be 

 easier and more clear-cut to determine a phase lag (if, indeed, a measurable one 

 exists) in the Indian Ocean, where the currents north of the equator reverse 

 with the changing monsoon conditions. 



5. Meridional Flow — Up welling 



Except at the ocean boundaries (where they are more properly treated as 

 eastern or western boundary currents), the north-south flow in equatorial 

 waters is weak. The presence of meridional flow has most often been described 

 on the basis of either (a) requirements in a theory of oceanic circulation or (b) 

 inferences from the distribution of properties (Defant, 1935, 1936 ; Montgomery, 

 1938; Sverdrup, 1947; Reid, 1948; Cromwell, 1953; Tchernia et al, 1958). 



Cromwell {loc. cit.) discusses the evidence for upwelling at the equator and 

 the lack of such evidence at the north edge of the countercurrent in the central 

 Pacific. The latter was suggested for the Atlantic by Defant (loc. cit). Cromwell 

 further suggests that there is a northward transport in the region of the equator 

 in the surface layer and a convergence at about 2°-3°N. His picture of the 

 meridional flow contrasts with the requirements of the Sverdrup theory which 

 requires a southward transport from about 5°N to 5°S, and in the region of 

 2°-3°N (which is in the boundary region between the South Equatorial Current 

 and the countercurrent) where there is neither divergence nor convergence 

 (Reid, 1948). It should be noted, however, that the Sverdrup theory is con- 

 cerned only with the integrated mass transport over a vertical water column, 

 and, since there is the possibility of opposing currents within the column, 

 these differences may not be irreconcilable. 



A. Geostrophic Flow in a Meridional Plane 



The possibility of a weak geostrophic flow in the meridional plane should not 

 be overlooked, and the observational data are not inconsistent with this 

 possibility. As previously noted, the sea surface slopes upward to the west in 

 the vicinity of the equator in both the Atlantic and Pacific (Fig. 6). This slope 

 is equivalent to a pressure force of 5 x 10~ 5 dyne/g. The balancing geostrophic 



