DEPTH OF THE SURFACE LAYER AS AN INDICATION OF 

 SURFACE CURRENTS. 



General. The usual procedure in determining the current 

 distribution is to determine the relative mass field from the 

 vertical temperature and salinity distribution. In the absence 

 of sufficient salinity data, it is often possible to obtain an 

 indication of the relative mass field from the depth of the 

 lighter warm surface layer. This results from the fact that 

 in the southern hemisphere the mass field and current field 

 are related in such a manner that the lighter water must occur 

 on the left of the current and the heavier water must occur on 

 the right of the current (observer facing downstream). Hence, 

 in the antarctic the slope of the semidis continuity surface at the 

 bottom of the surface layer (top of the thermocline) is associ- 

 ated with a relative current that flows in such a direction that 

 the deeper layer (hence lighter water) is on the left and the 

 shallower layer (hence heavier water) is on the right. 



This current -mass field is shown schematically in figure 17 

 which presents a vertical section taken across a current 

 which is shown as though flowing into the paper. The boundary 

 between the relatively warm surface water and the heavier 

 cold water below the thermocline is shown as sloping upward 

 to the right. Since column A has a greater percentage of 

 light surface water than column B, column B will weigh more 

 than column A. The conditions requiring the lighter water 

 on the left-hand side of a current in the southern hemisphere 

 are thus satisfied; hence the current flow is at right angles 

 to the section and the direction of flow is into the paper. 



Surface Currents from the Observed Temperature Struc- 

 ture. The layer depth was taken as the depth from the surface 

 to the top of the main seasonal thermocline. This depth was 

 averaged for each degree of latitude and longitude, and for 

 some regions, where the number of observations were few, 

 averages were taken over larger areas. The averaging tended 

 to smooth the data and hence to diminish the effect of the 

 small-scale, short-period changes in factors that alter the 

 layer depth (wind force, local heating and cooling, internal 

 waves, etc.), and to emphasize the more widespread, longer- 

 period trends. 



33 



