anticyclonic eddies. Within the upper 100 m (approximation of the mixed layer 

 depth) along the tanker sea lane between 2°N and 12°N during the monsoon 

 period (approximately three months) there was a heat loss of approximately 3 x 

 10'2 to 5 x lo''^ cal/cm as determined from the XBT temperature stations 

 (figure 8). Now the area of the region influenced by entrainment within the 

 eddy circulation where relatively good mixing is found (Bruce, 1968) extends 

 roughly 400 km offshore. Thus the total heat loss of the area would be about 

 10^0 cal if the tanker section is assumed to be representative of the area. 

 The rate of heat loss during the southwest monsoon is the same order of magni- 

 tude as that found in the western North Atlantic on an annual average (Bunker 

 and Worthington, 1976), although in the Somali region the loss occurs only 

 over the monsoon period and is regained again in the interim northern fall and 

 spring warming period between the northeast and southwest monsoons. 



Although there is a heat loss in the upper 100 m during the southwest 

 monsoon, at the same time because of the deepening of the isotherms below the 

 prime eddy there is a heat gain in the layers deeper than the mixed layer 

 (figures 9, 10 and 11). Thus the amount of actual heat lost to the atmosphere 

 would be reduced by the transfer of heat to deeper layers. After the south- 

 west monsoon as the eddy slowly weakens, as shown by the series of XBT 

 temperature sections, heat is released from these layers for three to four 

 months after the monsoon. There is evidence from the sections that the prime 

 eddy in fact might have been maintained from 1975 to 1976 (in particular see 

 22-27 January 1976, figure 2) through the northeast monsoon as suggested f 

 earlier measurements by Bruce and Volkmann (1969); however, it is not clea 

 that the continuity occurred during other years. 



rom 

 r 



The data indicate that the temperature of the mixed layer of the equatorial 

 water in the Somali Basin as well changes at the time the eddy field is built 

 up. Between l^N to 1°S during the southwest monsoon the near-surface water 

 along the tanker lane (480E to SO^E) becomes cooler resulting in a correspond- 

 ing drop in sea surface dynamic height (relative to 400 dbar) on the order of 

 0.10 to 0.15 dynamic meters (figure 12). In figure 2 the reduction of tempera- 

 ture in the mixed layer can be seen to occur after the spring warming period 

 each year. During the warming period the layer normally reaches 28°C to 30°C. 

 By July and August after the eddy field has developed the temperature drops to 

 about 25'-'C to 26°C in this region. After the cessation of the southwest 

 monsoon by November the layer had again warmed (27°C to 28°C). These changes 

 are clearly shown in the surface temperature values given in figure 6. It 

 seems that this change might well influence the strength and/or direction of 

 the equatorial undercurrent because a local slope of the sea surface which is 

 negative westward might occur. Taft and Knauss (1967) found no evidence of 

 the undercurrent on the western side of the Indian Ocean during the months of 

 the southwest monsoon. It should be noted that although there are only sur- 

 face salinity observations from these XBT sections, it turns out that the 

 hydrographic data from this equatorial region indicate that the probable 

 seasonal salinity variations would produce changes in the sea surface dynamic 

 height of only about 0.01 dynamic meters. 



3. WIND FIELD IN THE WESTERN INDIAN OCEAN 



Recent interest in the effect of monsoonal winds on the circulation of the 

 Indian Ocean has resulted in several programs of oceanographic measurement 



