imagery and shipboard hydrographic sections is significant but does not 

 explain the rapidly setting (300 to 350 cm sec"^) surface and near-surface 

 currents and the extraordinarily large values of mass transport. Swallow and 

 Bruce (1966) calculated geostrophic volume transports of 47.5 sverdrups in the 

 upper 200 m relative to 1000 dbar along a hydrographic section offshore from 

 the Somali Coast at S^N during the 1964 southwest monsoon season. Direct 

 current measurements along the same hydrographic section indicated volume 

 transports of 62.4 sverdrups in the upper 200 m. The measured volume trans- 

 ports during the 1964 southwest monsoon season were found to be supergeostrophic. 



In order to understand the dynamics of the Somali Current it is important 

 to examine the relative importance of the local and remote forcing. In 

 examining the effects of the low level Somali jet, Anderson and Rowlands 

 (1976) showed the initial relative predominance of the local forcing of the 

 prevailing strong southwesterly monsoon winds. In using nonlinear theory to 

 preclude excess leakage of energy equatorward via the Kelvin wave, it was 

 determined that the important feature of the local forcing mechanism was the 

 longshore component of the wind stress. The longshore component of the wind 

 stress is the explanation of the coastal upwelling along the Somali Coast and 

 the accompanying geostrophic northward currents. The northward flow is ini- 

 tially propagated equatorward via the Kelvin wave, but as the local northward 

 current builds up, the equatorward propagation is inhibited and the Kelvin 

 wave is slowed down. This serves to amplify the effects of the local wind 

 stress by preventing leakage of current energy into the equatorial Kelvin 

 wave. The remote forcing of the Somali Current is provided by the wind stress 

 field over the interior of the Indian Ocean. The effects of this remote wind 

 stress field are propagated westward along the equator by equatorial Kelvin 

 waves with the first baroclinic mode arriving at the Somali Coast in nine 

 days. For this reason the effects of the local wind stress are much more 

 immediate than those of the wind stress curl over the interior of the Indian 

 Ocean (Anderson and Rowlands, 1976). 



Although the volume transport of the fully developed Somali Current south 

 of Socotra Island is approximately 70 sverdrups and is thus comparable to that 

 found in the Gulf Stream, the Somali Current exhibits a strong increase in 

 volume transport downstream. This increase is particularly dramatic between 

 0.5°N and 2.5°N with an increase from 13 to 30 sverdrups calculated from 

 hydrographic sections taken during the 1964 southwest monsoon season (Diiing, 

 1978). Although one might be led to ascribe this pronounced increase in mass 

 transport to the local wind forcing of the southwest monsoon, the area between 

 0.5°N and 2.5°N is too close to the equator to contain the strong southwester- 

 ly monsoon winds prevalent along the central and northern Somali Coast. 

 Moreover, there is no surface current during the southwest monsoon season in 

 this area to supply the water required for the increase in volume transport. 



The rapid increase in volume transport of the Somali Current raises the 

 broader question of the water budget of the Somali Current/Arabian Sea circu- 

 lation system in general. If the rapid increase in volume transport in the 

 southern part of the Somali Current is caused by neither locally wind-induced 

 upwelling nor a strong westerly-setting equatorial current, then one must ask, 

 "Where does the water come from?" A more complete quantitative understanding 

 of the quantities of water recirculated in the large anticyclonic eddies and 

 advected into the Arabian Sea must be achieved. The exchange rates of water 



20 



