186 DISCOVERY REPORTS 



the formula is replaced by D = 3-67 *Jw 3 l<Js'm 6 (Sverdrup et al. 1946).) Even with force 8 winds the 

 depth D is of the order of 200-250 m., and therefore it is evident that only in the coastal region will 

 D have an effect upon the wind current. The region where the south-east trade prevails, outside the 

 continental shelf, will be void of any bottom-effects. 



Thus in the open ocean the south-east trade wind will produce a surface drift in a westerly direction 

 and a total transport of water to the south-west. The lighter surface-water will therefore be trans- 

 ported away from the African coastal region. The presence of the coastline, however, prevents any 

 surface replacement of this water in the coastal region, and therefore a vertical replacement or 

 upwelling must follow. This vertical movement in the coastal region will create a distribution of 

 density such that the heavier water lies against the coast, and a relative current will be produced 

 running northward along the coast. 



The intensity of this relative current will depend upon both the direction of wind and the direction 

 of the coastline. Now, in South-west Africa, the coast runs N.N. W. to S.S.E. and the trade wind is 

 south-easterly becoming more southerly in higher latitudes. The angle between the wind and the 

 coast increases, therefore, from south to north, and one might expect from this an intensification of 

 the relative current to the north. 



The total transport of the wind current is, however, directly proportional to the wind stress and 

 inversely proportional to the sine of the latitude, so the decreasing strength of the trade wind to the 

 north of the region will tend to reduce the intensity of the relative current. The ultimate development 

 of the latter will depend on the magnitude of these various effects. 



The winds in the coastal region must have an additional modifying influence on the relative current 

 set up by the trade wind. Furthermore, as we pass into the coastal region the sea bottom, rising to the 

 continental shelf, will have an increasing effect on the deflection of the wind current from the wind 

 stress. It is not practical, however, to calculate the deflection as the coast is approached, since the 

 wind also veers the nearer the coast one gets. 



Observed winds and hydrographical conditions 

 The synoptic charts of the South-west African Weather Bureau show that in September-October the 

 trade winds were consistently stronger than in February-March. Theoretically this would have given 

 rise to a greater relative current along the coast in the former months. 



To estimate the effect of the coastal winds upon this relative current the wind vectors for three 

 coastal stations have been plotted (Fig. 35) on the assumption that they would produce a surface drift 

 at 45 to the left of the wind direction. All such vectors producing a drift between a line parallel to 

 the coast to the north and a line normal to this offshore, have been considered positive, while all other 

 winds have been taken as negative. From this figure it can be seen that during the first survey the 

 coastal winds would only be expected to intensify the relative current off Luderitz Bay, little effect 

 being exercised by the light winds and calms at Walvis Bay and Orange river. On the second survey 

 in September-October there was a greater amount of coastal wind, particularly at Walvis Bay and 

 Luderitz Bay At Luderitz, it was predominantly positive and so would have intensified the relative 

 current, whiie at Walvis Bay the greater variation of direction might have been contributory to a more 

 rapid breakdown of the relative current into eddies. 



The relatively calm spells at Walvis Bay and Orange river in March are reflected in the absence of 

 convection layers in these sections while the greater wind stress in spring is associated with better 

 mixed inshore waters in that season (Table 5). 



A comparison of these features of the wind activity with the distribution of surface temperature 

 and salinity on the two surveys suggests that the localized intensification of upwelling and indeed the 



