548 Ejfect of Wind on the Mass Field and on the Density Current 



from another direction. Thus, for example, in a broad band of an oceanic region with 

 vertical increase of density and forming a channel around the earth in the Northern 

 Hemisphere, conditions will be more or less as follows. 



If there is a persistent wind in the direction of the channel the immediate effect of 

 the drift current (westerly wind) is to transport lighter surface water to the right 

 (south) side of the channel. In the top layers the isosteric surfaces can no longer be 

 horizontal and will adjust with an inclination from north to south in order to corres- 

 pond with the accumulation of lighter water on the right-hand side of the wind. A 

 solenoid field of this type will, however, produce a density current in the direction of 

 the wind in which the velocity will decrease with depth corresponding to a similar 

 decrease in the slope of the isosteres. At the same time, water will be piled up on the 

 right-hand (south) side of the channel and this will give rise to a gradient (Stau) 

 current in the direction of the wind. Its velocity will remain constant down to the lower 

 frictional depth. In this way the stratification will lead to a considerable complication 

 of the conditions and even more so if changes due to other factors (heating, cooling, 

 evaporation and others) must, too, be taken into consideration. 



It is doubtful whether a gradient (Stau) current will be generated in such a current 

 system. The displacement of the water masses in the top layer, where the solenoids 

 are numerous and which is superimposed on deep water where the solenoids are few, 

 may proceed so that the isobaric surfaces in the deep water remain horizontal (see 

 discussion on p. 483 and following pages). If the effect of the water accumulation (rise 

 in physical sea level) occurring on the right-hand side of the wind direction (Northern 

 Hemisphere) on the pressure field of the deeper water is compensated exactly by the 

 baroclinic mass distribution of the top layer there will be no gradient (Stau) current. 

 In actual practice, the relationship between the topography of the physical sea level 

 and the mass structure of the upper layers is usually satisfied so that any deep reaching 

 slope current is improbable. 



A complete theoretical treatment of the problem of currents in a baroclinic ocean 

 offers considerable mathematical difficulties, since it must take into account vertical 

 frictional effects, lateral mixing processes and boundary-surface conditions. In con- 

 nection with an investigation on the circulation of the antarctic circumpolar waters, 

 SvERDRUP (1933) has discussed the possibility of formation o{ o. steady drift current in 

 the presence of a baroclinic stratification of the water masses. He showed, in agree- 

 ment with the results of Ekman, that steady vertical circulations can hardly develop 

 in the ocean if only the effect of wind is taken into account. Due to the non-uniformity 

 of the wind field (divergences and convergences), and due to the boundaries between 

 different water bodies and the coasts, vertical circulations will be formed and will 

 produce changes in the mass field. However, since the density distribution in the sea 

 is usually a stationary one and apparently steady circulations still occur, it follows that 

 the effect of the vertical circulations produced by wind must be compensated by other 

 factors which affect the density. This gives emphasis to the great importance of these 

 factors for the development and maintenance of the oceanic circulation. Heating, 

 cooling, evaporation, precipitation and other factors thus take part indirectly in the 

 formation of the oceanic circulation. The convective sinking of cold waters in higher 

 latitudes plays an especially important part for the maintenance of vertical oceanic 

 circulations. 



