The Stratospheric Circulation 675 



in a short section after its passage through Drake's Passage. The entire energy dissipa- 

 tion and all the other disturbances occur at this point; in all the other sections of the 

 current course it is a simple frictionless geostrophic current. 



Stommel developed a simple model (Fig. 318Z7,(af) consisting ofa homogeneous ocean 

 of uniform depth surrounding a schematic Antarctic continent and only at one place 

 (indicated by the heavy radial line) a barrier closes Drake's Passage completely. The 

 zonal wind system assumed is also shown in Fig. 3186,(<^) with trade winds from the 

 equator to 30° S., westerlies from 30° S. to a little over 60° S. and further south a nar- 

 row zone of easterlies. The Ekman drift current transport is northwards in the 

 westerlies and southwards in the easterlies. Therefore a divergence zone exists between 

 about 55° and 50° S. and a convergence zone further north. Since there is a complete 

 barrier it is not difficult to maintain a wind-driven circulation. The meridional 

 components of this circulation are indicated by arrows in Fig. 3186,(fl) and the entire 

 circulation is shown in Fig. 318Z),(^)- At the western coast of the ocean (the eastern side 

 of the barrier) an intense western boundary current will be set up and this simple cir- 

 culation will be characterized by two immense gyres around the earth parallel to the 

 latitude circles. Stommel calculated that the transport in the southern gyre would be 

 somewhat more than 100 x 10^ m^ sec"^, and somewhat less in the northern gyre. 



In fact, however, the northern gyre is broken up by the African and by the Australian- 

 New Zealand land mass. If now the barrier between South America and the Antarctic is 

 broken in the manner indicated in Fig. 3 1 86, (c) then the transport lines will run through 

 this opening and a circulation to the east will develop at the southern rim of the 

 barrier. The flow through the passage still remains unexplained but without doubt 

 models can be devised in order to describe it. Stommel's explanation of the dynamics 

 of the Circumpolar Current is quite different from the previous explanations. He also 

 attempted to make this explanation more plausible by embedding this current system 

 under consideration into the system of the sub-Antarctic-Antarctic circulation. 



5. The Sub-Antarctic Intermediate Current 



The most important facts concerning the spread of the subpolar Antarctic inter- 

 mediate water as far as they can be deduced from the distribution of the oceano- 

 graphic factors have been described already in Pt. I, p. 173. This water type forms the 

 uppermost part of the oceanic stratosphere. The sinking at the polar convergence is 

 shown by all meridional salinity sections (see Pt. I; Fig. 62 for the Atlantic, p. 147, 

 Fig. 75 for the Indian Ocean and Fig. 76 for the Pacific, p. 172). The fact that this 

 process at the Antarctic convergence (see p. 669) occurs with about the same intensity 

 all round the earth indicates that at all places the sinking and the subsequent spread 

 of this water type are caused by the same factors. 



In the Northern Hemisphere the morphological configuration of the continents 

 interferes with the formation of an Arctic intermediate current and traces of it are 

 found only along the western side of the Atlantic. The weakly saline intermediate 

 current in the Atlantic is consequently not symmetrical about the equator and we 

 may only speak of a sub-Antarctic intermediate current here. In the Pacific the 

 northern current branch is almost as strong as the southern one and therefore in 

 the region of the thermal equator (6° to 8° N.) very similar water types come in contact 

 with each other. In the Atlantic the Antarctic branch is, however, so strongly 



