greater in the upper layers than in the lower (which also intensifies the mixing), causes exist which 

 retard the diffusion of heat upwards. It is not difficult to see that these causes consist of the fact 

 that the stability of the layers located above the warm layer is very great, while the stability of the 

 strata below it is very slight. 



LITERATURE: 45, 72, 77, 101, 164, 165. 



Section 148. Deep Water 



We have seen that water of positive temperature and high salinity in the Arctic Basin is con- 

 ventionally considered as Atlantic water. The conventionality of such a designation is particularly 

 marked in determination of the boundary between intermediate Atlantic and deep water which, as we 

 shall see below, is properly called Greenland water. 



Actually, all temperature- salinity curves constructed for observation stations of the Arctic 

 Basin show that homogeneity in respect to salinity (from 34.7 to 34.9 o/oo) commences approxi- 

 mately at depths of 300 to 400 m. The lower limit of positive temperatures is found, at a distance 

 from the shores, at depths of 700 to 800 m, while uniformity in temperature (from 0.5° to 0.9°) 

 commences at a depth of 1, 500 m. 



The fact must be noted, nevertheless, that the greatest temperature gradients below the level 

 of the Atlantic water axis are located everywhere at depths of 500 to 800 m. Thus the lower zero 

 isotherm coincides approximately with the lower limit of the large temperature gradients. There 

 is every basis for supposing that the large vertical temperature gradients are a result of compara- 

 tively large velocity gradients. We may consequently consider that in the Arctic Basin the Atlantic 

 water is separated from the deep Greenland water by a zone of lateral mixing. 



The origin of the deep water of the Arctic Basin was analyzed in detail by Nansen in connec- 

 tion with the Fr am observations . In addition, during the expedition on the Veslem in 1912, Nansen 

 made a series of observations for the purpose of checking his original conclusions and he arrived at 

 the same results. These conclusions are as follows: 



1. The deep water of the Arctic Basin is warmer than the deep water of the Greenland sea at 

 the same depths. Proceeding from this fact, Nansen came to the conclusion that between the north- 

 eastern end of Greenland and the northwestern end of Spitzbergen there must exist an underwater 

 ridge on which the depth must not exceed 1500 to 2000 m. This ridge must limit the water exchange 

 between the deep water of the Greenland Sea and the Arctic Basin and must explain the comparatively 

 high temperatures of the deep water of the Arctic Basin by comparison with the deep water of the 

 Greenland Sea. 



The existence of such a ridge is unquestionable but it has not yet been proven by direct obser- 

 vations. It has been decided to call it the Nansen ridge. 



2. The deep water of the Arctic Basin is formed in the Greenland Sea from Atlantic water at 

 approximately 75° north and at 0° longitude as a result of vertical winter circulation. It then passes 

 over the Nansen ridge and goes on to spread out along the bed of the Arctic Basin. 



Nansen illustrates his conception of the origin of the deep water of the Arctic Basin with a 

 diagram of an oceanographic section (figure 153), made from the Faeroe Islands north, approxi- 

 mately along the Greenwich meridian. 



416 



