ANTARCTIC SURFACE WATER 



The Antarctic surface layer of the South Atlantic Ocean has been described by the 

 author in an earlier report (1933). It is a shallow layer of cold water 100-250 m. in 

 thickness lying above a deeper and more extensive layer of warm highly saline water. 

 In winter it is practically homogeneous, but in summer, owing to the greater effect on 

 the surface water of the radiation from the sun and thaw-water from melting ice and 

 snow, there is a surface stratum which is much warmer and less saline than the rest of 

 the layer. The observations made during the circumpolar cruise in 1932-3 show that 

 there is a similar layer round the whole of the Southern Ocean ; it is bounded on the 

 south by the Antarctic Continent, and on the north by the Antarctic convergence. The 

 extent of the zone is shown on a circumpolar chart in Fig. 4 (p. 19). 



In some of the shelf-seas neighbouring the Antarctic Continent the surface layer is 

 not so well defined. The observations made by Brennecke (1921, p. 99) at Deutsch- 

 land St. 125, near Vahsel Bay, south of the Weddell Sea, and our observations in the 

 southern part of the Ross Sea (Sts. RS 19-29) and the Bransfield Strait (Clowes, 1934), 

 show that in these localities there are basins from which the warm deep water is wholly 

 or partly excluded by submarine ridges ; in such basins the water is completely or almost 

 homogeneous from the surface to the bottom in winter, and the surface water differs 

 from the deep water only in summer. At Sts. 1015-17 in section 1 (Plate II) the con- 

 ditions are typical of the Antarctic Zone, and the surface layer is separated from the 

 warm deep layer by a sharp discontinuity in which the temperature, salinity, and oxygen 

 content change rapidly with depth ; but at St. 1014 in the northern part of the Bransfield 

 Strait the properties of the surface water are not very different from those of the deep 

 water. Near the Antarctic Continent there are probably many other small basins with no 

 warm deep water and they are even found relatively far north in the Antarctic Zone, notably 

 in Douglas Strait between Cook and Thule Islands in the South Sandwich group (Kemp 

 and Nelson, 193 1, pp. 178-83), and on a smaller scale in Moranen and Drygalski Fjords 

 in South Georgia (Hart, 1934, pp. 213-14, and Mosby, 1934, pp. 109-111). 



The observations made during the circumpolar cruise give some indication of the 

 length of the season in which the surface stratum is warmer than the rest of the layer. 

 The data from the southern part of the Pacific Ocean show that in September and 

 October the layer is almost homogeneous : the only indication of a division into surface 

 and cold strata was found at St. 961 in section 15 (Plates XXXIV-XXXVI), but the 

 temperature difference, o-8° C, was too large to be regarded as due to the warming of 

 the water at the surface, and it was more probably a sign of a southward current at the 

 surface. The observations made along section 1 in the early part of November show that 

 a distinct surface stratum had developed. At Sts. 1015-16 in 59-57 S the surface 

 water was o-i to 0-2° C. warmer than the coldest water at 80-100 m., and at St. 1017 

 in 56 S the difference was as much as 0-7° C. The temperature curves for a station 

 north of Prince Olaf Harbour, South Georgia (Deacon, 1933, fig. 14), show that in 

 approximately 53 S the surface stratum begins to form in about the middle of October. 



