270 



Ice in the Sea 



pack-ice occurrence off Newfoundland and that off eastern Greenland. The Davis 

 Strait values over a series of nine sunspot periods show, however, that the ice amount 

 in the Davis Strait follows the sunspot cycle with a lag of 2 years rather well (Fig. 124). 

 The fluctuations in the pack ice in the area of the Newfoundland Banks are of course 

 directly connected with those in Davis Strait. They also parallel exactly the fluctua- 

 tions in icebergs in the same area. This is shown by the high correlation factor of 



120 5 

 80 I 

 40 I 



O 



s 



1820 



1640 



I860 1880 



Years 



1900 



1920 



Fig. 124. Relative sunspot numbers and smoothed values for the amount of ice in Davis 



Strait. (The latter is displaced two years to the left relative to the sunspot curve (9 full 



periods) (full line : sunspot number, dashed line : amount of ice.) 



+0-86 between the number of icebergs south of Newfoundland (48° N.) and the pack 

 ice off Newfoundland valid from February until May (47 years, Smith, 1926-7). 



For the Barents Sea particularly good ice statistics are available for areas in which 

 ice-measurements have been made by the Danish Institute during the years 1896-1916 

 (Nautik-Meteorol. Aarbog 1916). Wiese (1924) has used these in a study of the rela- 

 tionship between the occurrence of ice and variations in the atmospheric circulation. 

 He was able to show that the ice intensity in this sea from May to June depends largely 

 on the distribution of atmospheric pressure over the Norwegian Sea during the period 

 from January to the end of April and that a larger (smaller) atmospheric pressure 

 gradient directed from south-east to north-west between the Norwegian coast and the 

 axis of the low-pressure trough over the Norwegian Sea causes a decrease (increase) 

 in the ice coverage of the Barents Sea. By calculations from the regression equations 

 with the factors affecting the ice coverage, it is possible to obtain reliable ice prognoses 

 for this area. 



A very strong aperiodic change in the Arctic has been in progress since 1918. Since 

 the summer of that year there has been a general retreat of the ice limit, and at the same 

 time a warming up of the entire Arctic (Weickmann, 1942). This can be seen best 

 from the mean position of the ice limit from April to August in the two periods 

 1898-1922 (25-year mean) and 1929-38 (10-year mean) (Fig. 125). The especially 

 favourable conditions during the second period are very noticeable when compared 

 with those for the 25-year mean which can be regarded as normal. Bear Island, for 

 example, is normally still surrounded by ice in April and partly also in May. During 

 this second period it was ice-free during all months, and although the northern part 

 of Novaya Zembla is almost never ice-free the ice limit receded during the second 

 period almost to the northern tip in July and during August was only a little south 

 of Franz Josef Land and Wiese Island. 



