As we have seen, all these values vary within wide limits, even in the same region. This is 

 particularly apparent in the Atlantic regions of the arctic. During some years, the number of 

 freezing degree-days on Dickson Island (Ostrov Dikson) departs from the average by more than 

 15 per cent. However, the amplitude of variation of the number of freezing degree-days decreases 

 eastward and northward. The summer regime fluctuates just as greatly in individual regions, but 

 its amplitudes also decrease northward and eastward. 



For general considerations, however, let us assume that both the summer and winter regimes 

 remain unchanged in each individual region, and let us investigate the spatial distribution of these 

 elements . 



In the region of the Arctic Basin beyond the limits of the continental shoal, i.e., the main 

 region of perennial ice, the number of freezing degree-days evidently fluctuates within the limits 

 5, 000 and 8, 000. The summer regime varies more intensely. Thus, while the ice melted about 

 one meter in summer in the region of drift of the Tram and the Sedov near the Greenland Sea, 

 the ice melt did not exceed half a meter in the region of the pole, as observations made by station 

 "North Pole" showed. Evidently, summer melting amounts to only a few tens of centimeters in 

 the region between the North Pole and the Arctic Archipelago, where the Atlantic influence scarcely 

 penetrates . 



Thus, over the entire Arctic Basin, the number of freezing degree-days does not vary by 

 more than a few tens of per cents . Summer melting differs by hundreds of per cents from region 

 to region . 



However, when the ice is quite thick, even considerable changes in the number of freezing 

 degree-days have little effect on the ice thickness. Therefore the summer regime is the main 

 factor determining the maximum thickness of perennial ice accretion. 



Let us assume the number of freezing degree-days to be constant and let us use the simpler 

 formula. (6) for further discussion. Substituting R - 6000 in this formula, we find that during 

 summer melting 



AI = 100 cm, -Z"max ^ 265 cm, 



AJ = 50 cm, Jjjjg^ =480 cm, 



AI = 20 cm, -f max ~ 1185 cm. 



With the same number of freezing degree-days and with 10 cm summer melting, the maximum 

 ice thickness increases to almost 24 m. This evidently explains in part the formation of thick peren- 

 nial shore ice along the northern coasts of Greenland and the shelf ice along the coasts of the 

 antarctic . 



Thus, there is no doubt that with time the ice thickness gradually approaches the average for 

 a given region. The question arises : how many years does this take ? The corresponding formu- 

 las would be too complex, but a subsequent change in the ice thickness can be obtained easily using 

 the given formulas and the graph. As an example, let us assume that toward the beginning of win- 

 ter, ice of different thicknesses is brought into a certain region, where the number of freezing 

 degree-days is 6000, and summer melting 100 cm, and let us trace the changes of thickness from 

 year to year. Let us consider 265 cm to be the maximum ice thickness in the given region, in 

 other words, we shall use formulas (3) and (5) in the computations, ignoring the amount of summer 

 warming, which as yet is little known. I computed table 76 on the basis of just such assumptions. 



230 



