RESTRICTED 



width of the channel previously existing. If the wind is strong 

 enough, hummocks will be produced along a line approximately 

 perpendicular to the wind direction. 



The air temperature as influenced by the wind also has an effect on 

 the grouping of ice. If the wind which has regrouped ice is a cold 

 one, the lowered temperature may cause further freezing, so that the 

 masses may become joined by new formation. In this case the ice 

 would not be so readily broken up and regrouped by a change of wind. 

 On the other hand, if the weather is mild, cakes brought together by 

 a change of wind will not join together. 



The rate at which the different floes travel is not so much dependent 

 upon the size and depth of the floe as upon the nature of its surface. 

 Since the pack is made up of a conglomeration of young ice, old floes 

 which have been subjected to pressure, and icebergs, it varies radically 

 in resistance to wind and current. Surface irregularities, such as hum- 

 mocks and pressure ridges, act as sail areas, and the rate of movement 

 of a floe depends to a certain extent on the amount of hummocking in 

 proportion to the area and weight of the floe. iVs a result of previous 

 pressure, hummocked floes in turn become the cause of still further 

 pressure. When two floes are moving at different rates, either the dis- 

 tance between them is increased and a lane produced, or the distance 

 between them is decreased and the floes brought into physical contact. 

 In gaining momentum, larger floes will accelerate more slowly, but once 

 underway, they will carry their way long after smaller floes have 

 stopped moving. In the early stages, therefore, the large heavy floe 

 will be charged by smaller floes overtaking it ; in the later stages, it will 

 itself be the attacker of smaller floes in its path. Because of their size 

 and weight, the smaller floes will be disrupted and the floe surface 

 materially modified, thereby creating new possibilities of further dif- 

 ferences in speed. 



LAWS GOVERNING DRIFT 



While the general direction of the drift of icebergs over a long pe- 

 riod of time is known, it may not be possible to predict the drift of an 

 individual berg at a given place and time, for bergs lying close to- 

 gether have been observed to move in opposite directions. They move 

 under the influence of the prevailing current at the depth to which 

 they are submerged, which may often be in opposition to the existing 

 wind and sea or surface drift. The International Ice Patrol has had 

 considerable success in predicting the drift of bergs off Newfoundland 

 by determining the surface current patterns through the methods of 

 physical oceanography. 



RESTRICTED 33 



