P. BAFFIN BAY-DAVIS STRAIT LONG-RANGE ICE FORECAST FOR SEASON 1952-53. 



During early autumn, 1952, oceanographic obse nations were taken at 

 the locations shown on Figure 4» The method explained previously was 

 employed to forecast the ice formation and growth of these locations. 

 The dates of ice formation were computed and are shown beside each sta- 

 tion. Since the thermohaline structure at some of the stations was such 

 that no ice would be formed, it was possible to delineate the theoretical 

 extreme ice boundary shoivn as a heavy line in Figure 4« This boundary 

 agrees very well with the boundary which was actually observed in lat© 

 March and early April 1953* (shown as a dashed line in Figure 4) except 

 for the occurrence of ice south of Stations 39 and 40. It is possible 

 that ice which was observed at these two stations was not formed there, 

 but had drifted from nearby areas of ice formation,, The dispersion of 

 ice near the observed boundary in this area indicates that the ice 

 actually was formed outside the area. 



CONCLUSIONS 



The methods of Zubov and Defant for ice potential calculation have 

 proved in practice to give reasonable answers for open seas and for in- 

 shore areas where local variations in the physical properties of the 

 water are not large. In harbors and areas where runoff is important, 

 changes in salinity and density are so rapid that it is difficult to 

 attach a meaning to an average value of these parameters. Hence any 

 ice potential calculation based on one or a few observations in a harbor 

 may give a potential ice growth which is unreasonably high or low for a 

 single season. 



The forecast of ice growth based on the ice potential can be used 

 only during tho period when ice thickness is increasing,. No theory has 

 been included which accounts for decreasing ice thicknesses during the 

 breakup period. Indeed, no mathematical theory has yet been developed. 

 for the breakup of ice. 



Many refinements of the method presented in this paper are possible. 

 Some of them are discussed in a paper by Brown (1954). The method described 

 here was used to forecast during the 1952-53 ice seasonj the forecast proved 

 to be satisfactory. The most 'important advancement discussed in this report 

 is the use of the ice potential as a basis for forecasting ice growth. 



13 



