the sensible heat plus the ].aterit heat of .forraa,tion of ice„ Trie jxo- 

 cedurc for drawing ice climatological charts consists of two parts „ 

 First, the total he3,t loss for 3 plven ice potential pnd sensible heat 

 loss (assuming the cross section of the water column to be- 1 cKo^) is 

 found by adding 72 fOT. cal./cmr* of potential ice to the sensible heat 

 losso The quantity 72 gm. cal./ cm. 3 is derived by assuming the latent 

 heat 01" sea ice to be 80 gm. ca.l./gm. and the density of sea ice to , 

 be 0.9« Second^ it is necessary to postulate a heat loss which is 

 realistic in relation to the annual heat budget of the chosen area.. 

 After these two factors are determined, the chart is readily drawno 

 An example of this type chart is shorn in figure 1, which presents 

 isolines for the Beaufort Sea derived from data collected in summer 

 l?5lo Detailed discussion of al]. charts will be found in the follow- 

 ing;^ sectiono 



Limitations of this tjjpe chart are found in the requirement that 

 the ice potential be determined for open water areas where the effects 

 of advection are relatively minor, and in the often unrealistic as- 

 sumption that total heat loss is constant oyer a larg;e area. Since 

 the ice potential calculation assumes, negligible advective changes, 

 it is necessary to use hydrographic statioriS made in deep water av7ay 

 from ins?iore runoff » However, in the Beaufort Sea, it is difficult 

 to navi&;ate inside the Polar Pack, and most available data are in- 

 evitably derived from observations made in shallow inshore areas. 

 As a practical matter, it has proved necessary to use a minimum depth 

 of 50 meters in order not to restrict the information unduly. Where 

 more data are available, the minimura depth could be increased to 100 

 meters or even to 200 meters, with a consequent increase in reliability. 

 The assumption that heat loss is constant over the area is nearly 

 correct when applied to the Beaufort Sea, since its latitudinal extent 

 is only 5 dep-rees . However, if a simlar study were m.ade for Baffin 

 Bay, .for example, different heat losses n-dght be required for different 

 latitudes since the bay extends over a relatively wide range of latitude » 



The choice of a standard heat loss is made on the basis of a 

 reasonable heat loss, which will give average ice thickness values 

 corresponding to the total growth during a mnter. The reason for 

 choosing a large heat loss is to make sure that convection proceeds long 

 enough to penetrate Into laj^ers which are sufficiently deep to eliminate 

 ] ocal temperature variations, leaving only water m.ass variations. VJhen 

 a total heat loss of 20 kg. cal„/cmo'^ is used as in figirre 1, thiclmesses 



range .from a maximum of 220 cm. to a jninlm.uia of -LO cm. - Accordin.s to 

 Zubov (1938), average annual ice gro-.?t'h in the Arctic is about 200 cm. 

 Thus, the total heat loss used is reasonable, although arbritrary. 



- Negative ice potentials sx^e purely formalistic mathematical expressions 

 indicating that insufficient heat has been lost to prod.uce ice formation. 



-2- 



