834 POUNDER [sect. 7 



reliable because of the many uncertain factors involved in the calculations. A 

 good average figure is about 3.5 x 10~3 c.g.s. units. 



4. Electrical Properties 



Pure ice can be considered as a readily polarizable, poor conductor or as a 

 dielectric with a large loss factor. Mantis (1951) and Dorsey (1940) summarize 

 the extensive investigations of the electrical properties of ice. Sea-ice, with its 

 inclusions of ionic salt solutions, would be expected to show a higher electrical 

 conductivity and a markedly anisotropic conductivity because of the crystal 

 structure discussed earlier. The only extensive investigation, that of Dichtel 

 and Lundquist (1951), confirmed these conclusions. Pounder and Little (1959) 

 showed that the resistivity in annual sea-ice increases in the later months of 

 the winter owing to brine drainage. 



5. Growth and Disintegration of an Ice Cover 



Here it is proposed to summarize the physical factors and give references to 

 some of the many papers with empirical or theoretical growth and disintegra- 

 tion equations. Two distinct problems exist for the cooling period of fall and 

 winter : first, to predict the date of first ice formation and, second, to predict 

 the rate of ice growth. The first problem depends on both oceanographic and 

 meteorological conditions. Its solution is usually carried out using the "ice 

 potential" method introduced by Zubov (1938). Data from oceanographic 

 stations are used to find the depth of the thermocline and the salinity and 

 temperature distribution above it. Models of convective mixing in the water 

 and of heat transfer to the air are assumed. Climatological data permit a 

 prediction of the rate at which freezing exposure will accumulate and hence a 

 prediction of the date of first ice formation (Simpson, 1958). The freezing 

 exposure, Et, sometimes called the degree-days of frost, is the product of time 

 in days and the average difference between the air temperature and the freez- 

 ing point of sea-water. 



Once a cover of sea-ice forms the processes of heat removal change. The 

 latent heat of fusion of the sea-water is the principal heat source and contribu- 

 tions from the bulk of the water are small. In the Arctic, with the very light 

 snowfall, conduction through the ice cover is the limiting factor on growth. 

 Elsewhere, the amount of snow cover may be very important ; Holtsmark 

 (1955) measured growth rates for varying snow thicknesses. Stefan (1891) set 

 up the problem of ice growth mathematically and for a particular case obtained 

 a solution for the ice thickness h in the form, 



/i2 = {2klLd)Et, (3) 



where Et is the freezing exposure since first ice formation, k, L and d are the 

 thermal conductivity, latent heat and density of the ice respectively. This 

 solution ignores the snow cover, any radiation imbalance and heat transport 



