Aa the mixing model is developed, each layer will .yield a certain 

 amount of potential heat loss, q^, and eventually, when the temperatures 

 are low and the densities are high, each layer will yield a certain masa 

 of potential ice (£ ) associated with convective mixing down through the 

 corresponding layer. Thus it is possible to construct the very useful 

 "potential curve" (Figure 2) from corresponding 



I &Z=l,(h) and Q T (h)=Q T (h)-Q , wh ere V / V 2 ' (3) 



The potential curve is plotted, Q T starting from zero within the layer 

 where ice first appears. The remainder of the %% Q , which is lost before 

 ice is formed, determines the date of ice formation. The reason for 

 taking Q T as zero at this point becomes evident when one considers the 

 lower limits of the integrals in equation (13). Obviously some question 

 arises as to the exact quantity of heat that has been removed at the 

 time ice is first formed. There is, of course, some error introduced, 

 but it is not large as long as the water temperatures are near the 

 freezing point and the water column is divided into layers of suffi- 

 ciently small depth, for example five to ten meters. Recent work on 

 the same type of model as Defant's has yielded exact values of the 0^. 

 which must be removed before ice is formed, thus determining the proper 

 point to start accumulating the Qjp for the potential curve (Brown, 

 1954) . 



In order to illustrate the computation of the ice potential and 

 the drawing of the potential curve, a typical oceanographlc station 

 is used as an illustratioa. Station 37, at 66° N, 58° W, was occupied 

 on 3 October 1952. A ploc, of the oceanographlc variables at this sta- 

 tion is given in Figure 1, and the complete numerical data are shown 

 in Table 1. 



TABLE 1 

 Oceanographic Data for Station #37 

 (66°N, 58°W), 3 October 1952 



Depth 



Tempe rature 



Salinity 

 %o 



Density 

 Anomaly (°t) 



m. 



L. 



gm/liter 







0.56 



32.37 



25.98 



5 



0.10 



32.36 



26,00 



10 



0.61 



32.39 



25.99 



20 



0.73 



32.47 



26.05 



30 



-0.41 



32.78 



26.36 



50 



-1.65 



33.08 



26.64 



75 



-1.61 



33.69 



27.13 



99 



-2.13 



33.69 



27.14 



148 



-0.52 



34.04 



27.38 



197 



1.52 



34.21 



27.40 



296 



0.08 



34.34 



27.59 



394 



-1.71 



34.52 



27.81 



