SECT. 2] SMALL-SCALE INTERACTIONS 75 



From the drag-coefficient formula (22) which was adopted to fit the observa- 

 tions in Fig. 6, we can estimate the wind-profile coefficient Fm via equation (29). 

 The resulting Fm values for heights 4 m and 8 m are as follows : 



We can now combine the estimates of profile coefficients given above to 

 derive the bulk heat flux and evaporation coefficients, h and d, using (27) and 

 (28) for the neutral case. Let us restrict attention to evaporation, since the 

 measurements are not yet sufficiently refined to detect any difference between 

 values of Fh and Fe and, therefore, between h and d. Humidity measurements 

 have commonly been taken at a height of 4 m and winds at around 6 to 10 m. 

 Taking Fe (4 m) = 0.1 and (for a wind speed of 5 m/sec) Fm (8 m) = 0.092, then 

 (28) gives the bulk evaporation coefficient, (i = 0.00155. 



This value lies between two earlier estimates of evaporation from the oceans 

 (see Sverdrup, 1951): the average result from the Meteor Expedition is 20% 

 lower, while the result obtained by Jacobs from oceanic heat balances is some 

 37% higher. Of particular interest is a comparison with the result from an 

 intensive investigation at Lake Hefner, where Marciano and Harbeck (1954) 

 evaluated the bulk aerodynamic coefficient by matching against accurate 

 water-budget data. The Lake Hefner value is in fact 5% higher than that 

 quoted above, that is to say, the two are virtually equal. At two other lakes 

 (Harbeck et al., 1958; Webb, 1960), the Lake Hefner value has also been 

 confirmed ; and it seems to be indicated that this value of the bulk coefficient 

 is applicable to water expanses of a wide range of sizes from the open sea down 

 to lakes a mile or so across. Russian investigations reported by Budyko, 

 Berliand and Zubenok (1954) indicate again similar values: an approximate 

 value quoted for lakes is 3% higher than that derived here, and a value they 

 derive from heat-budget calculations over the oceans is 24% higher. 



Some discrepancy between the ocean heat-budget values and that derived 

 from the neutral profile is to be expected on account of the stability effect, the 

 former values having been deduced for ocean regions over which the average 

 condition is one of slightly unstable stratification. Furthermore, it is to be 

 remembered that the heat-budget values are obtained on the assumption, 

 needed to be able to utilize climatic averages, that 



{qs-qa)Ua = {qs-qa)Ua, 



where the bars denote mean values taken over an extended period of time. 



