The seasonal trend in Wyrtkl's Q(E) values 

 is consistent with those during 1963 to 1965. 

 Again, in view of the variability in Q(N) near 

 the Equator during 1963 to 1965, the trend of 

 Wyrtki's month-to-month values of Q(N) is not 

 expected to be similar to either of the 2 years. 



In contrast to the good agreement in average 

 values of Q(E) and Q(N) in the trade wind zone, 

 those in the equatorial region differ consider- 

 ably. The 1963 to 1965 average Q(E) is about 

 100 cal. cmT^day"', whereas Wyrtki's annual 

 average Q(E) is about 200 cal. cmT^ day"'. Sim- 

 ilar mean values of Q(N) are about 240 cal. 

 cm7^day"'and 160 cal. cm7^ day"' , respectively. 

 The difference in Q(E) is primarily due to the 

 use of different drag coefficients in the evapor- 

 ation equation. The average wind speed at lat. 

 2" N., long. 157° W., July 1963 to June 1965 

 (table B) was 4.7 m. seer' for which the drag 

 coefficient is 0.96 x 10"'. Wyrtkl used a con- 

 stant drag coefficient, Cp = 1.55 x 10"'. Thus, 

 the difference in drag coefficients accounts for 

 the major portion of the difference in the evap- 

 oration rates and therefore also In the differ- 

 ence of the net heat exchange across the sea 

 surface. 



Within the limitation discussed above, the re- 

 sults for the two sample areas show that the 

 heat exchange processes computed here are 



consistent with those based on long-term aver- 

 age data presented by Wyrtki. 



Interseason and Interyear Comparisons 



In a previous section it was shown that the 

 manner of processing affects mainly the heat of 

 evaporation and, therefore, also the results for 

 the net heat exchange across the sea surface. 

 Figures 3 and 4 show that, regardless of the 

 manner of computation, the month-to-month 

 trends in the heat of evaporation and the net heat 

 exchange across the sea surface are similar. 

 Here, then, the importance of absolute magni- 

 tudes in interseason and interyear comparisons 

 is examined. 



Results for the area, lat. 25° to 29° N., long. 

 140° to 144° W., which were previously used, 

 are also used to compute the differences be- 

 tween principal extremum values of Q(E|) and 

 Q(E2)--see figure 3--and Q(N|) and QCNg) (fig. 

 4). These differences, listed in table 3, show 

 that the trends of change (increasing or de- 

 creasing) are the same whether the exchange 

 processes are based on daily or mean monthly 

 meteorological properties. The magnitudes of 

 change are also similar but agreement tends 

 to be better in spring and summer than during 

 autumn and winter. 



400 



T — I — I — I — I — I — \ — r 



Q(E) 



SECKELI 1963-64) 



SECKEL (1964-65) 



WYRTKI 



S N D 



MONTHS 



ASONDJFMAMJ 

 MONTHS 



Figure 6. — The heat of evaporation, Q(E), and the net heat exchange 

 across the sea surface, Q(N), at lat. 2° N. , long. 157° W. , July 

 1963 to June 1965 (from table B) , and interpolated values from 

 climatic charts by Wyrtki (1966). 



