Although the actual values of these variations are not known 

 with certainty, tentative values may be given. A study of scale error 

 in BT's made by this Office in 1959 revealed an error of approximately 

 0,2° F. The reading error is estimated at about 0.1° F. The inter- 

 diurnal variation is small, probably no more than 0.2° F, but is con- 

 centrated near the surface because of wave action. Thus, after allow- 

 ing for these sources of variation, Table 3 shows that the major source 

 of variation must be presvimed to be advection. 



TABLE 3 



Hour (Z) 



Hourly Standard Deviations of Temperature After 

 Removal of Variations Other than Advection (° F) 



Surface 



20 ft 



kO ft 



6o ft 



80 ft 



100 ft 



0000 



.59 



.69 



.72 



.83 



.79 



.91 



0100 



.57 



.57 



.61 



.68 



.77 



.82 



0200 



.57 



.58 



.61 



.63 



.68 



.73 



0300 



.59 



.61 



.63 



,66 



.70 



.7^^ 



oUoo 



.59 



.59 



.59 



.61 



.65 



.68 



0500 



.65 



.65 



,6k 



.67 



.69 



.82 



0600 



.62 



.61 



.63 



.63 



.73 



.88 



0700 



.h5 



,h6 



M 



M 



.51 



.57 



0800 



.58 



.57 



.58 



.61 



.61 



.63 



0900 



.58 



.57 



.56 



.55 



.52 



.62 



1000 



.66 



.68 



.68 



.69 



.77 



.86 



1100 



.69 



.68 



.69 



.72 



.82 



.98 



1200 



.68 



.69 



.72 



.75 



.82 



.92 



1300 



.69 



.73 



.74 



.78 



.86 



.9h 



i4oo 



.58 



.60 



.60 



.65 



.72 



.81 



1500 



.58 



.59 



.58 



.63 



.73 



.84 



1600 



.51 



.46 



.45 



.61 



.61 



.72 



1700 



.58 



.9^ 



.55 



.60 



.75 



.81 



1800 



.71 



.65 



.66 



.68 



.74 



.81 



1900 



.59 



.55 



,62 



.57 



.76 



.86 



2000 



.67 



.65 



,66 



.78 



.81 



.83 



2100 



.75 



.70 



.67 



.76 



.82 



.88 



2200 



.77 



.77 



.72 



.73 



.76 



.83 



2300 



.77 



.77 



.72 



.76 



.73 



.83 



A standard deviation of 0.1° F is equivalent to I30 gra cal/cm^ in 

 a layer 20 feet thick. In Figure 3 it can be seen that the temperature 

 gradient for the top 20 feet decreased from 0.1° to 0° F between 230OZ 

 and 0200Z, or three hours, corresponding to a heat change of 22 gm 

 cal/cm^/hr. The gradient in the upper kO feet also changed from 0.1° 

 to 00 F between OlOOZ and 05OOZ, or four hovtrs, a rate of change of 32 

 gm cal/cm^/hr. Similarly the gradient in the top 60 feet was reduced 



18 



