1064 
equation [32] directly to determine the monthly aver- 
ages of Q. within the same five areas for which he 
determined Q,.6 He does not attempt to evaluate 
the equation empirically against the available obser- 
vational materials and he resorts to humidity data 
from land stations as a substitute for observations at 
sea. His monthly averages for Q, appear somewhat too 
large during summer (the three-month summer average 
for the five areas proves to be 116 cal em~? day). 
RATE OF TOTAL HEAT LOSS FROM THE 
OCEANS THROUGH CONVECTION 
The only method for directly estimating the total 
convective heat loss Q, from a water surface appears to 
be the one recently developed by Montgomery [17]. 
He presents a detailed discussion of methods for esti- 
mating the total eddy flux of heat from a moist surface 
East 180 West 
Ws 
ay 
MARINE METEOROLOGY 
tion of the vertical component of velocity, then the 
unit-area eddy flux of heat is (pVzh:) and 
(pVzhi) = Gh = Glen = Ws), (14) 
where k is the eddy diffuswity (Richardson), Cpa is the 
isobaric specific heat of dry air, 7. is the mean equiva- 
lent temperature at the surface, and 72 is the same 
quantity at a chosen, short distance above. For this 
purpose, the equivalent temperature of a sample of air is 
defined as the temperature of dry air having the same 
enthalpy. 
Montgomery’s method, however, has not yet been 
made applicable to the type of observational materials 
generally available in the marine climatic record. Fur- 
thermore, the meteorologist generally is more interested 
in the separate fractions of the energy exchange, Q, 
Fic. 3.—The annual values of the rate of total energy loss from the ocean surface through convection (Qa = Q; + Q.) over the 
North Atlantic and North Pacific, expressed in calories per square centimeter per day. 
by combining the effects of heat transfer by conduction 
and heat transfer by evaporation (or surface condensa- 
tion) and, after deriving several general expressions, he 
arrives at a more restricted equation which yields good 
approximations for Q, under certain conditions. These 
conditions are that the humidity and temperature ob- 
servations be obtained within 10 m of the water surface 
and that the pressure should be within about 10 per 
cent of standard pressure. He shows that if the specific 
enthalpy, including the latent heat of water vapor, is 
designated by h,, and if V; is a specially defined fluctua- 
5. The equation used is: 
590 X 2.15 X 102(em — ea)Wa 
Zo 2 
8.64 E (cise) | + 0.16Wa 
0 
Qe =z (18) 
cal em~ day, 
where LZ; = 590 cal g7. 
and @., than he is in their sum. Nevertheless, the 
method could facilitate computations of the rate of 
total “contact” heat loss from the sea surface and thus 
be useful to the oceanographer when he finds it con- 
venient to evaluate the sea-surface heat loss for special 
analytical purposes and when it is possible for him to 
obtain controlled meteorological observations. 
The only available large-scale computations of the 
total energy exchange between the ocean and atmos- 
phere have been presented in chart form for the North 
Pacific and North Atlantic [6, 10] and are based on the 
sum 
Qa = Qn =P Q.. (15) 
Therefore, they represent the summation of computa- 
tions which have already been described in the pre- 
ceding two sections. ; 
In their general configuration the charts for Q. are 
quite similar to the charts for Q., except that the 
