LARGE-SCALE ASPECTS OF ENERGY TRANSFORMATION OVER THE OCEANS 
west-east gradient in Q, is intensified due to the high 
values of Q;, over the western sides of the oceans and 
the low values over the eastern portions. The charts are 
also, in a qualitative manner, quite similar to those for 
Q, except that the latter show no maximum areas of 
sensible heat exchange within the tropical trade-wind 
region corresponding to the areas of maximum Q, in 
this belt. In the North Atlantic in winter, Q, reaches 
values as high as 937 cal cm~? day! within the Gulf 
Stream off the Virginia Capes. In the North Pacific the 
maximum value for Q, occurs in winter within the 
Kuroshio northeast of Formosa, coinciding with the 
area of maximum Q,, where a value of 728 cal em? day! 
is computed, with a secondary area of maximum Q, 
(718 cal em-? day!) between latitudes 35°N and 40°N 
approximately 350 miles east of Japan. The annual 
chart for Q, is given as Fig. 3. 
The seasonal values for Q, arranged by latitude zones 
are given in Table V. These data show that, except near 
1065 
entirely in the form of sensible heat; but in the case of 
the atmosphere only that fraction which appears as Q;, 
is immediately available for heating the air, the re- 
maining energy being latent in the form of water vapor. 
Therefore, in order to obtain a quantity for the atmos- 
phere which is equivalent to the quantity Q, for the 
oceans, it is necessary to consider the heat supplied to 
the atmosphere through the condensation of the water 
vapor. The writer has previously shown that the latter 
quantity can be obtained, completely, from data con- 
cerning precipitation amounts over the oceans [9]. 
The sensible heat equivalent (calories per square 
centimeter per unit time) of the precipitated water is 
given by 
Q, = LP, (16) 
where ZL; is the latent heat of condensation which, in 
this case, is correctly assumed to take place at the 
evaporation temperature t, and P is the rate of precipi- 
TaBLE V. SEASONAL VALUES oF Q, IN DirrERENT LatTiTupE ZONES (in cal em day *) 
North Atlantic North Pacific 
North latitude zone 
Dec.—Feb. Mar.—May June-Aug. Sept.—Nov. Dec.—Feb. Mar.—May June-Aug. Sept.-Nov. 
0° 5°. 136 159 193 166 173 133 197 197 
5°-10° 209 178 146 144 222 210 194 185 
10°-15° 241 204 184 165 244 251 202 193 
15°-20° 250 184 195 198 260 243 235 214 
20°=25° 226 146 160 199 302 229 210 246 
25°-30° 256 165 124 256 327 208 157 251 
30°-35° 284 168 103 242, 329 167 100 245 
35°-40° 437 175 94 293 374 178 74 274 
40°-45° 349 109 33 217 258 97 29 179 
45°-50° 287 94 —13 142 149 68 12 124 
50°-55° 355 122 31 171 156 86 19 143 
55°-60° 381 109 15 181 179 74 27 108 
the equator, Q, is at a maximum during winter and at 
a minimum during summer with the quantities tending 
to be greater during autumn than during spring. The 
seasonal variations are large in middle and high lati- 
tudes but small near the equator. Both oceans show 
secondary minima in the total energy exchange between 
latitudes 45°N and 50°N. 
The author [6] has previously pointed out that during 
winter the locations of the principal Northern Hemis- 
phere frontal zones as given by Petterssen [20] appear 
to correspond quite closely to the zones of maximum 
Q.; in every case the zones of frontogenesis lie to the 
northwest of a zone of maximum Q, with the major axes 
of the zones of maximum energy exchange parallel to 
the principal axes of the frontal zones. 
RATE OF TOTAL HEAT GAIN BY THE 
ATMOSPHERE THROUGH CONVECTION 
The quantity Q, discussed in the preceding section 
represents the rate of total energy loss (or gain) from 
the ocean surface by conduction to the atmosphere. 
From the standpoint of the sea surface, this loss is 
tation. One source of error which may be locally im- 
portant during the winter season is introduced through 
the assumption in (16) that all precipitation is in the 
form of liquid water. However, even if it is assumed that 
all of the water precipitated at the higher latitudes 
occurs in the form of unmelted snow or sleet, the total 
error cannot exceed about 12 per cent. Actually, only a 
fraction of the precipitation occurring over the oceans 
during winter will appear as snow, even at high lati- 
tudes, so the probable error is much less. 
Therefore, in order to complete the energy computa- 
tions for the atmosphere it is necessary to have seasonal 
data concerning precipitation amounts over the oceans. 
However, an examination of the literature reveals not 
only that no seasonal data on precipitation amounts 
over the oceans exist, but that even the few data on the 
annual amounts have been subject to considerable criti- 
cism. In order to obtain approximately correct annual 
precipitation values, the published regional data [14, 
27, 28, 29, 30] were made compatible and brought into 
agreement with Wiist’s more reliable zonal values [39] 
by the assignment of simple correction factors for each 
