EVAPORATION FROM THE OCEANS 
By H. U. SVERDRUP 
Norwegian Polar Institute, Oslo 
INTRODUCTION 
The problem of evaporation from the oceans has 
received little attention compared to that directed to- 
wards the study of evaporation from lakes and land 
surfaces. One of the reasons for this is that the results 
of the latter, being of great interest to the hydrological 
engineers, are of immediate economic importance. Fur- 
thermore, until recently, weather forecasting has been 
concerned primarily with forecasting over land areas, 
for which purpose it has been sufficient to know that 
air which has travelled over the oceans contains a 
great deal of moisture, but it has not been necessary to 
ask how this moisture has been added to the air. With 
the rapid development of air routes across the oceans, 
accurate weather forecasts for ocean areas are needed, 
and in this connection the question of how air masses 
are modified when passing over water becomes impor- 
tant. When dealing with this modification, knowledge 
of evaporation as a function of the meteorological con- 
ditions near the sea surface becomes indispensable. 
This development is reflected in the very manner in 
which the question of evaporation from the oceans 
has been dealt with during the last seventy-five years. 
Up to about 1930, most efforts were directed towards 
establishing the annual evaporation in different lati- 
tudes in order to examine the general water budget of 
the atmosphere, but during the last two decades the em- 
phasis has been placed on examinations of the evapora- 
tion under given meteorological conditions. Some ad- 
vances have been made, but these studies are still in 
their infancy. 
EVAPORATION FROM THE OCEANS 
DETERMINED BY EXTRAPOLATION 
OF VALUES FROM LAND 
In his discussion of the water budget of the atmos- 
phere, Briickner [3] tried to find the evaporation from 
the oceans by extrapolating values derived from ob- 
servations on land. He used observations from coastal 
regions particularly and obtained an average evapora- 
tion from all oceans between 70°N and 70°S of 110 
= 15 cm per year. He suggests that this value may be 
somewhat high because he has probably assumed too 
large an evaporation between 10°N and 10°S (see 
Table I). It is of interest to observe that, besides other 
material, Briickner made use of Russell’s map of 1888 
of the evaporation in the United States, from which he 
obtained the following values at the Atlantic and Pacific 
coasts: 
North Latitude Evaporation 
(deg) 
(cm per year) (mm per 24 hr) 
25-30 120 3.3 
30-40 98 2.7 
40-45 82 2.2 
45-50 51 1.4 
A comparison with the evaporation data in Table I 
shows that these values are in remarkable agreement 
with results that have been obtained by entirely differ- 
ent methods. 
DIRECT MEASUREMENTS OF EVAPORATION 
AT SEA 
History. Measurements of evaporation from pans 
have been carried out on board ship in limited number. 
The first observations of this type were made by Mohn 
[15] during the Norwegian North Atlantic Expedition, 
1876-78. Mohn used an evaporation pan that floated 
in a large container and determined the evaporation in 
12 hours by adding enough fresh water to the pan, 
at the end of the time interval, to bring the weight of 
pan and water back to its original amount. 
Tn subsequent experiments the evaporation has been 
determined by means of the increase in salinity during 
the period of exposure; in the earlier ones the salinity 
was determined by aerometric measurements of den- 
sity and in the later ones by chlorine titration. In 
several cases pans have been exposed simultaneously 
to wind and sunshine, to sunshine only, and to wind 
only, in order to examine the effects of exposure. How- 
ever, the results, which have been extensively discussed, 
have all been derived from observations obtained with 
the pan placed in as unprotected a position as possible. 
Nearly all the measurements referred to in the last 
paragraph were made in German investigations during 
the years 1892 to 1914, and most of them between 1908 
and 1914. They have been summarized and discussed 
by Wiist [29], who gives a list of the pertinent literature. 
Several measurements [10] were made during the last 
cruise of the Carnegie, but it is doubtful if such observa- 
tions will be continued in the future because the results 
are difficult to interpret and because other methods of 
approach have met with fair success. 
Discussion. Wiist [29] showed that the evaporation 
from pans on board ship depended principally on two 
variables, the wind velocity w, and the evaporation 
potential defined as 
P = (1/273)(e; — @). (1) 
In (1), T represents the absolute temperature of the 
water in the pan, e, the vapor pressure at the water 
surface, and e the vapor pressure in the air measured on 
board ship. The wind velocity, w, is the wind velocity 
measured or estimated on board ship. 
Regarding the vapor pressure at the water surface, 
it should be observed that the vapor pressure over 
sea water is somewhat lower than that over distilled 
water. If the latter is represented by ea, 
és = ea (1 — 0.000535), (2) 
1071 
