Evaporation from the Surface of the Sea and the Water Budget of the Earth 235 



observations, and the product of these quantities gives only an approximate idea of 

 the annual precipitation amount. In addition to these observations there are also 

 available precipitation records in coastal areas or from islands, but these are often 

 strongly affected by local topography so that there is usually a greater precipitation 

 amount than over the neighbouring oceanic regions. Therefore, the utmost critical 

 inspection and caution is needed in the use of records of coastal and island precipita- 

 tion for the construction of isohyeths for the oceans. However, with suitable allow- 

 ances, better numerical estimates can be made of the evaporation over the sea so that 

 the mean evaporation amount allows in return an estimate of the precipitation amount 

 over the oceans. 



The reverse applies on the land ; here the determination of the mean evaporation 

 involves almost insuperable difficulties, but a dense network of precipitation stations 

 can give the mean precipitation with suitable accuracy. In this way a complete picture 

 of the water budget of the sea, the land and of the total Earth can be obtained. Such a 

 summary however, does not give absolutely correct annual values since the accuracy 

 of each item in the water budget is not very great. It is thus of more importance to 

 enclose the different values within the most narrow limits possible so that the indi- 

 vidual values either support or exclude each other, in order to obtain maximum 

 probability. Table 90 summarizes the essential characteristics of the water budget 

 of the Earth. 



Table 90. Most probable water budget of the Earth 



The following points may be noted. The figures for precipitation are based on those 

 obtained by Meinardus (1934) from a most detailed investigation of the distribution 

 of precipitation over the Earth based on the precipitation charts for the oceans pre- 

 pared by Schott. The values for the land were taken directly. For the sea a correction 

 was applied based on the criticisms made by Wtisx (1936) of these charts. Meinardus 

 found a total precipitation over the oceans of 411-6 x 10^ km^ which corresponds to 

 a mean annual rainfall of 114 cm/year. The mean precipitation over the oceans 

 would thus be 1-7 times greater than over the land (67 cm/year), which can hardly 

 correspond to actual conditions. No doubt too much consideration of island and 

 coastal precipitation must have appreciably raised the precipitation amount over the 

 oceans. Calculation of the precipitation over the sea from the total evaporation over 

 the sea of 100 cm/year = 361 x 10^ km/year and the inflow from the land (Fritsche, 

 1906) of 37-1 X 10^ km^/year gives a correction factor of 0-79 for correcting the rain- 

 fall at coastal and insular stations to values for the undisturbed sea surface. The 

 coastal and insular values are thus on the average raised by about 20% by the effects 



