THE HYDROLOGIC CYCLE AND RIVER FORECASTING 
where P is average precipitation, H is the water re- 
turned to the atmosphere by evapotranspiration, AS 
is the increment of water added to or removed from 
storage in the basin, and # is the runoff volume leaving 
the basin via the streams. 
To balance equation (1) it must first be possible to 
determine the true average precipitation over the basin. 
This immediately involves problems in sampling, unless 
radar techniques can measure the integrated total over 
the basin. In level terrain, where the occurrence and 
distribution of precipitation are more or less random, 
the planning of an adequate precipitation-gage network 
depends on an understanding of the structure of storms 
ROM 
VEGETATION 
SA 
FROM LAKES 
AND RIVERS 
\ 
Fre 
TO SOIL “a 
= ____ TG 
TO LAKES & RIVERS r 
TO OCEAN 
et PRECIPITATION 
RW EVAPORATION pee 
1049 
titative solution of the water balance, is the unsolved 
problem of evaporation from land surfaces. 
For many years the only attempt to solve these 
problems at the practical level has been by the use of 
evaporation pans. Obviously, there is a vast difference 
between the evaporation from a pan 4 ft in diameter 
and 10 in. deep and that from a large reservoir. It 
has been generally accepted, on the basis of experi- 
ments by Rohwer [11], that the evaporation from 
reservoirs averages about 0.7 of that from the U.S. 
Weather Bureau Class A pan. It is known, however, 
that there are considerable variations in this coefficient 
depending on size, depth and exposure of the reservoir, 
MOIST AIR 
| 
INFILTRATION OF 
GROUND WATER 
Fie. 1.—Schematic diagram of the hydrologic cycle. 
and particularly of the variations of depth with dis- 
tance from the storm center. In mountainous regions 
the effect of topography must also be considered. Spreen 
[13] has suggested the use of empirical correlations 
between average annual precipitation and topographic 
parameters. Maps prepared from such correlations 
would provide a guide to the proper location of stations 
for better evaluation of annual or seasonal precipita- 
tion. Similar studies of individual storms in rugged 
terrain are still needed. 
Evaporation. The study of evaporation has presented 
more practical difficulties than most other items in the 
hydrologic balance. Storage reservoirs are designed to 
conserve water which would otherwise go to waste 
during the rainy season and to make it available when 
needed. There have been numerous instances in which 
the increased evaporation from the enlarged water 
surface created by the reservoir has exceeded the volume 
of usable water gained by storage, thus making the 
reservoir a liability rather than an asset. Less critical 
in many respects, but nevertheless preventing the quan- 
and time of year. Any quantitative application of pan 
data to soil evaporation is made difficult by the varying 
evaporation opportunity with variations in soil mois- 
ture. 
A technique for determining evaporation either by 
direct measurement of outgoing moisture or by compu- 
tation from the energy balance [10], turbulent transfer 
[14, 15], or other theory is urgently needed. However, 
evaporimeter data cannot be ignored, for they are all 
that will be available for many years. It has been 
shown that a good correlation exists between pan evap- 
oration and readily available meteorological data— 
temperature, dew point, and wind [7, 9]. Through such 
correlations, available evaporimeter data may serve 
to guide the extrapolation and application of data 
collected by other means. A more rational method for 
adjusting pan data to reservoir conditions may be 
found if water temperature, wind, and humidity are 
observed at both the pan and the reservoir site. One 
difficulty hampering a study of this problem is the 
fact that almost no completely watertight reservoir 
