The value of evapotranspiration (E. T.) observed by a lysimeter does not correspond to that from a field having large area in natural condition.
So it is worth making study to estimate E. T. from a field by the hydrological method without using a lysimeter.
The presnt data were obtained on sandy soil at the meteorological observation place and the neighbering orchard of the National Institute of Agricultural Sciences in Hiratsuka.
On the sandy land where surface run off does not occur, the following hydrological relations can be obtaind :
P=E+M+G… (1)
G= (G₂-G₁) +H·pa… (2)
where P : precipitation, E : evapotranspiration,
M : change of soil moisture,
G : recharge of ground water,
G1 : ground water flowing in to the locality from upper reach,
G2 : ground water flowing away from the locality to downstream reaches,
H : change of ground water level, pa. : air capacity of soil in the part of changing ground water level.
A small rainfall on dried soil or no rainfall gives, G_??_0. In this case, the decrease of groundwater level corresponds to (G₂-G₁), and a certain relation can be obtained between the various ground water flows and the ground water levels.
This is approximately applicable to the case where G≠0, making it possible to estimate the value of (G₂-G₁) in every month through the year.
Eventually, the evapotranspiration, E on sandy land can be obtained from (1) and (2).
Generally speaking, E of the sandy land is approximately from 600 to 700mm, and the ratio between E and the pan evaporation is almost equal to 0.7. This ratio is smaller in spring and larger in fall than other seasons.

View full abstract