Estimating Potential Evapotranspiration by the Priestley-Taylor Model

Abstract

The model of Priestley and Taylor (1972) for evaporation from saturated surface was examined in a radish field. This model contains a parameter (α) corresponding to a multiple of the equilibrium evapotranspiration. Since α is affected by some meteorological factors, it is important to make clear the characteristic of α in relation to these factors. In this paper, the characteristic of α is focused on relativity to the net radiation and the air temperature. The results can be summarized as follows:The amount of evapotranspiration calculated by using daytime meteorological data agreed closely with that measured by the lysimeter. However, the calculation by using 24 hour data was underestimated. It seemed that the underestimation depended greatly on the decreasing net radiation caused by long wave radiation exchange at night time.A seasonal variation of α which was calculated by using measured evapotranspiration and meteorological data was examined. The seasonal variation of α which was calculated by using daytime data cannot be recognized. However, α which was calculated by using 24 hour data increased after around October apparently. This increase in α was due to the rate of decreasing net radiation which was defined by Eq. (4). In case of use of α modified by Eq. (4), even if 24 hour data is used, the calculation is not underestimated.Finally, α in the case of daytime data was modified in connection with air temperature. The results indicated that α was affected very little by air temperature. The use of modified α brought about the rise of accuracy by only 0.8%.