A comparison among different modified Priestley and Taylor equations to calculate actual evapotranspiration with MODIS data

Priestley and Taylor's (1972) equation to estimate evapotranspiration (ET) stands for its simple form and data requirement. Although the original equation was developed for saturated surfaces, it has been widely extended to unsaturated surfaces. In this paper, different hypotheses to modify Priestley and Taylor's equation for unsaturated surfaces are compared. In general, ET models for unsaturated surfaces assume that the process is ruled by the available radiant energy and surface moisture or the atmospheric conditions or both surface and air states. The results presented here suggest that both atmospheric and surface variables should be jointly parameterized in order to obtain estimates with errors lower than 20%, which are common in ground observations. While surface condition parameterizations alone, such as those proposed by Barton (1979 Barton, I.J. 1979. A parameterization of the evaporation from nonsaturated surfaces. Journal of Applied Meteorology, 18: 43–47. [Crossref] , [Google Scholar]) and Jiang and Islam (2001 Jiang, L. and Islam, S. 2001. Estimation of surface evaporation map over southern Great Plains using remote sensing data. Water Resources Research, 37: 329–340. [Crossref], [Web of Science ®] , [Google Scholar]), show errors slightly larger than 20%, unaided atmospheric parameterization results in errors of about 50% of the mean ET. Our recently proposed model, with atmospheric and surface parameters, keeps the simple form of Priestley and Taylor's original equation while benefiting from remotely sensed data. Estimates with our model would have errors lower than 20%.