Remotely sensed evapotranspiration estimations have been used for the calculation of the water balance in hydrographical basins, the definition of aquifer recharge, and the monitoring of drought events. However, there are limitations on local scale applications, such as agricultural management and hydrological modeling that arise from the lack of spatial resolution or the low frequency of image acquisition of current polar satellite constellations. The present article puts forward a procedure to improve the spatial resolution of evapotranspiration estimations obtained from the Moderate Resolution Imaging Spectroradiometer (MODIS), from 1 km to 500 meters resolution. The procedure is based on the disaggregation of the thermal band before estimating of evapotranspiration. Comparison with the ASTER sensor, at 500 meters resolution, indicates a mean error of 25 W m-2 and a correlation coefficient of 0.95.
thermal; remote sensing; hydrological models
