Estimating Large Area Evapotranspiration from MODIS Data

Remotely sensed imagery of the Earth´s surface via satellite sensors provides information to estimate the spatial distribution of evapotranspiration (ET) especially in a wide range. The Surface Energy Balance Algorithm for Land (SEBAL) was used to derive ET maps from Moderate Resolution Imaging Spectroradiometer (MODIS) images. However, SEBAL contains a simple adjustment procedure for the windspeed from only one meteorological station. The Hai river basin with 54 meteorological stations covers 318, 200 square kilometers and the weather conditions is not uniform across the satellite image of this huge basin. Therefore, the most appropriate solution is to separate an image into several subareas based on the weather condition, and operate SEBAL model separately, using different cold and hot pixels and weather data. But based on the above method, the process for calculating 24-hour evapotranspiration (ET_daily) for 365 days in 2007 is too difficult and complex. So we propose a simplified method to calculate the friction velocity by wind speed. The result shows that the simplified method can be used to predict ET by the comparison of ET_daily predicted for August 13, 2007 in Hai river basin by two different methods.