Evaluation of the Effect of Pre‐processing of the Remotely Sensed Data on the Actual Evapotranspiration, Surface Soil Moisture Mapping by an Approach Using Landsat, DEM and Meteorological Data

Abstract A practical approach that combines Landsat TM, DEM and in situ conventional meteorological data to estimate the actual evapotranspiration (ET), and to predict the surface soil moisture content (SSMC) in a heterogeneous terrain was proposed and applied to the Urumqi River Basin, Tianshan, China. The successful application of the proposed approach to this meso‐scale watershed was found closely related to the procedures involved in the preprocessing of the remotely sensed data. The detailed investigation suggested that the atmospheric correction procedures is very sensitive to the resulted land cover classification, NDVI and surface albedo retrieved, which would ultimately affect the correct estimation of actual ET and accurate prediction of SSMC. This paper starts from the detailed description of the procedures in the preprocessing of the Landsat TM, DEM data and their combinations for the proposed ET, SSMC mapping approach. The different procedures were selected in the pre‐processing of the remotely sensed data to examine to what degree they will affect the final performance of the proposed approach on actual ET, SSMC estimations over the heterogeneous terrain on the Urumqi River Basin, Tianshan, China.