Integration of LSA SAF ET products and MODIS thermal infrared images for improved prediction of forest evapotranspiration

A recent paper has proposed the calibration of a water balance method (NDVI-Cws) at for improving the estimation of actual evapotranspiration (ETa) in forest areas. The calibration concerns the ecosystem sensitivity to water stress (WS) and is obtained using Land Surface Analysis Satellite Application Facility (LSA SAF) evapotranspiration products. The current work addresses the spatial resolution issue introduced by the large pixel size of these products (about 5 km) relying on the psychrometric constant theory, which postulates the existence of a linear relationship between differences in evapotranspiration fraction (ETf) and in land surface temperature (LST). This relationship allows the use of MODIS LST images for improving the spatial resolution of LSA SAF ETf estimates to about 1 km, which is currently investigated using datasets taken in Tuscany (Central Italy). Spatially enhanced ETf maps of June and August are yielded and assessed to estimate the variable impact of WS in the region. These ETf maps are then used for calibrating the NDVI-Cws method and predicting ETa in a forest site only partially affected by WS. The experimental results testify to the efficiency of the proposed method in spatially enhancing the LSA SAF ET products and improving ETa monitoring in fragmented forest areas.